Datasets:

BioLaySumm
/

BioLaySumm2025-PLOS

Dataset card Data Studio Files Files and versions Community

Split (3)

train · 24.8k rows

Search is not available for this dataset

article stringlengths 4.36k 149k	summary stringlengths 32 3.35k	section_headings listlengths 1 91	keywords listlengths 0 141	year stringclasses 13 values	title stringlengths 20 281
Thermotherapy is an accepted alternative therapy for new-world cutaneous leishmaniasis , but current heat-delivery modalities are too costly to be made widely available to endemic populations . We adapted a low-cost heat pack named the HECT-CL device that delivers safe , reliable , and renewable conduction heat . 25 patients with cutaneous leishmaniasis completed treatment with the device at an initial temperature of 52°C±2°C for 3 minutes to each lesion , repeated daily for 7 days , and were followed up for 6 months by direct observation . The overall definitive clinical cure rate was 60% . Concurrently , 13 patients meeting minimally significant exclusion criteria received identical compassionate use treatment with a cumulative definitive cure rate of 68 . 4% , 75% for those who had experienced CL relapse after prior antimonial treatment . Therapy was well tolerated . Reversible second-degree burns occurred in two patients and no bacterial super-infections were observed . HECT-CL is a promising treatment and deserves further study to verify its safety and efficacy as adjuvant and mono- therapy . Cutaneous leishmaniasis ( CL ) is a sand fly transmitted protozoan disease on the WHO list of Neglected Tropical Diseases with an estimated incidence of l . 5 million new cases yearly [1] . Peru hosts L . ( V ) braziliensis and L . ( V ) guyanensis , which are geographically distributed in the jungle , and L . ( V ) peruviana , which is distributed in the Andean highlands [2] . Administration of the first-line chemotherapy of pentavalent antimonials ( Sb5+ ) is burdened by a high cost , a 20-day treatment course under expert supervision , frequent side effects [3]–[6] , and rising clinical resistance [7]–[9] . Amphotericin B ( AMB ) is the only available second-line therapy in Peru , and is even more problematic with respect to cost , side effects , and treatment delivery [10] , [11] . Berman and Sacks demonstrated several pathogenic Leishmania species to be thermosensitive from 37 to 39°C in vitro [12] , [13] . Subsequently , thermotherapy ( TT ) has been evaluated in a variety of CL species and via a variety of heat-delivery modalities [14]–[17] . The ThermoMed™ device , which utilizes radio-frequency ( RF ) technology remains the most supported by randomized clinical trials and is WHO recommended as an alternative therapy for all American CL species [1] . In Peru , the high cost of this device , healthcare infrastructure limitations , and poverty in endemic areas has restricted its use to research and military settings [18]–[20] . We adapted a reliable , safe , and low-cost technology named the Hand-held Exothermic Crystallization Thermotherapy for Cutaneous Leishmaniasis ( HECT-CL ) . The HECT-CL is a sodium acetate heat pad calibrated to produce 52±2°C for greater than 3 minutes; it costs less than 3 dollars , is simple to use , and is rechargeable by boiling for recurrent reuse . A supersaturated sodium acetate solution and flexible metal disc are contained inside a sealed plastic pouch . Flexing the disc provides a nanoscopic nidus that nucleates an exothermic liquid-to-solid phase change reaction releasing a reliable maximum temperature at 52±2°C . The liquid phase is restored when boiled . Heat packs operating with equivalent technology have been sold commercially as hand warmers and for treatment of athletic injuries [21] , and sodium acetate is a food additive and is commonly infused intravenously with parental nutrition . In this study we evaluate the safety and efficacy of HECT-CL in in Peruvian patients with CL who had previous sodium stibogluconate ( SSG ) treatment and in those not previously treated with L . ( V ) peruviana infection The pilot study was performed at the Leishmaniasis Clinic , Institute of Tropical Medicine Alexander von Humboldt – Hospital Nacional Cayetano Heredia , in Lima , January through December 2011 . Twenty-five subjects with parasitologically confirmed CL who were likely infected in Peru were included . Exclusion criteria were ( 1 ) age under 8 or older than 80 years old , ( 2 ) facial lesions located less than 2 centimeters from mucosal surfaces ( such as the nose , mouth , eyes or ears ) , ( 3 ) maximum area of 15 cm2 ( diameter greater than 4 centimeters ) , ( 4 ) more than 4 lesions , ( 5 ) L . ( V ) braziliensis or L . ( V ) guyanensis disease without prior systemic therapy , ( 6 ) concomitant mucosal leishmaniasis ( ML ) , ( 7 ) severe or immunocompromising medical illness , ( 8 ) having received therapy for CL in the prior month , or ( 9 ) inability to commit to follow-up appointments for the proximal six months . The study received Institutional Review Board approval from both Tulane University School of Medicine and Hospital Nacional Cayetano Heredia . All adult participants provided written informed consent . Children less than 18 years of age were only included if written informed consent was provided by the participant's parent or guardian in addition to written agreement from the participant . This clinical trial was registered in Clinicaltrials . gov ( NCT01277796 ) . Subjects underwent a thorough history and physical examination with particular attention to exclude mucosal involvement . Lesions were measured using a digital caliper to record the largest extending diameter of and the corresponding perpendicular diameter . Standardized coded digital photographs were taken . Specimens for parasitologic testing were collected by scraping and non-invasive sampling , and diagnosis and speciation were confirmed by conventional PCR [22] . The primary outcome measure was efficacy of HECT-CL according to prior treatment status and the causative species ( L . ( V ) . braziliensis , L . ( V ) . peruviana , or L . ( V ) . guyanensis ) . Treatment response ( TR ) was staged using the following clinical criteria: M0: No improvement . Lesion remained active , having the same characteristics or becoming larger than prior to the start of treatment . MI: Size of the lesion decreased 50% in comparison with the initial lesion , with fewer inflammatory signs and discrete re-epithelialization . M2: Size of the lesion decreased between 50–90% in comparison with the initial lesion , and left few inflammatory signs . M3: Size of the lesion decreased more than 90% , with re-epithelialization and very little inflammation . M4: Complete re-epithelialization with a characteristic scar and no inflammation . Standardized clinical evolution was evaluated based on the aforementioned TR staging as: Secondary outcomes included pain , burn grade , and super-infection . Pain was evaluated by the Baker-Wong Likert emoticon-word rating scale [23] . Burn grade was noted as first-degree by the presence of erythema , or reversible ( partial thickness ) second-degree burn by painful reversible blistering . Super-infection was screened for by any persistent or progressive erythema or purulence . Each lesion was debrided and cleaned with sterile physiologic saline solution and the HECT-CL device activation temperature of 50–54°C was verified by an infrared thermometer ( Figure 1 , a–c ) . The HECT-CL device was applied only if temperature was in the range of 51–53°C given the ±1°C limitation in thermometer accuracy . The device was applied daily for 3 minutes in 1–3 fractions ( according to individual pain tolerance ) for 7 days; every application included at least 1 cm of “healthy skin” outside of the CL lesion . HECT-CL devices are malleable and the application area was delimited by careful hand application . Injectable lidocaine was available by request prior to direct application of HECT-CL . Local adverse events after HECT-CL were evaluated before and 30 minutes after heat therapy on days 1–7 and on scheduled clinical follow-up at 2 weeks , 1- , 2- , 3- , and 6 months . Study protocol allowed for HECT-CL interruption and initiation of a second-line therapy as soon as clinical suspicion of worsening was identified according to medical and ethical principles and oriented to reduce cosmetic implications especially in sensitive cosmetic lesions . CL was confirmed by direct amastigote identification on smear ( Giemsa staining ) and/or kDNA PCR . The principal causative species in Peru , L . ( V . ) braziliensis , L . ( V . ) peruviana , and L . ( V . ) guyanensis were differentiated by sequence targeted assays following the stepwise approach described by Veland , et al: a ) Mannose phosphate isomerase gene ( MPI ) which distinguishes L . ( V . ) peruviana from L . ( V . ) braziliensis and L . ( V . ) guyanensis , b ) The cysteine proteinase B ( Cpb ) gene which distinguishes between L . ( V . ) braziliensis and non-L . ( V . ) braziliensis species , c ) Heat shock protein 70 ( hsp70 ) which distinguishes between L . ( V . ) guyanensis and non-L . ( V . ) guyanensis species and d ) an 870-bp fragment of Leishmania glycoprotein of 63 kDa ( gp63 ) [24] . Statistical analyses were performed using STATA software , version 10 . 0 . Student's 2-tailed t test was used to compare the means of continuous variables ( i . e . , age , duration of disease , and number of lesions ) . Median values were compared using the Wilcoxon rank-sum test . Differences in clinical status were compared using the Fisher's exact test . Patients were analyzed as a group and then stratified according to prior treatment status . Differences were considered significant when p values were <0 . 05 . Parallel to the 25 patients enrolled in this Pilot study , 13 additional patients were treated with HECT-CL on compassionate grounds because they were deemed ineligible to receive Sb5+ or AMB due to one or more of the following: age less than 8 years old , a social or financial conflict for a 20-day treatment course , a medical contra-indication , and/or previous failed treatment prior to HECT-CL therapy . Each patient or his or her legal representative signed an informed consent for publication of clinical information , which was approved by the UPCH Institutional Review Board . The 13 additional patients were largely female ( 61 . 5% ) with a median age of 7 years ( Range: 3–52 ) who acquired the infection predominantly in highland regions ( 76 . 9% ) . All patients had L . ( V ) peruviana as the causative parasite ( eleven of whom had received prior Sb5+ ) , with a median duration of disease of 26 weeks ( range 8–106 weeks ) . The previously treated lesions generally had relapsed as nodules over the inactive post-treatment scar , later progressing with erythema and induration with a striking clinical similarity to Leishmaniasis Recidivans Cutis ( LRC ) described in old world CL [25] . At presentation these lesions tended to be multiple , nodular , and ulcerative: 12 of 13 patients had more than two lesions . At the end of HECT-CL therapy , day 7 , ten of 12 subjects had stage improvement and one patient demonstrated complete re-epithelialization ( M4 ) . Clinical cure ( M4 ) was subsequently observed in 6 of 13 patients at fifteen days after HECT-CL therapy end , 12 of 13 patients at 1 month , and all 13 patients at 2 months . However , at 3 months follow up , 1 patient experienced re-ulceration , and another patient the appearance of satellite subcutaneous nodules despite complete re-epithelialization of the treated ulcer . At 6 months after treatment , 11 of 13 patients experienced definitive clinical cure ( 84% ) . Post-hoc analysis including the 13 patients treated compassionately and the 25 pilot study subjects , demonstrates a definitive clinical cure rate of 68 . 4% ( 26/38 ) , 60% ( 9/15 ) in the NT group and 73 . 9% in the PT group . This study demonstrates that HECT-CL is safe and is promising to be efficacious . While the natural evolution of untreated New World CL is still poorly described , the rapid improvement observed with HECT-CL is striking . The definitive clinical cure rate reported here of 60–68 . 4% rivals the efficacy estimates of pentavalent antimonials for treatment CL in South America of 76 . 5% [26] , as well as broad variable efficacy of standard Radiofrequency thermotherapy ( RF-TT ) ranging from 38 to 90% with a variety of species [18]–[20] , [27] , [28] . The study also found HECT-CL to be associated with clinical cure of the three most common species in Peru , with the caveat that there may be regional variation in virulence and heat-susceptibility within species also endemic in other countries [29] , [30] . Clinical cure rates of reactivated lesions after Sb5+ treatment have been reported to be even lower than in treatment naïve patients [31] , suggesting that HECT-CL may be preferred in this setting . The HECT-CL safety profile was notable for a cumulative reversible 2nd degree burn rate of 0 . 8% , with 2 events out of 266 applications ( with fractionated heat deliveries grouped as a single application ) . In comparison , standard RF-TT can be complicated by up to a 93% 2nd degree burn rate and 19% wound infection rate per treatment [18] . While RF-TT heats dermal tissue directly , the HECT-CL conduction modality may enjoy a superior safety profile secondary to physiologic dermal-protective thermoregulation via vasodilatation and convection heat transfer . Of note , both episodes of HECT-CL reversible 2nd burn were observed after treatment initiation at 53°C±1°C , suggesting to limit initial treatment temperatures to 52±1°C in future investigations . While the RF-TT standard dosing delivers 50°C over 30 seconds accurately and precisely , optimal heat delivery has not been assessed by phase 2 trial . HECT-CL devices increased skin temperature in a fixed range of 41–45°C that was unquestionably therapeutic and considerably better tolerated than RF-TT . In the combined group reported here , 37 of 38 patients demonstrated initial improvement after HECT-CL application with an excellent safety profile , suggesting a role for elongated or recurrent treatment courses in the unsatisfactory lesion response . The reason why our pilot study used an extended therapy ( 7 days ) intervention contrasting with a single application of conventional RF-TT is due to concerns about the efficacy of a single application with HECT-CL , which was not tested before as a thermotherapy device for treatment of CL . Compared with the normal RF-TT device , there was no medical evidence to support that a single application of HECT-CL is similarly effective . For this reason it is necessary to design future controlled clinical trials considering modifications of current thermotherapy procedures to identify if prolongation or reduction of applications offers the best therapeutic outcomes . Future trials should be implemented by immunological studies to support or refute the systemic immune response hypothesis after local heat treatment and identified molecular and cellular phenomena during heat therapy . The observation of some recently infected NT patients developing satellite subcutaneous lesions despite complete resolution of the index lesion ( Fig . 4g–i ) could indicate these resulted from a high parasite load disease . This phenomenon was previously reported by Unger et . al in patients with young lesions treated with Sb5+ in whom failure rate was comparatively higher than in patients with well-established ulcerative lesions [32] . This hypothesis is supported by our post-hoc analysis findings in which subjects were primarily antimonial experienced with lesions that were focal , that likely contain a low parasite load , and demonstrate higher clinical cure rate with HECT-CL ( 84% ) than the pilot population . Still there is limited insight into the TT physical and biological mechanisms of action beyond the general concepts of heat-dependent parasite destruction augmented by local and systemic immunity [14] . Our findings highlight the importance of further study to better characterize species-specific amastigote heat-tolerance , parasite load and mechanism of extension , as well as host immunologic factors . In the meantime , our findings suggest the lesions most effectively treated with HECT-CL may be those with focal relapse after treatment , or older lesions ( >2 months ) with a lower parasite load . There is reasonable concern that local therapy may not protect against mucocutaneous progression in those patients infected with L . ( V ) . braziliensis . However , neither Sb5+ or other systemic treatments have been strongly demonstrated to protect against ML , and there is evidence that heat therapy may induce a systemic response [1] . All of our subjects infected with L . ( V ) . braziliensis or L . ( V ) . guyanensis had received prior standard Sb5+ treatment , and until ML protection is better characterized , caution should be exercised with rural field use of TT monotherapy in ML regions . In the meantime , TT remains a promising candidate for rural field use in Peruvian regions devoid of ML , and in the tertiary care centers to improve efficacy , safety , and cost of the current standard therapy . Our results suggest that L . ( V ) . braziliensis can be refractory to HECT-CL as used in our protocol . However , our L . ( V ) . braziliensis patients were not representative of typical L . ( V ) . braziliensis infections since all were included after receiving prior unsuccessful standard pentavalent antimonial treatment . Theoretically L . ( V ) . braziliensis might be more refractory to HECT-CL , as in the case of other therapies [29] , but due to the absence of molecular or immunological evidence supporting lower thermosensitivity in L . ( V ) braziliensis strains , we are unable to speculate about differences in treatment outcome according to the infecting strain . The potential benefit of a safe , low-cost , easy to use , and efficacious therapy is particularly relevant in low resource countries like Peru where standard Sb5+ or second-line non-liposomal AMB treatment course requires 20-days of directly observed treatment in a tertiary care setting . Such a commitment is often either prohibitive or financially catastrophic for low-income rural populations . Therefore , carefully designed exothermic crustallization conduction heat therapy should be explored as mono- or adjuvant therapy to make treatment more widely available in endemic regions , and to potentially shorten the duration of directly observed toxic and/or expensive therapy .	American cutaneous leishmaniasis is an endemic parasitic disease in Peru , with more than a reported thousand cases per year associated with significant disabilities and economic impact . In comparison to old-world cutaneous Leishmaniasis where the infection generally heals on its own , American cutaneous Leishmaniasis is a chronic skin disease and its treatment can be expensive , difficult to administer , and associated with significant adverse events . We developed a heat treatment using a novel and low-technology device named Hand-held Exothermic Crystallization Thermotherapy for Cutaneous Leishmaniasis ( HECT-CL ) . This device produces a stable thermal reaction ranging from 50–54°C which is considered therapeutic and has enormous benefits for its application in real world conditions , as it is extremely low cost , simple to use , and highly tolerable in patients receiving this treatment . Our pilot evaluation found cure rates close to those reported with the standard antimonial treatment and likely higher in cases of relapsing disease . Additionally , fewer severe adverse events were observed than reported with the use of other currently available heat therapies . Our objective is to show that HECT-CL therapy could be instituted in rural and real situations with limited health care infrastructure and with less cost and fewer side effects than standard antimonial treatment .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "leishmaniasis", "medicine", "neglected", "tropical", "diseases", "infectious", "diseases" ]	2013	Novel Low-Cost Thermotherapy for Cutaneous Leishmaniasis in Peru
Chikungunya is an endemo-epidemic infection , which is still considered as an emerging public health problem . The aim of this study was to evaluate in a 65+ population , the accuracy of two chikungunya screening scores that were developed in younger people . It was performed in the Martinique University Hospitals from retrospective cases . Patients were 65+ , admitted to acute care units , for suspected Chikungunya virus infection ( CVI ) in 2014 , with biological testing using Reverse Transcription Polymerase Chain Reaction . Mayotte tool and Reunion Island tool were also computed . Sensitivity , specificity , positive predictive value , negative predictive value , and Youden’s statistic were calculated . In all , 687 patients were included , 68% with confirmed CVI , and 32% with laboratory-unconfirmed CVI . Fever ( 73 . 1% ) and arthralgia ( 51 . 4% ) were the most frequent symptoms . Sensitivity ranged from 6% ( fever+headache ) to 49% ( fever+polyarthralgia ) ; and Youden’s index ranged from 1% ( fever + headache ) to 30% ( fever+polyarthralgia ) . PPV and NPV ranged from 70% to 95% , and from 32% to 43% , respectively . Performances were very poor for both tools , although specificity was good to excellent . Our results suggest that screening scores developed in young population are not accurate in identifying CVI in older people . Chikungunya virus infection ( CVI ) is still considered as an emerging public health problem in both tropical and temperate regions [1] . It is usually symptomatic and may have three phases: acute ( day ( D ) 1 to D21 ) , post-acute ( D21 to D90 ) , and chronic stage ( beyond D90 ) [2 , 3]; the latter two are sometimes absent . In the acute stage of infection , typical physical signs and symptoms of CVI are febrile illness associated with severe and debilitating polyarthralgia affecting the small joints . Severe functional disabilities characterise this phase . Other signs that can be observed include myalgia , headaches , or maculo-papular rash . In most cases , symptoms resolve within a few days with symptomatic treatment [2 , 3] . Prior to the outbreak of 2005–2006 in Reunion Island ( France ) , CVI was not considered to be life-threatening . Usually , the any cause overall mortality rate from CVI is considered to be low , comparable to that of seasonal influenza [4] . However , several studies have shown that mortality rates increased during the outbreak as compared to the same period in previous years [5–8] . Fatality increases in populations with atypical presentations , and the incidence of such atypical , serious or fatal cases increases with age . Indeed , age over 85 years has been shown to be associated with increased mortality [8] , and the mortality rate is five times higher in subjects aged 65 years or older ( 65+ ) than among those under 45 years [5] . On Reunion Island , excess mortality concerned mainly people aged 75 years or older ( 75+ ) [9 , 10] . Several comorbidities as well as increased age are linked with atypical presentation [11] . During epidemics , CVI prevalence rates are not fully known , and vary from 18% to 48% [12–14] . To meet patients’ needs , rapid and reliable diagnosis is required . Patients with CVI should be identified early , and receive appropriate care . Moreover , people with symptoms and signs consistent with CVI but who suffer from another type of disease must be diagnosed rapidly . Management without delay of differential diagnoses is essential . However , establishing a diagnosis of CVI in a simple and reliable way is very challenging . This concern is especially relevant to the frail elderly population . Furthermore , diagnosis based solely on physical examination may underestimate the magnitude of the epidemic [13] . The systematic use of biological diagnosis during an outbreak is not feasible , especially in low- and middle-income countries ( e . g . due to lack of access to laboratory testing , difficulties processing samples , delays in the treatment of patients , etc . ) . The use of predictive scores would thus be very helpful in this situation . During the outbreak in Mayotte and Reunion Island , two predictive scores were developed . Sissoko et al . [15] retrospectively derived a clinical score ( Mayotte tool ) in a population of children and young adults . This score was based on the pairing of fever with the four most common clinical signs ( polyarthralgia , myalgia , headaches , and back pain ) . More recently , Thiberville et al . [16] established a clinico-biological score ( Reunion Island tool ) from a population of patients aged 18 to 65 years . The performances of these scores were good , making them useful screening tools . However , they have not been evaluated in the elderly . Thus , we aimed to evaluate diagnostic performances of these two scores in a 65+ population , admitted to acute care units of Martinique University Hospital , with symptoms suggestive of CVI during the epidemic that occurred in 2014 . This was a diagnostic study performed in the University Hospital of Martinique ( French West Indies ) from retrospective cases . Eligible patients were aged 65 years or older , admitted to acute care units including the emergency department ( ED ) , for suspected CVI ( presence of fever or arthralgia at admission based on Rajapakse et al 2010 ) , from 10 January to 31 December 2014 , and who underwent biological testing using Reverse Transcription Polymerase Chain Reaction ( RT-PCR ) . Patients whose clinical and/or biological data were missing in their medical records , as well as those for whom it was not possible to compute either Mayotte tool or Reunion Island tool , were excluded . We recorded baseline characteristics including age , sex , time since onset of Chikungunya symptoms , as well as presence or absence of the following features: fever , arthralgia ( any of the following: knee , ankle , metacarpo-phalangeal joints , wrist , elbow , shoulder girdle , and pelvis ) , myalgia , digestive or neurological symptoms , and comorbidity burden ( assessed using Charlson’s comorbidity index [17] ) . The Charlson’s comorbidity index measures patient comorbidity using the tenth International Classification of Diseases Diagnoses Codes . Each comorbidity has a weight ( from 1 to 6 ) depending on its severity . The higher the score , the higher is the comorbidity burden . Biological testing included: white cells , neutrophils , lymphocytes , and RT-PCR . All patients included in this study had serum samples tested using RT-PCR with the RealStar® Chikungunya RT-PCR Kit ( Altona Diagnostics GmbH , Hamburg , Germany ) . We considered as confirmed CVI all suspected cases in whom biological confirmation was obtained by positive RT-PCR . The Mayotte tool and Reunion Island tool were calculated for all patients . The study was performed in accordance with the Declaration of Helsinki , and was approved by the “Commission Nationale de l’Informatique et des Libertés” ( CNIL ) : authorisation number 1898399 v 0 . Patient’s data was completely anonymised according to the CNIL requirements . All data was solely accessed and analysed retrospectively from the University Hospital of Martinique . The sample size was estimated based on the expected precision of sensitivity ( Se ) and specificity ( Sp ) confidence intervals . In a previous study [15] , the prevalence of symptomatic CVI was 28% ( 318/1154 ) . For an expected Se and Sp of 90% each , with a precision of 5% , and an alpha error of 5% , the estimated sample size was 192 for Se , and 494 for Sp . Therefore , we planned to include at least 494 patients . In the acute phase , RT-PCR was considered as the gold standard to identify subjects with or without CVI . Sensitivity ( % ) , specificity ( % ) , positive predictive value ( PPV , % ) , negative predictive value ( NPV , % ) , and Youden’s index ( J = Sensitivity ( % ) + Specificity ( % ) – 100 ) were estimated . Youden’s index is a single statistic that captures the performance of tests . Its value ranges from -100% ( totally useless test ) to 100% ( perfect test ) . Quantitative variables are described as mean ± standard deviation , and categorical variables as using number and percentage . Baseline characteristics were compared according to RT-PCR results using Student’s t-test ( continuous variables ) and chi2 test ( categorical variables ) Statistical analyses were performed using SAS release 9 . 4 ( SAS Institute Inc . , Cary , NC , USA ) . During the study period , 894 patients were potentially eligible . Among these , 207 were excluded . A flowchart of the study population is shown in Fig 1 . Excluded subjects did not significantly differ from subjects included in terms of age ( 79 . 0±8 . 0 vs . 80 . 4±8 . 0 years , respectively ) or sex ( 49% vs . 51% women , respectively ) . In all , 687 patients were considered in the present study . The mean Charlson’s comorbidity score was 1 . 7±1 . 9 . The average time between onset of symptoms and admission was 1 . 3±2 . 3 days . Clinical and biological characteristics at admission to hospital are presented in Table 1 . Fever ( 73 . 1% ) and arthralgia ( 51 . 4% ) were the most frequent symptoms . The knee ( 22 . 3% ) , and the ankle ( 19 . 1% ) were the most frequent sites of arthralgia . For biological characteristics , 77 . 9% of patients had a neutrophil count< 7500 , and 61 . 3% had a lymphocyte count <1000 . Patients with positive RT-PCR ( chik+ ) CVI ( n = 467 ) and patients with negative RT-PCR ( chik- ) CVI ( n = 220 ) did not differ significantly with respect to age ( 80 . 6±7 . 8 versus 80 . 0±8 . 3 , respectively; p = 0 . 33 ) , sex ( female sex 45 . 9% versus 52 . 9% , respectively; p = 0 . 09 ) , or Charlson’s comorbidity score ( 1 . 6±1 . 8 versus 1 . 7±1 . 9 , respectively; p = 0 . 73 ) . Performance indicators of the Mayotte tool and the Reunion Island tool are presented in Table 2 . Sensitivity ranged from 6% ( for fever+headache ) to 49% ( for fever+polyarthralgia ) . Youden’s index ranged from 1% ( for fever+headache ) to 30% ( for fever+polyarthralgia ) . PPV and NPV ranged from 70% to 95% , and from 32% to 43% , respectively . Our study shows that the diagnostic performance of two scores to screen for potential CVI , both developed in younger populations , is poor among older patients , as shown by the associated Youden’s index . While the specificity and the PPV of the scores are good to excellent , the sensitivity and NPV are mediocre , not to say poor . The specificity of the Mayotte tool [15] was 81% in our series , which was only slightly lower than the 89% reported in Sissoko’s seminal study . Regarding the Reunion Island tool [16] , its specificity in our series was excellent , at 97% , compared to 85% in the original population . Conversely , the sensitivity of both scores was poor in our series; at 49% for the combination of fever plus polyarthralgia ( for Mayotte tool ) , and 23% for Reunion Island tool . The authors of both these scores reported higher sensitivity ( 80% and 84% respectively ) . Using the clinical features score to compare three other pairs of symptoms found even lower sensitivity rates . These differences are likely due to the different clinical profiles observed in elderly subjects , which renders the use of scores developed in young populations perilous . In our study , the average age was 80 . 4 years , with an average comorbidity index of 1 . 7 , underlining the geriatric profile of our population . In the two scores we tested , the average age in the development cohorts were 27 . 2±16 . 8 years for the Mayotte tool , and 40 . 1±12 . 4 years for the Reunion Island tool . Indeed , Mayotte tool is based on signs of fever plus polyarthralgia , which were present in 83 . 6% of the chik+ patients . In our series , this pair of symptoms was only observed in 48 . 6% of chik+ cases . This variation in the clinical profile of elderly subjects has previously been reported by other authors , who suggested that the incidence of atypical , severe or fatal cases increases with age [5] . In the Reunion Island tool developed by Thiberville et al . [16] , the presence of fever and polyarthralgia were among the inclusion criteria , and therefore present in 100% of subjects . In our population , these two symptoms were found in 79 . 4% and 62 . 5% of chik+ patients , while we observed lymphopenia in 75 . 3% of chik+ subjects , compared to 79% in Thiberville’s study [16] . The symptom profile observed in our study was less specific , with fewer rheumatological symptoms than usually described in the semiology of CVI [3] . Modifications in clinical presentation in elderly people are frequently observed in general practice [18 , 19] . In many cases , the primary complaint is rarely directly related to the precipitating event . This phenomenon has been widely studied , and led to the modelling of clinical presentations in elderly subjects by Fried et al . [19] . Fried’s diagnostic models take account of comorbidities , as well as the influence of functional and psychosocial factors . Indeed , the classical model in which symptoms correspond to those habitually observed in the causal disease is rarely the norm . Frequently , the physician ( and/or the patient ) may attribute recent symptoms to a known disease , whereas the symptoms may in fact be the result of an acute affection . Fried and colleagues called this the attribution model and facilitating complaint , whereby the concern identified at presentation to medical care was not the major underlying problem . In another model , termed the causal chain model , an elderly subject , often frail with multiple diseases , experiences an acute event that disturbs the patient’s fragile health equilibrium , and subsequently precipitates a chain of complications that may mask the initial events and/or aggravate co-existing diseases . All of these models illustrate the complexity of establishing an accurate diagnosis in this special population , especially using signs that were initially observed in a younger population . Mediocre or poor sensitivity has major implications for the implementation of adequate treatment of CVI , even though treatment is mainly symptomatic . In older people , the problem is twofold . On one hand , sudden functional disability and loss of autonomy may lead to health complications ( falls , dehydration , pressure ulcer , delirium , etc . ) . On the other hand , CVI may aggravate chronic disorders with possible adverse outcomes . In addition , older people may present atypical signs , which expose them to inadequate patient care due to serial misdiagnoses ( differential diagnosis like dengue fever , leptospirosis , or bacterial infection ) . The lack of validated tools for use in elderly patients is a common problem in routine care . Although a small number of screening tools or predictive scores have been validated for use in the elderly ( e . g . the Mini Nutritional Assessment [20 , 21] , gait speed [22] , or the timed “Up and Go” test [23] ) , many other instruments are widely used on a daily basis to aid management of elderly populations without robust scientific evidence confirming their clinimetric properties ( e . g . the Wells score , or the Short Physical Performance Battery [24 , 25] . Our study presents several strengths . Firstly , the sample size is very large , and includes specifically elderly patients ( older age and higher comorbidity scores ) . The number of missing data per variable is also very low ( 3% at most ) . This provides a robust basis for results observed . The clinical and biological data were recorded by geriatric medicine and virology physicians from the hospital’s medical informatics system , with cross-checking from the patients’ medical records . Furthermore , confirmation of the diagnosis of CVI was obtained by RT-PCR using the same kits for all the subjects included in the study . Several limitations deserve to be addressed . We did not use serological testing to confirm CVI diagnosis . This could have impact in our results because people who have presented later their infection could have been misdiagnosed when using only RT-PCR . This would be very unlikely as patients for whom delay from onset symptoms to biological testing exceeded 48 hours were excluded from our study . The retrospective nature of the study could have been a limitation . Indeed , it would have been relevant to compare the performances of the Mayotte and Reunion tools in Martinique with the younger population they were developed in before comparing them in older population . Our population could be not representative of the overall elderly cases . The existing Mayotte tool and Reunion Island tool to predict CVI , developed in populations of younger patients , are not useful for the detection of CVI in 65+ patients . Population ageing and the likely recurrence of other epidemics of this virus justify the development of a specific clinical and/or clinico-biological score for elderly subjects in order to ensure early diagnosis and adequate management .	Chikungunya virus is an alpha-virus transmitted by Aedes egyptii or albopictus bites . This infection is still considered as an emerging public health problem . In the acute stage of infection , typical physical signs of Chikungunya virus infection are febrile illness associated with severe and debilitating polyarthralgia affecting the small joints . Several studies have shown that mortality rates increased during the outbreak . Age over 85 years has been shown to be associated with increased mortality , and the mortality rate is higher in 65+ subjects than among younger population . During epidemics , prevalence rates vary from 18% to 48% . Rapid and reliable diagnosis is required especially for frail elderly population . Diagnosis based solely on physical examination may underestimate the magnitude of the epidemic . The systematic use of biological diagnosis during an outbreak is not feasible , especially in low- and middle-income countries . The use of predictive scores would thus be very helpful in this situation .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion", "Conclusion" ]	[ "reverse", "transcriptase-polymerase", "chain", "reaction", "death", "rates", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "elderly", "chikungunya", "infection", "demography", "tropical", "diseases", "age", "groups", "signs", "and", "symptoms", "neglected", "tropical", "diseases", "molecular", "biology", "techniques", "research", "and", "analysis", "methods", "infectious", "diseases", "artificial", "gene", "amplification", "and", "extension", "molecular", "biology", "arthralgia", "people", "and", "places", "pain", "management", "diagnostic", "medicine", "fevers", "population", "groupings", "biology", "and", "life", "sciences", "viral", "diseases", "geriatrics", "polymerase", "chain", "reaction" ]	2017	Do Two Screening Tools for Chikungunya Virus Infection that were Developed among Younger Population Work Equally as Well in Patients Aged over 65 Years?
Continuous attractor networks are used to model the storage and representation of analog quantities , such as position of a visual stimulus . The storage of multiple continuous attractors in the same network has previously been studied in the context of self-position coding . Several uncorrelated maps of environments are stored in the synaptic connections , and a position in a given environment is represented by a localized pattern of neural activity in the corresponding map , driven by a spatially tuned input . Here we analyze networks storing a pair of correlated maps , or a morph sequence between two uncorrelated maps . We find a novel state in which the network activity is simultaneously localized in both maps . In this state , a fixed cue presented to the network does not determine uniquely the location of the bump , i . e . the response is unreliable , with neurons not always responding when their preferred input is present . When the tuned input varies smoothly in time , the neuronal responses become reliable and selective for the environment: the subset of neurons responsive to a moving input in one map changes almost completely in the other map . This form of remapping is a non-trivial transformation between the tuned input to the network and the resulting tuning curves of the neurons . The new state of the network could be related to the formation of direction selectivity in one-dimensional environments and hippocampal remapping . The applicability of the model is not confined to self-position representations; we show an instance of the network solving a simple delayed discrimination task . Multiple state variables can be encoded in the same network . An example is offered by the place representations of several environments [20] , [21] . To each environment corresponds a neural map which is encoded in the synaptic efficacies . Sensory inputs would then select the correct representation , i . e . both the environment and the position in the environment . The selected map wins the competition with the other maps stored in the network , and a localized pattern appears . In this case the network only maintains information about one of the several encoded state variables . A more peculiar property of multiple continuous attractors , is their ability to represent simultaneously the values of several state variables . This property was explored in [28] , where two partially overlapping neural populations ( representing discrete features ) , are assigned two uncorrelated maps . Another example is provided in the study of [29] , where a single network stores and represents simultaneously a continuous and discrete attractors . In principle , given the existence of multiple representations in different brain regions ( either one per region , or many in one region ) , a brain area downstream would necessarily encode several state variables . In light of a Hebbian interpretation on how this encoding takes place , it seems natural to distinguish between two cases . When multiple representations provide a simultaneous input to a region , the result is probably encoded multiplicatively [29] , or , in general , non-linearly . For inputs happening non concurrently , as for instance when walking through several rooms sequentially , an additive encoding of each room is expected [21] . In the following we will analyze additive encoding . The present contribution addresses the issue of encoding correlated maps . The motivations come from recent experimental results on place cells recording in morphed environments [30]–[32] , where place fields remapping along a sequence of morphed arenas is experimentally tested , and from theoretical and experimental studies concerning the morphing of discrete attractors [33]–[35] . In general , we would consider the encoding of manifolds , each of dimension , where . We will refer to a single manifold as a map , once a coordinate system is chosen . The use of uppercase ( e . g . ) or lowercase ( e . g . ) will distinguish between the whole map and a single point on it respectively . Given a pre-synaptic neuron indexed by , and a post-synaptic , the encoding of a single map is obtained using a synaptic matrix , and is such that a continuous attractor representation would arise if it were the only map . We assume , as mentioned above , that the complete encoding arises from a linear superposition of the matrices , . The statistical properties of the maps , and in particular the correlation between them , can be fully specified by providing the probability density . The general problem is too difficult to be studied analytically . Some results can be obtained for the case of uncorrelated maps on the same manifold [27] , though the system can be explored by simulating the full microscopic networks ( see e . g . [21] for the uncorrelated case and [36] for simulation results of the correlated case ) . In order to simplify the analysis , while retaining the basic structure of the problem , we focus on the case of representations , on a 1-dimensional circular manifold ( i . e . the ring model [16] , [37] ) . The correlation between the maps is constructed by limiting the distance between the single neuron locations on the two maps . We devise a simple method to generate a morph sequence between two uncorrelated maps , by linearly modifying the neurons locations between the original maps . This method also suggests a way to test the network response to the exposure of intermediate maps between the two stored correlated maps . For concreteness , one could think about maps of two similar circular arenas , and reason in term of spatial coding . In this context , we are interested in clarifying how the information about the position in the current environment is represented by the network , when varying the constitutive parameters of the model; And how the representation changes when the network is exposed to environments along a morph sequence . In the following we will describe with mean-field ( MF ) theory the attractor landscape of a network , i . e . the stable solutions in absence of any place specific input . We then consider the behavior of the solutions when a spatially tuned input is present . We will establish the approximate relationship between two strongly correlated maps and the encoding of a morph sequence between two reference rings , and study the behavior of the solutions in presence of a tuned input varying along the sequence . Finally we will verify the results with microscopic simulations of finite networks . The network properties can be tested experimentally to confirm ( or falsify ) the attractor hypothesis . In this Section we analyze the fixed point solutions of the system , and heuristically describe the region of stability of these solutions . A more rigorous description of the stability can be found in Methods - Stability . In Methods - Reduced dynamics we derive the dynamics of the order parameters from Eqs . 2 . We report here the result ( 5 ) where the function is defined as ( 6 ) i . e . the rescaled steady state activity profile Eq . 3 . Note that can be eliminated from the right hand sides of the Eqs . 5 , rotating the integration variable . This is possible because there is no spatial dependence in the external input to the network . The first four equations in Eqs . 5 can then be solved independently of the fifth one , since the right hand sides do not depend on . We show in Methods - Solutions properties that , once we have the solution for the variables ( ) , the last equation reduces to . We can thus restrict the analysis to four out of five equations in Eqs . 5 . The elimination of one angular degree of freedom is a consequence of the rotation invariant structure of the encoding , and is the hallmark of continuous attractors arising from spontaneous symmetry breaking . On the other hand , the integrals over in Eqs . 5 are not over the whole circle and we cannot rotate away . Before analyzing the fixed point solutions of the system described by Eqs . 5 , we briefly mention an uninteresting region in the parameters space which can be found also in the classical ring model . This region corresponds to the homogeneous solution , i . e . all the neurons in the network are active at a constant level , and can be obtained from Eq . 2 . The expression corresponding to the line of separation in the plane between the homogeneous solution and the spatially localized bump ( see Fig . 3A , curve surrounding the Homogeneous region ) , is ( 7 ) where . This result is obtained in Methods - Stability , see also below . Let us start by imposing , a restriction that will be addressed later on . The first tree equations at steady state from Eqs . 5 become then equations for the three order parameters : ( 8 ) The first two equations determine the shape of the bump . Given the map specific modulation in the coupling and the distance between the maps , we can derive from the first two equations the size of the bump and the order parameter , representing how close the network representations are to the stored environments and . The last equation gives us the amplitude of the network activity , which also depends on the parameter . As mentioned in Results - Phase diagram of the model , the order parameter can be chosen arbitrarily , due to the rotation invariance of the problem; for simplicity we choose . We deal first with the equation concerning the amplitude of the solution . Given that the activity can be rescaled by changing the value of the applied external current , we are not interested in actually solving the equation . The only requirement is that in order for the solution to be meaningful , i . e . no negative amplitudes are allowed . This requirement translates to a constraint on the inhibition : ( 9 ) We show with stability analysis ( Methods - Stability ) that the critical value , obtained by choosing the equality in the previous expression , corresponds to the onset of amplitude instability; given a choice for the parameters , which specifies the bump shape , for values of the inhibition weaker than the solution grows to infinity . This qualitative behavior was present also in the classical ring model . Fig . 3B shows the values of as a function of for various choices of . In order to stabilize the solutions , the inhibition must grow with increasing and decreasing . Note that it is reasonable to consider the previously mentioned homogeneous solution as a bump with maximal size . In this case the critical can be explicitly computed , and turns out to be . Now we focus on the possible solution . It is easy to see that when , the second of Eqs . 8 is automatically satisfied due to the symmetry of the integrand in ( and ) ; This means that the solution exists everywhere in the parameter space . The steady state activity Eq . 3 with ( and , our initial assumption ) reads ( 10 ) which corresponds to a packet of activity localized in the coordinate , and modulated in , see Fig . 2B for a plot of the activity profile . The remaining fixed point equation can be used to obtain . We refer to the case as a single ring solution; the ring is spanned by the freedom of choice in the angle . In this regime of activity the network is not able to represent separately the environments and , but only the middle environment described by . Even though the solution exists everywhere , it is destabilized in some regions of the parameter space , as shown in the phase diagram ( Fig . 3A , Single ring region ) . By looking at the maximal bump size , we can expect to reproduce the curve separating the homogeneous solution from the single ring . Inserting in the first of Eqs . 8 , it is possible in this case to compute explicitly the integral , which in fact yields Eq . 7 . In order to find the region of existence of the solutions with , we can solve numerically Eqs . 8 in the parameters plane . The result is shown in Fig . 4 , where the color code represents for a given choice of the parameters . It can be seen that there is only a narrow region of high ( low correlation ) and low where such a solution exists . It is important to note that the equations used to find are invariant under the symmetry . This means that both solutions ( ) representing map or are possible . The steady state activity profile in this case looks like: ( 11 ) Given the freedom of choice for the phase , each of this solutions lives on a ring; we call the solution , double ring . An instance of the network activity in this regime is shown in Fig . 2C . The curve separating representations preferring one of the two maps ( ) , and , can be obtained by expanding the second of Eqs . 8 to first order in : ( 12 ) where is the Heaviside step function , and . Dividing by , we get rid of the solution . By finding the zeros of the integral , we select the curve in the parameter space corresponding to the onset of existence of the double ring solution . This curve is shown in Fig . 4 . We have found that the stability of the double ring solution coincides , empirically , with the region of existence of such solution ( compare the phase diagram in Fig . 3A , Double ring region with Fig . 4 ) . Finally , we examine the meaning of the equation for , the order parameter linked to the location of the maximum of the bump in . We have assumed for simplicity , given that a rotation in the integrands in Eqs . 5 is in general not viable due to the restricted range of integration in . Note though , that when the size of the bump is small enough , it is possible to perform the rotation without affecting the value of the integrals; the only requirement is that the rotation keeps the bump from touching the boundaries . In Methods - Solutions properties we verify that there are no solutions with both and different from . We can therefore set in the steady state activity Eq . 3 , and impose the activity itself to be zero on the boundary to findThis equation corresponds to the curve of separation in the plane ( using the relationship , Eq . 8 ) between the single ring solution and a cylinder solution ( Fig . 3A , curve surrounding the C region ) . In this regime , in addition to the freedom of choice for the location of the bump in , the solution is also partially marginal in . The bump can be freely moved on a segment and a circle , defining a cylinder; the activity profile in this case is described by Eq . 4 , see an instance in Fig . 2D . This region extends in the high limit and covers the whole range of correlations . Despite the fact that each of the maps and defines a ring , it shouldn't come as a surprise that the topology of the attractor is a cylinder instead of a torus . The correlation between maps gives rise by definition to a cylinder structure , as can be seen for instance by inspecting Fig . 2B , II . It can be shown that when the cylinder solution degenerates in a torus; the bump of activity can be in any location of the coordinates ( hence , also in ) ) . This regime is linked to the observation of an activity bump simultaneously localized in two environments in network simulations [39] , and the study in [28] . Fig . 3 summarizes the results obtained so far . When is low , the only solutions is a constant level of activity which spreads over the whole network ( Homogeneous region ) . As is increased , the interplay between the short range excitation and long range inhibition creates a pattern of localized activity in the middle map ( Single ring , see also Fig . 2B ) or , if the correlation between maps is small enough , a localized pattern in either or ( Double ring , Fig . 2C ) . Intuitively , the network “remembers” the two maps separately ( , two solutions ) if they are weakly correlated ( ) . When the maps are more similar , the network represents just an average between them ( ) . The bump size decreases with increasing . When is further increased , instead of having a reduced size of the localized activity in just one of the maps , the presence of two stored maps in the synaptic structure and the inhibition produce a packet of activity which looks localized in both maps ( Cylinder solution , Fig . 2D ) . Three particular values of the distance deserve a special mention . The case , corresponding to the encoding of two identical maps , can be shown to be identical to the ring model [37] , as expected . In particular , besides the homogeneous solution and the amplitude instability region , the system can only exhibit the single ring solution . The case , corresponding to the encoding of two uncorrelated maps , does not have the single ring regime as a possible solution . The double ring solution in this case is depicted in Fig . 2A , where it can be seen that the bump is perfectly localized in either maps or , lacking any spatial tuning in the other map . This is the desired outcome in the “multi-chart” approach of [21] . The third case is . We will see in Results - Morphing maps that this case is closely related to the behavior of a network storing a morph sequence between two uncorrelated maps . As can be seen in the phase diagram , the double ring solution is not possible in this regime . How the environment , and the position in the environment , are represented by the network activity ? For the single ring ( Eq . 10 ) and the double ring ( Eq . 11 ) solutions , both characterized by , it is evident that the position is coded by the order parameter . The identity of the environment can only be represented with the ambiguity in the choice of the sign of when the network operates in the double ring regime . In the cylinder regime , it is not clear how the information about the environment is represented in the network , since now the solution is described by and . The following Section is mainly devoted to explore the link between the state variable ( eventually time-dependent ) in the active environment , and the behavior of the solution in this novel regime , by introducing a spatially tuned external input . Until now we considered the condition in which the only external input to the network , , was steady and uniform . Let us introduce a tuned input , for instance in map at position :For simplicity we assume the shape of the external input to be . The parameter measures the strength of the tuned component of the external input as a fraction of the constant baseline we adopted so far . In general what we are interested in , and what is experimentally observable , are the tuning curves of the neurons i . e . their profile of activity as a function of the input angle in the active environment . It is easy to see the effect on the dynamics of the order parameters ( Eqs . 5 ) when the location specific external current is inserted in the original dynamics for the network activity , Eq . 2 . The dynamics keeps the same form as in Eq . 5 , with the exception of the threshold-linear term in , which now reads ( 13 ) where correspond to the choice of map in the input , and for map . With the input at a constant location , one can see that a solution of Eqs . 5 for the single and double ring regime ( ) , is , i . e . the input pinpoints the location of the bump . This implies that , assuming a weak tuned input , the tuning curve of a neuron can be written in the single and double ring regime ( from Eqs . 10 , 11 ) as ( 14 ) and ( 15 ) respectively . The tuning curve in the single ring regime has a maximum for ( hence is the preferred angle for a neuron ) , independently of which map is being used in the external input , as can be seen from Eq . 14 . This implies that each neuron has identical tuning curves in both environments , and that the preferred angle of a neuron does not coincide with either the assigned or but with their average . For the double ring regime , the preferred angle assumes the form ( maximizing Eq . 15 in ) In this case each neuron has two different tuning curves according to the map used in the external input . The preferred angles coincide with the assigned ones ( ) only when the stored maps are uncorrelated ( , hence ) . In the cylinder regime ( , not necessarily ) , a solution for Eqs . 5 in presence of a tuned input is . For an input in map , , the tuning curve would then be proportional to ( from Eq . 4 ) Note that the dependence on means that the external stimulus does not determine completely the network activity , in contrast to what happens in the previously examined regimes . Neurons that respond maximally to the tuned input are then , and , hence . This means that the tuned external input pinpoints the location of the bump maximum in map but the bump is free to stabilize anywhere along the other map given the freedom of choice in ( see activity example in Fig . 2D ) . If several randomly selected external locations in one of the maps are presented to the network , once at time and starting from random initial conditions , the tuning curves would be an average over :where the allowed range for is , see Methods - Solutions properties . The cylinder regime extends the region of existence of two tuning curves per neurons to an higher correlation between the stored maps; the difference is that the coding becomes unreliable: during a single exposure to a given value of the input angle , a neuron could remain silent even if its average tuning curve would predict a response . When the representation refers to the location in an environment , it is natural to think about a smoothly varying location . With a moving input like , the tuning curve depends as before on which map is stimulated , but in a novel way . Assume for simplicity to start from a initial condition , corresponding to ( ) . A moving input in the map would tend to move the bump along that map ( i . e . increase the of the solution ) , while keeping constant ( hence the bump will move to ) . This movement is possible only until the bump reaches the part of configuration space not occupied by neurons due to the distance between maps , see Fig . 2D . At that point , the bump will start to move equally along and , maintaining , which is proportional to , and increasing ( proportional to ) . A similar scenario , but with , is obtained when stimulating the map . If the size of the bump is sufficiently small , this effect has dramatic consequences . The small bump will move along neurons with when a moving stimulus is presented in environment , and viceversa neurons with will be active only when the moving stimulus is presented in environment . As a consequence , neurons will essentially just have a tuning curve ( or field ) , only in one map , and will be silent in the other one . We refer to this phenomenon as dynamical pattern separation ( see Fig . 5 for an example ) . The separation of the activity patterns is essentially a dynamical phenomenon , dependent on the history of the inputs . The figure shows also the robustness of the dynamical pattern separation behavior to the addition of Gaussian -correlated noise in the external current ( see Methods - Numerical Methods ) . Note that neurons characterized by ( i . e . ) , will have tuning curves in the same location . The number of neurons with tuning curves in both environments grows with the size of the bump . Note though that by changing the sign of the velocity in the moving input , the behavior would reverse; neurons with positive ( negative ) would be active during a stimulation in map ( ) . In order to maintain the dynamical pattern separation and the analogy with place coding , one could think about two circular environments , as we did so far , with the additional constraint that the environments can only be traveled , for instance , in the counter-clockwise direction ( CCW ) . As an alternative , the two environments may be thought as the same circular arena , but traveled clockwise ( CW , environment ) and CCW ( ) ; this interpretation would give rise to place fields with directional selectivity ( see Discussion ) . The dynamical pattern separation is basically dependent on the history of the input ( positive or negative velocity ) , in addition to the identity of the map used in the stimulation . This history dependence is present also for non smooth time-dependent stimuli , as for instance the sequential presentation of stimuli with an intervening delay period . In this case the history dependence gives rise to a memory effect: the current location of the bump following a stimulation depends on the location attained after the previous stimulus presentation . Let us consider a basic example of this phenomenon , where the tuned external input is always presented in map . Consider for simplicity the state of the network being characterized by , as a result of the presentation of stimulus sometime in the past . If we now present a stimulus , the bump will move , through the shortest arc on the map , to the new location . Depending on the stimuli , this movement can happen in two ways . If the shortest arc from to is directed CCW , the bump will move with a positive velocity and will end up being located in the region ( as we previously saw in the case of moving tuned input ) . If the shortest arc is directed CW , then the movement will happen with a negative velocity , and the final location of the bump will be in the region . Hence , by looking at the activity resulting from the presentation of , we know whether the shortest way on the ring to it from is CW or CCW . A similar result can be obtained if the stimulus presentation alternates between map and . If we vary the manifolds on which the maps live , for example to segments instead of circles , the history dependence changes accordingly . For instance , on segments the activity would give us information about the second stimulus being greater/smaller than the first one ( see Discussion ) . In the next section we present a simple ( albeit artificial ) delayed discrimination task which the network can perform by exploiting the memory effect . Let us suppose to have a screen with a circle on it . A first stimulus ( a dot ) appears on the circle at some random location ( described by an angle , ) , for the duration of . This first stimulus is then removed for a delay period of . Then a second stimulus appears at another random angle ; the subject's task is to determine whether the shortest path on the circle from angle to is CW or CCW . The basic idea is that it is enough to look at the network activity ( location of the bump in the axis ) , to determine the relationship between the first and the second stimulus ( see Results - Tuned external input for a description of the idea ) . To test the ability of the network to solve this task , we numerically solve the dynamics for the order parameter ( Results - Phase diagram of the model ) with an external input ( Results - Tuned external input ) mimicking the presentation of the stimuli , for a sequence of trials . We used no inter-trial interval , i . e . the presentation of the second stimulus in the -th trial is immediately followed by the presentation of the first stimulus in trial . The time courses of the bump location on the axis ( ) in two example trials for which , are shown in Fig . 6A . When looking at the location of the bump in the axis at the end of a trial , there is a clear difference between the two cases of shortest CW , corresponding to positive ( in the specific example ) , or CCW arcs ( , where ) . Fig . 6B shows that the bump location at the end of trial , can be used to easily discriminate between the two possible answers ( except for the cases in which the first and second stimuli are relatively close to each other ) . Note that this result has been obtained without any activity reset to new initial conditions during the inter-trial intervals . How do the results described so far change when , instead of storing just two correlated maps , the network encodes a sequence of maps gradually morphed between two uncorrelated ones ? Let us start by constructing two random uncorrelated maps , and . We would like to define the intermediate maps as gradual rotations between the two extreme ones; since we are dealing with circles , the rotation should be performed along the shortest arc between and ( see Eq . 21 , Methods - Inverse transformation ) . We assume here to have already transformed the variables in such a way that we can write directly ( 16 ) where indexes the maps along the morph sequence . Hence a neuron with label in the first map , will rotate along the sequence to its location on the last map , following the shortest path on the circle . With this choice of the morphing procedure , each neuron is still characterized by just two quantities , its labels in the extreme maps . We store the whole morph sequence by a superposition of the synaptic structures generated in each map separately , as for the case of two correlated maps previously described . For the sake of analytical tractability , we study the resulting coupling in the limit ( 17 ) Introducing the definition of two uncorrelated maps ( Eq . ( 1 ) with ) into Eq . ( 16 ) , we can rewrite the angles in the intermediate maps as , We can now integrate Eq . ( 17 ) ( 18 ) Making use of the Euler formula for the functionit is possible to deriveThe first term of the infinite product in the Euler formula , or the first term in the limit sum , gives us . Comparing the coupling in Eq . ( 18 ) , and the one derived for two maps , Eq . ( 2 ) , we see that to first order , the synaptic coupling induced by the storage of the whole morph sequence , is equivalent to the storage of two correlated maps with . In Fig . 7 , we compare the network activity generated by the approximated coupling and the full result of Eq . 18 , when the external input is constant . The results are qualitatively similar but the full morph case reaches the cylinder regime for lower compared to the case . Note that the network storing the morph sequence shows the same dynamical pattern separation observed in the two maps case ( Fig . 8 ) , see next Section for a simulation example in a finite network with a finite number of encoded maps . The important difference , is that while the very correlations between maps forced the absence of neurons with certain labels , hence constraining the permissible region for a marginal solution in , here the neurons cover the entire ( ) space . The result is purely due to the process of storing multiple maps along the morph sequence . This morphing algorithm also yields a way of stimulating the network with positions in environments intermediate between and ( with or without the intermediate maps encoded in the network ) . It is sufficient to use as a place specific input what we had in Eq . 13This time , the suitable range for the variable indexing the morph sequence is the whole range , if using as an approximation for the morphed case , or the restricted if the network is storing just two correlated maps . In the reference frame defined by the original coordinates ( ) , a change in the stimulated environment corresponds to a rotation of the axis representing the maximal external input; between a vertical axis ( stimulus localized in environment , to an horizontal axis , stimulus localized in environment . ) In the experiment of [32] , the rat is trained until it develops two separate place coding for a single arena with different light configurations ( representing two distinct environments ) . The advantage of this setup is that it allows , for instance , to slowly morph the light configuration between the two environments familiar to the rat . The experimental results shows a sharp transition around the middle of the light morphing ( lasting ) between the place representation in light configuration and . A link to these experimental results is provided by the use of time-varying external environment , where represents the duration of the morphing and denote the upper and lower bounds of the range . An example usage of this protocol is shown in Fig . 8 for the approximated whole morph sequence storage , for two slightly correlated maps in the cylinder region of the parameter range and for the double ring regime . For each run we show the dynamics of the relevant order parameter for the regime under consideration , for the double ring case and for the cylinder solution . In addition , we numerically solve the dynamics for a moving stimulus in either environment or . We use this as a reference for computing , at each time step , the correlation coefficient between the network activity during the morphing protocol and the activity in the fixed environment . The transition is sharpest for the storage of two slightly correlated maps . Note that similar results would be obtained by testing the network separately in each environment of the sequence ( see e . g . [31] ) . The sharp transition is maintained when increasing the amplitude of the external tuned input , because a small tilt in the tuned input towards either map or is sufficient to generate the dynamical pattern separation described in the previous Section . The transition in the cylinder regime occurs few seconds later than the one occurring in the double ring regime , which in turn happens in the middle of the morphing ( ) . This delay is due to the time required for the bump to move from the region of to , or viceversa ( see also Fig . 5B ) . This result could be compared with the experimental results of [32] . The delay does not occur when testing the network in separate environments along the morph sequence . There are two additional observations to be made ( data not shown ) . The first one is related to the sharpness of the transition in the double ring regime; by further reducing the amplitude of the external input , the mean-field dynamics can produce a sharp transition between the environments representations , which is also delayed compared to the middle of the morphing period . The delay gets longer as the external input gets weaker , in extreme cases it happens just before the end of the morphing procedure . This sharp and delayed transition is not observed in microscopic simulations with up to neurons , since the weak input is not able to overcome the local inhomogeneities in which the bump is trapped ( see e . g . [18] ) . It is possible that in larger networks the transition can be observed . The fine-tuning of the external input strength required to have the transition around the middle of the sequence , makes the double ring regime a weaker candidate explanation for the experimental results of [32] compared to the cylinder regime . The second observation concerns the dependency of the transition parameters on the velocity of the moving external input . We have noticed that the transition becomes smoother and closer to the middle as the velocity of the simulated animal is reduced . The details of the transition in a realistic setting would depend on the velocity history of the animal . In order to verify that the results obtained in the previous Sections are not artifacts coming from our assumptions of having an infinite number of neurons ( and maps , referred to the morphing procedure ) we compare some of the MF predictions to simulations of networks with a finite number of neurons . Each neuron is assigned a random pair of labels ( , for the -th neuron ) , from which we create either two maps with distance , or a finite number of maps ( , for the -th map ) along the morph sequence between two uncorrelated references ( see Methods - Numerical Methods ) . In Fig . 9 we compare the order parameters from MF and estimated from simulations , at a fixed value of the distance between the maps and inhibition . Varying , the solution goes through the double ring , single ring and cylinder regime . The order parameter is particularly sensitive to the finite size of the network ( and the randomized maps , see [18] ) . Fig . 10 shows the time evolution of a network storing few maps from a morph sequence . This is the best example to show dynamical pattern separation at finite size , since it is less intuitive than the case of two correlated maps . From an arbitrary initial position , the bump of activity starts moving first towards negative ( increasing angles in map ) , then along increasing without changing its location in . Note that , despite the presence of neurons everywhere in the ( ) plane , the bump moves along an invisible barrier resulting from the storage of the morph sequence . We have also verified that all the qualitative behaviors , number and type of solutions , unreliable coding , dynamical pattern separation and memory effect , are maintained when moving from maps on rings , to segments ( either two correlated maps or morphed ) , as studied e . g . in [18] , [37] for the single map ( data not shown ) . Instead of having neurons arranged on a cylinder in the coordinates , as for the ring case ( see e . g . Fig . 2B , II ) , the geometry resulting from two correlated linear maps would be an infinite strip . A strong enough map-specific interaction would produce a bump localized in both maps . An external moving input in one of the maps would move the bump on the strip up to the boundary , and then the bump would crawl along such boundary . Depending on the direction of the moving input or the identity of the stimulated environment , the bump can settle either in “upper” of “lower” part of the strip as in the cylinder regime . We have studied a continuous attractor network model storing a pair of correlated maps . The storage of a morph sequence between two uncorrelated maps falls in this class of model , since it is approximately equivalent to the storage of two strongly correlated maps . The other relevant parameter for describing the possible network behaviors , beside the correlation between the maps , is the strength of map-specific interaction between neurons . The analysis of the solutions of the system with a weak tuned external input , reveals several interesting behaviors . When the correlation between the maps is weak , neurons have two different tuning curves corresponding to the stimulus presentation in different maps . The representation is reliable , in that the single neuron response is consistent between presentations . This is the operating regime which is usually considered useful in place coding applications . For higher correlations between the maps and weak map-specific interactions , each neuron possesses only one tuning curve , irrespectively of the stimulated map . In contrast to the previous regime , this one is rendered useless by the inability to represent fully the state of the external world , i . e . the identity of the environment in the context of place coding analogy . We find another , novel regime for strong interactions and for any amount of correlation between maps . The surprising aspect of this regime is that the state of the world does not uniquely determine the state of the network; there is an additional degree of freedom in the network representation . To a closer look , this additional freedom found in the novel regime is rich of consequences . When the external input location is randomly varied between presentations in one map , we can define the response of a neuron to a particular location as an average of the neuron activity over external input presentations in that location . In this context each neuron has different tuning curves relative to the different maps used in the stimulation , but the price to pay is unreliable coding; a neuron which should be active during a particular state of the world , could remain silent . When the location of the external input changes smoothly in time on one map , some neurons develop a selectivity to the direction of change . When the increase happens on the other map , another subset of neurons fires . The overlap between the two subsets may be arbitrarily small , depending on the parameters choice . Neurons active in both maps would have tuning curves around similar values of the external input location . We refer to this phenomenon as dynamical pattern separation . There is an ambiguity in the network representation , due to the fact that the subset of neurons activating with the increase of the external location in map , will also activate with a decrease of the location in map . There are three possible experimental contexts in which this ambiguity does not arise . A simple experimental context would arise if the input is tuned in only one of the two maps and the only parameter changing is the location of the external input . Given some state variable , like size and orientation of objects , or frequency of sound waves for instance , our model would produce respectively tuning for expansion/contraction , CW/CCW rotation and upward/downward frequency sweeps ( all experimentally observed , see e . g . [40] , [41] ) . Our model provides a unique way for producing selectivity for the direction of change of a state variable , given a selectivity for the variable itself . Both kind of responses give rise to another interesting phenomenon: The current representation of the state of the world is influenced by the preceding one , even with an intervening delay . It is possible to read out from the network the direction of change of the state variable . This property may be exploited when solving delayed discrimination tasks ( see [42] for data analysis and modeling in terms of remapping for a somatosensory discrimination task ) . A second experimental context is related to place coding; the two environments should be considered as two distinct circular arenas which can be traveled only in one direction . Experimental observations show that when an animal is exposed to two environments , the majority of place cells have a place field in only one of the two environments ( see e . g . [43] , [44] ) . A possible experiment to test the model would consist in training the animals in two well differentiated environments . After measuring the distance between preferred locations for neurons having tuning curves ( place fields ) in both environments , one could train the animals in intermediate environments , which would correspond to the storage of the morph sequence in the model . For the novel regime of the model , the disappearance of the place fields in one of the environment would be predicted for neurons with very different preferred locations , and the remaining fields will converge to a common representation . Alternatively the training could be performed by using the initial two environments , and then slowly changing them across several training days to increase their similarity . This would correspond to the storage of two correlated environments . A third experimental context is related to direction selectivity in place cells . Animals trained to shuttle back and forth in a one-dimensional track ( a segment or a circle ) , have place cells showing selectivity to the direction of motion . For instance a cell could be active in a certain region of the circular environment when the animal is moving clockwise , while being completely silent when the animal moves counterclockwise . The link with our model is provided by the simple observation that the same 1D track , but walked in opposite directions , correspond to two different environments . Dynamical pattern separation would produce directional selective neurons , while a neuron having place fields in both environments would have similar preferred locations . In [45] , place cells recorded from rats trained in a circular environment indeed showed bi-directional place fields in similar locations . There was however a systematic bias in the difference between the preferred locations in the CW and CCW directions of the majority of the bi-directional cells: place fields were displaced backward with respect to the direction of motion of the animal . We believe that this result , termed by the authors “prospective misalignment” , could be obtained in the context of our model in more than one way . One possibility is the introduction of an asymmetry in the synaptic connections ( following [46] ) , with the asymmetry determined by the emerging direction selectivity of the neurons . The spread of activity due to the asymmetry would activate neurons earlier compared to the symmetric case , reproducing the prospective misalignment . A similar result could be obtained with short-term synaptic plasticity , which is known to produce a moving bump of activity ( [47] ) . A third option could be the introduction of a systematic shift between the maps , possibly resulting from Hebbian learning of the configurations generated by the suggested asymmetry mechanisms . In the experiments of [32] , two environments correspond to two different light configurations in the same arena . A slow linear morph between light configurations results in a sharp transition from the population representation for one environment to the other . This is a promising experimental technique which is able to probe with unprecedented flexibility the dynamics of remapping between two environments or along a morph sequence [32] , and could serve as a fertile ground for our model's predictions , hence for testing the attractor hypothesis . We show that , in agreement with the experiment , the slow morph protocol produces sharp transitions due to dynamical pattern separation . This result is even more significant considering the acknowledged difficulties in reproducing sharp transitions between correlated maps in a “traditional” setting [36] . The model predicts a transition between representations slightly delayed compared to half of the morphing period; it remains to be seen whether this occurs also in the experiment . Our results can be related to experimental observations about changes in place representation between distinct environments . Two major classes of remapping have been observed when an animal is tested in two distinct environments: rate remapping , in which cells maintain the positions of their firing fields while differentially changing their amplitudes , and global remapping , where changes in firing location are observed in addition to firing rate modifications ( see e . g . [43] ) . Based on these properties , we could associate the double ring regime to the global remapping and the cylinder regime to the rate remapping . The model results can also be compared to experiments with sequences of continuously morphed environments . When animals explored intermediate environments , both sharp and smooth transitions in representations were observed in different experiments ( see [31] and [30] correspondingly ) . Our model exhibits both sharp transitions between the place representations corresponding to intermediate environments ( cylinder regime ) and smooth transitions ( double ring regime ) . The linkage of cylinder and double ring regimes to sharp and smooth transitions respectively , taken together with the above mentioned association between these two model regimes with global and rate remapping , would be against the hypothesis made in [30] that related global remapping and sharp transitions on one hand , and rate remapping with smooth transitions on the other . In the present form , our model cannot be made compatible with this hypothesis . Since both the recordings of [31] and [30] contained populations of neurons exhibiting different transition behaviors , we speculate that the introduction of an additional selectivity for the environments ( see below ) could help in resolving the contradiction . Rate remapping would then correspond to a mixed single ring-cylinder regime ( different subsets of the network would exhibit the different regimes ) , while global remapping would resemble a mix of the double ring and cylinder regimes . A future extension of the model would include neurons with some form of selectivity for the context; each neuron would then be characterized not only by its location on the two maps , but also by selectivity indexes measuring its “preference” for the maps ( e . g . [17] ) . This more realistic setting including selectivity would produce silent neurons and place fields with variable peak rates/widths even when storing a single map . A second issue to be addressed is how the network can learn the synaptic structure from its inputs . The long-term plasticity ( e . g . [33] , [34] ) , could bring the network through various operating regimes depending on the training protocol . This could impose additional constraints on the model and provide additional predictions . Finally , with the introduction of short-term plasticity [48]–[53] , the network could exhibit an even richer repertoire of dynamics . This extension of the model would be an important step towards the experimental results of [32] . In this study , it was observed that when there is a fast switch between the two light configurations , the population vector sometime oscillates between the place representation of the environments , before settling on the current one . Preliminary results coming from the introduction of short term facilitation and depression in a network exhibiting a double ring solution , show that is indeed possible to observe oscillations between place representations . A detailed analysis of this behavior will be matter for a future report . To solve numerically the MF dynamics described by Eq . 5 , we discretized on regular grid in The integrals in the rhs of the equations were estimated using a trapezoidal method . The system of ODEs were integrated with an adaptive 4-th order Runge Kutta scheme . The simulation of the microscopic networks , whose results are reported in Figs . ( 9 , 10 ) , were performed by solving numerically the system of ODEs ( 19 ) where indexes the neurons . The matrix is built by summing the single map encoding , whereTo obtain the labels characterizing each neuron , we first randomly generated a and used Eq . 1 for or Eq . 16 for . For the comparison of the simulation with the MF results in Fig . 9 , we estimated from the steady state activity ( compare with Eq . 22 ) ( 20 ) from which we constructed the estimates for the order parameters , using Eq . 24 . For the noisy simulations shown in Fig . 5B , we used a current-based version of the dynamics described by Eqs . 19:We then estimated the order parameters via Eqs . 20 , using the firing rates . The noise was introduced as an additional term in the currentwhere is a zero average , unit variance Gaussian -correlated noise . We used for the results in Fig . 5 . The numerical solution was obtained using the Euler-Maruyama integration scheme . The simulations performed in Fig . 7 , for a network storing the whole morph sequence , were carried out as follows . Substituting the synaptic coupling obtained in Eq . 18 with the one in Eq . 2 , it is possible to derive a dynamics for the “order function”following the same procedure of Methods - Reduced dynamics . An order parameter is defined exactly as in Eq . 22 . The steady state activity of such dynamicswas compared with Eq . 4 , for and ( in absence of a spatially tuned input is constant ) . The time constant was set to everywhere . The inverse transformation can be obtained from Eq . 1 , defining ( 21 ) The rotation in the first equation is just needed to select the shortest distance between two maps on a ring , and it is transparent for the connectivity given its periodicity . This rotation was implicitly assumed when defining the neurons locations along the morph sequence , Eq . 16 . A first reduction of the dynamics described by Eq . 2 is done using the first two Fourier components of the activity with respect to the two correlated maps and , rewritten in terms of center map and the distance using Eq . 1 . In line with [37] we define the following variables ( 22 ) The variable is just the average activity , while and measure the spatial modulation of the network activity , in the map and respectively . Intuitively their values tell us which angle of which map is instantaneously represented by the network . The dynamics of the network activity , and of the order parameters , becomes ( 23 ) withIt is convenient to introduce dimensionless combinations of the order parameters to better expose the structure of the solutions , and then derive the dynamics of these new order parameters . From the two complex variables and the real one , we construct five new variables ( 24 ) From Eqs . 23 , after some algebra , it is finally possible to obtain the dynamics of the new order parameters , Eq . 5 . In Methods - Phase diagram of the model , we mentioned that the equation from Eqs . 5 , i . e . is automatically satisfied once the solution for the other four order parameters has been found . This can be seen using the fact that , by definition , the imaginary part of the real numbers and is . Since at steady state , and ( Eqs . 22 ) , by computing we can prove the property . Another statement mentioned in Methods - Phase diagram of the model , is that fixed points solutions of Eqs . 5 with do not exist . Observing the shape of the network activity at steady state ( Eq . 3 ) ( setting the phase for convenience ) which we rewrite herewe would like to know , given the correlation between the stored maps and the bump size , how much we can move the bump along by increasing without having active neurons at . We first analyze the onset of the freedom of choice of , by requiring the bump to “fit” exactly the range; with a bigger bump , the only possible choice for would be , with a smaller bump it would be possible to move it along . Hence , posing , the activity at would beThe angle at which this activity is maximal isso the maximal activity at the boundaries isWe recognize the first term inside the transfer function to be positive , so the only way to obtain a vanishing activity is to have . From Fig . 4 it is possible to see that the double ring solutions have always size . In order to obtain the range of integration for used to compute the average tuning curve in the cylinder regime ( Results - Tuned external input ) , it is enough to consider the activity at its maximum in We want this bump in to at most touch the endpoints . Given that the half-width of the bump is , the allowed range for is . In order to study the stability of the homogeneous solution , corresponding to in Eq . 23 ( i . e . from 24 ) , we can either linearize Eq . 23 , or take a step back from the MF reduction which lead to Eq . 5 , so to avoid division by . We take the second approach and redefine one of the order parameter , . To study the stability of the solution , it is sufficient to look at the dynamics of and . Posing , it is easy to verify thatwhere the function is defined asThe matrix describing the linear dynamics for the vector of small perturbations around the solution readswhereTherefore , two conditions must be satisfied for the solution to be stable: ( amplitude instability ) and . Evaluating the integral in explicitly , we get the line of separation between the homogeneous solution and the localized bump ( Turing instability ) , expressed in Eq . 7 . For the single , double ring and cylinder solution we can linearize directly Eq . 5 , posing . The matrix associated with the dynamics of the vector , after using the fixed points equations ( Eq . 8 ) , is ( 25 ) We define ( 26 ) whereUsing the identity , to write the function ( Eq . 6 ) , we see thatThe fixed points equations can thus be rewritten in term of the quantities in Eq . 26 ( 27 ) Let us examine the single ring and cylinder solution , . Given the symmetry in the integrand , in this case . The stability matrix from Eq . 25 becomes then ( 28 ) It is immediately seen that the eigenvalue corresponding to a destabilization of changes sign when . Substituting for the expression in Eq . 27 , it is easy to verify that this reproduces the curve of separation between the single and double ring regime described by Eq . 12 . We analyze the remaining two eigenvalues by looking at the trace and the determinant of the sub-matrix in the subspace ( from 28 ) :Given that , we see immediately that the eigenvalues have the same sign for , and one of them changes sign when . Recall that is the onset of amplitude instability we introduced without proof in Eq . 9 . If we find that when the trace is negative , then we know that correspond to a destabilization of the solution . Using Eq . 27 , we see that imposing is equivalent to , and that the trace satisfiesThe numerator in the first term is non-negative ( ) . The denominator is simply , non-negative by definition . The denominator in the second term is , and we can write the numerator asFinally , we numerically verified that the region of existence of the double ring solution coincides with its stability region .	How is your position in an environment represented in the brain , and how does the representation distinguish between multiple environments ? One of the proposed answers relies on continuous attractor neural networks . Consider the web page of your campus map as a network of pixels . Every pixel is a neuron , and nearby pixels excite each other , while distant pairs are inhibited . As a result of their interactions , a bunch of close-by pixels will light up , indicating your current position as suggested by your web-cam ( the sensory input ) . When you travel to another campus , the common assumption holds that pixels are completely scrambled and the excitatory/inhibitory pattern of connections is summed to the existing one . Now these connections and the sensory input will activate the pixels corresponding to your location in the new campus . The active pixels will look like noise in the old map . But what if the campuses are similar , i . e . the pixels are not completely scrambled ? We show that the network has a novel way of distinguishing between the environments , by lighting up distinct subsets of pixels for each campus . This emergent selectivity for the environment could be a mechanism underlying hippocampal remapping and directional selectivity of place cells in 1D environments .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "neuroscience/theoretical", "neuroscience" ]	2010	Continuous Attractors with Morphed/Correlated Maps
The accuracy of protein structures , particularly their binding sites , is essential for the success of modeling protein complexes . Computationally inexpensive methodology is required for genome-wide modeling of such structures . For systematic evaluation of potential accuracy in high-throughput modeling of binding sites , a statistical analysis of target-template sequence alignments was performed for a representative set of protein complexes . For most of the complexes , alignments containing all residues of the interface were found . The full interface alignments were obtained even in the case of poor alignments where a relatively small part of the target sequence ( as low as 40% ) aligned to the template sequence , with a low overall alignment identity ( <30% ) . Although such poor overall alignments might be considered inadequate for modeling of whole proteins , the alignment of the interfaces was strong enough for docking . In the set of homology models built on these alignments , one third of those ranked 1 by a simple sequence identity criteria had RMSD<5 Å , the accuracy suitable for low-resolution template free docking . Such models corresponded to multi-domain target proteins , whereas for single-domain proteins the best models had 5 Å<RMSD<10 Å , the accuracy suitable for less sensitive structure-alignment methods . Overall , ∼50% of complexes with the interfaces modeled by high-throughput techniques had accuracy suitable for meaningful docking experiments . This percentage will grow with the increasing availability of co-crystallized protein-protein complexes . Protein interactions are a central component of life processes . The structural characterization of these interactions is essential for our ability to understand these processes and to utilize this knowledge in biology and medicine . Experimental approaches , primarily X-ray crystallography , are producing an increasing number of protein structures ( www . pdb . org ) , which to a certain extent are representative of a significant part of the “protein universe . ” However , the overall number of proteins by far exceeds the capabilities of the experimental structure-determination approaches [1] , [2] . The answer to this discrepancy is computational modeling of protein structures . The modeling not only can supply the vast majority of protein structures , but also , importantly , is indispensable for understanding the fundamental principles of protein structure and function . Computational structure prediction methodology historically started with ab initio approaches based on approximation of fundamental physical principles , and continues to develop in this direction for the goal of learning the principles of protein structure and function . However , for the purpose of predicting protein structures , it has largely evolved to comparative techniques based on experimentally determined structural templates ( to a significant extent due to the increasing availability of such templates ) . Such approaches are faster , more reliable , and provide accuracy increasingly comparable with experimental approaches [3] . A similar trend is underway in structural modeling of protein interactions - protein docking [4] , [5] . Because of the nature of the problem , the ab initio structure-based methods in docking ( prediction of the complex from known separate structures ) are relatively more reliable than those in individual protein modeling ( docking rigid-body approximation has only six degrees of freedom and has an established record of practical applications ) . However , the knowledge-based docking approaches , including the template based ones , are rapidly developing , following the increasing availability of the experimentally determined structures of protein-protein complexes , which generally are more difficult to determine than the structures of individual proteins [6]–[8] . It was established by studies based on different sets of proteins that proteins similar in sequence , fold and/or function share similar binding sites [9]–[12] . Quantitative guidelines for quality of homology modeling of protein complexes were provided by Aloy and others [13] where it was demonstrated that sequence identities >40% yield high similarity of protein-protein binding sites . The modeling techniques for proteins and protein complexes applicable to entire genomes have to be high-throughput by design . This reason , along with the still limited availability of templates , causes the modeling techniques to combine high-resolution approaches , when available and computationally feasible , with low-resolution capabilities , for broad coverage of the proteome/interactome . Such low-resolution approaches still are capable of predicting essential structural characteristics of proteins and protein interactions , including the binding sites [14]–[16] , macromolecular assemblies [17] and binding modes for protein-protein [18] , [19] and protein-ligand [20] complexes . For template based docking ( based on co-crystallized protein-protein templates ) , the degree of similarity to the templates is key to the accuracy of the docking . For ab initio , as well as some knowledge/template based docking techniques , the accuracy of the resulting structures is directly dependent on the accuracy of the individual participating proteins , which in its turn is based on the similarity to the templates of individual proteins . In both cases , the critical component affecting the docking outcome is the ability to model the structures of the binding sites . Although one can argue that the structure of the whole proteins is important in general , the binding sites are the parts that have a direct effect on the accuracy of the predicted complex . Earlier estimates showed that the binding site accuracy of ∼6 Å Cα RMSD is sufficient for low-resolution ab initio docking [19] ( <3 Å Cα RMSD for small ligand-receptor docking [20] ) , with even lower accuracy suitable for meaningful docking prediction by template based docking ( Sinha et al . in preparation ) . In the current study we present a systematic analysis of the sequence alignment and subsequent modeling accuracy of known protein-protein binding sites . The analysis is performed and validated on the Dockground comprehensive dataset of co-crystallized protein-protein complexes [21] . According to the purpose of this study ( the assessment of high-throughput modeling capabilities for genome-size systems ) the modeling was deliberately performed in a high-throughput fashion using standard alignment ( BLASTPGP [22] ) and comparative modeling ( NEST [23] ) programs , as opposed to more detailed and sophisticated ( but also more computationally expensive ) multi-template procedures . The results show that for a significant part of the proteins the binding sites can be modeled with accuracy that would ensure meaningful docking , even in cases of alignments considered poor for modeling of monomeric proteins . Thus , structural modeling of protein-protein interactions can often be performed by means simpler than those typically used for modeling of monomeric proteins , despite the fact that protein-protein interactions in general are on the next complexity level relative to individual proteins . However , further advancement of large scale , high-throughput docking requires progress in experimental determination of structural templates . To assess the potential quality of binding site modeling , the sequences of 658 two-chain complexes ( Table 1 ) were subjected to PSI-BLAST search for homologous sequences in the PDB data bank . The following alignments were excluded from the resulting pool: ( a ) statistically insignificant alignments with expectation value e>1 and ( b ) alignments with target/template difference <10 residues . The latter allowed us to avoid a bias in alignment statistics caused by overrepresentation of certain groups of the proteins and their mutants in PDB . The resulting 66 , 706 alignments were further analyzed in terms of the target sequence coverage q ( see Methods , Eq . 1 ) , and coverage of the target interface residues qint ( Eq . 2 ) , with an emphasis on alignments with qint = 100% ( hereafter referred to as full interface coverage , or FIC , alignments ) . A residue of the target complex was assigned to the interface if the distance between any atom of the residue and any atom of the other subunit in the complex was less than the sum of the van der Waals radii of the atoms plus the diameter of water molecule 2 . 8 Å . An alignment was considered FIC with a level of tolerance that allowed one target interface residue to be missing in the alignment . The analysis showed that 37 , 062 alignments , or 56 . 1% of the entire alignment pool , are FIC alignments . On the other hand , FIC alignments were observed for both monomers in alignments of 218 target complexes and for one of the monomers in additional 101 targets , which together constitute most ( 97% ) of the dataset . In the distribution of FIC alignments for different functional classes of proteins ( Table 2 ) , notably , but not surprisingly , antibody-antigen complexes representing a fraction ( 3 . 6% ) of the protein set , produce a significant part of all alignments ( 17 . 5% , or ∼970 alignments per target complex ) , with FIC alignments for both monomers in all 12 cases . Interestingly , in two other functional classes ( enzyme-inhibitor and cytokine receptor ) the FIC alignments were observed at least for one monomer in almost 100% of cases as well , with the only exception of 1e44 , for which PSI-BLAST did not find any homologous sequences in PDB . Out of 11 cases in the ‘other’ functional class , for which no FIC alignments were found , 8 cases had no statistically significant alignments . In 3 complexes ( 1o6s , 1tt5 , and 1zm2 ) the interface consisted of terminal residues only . Thus the interface coverage could have been significantly reduced by absence of these terminal residues in an alignment , which is often the case in local alignments . For further analysis we introduced parameter qmax , the maximal target sequence coverage in a subgroup of alignments and counted the number of alignments ( all or FIC only ) in subgroups corresponding to q≤qmax = 40 , 50 , 60 , 70 , 80 , 90 , and 100% ( the entire alignment pool ) . The results in Figure 1 show that even when the target sequence coverage does not exceed 40% , there is a significant number of FIC alignments ( 191 out of 9 , 358 alignments with qmax = 40% ) . Although these FIC alignments constitute ∼2% of alignments with qmax = 40% , they are still sufficient for statistical analysis . The absolute lengths of these alignments range from 32 to 220 residues ( for 86 and 631 residue proteins , respectively ) , covering from 8 to 40 interfacial residues . The quality of the alignments is rather poor ( the range of the expectation values is from 2×10−48 to 1 . 0 , the sequence identities vary from 6 . 5% to 39% , and the gaps constitute up to 32% of the alignments ) . Such short alignments are generally considered poor in homology modeling of monomeric proteins . However , they can arguably be used for accurate modeling of protein-protein interfaces if all residues of the target interface are present in the alignment . Such interface modeling would provide accuracy sufficient not only for a meaningful analysis of binding properties , but also for docking of 3D models of monomers . Such docking is important for large-scale modeling of protein-protein complexes because modeling based on homology to co-crystallized protein-protein complexes accounts for only 15–20% of all known interactions [24] , [25] . It is important to determine if FIC alignments have properties that distinguish them from the whole pool of alignments . The knowledge of such properties would help in “real” homology modeling where interface residues are not known in advance and only the information related to the alignment properties , such as alignment expectation value e , and/or alignment identity aiden and similarity asim ( Eq . 3 ) , is available . For this purpose we compared the distributions of e , aiden and asim for FIC alignments and for all alignments with maximum target sequence coverage qmax ( see Figure 2 ) . The results show that e-distributions ( data not shown ) do not differ significantly between the FIC alignments and all alignments , irrespective of qmax values with a weak tendency of the FIC alignments to have e values lower than those in the whole pool of alignments . This difference is small and can be hardly used in practical discrimination of the FIC alignments . The pattern of distributions of other alignment parameters is different ( Figure 2 ) . Whereas for the alignments with qmax = 100% there is no large difference between the FIC and all alignments ( Figure 2B , D ) , the FIC alignments with qmax = 40% show a distinguishable difference from all alignments ( Figure 2A , C ) . For example , the part of the FIC alignments with aiden between 15 and 20% ( 84 out of 191 ) is two times larger than for all alignments ( 2124 out of 9358; Figure 2A ) . This difference is even more pronounced for the asim distributions ( Figure 2C ) , where the part of alignments with asim between 15 and 20% is four times larger for the FIC alignments ( 33 out of 191 as opposed to 459 out of 9358 for all alignments ) . We can hypothesize that this is due to a larger evolutionary distance between the target and the template proteins in alignments containing only a small part of the target sequence . Binding sites tend to be more conserved than the rest of the surface in evolutionary related proteins [26] . Such proteins usually correspond to “good” alignments with high target sequence coverage and alignment identity . This assumption is indirectly supported by the distributions of all alignments shown in Figure 2B , D where the fraction of the FIC alignments is larger at higher values of alignment identities and similarities , whereas at lower aiden and asim the situation is opposite . Figure 3 shows the distributions , similar to those in Figure 2 , but only for the residues that belong to the target binding site ( these residues do not necessary form continuous stretches of the protein sequence ) . To avoid ambiguities in definition of interface identity and similarity ( Eq . 4 ) for the alignments with no or little interface coverage , only FIC alignments are considered . The distributions of interface identity iiden and similarity isim qualitatively are similar to distributions of aiden and asim . The main difference is the positions of distribution maxima , which are shifted towards smaller values , compared to corresponding maxima positions in the aiden and asim distributions . The largest difference is in the iiden distribution for the short alignments , with the maximum for iiden between 5 and 10% as opposed to 15 to 20% for the aiden distribution . The distributions for the interface residues are also slightly broader than corresponding distributions for the whole alignments . For example , the peak in aiden accounts for ∼20% of the alignments while corresponding peak in the iiden distribution amounts only to ∼15% of the alignments . This is consistent with the previous assumption that alignments with small target sequence coverage are observed for evolutionary distant proteins where interface conservation is not evident . It is important to note that there are significant parts of the alignments with no identity in binding site residues ( ∼6% for the whole pool of FIC alignments in Figure 3B , and ∼15% for the short FIC alignments in Figure 3A ) whereas there are no alignments with zero alignment identity overall ( Figures 2A , B ) . This result by itself is not surprising since alignments with no identical aligned residues have expectation value so high that they are considered statistically insignificant and are not included in the PSI-BLAST output . On the other hand , there are no alignments with zero similarity ( no similar residues at all ) for the short alignments ( Figure 3C ) and almost no such alignments ( <1% ) for the whole alignment pool ( Figure 3D ) . This suggests that even for proteins distant in evolution the interface conservation may play some role , although at more complex level than simple amino acid preservation . For practical modeling of protein complexes it is important to estimate if the interface residues are inside an alignment based on the alignment properties only . For this purpose we determined the number of FIC alignments having certain range of alignment identities/similarities ( with a window of 5% ) and the number of all alignments having the same range of identities/similarities values . The ratio of those two numbers gives a probability to find all interface residues inside an alignment ( or FIC alignment probability ) with given identity/similarity . The calculations performed for the alignments with qmax ranging from 40% to 100% did not find significant differences in the resulting trends . For better visualization ( lower statistical noise ) Figure 4 shows the FIC alignment probability as a function of alignment identity and similarity for the whole alignment pool ( qmax = 100% ) only . Because of representative nature of our dataset of complexes , we can argue that the observed trends in this dataset will hold in the general case . Thus , we can assume that for the alignments with identity >40% ( similarity >60% ) , the probability to find all interface residues in a given alignment is ≥80% . This observation relates to the above suggestion that in the alignments with higher identity/similarity , proteins are closely evolutionary related . It was demonstrated in previous studies of ion binding proteins [27] , mitochondrial carriers [28] , glycolitic enzymes [29] , cyclic dependent kinases [30] , and other protein families [26] , [31] that the binding sites in closely related proteins are more conserved than the rest of the surface . Thus , the alignment programs ( such as PSI-BLAST used in this study ) more reliably identify these highly conserved regions , increasing chances to have full binding sites inside an alignment irrespectively of the alignment length . One can argue that this is a nonessential observation since it is well established in homology modeling of individual proteins that model building from the alignment with identity >40% is a trivial task since the fraction of correctly aligned residues in such alignments is approaching 100% ( e . g . , see Fig . 1B in Ref . [32] ) . However , the importance of our finding is that it provides a simple recipe for evaluating suitability of a particular alignment for building partial homology model of a protein complex of interest with good accuracy in the interface region . As mentioned above , there is a significant amount of alignments with low target sequence coverage containing all residues belonging to the interface of the target complex . To assess if such short alignments are useful for structural modeling of protein complexes , we built the structural models and estimated their quality in terms of interface RMSD between the model and the native structures ( see Methods ) for all FIC alignments with a certain maximum target sequence coverage qmax . To avoid ambiguities caused by possible absence of parts or even all of the interface residues in partial models , the study is restricted to FIC alignments and RMSD of the binding sites atoms . Also we focused on the extreme case of qmax = 40% , although modeling was performed for the alignments with qmax = 50% and 60% as well , with results being qualitatively similar to those for the qmax = 40% . Among the alignments considered , there were no cases for direct homology modeling where sequences of monomers in the target complex are aligned with the sequences from a template complex . The identities of aligned sequence parts in the alignments used to build the models in all cases were well below 40% , which puts them in the “twilight” zone of homology modeling of protein complexes [13] . There were 191 FIC alignments with qmax = 40% for 26 target sequences , among which two were from antibody-antigen complexes , three from enzyme-inhibitor complexes , and the rest from the “other” functional group . This distribution shows no overrepresentation of functional groups compared to the entire dataset . Models were built for all 191 alignments . However , for further analysis we chose a single model per target sequence , based on the highest identity of aligned sequence parts ( top model ) . The results are presented in Table 3 . For seven target complexes ( ∼27% ) the top model had interface RMSD<5 Å , which is in line with the estimates of the binding site accuracy needed for meaningful docking predictions [19] . For five complexes , interface RMSD was between 5 Å and 10 Å , which according to the estimates of the docking funnel size [33] , can produce near-native matches . Thus we define them as acceptable accuracy models of the monomers ( not to be confused with the acceptable accuracy models of the complexes in the CAPRI evaluation http://www . ebi . ac . uk/msd-srv/capri ) . The FIC alignments were detected in 50% of the complexes with overall alignments considered unsuitable for homology modeling of monomeric proteins . Interestingly , the expectation value of the alignment does not appear to be an appropriate parameter to assess the quality of the resulting model , since in all cases the alignment for the best model did not have the lowest e-value among FIC alignments , although the lowest e-value observed for the top models alignments was 10−47 ( 1gxd , chain A ) . For 17 target sequences , the top model was found to be also the best model , i . e . model with the lowest interface RMSD . Among 9 cases with different top and best models , only in two cases interface RMSD values were significantly different ( the top and the best models in different quality categories; data shown in Table 3 in bold ) . The data in Table 3 indicate that all FIC alignments for the top models have low sequence and interface identity/similarity , which suggests that target and template proteins in those alignments are evolutionary remote ( see discussion in previous sections ) . Thus , it is interesting to analyze whether there is a preference of target and template proteins in alignments to be from the same organism or from different species . Our analysis suggests no such preference since for good and acceptable models there were 6 target-template pairs from the same organism and 9 pairs from different organisms ( corresponding numbers for the wrong models are 5 and 8 ) . This does not support a conclusion from an earlier study [34] that protein-protein interactions are more conserved within one species than across the species . However a statistical analysis on a much larger pool of data is needed to reach a more definite assessment ( work currently in progress ) . Figure 5 shows examples of the models , including those for which the target and the template sequences are from the same and from different organisms . One interesting similarity in both cases ( Figures 5A and 5B ) is that the target proteins have two clearly distinguishable domains and the model structure covers a significant portion of one of the domains , which exclusively participates in the interaction with the other monomer ( not shown for clarity ) . In fact , this feature is common to all good-accuracy models ( interface RMSD<5 Å ) . The data on the binding domain coverage is provided in Table 3 ( where applicable ) . It shows that there is no clear correlation between the binding domain coverage ( although it is higher than the entire sequence coverage ) and the model quality . Acceptable accuracy models are built for the single domain proteins as well . Figure 5C shows an example of such model . The implication for practical modeling is that if the target protein is predicted to have a domain structure , then it is likely that the accuracy of the homology models produced on the basis of the “bad” alignments will be sufficient to perform a meaningful template-free docking . On the other hand , for homology models of single-domain proteins , methods less sensitive to structural inaccuracies ( e . g . , structural alignment ) should be used . This assessment is supported by a comprehensive study of the template free docking ability to tolerate structural inaccuracies [19] , which showed that low-resolution structural features of protein–protein interactions can be determined for a significant percentage of complexes of highly inaccurate protein models ( typically up to 6 Å RMSD from the native structure of the monomer ) . The results were further supported by recent studies of antibody-antigen docking of homology models , which concluded that the homology models yield medium-to-high quality of docking predictions [35] . Further confirmation came in the recent study by Aloy et al . [36] on the structural modeling of yeast interactome where it was found that the use of homology models in docking does not lead to a critical loss of accuracy ( assessed by extrapolation of docking results for the unbound X-ray structures ) . Our preliminary results on the benchmarking of the template free docking of the modeled structures was performed using GRAMM procedure , according to the goal of this study in the high-throughput fashion that does not involve computationally expensive scoring and structural refinement . The low-resolution criterion for success was: a match with the ligand interface RMSD<8 Å in the top 100 predictions . This RMSD value corresponds to the characteristic size of the binding funnel [33] . Such low-resolution predictions from the coarse-grained global scan are located within the binding funnel and can be further locally refined within the funnel . Higher-resolution docking , and the corresponding more strict success criteria ( such as those used in CAPRI ) , in addition to longer computational times , require higher , non-high-throughput accuracy of the binding site modeling , which is outside the scope of this study . The current study is aimed at the models of poor quality that still preserve the acceptable accuracy of the binding site . According to the above criterion , the success rate for the modeled proteins dropped to 23% from the similarly obtained 43% for the unbound X-ray proteins . However , such success rate is significant for the genome-wide studies . A systematic assessment of docking application to modeled structures of different accuracy is currently in progress . Table 3 also includes data on the failed modeling ( interface RMSD>10 Å ) . Figure 6D shows an example of such model . The target native structure has the domain structure similar to the successful models described above . The main reason for the incorrect modeling of the interface region is presence of a long stretch of gaps on the template side in the alignment . This is the reason for the incorrect loop ( indicated by arrow in Figure 5D ) , modeled without a template in the vicinity of the interface , which resulted in position shift of the interface residues in the model compared to the native structure ( yellow and blue meshes in Figure 5D ) . Another typical reason for large interface RMSD is the native structure interface having no secondary structure elements ( e . g . , a loop in enzyme-inhibitor complexes ) , but the fragment is modeled on a template with distinct secondary structure elements . A large difference between quaternary structures of the native target and the template structures also may lead to large shift of interface residues in the model , even if these residues belong to the same secondary structure elements as in the native structure . Analysis of organism and functional annotations ( Table 3 ) revealed that both target and template proteins are from the species spanning the entire universe of life - viruses , archaea , bacteria , lower ( fungi ) and higher ( plants and mammals ) eukaryotes - and participate in a broad range of biochemical processes . Moreover , there is no clear correlation between source organisms of the target-template pair or the biochemical pathways in which they participate . There are correct models with the target and the template from evolutionary distant organisms ( e . g . , mammals and archaea ) , as well as incorrect models with the target and the template from evolutionary close organisms or even the same organism . Similarly , no such correlation was found for the functions of the target and the template proteins , although the functional assignment has limited reliability . This suggests that the current ability to model complexes may not be restricted to certain species and/or functions . However , statistical analysis of a much larger protein interactions dataset , when it becomes available , would be necessary to draw more definite conclusions . For systematic evaluation of potential accuracy in high-throughput modeling of binding sites , local sequence alignments were performed in a representative set of protein-protein complexes . The results indicate that for the majority ( 97% ) of the target sequences there is at least one alignment containing all residues belonging to the interface of the target complex ( FIC alignments ) . Significant number of the FIC alignments was observed even when only ∼40% of the target sequence is aligned against the template . The results suggest a simple graphical function for evaluating the probability of finding all interface residues inside a local alignment when only the alignment information is known . Homology models of the interfaces in target monomers were built based on the FIC alignments with query target sequence coverage <40% . A simple scheme of model ranking based on the alignment identity showed that in ∼50% of cases the structural models have accuracy high enough for protein docking . Alignments that contain only a small portion of the target sequence and have low sequence identity are usually considered poor in modeling of individual proteins . They are used primarily in elaborate and computationally expensive techniques hardly applicable on genome-wide scale . Our results suggest that for the genome-wide structural modeling of protein interactions , simpler and less computationally expensive techniques based on the use of single , local sequence alignment , may yield satisfactory results , given that the interface residues are reliably identified in the alignment . Current methods for predicting protein-protein binding sites based on sequence information alone have limited accuracy ( e . g . Refs . [37] , [38] ) . However , because of the on-going significant community efforts in this direction , one may expect emergence of more accurate methods in the near future . A straightforward template-based modeling of protein complexes is possible on the basis of a co-crystallized template complex . However , previous studies [24] , [25] demonstrated that this technique could account only for ∼15–20% of all known interactions , whereas the rest of the protein complexes have to be modeled by other techniques . One possible direction is independent modeling of individual monomers on different templates with further application of docking ( either template free or based on structure alignment ) to these models . Earlier studies ( e . g . Refs [19] , [35] , [39] and others ) , as well as the results of this work suggested feasibility of this scenario . However more systematic and comprehensive studies are needed for quantitative guidelines of applicability of the homology models in large-scale structural modeling of protein-protein interactions ( study currently in the progress ) . Hetero-complexes with known 3D structures available in PDB were used in the study . To avoid bias caused by overrepresentation of certain protein families in PDB , we used the representative set of protein complexes from the Dockground resource [21] , manually selected and purged at 30% sequence identity level . Out of 523 complexes in the dataset , we further excluded structures with multi-chain interactions and those with large structural defects in the vicinity of the interface , which allowed us to avoid ambiguities in determining binding site residues . The final set consisted of 329 two-chain non-obligate complexes shown in Table 1 ( 63 enzyme-inhibitor , 12 antibody-antigen , 25 cytokine receptors , and 229 other complexes ) . This set is based on all protein structures available in PDB; thus the results are not dataset-dependent . For 658 sequences in the dataset , the search for sequence homologues was performed by PSI-BLAST [22] implemented in the program BLASTPGP . To broaden the pool of potential templates , the maximum number of hits was set to 2000 , with all other parameters set to default values . To obtain the checkpoint file ( the position specific scoring matrix PSSM ) [22] , the search was performed against all sequences in the non-redundant database of sequences ( www . ncbi . nlm . nih . gov ) with the substitution matrix BLOSUM62 [40] with five iterations . The checkpoint file was used in sequential PSI-BLAST run against all non-redundant sequences in PDB . The 3D models from the PSI-BLAST sequence alignments were built by program NEST from the JACKAL package developed in Honig's lab [23] using default parameters . Large errors in some template files were repaired by the program PROFIX from the same package . The NEST program was chosen over other popular modeling programs because it yields reliable models fast enough to be used in large-scale calculations ( e . g . , according to benchmarking of various homology modeling programs [41] ) and can be easily incorporated into automatic scripts for generating and updating databases of structural models currently under development in the lab . Since sequence alignments produced by PSI-BLAST are local by design [22] , not all residues of the target sequence are present in the alignment . Thus for the analysis of the alignments we defined the target sequence coverage ( 1 ) and , similarly , the interface coverage ( 2 ) Where and are the numbers of all target residues and the target interface residues , respectively , in the alignment; and are the total numbers of all residues and the interface residues , correspondingly , in the entire target sequence . We did not analyze whether the template is multi- or monomeric ( although the data is available in Table 3 ) since our goal was to determine the general usefulness of short sequence alignments in binding site modeling , rather than traditional homology modeling of protein complexes where both target and template are multimers . When the target had the multi-domain structure , we also calculated the domain coverage qdom using formula ( 1 ) , where Nali is the total number of the target residues inside the binding domain . The alignments were further analyzed with respect to the alignment e-value as well as their identity and similarity , defined as ( 3 ) where Lali is the length of the alignment ( number of target residues in an alignment plus gaps in the aligned target sequence ) , Niden is the number of aligned identical residue pairs , and Npos is the number of aligned residues pairs for which substitution matrix displays a positive number ( evolutionary favorable substitutions ) . Similarly , the identity and similarity of the interface residues inside an alignment was defined as ( 4 ) Where ( ) are the number of aligned identical ( positive ) residue pairs where the residue on the target side belongs to the target complex interface , and is the total number of the interface target residues in the alignment . To evaluate the quality of the resulting homology model , we calculated the root-mean square distance between Cα atoms of the interface residues ( interface RMSD ) , with the native structure of the monomer and its model superimposed by the program TM-align [42] . This measure is different from the RMSD used in the CAPRI evaluation [5] , where it is calculated between the interface atoms of the ligand in the native and in the docked matches , after structural superimposition of the receptors . Other widely used modeling quality criteria , such as sensitivity and specificity , are not applicable to our study because they involve true and false-positive/negative predictions that can be defined either for binary predictions of the fact of protein interactions ( which is not the case in our study ) or in the case of full modeled complex structure with both monomers present .	Protein-protein interactions play a central role in life processes at the molecular level . The structural information on these interactions is essential for our understanding of these processes and our ability to design drugs to cure diseases . Limitations of experimental techniques to determine the structure of protein-protein complexes leave the vast majority of these complexes to be determined by computational modeling . The modeling is also important for revealing the mechanisms of the complex formation . The 3D modeling of protein complexes ( protein docking ) relies on the structure of the individual proteins for the prediction of their assembly . Thus the structural accuracy of the individual proteins , which often are models themselves , is critical for the docking . For the docking purposes , the accuracy of the binding sites is obviously essential , whereas the accuracy of the non-binding regions is less critical . In our study , we systematically analyze the accuracy of the binding sites in protein models produced by high-throughput techniques suitable for large-scale ( e . g . , genome-wide ) studies . The results indicate that this accuracy is adequate for the low- to medium-resolution docking of a significant part of known protein-protein complexes .	[ "Abstract", "Introduction", "Results/Discussion", "Methods" ]	[ "computational", "biology/macromolecular", "structure", "analysis", "biophysics/macromolecular", "assemblies", "and", "machines", "biophysics/structural", "genomics", "biophysics/biomacromolecule-ligand", "interactions", "computational", "biology/genomics" ]	2010	Accuracy of Protein-Protein Binding Sites in High-Throughput Template-Based Modeling
Perceptual decision-making relies on the gradual accumulation of noisy sensory evidence . It is often assumed that such decisions are degraded by adding noise to a stimulus , or to the neural systems involved in the decision making process itself . But it has been suggested that adding an optimal amount of noise can , under appropriate conditions , enhance the quality of subthreshold signals in nonlinear systems , a phenomenon known as stochastic resonance . Here we asked whether perceptual decisions made by human observers obey these stochastic resonance principles , by adding noise directly to the visual cortex using transcranial random noise stimulation ( tRNS ) while participants judged the direction of coherent motion in random-dot kinematograms presented at the fovea . We found that adding tRNS bilaterally to visual cortex enhanced decision-making when stimuli were just below perceptual threshold , but not when they were well below or above threshold . We modelled the data under a drift diffusion framework , and showed that bilateral tRNS selectively increased the drift rate parameter , which indexes the rate of evidence accumulation . Our study is the first to provide causal evidence that perceptual decision-making is susceptible to a stochastic resonance effect induced by tRNS , and to show that this effect arises from selective enhancement of the rate of evidence accumulation for sub-threshold sensory events . Noise is an intrinsic property of all biological systems [1] . Typically , noise is viewed as being detrimental for neuronal computations and the behaviors they regulate [1 , 2] , including decision-making [3] . A key limiting factor in decision-making arises from noisy representations of sensory evidence in the brain [4 , 5] . On this view , noisy sensory information representations are not optimal , and this leads to errors in decisions . However , small amounts of noise added to a nonlinear system can increase stimulus quality by increasing the signal-to-noise ratio ( SNR ) [6] . This phenomenon is known as stochastic resonance ( Fig 1 ) , and its expression has been demonstrated in different sensory modalities [7–9] . Stochastic resonance occurs when an optimal amount of noise is added to a sub-threshold signal , which makes the signal cross a threshold and therefore enhances detection performance ( Fig 1 ) [10–13] . Neurophysiologically , adding an optimal amount of noise to a subthreshold signal pushes otherwise silent sensory neurons above the spiking threshold [7 , 15 , 16] . A common way of adding noise in a stochastic resonance context is to add it directly to the sensory stimulus . In such cases , however , the noise might simply increase peripheral receptor sensitivity [17] , which would not address the question of whether central neural processes in decision-making are sensitive to a stochastic resonance mechanism . Recently , we showed that it is possible to induce a stochastic resonance effect in a simple detection task when noise is added to the visual cortex with transcranial random noise stimulation ( tRNS [18] ) . Although the underlying neural mechanisms are not completely understood [19] , single unit recordings in visual cortex have revealed an increase in the SNR of neuronal spiking when an optimal level of noise is applied to a visual stimulus [20] , consistent with a stochastic resonance mechanism . This is likely due to the recruitment of voltage-gated sodium channels by the noise [21–23] . Our previous study [18] , together with several related investigations [24 , 25] , point to stochastic resonance as an underlying mechanism by which non-invasive brain stimulation can enhance behavioural performance when it is applied concurrently during task performance . Here we go beyond these findings by asking whether higher-level perceptual decisions in a random-dot-motion ( RDM ) task ( see Fig 2 ) are susceptible to a stochastic resonance effect . The RDM task has been widely used in studies of perceptual decision-making , and has well characterized neural correlates [26 , 27] . A recent study showed that RDM judgements are affected when noise is added peripherally to a visual display [28] , but it remains unclear whether an analogous effect arises for noise administered centrally ( i . e . , to the cerebral cortex ) . In addition to measuring the influence of central noise on perceptual decisions , we also investigated which aspects of the decision process itself are sensitive to stochastic resonance using drift diffusion modelling ( DDM; see Fig 2 [29 , 30] ) . Such modelling approaches have been very successful in describing both human and animal behavior [29] . Under the DDM framework , performance improvements can occur via a change in the decision criterion ( i . e . , the bound separation ) , or through an increase in the rate of evidence accumulation ( i . e . , the drift rate; [29] ) . In the current study , an increase in bound separation would suggest that the stochastic resonance effect is driven by a change in the decision criterion , whereas an increase in drift rate would suggest an improvement in the quality of sensory evidence on which the decision is based . In addition , if the stochastic resonance model applies to perceptual decision-making , then the addition of relatively small amounts of noise should enhance discrimination performance for coherent motion trials in which the signal is just below threshold , but not for trials in which the signal is well below or above threshold [8] . Finally , previous brain imaging studies have shown that visual motion stimuli elicit bilateral activation of extrastriate visual cortex [31–33] . By contrast , application of non-invasive brain stimulation over left hemisphere visual areas has been shown to have larger effects on motion discrimination than equivalent stimulation over right hemisphere regions [34–36] . We therefore applied tRNS over visual cortex bilaterally and unilaterally ( left and right ) , across separate experiments , to determine whether any stochastic resonance effect can be induced by stimulating the two hemispheres in combination or alone . Our results show that adding an optimal amount of noise to the visual cortex bilaterally enhances perceptual decision making in the RDM task , consistent with the stochastic resonance hypothesis . Performance deteriorated with larger amounts of noise , and the effect was not evident during unilateral hemispheric stimulation . Modelling of observers’ responses under the drift diffusion framework revealed that the improvement in performance with optimal noise was associated with a reliable increase in the drift rate parameter , implying an increase in the rate of evidence accumulation . In Experiment 1 , we stimulated the visual cortex bilaterally with tRNS in 15 participants . The coherence levels of 3% and 6% were subthreshold for both the group as a whole , and for the individual observers ( average detection performance < 63% ) , i . e . , performance was below the detection threshold , which corresponded to 75% correct in our task . For each observer we determined an individual discrimination threshold in the noise-free trials , and showed that this was above 6% coherence in all individuals . As shown in the left panel in Fig 3 , for the 6% coherence condition , which was just below threshold in the no-tRNS condition , motion discrimination performance improved when tRNS was applied at a relatively low intensity , whereas performance remained unaffected for the other coherence levels and noise intensities . For the analysis , we calculated the group %correct-choice-index ( %CCI ) for each coherence level and each tRNS intensity by dividing the %correct motion-direction responses under tRNS by the %correct responses when no tRNS was applied ( baseline ) , as given in the following formula: ( %CCI ) =%Corr ( i ) /%Corr ( zeronoise ) where i denotes each of the 4 tested noise intensities . There was a significant interaction between coherence level and tRNS-intensity ( F ( 12 , 156 ) = 2 . 47 p < 0 . 01 , Cohen’s f = 0 . 43 ) . To isolate the source of this interaction , one-way ANOVAs were conducted for each coherence level separately . For the 6% coherence condition only ( red symbols in Fig 3 ) , performance was significantly affected by the different tRNS intensities ( F ( 3 , 39 ) = 3 . 56 p = 0 . 02 Cohen’s f = 0 . 52 ) . There were no other significant main effects or interactions for the coherence conditions of 3% , 12% , 25% or 50% . Post-hoc tests were conducted to compare performance in the 6% coherence condition at each noise level against the baseline . All p-values were corrected for multiple comparisons . These comparisons revealed that a tRNS intensity of . 25mA significantly enhanced motion discrimination performance relative to baseline ( t ( 13 ) = 3 . 39 pcorrected < 0 . 02 ) . A similar enhancement was evident for the 6% coherence level at an intensity of . 375mA , but this effect did not survive our stringent correction for multiple comparisons , ( t ( 13 ) = 2 . 53 , pcorrected > 0 . 1 ) . These results suggest that perceptual decision-making for sensory stimuli that are just below threshold can be improved by adding a small amount of neural noise over bilateral visual cortex , consistent with predictions arising from the stochastic resonance hypothesis [8] . We used the hierarchical drift diffusion model ( ( HDDM , [37] ) to determine which aspect of decision-making was affected by tRNS . As shown in the right panel of Fig 3 , the drift rate was markedly affected by tRNS for the 6% coherence condition , whereas it was relatively unaffected for the remaining coherence levels . We submitted the drift-rate parameter to a 5 x 4 repeated measures ANOVA . This analysis revealed a significant main effect of tRNS-intensity ( F ( 3 , 39 ) = 2 . 85 , p = 0 . 049 ) and of coherence level ( F ( 4 , 52 ) = 3 . 18 , p = 0 . 02 on drift rate , as well as a significant tRNS-intensity x coherence level interaction ( F ( 12 , 156 ) = 3 . 22 , p < . 01 , Cohen’s f = 0 . 47 ) . To isolate the source of the significant interaction , one-way ANOVAs were conducted for each coherence level separately . Consistent with the behavioral data , there was a significant effect of tRNS intensity on the drift rate in the 6% coherence condition ( F ( 3 , 39 ) = 5 . 63 , p < . 01 , Cohen’s f = . 58 ) , but no significant effects for the other coherence levels were observed ( 3% , 12% , 25% , 50% ) . Adding higher amounts of noise to the 6% coherence condition resulted in a decrease in both behavioural performance and the drift rate ( see Fig 3 ) . This inverted U-shaped relationship between performance and noise level is a key signature of the stochastic resonance effect [8 , 38] . Post-hoc tests were conducted to compare performance in the 6% coherence condition against the baseline for each noise level . For the tRNS intensity of . 25mA , the drift rate for the 6% coherence condition was significantly higher than baseline ( t ( 13 ) = 3 . 44 , pcorrected < 0 . 02 , corrected for multiple-comparisons ) . A similar benefit for the 6% coherence condition was apparent for the tRNS intensity of . 375mA , but this effect did not survive correction for multiple comparisons ( t ( 13 ) = 2 . 55 , p = 0 . 1 ) . Separate 5 x 4 repeated measures ANOVAs revealed no significant effects for the bound-separation parameter ( all p > 0 . 06 ) , and no significant effects for non-decision time ( all p > 0 . 13 ) . Previous studies of visual motion discrimination have shown reliable effects of offline transcranial electrical stimulation—as opposed to the online effects reported here—following unilateral stimulation of left or right visual cortex in isolation [34 , 39 , 40] . We therefore conducted two further experiments to determine whether the stochastic resonance effects we observed for bilateral tRNS in Experiment 1 also arise for unilateral visual stimulation . We also modelled the current spread for the electrode montage used in each experiment using the SimNibs toolbox [41] . The modelling results revealed that the bilateral electrode montage affected the visual cortex in both hemispheres , whereas the unilateral configurations affected one hemisphere ( left or right ) only ( see Fig 4 ) . Fig 5A and 5B show the behavioral results for Experiments 2 and 3 . Neither left nor right unilateral tRNS influenced visual discrimination performance or the drift rate derived from the HDDM . To characterize these data statistically , we employed the same analytic approach as in Experiment 1 ( bilateral stimulation ) , for both the behavioral data and the drift diffusion modelling . There was no significant interaction between stimulation intensity and coherence level for either left unilateral or right unilateral visual cortex stimulation ( p > . 05 for all key comparisons ) . Thus , there was no evidence for a stochastic resonance effect as observed during bilateral stimulation in Experiment 1 ( see also S1 Fig ) . We found that adding an optimal amount of noise bilaterally to the visual cortex can enhance perceptual decision-making in a motion discrimination task , particularly for stimuli that are just subthreshold ( 6% coherence ) , as predicted by the stochastic resonance hypothesis [8] . By contrast , there was no reliable effect of tRNS on stimuli that were above or well below threshold , again consistent with the stochastic resonance account . When modeled as a drift-diffusion process , the tRNS-induced performance improvement for 6% coherence displays coincided with an increase in the rate of evidence accumulation for these displays only , reflected as a change in the model’s drift-rate parameter . The same model revealed no change in either bound-separation or non-decision time , suggesting that an optimal level of neural noise exclusively improves perceptual decision-making by enhancing sensory information quality , consistent with a stochastic resonance account [7–9] . Our results cannot be attributed to a speed-accuracy trade-off in observers’ responses , as the DDM controls for any such effects [42] . All tRNS intensities and motion coherence levels were randomized over participants to account for any fatigue , aftereffects or learning effects across conditions . It has been demonstrated that continuous application of at least 5-minutes of tRNS over the motor cortex can increase motor cortex excitability [43] . The effects we report here are unlikely to be due to changes in general cortical excitability , however , as it has previously been demonstrated that cathodal tDCS influences neuronal processing in motion sensitive areas , irrespective of the coherence level of visual stimuli [44] . By contrast , here we found a specific effect of tRNS on perceptual judgements for subthreshold motion coherence levels only . There was no evidence for a stochastic resonance effect when noise was applied unilaterally to the visual cortex . The absence of an enhancement effect for unilateral tRNS was not due to differences in baseline performance between the groups: discrimination performance in the 6% coherence condition was similar across experiments ( Experiment 1–60%; Experiment 2–58%; Experiment 3–57% ) . Modelling of the electrical field for each electrode montage ( Fig 4 ) indicated a higher peak current when the tRNS was applied bilaterally than in the unilateral stimulation conditions . It is unlikely that this apparent difference in current densities prevented a stochastic resonance effect for the unilateral stimulation conditions , however , because the same absolute current densities during bilateral stimulation were also reached during unilateral stimulation , but simply at higher tRNS intensities ( see S1 Table ) . The visual stimuli employed in our motion discrimination task were always presented in the centre of the screen ( i . e . , at the fovea ) , and thus would have been processed initially by visual areas in both the left and right hemispheres [45] as early cortical areas receive visual information from the contralateral hemifield . It is also known that area V5/MT receives information from both visual hemifields [46 , 47] . It is likely , therefore , that in the motion discrimination task employed here , areas V1 and V5/MT in both hemispheres would need to be recruited for successful performance . Based on our findings , it seems reasonable to hypothesise that visual areas in both hemispheres must be stimulated concurrently with tRNS for the stochastic resonance effect to occur . A study by Boulinguez and colleagues suggests that most human observers have a right hemisphere dominance for processing of visual motion stimuli , and non-invasive brain stimulation can enhance these individual asymmetries [48] . We did not test our participants for the presence of such asymmetries for visual motion perception , but it is possible that the absence of a stochastic resonance effect with unilateral stimulation was due to a mixture of right- and left-hemisphere dominant individuals in our sample . Because of the relatively diffuse nature of transcranial electrical stimulation in general [49] , it is not possible to determine the specific anatomical regions that mediate the stochastic resonance effect we observed . The primary visual cortex ( V1 ) [50] and motion area V5/MT are both crucial for the processing of dynamically moving visual stimuli [51–53] . These two areas are highly interconnected , so our bilateral stimulation protocol might have impacted motion processing in area V5/MT , enhanced signal quality in area V1 , or both . Further work using more focal stimulation techniques ( e . g . , transcranial magnetic stimulation ) will be needed to pinpoint the visual areas responsible for the stochastic resonance effects we report here . Recently , animal work has shown that optogenetic-noise-photostimulation of the barrel cortex in mice enhances both evoked-field and spike-firing responses to mechanical stimulation of the whiskers [54 , 55] . Optogenetic-noise-photostimulation could be used in combination with a decision task in mice [56] to further investigate the mechanism underlying our observed behavioral effect in human perceptual-decision making . Our results are in line with recent work which employed a similar motion-discrimination task to show that decision-making is sensitive to the addition of external noise to visual motion stimuli [28] . Future studies could investigate whether there is an interaction between external noise added to a visual motion stimulus , as used in [28] , and central noise delivered via tRNS over the visual cortex . If external and central noise affect a common underlying mechanism , then their combination should yield an interacting influence on the SR effect . By contrast , if external and central noise have separate underlying causes , their influence on the SR effect should vary independently . Our findings suggest that a stochastic resonance effect can be induced in a decision-making task when noise is directly applied to the visual cortex with tRNS [24 , 25] . Moreover , ours is the first study to show that this stochastic resonance effect enhances the quality of information processing as indicated by an accelerated rate of evidence accumulation . Many daily activities depend on our ability to decide upon appropriate actions based on available sensory information , e . g . , judging the speed of oncoming traffic to decide whether it is safe to cross the road . Even subtle impairments of perceptual decision making are likely to have a negative impact on daily functioning . Our findings could be applied to enhance perceptual decision making in people with developmental [57] or acquired [58] neurological impairments , in the elderly [59] , or even potentially amongst those in specialised professional and sports settings . The study was approved by The University of Queensland Human Research Ethics Committee and the Kantonale Ethik Komission Zurich , Switzerland , and was conducted in accordance with the Declaration of Helsinki . To select an appropriate sample size for the study , we conducted a power analysis ( G*Power version 3 . 1 . 3 , [60] ) . This indicated that a sample of 10 participants would be sufficient to detect a significant effect on discrimination performance in a repeated-measures ANOVA with a power of 0 . 80 for an α level of 0 . 05 . This estimate was based on an effect size ( Cohen’s d: 0 . 77 ) derived from our previous work on the influence of tRNS on detecting weak visual signals [18] . We chose to err on the side of caution , and tested 15 participants in each of the three experiments ( bilateral , unilateral left and unilateral right stimulation ) . Thus , a total of 45 healthy adults participated in the study overall ( 28 males , aged: 18–27 years , mean age = 22 . 5 years ) . All participants had normal or corrected-to-normal vision , and met the inclusion criteria for tRNS as assessed by a checklist prior to the experiment [61] . Written informed consent was obtained for all participants . Each participant received four tRNS noise intensities twice ( . 25mA , . 375mA , . 5mA and . 75mA; all delivered at frequencies between 100 and 640 Hz ) . Noise intensity order was randomized across participants . The tRNS was applied with a 0mA offset , and was applied for 20 trials followed by 20 trials of no-stimulation . This order was counterbalanced across participants . The tRNS was delivered via a battery-driven electrical stimulator ( version DC-Stimulator PLUS , neuro-Conn ) . The maximum current density was 46 . 87 μA/cm2 , which is well within published safety limits [60] . Electroconductive gel was applied to the contact side of the electrode ( 4 x 4 cm ) to reduce skin impedance . In Experiment 1 , the visual cortex was stimulated bilaterally , with electrodes placed 3 . 5 cm above the inion and 6 . 5 cm left and right of the midline in the sagittal plane . These coordinates were selected based on previous brain imaging and stimulation studies that investigated the offline aftereffects of transcranial current stimulation ( tCS ) on a motion detection task [35 , 62–66] . In Experiment 2 , the stimulation electrode was placed over the left visual cortex ( positioned as described for Experiment 1 ) , and the reference electrode was placed over the vertex ( Cz in the 10–20 EEG-system ) . In Experiment 3 , the stimulation electrode was placed at the homologous location over the right visual cortex , and the reference electrode was placed at the vertex as in Experiment 2 . All experiments took place in a dark and quiet room . Visual stimuli were generated using Matlab 8 . 0 ( 2012b ) and the Psychophysics Toolbox [67–69] , and were presented using a Dell PC ( T3400 ) running Windows XP with a NVIDIA Quadro FX 1700 graphics card . Stimuli were presented on an Asus VG428QE color monitor with a resolution of 1920x1080 pixels , and a refresh rate of 60 Hz . The luminance of the monitor was gamma-corrected with a maximum intensity of 316 . 5 cd/m2 and minimum of 0 . 33 cd/m2 . Viewing distance was maintained at 62 cm using a chinrest , meaning the display subtended 46° x 27° ( 1 . 5’ per pixel ) . We employed a two-alternative , forced-choice random-dot motion discrimination task [51 , 70] in which participants judged the direction ( leftward or rightward ) of the coherently moving dots as quickly and as accurately as possible . Each block of 20 trials began with the presentation of a central fixation cross ( 2 s ) . On each trial , the fixation cross was presented for 200 ms . The motion stimulus then appeared , and consisted of 100 square dots ( 3 x 3’ ) within an aperture ( radius 4 . 1° ) at the centre of the screen . The dots were randomly positioned within the aperture on the first frame before rigidly translating at 1 . 5 deg/s . If a dot was going to move outside the aperture on the next frame , it was wrapped to the opposite side of the aperture . The dot-motion display remained visible until response , up to a maximum duration of 3 s . Participants indicated their choices by pressing the left or right ‘shift’ key on a standard keyboard with the left or right index finger , respectively . If the participants did not respond within 3 s , the motion stimulus was extinguished , and the trial was counted as incorrect and excluded from further analysis . Participants were provided with immediate auditory feedback . A low-pitched tone indicated a correct response , a high-pitched tone an incorrect response , and a prolonged low-pitched tone indicated a response that was too slow ( i . e . , > 3 s ) . A new trial commenced 2 s after the previous response . A method of constant stimuli was used to determine global motion sensitivity . A proportion of the dots moved coherently to the left or right , and the remaining dots moved in random directions . Thus , for example , a coherence level of 3% indicates a display in which 3% of the dots translated coherently ( left or right , depending on the trial ) , while the remaining 97% of dots moved in random directions . Five logarithmically spaced coherence levels ( 3% , 6% , 12% , 25% and 50% ) were chosen , consistent with previous work [71] . The dots had a limited lifetime of 5 frames . In keeping with a common convention [72] , half of the dots were black and half of the dots were white , all of which were presented on a mid-gray background . To measure the effects of tRNS on visual motion discrimination , participants performed 10 blocks of 200 trials each , with different tRNS intensities . The first and the last blocks contained no tRNS . The four tRNS levels ( . 25mA , . 375mA , . 5mA and . 75mA ) were applied twice each in blocks 2–9 , in random order . The first block served as practice , and the data obtained were not included in the analyses . Each block contained 200 motion discrimination trials , with an equal number of presentations of each motion coherence level , presented in a pseudo-randomized order ( the total length of each block was ~ 6 mins ) . To minimize any build-up of tRNS effects , stimulation was applied for 20 trials before being turned off for the next 20 trials within each block . Coherence levels for stimulator-on and stimulator-off trials were balanced for each observer , and were combined during data analysis . Including electrode setup and data collection , the entire experiment took around 90 minutes per participant . The same statistical procedures were used in all three experiments . In each experiment , one participant was excluded ( 3 in total ) because the individual did not reach 80% correct responses in the highest coherence condition . The α level was set to 0 . 05 for all tests , adjusting for multiple comparisons using the Bonferroni correction . We used the same procedure to quantify the stochastic resonance effect as in our previous paper [18] . By normalizing the data to the mean of the noise-free trials , which were interspersed with active tRNS trials throughout the experiment , we were able to rule out any contribution from practice , learning or fatigue . The normalized behavioral data were subjected to a repeated-measures ANOVA with the factors of coherence level ( 5 levels: 3% , 6% , 12% , 25% and 50% ) and tRNS-intensity ( 4 levels: . 25mA , . 375mA , . 5mA and . 75mA ) . Drift diffusion modeling ( DDM ) has been employed widely to disentangle the different component processes involved in simple decision-making tasks [29 , 73] . It captures three distinct stages of the decision process: ( i ) boundary separation , which indicates how much evidence must be accumulated before a response is made; ( ii ) information accumulation rate ( ‘drift rate’ ) , which is a measure of how rapidly evidence is accumulated and depends on the quality of evidence in the stimulus , such that easier decisions result in a higher drift rate; and ( iii ) non-decision time , which is the time required to encode the stimulus and execute an appropriate motor response [29] . We used the hierarchical drift diffusion model ( HDDM ) to fit the DDM parameters to the data [37] . The HDDM uses a Bayesian method for estimating the DDM parameters , which allows simultaneous estimation of group and subject parameters . A benefit of the HDDM is that it outperforms other approaches when a small number of trials is available [74] . We took a similar approach in our implementation of the HDDM as Herz and colleagues [75] . We fixed the starting parameter , z ( also known as the bias parameter ) , to 0 . 5 , which is chance level in a 2-AFC task . We modelled the data with the drift rate , bound separation and non-decision time as free parameters . We obtained parameter estimates for the conditions noise-on/noise-off , coherence level and tRNS intensity . We normalized the obtained parameters to the zero-noise ( no tRNS ) trials . This normalization procedure was the same as for the correct choice index ( CCI ) data . Markov-chain Monte Carlo sampling methods were used for accurate Bayesian approximation of the posterior distribution of parameters ( generating 20 , 000 samples , discarding 10 , 000 samples as burn-in , and keeping every fifth subsequent sample ) . We visually inspected all traces of model parameters , their autocorrelation and computed the R-hat ( Gelman-Rubin ) convergence statistics to ensure that the models had properly converged [37] . All R-hat values were below 1 . 1 , verifying that convergence had been achieved [76] . For each experiment , we plotted observed and predicted RTs for the 10 , 30 , 50 , 70 and 90 percentile of trials ( i . e . , for the fastest 10% of trials , fastest 30% of trials , etc . ) against the cumulative probability ( see S2 Fig ) . These results indicated that the HDDM provided a good prediction of the observed data . The parameter estimates for bound separation and non-decision time ( NDT ) are shown in S1 Fig . As a sanity check we also plotted the drift rate against motion coherence level ( S3 Fig ) . As expected , the drift rate increased with increasing motion coherence . This provides further confirmation that our model provided an appropriate fit to the data . We used the SimNibs toolbox to model current flow in the brain [41] . The modelling results revealed that the bilateral electrode setup affected the visual cortex in both hemispheres , whereas the unilateral stimulation affected one hemisphere ( left or right ) only ( see Fig 4 ) . The SimNibs modelling approach does not provide any frequency-specific information . To determine whether the chosen tRNS frequencies ( 100–640 Hz ) reached the cortex , we estimated the electrical field strength at frequencies between 100 and 500 Hz , in steps of 50 Hz , with Spheres 2 . 0 [77] . The estimated electrical field strengths can polarize somatic membranes ( polarization <0 . 3 mV per V/m electrical field [78] ) and modulate network activity at low stimulation intensities ( 0 . 2 V/m , [79 , 80] ) . The electrical field strengths obtained with this modelling approach are estimates of the amount of current that reached the cortex ( see S1 Table ) . A recent study suggested that these results might be overestimated due to possible inaccurate resistance estimates for different tissues [81] , but even very low electrical fields ( 0 . 2 V/m ) are able to influence network activity . Analysis of the baseline data in all three experiments revealed no significant interaction between coherence level and tRNS intensity ( repeated-measures ANOVA with a within-subjects factor of coherence level and between-subjects factor of experiment , F ( 2 , 39 ) = 1 . 15 , p > . 32 ) , suggesting that the stochastic resonance-effect observed in Experiment 1 was not driven by differences in baseline performance between the three experiments . Across all three experiments , there was a highly significant main effect of coherence level on performance , as expected . For completeness , we also report here a small number of significant main effects which seem to be unrelated to the central stochastic resonance hypothesis under examination in this study . First , there was a small but consistent main effect of tRNS intensity on accuracy during right visual cortex stimulation , F ( 3 , 39 ) = 3 . 13 p = . 036 , Cohen’s f = 0 . 49 . Post-hoc contrasts revealed that this effect was driven by overall poorer performance for the . 25mA tRNS intensity , regardless of motion coherence level , t ( 69 ) = -2 . 78 pcorrected < 0 . 03 . This decrease in performance was mirrored by a significant main effect of tRNS-intensity on drift rate ( see S1 Fig ) , ( F ( 3 , 39 ) = 4 . 54 p < . 01 , Cohen’s f = 0 . 59 , which was again specific to the . 25mA tRNS intensity , ( t ( 69 ) = 2 . 67 pcorrected = . 04 ) , regardless of motion coherence level . Second , there was a significant main effect of coherence level on bound-separation during stimulation of the right visual cortex , F ( 4 , 52 ) = 3 . 09 p = . 024 , Cohen’s f = 0 . 4 ( see S1 Fig ) . Post-hoc tests showed that the bounds were significantly closer together for the highest ( 50% ) coherence condition , t ( 55 ) = 3 . 16 pcorrected < . 04 , relative to baseline ) , but there were no significant effects on bound separation for the other coherence levels . Although these effects were statistically significant , there was no interaction between tRNS-intensity and stimulus coherence level , which is a hallmark of the stochastic resonance effect . Moreover , it is important to note that these unspecific effects only occurred for right visual cortex stimulation . In that experiment there was no evidence for a stochastic resonance effect .	Noise is usually thought of as being detrimental for perception and decision-making , but recent work has revealed that under certain circumstances simple detection performance can be enhanced when an optimal amount of noise is applied to the visual cortex non-invasively using tRNS . Here we asked whether this stochastic resonance effect also applies to a higher level perceptual decision-making task . We found that adding an optimal level of neural noise to the visual cortex bilaterally enhanced decision-making , specifically for below-threshold stimuli , consistent with a stochastic resonance effect . Computational modelling under a hierarchical drift diffusion framework revealed that the enhancement of observers’ decision making with optimal noise arose from an increase in the rate of perceptual evidence accumulation . The findings provide new evidence in support of stochastic resonance as a neural mechanism through which weak stimuli can influence perceptual decisions , and suggest a novel target for interventions in neurological patients .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "medicine", "and", "health", "sciences", "decision", "making", "brain", "social", "sciences", "cerebral", "hemispheres", "neuroscience", "surgical", "and", "invasive", "medical", "procedures", "left", "hemisphere", "cognitive", "psychology", "functional", "electrical", "stimulation", "cognition", "right", "hemisphere", "vision", "visual", "cortex", "psychology", "anatomy", "biology", "and", "life", "sciences", "sensory", "perception", "cognitive", "science" ]	2018	Stochastic resonance enhances the rate of evidence accumulation during combined brain stimulation and perceptual decision-making
Taenia solium , a zoonotic infection transmitted between humans and pigs , is considered an emerging infection in Sub-Saharan Africa , yet individual and community-level factors associated with the human infection with the larval stages ( cysticercosis ) are not well understood . This study aims to estimate the magnitude of association of individual-level and village-level factors with current human cysticercosis in 60 villages located in three Provinces of Burkina Faso . Baseline cross-sectional data collected between February 2011 and January 2012 from a large community randomized-control trial were used . A total of 3609 individuals provided serum samples to assess current infection with cysticercosis . The association between individual and village-level factors and the prevalence of current infection with cysticercosis was estimated using Bayesian hierarchical logistic models . Diffuse priors were used for all regression coefficients . The prevalence of current cysticercosis varied across provinces and villages ranging from 0% to 11 . 5% . The results obtained suggest that increased age , being male and consuming pork as well as a larger proportion of roaming pigs and percentage of sand in the soil measured at the village level were associated with higher prevalences of infection . Furthermore , consuming pork at another village market had the highest increased prevalence odds of current infection . Having access to a latrine , living in a household with higher wealth quintiles and a higher soil pH measured at the village level decreased the prevalence odds of cysticercosis . This is the first large-scale study to examine the association between variables measured at the individual- , household- , and village-level and the prevalence odds of cysticercosis in humans . Factors linked to people , pigs , and the environment were of importance , which further supports the need for a One Health approach to control cysticercosis infection . Taenia solium cysticercosis , a zoonotic infection transmitted between humans and pigs , is considered as an emerging infection in Sub-Saharan Africa . In the Sahel and West Africa region , the pig population has increased by 23% between 1985 and 2005 , the largest increase in all animal populations during that period [1] . The pig population more than doubled in Burkina Faso between 2000 and 2008 [2] . Nearly 98% of pigs in the country are raised in a traditional manner in small holder farming communities , and left free roaming to fetch their own food [2] . The Joint Monitoring Programme ( UNICEF/WHO ) estimated the percentage of improved sanitation in rural Burkina Faso to be 8% at the end of 2008 , far below the target of 54% set for 2015 by the Burkina Faso National Program for Drinking Water Supply and Sanitation [3] . The increase in primarily traditionally raised pig population and the lack of improvement in sanitation in rural areas are ideal conditions for the spread and maintenance of T . solium infection in humans and pigs . This is likely to have great consequences on public health since humans , when infected with the larval stages of the infection , may develop neurocysticercosis ( NCC ) . NCC is a preventable cause of multiple neurological manifestations including epilepsy , severe chronic headaches , and focal deficits , to name but a few [4–6] . Serological studies using tests to detect antigens or antibodies have demonstrated the presence of the infection in humans [7–9 , reviewed by 10] and pigs ( reviewed by [11] ) in several countries of West Africa , but with some large variation in estimates from country to country and from village to village within countries . Part of the variation could be explained by the use of antibody-detecting tests such as the EITB [12] , which measure past exposure to and current infection with living metacestodes , and antigen-detecting tests [13 , 14] , which measure current infection with living metacestodes . For example , in Burkina Faso , no human was found to have a strong positive AgELISA result in a village where very few pigs were present , while the prevalence of strong positives in humans was between 1 . 4% and 10 . 3% and in pigs between 32 . 5% and 39 . 6% in two villages where pigs were raised [15 , 16] . Previous studies conducted in West Africa have included a small number of participants or of villages , limiting the ability to detect associations , and especially weaker ones , between village-level as well as individual-level variables and infection . A better understanding of individual and community-level factors associated with infection is essential to developing effective control strategies to reduce the burden of this devastating disease . The aim of this study is to estimate the magnitude of association of individual- , household- and village-level factors with current human cysticercosis infection in 60 villages located in three Provinces of Burkina Faso . This study reports on the baseline cross-sectional component of a large community randomized-control trial aimed at estimating the effectiveness of an educational package on reducing the cumulative incidence of current infection of human and porcine cysticercosis . The baseline cross-sectional study took place between February 2011 and January 2012 . The provinces of Nayala ( Region of Boucle du Mouhon ) , Boulkiemdé and Sanguié ( Region of Centre-Ouest ) were selected for inclusion in the study . Boulkiemdé and Sanguié are among the provinces with the largest number of pigs in the country ( 191 , 438 and 145 , 923 heads respectively in 2010 [17] . Nayala has an estimated pig population of 41 , 521 and was selected because of local reports that humans tend to defecate in pigsties where most pigs are kept during the rainy season and because it was neighbouring the other two provinces . In each province , all departments where pigs were raised ( 30 of 31 departments ) were selected . Within each department , two villages meeting the eligibility criteria were selected at random . The exclusion criteria were: being located on a National or Provincial road; being the Chef-Lieu ( Capital ) of the Region or of the Province; being located within 20 km of Koudougou or Ouagadougou . For the purpose of the parent community-based trial , the inclusion criteria were: having a population of at least 1000 people at the 2006 census; being present on the map from the Institut Géographique du Burkina ( from the year 2000 ) ; being separated from another participating village by at least 5 kilometers . A third village meeting all the eligibility criteria was also selected at random in each department as a replacement if the village initially sampled were to be found not eligible during the field visit ( for example , too few households raising pigs , refusal from the village leaders which happened in one instance ) . Only one village met the eligibility criteria in the Department of Zamo ( Province of Sanguié ) . A village located in the Department of Pouni but right on the border with Zamo was selected as the second village for that Department . The location of all participating villages is illustrated on Fig 1 . A concession ( compound ) is defined here as a group of households living in a residential development , often fenced , where the authority of a concession chief is recognized . A household is defined as a socio-economic unit where members live together , share resources and satisfy the food and other essential needs of all members . In general , a household includes a man , his wife or wives , their children , and any other person who shares resources with other members of the household . In each village , the research team of five or six interviewers started by visiting the concession of the village Chief . The field team moved counter clockwise to enumerate each concession . In each concession , starting on the right from the entrance , the field team enumerated each household , as well as the number of members and if someone raised piglets less than 12 months old or reproductive sows in each household . A stratified random sampling approach was adopted to select concessions . The strata were the presence of reproductive sows , piglets , or no pigs . First , the list of all concessions where there was at least one household raising sows was made and corresponding concession numbers were placed in a bag . A village member was asked to sample 10 numbers from the bag . If 10 or less concessions were raising sows , all of those concessions were invited to participate . Next , the numbers of all concessions where at least one household raised piglets aged 12 months or younger were placed in a bag , including concessions where sows were also raised but that were not selected in the previous step . A village member was asked to sample 30 additional numbers from the bag . If 30 or less concessions raised piglets less than 12 months of age , all of those concessions were invited to participate . Lastly , the numbers of all concessions not yet sampled were placed in a bag . A village member was asked to sample an additional 40 numbers . This resulted in the sampling of a total of 80 concessions in each village , with at least 10 concessions with sows and at least 30 concessions with piglets aged less than 12 months , except in a few villages in Nayala where there were less than 40 concessions raising pigs in the whole village . Fig 2 summarizes the sampling strategy for the concessions . Only one Chief of a concession refused participation in a village of Boulkiemdé and was replaced by another concession in the village . Once the Chief of a sampled concession gave his/her consent to participate in the study , he was asked to enumerate all households in the concession with the names of their Heads . Numbers corresponding to the households were placed in a bag and the chief of the concession was asked to sample one number from the bag , or , if there was only one household in the concession , that household was selected . The Chief of the selected household was asked for his consent to participate in the study and to enumerate all members of his household . Only one Chief of household refused participation in Boulkiemdé and asked that his father’s household be sampled instead of his , as a mark of respect for his elder . All household Chiefs consented to participate . Each eligible household member was allocated a number . The inclusion criteria for participants were that they were at least 5 years old , had lived in the village for at least 12 months and were not planning to move in the next three years . The head of the household was asked to pick a number from the bag . The person with the corresponding number was asked for his/her consent to participate in the study which included a screening questionnaire on epilepsy , progressively worsening severe chronic headaches as well as a blood sample to test for the presence of antigens to the larval stages of cysticercosis . Participation in the study involved answering the screening questionnaire and having blood collected three times in a three-year period for the purpose of the parent community-based randomized controlled trial . The sampling process ( starting with the concession sampling ) was repeated until 60 people consented to being screened and having blood collected three times over three years duration of the randomized trial . If the sampled person refused to consent to the serological component of the study , s/he was asked for his/her consent to participate in the screening-only study for a maximum of 20 screening-only participants per village . If the sampled individual did not consent to either the serological or screening study , the head of the household was asked to sample another number from the bag for that household until a consenting individual was found . There were a few refusals to the serological participation , and all cases were replaced as described above . No one refused to participate to the neurological component of the study . Each consenting participant was asked to answer an individual questionnaire including questions about socio-demographic factors , pork consumption behavior , knowledge of T . solium infection and life cycle , and screening questions on seizures , epilepsy , and severe chronic headaches . Since the focus of the current study is to estimate the magnitude of association between potential risk factors and current cysticercus infection , results from screening for epilepsy and progressively worsening severe chronic headaches are not reported . In addition , the Chief of each participating household was asked to list assets owned by members of the household ( ie bicycles , carts , livestock , etc ) . The mother of each household was asked questions about preparation of pork and access and use of latrines by members of the household . The building material of the house’s roof , floor and walls was also measured at that time . In each village , the owners of at most 10 sows and 30 piglets aged less than 12 months were interviewed regarding their pig management practices ( at most 40 pig owners per village ) . All questionnaires were conducted and recorded on Personal Digital Assistants ( PDAs ) . The PDAs recorded the geographical coordinates of each concession . We used indicators of wealth as suggested in Gwatkin et al . ( 2000 ) [18] to estimate the wealth quintile of each household . Several wealth indicators were missing and therefore imputed using the missMDA package in R [19 , 20] . Principle Component Analysis was run in R using the PCA command of the FactoMineR package [21] . The imputed and known were exported back to Stata where PCA analysis was conducted and the fitted values were used to obtain the percentiles that were used to classify the households into 10 wealth groups . Between 18 March and 25 November 2014 , the field team returned to each village to take soil samples to measure soil composition and pH using the LaMotte Soil Texture Unit test ( code 1067 ) and LaMotte pH Test kit ( code 5024 ) , respectively . The soil composition test estimates the percentage of sand , silt and clay in the soil . In each village , a soil sample was taken in each of the four cardinal directions ( within 2 km ) using a plastic pot filled to 125 ml , having scraped the soil surface to a depth of about 5 to 10 cm . The four samples were then thoroughly mixed , dried and cleaned of coarse elements such as stones . A subsample of this mixture ( about 125 ml ) was then taken and stored for analysis of soil composition and pH , according to LaMotte procedure . After an average of 7 weeks ( range of 0 to 140 days ) , a physician went to each village to take blood samples from the 60 participants who consented to the serological component of the study and to examine all participants who had screened positive for seizures , epilepsy or severe chronic headache . Any participant confirmed as having epileptic seizures , epilepsy or severe chronic headache who had not initially consented to the serological follow-up were asked to provide a blood sample for ( clinical ) diagnosis purposes . However , to avoid an over-selection of people with neurological symptoms in our analysis , only those who initially consented to the serological analysis component of the study are included in the analysis . Blood samples of 8 ml were drawn by venipuncture using a syringe , preferably at the antebrachium vein , using 10 ml Venosafe serum gel tubes . The tubes were placed in a cooler , left to decant , and the sera were collected and put in two pre-labelled tubes at the end of each day or the following day . The sera were placed in freezers ( -20°C ) at most three days after the blood draw . The sera were brought to the IRSS ( Institut de Recherche en Sciences de la Santé ) in Bobo-Dioulasso every 4 to 8 weeks and kept thereon in a freezer at -20°C until analyses took place . The serum samples were tested for presence of excretory secretory circulating antigens of the metacestode of T . solium using the B158/B60 enzyme-linked immunosorbent assay ( ELISA ) [13] . The test was found to have a sensitivity of 90% ( 95%BCI: 80–99% ) and a specificity of 98% ( 95%BCI: 90–99% ) to detect current infection in a study conducted in Ecuador [22] . Descriptive analyses on the study population were first conducted , followed by assessing the association between each potential risk factor and the prevalence of current cysticercus infection at the individual-level and at the village-level . Household characteristics measured through the mother and chief questionnaires were attributed to each individual since only one individual was sampled per household ( and concession ) . Consequently household-level variables were included at the individual-level in all analyses . All descriptive analyses were conducted in Stata 13 . 1 . Data obtained from pig owners were considered as village-level variables . These included the percentage of pig owners letting their animals roam all seasons , practicing home slaughtering of pigs and asking for inspection at the time of slaughter . Soil composition , soil pH and the season when human samples were collected ( dry vs wet ) were also considered as village-level variables . The effect of the coverage of the 2012 filiariasis mass drug delivery campaign , which provided albendazole ( 400 mg ) with ivermectin , obtained from the Ministry of Health , was also explored at the village-level . The effects of variables which may impact the contamination of the environment with human feces were explored at the individual-level and at the village-level . These included the use of latrines to defecate reported by interviewees , the access to a latrine as reported by the mother , the household wealth quintile ( as an indicator of general hygiene and sanitation ) , the knowledge about taeniasis including report of self-infection , and the reporting of pork consumption at home and outside the home . Individual-level variables explored that were not likely to directly influence environmental contamination included age , gender , education and occupation , although the effect of age was modelled separately for each province . Age was categorized because it was not linear in the logits . The only concession-level variable explored was the type of concession sampled ( i . e . sow , piglet or other ) . Given the sampling strategy , concession , household and individual characteristics were all attributed to the individual-level in the analysis . To take the stratified nature of the sampling into account , models including the type of concession sampled were run but did not modify the estimated medians and 95%BCI of the estimates . Hence , results from the simpler model are being reported here . Bayesian hierarchical logistic models were fitted to estimate the prevalence odds ratios between each variable of interest and the prevalence of current cysticercus infection . Some models were also run with Bayesian hierarchical log-binomial models and resulted in similar estimates when convergence was achieved . At the first level , current cysticercus infection was assumed to follow a Bernoulli distribution . The logit of this distribution was modeled using the individual-level variables and village-level random-effect intercepts . At the second level , each village-level intercept was assumed to follow a normal distribution . The mean of the random-effect intercepts were modelled as a linear regression using the village-level variables , including the province in some models . The effect of the provinces with a random-effect on villages was not important , and therefore all presented models exclude province effects . Diffuse priors were used for all regression coefficients . Missing independent variable values were imputed using the mean value of non-missing data , assuming an ignorable missingness mechanism . When the mean values varied by province , province-level means were used for imputation . Fit was measured by comparing deviances . Convergence was assessed by looking at the history and b Rubin Gelman plots . Some of the more complex models required large numbers of iterations and thinning of 100 to obtain stable estimates . All models were run in WinBUGS [23] . The protocol and consent forms were approved by the University of Oklahoma Health Sciences Center Institutional Review Board and by the Centre MURAZ ethical review panel ( Burkina Faso ) . All participants were read the consent form and any questions they had were answered to the best knowledge of the field staff . Each participant was given a bar of soap to thank them for their time . Information about the study provided in the consent forms were read and explained to each potential participant ( mother of the household , chief of the household , participant , pig owners ) by the field workers who took the time to answer all questions . Consenting participants signed the consent forms when able or put a cross when not . All consents were witnessed by a local villager . Children aged more than 10 were asked for their assent , but parents consented for all children aged less than 18 years old . A total of 4795 villagers consented to being screened for epilepsy and severe chronic headaches three times during the course of the parent community-based randomized trial . All three mother , chief of the household and individual questionnaires were missing for three individuals who were excluded from the analyses . Of the remaining 4792 participants , 4788 , 4772 and 4778 had information from the chief questionnaire , mother questionnaire or individual questionnaire available , respectively . A total of 3609 participants consented to participate in the serological component of the parent randomized trial and provided sufficient blood at baseline to be analyzed . The characteristics of individuals participating in the serological and screening-only component of the study are described in Table 1 . Since individuals living in concessions where pigs were being raised were first asked to participate in the serological component of the study , there was a larger proportion of participants who provided blood who either raised pigs or consumed pork . The participation proportion was similar according to other characteristics although females and more educated people tended to be more likely to consent to the serological follow-up component of the study . A total of 120 individuals tested positive for current infection with cysticerci . The prevalence of current cysticercosis varied considerably across provinces and villages ( Figs 1 and 3 ) ranging from 0% to 11 . 5% , although the 95%CI were wide . In the Province of Sanguié , no individual tested positive in seven ( 35% ) of the 20 villages studied . In contrast , this was the case in only three ( 10% ) and one ( 10% ) villages in Boulkiemdé and Nayala , respectively . Table 2 provides estimates of the prevalence according to different individual-level characteristics of the participants as well as their associated prevalence proportion ratios . The univariate analyses suggested that older males , people living in a household with lower wealth quintiles and those consuming pork had higher prevalences of infection . In addition , access to a latrine as reported by the mother of the household was associated with a reduced prevalence of infection . Table 3 shows the linear regression coefficients between the prevalence of infection in each village and the village-level variables . The percentage of participants reporting eating pork , and particularly those eating pork in someone’s or their household , and the percentage of participants reporting having had a tapeworm infection were associated with a slight increase in the village-level prevalence of infection . The percentages of silt in the soil and of sand in the soil were associated with a decrease and increase in the prevalence , respectively . Table 4 shows the results of three candidate models with the lowest deviances . Models 2 and 3 include soil indicators which were measured up to 36 months after the baseline visit while model 1 does not include these variables . The individual-level variables were common to all models and all resulted in similar magnitudes of association with the prevalence odds of infection . Being aged more than 50 years old had a stronger effect on the prevalence odds in Boulkiemdé than in the other two provinces . Males had cysticercosis prevalence odds of nearly 2 . 6 when compared to females . Eating pork at another village market had the strongest effect , while eating pork at the village market or at home also increased the prevalence odds of infection when compared to never having eaten pork . Living in a household with higher wealth quintiles and having access to a latrine both decreased the prevalence odds of infection , with those not having access to a latrine having a prevalence odds of about 2 . 8 times higher than those having access to a latrine . The effect of a self-history of taeniosis and knowledge of taeniosis became negligible in models adjusted for age ( ie age was a strong confounder of taeniosis history and knowledge ) . The major differences in the three models come from the inclusion of the type of soil and soil pH . When these variables are excluded , the percentage of pigs not penned during the rainy season ( i . e . left roaming or tethered ) led to a weaker association with the prevalence odds of cysticercosis . An increase in the alkalinity of the soil was associated with a lower prevalence odds of cysticercosis . This effect was only noted when the percentage of silt or sand in the soil was also included in the model . We present here the effect of the percentage of sand , but the percentage of silt had an opposite effect of similar magnitude to that of sand . This cross-sectional study is the most widespread ever conducted , estimating factors associated with the prevalence of current infection of human cysticercosis . Our study is unique in its inclusion of 60 villages located in three provinces and the evaluation of over 3600 people living in these villages . The inclusion of only one individual per household and concession reduced the dependence among observations , thus maximizing the power to detect individual and household-level factors associated with the prevalence . Our hierarchical model also includes potential risk factors measured at the individual- and village-level , thus respecting the sample size of each unit . Finally , by including the village as random-effects , and having explored the impact of incorporating the type of concession which was part of the sampling scheme , we are effectively using a model-dependent approach to adjusting for the sampling scheme , thus reducing the potential biases which may be introduced by clustered sampling [24] . To our knowledge , only two studies conducted in Sub-Saharan Africa had adjusted for the cluster nature of the sampling or the infection [7 , 25] . We found that the prevalence of current infection with cysticercosis varied from 0% to 11 . 5% in the sampled 60 villages . These villages were sampled with the goal of conducting a community-based randomized controlled trial and participants were selected based on the presence of pig raising in their household . Therefore , the overall prevalence cannot be generalized to the three provinces nor to the country as a whole . Nonetheless , the village-level prevalences of current infection with cysticerci measured with the AgELISA are within the range of those reported by others using the same test in community-based studies conducted in three rural communities of West Cameroon ( from 0 . 4% ( 0 . 2%;0 . 7% ) to 3% ( 0 . 3%;11 . 2% ) depending on the locality ) [7] , one small village in Sénégal ( with 7 . 7% ( 5 . 3%-10 . 7% ) ) [8] , and 20 villages in Zambia ( with 5 . 8% ( 4 . 1%-7 . 5% ) [26] . Other community-based studies have reported higher overall prevalences of current cysticercus infection ( ie not village-specific ) in one village in the Democratic Republic of Congo ( with 21 . 6% ( 18 . 2%-25 . 0% ) [25] and 13 villages in Tanzania where very high porcine cysticercosis prevalence levels had been reported ( with 16 . 7% ( 14 . 2%;19 . 2% ) ) [27] . A unique characteristic of our study is its ability to explore between-villages variation in prevalence . Although the villages were sampled with a set of inclusion criteria necessary for the randomized trial , considerable variation in the prevalence was observed among them as well as among the three sampled provinces , although the credible intervals were wide . Such variation in areas was also observed in a study conducted in three areas of West Cameroon [7] , different districts of a village in the Democratic Republic of Congo [25] , and six departments of Bénin [9] , although the latter study used an EITB to detect exposure to infection [12] instead of current infection . This confirms observations by others that current cysticercus infections in humans occur in clusters [26 , 28–30] , often around taeniosis carriers . The very clustered nature of cysticercosis calls for great care in attempting to generalize results from studies conducted in a small number of villages or communities to larger areas or to a country . The individual-level factors found to be associated with the prevalence odds of current infection are similar to those reported in other community-based studies conducted in Sub-Saharan Africa . An increased prevalence odds of current cysticercus infection in adults aged 30 or more as compared to individuals aged 7–30 years old was observed in all three provinces . This confirms what was observed in a study of 720 participants living in 20 villages of Zambia where the prevalence odds increased after the age of 30 years old when compared to those aged 0–9 years old [26] and a study in Tanzania where the prevalence odds was increased in individuals aged 36 or more when compared to those aged 15–25 years old [27] . Moreover , in the province of Boulkiemdé , a further increase in the prevalence odds was observed in people aged 50 years old or more . Such increase in prevalence in older people has been observed in studies conducted in the Democratic Republic of Congo ( POR of 2 . 8 95%CI: 1 . 14; 3 . 81 for those aged 70 or more compared to those aged 0–9 years old ) [25] and in West Cameroon ( seroprevalence of 2% in those aged 46 years old or more and 0 . 1% in those aged 15 or less ) [7] . A study conducted in 1989 in Bénin found an age-pattern of sero-prevalence of exposure to cysticercosis which is very similar to that observed in Boulkiemdé , with an initial increase at 30 years of age , followed by a further increase after 50 years old [9] . The Zambian study did seem to show a tendency for higher prevalence of current cysticercosis after the age of 50 , but the small sample size in older age groups may have reduced the power to detect such increase [26] . A study conducted in Ecuador suggested that the increase in current infection in older age could be linked to reduced immunity in older age groups [31] . This is further supported by findings from a cohort study conducted in Zambia which showed that while sero-reversion rates were higher than seroconversion rates in people aged less than 60 , such difference disappeared in people aged 60 years old or more [32] . The reason why an additional increase in older ages was only observed in one province is difficult to explain and may require future studies , but may be linked to differential at risk behavior in this group of older men not captured elsewhere . The increase of prevalence odds of current infection with or exposure to cysticercosis in males has been reported by some community-based studies conducted in Sub-Saharan Africa [9 , 15 , 25 , 33] , but not all [7 , 8 , 26] . The study conducted in the Mbozi district of Tanzania reported an impact of gender on exposure to cysticercosis measured with the EITB , but the reporting is problematic because the gender among whom the prevalence odds is reported to be increased is inconsistent ( males in the results section and females in the discussion section and Table 3 ) [27] . In that same study , the effect of washing hands by dipping in multiusers buckets was associated with a decrease prevalence odds of exposure to infection and an increased prevalence odds of current infection , casting doubts about the reported results . The difference in results between studies could be real , or could be attributable to the reporting of crude associations in some studies [9] and that resulting from multivariable analyses in others [7 , 8 , 15 , 25 , 26 , 33] . In our study , the association between being male and the prevalence odds persisted after adjusting for pork consumption , wealth quintile of the household , and access to a latrine . This confirms that factors which bring males to get exposed to T . solium eggs more than females , such as poor hand hygiene or consumption of fresh produce such as fruits and vegetable that have not been cleaned properly , or eating meals outside the home that could be prepared by foodhandlers infected with taeniasis may be playing a role . Behavioral studies comparing male and female food consumption and hand hygiene behavior are warranted to develop future interventions . Where pork is consumed was found to have an impact on current infection with cysticerci . Since cysticercosis can be acquired by either self-infection or through food or hands contaminated with the eggs of T . solium , and since the effect of a self-reported history of taeniosis became negligible once gender and age were included in the model , contamination of food with T . solium eggs may play an important role in this population . To our knowledge , this is the first time that the effect of where the pork is consumed is reported . In Sub-Saharan Africa , only one study in Zambia found that not boiling pork before its consumption played a role among older females based on a classification tree model [33] . Other studies conducted in Latin America , although in univariate analyses , had reported an association between pork consumption and exposure to cysticercosis [34 , 35] . Having access to a latrine was associated with reduced prevalence odds of current infection with cysticerci . This had been mentioned in the small scale study of Secka et al . conducted in Sénégal [8] , but was not confirmed in a multivariable analysis . It was also observed in relation to the prevalence of exposure to infection in univariate analyses in Colombia [35 , 36] and Honduras [37] . The effect of having access to a latrine by family members had a stronger effect than that of participants declaring that they used a latrine to defecate . This is in agreement with a recent qualitative research conducted in Eastern Zambia which showed that latrines are usually considered as public among neighbors [38] . Our results support that access to latrines is an indicator of environmental contamination with taeniid eggs around the household . Poor living conditions , as indicated by the three lowest wealth quintiles in our study , have also been reported to be associated with increased prevalence odds in previous studies , in these cases using univariate analyses . In an urban area of Honduras , Sanchez et al . [37] reported that several indicators of poor living conditions such as raising pigs , lack of potable water , lack of sanitary toilets and earthen floor were associated with an increased prevalence odds of exposure to cysticercosis . Poor hygienic conditions of the household was associated with an increased prevalence odds of exposure to infection in the area of Morelos , Mexico [34] . The inclusion of 60 villages in our study allowed us to assess the village-level effect of the type of pig management on the prevalence odds of current cysticercus infection . To our knowledge , this is the first time that a study has a number of sampled villages large enough to explore and measure village-level factors . Previous studies conducted in Latin America had found that ownership of pigs was associated with the prevalence odds of exposure to infection [28 , 35–37]; we found that how pigs are managed at the village-level has an impact . No previous studies conducted in Sub-Saharan Africa had found such effect . Indeed , ownership of pigs by the respondent did not yield an association with the prevalence of cysticercosis while pig ownership at the household level did in univariate analyses . This should be considered in future studies . In the study areas , almost all pigs were left to roam during the dry season . However , villages where a larger percentage of pigs were not penned during the rainy season had higher prevalence odds of current cysticercosis . This design also allowed us to find an association between the type of soil and the soil pH and the prevalence odds of current infection . An increase in the pH of the soil was associated with lower village-level prevalence odds of current infection . In an experimental study of inactivation of T . solium eggs in different temperature , dryness and pH conditions , an increase in alkalinity in an alkaline environment ( pH from 12 . 1 to 12 . 7 ) was linked to an increase in inactivation rates of the eggs while the opposite was true for in an acidic environment ( pH from 5 . 1 to 5 . 5 ) [39] . The soil pH in our study villages was at an average of 6 . 8 with a range from 5 . 4 to 8 . 2 . The soil was therefore nearly neutral on average , and it is difficult to say if the results are consistent to these found in [39] . However , it could be hypothesize that the eggs are more tolerant to the generally more acidic environment of the gastro-intestinal tract , which could favor better egg survival in slightly more acidic soil . An increase in the percentage of sand in the soil was associated with an increased village-level prevalence odds of current infection . Perhaps taeniid eggs are more easily disseminated from sandy soil onto vegetables and water through wind . The fact that the soil was sampled nearly three years after the baseline study may also have an impact , although it is unlikely for the soil to change its pH extensively through time . A self-report of a history of infection with taeniosis was associated with current infection with cysticerci only in univariate analyses . This association was confounded by age , and became non-significant in the hierarchical model also including pork management variable . Others had found associations between the self-report of taeniosis and exposure to cysticercosis in univariate analyses [34] . This underlines the importance of conducting multivariate analyses to identify factors with the strongest impact on infection . This study had some limitations . The most important one is that the soil samples were collected nearly three years after the baseline study , therefore , these associations should be interpreted with great care . This is why results were presented with and without including the soil analyses . Future studies using follow-up data of the randomized trial among the control group should be able to confirm ( or not ) this association . Only one person per household was sampled and therefore , factors affecting clusters within households could not be evaluated . In conclusion , this study is the first to assess the association between several individual- , household , and village-level variables and the prevalence odds of cysticercosis in humans at a large scale . We found that factors linked to people , pigs , and the environment were of importance . This further supports the need for a One Health approach to control this infection .	Taenia solium is an infection that is transmitted between pigs and humans . Humans may get infected with the larvae of Taenia solium , which results in cysticercosis , an infection common in pig farming communities where there is poor sanitation and free roaming pigs . Most published studies on this infection have included less participants covering a restricted geographic area , thereby resulting in a limited understanding of the important risk factors for infection . Our study aimed to examine important individual- , household- and village-level characteristics associated with current infection using baseline data from 3609 participants living in 60 villages across three provinces in Burkina Faso . Blood samples from village participants were taken to assess whether they were infected with cysticercosis . We found that eating pork , especially in other village’s markets , being older and male , living in a poorer household , not having access to a latrine , and living in a village where a larger percentage of pigs are left roaming were associated with infection . Soil pH and composition may also play a role in infection . Our results suggest that interventions that include human and veterinary health as well as environmental components should be considered to effectively control cysticercosis in such settings .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[]	2015	Prevalence of and Factors Associated with Human Cysticercosis in 60 Villages in Three Provinces of Burkina Faso
Whipworms and blood flukes combined infect almost one billion people in developing countries . Only a handful of anthelmintic drugs are currently available to treat these infections effectively; there is therefore an urgent need for new generations of anthelmintic compounds . Medicinal plants have presented as a viable source of new parasiticides . Ajania nubigena , the Bhutanese daisy , has been used in Bhutanese traditional medicine for treating various diseases and our previous studies revealed that small molecules from this plant have antimalarial properties . Encouraged by these findings , we screened four major compounds isolated from A . nubigena for their anthelmintic properties . Here we studied four major compounds derived from A . nubigena for their anthelmintic properties against the nematode whipworm Trichuris muris and the platyhelminth blood fluke Schistosoma mansoni using the xWORM assay technique . Of four compounds tested , two compounds—luteolin ( 3 ) and ( 3R , 6R ) -linalool oxide acetate ( 1 ) —showed dual anthelmintic activity against S . mansoni ( IC50 range = 5 . 8–36 . 9 μg/mL ) and T . muris ( IC50 range = 9 . 7–20 . 4 μg/mL ) . Using scanning electron microscopy , we determined luteolin as the most efficacious compound against both parasites and additionally was found effective against the schistosomula , the infective stage of S . mansoni ( IC50 = 13 . 3 μg/mL ) . Luteolin induced tegumental damage to S . mansoni and affected the cuticle , bacillary bands and bacillary glands of T . muris . Our in vivo assessment of luteolin ( 3 ) against T . muris infection at a single oral dosing of 100 mg/kg , despite being significantly ( 27 . 6% ) better than the untreated control group , was markedly weaker than mebendazole ( 93 . 1% ) in reducing the worm burden in mice . Among the four compounds tested , luteolin demonstrated the best broad-spectrum activity against two different helminths—T . muris and S . mansoni—and was effective against juvenile schistosomes , the stage that is refractory to the current gold standard drug , praziquantel . Medicinal chemistry optimisation including cytotoxicity analysis , analogue development and structure-activity relationship studies are warranted and could lead to the identification of more potent chemical entities for the control of parasitic helminths of humans and animals . The World Health Organization ( WHO ) recognises 17 different ‘neglected tropical diseases’ ( NTDs ) that affect more than 1 . 4 billion people in 149 countries [1] . Helminth infections caused by roundworms ( nematodes ) and flatworms ( platyhelminths ) comprise the largest group of NTDs [2] . Whipworms ( Nematoda ) cause trichuriasis and infect about 800 million people worldwide , second among the nematodes only to Ascaris infection [3] . The schistosome blood flukes ( Platyhelminthes ) cause schistosomiasis , a disease that afflicts more than 240 million individuals and kills hundreds of thousands each year [4] . A variety of approaches have been employed to combat these infections including education , vector control , sanitation and hygiene , behavioural change and mass drug administration ( MDA ) programs . Various in vitro and animal model studies have highlighted the repurposing of existing drugs and discovery and development efforts for new drugs [2 , 5] but all things considered , the pipeline for the next generation of anthelmintic drugs is sparse . Indeed , a systematic assessment of databases of drug regulatory authorities and the WHO , as well as clinical trial registries , revealed that no new antiparasitic drugs have been approved during the last decade [6] . There are only a handful of anthelmintic drugs on the market , some of which have unwanted side effects or achieve poor cure rates due to primary drug resistance developing in the parasites [7–9] . For example , praziquantel , which is the sole frontline drug used in the mass treatment of schistosomiasis , is efficacious but has many disadvantages: a ) it is ineffective against juvenile stages of the parasite , b ) reduced efficacy has been reported in field studies [10] , c ) there is a strong possibility that praziquantel resistance could appear if sufficient selection pressure is applied and mass drug administration is continued [11] , and d ) its active ( S ) -enantiomer and inactive ( R ) -enantiomer components remain inseparable in the production process , rendering bulky tablets that discourages patients from taking the right doses or the complete dosing regimen , which could trigger the development of drug resistance [12] . Until new arsenals of safe and effective drugs and/or vaccines are made available , helminth infections will continue to affect the world’s most impoverished populations , causing significant morbidity and mortality worldwide . While new drugs can be developed synthetically , natural products—especially the medicinal plants—have been an important pool of antiparasitic drugs . Quinine and artemisinin discovered from medicinal plants continue to save the lives of millions of people worldwide . As such , the notion of therapeutics derived from medicinal plants has re-surfaced [13] . Crude extracts and compounds of plant origin have been demonstrated to possess broad biological activities in in vitro and ex vivo assays and animal models of parasitic infections [14–19] . Edwards et al . [20] showed that 7-keto-sempervirol isolated from the boxthorn from which goji berries are harvested , Lycium chinense , was effective against Schistosoma mansoni and Fasciola hepatica . A compound that displays such broad anti-parasitic activity against various life stages of multiple parasites is highly desirable . Extracts of the Bhutanese medicinal plant from the flowering daisy family , Ajania nubigena ( Syn . Tanacetum nubigenum DC . ) have been previously shown to possess broad biological activities including antiparasitic effects against Plasmodium falciparum and antimicrobial properties [21] . It is locally known as m . khan-d . kar and has been used in Bhutanese traditional medicine ( derived from Tibetan scholarly medicine ) for thousands of years as incense and for treating an array of conditions and infections including wounds , bleeding and swelling [21] . Although this plant is not specifically indicated for treating intestinal worms , the decoction of its closely related species , Tanacetum parthenium L . ( feverfew ) and Tanacetum dolichophyllum Kitam has been traditionally used by the Ladakhis Amchis ( medical system derived from Tibetan medicine and similar to Bhutanese traditional medicine ) [22] and Costa Ricans healers [23] against intestinal worms . These plants have reserves of highly aromatic essentials oils that have evolved to aid in plant protection and competition against plant parasites and herbivorous insects . Chemically , these plants contain similar chemotypes including sesquiterpenes and flavonoids [23] . Encouraged by these lead information , we have investigated the anthelmintic properties of four compounds isolated from the Bhutanese A . nubigena against two of the most important genera of human helminth parasites , the nematode whipworm ( Trichuris ) and the platyhelminth blood fluke ( Schistosoma ) . To monitor worm viability we used xWORM , a technique that monitors helminth motility in real time using xCELLigence [24–25] . The advantage of using xWORM over other methods is that it enables high-throughput screening of a large number of compounds in a fully automated , label-free manner . The aerial part of wild Ajania nubigena was collected from alpine mountains ( altitude range of 3600–4800 meters above sea level ) of Lingzhi , Bhutan in August 2009 . The collected plant material was air-dried and a herbarium specimen with voucher number 73 was deposited at the herbarium collection section of Menjong Sorig Pharmaceuticals , Ministry of Health , Bhutan . The air-dried plant material ( 2 kg ) was chopped into small pieces and was repeatedly extracted with methanol ( AR/HPLC grade , 3 L over 48 h ) . The extract was filtered and then concentrated using a Buchi rotary evaporator to generate a crude methanol ( MeOH ) extract ( 58 . 2 g ) . The isolation technique was described previously [21] . MeOH extract was dissolved in MeOH:H2O ( 200 mL , 1:9 ) and then fractionated with hexane followed by ethyl acetate to obtain the hexane extract ( 28 . 0 g ) and the ethyl acetate extract ( 12 . 5 g ) , respectively . Subsequently , essential oil ( EO ) extraction was performed using hydro-distillation ( 60°C ) . One kg of dried plant material yielded 7 mL of pale green EO . The crude MeOH extract and EO were subjected to extensive natural products isolation processes . Flash column chromatography packed with Merck Kieselgel 60 PF254 and pre-coated silica plates ( 0 . 2 mm silica thickness , Merck ) were used for repeated separation and purification of compounds . UV light ( short wavelength of 254 nm , long wavelength of 366 nm ) and ceric ammonium molybdate ( CAM ) were used for visualization and detection of compounds on Thin Layer Chromatography ( TLC ) plates . Eight compounds were isolated and characterised in total from the MeOH and EO extracts using Infrared ( IR ) Spectroscopy , Mass Spectrometry ( ESI-MS , HR-EI-MS ) , Gas Chromatography Mass Spectrometry ( GCMS ) , and Nuclear Magnetic Resonance ( NMR-1H , 13C , gCOSY , gNOESY , TOCSY , gHSQC and gHMBC ) [21] . In this study , we have selected four major compounds whose structures are produced in Fig 1: ( 3R , 6R ) -linalool oxide acetate ( 1 ) , ( E ) -spiroether ( 2 ) , luteolin ( 3 ) and luteolin-7-O-β-D-glucopyranoside ( 4 ) . The stock solutions of the four test compounds were prepared at the concentration of 100 mg/mL in DMSO and then subsequently diluted them with respective tissue culture media to make 10x solutions . 20 μl of 10× drug concentration was added to 180 μl of media containing helminths in the E-plate wells . Control worms were cultured in the presence of DMSO equivalent to that used for the highest drug concentration; this group was used to determine 100% motility . For schistosomula drug assays , stock solutions were diluted in culture media with two-fold dilutions with in-well concentrations of ( 2–1000 μg/mL ) . S . mansoni cercariae were shed from infected Biomphalaria glabrata snails ( Biomedical Research Institute , MD , USA ) by exposure to light at 26°C for 2 hours and used to infect 12–14 week old male BALB/c mice ( 120 cercariae/mouse ) by abdominal penetration [26] . Adult flukes were perfused from the mesenteries 7 weeks post-infection and then transferred to Basch medium for culturing as previously reported [27] . To monitor the effects of drugs on S . mansoni schistosomula , shedding of cercariae from snails and subsequent in vitro transformation to schistosomula was performed as described by Peak et al [28] . Genetically susceptible mice ( STAT6-1- ) were orally infected with T . muris eggs ( 200 μL volume containing ~ 200 eggs ) and sacrificed after 4 weeks . Adult worms were harvested from the caecum , washed with PBS/2x antibiotic/antimycotic ( AA ) and resuspended in 100 μl of RPMI containing 10% foetal calf serum and AA ( culture medium ) then transferred to E-plates for motility assessment using the xWORM assay . The trematocidal effects of test compounds against S . mansoni were evaluated using an xCELLigence SP system ( ACEA Biosciences ) as described by us [25] . Adult flukes ( 1 fluke per well ) were placed in triplicate for each compound into 96 well E-plates ( ACEA Biosciences ) containing 180 μl of culture medium and cultured overnight at 37°C with 5% CO2 to obtain a baseline motility reading . Test compounds were added to E-plates and motility was monitored for 12–40 hr . The inter-well spaces of E-plates were filled with 100 μL culture media . All experiments were carried out as per manufacturer’s instructions with 15 sec read intervals using the real time cell assay ( RTCA ) software ( ACEA Biosciences ) as described previously [24–25] . Similarly , the nematocidal effects of test compounds against T . muris were assessed using the same xCELLigence SP system as described above . We determined the optimal culture duration and worm concentration to maximize the signal to noise ratio using the xWORM technique for the first time with T . muris . Different numbers of adult T . muris ( 2 , 4 and 8 ) were added to individual E-plate wells and motility was monitored overnight . Four worms of mixed gender per well in a final volume of 200 μl of culture medium was determined to be optimal for this study . The E-plates containing worms were treated with prepared concentrations of the test compounds and were monitored using the xCELLigence SP system for 12–40 hr . Inter-well spaces were filled with 100 μL of culture medium or PBS to prevent evaporation . Each set of conditions was monitored in triplicate . The 96 well plates containing culture media were loaded with schistosomula ( 100 μL volume containing ~ 100 schistosomula ) in triplicate and treated with the test compounds at various in-well concentrations of 2–1000 μg/mL . Plates were cultured at 37°C with 5% CO2 for 12–40 hr and were finally stained with trypan blue solution to assess final viability after treatment . The stained schistosomula were observed by light microscopy , and live and dead flukes in each well were counted manually and 50% inhibitory concentration ( IC50 ) values were obtained . The IC50 values of test compounds were determined based on the motility index for adult worms as described by us [25] . Briefly , motility index was calculated as the standard deviation ( SD ) over 800 data points ( i . e . 4 readings per min for 200 min ) of the cell index ( CI ) difference from the rolling average over 20 data points ( 10 proceeding and preceding CI values—5 min total ) . One hundred percent motility was determined from the average motility index of the untreated wells , while 0% motility was determined from a media only well ( no worms present ) . The motility index averaged over 100 data points ( 25 min ) was converted to percentage motility and this figure was used in GraphPad Prism 6 . 0 to calculate dose response curves . We used a log ( test compound concentration ) vs normalised response ( 100%–0% ) formula , with variable slope when data were sufficient or set -1 hill slope when data was limited , and automatic removal of outliers ( with default ROUT coefficient used: Q = 1 . 0% ) . IC50 values for each dose concentration were calculated at 1 hr , 6 hr , and 12 hr post-treatment of the worms with the test compounds . Compounds with IC50 values of higher than 100 μg/mL were considered ineffective in this study . Statistical analyses were undertaken using GraphPad Prism 6 . 0 . When data were sufficient to use the variable slope analysis , the Hill Slope and the Log IC50 value were together compared for significant differences using an extra sum-of squares F-test . For the in vivo mouse experiments , 1-way ANOVA with Holm-Sidak’s multiple comparisons test was used for determining significance p-values . Worms treated with the test compounds were prepared for scanning electron microcopy ( SEM ) as follows: a ) fixed in 3% gluteraldehyde in Sorensen’s buffer overnight , b ) dehydrated for 15 min in a graded ethanol series ( 50% , 60% , 70% , 80% , 90% , 100% ) , mixture of ethanol and hexamethyldisilizane ( HMDS ) ( 1:1 ratio ) and then finally with pure HMDS ( 100% ) , c ) the dehydrated worms were covered and left overnight in a fume hood to allow the HMDS to evaporate . Completely dried worms were placed on an aluminum stub ( at least three worms from each treatment regimen ) , sputtered with gold and visualized using a JEOL JSM scanning electron microscope operating at 10 kV . Each worm on a stub was scanned from head to tail to determine if the compounds had altered its gross morphology . Digital image acquisition was performed on the affected region of the worms using Semaphore software . Four to five week-old STAT6-1- mice were grouped ( each group with 9 mice ) as: solvent control , positive control and luteolin ( 3 ) . Each mouse was orally infected with 200 μl of PBS containing approximately 200 live embyronated eggs of T . muris . These mice were housed for 4 weeks with constant access to water and pelleted food . After 4 weeks , luteolin ( 3 ) and the positive control drug ( mebendazole ) prepared in 1% DMSO/PBS were orally administered at a single dose of 100 mg/kg ( 9 mice in total for each group ) as per the protocol [29–30] . Five days after one dose of oral treatment the mice were sacrificed , worms were harvested from the caecum , and counted manually using light microscopy . The recorded numbers of worms were averaged to find the percentage reduction in worm burden for each group of mice . The permit to collect medicinal plants from the park management areas around Lingzhi , Bhutan was obtained from the Department of Forest , Ministry of Agriculture and Forestry in Bhutan . The material transfer agreement and approval was sought from the National Biodiversity Centre of Bhutan . MeOH extracts of the plants were transported to Australia with prior approval from the Bhutan Agriculture and Food Regulatory Authority , University of Wollongong and sample inspections by Australian Quarantine & Inspection Service in 2010 . The James Cook University ( JCU ) animal ethics committee approved all experimental work involving animals ( Ethic approval number A2213 ) . Mice infected with S . mansoni and T . muris were raised in cages in the JCU animal facility for 4–7 weeks in compliance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes , 7th edition , 2007 and the Queensland Animal Care and Protection Act 2001 . Mice were kept under normal conditions at regulated temperature ( 22°C ) and lighting ( 12 hr light/dark cycle ) with free access to pelleted food and water . All reasonable efforts were made to minimise the suffering of the mice . Of the four compounds tested , ( 3R , 6R ) -linalool oxide acetate ( 1 ) , luteolin ( 3 ) and luteolin-7-O-β-D-glucopyranoside ( 4 ) showed anti-schistosome dose-dependent anti-schistosome effects ( Fig 2A ) . At the highest concentration tested ( 1 mg/mL ) all three compounds killed flukes within 1–12 hr . Lower drug concentrations , however , took longer to kill flukes reflected by higher motility index values ( Fig 2A ) . When the IC50 values of each compound were averaged or combined ( calculated for the dose concentrations of 0 . 1–1000 μg/mL ) for each time point ( 1 hr , 6 hr and 12 hr ) , luteolin ( 3 ) and luteolin-7-O-β-D-glucopyranoside ( 4 ) appeared to be fast acting on worms as their IC50 values did not change significantly between the initial and final 12 hr dosing time points ( Fig 2B ) . On the other hand , ( 3R , 6R ) -linalool oxide acetate ( 1 ) was slow-acting , showing two-fold decreases in IC50 at each 6 hr time point . Of the compounds assessed , luteolin ( 3 ) exhibited significantly better trematocidal activity at both 6 hr and 12 hr time points with IC50 values of 4 . 6 μg/mL and 5 . 8 μg/mL , respectively . The compounds that were active against adult S . mansoni were also tested against the intra-mammalian larval stage , the schistosomulum . Monitoring of schistosomula survival in the presence of different drug concentration using Trypan blue exclusion showed that luteolin ( 3 ) started to show lethal effects at the lowest dilution of 3 . 91 μg/mL and achieved 98–100% killing at the dilution of 31 . 3 μg/mL . ( 3R , 6R ) -linalool oxide acetate ( 1 ) however started to show lethal effects at a concentration of 125 μg/mL and only achieved about 43% killing at the maximum dose tested of 250 μg/mL . Schistosomula treated with 1% DMSO alone ( solvent control ) had 100% survival as measured by Trypan blue exclusion . A dose response curve of schistosomula survival after treatment with luteolin ( 3 ) revealed an IC50 of 13 . 3 μg/ml ( Fig 3B ) . Luteolin ( 3 ) demonstrated the best anthelmintic activity when motility was assessed using xWORM . The effect of this compound and praziquantel on adult fluke morphology at 4–20 μg/mL concentration , with particular emphasis on the tegument , was assessed by SEM . Adult flukes treated with praziquantel adopted a tightly coiled appearance due to contraction . Both male and female flukes treated with luteolin ( 3 ) were contracted and coiled compared to control flukes cultured in 1% DMSO in media , but not as tightly coiled as praziquantel-treated parasites ( Fig 4A–4F ) . Observation of fluke teguments by SEM under high magnification revealed that the tegument of DMSO-treated flukes displayed numerous tubercles , with well-formed spines in the males ( Fig 5A ) and clearly defined surface grooves with sensory papillae in females ( Fig 5B ) , and oral suckers with clearly defined pits containing sharp spines ( Fig 5C ) . Flukes treated with praziquantel ( Fig 5D–5F ) and luteolin ( 3 ) ( Fig 5G–5I ) exhibited severe morphological alterations of the tegument . At the lowest concentration tested of 4 μg/mL we observed clear tegumental damage induced by luteolin ( 3 ) ( Fig 5G–5I ) , similar to that induced by praziquantel ( Fig 5D–5F ) . While male flukes ( Fig 5G ) suffered partial loss of pits and their encased spines , female flukes ( Fig 5E ) showed surface erosion , constriction of grooves , bursting of small sensory papillae and formation of holes on the tegument . At higher concentrations of 20–1000 μg/mL , luteolin ( 3 ) and praziquantel completely destroyed the body surfaces and exhibited erosion of tubercles , oral and ventral suckers ( S1 Fig ) . Prior to testing the nematocidal effects of ( 3R , 6R ) -linalool oxide acetate ( 1 ) , ( E ) -spiroether ( 2 ) , luteolin ( 3 ) and luteolin-7-O-β-D-glucopyranoside ( 4 ) , we standardized the culturing conditions of adult T . muris for the xWORM assay . E-plate wells containing four adult worms ( both males and females ) yielded optimal motility signals ( S2 Fig ) , and was the condition selected for subsequent anthelmintic screening of the test compounds . ( 3R , 6R ) -linalool oxide acetate ( 1 ) and luteolin ( 3 ) showed the best anti-Trichuris activity with IC50 values of 20 . 4 μg/mL and 9 . 7 μg/mL , respectively , calculated on cell motility index at the 12 hr time point ( Fig 6A ) . The IC50 values of the four compounds tested here were obtained using xWORM and calculated at 1 hr , 6 hr and 12 hr time points ( Fig 6B ) . Luteolin ( 3 ) was the most efficacious drug in terms of reduced motility of T . muris , exhibiting the lowest or equally low IC50 values at all time points and a final 12 hr value of 9 . 7 μg/mL . Based on the efficacy of luteolin ( 3 ) at reducing Trichuris motility , we examined the morphological changes in the cuticle induced by this compound using the SEM protocols described by Stepek et al [30] and Tritten et al [31] specific to T . muris . Live adult T . muris ( mixed sexes ) were treated for 48 hr with luteolin ( 3 ) or mebendazole at dose concentrations of 200–1000 μg/mL . Morphological changes were observed towards the anterior end of the worms in the form of partially damaged bacillary band/glands and adjacent cuticle . The bacillary band is a specialized row of longitudinal cells of some nematodes consisting of glandular and non-glandular cells . These bands host the glands . Worms treated with DMSO/media ( solvent control ) alone had a moderately coiled appearance with a smooth cuticle displaying knitted parallel segmental joins and minimal shrinkage of bands ( Fig 7A–7C ) in comparison to the groups treated with mebendazole ( Fig 7D–7F ) and luteolin ( 3 ) ( Fig 7G–7I ) . At higher magnification we observed that the luteolin-treated worms exhibited blister-like formations on the surface of the cuticle , swelling and loosening of cuticle seams/grooves near the bacillary glands ( Fig 7I ) . These morphological changes were similar to that of the mebendazole-treated worms ( Fig 7F ) . Based on the significant in vitro nematocidal activity demonstrated by luteolin ( 3 ) , this compound was further assessed for its anti-Trichuris effect in vivo using a mouse model of T . muris infection [29–31] . Four weeks post-infection with T . muris eggs , mice were administered with a single 100 mg/kg oral dose of luteolin ( 3 ) . When mice were sacrificed 5 days later we observed that a single treatment of this compound resulted in a 27 . 6% reduction in worm burdens ( 249 worms ) ; luteolin ( 3 ) –treated mice = 651 worms; DMSO-treated mice = 900 worms; P = 0 . 0087 ) ( Fig 8 ) . The positive control drug , mebendazole , reduced worm burdens by 93 . 1% ( mebendazole treated mice = 50 worms ) . The activity of luteolin ( 3 ) against T . muris , despite being significantly better than the untreated control group , was markedly weaker than mebendazole in the mouse model . Mice treated with luteolin ( 3 ) did not show signs of toxicity at any of the concentrations tested . Globally , helminth infections caused by roundworms ( nematodes ) and flatworms ( platyhelminths ) comprise the largest group of NTDs [2] . Schistosomiasis ( caused by the platyhelminth blood flukes ) and trichuriasis ( caused by the nematode whipworms ) affect about 240 million and 800 million people , respectively [3–4] . While schistosomiasis is transmitted through infected snails and water , trichuriasis is transmitted through soil and is therefore referred to as a soil-transmitted helminth infections ( STHI ) . The STHI are among the most common infections worldwide and affect the poorest and most deprived communities where sanitation is poor . Sole reliance on praziquantel for schistosomiasis , and only a very small number of drugs ( some with poor cure rates ) for STHI , has precipitated the need for new anthelmintic drugs to treat parasites that infect both humans and animals [7] . In this context , medicinal plants present a viable source of novel anthelmintic compounds . Indeed , many anti-parasitic drugs including quinine , chloroquine , artemisinin and atovaquone were originally purified from plants . Our initial study , involving the crude CHCl3 extract of A . nubigena ( syn . Tanacetum nubigenum DC . ) and its compounds luteolin and luteolin-7-O-β-D-glucopyranoside , showed significant antiparasitic activities against the multidrug resistant K1CB1 strain and chloroquine sensitive TM4/8 . 2 strain of Plasmodium falciparum [16 , 21] . Interestingly , this plant and its close relatives have been used in the ethnomedicines for treating arrays of disorders including wound , bleeding and worm infection [16 , 21–23] . In this study , we have demonstrated that compounds linalool oxide acetate ( 1 ) and luteolin ( 3 ) had significant trematocidal activity against S . mansoni and nematocidal activity against T . muris . These compounds are simple small secondary plant metabolites belonging to the terpenes and flavonoids . Luteolin ( 3 ) was the most active compound against both parasites with IC50 values of 5 . 8 μg/mL against S . mansoni and 9 . 7 μg/mL against T . muris calculated at the 12 hr time point . It also effectively killed schistosomula with an IC50 value of 13 . 3 μg/mL . Intriguingly , this same compound has been shown to have significant anti-malarial activity against P . falciparum TM4/8 . 2 ( chloroquine-antifolate sensitive strain ) and K1CB1 ( multidrug resistant strain ) [21] . New anti-parasitic drugs require excellent safety and therapeutic profiles , should exhibit broad spectrum activity against different types of infections , and also display significant activity against different developmental stages of parasites . Current anthelmintic drugs are generally effective at treating single stages of target helminths . For example , praziquantel is effective against adult stage schistosomes but not schistosomula/cercariae . Therefore , finding a broad-spectrum drug that could treat multiple diseases or multiple life stages is desirable when treating large populations in resource-poor settings . Luteolin ( 3 ) met these criteria in that it has anti-malarial [21] , anti-fluke and anti-whipworm properties . In addition , this compound was effective in killing schistosomula , the stage of S . mansoni that is unaffected by praziquantel . While our findings do not specifically address the mechanism of action of these anthelmintic compounds , we showed that luteolin ( 3 ) is capable of damaging the outer surface membranes of the parasites–the fluke tegument and the nematode cuticle and associated glandular structures , and worms adopted a coiled state of paralysis . Previous studies on S . mansoni have demonstrated that the tegument plays essential roles in many processes at the host-parasite interface , and numerous molecular pathways that are represented at the host-parasite boundary are anti-parasitic drug targets [5 , 32] . SEM has been used to demonstrate the mechanisms of many anti-parasitic drugs , including oxamniquine , praziquantel , mefloquine , mebendazole and artemisinin [20 , 32–39] . Schistosomes treated with these drugs displayed vacuolization or bubble-like-lesions , surface erosion , destruction of tubercles and tissues , loss of sensory papillae , and pore formation leading to death . SEM has also been used to reveal cuticular damage in nematodes , particularly for the benzimidazole class of drugs [40–41] . Stepek et al [30] and Tritten et al [31] first used the plant extracts and nitazoxanide to demonstrate the mechanisms of the antiparasitic action against T . muris . These same morphological changes were observed in adult S . mansoni and T . muris that were treated with luteolin ( 3 ) , suggesting that the mechanism of action is similar to at least some of the existing anthelmintic drugs . For T . muris , luteolin ( 3 ) and mebendazole mainly affected the anterior bacillary band and surrounding glands . The anterior ventrolateral face of the worm contain the glandular pores , bacillary band sensory cells and glands , and the stichosome cells which helps in the formation of the syncytial feeding site in the host , and plays an important role in the excretion of digestive enzymes , pre-digestion and nutrient uptake in Trichuris [42–45] . Luteolin ( 3 ) , with a LogP value of 2 . 6 , meets the Lipinski rule of 5 criterion for drug-likeness [46] . Generally , compounds with LogP values in the range of 2–3 are more likely to diffuse/permeate through the cell membrane of an organism , and therefore enabling them to interact with target receptors . There was no structural similarity between our active compounds and the currently used anthelmintic drugs , which suggest that compounds that damage the worm surface do not necessarily have similar structural scaffolds . It should be noted that our in vivo assessment of luteolin ( 3 ) against T . muris infection at a single oral dosing of 100 mg/kg , despite being significantly ( 27 . 6% ) better than the untreated control group , was markedly weaker than mebendazole ( 93 . 1% ) in reducing the worm burden in mice . Moreover , the mice showed no signs of ill health , suggesting that luteolin ( 3 ) is not overtly toxic and allows future studies to explore the efficacy of multiple treatments . We did not assess in vivo efficacy of luteolin ( 3 ) against S . mansoni in the mouse model . Future work will entail synthesis of luteolin and its derivatised compounds in a thorough in vivo assessment of efficacy in mouse models of schistosomiasis and trichuriasis , as well as other soil-transmitted helminth infections .	Schistosomiasis and trichuriasis affects millions of people worldwide and are caused by blood flukes and whipworms , respectively . Only a handful of anthelmintic drugs exist to treat these infections and the pipeline for the next generation of anthelmintic drugs is sparse , precipitating the need for new drug development . In this context , medicinal plants present a viable source of novel anthelmintic compounds . This inspired us to study the selected naturally occurring compounds derived from a Bhutanese daisy medicinal plant , Ajania nubigena for their anthelmintic activities . Here , using the xWORM motility assay , we demonstrate that two compounds , luteolin ( 3 ) and ( 3R , 6R ) -linalool oxide acetate ( 1 ) , display significant broad-spectrum anthelmintic activity against two of the most important genera of human helminth parasites , the nematode whipworm and the platyhelminth blood fluke . Luteolin exhibited the best activities with IC50 values of 5 . 8 μg/mL against schistosomes and 9 . 7 μg/mL against whipworms . Using scanning electron microscopy we showed that luteolin damages the tegument of blood flukes and induces abnormalities in the bacillary bands/glands and cuticles of whipworms . Intriguingly , our previous study showed that luteolin ( 3 ) was effective against multi-drug resistant Plasmodium falciparum malaria . Due to its broad-spectrum anti-parasitic activities , luteolin ( 3 ) is a desirable drug lead scaffold , which could be used for developing effective compounds to control and treat numerous tropical diseases .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "schistosoma", "invertebrates", "schistosoma", "mansoni", "medicine", "and", "health", "sciences", "helminths", "tropical", "diseases", "parasitic", "diseases", "animals", "nematode", "infections", "animal", "models", "model", "organisms", "pharmaceutics", "medicinal", "plants", "neglected", "tropical", "diseases", "pharmacology", "plants", "research", "and", "analysis", "methods", "mouse", "models", "helminth", "infections", "trichuriasis", "drug", "research", "and", "development", "biology", "and", "life", "sciences", "drug", "therapy", "soil-transmitted", "helminthiases", "organisms" ]	2016	Compounds Derived from the Bhutanese Daisy, Ajania nubigena, Demonstrate Dual Anthelmintic Activity against Schistosoma mansoni and Trichuris muris
Livestock populations can be used to study recessive defects caused by deleterious alleles . The frequency of deleterious alleles including recessive lethal alleles can stay at high or moderate frequency within a population , especially if recessive lethal alleles exhibit an advantage for favourable traits in heterozygotes . In this study , we report such a recessive lethal deletion of 212kb ( del ) within the BBS9 gene in a breeding population of pigs . The deletion produces a truncated BBS9 protein expected to cause a complete loss-of-function , and we find a reduction of approximately 20% on the total number of piglets born from carrier by carrier matings . Homozygous del/del animals die mid- to late-gestation , as observed from high increase in numbers of mummified piglets resulting from carrier-by-carrier crosses . The moderate 10 . 8% carrier frequency ( 5 . 4% allele frequency ) in this pig population suggests an advantage on a favourable trait in heterozygotes . Indeed , heterozygous carriers exhibit increased growth rate , an important selection trait in pig breeding . Increased growth and appetite together with a lower birth weight for carriers of the BBS9 null allele in pigs is analogous to the phenotype described in human and mouse for ( naturally occurring ) BBS9 null-mutants . We show that fetal death , however , is induced by reduced expression of the downstream BMPER gene , an essential gene for normal foetal development . In conclusion , this study describes a lethal 212kb deletion with pleiotropic effects on two different genes , one resulting in fetal death in homozygous state ( BMPER ) , and the other increasing growth ( BBS9 ) in heterozygous state . We provide strong evidence for balancing selection resulting in an unexpected high frequency of a lethal allele in the population . This study shows that the large amounts of genomic and phenotypic data routinely generated in modern commercial breeding programs deliver a powerful tool to monitor and control lethal alleles much more efficiently . Domesticated animals are excellent models to study the effect of inbreeding on fitness , and the role of selection in inbreeding depression . Breeding of domesticated animals increases inbreeding by applying artificial insemination that allows breeding populations to be sired by a small number of elite males . The frequency of deleterious alleles including recessive lethal alleles can rise in populations as a consequence of drift due to small effective population size , but also due to selection [1] . Inherited defects usually derive from unique “founder” mutations [2] . Especially in cattle breeds , several high frequency lethal alleles have been described [3 , 4] reaching carrier frequencies up to 32% [5] , that can be traced back to prime bulls that were used extensively in the past decades . However , the effect of individual sires on the population depends on the breeding goal and the structure of the breeding program . In cattle breeding , the genetic contribution of a single bull can be extreme , producing up to hundreds of thousands of daughters . In pig breeding , however , drift effects are expected to be less severe because recessive lethal alleles from founder boars are less likely to rise in frequency very rapidly , because of a lower male selection intensity compared to cattle breeding [6] . The role of random drift and/or selection in increasing the frequency of deleterious variants is complex . When effective population size is small , drift effects can result in less effective selection [7] . Interestingly , the number of lethal variants found at relatively high frequency in commercial pig populations appears to be low [8–10] . The relative paucity of high-frequency deleterious alleles in pig and chicken , species that generally show a more gender-balanced selection [6 , 11] , and larger effective population size compared to cattle breeds [12 , 13] , raises the question why still some alleles rise to moderate frequency despite having a very clear adverse effect . Heterozygote advantage for traits selected in commercial populations provides a tantalizing alternative hypothesis [14] . In cattle , various instances of balancing selection have been described , driving deleterious alleles to higher population frequencies [5 , 15] . In pigs , similar observations were made involving a transposable element ( L1 ) insertion with positive effect on litter size , but negative consequences for boar fertility [16] . In a previous study we identified various recessive lethal alleles in three pig breeds [8] , but the majority of these lethal alleles were found at low frequencies . One recessive lethal haplotype , however , was found at moderate frequency ( ~9% carrier frequency ) causing a significant increase in foetal mortality at mid- to late-gestation and resulting in a high fraction of mummified piglets in a Large White commercial population . The strong deleterious nature of the allele and the high frequency suggests a factor other than drift driving this haplotype to high frequency . In this study , we report evidence of balancing selection on a recessive lethal 212kb deletion within the BBS9 gene with antagonistic effects on fertility and growth . The allele affects fertility by causing early fetal death in homozygous progeny , resulting in mummified piglets . The same allele increases growth rate and feed intake for carrier animals compared to non-carrier animals . We propose that the deletion is maintained at moderate frequency in the Large White breed because of its association with this positive effect , despite it being lethal in homozygous state . Genomic loci that harbour recessive lethal alleles can be identified by searching for haplotypes showing reduced or missing homozygosity . In this study , we analysed a previously identified recessive lethal haplotype on pig chromosome 18 ( SSC18: 39 . 25–40 . 1 Mb ) using 23 , 722 Large White animals from a single purebred sow line genotyped on the Porcine50K SNPchip ( Sscrofa11 . 1 build ) . The haplotype frequency is estimated at 5 . 4% ( 10 . 8% carrier frequency , Table 1 ) , showing that the haplotype is segregating at moderate frequency in this Large White population . In total , we expect 55 homozygote carriers for the SSC18 haplotype within the population . However , no homozygous del/del animals were observed , supporting that all copies of the haplotype carry the recessive lethal variant exhibiting complete penetrance for homozygous animals . We also observe a significant reduction in total number born ( 19 . 5% ) and liveborn individuals ( 19 . 3% ) for carrier-by-carrier matings ( CxC ) compared to carrier-by-non-carrier matings ( CxNC ) . Moreover , we found an approximate fivefold increase in mummified piglets ( Table 1 ) . The difference between stillborn and mummies lies in the moment the foetus dies: The term ‘mummy’ is used for a foetus that dies mid-to-late-gestation ( e . g . second to third trimester ) and is subsequently encapsulated and desiccated during the remainder time of the pregnancy . A foetus that dies near the end of gestation or perinatally is identified as ‘stillborn’ . The reduction in total number born is slightly lower than the expected 25% based on the 1:2:1 genotype distribution expected from CxC matings . About 73% of the CxC progeny is heterozygous for the SSC18 haplotype , corresponding to the 1:2 genotype ratio expected for CxC matings that lack homozygous offspring , significantly different compared to the normal 1:2:1 Mendelian ratio ( p = <0 . 00001 ) . Based on the carrier frequency , we estimate that about 1 . 17% of the litters within this breed are affected by the SSC18 haplotype , producing affected animals ( ‘mummies’ ) , and resulting in reduced litter sizes ( on average 3 . 08 piglets per CxC litter ) . We tracked five recent CxC matings . Four pregnancies reached full term , while one resulted in spontaneous early abortion of the entire litter ( Table 2 ) . The four full-term litters produced 49 liveborn , 7 stillborn , and 14 mummified piglets . Each of these four litters produced at least 2 mummified piglets ( maximum 5 ) , significantly more than what is normally observed in this breed ( on average 0 . 35 mummified piglets per litter , p = 0 . 0027 ) . Among the total of 48 genotyped liveborn and stillborn siblings ( 8 siblings were not genotyped ) , 16 were non-carriers , 30 were heterozygous ( 62 . 5% ) , and two were homozygous for the SSC18 haplotype , close to the expected 1:2 genotype ratio caused by missing homozygous offspring ( S1 and S2 Tables ) . Among the two "fresh born" homozygous animals ( i . e . piglets surviving at least until around birth ) , one was a stillborn piglet , the other was a liveborn but very weak piglet , that died shortly after birth . We confirmed the homozygous status for two mummified piglets with sufficient call rate ( call rate > 0 . 8 , S1 Table ) , the other mummified piglets yielded insufficient DNA quality to perform genotyping and phasing ( call rate < 0 . 8 , S1 and S2 Tables ) . Next , we collected eight mummified piglets from one farm for phenotypic evaluation ( including X-rays , S1 Fig ) , the other six mummified piglets were measured ( length ) , but not stored . The approximate age when a mummified pig has died can be determined based on the length ( crown to rump ) and weight . The majority of the mummified piglets die approximately in the second half of the second trimester of pregnancy ( 50–70 days ) , based on the length ( 100–200 mm ) and weight ( 100–190 gram ) of the mummified piglets ( S1 and S2 Tables , S1 Fig ) . Three mummified piglets from one litter ( litter ID: CC4 ) died later in gestation as was evident from a larger size and weight ( S1 Table ) . However , we cannot confirm the homozygous status for the SSC18 haplotype , since these animals could not be successfully genotyped due to poor DNA quality . Together these results support a broad range in the time of death between homozygous animals ( supporting variation in penetrance ) , ranging from 50 days in gestation to 24 hours post-partum . To identify candidate causal mutations , we analysed whole genome sequence data from 73 individuals from the same Large White population and identified 10 carrier animals for the SSC18 haplotype ( S5 Table ) . We first annotated loss-of-function and ( deleterious ) missense mutations within and surrounding the haplotype region ( +/- 5 Mb ) uniquely found in the SSC18 haplotype carriers . However , none of the mutations were predicted to have high impact ( Variant Effect Predictor , build 90 [17] ) . Next , we assessed the presence of structural variation within the same region and identified a large deletion in complete LD with the SSC18 haplotype of approximately 212kb ( position 39 , 817 , 373 to 40 , 029 , 300 ) , spanning a part of the BBS9 gene ( Fig 1A and 1B ) . The deletion is supported by both split-reads and discordantly mapped pairs in carrier samples ( S2 Fig ) . Moreover , carrier animals show reduced signal intensities ( referred to as Log R Ratio; Fig 1A , S3 Fig ) , and increased homozygosity for four markers on the Porcine50K SNPchip located within deletion , caused by the absence of a second haplotype for the deletion region . In addition , several markers neighbouring the deletion show an excess of heterozygosity , caused by the absence of homozygous del/del animals . We analysed RNA-seq data from one carrier animal in eight different tissue types ( sample: PigWur166 , S6 Table ) to investigate the impact of the deletion on the expression of BBS9 . Moderate gene expression levels for BBS9 were observed for the majority of the examined tissues , except for muscle , and with highest gene expression in testis ( S6 Table ) . We evaluated the effect of the deletion on the BBS9 mRNA and show that the deletion induces skipping of 4 coding , and 4 3'UTR exons for the BBS9 canonical transcript ( Fig 2 , RefSeq ID: XM_021079336 . 1 ) , resulting in direct splicing from exon 19 to exon 28 ( 3'UTR ) . The mutant transcript results in a frameshift introducing 11 novel amino acids before a premature stop codon , generating a truncated BBS9 protein of 694 amino acids ( including 11 novel amino acids ) instead of the wild type 865 amino acids . This truncated BBS9 protein will likely be non-functional ( Fig 2 ) , supported by pathogenic mutations identified in humans affecting the same C-terminal tail of the BBS9 protein [18 , 19] . Moreover , the affected protein coding exons exhibit a negative subRVIS score , indicating intolerance to loss-of-function mutations [20] . Finally , we evaluated the expression of BBS9 using a RT-qPCR on 8 carrier and 10 non-carrier samples from whole blood using primers that target exons located within the deletion . The results show a 50% lower expression of the wild-type BBS9 gene in carrier animals ( S4 Fig ) . To evaluate the impact of the deletion on the downstream BMPER gene we investigated possible allelic imbalance for the BMPER gene within the same carrier animal . The BMPER gene is highly expressed in lung , while moderately expressed in the other tissue types ( S6 Table ) . One heterozygous coding synonymous mutation within the fourth exon of the BMPER canonical transcript ( XM_013990842 . 2 ) was used to test for allelic imbalance . Interestingly , we observed a three-fold higher expression of the BMPER allele for the wild-type haplotype ( T allele ) compared to the del haplotype ( in lung tissue , Table 3 ) . By contrast , three homozygous wild-type animals showed no allele specific differences in expression for the BMPER gene ( S7 Table ) , suggesting that the region affected by the 212kb deletion contains BMPER cis-regulatory elements . To support the presence of BMPER regulatory elements within the deletion we aligned liver ChipSeq ( H3K27Ac , H3K4Me3 ) data [21] to the Sscrofa11 . 1 genome build . Two strong enhancer peaks are observed within the deletion region , while only weak signals are observed outside the deletion region ( S5 Fig ) . In addition , the sequence of the 212kb deletion was mapped to the human genome to identify the homologous sequence on the human genome ( GRCh38: Chr7:33 . 50–33 . 71 ) . This region contains several conserved regulatory elements , identified from the Regulatory Element Database [22] , one non-coding RNA ( LOC105375227 ) , and several enhancer sites , of which at least two are annotated to enhance BMPER expression according to the human EnhancerAtlas [23] . To investigate the origin of the deletion , we analysed the frequency of the deletion over the last decades . The first born animals within our genotyped set are from February 2006 , allowing the tracking of the frequency of the deletion over the past decade . The number of genotyped animals was lower in the period 2006–2010 . However , we genotyped over 320 animals in the ( live ) population from 2008 onwards , providing reliable frequency estimates ( S8 Table ) . The SSC18 haplotype carrier frequency was high ( >15% ) over the period 2006–2010 ( maximum 20% in 2008 ) and then decreased to a relative stable ~10% carrier frequency from 2012 onwards ( Fig 3 ) . The Large White population under study has been created out of the consolidation of a number of Dutch breeding organizations around the turn of the last century [24] . During the consolidation phase , which resulted in merging of populations and phasing out of other populations , sperm of breeding boars was deposited at the Dutch Centre for Genetic Resources ( CGN ) . The current Large White pure line descends from two different populations , the StamBoek-Z and the Dumeco-W line [24] . Both breeds were merged around 2003 to form the current Large White breeding line . From the 11 StamBoek-Z boars available at CGN , none were carrier of the deletion . However , from the 56 genotyped Dumeco-W boars available at CGN , five were carrier for the deletion haplotype ( 8 . 9% ) . These boars were born in 2000 and 2001 ( S9 Table ) , showing that the deletion derives from this ancestral line and has been maintained in the Large White population for the past eighteen years ( ~ 15 generations ) . We examined whether the current carrier frequency is purely the result of genetic drift , or whether carriers exhibit selective advantage for important traits within the breeding program . We first simulated genetic drift for a lethal recessive allele in the current Large White population ( S7 Fig ) . The results show that lethal alleles can reach allele frequencies up to about 10% by drift alone ( although extremely rare ) , at which the lack of homozygotes is preventing further increase . Next , we tested whether deletion carriers exhibit heterozygote advantage , by performing an association study for both carrier and non-carrier animals using deregressed estimated breeding values ( DEBVs ) for 16 production traits available from the Topigs-Norsvin breeding program ( Table 4 ) . The carriers grow faster ( TGR and LGR ) , have smaller loin depth ( LDE ) , produce litters that are lighter ( LBW ) , show higher mortality in their litters ( LMO ) and have a higher feed intake ( DFI ) when compared to the non-carriers . Selection on growth has not significantly changed in the last decade , and there is consistent increase in genetic progress for growth in this time period ( S8 Fig ) . To further support a balancing scenario , we evaluated the difference in the total selection index ( TSI ) between the carrier and non-carrier group for all animals born in 2017 . Animals are ranked based on this selection index to select the top animals to produce the next generation . We observe a 2 . 7% higher TSI ( on average ) for carriers compared to non-carriers ( S11 Table ) , caused by the positive effect on growth , that outweighs the negative effect on other traits . Next , we simulated the long term effect on the SSC18 carrier frequency based on the current heterozygous advantage and frequency ( Fig 4 , S9 Fig ) . We observe a decrease in carrier frequency in the first generations due to the loss of homozygotes , which outweighs the heterozygous advantage perceived in the selection index . However , at approximately 6% carrier frequency , the heterozygous advantage compensated for the loss of homozygous offspring , reaching a trade-off at this point . Moreover , carriers show 12 . 4% higher breeding values for growth compared to non-carriers ( S11 Table ) , and we show that the carrier frequency can rise up to 22% if selection would be exclusively on growth ( S10 Fig ) . Together these results support a balancing selection scenario showing heterozygote advantage for growth rate ( Fig 5 ) , an important selection trait in the pig breeding industry . In this study we report a 212 kb deletion with antagonistic effects on fertility and growth . We show that homozygotes for the deletion die around mid- to late-gestation , becoming mummified . Compared to other lethal alleles identified in this population , the deletion seems to be maintained at moderate frequency ( 10 . 8% ) in the population . This moderate carrier frequency is likely not a result of random drift effects , as heterozygotes for the deletion-haplotype show , despite a lower birth weight , increased growth rate , and feed intake , important traits in the breeding goal . The balancing scenario observed , most likely , is a consequence of pleiotropic effects of the deletion on two different genes affecting fertility ( BMPER ) and growth ( BBS9 ) . The large amount of genotype data accumulating in modern breeding schemes applying genomic selection in combination with the large amount of phenotypic data deliver a powerful tool to monitor and control deleterious alleles much more efficiently . Samples collected for DNA extraction were only used for routine diagnostic purpose of the breeding programs , and not specifically for the purpose of this project . Therefore , approval of an ethics committee was not mandatory . Sample collection and data recording were conducted strictly according to the Dutch law on animal protection and welfare ( Gezondheids- en welzijnswet voor dieren ) . The dataset consists of 23 , 722 purebred Large White animals . The animals were genotyped on the Illumina GeneSeek custom 50K SNP chip ( Lincoln , NE , USA ) . Animals with a frequency of missing genotypes > 0 . 20 were removed . We discard markers that did not meet following filtering criteria: A minimum call rate of 0 . 85 , a minor allele frequency > 0 . 01 , and a Hardy-Weinberg proportions exact test p-value below P < 10−6 . Moreover , markers with unknown location on the Sscrofa11 . 1 genome build [38] were discarded , leaving 42 , 288 markers after filtering . All steps were performed in Plink v1 . 90b3 . 30 [39] . We performed haplotype phasing and imputation of missing sites in Beagle4 . 1 with parameter for effective population size set to 195 , other settings were default [40] . Reference and test phased VCF files were merged using bcftools 1 . 3–27-gf31e888 [41] . We tested the SS18 haplotype for the expected number of homozygotes using both parents haplotype information ( sire , and dam ) with the formula described in Fritz et al . , 2013 [42] . An exact binomial test was applied to test the number of observed homozygotes with the number of expected homozygotes . The haplotype was considered significantly depleted if P < 5 × 10−3 . The difference in Mendelian ratios for CxC compared to CxNC matings was tested using a Chi-Square test . To genotype animals directly for the SSC18 deletion , we first calculated LRR normalized signal intensities using PennCNV analysis software [43] . We built a classifier with 5 features: the LRR signal intensities for the four overlapping markers within the deletion ( WU_10 . 2_18_43630319 , WU_10 . 2_18_43773633 , WU_10 . 2_18_43778188 , WU_10 . 2_18_43803484 ) , and the average LRR signal intensity over these four markers . Next , we applied logistic regression to distinguish carrier from non-carrier animals using the sci-kit learn Python library [44] ( S6 Fig ) . We examined the SSC18 haplotype for records on TNB , NSB , and MUM listed for all C x C , and C x NC matings identified in the phenotypic records , the order of C x NC matings does not reflect the sex of the parent animal and is both carrier boar and carrier sow combined . We used a Welch’s t-test to assess whether the phenotypes from the C x C matings differ significantly from C x NC matings . A p-value < 0 . 05 was considered significant . The dataset consists of 73 whole genome sequenced Large White individuals with a total volume of 1 . 77 Tbp ( tera base pairs ) from 15 . 539 billion paired-end reads , ranging from 100–150 bp in length ( S5 Table ) . The data was sequenced on Illumina Hiseq 2000 . We used sickle software for quality trimming of raw reads . Next we aligned the sequences to the Sscrofa11 . 1 genome build [38] using BWA-MEM version 0 . 7 . 15 [45] with an average mappability of 96 . 11% and a sample coverage ranging from 6 . 6–22 . 7X ( 10X average ) . Samtools dedup function was used to remove PCR duplicates [41] . GATK IndelRealigner was used to perform local realignments around indels [46] . Variant calling was performed with Freebayes v1 . 1 . 0 with following settings:—min-base-quality 10—min-alternate-fraction 0 . 2—haplotype-length 0—min-alternate-count 2 [47] . Variants with phred quality score < 20 , and within 3 bp of an indel were discarded [41] . Variants were annotated using the Ensembl variant effect predictor ( VEP , release 90 ) [17] . The impact of missense variants was predicted using SIFT [48] . The sequenced population was phased using Beagle4 . 1 [40] . Analysis on structural variation ( SV ) was performed using Lumpy with default settings [49] , taking the aligned BAM files as input . Coverage information was calculated for predicted SV events using samtools depth [41] , and added to the VCF format tag using PyVCF . Alignments and SV events were visualized using the JBrowse genome viewer version 1 . 12 . 1 [50] . We analyzed RNA-seq data on eight different tissues in one SSC18 carrier animal ( sample: PigWUR166 ) . In addition , we analyzed two other pigs from Duroc , and Pietrain genetic background on five different tissues . RNA-seq reads were aligned to the Sscrofa11 . 1 genome build using STAR 2 . 5 . 3a [51] , generation of transcripts and gene expression levels were achieved with Cufflinks v2 . 2 . 1 [52] . We applied the following steps to examine allele specific expression: First , samtools [41] was used to extract uniquely mapped reads from the BAM alignment files . Next , WASP [53] was used to reduce the mapping ( reference sequence ) bias . Then , GATKASEreadcounter [46] was used to obtain read counts for reference and alternative alleles at each SNP position . Lastly , a two-sided binomial test with p = 0 . 5 ( assuming no bias ) and Benjamini-Hochberg false discovery rate ( FDR ) correction were performed in R v . 3 . 4 at each variant position using the Stats package . The variants with FDR adjusted p-value < 0 . 05 were considered as allele specific expression variants . Visual examination of the alignments and transcripts was performed in JBrowse [50] . RNA was extracted from frozen whole blood using the Nucleospin RNA blood kit from Machery Nagel . cDNA was synthesized using Superscript II Reverse Transcriptase ( Invitrogen ) with RNA input ~100ng . RT-qPCR was started with: 3 . 75ul cDNA ( 1:1 ) , 1 . 25ul primer forward ( 2uM ) , 1 . 25ul primer reverse ( 2uM ) , and 6 . 25ul MESA blue mix ( Eurogentec ) . RT-qPCR was then performed with a QuantStudio 5 system using the comparative Ct ( delta delta Ct ) method with GAPDH as housekeeping gene for normalization . Reaction was performed as follows: Data was analysed with the Quantstudio Design & Analysis Software v . 1 . 4 . 3 . All primers and results are listed in S12 and S13 Tables . We downloaded three H3K27Ac , and three H3K4me3 libraries ( ArrayExpress accession number: E-MTAB-2633 ) from liver tissue from three male pig samples described by Villar et al . 2015 [21] . Data was aligned using BWA-mem [45] and visualized in JBrowse [50] . We analyzed the frequency of the SSC18 haplotype per half-year starting from 01-jul-2006 . We assessed the frequency based on total population ( live animals ) on each time point by looking at the proportion of carrier and non-carrier animals in the population . The number of animals per time point are provided in S8 Table . We used a One-Way Repeated Measures ANOVA to test whether the frequency differs over time . In this study , we evaluated 16 traits used in the Large White breeding program . Deregressed estimated breeding values ( DEBV ) were used as a response variable for each trait under study . The estimated breeding value ( EBV ) was separately deregressed for each trait using the methodology described by Garrick et al [54] . The EBV of each animal was obtained from the routine genetic evaluation by Topigs Norsvin using an animal model . The reliabilities per animal for the purpose of deregression were extracted from the genetic evaluation based on the methodology of Tier & Meyer [55] . The heritabilities used for the deregression were also extracted from the routine genetic evaluation . Parent average effects were also removed as part of the deregression process to obtain more accurate estimates of the genetic merit of each individual . Finally , weighting factors based on the estimated reliability of the DEBV were also estimated according to Garrick et al [54] using a value of 0 . 5 for the scalar c . To ensure the quality of the DEBV , only animals with a w higher than not equal to zero and a reliability of the DEBV greater than 0 . 20 were used in the association analyses . The reliability of the DEBV was obtained according to Garrick et al [54] . Association analyses were performed using the software ASREML [56] applying the following model: DEBVijω=μ+Ri+aj+eij , where DEBVij is the observed DEBV for the animal j , w is weighting factor for the residual , μ is the overall DEBV mean of the population , Ri is the carrier status of the lethal allele i , aj is the additive genetic effect estimated using a pedigree-based relationship matrix , and eij the residual error . We simulated changes in allele frequency across multiple populations under the model of Wright [57] . Each allele is associated with a fitness , and we set the fitness to zero for homozygotes ( for lethal recessive allele ) and fitness to 1 ( no negative fitness effect ) for carriers and non-carriers . We assume constant population size through time , and matings are simulated randomly at each generation . Changes in allele frequencies are plotted using the R package driftR ( https://github . com/cjbattey/driftR ) . Within each generation the top 5% of boars , and top 25% of gilts ( based on the TSI selection index value ) are used to produce the next generation . We first calculated the average TSI , and estimated breeding values for six important traits in the breeding line ( S11 Table ) . Next , we used the ratio of carrier TSI over non-carrier TSI to estimate the selective advantage in the breeding program . Next , we simulated the long-term allele frequency change ( assuming random matings ) based on the selective advantage , and the loss of homozygous animals using the Hardy-Weinberg principle . Similar analysis was performed using the selective advantage on growth exclusively .	We report a large deletion within the BBS9 gene that induces late fetal mortality in homozygous affected animals in a commercial pig population . This late fetal mortality causes the fetus to become encapsulated and desiccated during the remaining time of the pregnancy , a process called mummification . The unusually high carrier frequency for this lethal deletion ( 10 . 8% ) likely results from its strong positive association with growth rate in heterozygous individuals , an important selection trait in the pig breeding industry . Interestingly , we show that the positive effect on growth is induced by a heterozygous loss-of-function of the BBS9 gene , associated with obesity in human and mouse . However , late fetal mortality is induced by insufficient expression of the BMPER gene located directly downstream of the deletion which affects its regulatory elements required for gene expression . Together , our study shows an unique example of allelic pleiotropy in which one allele ( deletion ) is responsible for both increased growth and late fetal mortality by affecting two different genes .	[ "Abstract", "Introduction", "Results", "Discussion", "Material", "and", "methods" ]	[ "body", "weight", "medicine", "and", "health", "sciences", "engineering", "and", "technology", "vertebrates", "alleles", "animals", "genetic", "mapping", "mammals", "telecommunications", "physiological", "parameters", "birth", "weight", "swine", "gene", "expression", "genetic", "loci", "haplotypes", "eukaryota", "natural", "selection", "heredity", "physiology", "genetics", "biology", "and", "life", "sciences", "carrier", "frequencies", "evolutionary", "biology", "amniotes", "evolutionary", "processes", "organisms" ]	2018	Balancing selection on a recessive lethal deletion with pleiotropic effects on two neighboring genes in the porcine genome
To date , the genome-wide association study ( GWAS ) is the primary tool to identify genetic variants that cause phenotypic variation . As GWAS analyses are generally univariate in nature , multivariate phenotypic information is usually reduced to a single composite score . This practice often results in loss of statistical power to detect causal variants . Multivariate genotype–phenotype methods do exist but attain maximal power only in special circumstances . Here , we present a new multivariate method that we refer to as TATES ( Trait-based Association Test that uses Extended Simes procedure ) , inspired by the GATES procedure proposed by Li et al ( 2011 ) . For each component of a multivariate trait , TATES combines p-values obtained in standard univariate GWAS to acquire one trait-based p-value , while correcting for correlations between components . Extensive simulations , probing a wide variety of genotype–phenotype models , show that TATES's false positive rate is correct , and that TATES's statistical power to detect causal variants explaining 0 . 5% of the variance can be 2 . 5–9 times higher than the power of univariate tests based on composite scores and 1 . 5–2 times higher than the power of the standard MANOVA . Unlike other multivariate methods , TATES detects both genetic variants that are common to multiple phenotypes and genetic variants that are specific to a single phenotype , i . e . TATES provides a more complete view of the genetic architecture of complex traits . As the actual causal genotype–phenotype model is usually unknown and probably phenotypically and genetically complex , TATES , available as an open source program , constitutes a powerful new multivariate strategy that allows researchers to identify novel causal variants , while the complexity of traits is no longer a limiting factor . Genome-wide association studies ( GWAS ) are currently the primary tool to identify genetic variants ( GVs ) underlying phenotypic variation . GWAS are generally univariate in nature , i . e . , focus on a single phenotype . This means that researchers , prior to analyses , often reduce available , originally multivariate , phenotypic information ( e . g . , information on multiple questions from a diagnostic interview or questionnaire , or multiple items in a test ) to a single phenotypic composite score , such as a continuous sum score or binary case-control status ( the latter is often based on the number of endorsed symptoms , i . e . , effectively a dichotomized sum score ) . Such univariate conceptualisations are consistent with the practical and diagnostic definitions employed in psychology and medicine of traits like depression , cognition , Type I diabetes , and asthma . However , whether they represent informative entities with respect to biological aetiology is questionable [1] . Many acknowledge the possible genetic heterogeneity of psychological and medical traits [2]–[3] . This heterogeneity implies that distinct GVs may give rise to the same univariate trait score , and that the same GV may have different behavioral manifestations , depending on genetic background and environmental exposure . It also implies that phenotypes ( e . g . , symptoms , items , subtests ) may be affected by different GVs . To appreciate this , consider diagnostic indicators of asthma , like spirometric measures , serum total IgE , and fractional exhaled nitric oxide . These measures are phenotypically correlated and all associated with asthma diagnosis , yet their genetic architecture may differ . When GWAS is subsequently conducted on asthma case-control status , however , both the plausible phenotypic and genetic heterogeneity of the trait is discarded . Likewise , depression symptoms like worrying , insomnia , and feeling lonely or irritable , and metabolic syndrome related measures like waist-to-hip ratio , fasting glucose levels , triglycerides , and high-density lipoprotein , are phenotypically correlated yet need not be subject to the same GVs . That is , while the conceptual multidimensionality of traits is often acknowledged in the phenotypic instruments – e . g . by including measures of multiple symptoms for disease traits , or multiple subtests to cover distinguishable dimensions of complex traits ( e . g . , spatial and verbal ability , memory , and general knowledge in cognition ) - this phenotypic resolution is lost when the multivariate phenotypic information is subsequently reduced to a univariate composite score . As we often do not know how a causal GV impinges on a phenotype , determining the most informative operationalisation of a trait for gene-finding purposes poses a challenge . Multiple studies [4]–[7] have shown that phenotypic data reduction , such as case-control status phenotypes or sum scores calculated across all distinguishable phenotypes , results in a considerable loss of statistical power to detect GVs in all but the special circumstance that 1 ) a single phenotypic dimension underlies the variance-covariance structure of the multivariate phenotypes ( i . e . , single common factor model ) , and 2 ) the GV directly affects this dimension ( schematic representation Figure 1a ) . In this ideal unidimensional model , the underlying phenotypic dimension mediates the relationship between the GV and the multivariate phenotypes , and the univariate sum score is a good approximation of this dimension . However , many other genotype-phenotype models are plausible . For instance , the model could be multi-dimensional rather than unidimensional ( Figure 1b–1c ) , and the GV effect could be specific to one of the phenotypes , rather than on the latent dimension ( Figure 1d–1e ) . Recently , the field of psychology has witnessed a shift towards network models , in which relations between individual phenotypes are not believed to result from shared causal latent factors , but rather originate in direct causal influences between phenotypes over time [8]–[10] . For instance , from a network perspective , symptoms like worrying , sleeplessness and agitation are not viewed as manifestations of the latent dimension depression , but as directly and causally related: worrying interferes with sleep , and lack of sleep causes agitation . In such network models , which obviate the need to invoke latent dimensions , each phenotype could be affected by different GVs ( Figure 1f ) . In all these alternative genotype-phenotype models , univariate conceptualisations like sum scores and case-control status result in substantial loss of power to detect underlying GVs . One way to avoid the potential loss of power associated with univariate conceptualisations of complex heterogeneous traits , is to adopt a multivariate method , which accommodates the originally multivariate nature of the phenotypic measure . Exploratory multivariate strategies , developed in GWAS context , include MultiPhen [11] , and canonical correlation analysis [12] , which is included in the GWAS software PLINK [13] ( as canonical correlation analysis is identical to MANOVA with one GV treated as additive codominant ( i . e . , covariate ) , we use the term MANOVA here ) . MultiPhen uses ordinal regression to regress 0/1/2-coded GVs on a collection of phenotypes of any measurement nature ( i . e . , continuous , dichotomous , ordinal ) , and applies one omnibus test to test whether all regression weights in the model are together significantly different from zero . MultiPhen has been shown to outperform MANOVA when minor allele frequency ( MAF ) is low and the phenotypes are case-control status or non-normally distributed continuous variables [11] . Under most circumstances , however , MultiPhen and MANOVA yield very similar results in terms of power to detect causal GVs . A drawback of these multivariate methods is that their power depends on the specific configuration of phenotypic correlations and on the location of the GV effect ( e . g . , on the latent dimension , or specific to one of many correlated phenotypes ) . For instance , when the ideal model ( Figure 1a ) holds , MANOVA is decidedly less powerful than univariate analyses based on sum scores . MANOVA , however , easily outperforms the sum score approach when the GV affects only one of multiple strongly correlated variables ( e . g . , Figure 1d–1f ) [4]–[5] , [14] . As prior knowledge about the exact location of the GV effect in a multivariate system is usually lacking , a computationally efficient multivariate procedure that performs well in many different circumstances is required to increase the success of future GWAS . Here , we introduce a new multivariate technique called TATES: Trait-based Association Test that uses Extended Simes procedure . TATES is based on the GATES procedure [15] , which was developed to combine p-values of individual SNPs located within the same gene into one gene-based p-value PG ( where the gene is considered a more attractive unit of analysis for association studies than the SNP because genes are the functional units in the genome ) . Similarly , for individual phenotypes characterizing a trait ( e . g . , items or symptoms ) , TATES combines the p-values obtained in standard univariate GWAS to arrive at a global trait-based p-value PT , while correcting for the observed correlational structure between the phenotypes . Here we show that TATES has correct false positive ( type-I error ) rate , and that TATES picks up both phenotype-specific genetic effects as well as genetic effects that are common to multiple correlated phenotypes . Through extensive simulations , probing a wide variety of genotype-phenotype models , we demonstrate under which circumstances TATES outperforms analyses based on sum scores and MANOVA/MultiPhen with respect to the statistical power to detect causal GVs . The TATES method is described in detail in the Materials and Methods section . Briefly , TATES requires the m×n p-values of the regression of m phenotypic variables on n GVs , and the m×m correlation matrix of the phenotypes . The regression of the phenotypes on the GVs can be conducted in standard software packages like PLINK , Mach2dat/qtl , SNPtest , and Gen/ProbABEL [13] , [16]–[20] , which are fast , facilitate quality control , and can correct for population stratification . For samples that include related individuals , analyses could be conducted using PLINK ( where the –mperm option should not vary over the m phenotypes to assure that the p-values used in TATES have similar accuracy ) , ABLE , PBAT or Merlin-offline [13] , [16]–[17] , [21]–[22] . For each GV , TATES sorts the m p-values ascendingly . To derive from these m p-values one trait-based p-value PT for each of the n GVs , TATES takes into account that the m phenotypes , and thus the m p-values , are correlated . In an iterative procedure , TATES weighs the jth p-value in the 1 to m sorted p-values with me/mej , where me is the effective number of independent p-values among all m variables , and mej the effective number of p-values among the top j p-values . The weight me is a function of m , and the sum of those eigenvalues larger than 1 of the m×m correlation matrix of the p-values . Similarly , mej is a function of j and the sum of the eigenvalues larger than 1 based on the j×j correlation matrix of the top j p-values . The correlation matrix of the m p-values is approximated from the observed correlation matrix between the m phenotypes using a 6th order polynomial ( coefficient of determination R2 = . 992 , see Materials and Methods and Figure S1 ) . For each of the n GVs , the trait-based TATES p-value PT equals the smallest weighted p-value , with the null-hypothesis that none of the phenotypes is associated with the GV , and the alternative hypothesis that at least one of the phenotypes is associated with the GV . The TATES procedure is implemented in a Fortran 77 program and an R script , both of which are freely available from the website ( http://ctglab . nl/software ) . The Fortran program takes less than 1 minute to calculate the TATES trait-based p-values PT for 12 phenotypes and 437 , 598 GVs on an ordinary desktop computer with Intel ( R ) Core ( TM ) 2 Duo CPU 2 . 99 GHz , RAM 2 . 94 GB , and 32-bit Windows XP Professional Version 2002 . To study the false positive rate and the power to detect GVs using TATES , we simulated genotype-phenotype data for 2000 subjects and 20 phenotypes ( standard normally distributed unless stated otherwise ) according to various scenarios that are illustrated in Figure 1a–1f . Specifically , the phenotypic correlation structure was due to one underlying common factor ( or dimension , Figure 1a , 1e ) , multiple underlying common factors ( Figure 1b–1d ) , or to a network model , in which correlations between phenotypes are due to direct , mutual relations between the components ( Figure 1f ) . Within these phenotypic correlational settings , the GV affects multiple phenotypes via the common factor ( Figure 1a , b , c ) , or affects a single component directly ( Figure 1d–1f ) . For each scenario , we simulated GVs ( MAF of . 50 ) with effect sizes ranging from 0 to 1% explained variance . The false positive rate was also studied given MAF = . 05 and N = 12000 . Simulations are described in detail in the Materials and Methods section . In each scenario , the 20 simulated phenotypes were either a ) summed and the sum score was regressed on the GV , b ) subjected to a 1-factor model to calculate Thompson's factor scores [23] , which were regressed on the GV , c ) subjected to a MANOVA with the GV as covariate ( canonical correlation analysis ) , d ) subjected to MultiPhen ( regressing the GV on all 20 phenotypes in a multivariate ordinal regression model ) , or e ) individually regressed on the GV ( using logistic or ordinal regression where appropriate ) . The last procedure yielded 20 p-values per simulated GV , which were then combined into 1 overall trait-based p-value PT using TATES . In addition , we compared the performance of TATES to that of various other published procedures for combining p-values , limiting our comparison to procedures that , like TATES , do not require permutation , i . e . , Fisher's combination test , Lancaster's weighted Fisher test , the Z-transform test , and the original Simes procedure ( see Text S1 ) . All data simulations and subsequent analyses were repeated 2000 times . We counted the number of times that the GV effect was detected given α = . 05 . The results of all simulated scenarios are presented in detail in Tables S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , S12 . The false positive rates of TATES , the sum score and factor scores procedures , MANOVA , and MultiPhen were correct given our simulation settings with both MAF = 50% and MAF = 5% , while the original Simes procedure proved slightly conservative , if the phenotypes were highly correlated . ( Note that the false positive rate of MANOVA is known to be inaccurate if the GV has low MAF ( . 5 or 5% ) and the phenotypic data are dichotomous or non-normally distributed [11] ) . In contrast , the false positive rate of the Fisher combination test , Lancaster's weighted Fisher test , and the Z-transform test , which do not account for correlations between the 20 phenotypes , was often highly inflated ( up to 20% , depending on the magnitude of the phenotypic correlations ) . Power results for these methods are therefore not discussed here ( but see Tables S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , S12 ) . Since power results of the factor scores , MultiPhen , and the original Simes procedure were quite similar to those of the sum scores , MANOVA , and TATES , respectively , these are not discussed here ( but see Tables S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , S12 ) . Figure 1g illustrates the power results of 12 selected simulation scenarios for the sum score procedure , MANOVA and TATES , given a GV explaining . 5% of the phenotypic variance . As expected [4]–[6] , the sum score procedure has excellent power to detect the GV , if the phenotypic data are generated according to a 1-common factor model , and the GV effect is on this factor ( Figure 1g: A1–A3 ) . However , if either the location of the GV effect or the data-generating process is different , the power of the univariate sum score procedure drops to levels often < . 10 ( Figure 1g: B1 , C1 , D1 , E1–3 , F1–3 ) . In 9 out of the 12 scenarios we considered , the power of TATES was 2 . 5 to 9 times higher than the power of the sum score procedure . As expected [4]–[5] , [14] , the power of MANOVA is especially high if the GV effect is specific to only one of many highly correlated phenotypes ( Figure 1g: D1 , E1 ) . The power of MANOVA drops if the phenotypic correlations are lower , or if multiple phenotypes are subject to the GV effect . In contrast , TATES is only slightly less powerful than the sum score procedure if the phenotypes correlate substantially ( Figure 1g: A1 , A2 ) , and clearly more powerful than MANOVA in this condition . TATES outperforms both other procedures if the GV affects multiple , but not all correlated phenotypes ( power TATES is 1 . 5–2 times higher , Figure 1g: B1 , C1 ) , and is approximately as good as , or better than , MANOVA , if the GV effect is specific to one of multiple phenotypes that correlate . 30 or lower ( Figure 1g: E2 , E3 , F1–3 ) . In 7 of our 12 scenarios , the power of TATES was 1 . 5 to 2 times higher than the power of MANOVA . As the original Simes procedure does not take into account the correlations among the p values ( originating in the phenotypic correlations ) , TATES is expected to increasingly outperform Simes as the phenotypic correlations increase . Given low to modest phenotypic correlations , the gain in power acquired with TATES varies from low ( 1% ) to modest ( 9% ) ( the latter observed in a 4-factor model with a phenotype-specific GV effect; Table S12 ) . Additional simulations ( Tables S13 , S14 , S15 , S16 , S17 , S18 ) , however , show that , as phenotypic intercorrelations increase in magnitude ( . 75 , . 85 , . 95 ) , the power of TATES can be as much as 10%–19% higher than the power of the Simes procedure , with TATES especially being more powerful when the GV effect is specific to one of multiple correlated phenotype . As TATES is comparable to Simes in computational effort , phenotypes within a trait are almost invariably correlated , and the location of the GV effect is generally unknown ( i . e . , could be phenotype-specific ) , one is well-advised to adopt TATES . Finally , we studied the effect of 10% missingness completely at random ( MCAR ) or 10% blockwise missingness on the power to detect GVs in three different genotype-phenotype models ( see Materials and Methods for details and Tables S19 , S20 , S21 , S22 , S23 , S24 , S25 , S26 ) . Power was hardly affected in 1-factor models with the GV effect on the factor . However , if the GV effect was specific to one of the phenotypes ( either in factor models or network models ) , the power of MANOVA usually showed a 5–6% larger drop in power compared to Simes and TATES . Only when the GV effect was specific to a phenotype showing blockwise missingness was the drop in power of Simes and TATES similar to , or even slightly higher ( 2–3% ) than , the power drop observed for MANOVA . We have presented TATES , a new , computationally feasible multivariate method to test genotype-phenotype relations . The false positive rate of TATES is correct for varying MAF , even if the multiple phenotypes are substantially correlated . Through simulations , we showed that TATES outperforms standard univariate analyses , unless the data-generating process is a unidimensional factor model and the GV affects the factor . TATES is only outperformed by MANOVA in the particular condition that the GV affects only one of multiple strongly correlated phenotypes . Multivariate genotype-phenotype analyses are important for several reasons . First , most complex traits , such as cognitive ability , personality , problem behavior in humans [24]–[26] , and anxiety in mice [27] , are multi-dimensional , i . e . , multiple common factors are required to describe the variance-covariance structure . Given this multidimensionality , multivariate genotype-phenotype analyses are indicated , as standard univariate analyses cannot accommodate genetic heterogeneity of subdimensions . Second , phenotypically distinguishable subdimensions need not correspond simply to genetic dimensions , and the information to parse a trait into genetically informative subdimensions is usually lacking . Consequently , researchers often focus on those GVs that are common to all subdimensions by studying a single , “general” composite measure . A simple , but deficient alternative is to conduct a series of independent univariate association studies without correcting for the dependency between the results caused by the correlations between the phenotypes . TATES offers a simply method to correct for this relatedness , while identifying GVs that are common to multiple phenotypes and GVs that are phenotype specific . As such TATES provides a more complete view of the genetic architecture of complex traits . Third , it is often unclear which phenotype ( s ) or combination of phenotypes will maximize the probability of unraveling the genetic architecture of a complex trait . For example , in studying a complex trait like schizophrenia , quantitative cognitive traits that are commonly affected in schizophrenia patients ( e . g . , attention , mental flexibility , memory , sensorimotor processing ) may be better candidates for unraveling the genetic architecture of schizophrenia than schizophrenia affection status [28] . Multivariate techniques obviate the need to focus on one phenotype , and help to chart both genetic overlap and genetic uniqueness of related traits . Such information on genetic similarity and dissimilarity of phenotypes may ultimately help uncover the actual disease mechanism . TATES allows researchers to test their genetic associations efficiently using standard GWAS software . In addition , TATES' reliance on p-value information assures that phenotypes of different measurement levels ( e . g . , dichotomous , ordinal , continuous ) can easily be analyzed simultaneously , and that contrasting effects ( i . e . , GVs affect some phenotypes positively , some negatively ) do not influence the power of the method . Note that the plausibility of contrasting genetic effects does not only depend on the magnitude of the phenotypic correlations and the effect size of the GV ( i . e . , the correlation matrix between the phenotypes and the GV should remain positive definite ) , but also on the underlying genotype-phenotype model . For instance , if the true genotype-phenotype model is a 1-factor model with the GV effect on the factor , the direction of the effect of the GV must be identical for all phenotypes ( assuming that all phenotypes are coded such that higher scores imply higher trait levels ) . Yet , if the true genotype-phenotype model is a network model , contrasting GV effects are unproblematic from a statistical point of view . Whether contrasting effects are plausible from a biological perspective depends on the actual functional role of the GV . For instance , symptoms like blunted affect and agitated mood can both be positive indicators of depression on a population level , but their biochemical basis may be antagonistic , making contrasting GV effects for these symptoms both statistically and biologically possible . TATES cannot be used directly to test specific hypotheses concerning the underlying genotype-phenotype model . However , as TATES outputs the p-values from the original univariate GWAS analyses along with TATES' trait-based p-values , further inspection of the pattern of significant univariate tests that drive the significant TATES p-value can be informative . For instance , if a significant TATES p-value is driven by an association with only one of the multiple phenotypes , then the true genotype-phenotype model is unlikely to be a 1-factor model with the GV effect on the factor . The more these phenotype-specific GV effects are observed , the stronger the indication that the trait under study is genetically heterogeneous . This , again , implies that multivariate approaches , in which the heterogeneity is accommodated , are more likely to reveal the genetic architecture of that trait than the standard approach based on univariate composite scores . Furthermore , if one aspires to actually test specific hypotheses concerning the underlying genotype-phenotype model , TATES can be used as a filter to reduce the number of SNPs to a computationally manageable set . The exact location and role of the selected SNPs may then be studied in detail in appropriate multivariate models [4]–[5] . Finally , TATES facilitates the study of the genetic overlap between phenotypes in different domains , for example medical and psychiatric disorders that show high comorbidity , and yet are generally studied separately . Studying behavioural profiles [29] rather than single phenotypes , and phenotypes spanning multiple levels of organisation ( e . g . , behaviour , morphology , physiology ) , advances analysis of the “phenome” ( the phenotype as a whole , on an organism-wide scale ) [30] . Here , TATES is a useful tool , as it is hypothesis- and model-free , and deals with the high phenotypic dimensionality by combining the univariate analyses while correcting for the relatedness between phenomic dimensions . Furthermore , in a highly dimensional phenotypic context , the fact that one does not need to know the underlying data-generating model , or the mechanism causing comorbidity/association between the individual phenotypes in the analysis , is an attractive feature of TATES . To summarize , TATES is an efficient multivariate method for combining p-values across different , correlated phenotypes in genotype-phenotype analyses . In the context of gene-finding studies , TATES allows researchers to test genetic associations without a priori data reduction or commitment to one phenotypic or genetic model . As the actual phenotypic and genetic architecture of traits is usually unknown and probably complex , an exploratory multivariate procedure like TATES provides a viable and , as simulations show , powerful new strategy . Suppose m phenotypes are measured as indicators of one trait , e . g . , individual symptoms within a disorder , items within a test , or multiple measures of one trait using different instruments ( e . g . , open-field test , a light-dark box , and an elevated plus maze to measure anxiety in mice ) . Rather than combining these m phenotypes into one general phenotype , we test the association between all m phenotypes and all n genotyped genetic variants ( GVs ) using a statistically appropriate method ( e . g . , linear or logistic regression ) . Let p ( 1 ) …p ( m ) be the ascending p-values of the m phenotypes for a given GV . TATES combines within each GV the m phenotype-specific p-values to obtain one overall trait-based p-value PT as follows: ( 1 ) where me denotes the effective number of independent p-values of all m phenotypes for a given GV , and mej the effective number of p-values among the top j p-values , where j runs from 1 to m , and pj denotes the jth p-value in the list of ordered p-values . PT is thus the smallest weighted p-value , associated with the null hypothesis that none of the phenotypes is associated with the GV , and the alternative hypothesis that at least one of the phenotypes is associated with the GV . Following Li et al [15] , we obtain an estimate of the effective number of p-values mej through a correction based on eigenvalue decomposition of the m×m correlation matrix ρ between the p-values associated with the m phenotypes . The effective number of p-values mej for the top j p-values is calculated as: ( 2 ) where j is the number of top j p-values , λi denotes the ith eigenvalue , and I ( λi−1 ) is an indicator function taking on value 0 if λi≤1 and 1 if λi>1 . That is , the effective number of p-values mej is calculated as the observed number of p-values j minus the sum of the difference between the eigenvalues λi and 1 for those eigenvalues λi>1 . If the j phenotypes are all uncorrelated , then all j eigenvalues equal 1 , and mej = j−0 = j . In contrast , if the j phenotypes are perfectly correlated , then the first eigenvalue equals j , and the other eigenvalues equal 0 , rendering mej = j− ( j−1 ) = 1 ( i . e . , j perfectly correlated phenotypes represent only 1 unique unit of information ) . In practice , phenotypes show intercorrelations of variable magnitude ( but not 0 or 1 ) , so the effective number of p-values mej will usually be smaller than j , but greater than 1 . Note that me is equal to mej for the case that j = m , i . e . , when the selection of top phenotypes covers all phenotypes . The m×m correlation matrix ρ between the p-values is not observed in practice . Following Li et al [15] , we used simulation to show that matrix ρ can be accurately approximated through the m×m correlation matrix r between the phenotypes . We simulated 55 continuous standard normally distributed phenotypes whose intercorrelations ranged between − . 90 and . 90 , and a GV ( MAF = . 5 ) that was simulated to be unrelated to the 55 phenotypes . The association between the GV and all phenotypes was tested , yielding 55 p-values , and this simulation was run 10 , 000 times . We then calculated , across the 10 , 000 simulations , the mean pair-wise correlations between the 55 phenotypes ( i . e . , ( 5555−55 ) /2 = 1485 correlations ) , and the mean pair-wise correlations between the p-values . Regressing the vector of correlations between the p-values on the vector of correlations between the phenotypes , we obtain the 6th order polynomial ρ = −0 . 0008−0 . 0023r+0 . 6226r2+0 . 0149r3+0 . 1095r4−0 . 0219r5+0 . 2179r6 ( coefficient of determination R2 = . 992; see Figure S1 ) , allowing accurate approximation of the correlations between the p-values from the observed correlations between the phenotypes . The thus obtained matrix ρ is subjected to the eigenvalue decomposition in Eq . 2 . To determine the circumstances in which TATES outperforms the original Simes procedure , we conducted six additional simulations . While the original Simes procedure corrects for the observed number of p-values , TATES corrects for the effective number of p-values , by taking the correlations between the p-values into account . The difference in terms of power between Simes and TATES is thus expected to be larger as the correlations between the p-values ( phenotypes ) are stronger ( i . e . , the effective number becomes smaller ) . To illustrate the difference in power between TATES and Simes , we simulated phenotypic data according to 1-factor Rasch models , with factor loadings of . 8660 , . 9220 , or . 9747 , indicating correlations of . 75 , . 85 and . 95 between the phenotypes , respectively . The GV effect was modeled either on the latent factor ( like Figure 1a; Tables S13 , S14 , S15 ) , or directly on one of the 20 phenotypes ( like Figure 1e; Tables S16 , S17 , S18 ) . To study the effect of missingness in the phenotypic data on the power to detect GVs , we conducted eight simulation studies in which we studied two types of missingness in five different genotype-phenotype models . The effect of missingness completely at random ( MCAR ) was studied by simulating data in which each of the 20 simulated phenotypes had 10% missingness distributed randomly across individuals . With 2000 subjects and 20 phenotypes , this results in ∼4000 missing values ( i . e . , 10% of the total of 40000 observations ) . In addition , we studied the effect of blockwise missingness; 400 randomly selected subjects in each simulated file had valid data only for the first 10 of 20 phenotypes ( e . g . , comparable to the situation that data of two samples are combined: in sample 1 ( N = 1600 ) , a full 20-item questionnaire is administered , while in sample 2 ( N = 400 ) , the abbreviated version of 10 items is administered ) . This results again in 4000 missing values , i . e . , the amount of missingness is the same across the two missingness scenarios , but the distribution is different . The effect of these two types of missingness was studied in three genotype-phenotype models: 1 ) 1-factor Rasch model with the GV effect on the factor ( Figure 1a; Tables S19 , S20 ) , 2 ) 1-factor Rasch model with the GV effect specific to one phenotype ( Figure 1e; specific phenotype not showing blockwise missingness; Tables S21 , S22 , or showing blockwise missingness; Table S23 ) , 3 ) network model with the GV effect specific to one phenotype ( Figure 1f; specific phenotype not showing blockwise missingness; Tables S24 , S25 , or showing blockwise missingness; Table S26 ) . In all these models , the 20 phenotypes correlated . 56 ( power results including missingness can thus be compared to power results concerning the same models without missingness presented in Tables S2 , S4 and S7 ) . Note that equal correlations between all phenotypes represents the ideal situation in which all phenotypes are equally reliable , i . e . , the effect of the missingness only depends on the pattern of missingness , not e . g . on the reliability of the individual phenotypes . In subsequent analyses , missingness was handled in two ways . The missing values were either imputed using mean imputation ( i . e . , missing values are imputed with the sample mean of the appropriate phenotype ) . This type of imputation , which was done for MANOVA , sum score , Simes and TATES , is standard in MultiPhen [11] and canonical correlation analysis in Plink [13] . Alternatively , the analyses were based on all available valid data . The sum score was then calculated as a weighted sum ( i . e . , the sum of all available data , divided by the total number of available data ) . For Simes and TATES , the univariate tests were based on all available data , and the p-values , now due to the missingness based on different sample sizes , were combined as usual . ( Whether a correction is required to deal with the fact that the p-values are based on different sample sizes , is open to debate . In theory , the test statistic , and thus the p-value , already take N into account . In practice , however , a procedure that weights for the sample size can be more powerful [34] . We tried one type of weighting for Simes and TATES , in which each p-value was weighted by dfmax/dfj , where dfmax denotes the maximal number of degrees of freedom ( i . e . , sample size ) of the 20 simulated phenotypes , and dfj denotes the number of degrees of freedom for the jth phenotype in the set of 1…20 . This way , the p-value belonging to the largest sample was weighted by dfmax/dfmax = 1 , while the other p-values were weighted by dfmax/dfj and as dfj is always <dfmax the weight is thus >1 , i . e . , p-values derived from small samples were adjusted upwards and are therefore less likely to be the minimal p-value chosen by Simes or TATES . ) MANOVA was not conducted on all available data because in standard MANOVA , cases are excluded listwise , resulting in a very low sample size when missingness is MCAR . In theory , fitting MANOVA on the raw data using Full Information Maximum Likelihood ( FIML ) is possible in software like LISREL , Mx , or Mplus [33] , [35]–[36] , but this is time consuming in a genome-wide context . Here , we chose to stick to the common practice of MultiPhen [11] and Plink [13] , which is mean imputation .	The genome-wide association study ( GWAS ) is the primary tool to identify genetic variants that cause phenotypic variation . As GWAS methods are generally univariate in nature , multivariate phenotypic information is usually reduced to a single composite score , which frequently results in a considerable loss of statistical power to detect causal variants . Multivariate genotype–phenotype methods do exist but attain maximal power only in special circumstances . We present a new multivariate method called TATES ( Trait-based Association Test that uses Extended Simes procedure ) . Extensive simulations show that TATES's false positive rate is correct , and that TATES's statistical power to detect causal variants explaining 0 . 5% of the variance can be 2 . 5–9 times higher than the power of univariate tests of composite scores and 1 . 5–2 times higher than the power of the standard MANOVA . Unlike other multivariate methods , TATES uncovers both genetic variants that are common to multiple phenotypes as well as phenotype specific variants . TATES thus provides a more complete view of the genetic architecture of complex traits and constitutes a powerful new multivariate strategy that allows researchers to identify novel causal variants .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genome-wide", "association", "studies", "social", "and", "behavioral", "sciences", "quantitative", "traits", "biology", "psychometrics", "psychology", "trait", "locus", "phenotypes", "heredity", "genetic", "association", "studies", "genetics", "human", "genetics", "genetics", "and", "genomics", "complex", "traits" ]	2013	TATES: Efficient Multivariate Genotype-Phenotype Analysis for Genome-Wide Association Studies
The pancreaticobiliary ductal system connects the liver and pancreas to the intestine . It is composed of the hepatopancreatic ductal ( HPD ) system as well as the intrahepatic biliary ducts and the intrapancreatic ducts . Despite its physiological importance , the development of the pancreaticobiliary ductal system remains poorly understood . The SRY-related transcription factor SOX9 is expressed in the mammalian pancreaticobiliary ductal system , but the perinatal lethality of Sox9 heterozygous mice makes loss-of-function analyses challenging . We turned to the zebrafish to assess the role of SOX9 in pancreaticobiliary ductal system development . We first show that zebrafish sox9b recapitulates the expression pattern of mouse Sox9 in the pancreaticobiliary ductal system and use a nonsense allele of sox9b , sox9bfh313 , to dissect its function in the morphogenesis of this structure . Strikingly , sox9bfh313 homozygous mutants survive to adulthood and exhibit cholestasis associated with hepatic and pancreatic duct proliferation , cyst formation , and fibrosis . Analysis of sox9bfh313 mutant embryos and larvae reveals that the HPD cells appear to mis-differentiate towards hepatic and/or pancreatic fates , resulting in a dysmorphic structure . The intrahepatic biliary cells are specified but fail to assemble into a functional network . Similarly , intrapancreatic duct formation is severely impaired in sox9bfh313 mutants , while the embryonic endocrine and acinar compartments appear unaffected . The defects in the intrahepatic and intrapancreatic ducts of sox9bfh313 mutants worsen during larval and juvenile stages , prompting the adult phenotype . We further show that Sox9b interacts with Notch signaling to regulate intrahepatic biliary network formation: sox9b expression is positively regulated by Notch signaling , while Sox9b function is required to maintain Notch signaling in the intrahepatic biliary cells . Together , these data reveal key roles for SOX9 in the morphogenesis of the pancreaticobiliary ductal system , and they cast human Sox9 as a candidate gene for pancreaticobiliary duct malformation-related pathologies . The pancreaticobiliary ductal system refers to the complex network of ducts that compose the hepatopancreatic ductal ( HPD ) system as well as the intrapancreatic and intrahepatic ductal networks . The HPD system consists of the extrahepatic duct , cystic duct , gallbladder , common bile duct , and extrapancreatic duct . It connects to the intrahepatic biliary ducts to enable bile flow and storage . The intrapancreatic ducts collect the digestive enzymes secreted by the pancreatic acinar cells . Pancreatic juice and bile flow to the hepatopancreatic ampulla to be released into the intestine and allow digestion and absorption of nutrients [1] . Malformations of the pancreaticobiliary ductal system impair the function of digestive organs and are associated with various congenital conditions whose causes are mostly unknown . In mammals , the transcription factor Sox17 is specifically expressed in a segment of the ventral foregut from which the pancreaticobiliary ductal system derives [2] . This factor has been shown to be a master regulator of pancreaticobiliary ductal system formation by specifying , in conjunction with Hhex and Pdx1 , different lineages of the liver , pancreas and HPD system [2] . The liver is specified as a group of cells that expresses Hhex but not Sox17 or Pdx1 . The intrahepatic biliary network requires several signaling pathways including TGFβ , Notch and Wnt , to differentiate and mature ( for a review , see [3] ) . In particular , Notch signaling has been shown to promote intrahepatic biliary differentiation and tubulogenesis [4]–[9] . Adjacent to the liver domain , cells expressing both Sox17 and Pdx1 delineate a domain that gives rise to the HPD system and pancreas [2] . After lineage segregation , Sox17+/Pdx1− cells give rise to the HPD system under the regulation of downstream factors such as HNF6 , HNF1β and Hhex [2] which themselves have been shown to play important roles in the development of the HPD system [10]–[12] . As for the intrapancreatic ducts , they arise from a subset of Pdx1+/Sox17− cells that also express HNF6 and HNF1β [13] , [14] . These two transcription factors regulate duct tubulogenesis as well as the differentiation of the epithelial cells lining the ducts [14] . The HPD system in zebrafish is morphologically similar to the one in amniotes . As in mammals , the zebrafish HPD system develops from a specific domain within the foregut endoderm that lies between the emerging liver and ventral pancreas [15] . HPD system patterning and differentiation depend on Fgf10 signaling [15] , whereas the specification of the liver and ventral pancreas is regulated by the transcription factors Prox1 [16] , [17] and Pdx1 [18] , respectively . In the liver , hepatocytes and intrahepatic biliary cells derive from bipotential hepatoblasts [19] . Multiple genes encoding Jagged ligands and Notch receptors are expressed in the zebrafish liver during intrahepatic biliary duct formation [20] . Perturbation of Jagged-mediated Notch signaling impairs differentiation and morphogenesis of the intrahepatic biliary cells , whereas constitutive Notch activation induces ectopic bile duct formation [20] , [21] . These studies support an evolutionarily conserved role for Notch signaling in intrahepatic duct development . Regarding the intrapancreatic ducts , live-imaging analyses of the Tg ( Nkx2 . 2a ( 3 . 5kb ) :GFP ) line revealed that they derive from cells in the ventral pancreatic bud that migrate towards the pancreatic islet to initiate the formation of a branched network [18] , [22] . Molecular mechanisms regulating the development of the intrapancreatic ducts remain poorly understood . Studies in mouse have shown that the transcription factor SOX9 is expressed in the intrahepatic and intrapancreatic ducts , as well as in the developing HPD system including the common bile duct , gallbladder and hepatopancreatic ampulla [23]–[25] . Sox9 belongs to the SRY-related box ( SOX ) gene family that encodes transcription factors containing an HMG box DNA-binding domain . In humans , SOX9 is expressed in the fetal brain , liver , testis and skeletal tissue [26] . Haploinsufficiency of SOX9 is associated with campomelic dysplasia ( CD , OMIM #114290 ) , which is characterized by severe skeletal malformations and sex reversal [26] , [27] . More recently , it has been shown that SOX9 is also expressed in the early human fetal pancreas and analysis of CD individuals have revealed pancreatic defects including islet hypoplasia and reduction of hormone expression [28] . Consistent with the defects observed in CD patients , heterozygous knock-out mice are perinatal lethal due to skeletal abnormalities [29] . Conditional knock-out mice have been generated to study SOX9 function: pancreas-specific inactivation of Sox9 using Pdx1:Cre reveals a critical role in the maintenance of the pancreatic progenitor pool [25] , whereas liver-specific inactivation of Sox9 using Albumin/α-fetoprotein ( Alfp ) :Cre shows that it is required for the timely maturation of asymmetrical structures to symmetrical biliary ducts [23] . A potential role for SOX9 in HPD development has not yet been investigated due to the lack of a HPD-specific Cre line . The zebrafish genome contains two sox9 orthologs , sox9a and sox9b , which exhibit partially overlapping expression patterns in the craniofacial cartilage , otic placodes and pectoral appendages [30] . Null mutants of sox9a exhibit cartilage defects that mimic those observed in human CD [31] . Although a similar phenotype has been reported for the sox9bb971 mutant [30] , the chromosomal deletion which underlies the b971 lesion removes eleven other genes , greatly limiting the use of this allele to study the function of Sox9b . Here , we dissect the requirement for Sox9b in the development of the pancreaticobiliary ductal system in zebrafish . We show that similar to mammalian Sox9 , zebrafish sox9b is expressed in the pancreaticobiliary ductal system . Detailed phenotypic analysis of a sox9b TILLING mutant reveals that Sox9b regulates the formation of the HPD system as well as the morphogenesis of the intrapancreatic and intrahepatic ducts . Strikingly , the pancreaticobiliary phenotypes observed in larvae worsen during juvenile stages and lead to cholestasis in the homozygous mutant adult fish . We also observed a positive feedback loop between Sox9b and Notch signaling in the developing intrahepatic biliary cells: Notch signaling regulates sox9b expression , and in turn Sox9b is required to maintain Notch activity in the intrahepatic biliary cells . Intrigued by the recent data revealing Sox9 expression in the ductal trees of the liver and pancreas as well as in the HPD system in mouse [24] , we analyzed the expression pattern of sox9b in zebrafish by in situ hybridization . We found that in addition to the head region and pectoral fins , sox9b is specifically expressed in the pancreaticobiliary ductal system ( Figure 1A–1D ) . At 30 hours post fertilization ( hpf ) , sox9b is expressed in a segment of the foregut endoderm ( bracket , Figure 1A ) that appears to give rise to the liver bud ( arrow , Figure 1B ) and the HPD primordium ( bracket , Figure 1B ) . At 60 hpf , sox9b expression becomes evident in the intrahepatic ducts ( arrow , Figure 1C ) and then extends to the extra- and intrapancreatic ducts ( white arrow , Figure 1D ) . In contrast to sox9b , sox9a does not appear to be expressed in the pancreaticobiliary ductal system in zebrafish ( Figure S1 , left panel ) . These data show that zebrafish sox9b recapitulates the expression pattern of mammalian Sox9 in the intrapancreatic and intrahepatic ducts as well as in the HPD system [23]–[25] . To investigate the potential role of Sox9b in the formation of the pancreaticobiliary ductal system , we isolated a novel mutation in sox9b , sox9bfh313 , in collaboration with the Zebrafish TILLING Consortium . Contrary to sox9bb971 which consists of a deletion of the lower tip of linkage group 3 [30] , sox9bfh313 is a point mutation located in the first exon of sox9b ( Figure 1E ) . The A to T transversion at position 302 leads to a premature stop codon at amino acid Lys68 . This nonsense mutation likely leads to the synthesis of a truncated protein that lacks the HMG-box DNA binding domain ( Figure 1F ) and therefore would be non-functional . In situ hybridization analyses revealed a substantial decrease of sox9b expression in sox9bfh313 mutants at 72 hpf ( data not shown ) , possibly via nonsense mediated mRNA decay . In order to analyze a potential redundancy between sox9 gene functions in zebrafish , we examined the expression of sox9a in sox9bfh313 mutants ( Figure S1 ) . As in wild-type , sox9a expression appears to be excluded from the digestive organs in sox9bfh313 mutants , suggesting that sox9a expression does not compensate for the reduction of Sox9b function in these mutants . Hence , sox9bfh313 is the first point mutation described for this gene in zebrafish and is likely to represent a severe loss-of-function allele . In contrast to a previous report describing the phenotypes of sox9bb971 mutants and sox9b morpholino-injected embryos [30] , homozygous sox9bfh313 mutants exhibit a normal external morphology and do not show a curly-down body axis or craniofacial defects ( data not shown ) . sox9b mutants survive to the adult stage but are much thinner than their wild-type or heterozygous siblings ( data not shown ) . Dissection of the digestive system of 5-month old sox9b mutants revealed preserved anatomical relationships , including a three-lobed liver and correctly-looped intestine; however , both the liver and pancreas were strikingly dark green suggesting abnormal bile accumulation ( cholestasis ) in these organs ( Figure 2A–2B ) . Hematoxylin-and-eosin staining of histological sections of mutant organs showed that both organs exhibited lesions with extensive proliferation and dilation of the ducts , which were surrounded by fibrotic tissue ( Figure 2C–2D″ ) . Interestingly , in the liver , ductal defects were restricted to the region that connects to the extrahepatic ductal system ( dashed rectangle , Figure 2D ) whereas the rest of the organ was much less affected . In contrast , ductal defects in the pancreas were present throughout the organ and worsened towards its distal part ( Figure 2D ) . In the pancreas , the acinar compartment was greatly reduced and secondary islets could not be detected in the sections examined ( Figure 2D″ ) . Due to the robust and highly conserved expression of sox9b in the pancreaticobiliary ductal system and the striking liver and pancreas phenotypes seen in the adult mutant fish , we decided to further investigate the roles of Sox9b in the development of these tissues . In zebrafish , the HPD system exhibits unique gene expression profiles that separate it from the liver and pancreas starting at early developmental stages [15] . At 50 hpf , the primordium of the HPD system can be distinguished by strong labeling with the 2F11 antibody , whose antigen remains to be identified [15] , [32] ( bracket , Figure 3A , 3A′ ) , and low expression of the transcription factor Prox1 [15] ( bracket , Figure 3C ) . In contrast , the liver and pancreas exhibit moderate labeling of 2F11 ( Figure 3A , 3A′ ) , but high expression of Prox1 ( Figure 3C ) . In sox9b mutants , 2F11 labeling was mostly absent from the region where the presumptive HPD primordium resides ( bracket , Figure 3B , 3B′ ) . Moreover , we observed elevated expression of Prox1 in the same region ( bracket , Figure 3D ) . 2F11 labeling showed that by 80 hpf , the HPD system in wild-type larvae has developed into different compartments , including the extrahepatic duct , cystic duct , common bile duct , and gallbladder [15] ( Figure 3E , 3E′ ) . At the equivalent stage , the differentiation of the HPD system had partially recovered in sox9b mutants as suggested by 2F11 labeling . However , it was severely dysmorphic , with no clear morphological distinction between the cystic duct , extrahepatic duct , and common bile duct ( Figure 3F , 3F′ ) . Furthermore , the mutant HPD system often seemed to intrude into the liver ( Figure 3F ) , which was never observed in wild-type larvae . These data indicate that the HPD primordium in sox9b mutants exhibits patterning and differentiation defects . Concordant with the dysmorphic HPD system , the gallbladder in sox9b mutants was often indistinguishable based on morphology ( Figure 3F and Figure S3A , S3B ) . We analyzed the expression of sox17 which marks the gallbladder and its primordium from 36 hpf to 5 days post-fertilization ( dpf ) [33] , and found that it was greatly reduced or absent in sox9b mutants at 52 hpf ( Figure 3G ) and that it did not recover during later development ( Figure 3H ) . This defect in sox17 expression supports the notion that gallbladder development is severely impaired in sox9b mutants . We then addressed the role of Sox9b in intrapancreatic duct formation by using the double transgenic line Tg ( Tp1bglob:GFP ) ;Tg ( Tp1bglob:H2B-mCherry ) that expresses both GFP and H2B-mCherry under the control of a Notch-responsive element [34] , [35] . This line allows the visualization of the shape and nuclei of the intrapancreatic duct cells , as indicated by the overlapping expression of these fluorescent proteins with ductal markers such as E-cadherin and 2F11 [34] , [35] . Intrapancreatic ducts derive from cells within the ventral pancreatic bud that migrate towards , and eventually surround , the principal islet at 48 hpf [22] . From 60 hpf , ductal progenitors start to migrate caudally to form a row of cells that give rise to the main intrapancreatic duct [22] ( Figure 4A–4A′ ) . The migration of the ductal progenitors did not seem to be impaired in sox9b mutants; however , the number of cells within the intrapancreatic ducts was significantly reduced ( Figure 4B–4B′ , 4G ) . In wild-type larvae , at 100 hpf , the pancreatic tail keeps elongating , the number of ductal cells has slightly increased ( Figure 4C–4C′ , 4G ) and secondary branches ( arrowheads , Figure 4C–4C′ ) start to form from the main duct . In contrast , in sox9b mutants , the number of ductal cells did not increase from 80 to 100 hpf , and no secondary branches appeared , resulting in a primitive ductal system ( Figure 4D–4D′ , 4G ) . At 120 hpf , the ductal network in wild-type larvae has become more complex with numerous secondary branches ( arrowheads , Figure 4E″ ) spreading over the acinar compartment ( Figure 4E–4E′″ ) . In contrast , the intrapancreatic ductal system in sox9b mutants remained poorly developed and clusters of cells could be observed along the main duct ( Figure 4F–4F′″ ) , which was still devoid of secondary branches . These data indicate that fewer intrapancreatic duct cells differentiate in the mutants and those that do fail to undergo branching morphogenesis . Furthermore , the number of ductal cells in sox9b mutants did not increase as in wild-types . Such a defect is likely due to a problem with cell differentiation as we did not observe any obvious differences in ductal cell proliferation or survival between wild-types and sox9b mutants ( data not shown ) . Given that the adult mutant pancreas exhibits a loss of acinar and potentially endocrine tissues , we analyzed the formation of these compartments during larval and juvenile stages . As assessed by Elastase staining , sox9b mutants showed apparently normal proportion of pancreatic acinar tissue at all stages analyzed ( Figure 4B , 4D , 4F and Figure S2B″ , S2D″ , S2F″ ) . As for the endocrine tissue , we investigated the morphology of the primary islet by analyzing TgBAC ( neurod:GFP ) expression which marks early endocrine cells [36] . At late larval ( 7 dpf ) as well as juvenile ( 2 and 3 weeks ) stages , the area of the primary islet appeared similar in wild-type and sox9b mutant animals ( Figure 4H ) , suggesting that Sox9b is not required for primary islet formation . At 4 weeks of age , the sox9b mutant primary islet was half the size of the wild-type primary islet ( Figure 4H ) . However , it is important to note that at this stage , sox9b mutant pancreata were also much less developed than wild-type pancreata ( Figure S2E′″ , S2F′″ ) . Indeed , sox9b juvenile mutants often exhibit growth retardation compared to wild-types and thus , the smaller size of the primary islet could be attributed to an overall growth defect . In addition to the primary islet , we investigated the formation of secondary islets that arise from progenitors in the intrapancreatic ducts [34] , [35] and that , during larval and juvenile stages , appear as small clusters of delaminated cells [34] . We decided to also count single TgBAC ( neurod:GFP ) -positive cells that recently delaminated from the ducts and assumed an endocrine fate . Hence , counting the number of TgBAC ( neurod:GFP ) -positive cells/clusters along the intrapancreatic ducts ( Figure S2A′–S2F′ ) , we observed a difference between wild-type and sox9b mutant animals at two weeks of age . At 3 and 4 weeks of age , this difference became more pronounced with respectively a 50% and 80% decrease in TgBAC ( neurod:GFP ) -positive cell/cluster number in sox9b mutants ( Figure 4I ) . Given that the mutant intrapancreatic ductal network remained primitive and failed to expand at juvenile stages ( Figure S2D′″ , S2F′″ ) , the defect in secondary islet formation could be related to the lower number of progenitors within the pancreas . Altogether , these data indicate that Sox9b function is required for the development of the intrapancreatic ductal system as well as - directly or indirectly - for the formation of secondary islets . To examine the role of Sox9b in intrahepatic biliary development , we used the Tg ( Tp1bglob:GFP ) line which also marks the intrahepatic biliary cells [21] , [35] . During zebrafish liver development , the intrahepatic biliary cells undergo significant morphological changes , whereby these initially contiguous cells separate from one another and interconnect via cytoplasmic processes [21] . By 96 hpf , the wild-type intrahepatic biliary system is composed of a lattice-like network of long ducts joined by short interconnecting ducts [21] ( Figure 5A ) . At the equivalent stage , sox9b mutant livers contained similar numbers of intrahepatic biliary cells and hepatocytes as wild-type ( data not shown ) , suggesting that differentiation of the biliary cells is not affected in the mutants . We did not detect any apoptosis of the biliary cells in wild-type or mutant larvae . Strikingly , we observed that most of the biliary cells in the mutant livers failed to separate from one another ( Figure 5B ) . We quantified the percentage of single intrahepatic biliary cells versus cells in cluster of two , three or four and more cells , and found a significant decrease in the percentage of single intrahepatic biliary cells in sox9b mutants compared to wild-types concomitant with a significant increase in the percentage of cells in clusters of four and more cells ( Figure 5C ) . Moreover , the long bile ducts in the mutants appeared to be wider than those in wild-types ( diameters of the mutant ducts: 3 . 5 µm or wider; wild-type ducts: 2 . 5 µm or thinner ) , and were less branched ( Figure 5D ) . We then used the Tg ( fabp10:ras-GFP ) line [37] to analyze hepatocyte organization , and co-labeled the animals with an antibody against the bile transporter BSEP to mark the bile canaliculi [38] ( Figure 5E , 5G ) . At 96 hpf , hepatocytes in wild-type livers are arranged as tubules surrounding intrahepatic biliary ducts [20] ( Figure 5E ) . Bile canaliculi are located on the hepatocyte apical membrane which can be marked by the activated leukocyte cell adhesion molecule Alcam [39] . However , in sox9b mutants , hepatocytes often formed spherical rosettes with bile canaliculi and Alcam expression located in the center ( Figure 5G ) . This phenotype coincided with the aberrant clustering of intrahepatic biliary cells . Moreover , we found that the canaliculi in sox9b mutants appeared to be shorter and wider compared to wild-types ( arrows , Figure 5F , 5H ) , which is consistent with recent data showing the highly coordinated development of intrahepatic biliary cells and bile canaliculi [21] . In liver-specific Sox9-inactivated mice , intrahepatic biliary duct morphogenesis is delayed until birth [23] , which incited us to track the development of the intrahepatic biliary system in wild-types and sox9b mutants during juvenile stages . We found that intrahepatic biliary duct morphogenesis did not recover in sox9b mutants and that these animals did not generate morphologically normal bile ducts ( Figure S2G–S2L ) . These data show that , despite a conserved requirement for SOX9 in intrahepatic biliary duct development , zebrafish sox9b mutants exhibit a much more severe intrahepatic biliary duct phenotype than the liver-specific knockout mouse model . To determine whether the cholestasis-like phenotype observed in adult sox9b mutants occurred during early larval development , we administered fluorescent lipid analogs used to visualize bile transport [40] to 6 dpf-old wild-type and mutant larvae . These fluorescent analogs consist of fatty acids with acyl chains of 5- and 2-carbons ( C5:0 and C2:0 , respectively ) tagged with the BODIPY fluorophore . We selected these two analogs because of the different cells and subcellular details each analog reveals following ingestion . BODIPY-FL C5:0 reveals a high degree of subcellular detail in hepatocytes and acinar cells , such as lipid droplets and zymogen granules , as well as in the ductal networks in the liver and pancreas . The shorter BODIPY-FL C2:0 illuminates the hepatic and pancreatic ducts , as well as the gallbladder providing a functional readout of gallbladder and ductal integrity . Wild-type larvae fed BODIPY-FL C2:0 exhibited a strong fluorescence signal in their gallbladders ( Figure S3A ) , indicating that bile production , drainage and accumulation was normal . Conversely , no gallbladder BODIPY-FL C2:0 signal was observed in sox9b mutants , consistent with their defective gallbladder development ( Figure S3B ) . In the pancreas , BODIPY-FL C2:0 fluorescence was detected throughout the entire intrapancreatic ductal network in wild-type larvae ( Figure S3E , S3E′ ) , while it was restricted to the anterior region of the pancreas in sox9b mutants ( Figure S3F , S3F′ ) , suggesting that the distal intrapancreatic ducts were not functional . Moreover , we observed large pools of fluorescent fluid accumulating in the mutants' pancreatic tail ( Figure S3F ) , which is not typically observed in wild-type larvae unless the gallbladder ruptures . This abnormal extracellular accumulation of fluid ( likely pancreatic juice or bile ) in and around sox9b mutant pancreata is consistent with their malformed pancreaticobiliary ductal system . Administering BODIPY-FL C5:0 to sox9b mutant livers confirmed the dilation and lack of branching morphogenesis of the intrahepatic biliary ducts described above as well as the deformation of bile canaliculi in hepatocytes ( Figure S3C , S3C′ , S3D , S3D′ ) . Taken together , these data support the notion that the cholestasis-like phenotype observed in the adult sox9b mutants results from defects in early ductal morphogenesis . The defects in intrahepatic biliary ducts and bile canaliculi observed in sox9b mutants are strikingly similar to those reported in the mouse and zebrafish models of Notch deficiency [6] , [9] , [20] , [21] , [41] . In particular , it has been shown in zebrafish that Notch signaling directs the segregation of intrahepatic biliary cells between 70 and 96 hpf [21] . Given that the intrahepatic biliary cells in sox9b mutant livers fail to separate from one another , we hypothesized that Sox9b interacts with Notch signaling to regulate the morphogenesis of the intrahepatic biliary ducts . To test whether inhibiting Notch signaling affects sox9b expression , we treated wild-type and sox9b heterozygous animals with a low dose of the γ-secretase inhibitor DAPT from 75 to 99 hpf [42] , and assessed sox9b expression by in situ hybridization ( Figure 6A–6E ) . Such DAPT treatment caused a reduction in sox9b expression which was more pronounced in sox9b heterozygotes than in wild-types . We also performed the reverse experiment by using Tg ( hsp70l:Gal4 ) ;Tg ( UAS:myc-Notch1a-intra ) hemizygous larvae to induce ubiquitous overexpression of the Notch intracellular domain ( NICD ) upon heat-shock treatment [43] . These animals were heat-shocked at 80 hpf and sox9b expression was analyzed 26 hours later by in situ hybridization . The heat-shock treatment efficiently induced overactivation of Notch signaling activity in the Tg ( Tp1bglob:GFP ) ; Tg ( Tp1bglob:H2B-mCherry ) larvae ( Figure 6H , 6I ) . We observed an increase in sox9b expression throughout the pancreas , liver and HPD system ( bracket ) in the double–transgenic larvae compared to their single-transgenic control siblings ( Figure 6F , 6G ) . Expression of sox9b could even be detected in the gallbladder ( arrowhead , Figure 6G ) , which was not seen in control larvae ( Figure 6F ) . This increase in sox9b expression is unlikely due to an NICD-induced proliferation of sox9b-positive cells because we could already detect higher levels of sox9b expression as early as four hours after heat-shock treatment ( data not shown ) . Taken together , these loss- and gain-of-function analyses reveal that Notch signaling regulates sox9b expression during intrahepatic biliary duct morphogenesis . These data are consistent with studies in mouse showing that Notch1 can directly bind to the Sox9 promoter [9] . To further analyze the dynamics of Notch signaling in sox9b mutants , we utilized Tg ( Tp1bglob:H2B-mCherry ) ;Tg ( Tp1bglob:VenusPest ) animals in which the Notch-responsive element drives the expression of both H2B-mCherry and VenusPest fluorescent proteins . Contrary to H2B-mCherry which is very stable , the destabilized fluorescent protein VenusPest has a short half-life ( 2 hours for GFP-Pest in mammalian cells ) [44] . Thus , the Tg ( Tp1bglob:H2B-mCherry ) ;Tg ( Tp1bglob:VenusPest ) -double positive cells are currently Notch responsive , whereas the Tg ( Tp1bglob:H2B-mCherry ) -positive;Tg ( Tp1bglob:VenusPest ) -negative cells were positive for Notch signaling in their recent past but have since switched it off [34] . In wild-type and sox9b heterozygous livers , the expression of Tg ( Tp1bglob:H2B-mCherry ) and Tg ( Tp1bglob:VenusPest ) largely overlapped at 75 hpf ( Figure 7A , 7A′ , 7G , and data not shown ) . Between 99 and 123 hpf , a small proportion of intrahepatic biliary cells switched off Notch signaling and became Tg ( Tp1bglog:H2B-mCherry ) -single positive ( Figure 7B–7C′ , G ) . Up to 99 hpf , sox9b mutant livers exhibited a similar pattern of Notch signaling activity as wild-type with a clear overlap between Tg ( Tp1bglob:H2B-mCherry ) and Tg ( Tp1bglob:VenusPest ) expression ( Figure 7D–7E′ , 7G ) . However , at 123 hpf , the proportion of Tg ( Tp1bglog:H2B-mCherry ) -single positive cells was significantly higher in sox9b mutants compared to wild-types or heterozygotes ( Figure 7F , 7F′ , 7G ) , suggesting that sox9b mutants fail to maintain Notch signaling in the intrahepatic biliary cells . Interestingly , we did not observe any obvious phenotype when assessing Notch signaling activity in the mutant intrapancreatic ducts . To follow up this observation , we treated wild-type and mutant larvae with 50 µM DAPT , a dose that only partially inhibits Notch signaling . Upon treatment between 106 and 154 hpf , DAPT-treated wild-type larvae showed a specific loss of Notch signaling activity in the proximal ( p ) rather than the distal ( d ) region of the liver ( Figure S4C ) . In DAPT-treated mutant larvae , we observed a drastic loss of Notch signaling throughout the entire liver ( Figure S4D , S4E ) , indicating that sox9b mutant biliary cells are more likely to lose Notch signaling activity than wild-type cells . Altogether , these data indicate that Sox9b is required for the maintenance but not the initiation of Notch signaling in the intrahepatic biliary cells . Considering our previous data showing that Notch signaling regulates sox9b expression , we hypothesize that Notch and Sox9b interact in a positive feedback loop to ensure the development of the intrahepatic biliary network . To better understand the biological significance of Notch responsiveness during intrahepatic biliary duct morphogenesis , we compared the distribution of Tg ( Tp1bglob:H2B-mCherry ) -single positive cells and Tg ( Tp1bglob:H2B-mCherry ) ;Tg ( Tp1bglob:VenusPest ) -double positive cells in wild-type livers . At 123 hpf , 89% of the double-positive cells existed as individual cells connecting to one another through cellular extensions ( Figure 7C′; 467 cells in 5 larvae were analyzed ) . On the other hand , 60% of the Tg ( Tp1bglob:H2B-mCherry ) -single positive cells , which had switched off Notch signaling , were intermingled with each other to form larger groups ( Figure 7C′ , arrows; 110 cells in 5 larvae were analyzed ) . Interestingly , these clusters of Tg ( Tp1bglob:H2B-mCherry ) -single positive cells were mostly present in the multicellular large bile ducts contiguous with the extrahepatic duct , whereas the individual Tg ( Tp1bglob:H2B-mCherry ) ;Tg ( Tp1bglob:VenusPest ) -double positive cells were localized in the distal part of the liver and formed smaller bile ducts ( Figure 7C′ ) . These data suggest that Notch responsiveness correlates with the relative position of the intrahepatic biliary cells , with the cells turning off Notch signaling forming the large bile ducts in the proximal region of the liver . In sox9b mutant livers , we observed more clusters of Tg ( Tp1bglob:H2B-mCherry ) -single positive cells in both the proximal and distal regions ( Figure 7F′ , arrows; 243 cells in 6 larvae were analyzed ) . We then addressed whether Notch responsiveness was related to the proliferation status of the intrahepatic biliary cells , which would correlate with the increase in ductal structures observed in mutant adults . We incubated wild-type and mutant animals with the replication marker 5-ethynyl-2′-deoxyuridine ( EdU ) during two intervals of larval development , and analyzed EdU incorporation in Tg ( Tp1bglob:H2B-mCherry ) -single positive and Tg ( Tp1bglob:H2B-mCherry ) ;Tg ( Tp1bglob:VenusPest ) -double positive cells ( Figure 7H ) . In wild-type larvae , approximately 30% of the Tg ( Tp1bglob:H2B-mCherry ) -single positive cells incorporated EdU after incubation from 96 to 120 hpf . The Tg ( Tp1bglob:H2B-mCherry ) ;Tg ( Tp1bglob:VenusPest ) -double positive cells exhibited a slightly higher percentage of EdU incorporation , although the difference between these two cell populations was not statistically significant ( p>0 . 08 ) . Similar rates of EdU incorporation were observed when we incubated the animals from 120 to 144 hpf . In sox9b mutants , we detected an increase in EdU incorporation in both the single and double positive cells compared to wild-type ( Figure 7H ) , with the increase being more pronounced in the Tg ( Tp1bglob:H2B-mCherry ) -single positive cells than in the double positive cells . To further support the hypothesis that loss of Notch signaling correlates with increased proliferation of biliary cells , we found that partial inhibition of Notch signaling in wild-type larvae by a low dose DAPT treatment led to an increase in biliary cell proliferation similar to that observed in sox9b mutants ( data not shown ) . Hence , taken together , these data suggest that the reduction in Notch signaling in sox9b mutants promotes the clustering and proliferation of the intrahepatic biliary cells , which is consistent with the biliary duct defects observed in the mutant adults . In this study , we analyzed a novel sox9b mutant in zebrafish , revealing for the first time that global loss-of-function of Sox9b severely impairs the development of the pancreaticobiliary ductal system . In particular , we showed that in the mutant animals , the HPD system is malformed , and the intrahepatic and intrapancreatic ducts fail to form a functional ductal network . We also uncovered the existence of a Notch-Sox9b positive feedback loop that is crucial for intrahepatic biliary duct development . Our study thus brings new insights into our understanding of pancreaticobiliary ductal system formation , an important but understudied process . We showed that loss of Sox9b function leads to mispatterning of the HPD primordium . Fgf10 signaling also plays a pivotal role in the formation of the HPD system in zebrafish [15] . Zebrafish fgf10 is expressed in the mesenchyme surrounding the HPD system and intestine but not that surrounding the liver or pancreas [17] . fgf10 mutants show a dysmorphic HPD with a reduction or loss of the common bile duct as well as reduced extrapancreatic and extrahepatic ducts . Moreover , fgf10 mutants misexpress hepatic markers such as Prox1 and Hnf4α in their HPD system and pancreas , leading to the ectopic differentiation of some cells in these organs towards a liver fate . Based on the similarities of the fgf10 and sox9b mutant phenotypes , it is possible that Fgf10 and Sox9b interact; for example , Fgf10 signaling could induce or maintain sox9b expression in the HPD primordium to modulate the patterning of this tissue . However , we did not observe any obvious change in sox9b expression in the HPD upon Fgf receptor pharmacological inhibition treatment , nor in fgf10 expression in the surrounding mesenchyme in sox9b mutants . Thus , Fgf10 and Sox9b functions might intersect in other ways . Our data show that Sox9b is involved in gallbladder development , whose primordium specifically expresses sox17 . sox17 is expressed in all endodermal cells during gastrulation [45] and starts to be reexpressed at 36 hpf in a small region of the liver close to the extrahepatic duct [33] . It is then detected in the gallbladder at 60 hpf where it persists until 5 dpf . It will be interesting to determine whether Sox9b directly regulates sox17 expression and also to identify the additional factors involved in inducing sox17 expression in a subset of sox9b positive cells . The intrapancreatic ductal network is severely disrupted in sox9b mutant larvae , leading to the formation of dilated ducts surrounded by fibrotic tissue in mutant adults . At first glance , this adult phenotype is reminiscent of the one caused by pancreas-specific inactivation of mouse Sox9 [25]; however , given that the pancreatic remnants described in the mouse mutants come from unrecombined Sox9+ progenitor cells , the zebrafish adult phenotype appears much less severe than the one seen in mouse . The discrepancy between the two models could be explained by differences during early pancreas development . Indeed , mouse SOX9 is expressed in pluripotent pancreatic progenitors and is required to stimulate their proliferation and survival [25] . However , in zebrafish at the earliest stages of pancreas development , endocrine cells derive first from the endodermal epithelium [18] and then from the extrapancreatic duct [46]; only during larval stages , do endocrine cells derive from progenitors within the intrapancreatic ducts [34] , likely following mechanisms similar to those regulating the secondary transition in mouse . Analysis of zebrafish sox9b expression at early stages suggests that it is present in a subset of ventral pancreatic bud cells , which may correspond to the ductal progenitors . If this interpretation is correct , the lack of Sox9b function in zebrafish would first impair these ductal progenitors , leading to a reduced number of intrapancreatic ductal cells . The first two waves of endocrine cells , as well as the acinar cells , which originate from sox9b negative tissue , would therefore not be affected in a sox9b mutant . Loss of Sox9b function in zebrafish severely impairs the development of the intrahepatic biliary network including the morphogenesis of the bile canaliculi . These phenotypes are more severe than those seen in liver-specific SOX9-depleted mice [23] . During mouse liver development , SOX9 expression is first detected at E10 . 5 in the endodermal cells lining the lumen of the liver diverticulum [23] . This expression is lost as the liver cells migrate into the septum transversum , but re-emerges at E11 . 5 in cells that form the ductal plate . The Alfp:Cre line that was used to recombine the Sox9 locus becomes active at E11 . 5 , thus likely only inhibiting the second phase of SOX9 expression . Therefore , the phenotypic differences between the zebrafish sox9b mutant and the mouse model might be related to the consequences of an earlier depletion in Sox9b in zebrafish than in mouse , illustrating the value of the zebrafish Sox9b global loss-of-function model to uncover functions of this critical transcriptional regulator . However , differences in the expression or function between sox9b ( zebrafish ) and Sox9 ( mouse ) may also explain the phenotypic differences between the two models . Intrahepatic biliary cells in sox9b mutant livers fail to segregate from one another and remain clustered , leading to a primitive ductal network . Such defects are strikingly similar to Notch-deficient zebrafish and mouse models [6] , [9] , [20] , [21] , [41] which phenocopy the human Alagille syndrome ( OMIM#118450 ) , which itself is associated with JAGGED1 and NOTCH2 mutations . Notably , this developmental disorder is characterized by cholestasis due to a paucity of biliary ducts . Our data indicate that Sox9b interacts with Notch signaling in a positive feedback loop to regulate intrahepatic biliary duct morphogenesis . Studies in mouse have provided possible mechanisms underlying this Sox9-Notch crosstalk: the Sox9 promoter displays ten consensus Rbpj binding sites and is a direct target of Notch signaling [9] . In addition , SOX9 modulates Notch signaling by positively regulating the expression of the Notch downstream target gene Hes1 in the liver [23] as well as in other organs such as the pancreas [25] . These mechanisms are likely to be at play in zebrafish as well , and it will be interesting to delve deeper into the complexities of this positive feedback loop . Intriguingly , we found that the biliary cells in the proximal region of wild-type livers tend to lose Notch signaling more quickly and are more susceptible to Notch inhibition than the distal cells . Notably , the biliary cells in the proximal region form large ducts whereas the distal cells form small ducts . In sox9b mutants , loss of Sox9b function tempers Notch signaling in all biliary cells . Consequently , the mutant livers exhibit aberrant clusters of Tg ( Tp1bglob:H2B-mCherry ) -single positive cells in their distal region , suggesting that lack of Sox9b-mediated maintenance of Notch signaling could promote the formation of large ectopic ducts in distal regions of the liver . Interestingly , it has been shown in mouse that SOX9 , which is expressed in all biliary cells at E18 . 5 , persists in small ducts but regresses from large ducts after birth [23] . Therefore , it is possible that in wild-type animals , the loss of Notch signaling in the proximal region of the liver induces the local loss of sox9b expression , leading to the formation of large bile ducts . To begin to test this hypothesis , it will be necessary to generate a Sox9b antibody in order to examine Sox9b expression at cellular resolution . The identification of markers that distinguish large and small bile ducts would also greatly facilitate such studies . Our data suggest that in addition to regulating biliary morphogenesis , the Notch-Sox9b module also influences the proliferation of biliary cells . In sox9b mutants , the biliary cells that turn off Notch signaling exhibit a higher proliferation rate . Recent lineage tracing studies in mouse have shown that Sox9 is expressed in liver progenitor cells that reside within the biliary ducts [47] , [48] . It will be interesting to determine whether zebrafish sox9b is also expressed in liver progenitor cells , whether loss of Sox9b function affects their proliferation rate , and investigate into the underlying mechanisms . In addition to bringing new insights into our understanding of the development of the pancreaticobiliary ductal system , the analysis of zebrafish sox9b mutants should lead one to consider SOX9 as a candidate gene for human diseases associated with HPD , intrapancreatic or intrahepatic duct malformations . In particular , numerous cases of congenital non-syndromic or syndromic extrahepatic biliary atresia have been reported [49] and their causes remain unknown [50] . These conditions likely have multifactorial causes and do not display simple Mendelian inheritance . Given the expression pattern of SOX9 in mammals [24] , [26] , [28] as well as the phenotypes caused by loss of SOX9 function in zebrafish and mouse , SOX9 is therefore an interesting gene to sequence in those patients . Campomelic dysplasia has been shown to be essentially associated with heterozygous mutations that are predicted to severely disrupt SOX9 protein structure and function [51]; but milder lesions could be associated with pancreaticobiliary duct malformations and contribute to the onset or severity of these malformations without necessarily impairing skeletal development . Embryos and adult fish were raised and maintained under standard laboratory conditions [52] . The sox9bfh313 heterozygote was crossed with Tg ( fabp10:ras-GFP ) s942 [37] , Tg ( Tp1bglob:GFP ) um14 [35] , Tg ( Tp1bglob:H2B-mCherry ) s939 , Tg ( Tp1bglob:VenusPest ) s940 [34] , TgBAC ( neurod:GFP ) nl1 [53] and genotyped according to the TILLING center protocol with AcuI or SfcI ( http://labs . fhcrc . org/moens/Tilling_Mutants/sox9b/allele_1 . html ) . We also used Tg ( hsp70l:Gal4 ) 1 . 5kca4 and Tg ( UAS:myc-Notch1a-intra ) kca3 [43] hemizygous or double hemizygous fish . Whole-mount in situ hybridizations were performed as described previously [54] using sox9b [30] and sox17 [45] probes . Animals were photographed with a Zeiss Axioplan using an Axiocam digital camera . Immunohistochemistry on whole-mount animals or cryosections was performed as previously described [16] , using the following antibodies: chicken polyclonal anti-GFP ( 1∶1000; Aves Labs , Tigard , OR , USA ) , rabbit polyclonal anti-Prox1 ( 1∶1000; Chemicon , Billerica , MA , USA ) , mouse monoclonal 2F11 ( 1∶1000; Abcam , Cambridge , UK ) , rabbit polyclonal anti-dsRed ( 1∶500; Clontech , Mountain View , CA , USA ) , rabbit polyclonal anti-ABCB11/BSEP ( 1∶1000; Kamiya Biomedical ) , mouse monoclonal anti-Alcam/Zn8 ( 1∶20; ZIRC ) , rabbit polyclonal anti-elastase ( 1∶200; Millipore AB1216 ) and fluorescently conjugated Alexa antibodies ( 1∶250; Molecular Probes , Carlsbad , CA , USA ) . Samples were imaged on a Zeiss Pascal confocal microscope . The width of the bile ducts was measured using the “local thickness” function in Fiji software . Heat-shock treatments of Tg ( hsp70l:Gal4 ) 1 . 5kca4 larvae were performed at 38°C as described [16] . To inhibit Notch signaling , larvae were treated with 20 µM or 50 µM DAPT ( Sigma ) in egg water [42] . Control larvae from the same batch were treated with 0 . 4% DMSO in egg water . Statistical analyses were performed using the Student's two-tailed t-test . Five-months old zebrafish ( one wild-type and one mutant ) were euthanized and their digestive systems were dissected and fixed overnight with formalin at 4°C . The samples were embedded in paraffin , cut into 5 µm sections , and stained with hematoxylin and eosin . 6–7 dpf larvae were fed with BODIPY C2 . 0 or BODIPY C5 . 0 as described in [40] for 6–8 hours before being mounted and imaged live . The larvae were subsequently genotyped and 9 wild-type and 9 sox9b mutant animals were analyzed . To assess the proliferation of the intrahepatic biliary cells , wild-type and sox9b mutant larvae were incubated in 7 µM EdU dissolved in egg water during the stages indicated . Control larvae collected from the same batch were treated with 1 . 7% DMSO . Animals were fixed after incubation and processed using the Click-iT EdU Imaging Kit ( Invitrogen ) . Quantification of EdU incorporation was conducted using the Cell Counter plug-in in ImageJ .	The liver and pancreas function as exocrine glands that secrete bile and pancreatic juice , respectively , to aid the digestion and absorption of nutrients . These fluids reach the intestine via the pancreaticobiliary ductal system , a complex network of ducts . Despite its pivotal role , the development of this ductal system is poorly understood . We have discovered that the zebrafish transcription factor gene sox9b , like its mammalian ortholog , is specifically expressed in the pancreaticobiliary ductal system . The perinatal lethality of Sox9 heterozygous mice makes the analysis of SOX9 function challenging; thus , we turned to the zebrafish to analyze the role of SOX9 in pancreaticobiliary ductal system development . We found that zebrafish sox9b mutants , which survive to adulthood , display defects in the morphogenesis of this ductal network: the intrahepatic and intrapancreatic ducts fail to form a branched network , whereas the ducts connecting the liver and pancreas to the intestine are malformed . These ductal defects affect bile transport and lead to cholestasis in adult mutant fish . At the molecular level , Sox9b interacts with the Notch signaling pathway to regulate the development of the intrahepatic biliary network . Therefore , our work in zebrafish reveals a broad and complex role for SOX9 in pancreaticobiliary ductal system morphogenesis .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "animal", "models", "developmental", "biology", "zebrafish", "model", "organisms", "genetics", "biology", "morphogenesis", "genetics", "and", "genomics", "gene", "function" ]	2012	sox9b Is a Key Regulator of Pancreaticobiliary Ductal System Development
Mitigation of a severe influenza pandemic can be achieved using a range of interventions to reduce transmission . Interventions can reduce the impact of an outbreak and buy time until vaccines are developed , but they may have high social and economic costs . The non-linear effect on the epidemic dynamics means that suitable strategies crucially depend on the precise aim of the intervention . National pandemic influenza plans rarely contain clear statements of policy objectives or prioritization of potentially conflicting aims , such as minimizing mortality ( depending on the severity of a pandemic ) or peak prevalence or limiting the socio-economic burden of contact-reducing interventions . We use epidemiological models of influenza A to investigate how contact-reducing interventions and availability of antiviral drugs or pre-pandemic vaccines contribute to achieving particular policy objectives . Our analyses show that the ideal strategy depends on the aim of an intervention and that the achievement of one policy objective may preclude success with others , e . g . , constraining peak demand for public health resources may lengthen the duration of the epidemic and hence its economic and social impact . Constraining total case numbers can be achieved by a range of strategies , whereas strategies which additionally constrain peak demand for services require a more sophisticated intervention . If , for example , there are multiple objectives which must be achieved prior to the availability of a pandemic vaccine ( i . e . , a time-limited intervention ) , our analysis shows that interventions should be implemented several weeks into the epidemic , not at the very start . This observation is shown to be robust across a range of constraints and for uncertainty in estimates of both R0 and the timing of vaccine availability . These analyses highlight the need for more precise statements of policy objectives and their assumed consequences when planning and implementing strategies to mitigate the impact of an influenza pandemic . In the event of the emergence of a new human influenza A strain with a high case fatality rate indicating the possibility of a global pandemic with severe impact , control strategies primarily aim at limiting morbidity and mortality rather than halting transmission completely . This is because transmission of influenza A is difficult to block due to its short generation time and efficient transmission characteristics [1] . In the early days of the H1N1 influenza pandemic in Mexico in 2009 [2] , social distancing measures were implemented with the aim of slowing the epidemic during its early stages . For any future pandemic of a directly-transmitted infectious agent , it is expected that similar strategies will be used in high resource settings while the pathogen is being identified , epidemiological studies to both characterize transmission [3] , [4] , [5] , [6] , [7] and determine pathogenicity are completed [8] , [9] and strain-specific control options , such as vaccines , are being developed [10] , [11] . For influenza , policy options are clearly outlined in national pandemic plans , but there is rarely any clear statement of policy objectives [12] . The problem is that these different objectives are potentially conflicting in their effects , and clear prioritisation is therefore necessary . Is the aim to minimize mortality and morbidity , is it to limit the peak prevalence of serious disease so that public health resources are not overwhelmed or is it to minimise the impact of the intervention on society and economy ? In this paper we form a framework for policy makers to consider these potentially conflicting objectives . A number of studies have investigated the role of targeted interventions at different phases of the epidemic based on mathematical models which include various levels of population structure and spatial complexity [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] . However , none of these studies have addressed how multiple policy objectives are met by the common interventions , or how a clear statement of the key policy aims guides which set of interventions work best . It is typically assumed by policy makers that the more intervention measures implemented as early as possible in the course of the epidemic the better the outcome in terms of mitigation . Reservations about this strategic approach rest on the costs , and societal impact plus economic implications of sustaining control measures over a long period of time . In recognition of this the USA pandemic plan , for instance , mentions a maximum duration of 12 weeks for many transmission-reducing interventions [21] . However , there has been no quantitative analysis of when such an intervention should be initiated . Should it be as soon as the first cases are discovered , or later in the outbreak when more cases have arisen ? Neither has it been acknowledged that planned levels of coverage with antiviral treatment or pre-pandemic vaccines may implicitly determine the magnitude of social distancing interventions required . Studies have shown that during the 1918–19 influenza pandemic public health control strategies and changes in population contact rates lowered transmission rates and reduced mortality and case numbers [22] , [23] . Similar measures were arguably effective for H1N1 in Mexico in 2009 [3] . Strategies used then , and to be considered in future , include social distancing measures , such as school closures [24] , [25] , restaurant and cinema closures [26] , and transport restrictions [27] , [28] , [29] . There are a number of other measures , such as hand washing and the use of face masks [30] , which may reduce contact rates [31] , [32] . Transmission will also be affected by changes in human behaviour in response to a pandemic , as was observed in travel and mixing patterns during the severe acute respiratory syndrome ( SARS ) outbreak in 2003 [29] . Within the last 100 years , there have been two international outbreaks of a directly transmitted pathogen with high case fatality rates in which social distancing measures were implemented . The first was the influenza pandemic of 1918 , where non-pharmaceutical public health strategies were effective at reducing morbidity and mortality in a number of settings [22] . However , the impact of these interventions on transmission was highly variable . An analysis of cities in the USA showed reductions in transmission ranged from approximately 0–60% ( Figure 1 ) . These interventions were held in place from 1 week to 3 months . One might expect that interventions with higher impact were held in place for shorter time , but there was no systematic relationship between the duration and the impact of interventions ( Figure 1 , black circles ) . During the SARS outbreak of 2003 , the aim of intervention strategies was to eliminate transmission , not only to mitigate the effects of the epidemic . Elimination was possible due to the characteristics of the virus - post-symptomatic transmission and a long generation time [1] . Large scale reductions in the transmission rate of SARS ( >70% , Figure 1 , [33] ) were brought about by a number of public health interventions . These interventions were held in place for several weeks . The small amount of data available perhaps suggest a trend towards lower impact interventions being held in place for longer to achieve elimination ( Figure 1 , open triangles ) , but an important driver of the duration of these interventions was the number of cases that were present when the interventions started . These empirical data from two severe outbreaks suggest that moderate reductions in influenza transmission can be achieved and maintained at a population level for a number of weeks . The impact of any particular intervention is difficult to estimate from past epidemics due to variation in the viral strain and its transmission properties , and due to the concurrent effects of many different behavioural responses and government led initiatives . Planning therefore depends increasingly on the predictions of mathematical models of viral spread that permit analyses of the potential impact of various interventions , alone or in combination [13] , [34] , [35] , [36] , [37] , [38] , [39] . In this paper we consider the effectiveness of contact-reducing interventions during the first six months after the initial cases , before a pandemic vaccine is available , and evaluate optimum interventions for a range of policy objectives or constraints , such as a limited stockpile of treatments or non-specific vaccine . Analyses are based on a mathematical model of virus transmission and the impact of control measures . We focus on the identification of policies that minimise peak demand for public health services and those which minimise the potential costs or socio-economic impact as evaluated by a simple cost function . This paper is not designed to give specific policy guidance . Box 1 outlines a number of factors which should be considered in designing policy which are not covered here . Our aim is to develop an understanding of how different policy objectives determine the optimal mix , timing of introduction and duration of implementation of the available mitigation strategies . All results have been obtained with a model based on the well-known deterministic SIR-model , that has proven its value in many studies of infectious diseases [40] , [41] . We parameterized the model with a mean infectious period of 2 . 6 days ( recovery rate γ = 1/2 . 6 ) , and a basic reproduction number R0 = 1 . 8 ( see Ferguson et al [13] ) with a population of size n = 58 . 1 million . The population was subdivided into proportions of the population in the classes of x susceptibles , y infectives , and z immunes , with dynamics given by ( 1 ) The parameter is the transmission rate , i . e . the number of contacts an infective has per day in which the infection is passed on , and has the baseline value . Simulations were started with 1 infective , n−1 susceptibles , and no immunes . We investigated the impact of a social distancing intervention on transmission through a constant reduction in transmission , , resulting from an unspecified combination of public health measures , maintained over a time period , D . In model terms , the transmission rate was assumed to change during intervention from the baseline rate to a reduced rate . This happened from , the start of the intervention , until , the end of the intervention of duration D . For the duration we considered three options , first an intervention that is kept in place indefinitely , second an intervention with a fixed duration of twelve weeks , which is the maximum duration mentioned in the USA national pandemic plan [21] , and third an intervention until the a pandemic specific vaccine is available , after six months . In the ‘indefinite’ scenario , the duration of the epidemic was formally defined as the time until . Transmission-reducing public health interventions for influenza are unlikely to completely halt transmission [22] , [23] , [42] , [43] . It is most likely that mitigation strategies will be ‘sub-critical’ interventions which reduce the effective reproduction ratio ( the mean number of new infections per infected individual ) towards , but not below , 1 . Thus , we assumed that . Numerical simulations of the model were used to evaluate the impact of the interventions twelve months after the first case . Impact is primarily measured by ( the reduction in ) the total number of cases . We also evaluated two other measures of effectiveness: firstly , the ( reduction in ) peak prevalence , since high prevalence may overwhelm public health facilities and as such increase both morbidity and mortality; and secondly , the socio-economic costs of the interventions , determined both by the level of intervention and the duration they are in place , calculated as the simple cost function . Many countries have stockpiled antiviral drugs in preparation for an influenza pandemic [12] . Whilst these may be used prophylactically to reduce transmission [35] , [44] , [45] , most pandemic strategies advocate the use of antivirals to treat cases of infection or to treat those cases where other risk factors suggest that disease severity may be high [46] . The treatment of cases will reduce morbidity and mortality and has been shown to be cost-effective for high risk patients [47] . We focus on the treatment of cases in combination with transmission-reducing intervention as above . We make the assumption that treatment of cases does not affect transmission . The assumption is made firstly because drugs are given upon case notification , which is when much infectiousness may have passed [1] , and secondly because symptomatic patients will be advised to remain at home reducing their contacts . The additional transmission reduction in transmission due to antivirals will thus be minor . The use of antivirals for severely ill patients could have implications for occupancy and therefore availability of isolation units and high dependency beds . Whilst this might change the infectious profile of the few severely ill patients who would have access to these facilities , it does not affect the majority of cases and detailed consideration of these logistics is outside the scope of this study . In addition , we do not include the possible effect of mass treatment on resistance [48] and therefore on the efficacy of the drugs . Consideration of these effects may lead to a range of different policy objectives , taking into account combination therapy or sequential deployment of different lines of therapy [49] . As well as stockpiling antivirals , it may be possible to reduce transmission and severity of disease by stockpiling a partially-protective pre-pandemic vaccine in advance of the pandemic [12] . Even partially effective vaccines can have large beneficial effects because the unvaccinated are indirectly protected from infection by those portions of the vaccinated population who are not infected or are less severely affected and possibly have reduced infectiousness ( ‘herd’ immunity – see [40] ) . Use of an imperfect vaccine can , however , also lead to increased incidence if reductions in infectiousness are associated with corresponding increases in the infectious period [50] , [51] . Effectiveness estimates for a pre-pandemic vaccine are not available , but evidence from cross-protection studies led to the assumption that both susceptibility to infection and infectiousness may be reduced by 30% [13] , [52] . The duration of infectiousness is assumed to be unchanged , precluding any increased incidence in the presence of the vaccine . We evaluate a partial vaccination strategy , in combination with a transmission reducing intervention , aiming to keep the number of unvaccinated cases ( epidemic size ) less than 25% of the population . To consider vaccination with a pre-pandemic vaccine , the transmission model was adjusted to include infection of vaccinated individuals: ( 2 ) In this adjusted model , , , and are the proportion of vaccinated individuals , and ( = 0 . 7 ) and ( = 0 . 7 ) are the relative infectiousness and susceptibility of vaccinated versus unvaccinated individuals . It is assumed that vaccinated cases would not require treatment , and therefore were not included in the epidemic size or peak prevalence . Simulations were carried out with a vaccine coverage of 10% , starting with one unvaccinated infective . To place our results in a more realistic context whilst not giving precise policy guidance , we consider two scenarios for pandemic planning in high resource settings . They are scenarios which are covered in a number of pandemic plans . We will outline the range of interventions which can achieve these aims . Scenario 1: A strain-specific vaccine is expected to be available within 6 months of the start of a pandemic . In order to minimize morbidity and mortality , social-distancing interventions will be used to ‘buy time’ until the vaccine is available . Antiviral drugs are available to treat symptomatic cases with a stockpile for up to 25% of the population . Social-distancing interventions will be used to ensure that symptomatic cases are kept below this level and to minimize socio-economic impact and peak demand for hospital and other public health services by minimizing prevalence in the population . Scenario 2: This scenario is very similar to Scenario 1 , except that in addition a pre-pandemic vaccine is available which can be rapidly rolled out to 10% of the population . The question of interest will be the extent to which the pre-pandemic vaccine will reduce the level of intervention required . Since we are considering interventions implemented early in the epidemic , key epidemiological parameters may still be in the process of being estimated . Therefore , we investigated which strategies are least sensitive to incorrect estimation of R0 , i . e . R0 = 1 . 7 or 2 . 0 . In addition , availability of a pandemic vaccine may be delayed , or the pre-pandemic vaccine may be less effective than anticipated , so we ran our simulations out to an eight-month period and with a vaccine efficacy of ( 50% less reduction in transmission ) . One possible policy choice is to maintain an intervention irrespective of cost until the last case has recovered from the disease . This will always reduce the total number of cases and peak prevalence . These quantities can be expressed or approximated by analytical expressions , which we derive in Text S1 and illustrate using numerical simulations . The final proportion of the population affected by an unconstrained epidemic , aNI , is given by solving [40] , [41] ( 3 ) The final size increases monotonically with increasing R0 and does not depend on the generation time of the infection [40] . For a long term intervention , implemented at T1 and held in place until there are no cases ( Figure 2 ) , the final epidemic size , ( proportion of the population who have been infected ) is given by ( 4 ) where is cumulative incidence up to time T1 . In the exponential growth phase , the cumulative incidence can be approximated by ( 5 ) where r is the epidemic growth rate , given by . For our parameter values , this approximation works well until about T1 = 49 days ( 7 weeks ) , when equation ( 5 ) overestimates by 22% . The final epidemic size decreases monotonically as the timing of the intervention , T1 , becomes earlier , and as the size of the intervention , , becomes larger ( Figure 2 ) . However , before week 5 is very small , so interventions starting earlier do not have much effect ( Figure 2 ) . In the absence of an intervention the maximum prevalence occurs when , or when , and the maximum prevalence is ( using the equations above and approximations to the initial conditions ) is [40] ( 6 ) which increases with increasing R0 , and , as with the unconstrained epidemic size , does not depend on the generation time . In the presence of the intervention , maximum prevalence is dependent on the proportion of the population who are still susceptible at the time of the intervention . If the intervention is initiated before the peak in the unconstrained epidemic , and if cumulative incidence is sufficiently high and the proportion of the population still susceptible at the start of the intervention is less than , then peak prevalence will be at the start of the intervention , . On the other hand , if the cumulative incidence is less than there will be a peak during the intervention ( Figure 2 ) , which is given by ( 7 ) If the intervention is initiated after the peak of the unconstrained epidemic , then there will not be another peak in prevalence during the intervention , since there will be too few susceptible individuals . These analytical results can be used to understand the effect of an intervention on the final size and peak prevalence , but we do not have neat expressions for the resulting duration of the whole epidemic ( time until final case recovers ) when an intervention is in place , and therefore we turn to simulation ( Figure 2 ) . The higher the transmission rate , the shorter the epidemic , which may be a desirable policy outcome . For influenza-like parameters , a few weeks delay may have only moderate deleterious consequences for peak prevalence , peak incidence or epidemic size ( Figure 2 ) . This delay will result in higher peak prevalence , but it will also result in a considerably shorter epidemic than an early intervention ( Figure 2A circular inset and 2B ) . This may be a desirable outcome in economic terms . The level of reduction in transmission has similar effects , where a more effective intervention put in place early in the epidemic will lead to the smallest epidemic size and peak prevalence , but the longest epidemic duration ( Figure 2C and D ) . In brief , the earlier a long term intervention is put in place and the more effective it is at reducing transmission , the greater the beneficial effect in terms of total epidemic size and peak prevalence . Interventions of this kind are likely to be the most costly , and , counter-intuitively , may have to be held in place the longest . A strong argument to start an intervention early , however , is that the epidemic peak occurs later for early interventions ( Figure 2A ) , allowing time to prepare public health facilities , to manufacture a strain specific vaccine and because there is great uncertainty about severity in the early stages of an outbreak [8] . The drawbacks of a long intervention period are recognised in the USA national pandemic plan , where a maximum duration of 12 weeks intervention is anticipated - another policy choice we considered . As above , we first consider some analytical expressions , and illustrate them using numerical simulation . For a single short term intervention from to , the final epidemic size , , is given by ( 8 ) Note that , although can still be approximated during the exponential phase of the epidemic ( equation ( 5 ) ) , we cannot approximate . In this case , the relationship between the final epidemic size and intervention parameters is more complex because cumulative incidence at the time the intervention is lifted depends both on cumulative incidence at the time the intervention is initiated and the size of the intervention , . For example , if the duration of the intervention and its starting time are fixed , the epidemic size is optimized for intermediate values of the size of the intervention , ( Figure 3B , D ) . With a short-term intervention , there are three possible maximum prevalence points . Firstly , prior to the intervention ( equation ( 6 ) ) , during the intervention ( equation ( 7 ) ) , or after the intervention ( 9 ) ( note that ) . The peak value could also occur at the point at which the intervention starts , i . e . when . The conditions for each peak being the maximum are given in Table 1 . A large magnitude intervention ( large ) may actually be deleterious , leading to a larger resurgence in prevalence after the intervention than an intervention with a smaller reduction in transmission . With a short-term intervention , there is no longer a monotonic relationship between the policy outcomes and the magnitude and length of the intervention . Therefore strategies which contain the epidemic size below certain levels are unlikely to be the same interventions which contain peak prevalence below particular targets . For influenza-like parameters a 12-week intervention will almost certainly lead to a resurgence of the epidemic once the controls are lifted ( Figure 3A , C ) . If peak prevalence is very much lower during the intervention than it would be with no intervention , the implemented policy may even result in almost no change in the total epidemic size ( Figure 3 ) . For late , or less effective , interventions , prevalence during the intervention is higher than for early , or more effective , interventions , , resulting in fewer susceptible individuals remaining when the intervention is lifted . In this case the second peak is smaller , and reductions in total epidemic size are larger ( Figure 3 ) . For short term interventions , in contrast to long-term strategies , peak prevalence , peak incidence , and epidemic size cannot all be minimized by the same strategy . For instance , a 33% reduction in transmission timed to minimise total epidemic size ( Figure 3A , B , initiated week 7 ) may not be the intervention which minimises peak prevalence ( Figure 3A , B , initiated week 6 ) . Both these strategies have small and late resurgent epidemics ( Figure 3A , circular inset ) , with cases beyond the end of the year . Similarly , an intervention initiated at week 5 may minimise peak prevalence for a 33% reduction in transmission ( Figure 3C , D ) , or minimize epidemic size with a 22% reduction in transmission ( Figure 3C , D ) , but neither of these strategies are optimal if the aim is to have the epidemic exhaust itself most rapidly , with the quickest epidemic being the one without any intervention . The intervention always reduces peak prevalence from what it would have been in the absence of an intervention . However , which particular value is the peak value is determined by the timing of the intervention and the magnitude of the intervention ( Table 1 ) . Each of these vary according to the characteristics of the intervention , and the underlying epidemic . For a fixed starting time and duration , there is a non-linear relationship between peak prevalence and the reduction in transmission , ( Figure 3 ) . The value of for which peak prevalence is minimized is almost certainly not that at which the total epidemic size is minimized ( Figure 3 ) . It is not possible to achieve a symptomatic epidemic size of 25% of the population with a 12 week intervention for these parameter values . We therefore consider a scenario in which an intervention is initiated in the first weeks or months of the outbreak and held in place until 6 months after the start of the outbreak . Many different interventions can be used to constrain the epidemic size to 25% of the population . They range from an early intervention with a mild reduction in transmission , to a late , more impactful intervention ( Figure 4A ) . To achieve this aim whilst minimising peak prevalence it is not necessary to initiate the intervention early , in fact a delay may even be beneficial ( Figure 4B ) . But , the intervention must start before 7 weeks ( for these parameter values ) , when the number of cases prior to the intervention becomes large . If we evaluate the socio-economic ‘cost’ of these interventions as a simple product of the duration of the intervention and the reduction in transmission achieved , a delay also reduces the costs of the intervention , and the ideal intervention is more clearly defined ( Figure 4B ) . Delay is valuable because transmission is being reduced , not eliminated , and therefore some of the effort in constraining the epidemic at the early stages is redundant . Choices about intervention policy will be made early in the epidemic when parameters are uncertain . For example , R0 and the date of availability of the vaccine could be over or under estimated . Of course , designing this intervention based on an overestimate of R0 means that the epidemic is smaller than expected , and so the intervention is too large and there are fewer cases overall ( Figure 4C ) . An underestimate in R0 means that the epidemic is larger than expected and so the intervention is not large enough to contain the epidemic and there are more cases than expected ( Figure 4C ) . In either of these cases , the intervention would have to be adjusted during the outbreak . If the ‘optimum’ intervention , which minimised peak prevalence , is chosen , it is more robust to changes in R0 than the other options ( Figure 4C ) . A delay in the availability of vaccine increases the number of cases , but picking a late intervention minimises this effect . Use of an imperfect vaccine for only 10% of the population results in a slower epidemic with fewer cases ( Figure 5 ) . The use of a pre-pandemic vaccine means that interventions which contain the total number of cases and peak prevalence can be rolled out later ( Figure 5A ) , compared to the non-vaccination scenario . Also , as can be seen from the simple cost function ( Figure 5B ) , the level of intervention can be reduced if pre-pandemic vaccines are used . The true economic value of this reduction in costs depends on the relative costs of vaccination , cases and interventions . The general picture remains the same as without vaccination . To minimize peak prevalence , the intervention should be initiated earlier than to minimize costs , but both objectives require interventions that commence several weeks into the epidemic growth phase ( Figure 5B ) . Sensitivity to the value of R0 or the effectiveness of the pre-pandemic vaccine highlights that once again the most robust strategies are those that are minimize peak prevalence ( Figure 5C ) . In the absence of detailed analyses , it is often argued that epidemic outbreak control is best achieved by putting all mitigation options into play as early as is feasible . There may be delays before control strategies are implemented due to difficulties in identifying the early stages of a novel outbreak [53] , as well as other logistical , political and economic constraints . Of course , if interventions are held in place until a pandemic vaccine is available a greater level of reduction and earlier start of intervention will result in fewer cases , and a lower peak prevalence and incidence if intervention starts before the peak . However , not only are the costs of an intervention held for a long time likely to be high , but high demand for health services will be extended over a longer time period . Our results indicate that an intervention starting at a few weeks into the epidemic is almost as effective at reducing epidemic size and peak prevalence as one starting at week 0 . As such , given that the social and economic burden will be greater when starting earlier , starting a little bit later may be a better policy option . However , this will crucially depend on the socio-economic costs of both cases and interventions and on the estimated severity of the epidemic , which may be uncertain in the early stages of the epidemic [8] . As noted in the introduction , the drawbacks of a long intervention period are recognised in the USA national pandemic plan , where a maximum intervention duration of twelve weeks is anticipated [21] . Using a twelve-week intervention , we have illustrated how the introduction of a short term intervention complicates the dynamics and increases the potential for conflict between policy aims . Interventions of limited duration are very likely to result in a resurgence of the epidemic once they are lifted , unless it is imposed late in the epidemic or with low effectiveness . However , the height of this resurgence can be managed . A twelve-week interventions minimizing peak logistical pressure ( peak prevalence and incidence ) need not be very strong but require a timely start . On the other hand , an intervention that minimizes total epidemic size needs to be stronger and can start later , preventing a second peak . A number of American cities experience a second peak in mortality following the lifting of interventions during the 1918 pandemic [22] , [43] . Re-analyses of a number of cities showed that multiple interventions were more effective at controlling transmission than single interventions [43] . In addition , it was found that the later multiple interventions were implemented , the less effective they were in reducing mortality [22] , [43] . This was most notable when controls were implemented when excess mortality was higher than ∼100 per 100 , 000 [22] . This conclusion cannot be so easily drawn in epidemics for which interventions were initiated prior to this threshold [22] . Here , we have shown that for short term interventions implemented during this early part of the epidemic earlier commencement is not always better , and that the outcome is highly sensitive to the timing and effectiveness of interventions . Our two scenarios for policy design illustrate that applying one objective and then another sequentially ( e . g . limiting total cases and then minimising peak prevalence for that epidemic size ) can be used to resolve potentially conflicting aims . Our results also show that the most extreme and earliest mitigation interventions are not always the best , and not always the least costly . It has not previously been highlighted that the level of stockpiles will quantitatively affect the required magnitude of social-distancing interventions so that all those who require treatment will receive it . Any level of stockpiled antiviral drugs will reduce morbidity and mortality and therefore reduces the need for transmission-reducing interventions , as not all cases need to be prevented , but the availability of drugs means that demand for these drugs should not exceed supply . In addition , our results illustrate that even low coverage with imperfect vaccines can lead to reductions in the required interventions level to meet a defined objective for control . There are many complexities involved in quantifying the effect of interventions which are not included here , the complexities of transmission by age and spatial heterogeneities , the likely behavioral changes during an epidemic that affect transmission , seasonal variation in transmission , the logistics of delivery of pre-pandemic vaccines and drugs , the economic costs of an outbreak and potential development of resistance to antiviral drugs . Detailed investigations are required to tailor general policies to particular settings , and therefore we are not attempting to make quantitative policy recommendations ( see Box 1 ) . However , uncertainties with regard to characteristics of the next pandemic strain will make it difficult in general to do very detailed optimization analyses . Decisions on stockpiling must be based on knowledge from previous pandemics and seasonal influenza , but when a pandemic is at hand one has to work with the stockpiles available . Intervention measures can be additionally imposed if a shortage of drugs is expected , or lifted to reduce the impact of intervention on society and economy , if drug supplies permit . Our analyses show that there is indeed some time to choose the appropriate level of control , as very early commencement of intervention is hardly ever optimal for these time-limited interventions . Our analyses also illustrates that even a simple inclusion of ‘costs’ changes what is optimal by comparison with analyses that are just based on impact on epidemiological measures . Economic costs typically enter the equations in a non-linear term as indicated in our model formulation . However , including empirically derived cost functions will probably lead to the inclusion of more highly non-linear functions . This highlights the need to include more robust economic constraints into future epidemiological model analyses for public health policy support . In our view , this is a more urgent need than that of increasing the complexity of epidemiological description within models of infectious disease control . Concomitantly , there is the associated need for measurement of the appropriate cost functions . Data is available for both drug and vaccine purchase but this is regarded as confidential at present as neither the pharmaceutical industry nor government health departments are keen to say how much was paid per dose as a function of total volume purchased . Future research must address the detail of cost and benefit , both in terms of measurement of direct and indirect socio-economic costs , the costs of stockpiling and the benefits of reducing the impact of the epidemic and in terms of using a template for analysis that reflects the dynamics of virus transmission and the impact of control measures . In our model we have considered contact-reducing interventions , the use of antiviral medication , and vaccination with a pre-pandemic vaccine . For insight into the effect of other control options , it is useful to understand what characterizes these three particular control measures . Antivirals work on the individual level , contact reduction on the population level , and vaccination on both . Contact reduction and vaccination are preventive measures , whereas treatment is reactive . Treatment and vaccines require stockpiling , and both are flexible with respect to possible timings of introduction during the epidemic . Contact reduction is flexible in both planning and timing , but has major implications for the normal functioning of society . This flexibility implies that a broad range of more complex strategies could be envisaged , for example implementing and lifting a hierarchy of controls in response to the dynamics of the epidemic and importation of cases . However , the simple scenarios illustrated here highlight the complexities in selecting the best intervention policy , in terms of magnitude , timing and duration of interventions . The optimum intervention in terms of minimising peak logistical pressures ( peak prevalence or incidence ) , may not be the same as one which minimises total epidemic size , and will almost certainly not be the one minimising direct social or economic impact from the intervention itself . The aims of a public health intervention policy must therefore be clearly defined , so that in the early phase of a pandemic sufficient resources can be put into characterizing the virus strain and measuring key epidemiological parameters as an essential template for decisions on what is the optimal mitigation strategy .	In the event of an influenza pandemic which has high mortality and the potential to spread rapidly , such as the 1918–19 pandemic , there are a number of non-pharmaceutical public health control options available to reduce transmission in the community and mitigate the effects of the pandemic . These include reducing social contacts by closing schools or postponing public events , and encouraging hand washing and the use of masks . These interventions will not only have a non-intuitive impact on the epidemic dynamics , but they will also have direct and indirect social and economic costs , which mean that governments will only want to use them for a limited amount of time . We use simulations to show that limited-time interventions that achieve one aim , e . g . , contain the total number of cases below some maximum number of treatments available , are not the same as those that achieve another , e . g . , minimize peak demand for health care services . If multiple aims are defined simultaneously , we often see that the optimal intervention need not commence immediately but can begin a few weeks into the epidemic . Our research demonstrates the importance of tailoring pandemic plans to defined policy targets with some flexibility to allow for uncertainty in the characteristics of the pandemic .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "public", "health", "and", "epidemiology/infectious", "diseases", "infectious", "diseases/epidemiology", "and", "control", "of", "infectious", "diseases" ]	2011	Mitigation Strategies for Pandemic Influenza A: Balancing Conflicting Policy Objectives
Despite the described central role of jasmonate signaling in plant defense against necrotrophic pathogens , the existence of intraspecific variation in pathogen capacity to activate or evade plant jasmonate-mediated defenses is rarely considered . Experimental infection of jasmonate-deficient and jasmonate-insensitive Arabidopsis thaliana with diverse isolates of the necrotrophic fungal pathogen Botrytis cinerea revealed pathogen variation for virulence inhibition by jasmonate-mediated plant defenses and induction of plant defense metabolites . Comparison of the transcriptional effects of infection by two distinct B . cinerea isolates showed only minor differences in transcriptional responses of wild-type plants , but notable isolate-specific transcript differences in jasmonate-insensitive plants . These transcriptional differences suggest B . cinerea activation of plant defenses that require plant jasmonate signaling for activity in response to only one of the two B . cinerea isolates tested . Thus , similar infection phenotypes observed in wild-type plants result from different signaling interactions with the plant that are likely integrated by jasmonate signaling . Jasmonate-mediated signaling controls diverse aspects of plant growth and defense . In particular , jasmonate signaling exerts a major influence on plant response to wounding , chewing insects , and necrotrophic pathogens such as Botrytis cinerea , Alternaria brassicicola , Plectosphaerella cucumerina , and Sclerotinia sclerotiorum [1]–[6] . Appropriate plant responses to these diverse stimuli are believed to be tailored by cross-talk between jasmonate and other hormone signals , such as salicylic acid ( SA ) , ethylene , and abscisic acid ( ABA ) [7]–[14] . Jasmonate signaling therefore does not mediate plant defense in isolation , but as part of a network of signals with the potential for positive and negative interactions . These signals include inputs from the pathogen that may influence the plant's defense response with positive or negative outcomes for the plant . Two major pathogen classes are roughly delineated by the pathogen's “lifestyle”: biotrophic pathogens infect living host cells and necrotrophic pathogens kill cells prior to consuming them [15]–[17] . This difference in the pathogen's mode of attack strongly influences which signaling networks mediate the plant response . Plant responses to biotrophic pathogens are largely mediated by salicylate signaling with an emphasis on specific recognition of pathogen effectors by the products of plant resistance ( R ) genes , often characterized by nucleotide binding sites and leucine-rich repeats [18] , [19] . Plant responses to necrotrophic pathogens appear to be mediated by a complex web of signaling dominated by jasmonates and ethylene [20]–[22] . Specific recognition of necrotrophic pathogens by the products of plant R genes is currently unknown , although recent identification of a gene possessing structural similarities to R-genes as the molecular basis of a quantitative trait locus ( QTL ) affecting resistance of Arabidopsis thaliana to multiple necrotrophic and hemibiotrophic pathogens has been suggested to link mechanisms of defense against biotrophic and necrotrophic pathogens [23] . While plants respond to biotrophic and necrotrophic pathogens via different signaling systems , these systems activate common defense responses , such as the production of the A . thaliana defense metabolite , camalexin . Thus , common responses may be controlled by distinct regulatory networks . The simplified statement that biotrophic and necrotrophic pathogens activate distinct , but overlapping , defense signaling pathways is largely based on observation of single genotypes of the respective pathogens . Yet biotrophic pathogen species exhibit considerable variation in activation of plant defense signaling . This biotroph variation is largely associated with diversity in the R-gene mediated specificity of plant-pathogen recognition , a phenomenon not documented for necrotrophic pathogens [24]–[26] . Examples of naturally occurring intraspecific pathogen variation affecting plant defense against necrotrophs include variation in toxin production by pathogens and variation in pathogen tolerance or detoxification of plant-produced defense compounds [27]–[31] . While activating plant defense signaling should logically hinder infection , pathogens may manipulate plant defense signaling to improve pathogenesis by diverting plant resources toward defense strategies that are less effective against , or actually increase sensitivity to , the pathogen . Pathogens are known to produce plant hormones or analogues such as coronatine , gibberellins or ABA , and the production of these compounds has been associated with virulence [32]–[34] . Interestingly , the ability to produce these compounds may vary among isolates of the same pathogen species as shown by a survey of 95 strains of Pseudomonas syringae where only 15% assayed positively for coronatine production [35] . While production of ABA by pathogenic fungi has not been as extensively assayed , ABA-overproducing and ABA-deficient B . cinerea strains have been described [36] . In addition , some B . cinerea isolates produce ethylene [37] . Thus , while elements of plant defense signaling may be associated with resistance to particular pathogens , pathogen variation in activation , manipulation , and response to plant defense signaling may alter these associations . Despite available literature suggesting that B . cinerea natural diversity could impact plant defense signaling , this diversity has not been routinely integrated into studies of plant—pathogen interaction . Unlike many pathogens that possess shorter or longer biotrophic stages , B . cinerea is identified as an unambiguously necrotrophic pathogen [17] , [20] . This ascomycete fungus occupies broad geographic and host ranges and exhibits a high degree of genetic and phenotypic variability [38]–[40] . However , this variation has been little explored in the context of plant defense signaling . Testing the interaction between a collection of B . cinerea isolates and A . thaliana mutant genotypes with defined deficiencies in jasmonate signaling revealed significant variation in plant response to B . cinerea isolates that was not apparent in wild-type plants . This included variation in lesion phenotype , altered mRNA transcript accumulation responses , and variation in accumulation of the A . thaliana defense metabolite camalexin . An unexpected dependency of camalexin accumulation in response to B . cinerea infection on intact jasmonate signaling was also revealed . The results presented here , while not contradicting the accepted view that jasmonate-mediated defense is vital for plant resistance to B . cinerea , suggest that additional pathways modulate A . thaliana—B . cinerea interactions . Finally , the architecture of plant defense signaling networks that provide resistance to necrotrophic pathogens is not static , and will vary with the pathogen genotype investigated . To test effects of jasmonate-mediated plant defense on diverse B . cinerea isolates , A . thaliana leaves of the aos genotype ( deficient in jasmonate biosynthesis ) and its corresponding wild-type were inoculated with 10 diverse B . cinerea isolates , two abiotic elicitors ( acifluorfen and AgNO3 ) , or a mock inoculation ( Table 1 ) [41] . Visible initiation of leaf necrotic lesions was observed between 24 and 48 hours post inoculation with B . cinerea . While tissue necrosis of aos plants initiated within a time frame similar to wild-type plants , lesions expanded more rapidly in aos plants , with near total consumption of the leaf by B . cinerea between 72 and 96hpi . aos mutant leaves failed to develop the zone of chlorosis surrounding the developing lesion that is often observed in B . cinerea infections ( Figure 1A ) . A comparison of camalexin accumulation in wild-type versus aos leaves induced by 10 B . cinerea isolates revealed significant diversity ( Figure 2 ) . Among the B . cinerea isolate treatments tested , camalexin accumulation in aos leaves ranged from 5% to 50% of camalexin accumulation in wild-type leaves , with a median camalexin accumulation among B . cinerea infections of 14% wild-type levels . Mock treatment , acifluorfen , and AgNO3 induced camalexin in aos leaves at 5–7% wild-type levels . In no case was the absence of jasmonate synthesis associated with increased camalexin accumulation . To explore the observed pathogen variation in interaction with jasmonate-deficient genotypes and activation of metabolic defense , the two B . cinerea isolates inducing camalexin accumulation in the aos leaves at the highest and lowest levels relative to wild-type leaves , BcGrape ( 5% ) and Bc83-2 ( 50% ) were used for further experiments . One hypothesis that could explain the differential accumulation of camalexin in BcGrape and Bc83-2 infected jasmonate-deficient plants is that one of the B . cinerea isolates produces a molecule that stimulates the intact jasmonate perception in the A . thaliana aos mutant . To determine whether plant deficiencies in jasmonate synthesis and jasmonate perception create similar infection phenotypes and show fully overlapping effects on plant defense signaling , we generated a double mutant containing both aos and the coronatine-insensitive 1 ( coi1 ) mutation that confers deficiency in jasmonate perception [42] . A population segregating both coi1 and aos mutations was experimentally infected with BcGrape and Bc83-2 . coi1 aos double mutant plants displayed infection phenotypes for both tested isolates that did not differ significantly from those observed in either the single mutant coi1 or aos plants ( Figure 3 ) . Both the coi1 and aos mutations appear recessive for these phenotypes , as infection phenotypes of plants heterozygous for either or both mutations tested did not differ significantly from homozygous wild-type plants ( data not shown ) . The similarity of coi1 and aos phenotypes suggested that camalexin accumulation in jasmonate-deficient plant genotypes infected with Bc83-2 is not likely mediated by isolate-specific production of a metabolite with jasmonate-like coi1 dependent activity similar to coronatine [43] . Testing this segregating population also showed that the glabrous ( gl1 ) mutation , present in the aos mutant background and thus segregating in the aos×coi1 F2 population , had no significant effect on lesion size or camalexin accumulation [44] . We additionally tested a downstream component of the JA pathway , utilizing JAZ1Δ3 mutant plants . These plants produce a modified version of the JAZ1 protein that confers a dominant jasmonate-insensitive phenotype . The JAZ1Δ3 mutant plants showed defects in B . cinerea mediated camalexin induction similar to aos and coi1 plants , but with a less-dramatic increase in lesion size ( Figure S1 ) . These defense responses showed similar B . cinerea isolate dependency to that observed in aos and coi1 . Thus , B . cinerea isolates vary in their stimulation of signaling networks within A . thaliana as demonstrated by the ability of Bc83-2 to induce moderate camalexin levels in the absence of a functional jasmonate signaling pathway ( Figure 3 ) . The A . thaliana Phytoalexin Deficient 3 ( PAD3 ) locus encodes a cytochrome P450 enzyme catalyzing the final steps of camalexin biosynthesis [45] . The increased susceptibility of pad3 mutants to necrotrophic pathogens has supported the conclusion that camalexin is an important defense against these pathogens [28] , [46] . We showed that camalexin accumulation depends in part upon an intact jasmonate signaling pathway ( Figures 2 and 3 ) . To evaluate the extent that increased susceptibility of jasmonate-insensitive A . thaliana genotypes is due to decreased camalexin accumulation in these mutants , we measured development of necrotic lesions and camalexin accumulation in experimentally-infected Col-0 ( wild-type ) , coi1 , pad3 , and coi1 pad3 double mutant plants ( Figure 4 ) . Lesion size at 72hpi did not differ between coi1 and coi1 pad3 plants , but both of these genotypes developed significantly larger lesions than pad3 single mutants , indicating that camalexin deficiency explains a significant fraction of , but not the entire increase in , susceptibility of jasmonate mutants to B . cinerea ( Figure 4A ) . As anticipated , pad3 and coi1 pad3 plants did not accumulate measurable amounts of camalexin ( Figure 4B ) . This observation shows that camalexin accumulation in jasmonate mutants infected with Bc83-2 is not due to a previously-undescribed camalexin biosynthetic capacity in B . cinerea . Further , the similarity in lesion size between pad3 mutant plants infected with BcGrape and Bc83-2 suggests that the difference in susceptibility of jasmonate mutants to these two isolates is not explained by camalexin accumulation in jasmonate mutants infected with Bc83-2 . While BcGrape and Bc83-2 induced similar levels of necrosis on wild-type and pad3 plants , lesions produced by Bc83-2 on coi1 and coi1 pad3 plants were significantly smaller than those produced by BcGrape , supporting our observations that jasmonate deficiency had comparatively less impact on plant susceptibility to Bc83-2 ( Figures 2 and 4 ) . Consistent with previous experiments , coi1 plants infected with BcGrape accumulated extremely low levels of camalexin that did not significantly differ from levels accumulated in pad3 mutants , and coi1 plants infected with Bc83-2 accumulated camalexin at levels significantly lower than wild-type but significantly greater than pad3 mutant plants ( Figure 4 ) . In combination , this shows that while camalexin is a large component of the jasmonate-mediated defense against B . cinerea , its accumulation does not explain the differential virulence of Bc83-2 and BcGrape on jasmonate-deficient A . thaliana . To determine whether observed differences in camalexin accumulation and lesion growth between B . cinerea treatments were associated with differences in the timing of plant response , time course experiments were conducted using wild-type ( COI1/COI1 ) and coi1 mutant plants ( Figure 5 ) . B . cinerea isolates BcGrape and Bc83-2 produced similarly-sized necrotic lesions on wild-type leaves at 48 hpi , but lesions produced by BcGrape infection of coi1 leaves rapidly expanded starting at 40–48 hpi . Bc83-2 showed an increase in induced necrosis on coi1 leaves that was less dramatic than shown by BcGrape but still significantly larger than necroses formed on wild-type leaves . By 32 hpi , camalexin was significantly induced in wild-type but not coi1 leaves ( Figure 5B ) . Camalexin accumulation at all time points after 24 hpi was highest in wild-type leaves infected with Bc83-2 . coi1 infected with Bc83-2 showed consistently higher levels of camalexin than coi1 infected with BcGrape . Thus , the difference in camalexin response or virulence between Bc83-2 and BcGrape does not appear to be solely an issue of infection timing but rather variation in pathogen interaction with the plant . To explore mechanisms controlling altered accumulation of camalexin in jasmonate deficient plants as well as differences between B . cinerea treatments , we examined transcript levels of PAD3 and CYP71A13 . These genes encode enzymes which catalyze respectively the first committed step and the final steps in camalexin biosynthesis [45] , [47] , [48] . Relative levels of PAD3 and CYP71A13 transcripts were measured at 24 and 48 hours post-inoculation , time points flanking the observed onset of camalexin accumulation ( Figure 5B ) . PAD3 transcript levels were low but detectable at 24 hours post inoculation ( Figure 6A ) . At 48 hpi , all B . cinerea treated samples showed significantly increased PAD3 transcript accumulation compared to mock treatments . Samples from coi1 mutants showed less induction of PAD3 than wild-type samples but the reduction was not commensurate with the observed decrease in metabolite accumulation . While camalexin accumulation was nearly abolished in coi1 infected with BcGrape , PAD3 transcript was reduced by only half . Further , Bc83-2 infection is associated with relatively higher camalexin accumulation in coi1 , but significantly lower PAD3 transcript accumulation in coi1 compared to BcGrape infected coi1 . CYP71A13 transcript accumulation showed a similar pattern ( Figures S2 and S4 ) . Lack of correlation between PAD3 transcript accumulation and camalexin accumulation measured from the same tissue pool contrasts with previous reports that PAD3 transcript and camalexin accumulation are highly correlated ( Figure 6B ) [45] . B . cinerea infection with diverse isolates thus reveals evidence of additional regulation of camalexin biosynthesis , beyond transcriptional regulation of known biosynthetic genes . As camalexin accumulation during B . cinerea infection occurs primarily within the plant tissue immediately bordering the developing lesion , it is possible that the spatial distribution of camalexin biosynthetic transcript within an infected leaf may be more relevant to camalexin accumulation than total transcript accumulation within a leaf [30] . To visualize effects of jasmonate insensitivity and B . cinerea isolate differences on the pattern of transcript accumulation of the camalexin biosynthetic enzyme CYP79B2 , we crossed a CYP79B2 promoter-GUS fusion transgene into a coi1 background . CYP79B2 catalyzes the conversion of tryptophan to indole-3-acetaldoxime during camalexin biosynthesis in planta [45] , [49] . Leaves from homozygous wild-type and coi1 plants showing GUS activity were inoculated with B . cinerea isolates BcGrape and Bc83-2 . Wild-type leaves infected with either B . cinerea isolate showed blue staining indicative of GUS activity in a narrow zone bordering the lesion , consistent with previous studies showing that camalexin accumulates primarily within this zone ( Figure 1 ) [30] . coi1 leaves showed a dramatic difference in staining pattern between BcGrape and Bc83-2 infections , with BcGrape-infected coi1 leaves showing patterns of GUS activity similar to those seen in wild-type plants , and Bc83-2 infected coi1 leaves showing intense blue staining within the area visually defined as the necrotic lesion . This intense staining was not associated with increased accumulation of CYP79B2 transcript in coi1 leaves infected with Bc83-2 ( Figure S3 ) . The presence of the ProCYP79B2:GUS transgene did not significantly affect camalexin accumulation compared to plants without the transgene from the same segregating F2 population . A possible explanation for the above observation is that there is less cell death within the Bc83-2 lesion in comparison to BcGrape . We stained infected leaves with a vital stain , Trypan Blue , to compare patterns of cell death associated with infection by the two isolates on wild-type and coi1 leaves . This showed similarly sized halos of plant cell death surrounding the BcGrape and Bc83-2 lesions on both wild-type and coi1 leaves that was a lighter color in the coi1 lesions ( Figure 7 ) . Interestingly , these areas contained no detectable fungal cells , suggesting that plant cell death can be caused by mobile plant or fungal signals . No living or dead plant cells were visible within the hyphal mass , suggesting that B . cinerea rapidly consumes material in this region and that the observed difference in camalexin accumulation is not due to differential presence of plant cells . These results suggest that the observed GUS staining pattern is caused by persistence of plant-produced protein within the Bc83-2 lesion , rather than active transcription and translation from the plant genome within the Bc83-2 lesion , implying that the absence of a functional jasmonate signaling network alters the ability of Bc83-2 to degrade or disperse proteins . Trypan Blue staining also showed that the two isolates have different growth habits independent of the plant genotypes tested . Bc83-2 hyphae grew at higher density with a well-defined boundary to the hyphal mass , while BcGrape hyphae grew more sparsely with isolated probing hyphae that grow into the surrounding plant issue . We further compared the infection phenotypes of BcGrape and Bc83-2 using staining for H2O2 accumulation ( DAB ) . On wild-type A . thaliana leaves , infection by either tested B . cinerea isolate was associated with diffuse H2O2 generation within and around the lesion , suggesting that both the plant and fungus generate H2O2 . In contrast , Bc83-2 caused a strong halo of H2O2 surrounding the developing lesion on coi1 whereas the BcGrape lesions were associated with a H2O2 accumulation pattern similar to that observed in wild-type leaves ( Figure 7 ) . As generation of reactive oxygen species , including H2O2 , is associated with production of camalexin , the observed pattern of H202 accumulation supports our earlier observation that Bc83-2 induces camalexin via a jasmonate-independent mechanism that is lacking in BcGrape infections . Interestingly , this staining also showed that infection by BcGrape is associated with a systemic accumulation of H2O2 in trichomes that was independent of plant jasmonate perception and not seen in leaves infected with Bc83-2 ( Figure 7 ) . These B . cinerea isolates elicit distinct defense responses from plants that include both jasmonate-dependent and jasmonate-independent phenotypes , suggesting both the danger of oversimplifying models of plant—“B . cinerea” interaction and the rich potential of intraspecific studies of this pathogen . To identify additional differences in plant transcriptional response to these two B . cinerea isolates and build hypotheses regarding the molecular basis of differences in infection phenotype , whole-genome transcriptional profiles of A . thaliana leaves inoculated with B . cinerea isolates BcGrape or Bc83-2 were compared to each other and to control leaves using both wild-type and jasmonate-insensitive ( coi1 ) plants . Based on directed transcript measurements , where induction of camalexin biosynthetic and other defense-associated transcripts was not detected until 48hpi , transcriptional profiling was performed on samples from this 48hpi time point ( Figures 6 and S2 ) . Additionally , both B . cinerea isolates had initiated lesions by 48hpi , but lesions at this time point , arising from a single inoculation droplet per leaf , occupy only a small portion of the total leaf area and do not show the large differences in lesion size observed on coi1 leaves at later time points ( Figure 5 ) . Estimates of transcript accumulation obtained from arrays were highly consistent with targeted transcript measures obtained via quantitative RT-PCR , with significant Pearson correlation coefficients ranging from 0 . 76 to 0 . 91 ( Figure S4 ) . Array data are provided as Dataset S1 . Of 22810 transcripts represented on the arrays , over half ( 12 , 999 ) showed significant effects for the model transcript = genotype + treatment + ( genotype × treatment ) even after false-discovery adjustments . The majority ( 11 , 989 ) of these statistically significant transcript changes were associated with treatment where most of these transcripts differed between B . cinerea-infected and control leaves , rather than between leaves infected with the two B . cinerea isolates . We therefore describe statistically significant plant responses consistent between both pathogen isolates as responsive to “B . cinerea” . Transcript accumulation from 1458 genes of the B . cinerea-responsive loci identified above showed greater than 2-fold increase in response to B . cinerea infection , while transcripts from 1602 genes showed more than 2-fold decrease relative to control samples . Differences in transcript abundance between wild-type and coi1 plants as well as between B . cinerea-inoculated and control plants showed overlap with previous studies [50] , [51] . All known enzymes of the camalexin biosynthetic pathway were upregulated by B . cinerea infection , with CYP71A13 and PAD3 respectively showing 124-fold and 67-fold increases in B . cinerea infected leaves . An additional five transcripts contributing to biosynthesis of the camalexin precursor , tryptophan , were also upregulated in response to B . cinerea , but less dramatically than camalexin biosynthetic genes ( Table S1 ) . Other transcripts showing greater than 2-fold transcriptional effects of B . cinerea infection that have been previously identified as contributing to plant defense against fungal pathogens included a camalexin regulator ( PAD4 ) , the MYB transcription factor botrytis-susceptible 1 ( BOS1 ) , phenylalanine ammonia lyase ( PAL1 ) , polygalacturonase-inhibiting protein ( PGIP1 ) , and pathogenesis response proteins ( PR1 , PR4 , and PR5 ) ( Table S1 ) . Transcripts of PDF1 . 2a and VSP2 , considered markers for jasmonate signaling , were detected only at extremely low levels in both B . cinerea-infected and control leaves from coi1 plants , further supporting our conclusion that camalexin accumulation in jasmonate mutants infected with Bc83-2 is not attributable to isolate-specific pathogen-mediated jasmonate signaling independent of coi1 and aos ( Figures S2 and S4 ) [52] , [53] . Differences in transcript accumulation after infection by BcGrape or Bc83-2 were generally similar in direction of effect between wild-type and coi1 leaves but of greater magnitude in coi1 leaves . Of 824 transcripts showing differential accumulation in response to the two tested B . cinerea isolates , 787 show larger differences in coi1 leaves than wild-type ( Table S2 ) . While this correlates with lesion development at later time points , lesion sizes at 48 hours do not significantly differ among genotype×isolate combinations ( Figure 5 ) . To identify patterns in these transcript differences that might enhance our understanding of the biology of A . thaliana response to B . cinerea , we clustered these transcripts by similarity of normalized transcript levels . This identified two large groups of transcripts , those showing relative increases in transcript level in response to B . cinerea ( clusters 1–3 ) and those relatively decreased in B . cinerea-infected leaves ( clusters 4–6 ) ( Figure 9 , Table S2 ) . Subsequent clustering of transcript profiles for these loci by genotype and treatment suggested that BcGrape and Bc83-2 infections exert similar transcriptional effects on wild-type leaves . This contrasts with a dramatic difference in transcript patterns observed in coi1 samples , where Bc83-2 infected coi1 leaves showed transcript patterns similar to mock-inoculated samples while BcGrape infected coi1 were more transcriptionally similar to infected wild-type samples . This echoes the pattern observed for the putative DETOX network for A . thaliana response to B . cinerea , and suggests that jasmonate has opposing transcriptional effects on a set of genes in response to these two B . cinerea isolates ( Figure 8 ) . Wild-type ( Col-0 ) A . thaliana leaves showed similar responses to the two B . cinerea isolates used in these experiments ( Figures 1 and 3 ) . These included not only visual and biochemical symptoms ( leaf necrosis and camalexin accumulation ) , but also transcriptional responses to infection ( Figures 6 and S2 ) . Comparison with an earlier transcriptional profiling dataset revealed that an unnamed B . cinerea isolate showed similar effects to this experiment: of 7718 transcripts described as significantly responding to B . cinerea treatment , 6465 showed a significant effect of B . cinerea treatment in the experiments described here [51] . Of these , 6107 transcripts showed the same directionality of B . cinerea effect . Where the effects of the three B . cinerea isolates represented in these two datasets disagree , no single isolate appears to be an outlier . This suggests that , while B . cinerea isolates elicit different transcriptional responses from wild-type A . thaliana , comparison among datasets reveals a consistent transcriptional signature of B . cinerea infection . While infection of wild-type plants with genetically and phenotypically distinct B . cinerea isolates elicited very similar plant responses , infection phenotypes displayed by jasmonate-deficient plants indicate that the phenotypic similarity observed in wild-type plants must be produced by different mechanisms . In particular , the isolate Bc83-2 induces camalexin accumulation both via jasmonate signaling and an additional pathway that is either not induced or specifically blocked by BcGrape infection . Examining differences in transcription between A . thaliana leaves infected with these B . cinerea isolates revealed that transcriptional responses to these isolates differed more dramatically in jasmonate-insensitive coi1 plants than in the wild-type background ( Figure 9 , Table S2 ) . This suggests that jasmonates are not only important signaling components but also integrators of signals from diverse pathogen genotypes into consistent plant defense responses . The visually distinctive lesion phenotype produced by Bc83-2 infection of jasmonate-deficient A . thaliana genotypes , coupled with the persistence of plant-produced GUS activity within the lesion produced by Bc83-2 on coi1 mutants , initially suggested that the mechanisms by which this isolate induces plant death may be jasmonate-dependent ( Figure 1 ) . However , vital staining indicated that the B . cinerea isolates caused similar patterns of plant cell death in leaves of both wild-type and jasmonate-insensitive plants ( Figure 7 ) . Thus , the observed differences in plant transcriptional response to these pathogen isolates are not likely linked to simple differences in the number of living cells in the leaf , but instead result from differences in plant—pathogen communications , potentially including plant perception of pathogen-induced damage and pathogen metabolism of dead plant tissues . Despite similarities in lesion development and transcriptional effects on wild-type plants , the two B . cinerea isolates tested in this study , BcGrape and Bc83-2 , show differing interactions with plant response networks that are masked by the response of an intact plant jasmonate signaling pathway . These differences are revealed in mutants deficient in jasmonate biosynthesis and several aspects of jasmonate signaling , most strikingly by quantitative differences in camalexin accumulation in jasmonate-deficient A . thaliana leaves infected with these pathogen isolates . Examination of transcriptional response to B . cinerea infection in plants with impaired jasmonate signaling has revealed the involvement of at least two groups of co-regulated loci not previously associated with plant defense responses . Exploration of the function of these putative networks in A . thaliana defense against B . cinerea and other pathogens may provide novel insight into mechanisms of plant defense . A . thaliana mutants deficient in jasmonate biosynthesis , allene oxide synthase ( aos ) , and biosynthesis of camalexin , phytoalexin deficient 3 ( pad3-1 ) , were obtained from the Arabidopsis Biological Resource Center ( www . biosci . ohio-state . edu/pcmb/Facilities/abrc/abrchome . htm ) [41] , [45] . All mutant lines were in the Col-0 genetic background , with aos mutants additionally containing the visible marker gl1 . The presence of a mutant aos allele was determined by PCR using gene-specific and insert-specific primers [41] . A . thaliana segregating the coronatine insensitive 1 ( coi1-1 ) mutation , conferring deficiency in jasmonate perception , was obtained from J . Glazebrook , University of Minnesota [42] . Homozygous coi1-1 plants were identified using a CAPS marker; a 531bp fragment of At2g39940 ( COI1 ) contains an Xcm1 restriction site that is abolished by the coi1-1 mutation [42] . Plants with coi1 aos double mutant genotypes were generated by fertilizing aos plants with pollen from COI1/coi1 heterozygous plants . F1 progeny were genotyped to select COI1/coi1-1 heterozygotes; these were allowed to self-pollinate and B . cinerea lesion growth and camalexin accumulation phenotypes were determined for a segregating F2 population . ProCYP79B2:GUS contains a transgenic fusion of the CYP79B2 promoter to a β-glucuronidase reporter [87] . ProCYP79B2:GUS coi1-1 plants were generated by fertilizing male-sterile coi1-1 flowers with ProCYP79B2:GUS pollen , allowing F1 plants to self-pollinate , and selecting appropriate genotypes from the F2 segregants . A . thaliana containing the JAZ1Δ3::GUS transgene , conferring a dominant jasmonate-insensitive phenotype , was obtained from G . Howe , Michigan State University [88] . Plants for all experiments were grown in 36-cell flats ( approximately 120cm3 soil per cell ) in a growth chamber at 12h∶12h light∶dark , 22°C , 50–60% RH , and ∼150µE light intensity . Seed was sown on soil ( Sunshine Mix #1 , Sun Gro Horticulture Ltd . , Bellevue WA ) and thinned to one plant per cell at three days post-germination . Genotypes compared within an experiment were systematically interspersed within flats . Plants were sub-irrigated twice weekly with deionized water . Experiments were conducted with mature , non-bolting rosette plants at 5–6 weeks post-planting . Source and reference data for B . cinerea isolates used in this study are provided in Table 1 . Preliminary experiments compared infection phenotypes of whole rosettes ( detached from the root approximately 0 . 5cm below the soil surface and placed on agar ) with observations of detached single leaves; no differences in measured phenotypes were observed ( Figure 1 ) . Further experiments used detached rosette leaves , inoculated with B . cinerea spores as previously described [89] . Inoculum was freshly prepared for each experiment from concentrated spore stocks stored at −20°C in 25% glycerol . Leaves were inoculated with 5µl droplets of spore suspension ( 5×105 spores/ml in half-strength filtered organic grape juice ) ( Santa Cruz Organics , California USA ) . Digital photographs were analyzed using Image J to measure lesion area [89] , [90] . Control leaves ( mock ) were inoculated with half-strength grape juice . Abiotic elicitors of camalexin were 5mM AgNO3 and 10µM acifluorfen ( Sigma-Aldrich , St . Louis , MO USA ) , applied as four 5µl droplets per leaf to one side of the midvein . Staining of ProCYP79B2:GUS leaves for GUS activity at 72 hours post-inoculation was performed as described [91] . Staining of wild-type and coi1 leaves for cell death ( Trypan Blue ) and H2O2 accumulation ( DAB ) at 72 hours post-inoculation was performed as described [92] , [93] . Camalexin was extracted in 90% MeOH and quantified via HPLC as previously described [30] . Whole leaves were collected in 500µl 90% MeOH in 96- deep-well plates and stored at −20°C until extraction and analysis , except tissue samples used for transcript measurements where fresh tissue was frozen in liquid nitrogen , ground without solvent , and separate aliquots of frozen tissue were removed for RNA isolation and camalexin extraction . Camalexin measurements are standardized by tissue weight ( g ) or leaf area ( cm2 ) ; leaf weight and area are highly correlated within the A . thaliana genotypes used for these experiments . Seed from heterozygous COI1/coi1 A . thaliana was grown as described ( “Plant Growth Conditions” ) and genotyped 2–3 days prior to experiments . DNA was isolated from the first true leaves to minimize stress to the plant and maximize leaf tissue available for experiments . Eight leaves were detached from each homozygous wild-type or coi1 plant , such that each plant contributed one leaf per B . cinerea isolate ( BcGrape vs . Bc83-2 ) ×time point ( 24 , 32 , 48 , and 72 hours post-inoculation ) combination . At each time point , leaves were photographed and six to eight leaves per plant genotype×B . cinerea isolate combination were collected individually into 90% MeOH and processed as described ( “Camalexin measurements” ) . Comparisons of lesion and camalexin data for the experiments described above were performed using a 2-way factorial ANOVA model with classes plant genotype and treatment ( Table 1 ) . A genotype×treatment interaction term was included in the model . Specific comparisons of least-squares means were evaluated for significance using Tukey's HSD adjusted p-values . Time course experiments were analyzed similarly , but including time point as an additional class variable . These analyses were conducted in SAS ( Version 9 . 1 , SAS Systems , Cary NC USA ) .	While many important elements of plant defense signaling have been identified , the function of these defense signaling pathways may mask additional variation in the plant–pathogen interaction , including both pathogen variation and variation in downstream plant defense responses . Jasmonate plant hormones contribute to both plant development and defense , including plant defense against necrotrophic fungal pathogens such as the grey mold Botrytis cinerea . Ten diverse B . cinerea isolates all showed increased virulence and decreased induction of a plant antimicrobial metabolite in experimental infections of Arabidopsis thaliana lacking functional jasmonate signaling . Yet within this consistent result , B . cinerea isolates varied considerably . Through comparing the transcript profiles of A . thaliana infected with the two most disparate B . cinerea isolates , we found that wild-type plants showed similar transcriptional responses to infection with these two isolates , but the absence of functional jasmonate signaling revealed dramatic differences in plant response , including groups of co-regulated genes that may participate in undescribed plant response networks . Jasmonate signaling appears to integrate plant responses to diverse pathogen inputs , and its absence may reveal novel aspects of plant–pathogen interaction .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "plant", "biology/plant-biotic", "interactions", "genetics", "and", "genomics/genetics", "of", "disease", "plant", "biology/plant", "genetics", "and", "gene", "expression", "genetics", "and", "genomics/plant", "genetics", "and", "gene", "expression" ]	2010	Deficiencies in Jasmonate-Mediated Plant Defense Reveal Quantitative Variation in Botrytis cinerea Pathogenesis
The essential biological properties of proteins—folding , biochemical activities , and the capacity to adapt—arise from the global pattern of interactions between amino acid residues . The statistical coupling analysis ( SCA ) is an approach to defining this pattern that involves the study of amino acid coevolution in an ensemble of sequences comprising a protein family . This approach indicates a functional architecture within proteins in which the basic units are coupled networks of amino acids termed sectors . This evolution-based decomposition has potential for new understandings of the structural basis for protein function . To facilitate its usage , we present here the principles and practice of the SCA and introduce new methods for sector analysis in a python-based software package ( pySCA ) . We show that the pattern of amino acid interactions within sectors is linked to the divergence of functional lineages in a multiple sequence alignment—a model for how sector properties might be differentially tuned in members of a protein family . This work provides new tools for studying proteins and for generally testing the concept of sectors as the principal units of function and adaptive variation . The amino acid sequence of a protein reflects the selective constraints underlying its fitness and , more generally , the evolutionary history that led to its formation [1] . A central problem is to decode this information from the sequence , and thus understand both the “architecture” of natural proteins , and the process by which they evolve . With the dramatic expansion of the sequence databases , a powerful strategy is to carry out statistical analyses of the evolutionary record of a protein family [2–6] . With the assumption that the principal constraints underlying folding , function , and other aspects of fitness are conserved during evolution , the idea is to start with an ensemble of homologous sequences , make a multiple sequence alignment , and compute a matrix of correlations between sequence positions—the expected statistical signature of couplings between amino acids . Using mathematical analyses that explore different aspects of this matrix [7 , 8] , studies have exposed tertiary structural contacts in protein structures ( Direct Coupling Analysis , or DCA , [4 , 9] ) , determinants of binding specificity in paralogous protein complexes [5] , and larger , collectively evolving functional networks of amino acids termed “protein sectors” ( Statistical Coupling Analysis , or SCA [10] . These different approaches suggest a hierarchy of information contained in protein sequences that ranges from local constraints that come from direct contacts between amino acids in protein structures to global constraints that come from the cooperative action of many amino acids distributed through the protein structure . Sectors are interesting since they may represent the structural basis for functional properties such as signal transmission within [3 , 6 , 11–14] and between [15–17] proteins , allosteric regulation [6 , 15 , 18–20] , the collective dynamics associated with catalytic reactions [16] , and the capacity of proteins to adapt [21] . In addition , experiments show that reconstituting sectors is sufficient to build artificial proteins that fold and function in a manner similar to their natural counterparts [22–24] . Thus , the quantitative analysis of coevolution provides a powerful approach for generating new hypotheses about the physics and evolution of protein folding and function . These results imply that together with structure determination and functional measurements , the evolution-based decomposition of proteins should be a routine process in our study of proteins . However , the analysis of coevolution poses non-trivial challenges , both conceptually and technically . Conceptually , coevolution is the statistical consequence of the cooperative contribution of amino acid positions to organismal fitness , a property whose relationship to known structural or biochemical properties of proteins remains open for study . Indeed , there is no pre-existing model of physical couplings of amino acids with which to validate patterns of coevolution . Thus , the goal of coevolution based methods is to produce models for the pattern of constraints between amino acids that can then be experimentally tested for structural , biochemical , and evolutionary meaning . Technically , the analysis of coevolution is complicated by both the limited and biased sampling of sequences comprising a protein family . Thus , empirical correlations deduced from multiple sequence alignments do not always reflect coevolution . Interestingly , the complexities in sequence sampling can represent both sources of noise and useful signal in decomposing protein structures , and it is essential to understand these issues in effectively using methods of coevolution . The DCA approach for mapping amino acid contacts has been well-described by analogy with established theory in statistical physics [4] . Here , we present the principles and implementation of the SCA method for identifying sectors and introduce new tools for understanding the global patterns of coevolution between amino acid positions . The methods are implemented in an open-source python-based software package that is available to the scientific community , and illustrated in the main text using the small G protein family of nucleotide-dependent switches [25 , 26] and the S1A family of serine proteases [27 , 28] . Technical modifications from previous implementations of SCA are indicated in the main text and summarized in the the S1 Table . In prior work , we examined just the broadest level of coevolution to define sectors—quasi-independent groups of coevolving amino acids [10] . We now go beyond this top-level decomposition to reveal a more elaborate internal architecture for sectors in which subgroups of amino acids diverge along functional , and sometimes phylogenetic , subfamilies within the sequence alignment . Overall , this work provides a necessary foundation for broad testing of the concept of protein sectors . The evolutionary conservation of a sequence position is estimated from the deviation of the observed distribution of amino acids at this position from a background distribution expected by neutral drift . A simple mathematical quantity that captures this concept is D i a = f i a ln f i a q a + ( 1 - f i a ) ln 1 - f i a 1 - q a , ( 5 ) where f i a is the observed frequency of amino acid a at position i in the alignment and qa is the background expectation ( see S1 Text , section B for derivation ) . D i a is known as the Kullback-Leibler relative entropy [36] and indicates how unlikely the observed frequency of amino acid a at position i would be if a occurred randomly with probability qa—a quantitative measure of position-specific conservation . Note that D i a = 0 only when f i a = q a and increases more and more steeply as f i a deviates from qa ( Fig 2 ) , consistent with intuition that a measure of conservation should non-linearly describe the divergence of the observed distribution of amino acids from their expected values . An underlying assumption in the derivation of the relative entropy is that the sampling of sequences in the alignment is unbiased , a condition that , to varying extent , is violated by the tree-like phylogenetic structure of real alignments . But without validated models for protein evolution that can provide a basis for more accurate measures of conservation , this choice reflects the simplest definition that satisfies the general principle of conservation . Finally , Eq ( 5 ) gives the conservation of each amino acid a at each position i , but an overall positional conservation Di can be defined following the same principles ( Fig 3A , and see S1 Text , section C ) . Analysis of the spatial pattern of positional conservation generally leads to a simple conclusion: the solvent inaccessible core of proteins and functional surfaces tend to be more conserved and the remainder of the surface is less conserved ( Fig 3A and 3B ) [10 , 37 , 38] . Thus , positional conservation in sequence alignments reflects well-known properties of protein three-dimensional structures . The cooperativity of amino acids in specifying protein folding and function implies that the concept of positional conservation of individual positions should at least be extended to a concept of pairwise conservation , reporting coevolution between positions in a protein family . Given the alignment , a measure of correlation of the pair of amino acids ( a , b ) at positions ( i , j ) is given by the difference of their joint frequency f i j a b and that expected in absence of correlation , f i a f j b . Computed for all pairs of amino acids in the alignment , this defines a covariance matrix C i j a b = f i j a b - f i a f j b . ( 6 ) Alternatively , statistical dependency can be quantified by the mutual information , whose origin is similar to the relative entropy [5 , 36] . However , both the covariance matrix and the mutual information report deviations from independence given the frequencies f i a , and do not take into account the evolutionary relevance of observing those frequencies . In the current implementation of SCA , the approach is to perform a first-order perturbation analysis on the multiple sequence alignment in which we compute the correlated conservation of pairs of amino acids . To explain , consider that many alignments A are available for the same protein family . We can then define relative entropies D i , A a—our measure of positional conservation—for each alignment A , and compute their correlations over the ensemble of alignments by C ˜ i j a b = ⟨ D i , A a D j , A b ⟩ A - ⟨ D i , A a ⟩ A ⟨ D j , A b ⟩ A , ( 7 ) where the angled brackets indicate averages over the A alignments . In practice , many such alignments can be obtained by bootstrap resampling the original alignment [39]; for instance , a procedure known as “jackknife resampling” consists of successively removing each sequence s from the original alignment to create a collection of M sub-alignments . A perturbative expansion of D i a as a function of f i a for the jackknife resampling process shows that Eq ( 7 ) yields a covariance matrix that has the form C ˜ i j a b = ϕ i a ϕ j b C i j a b , ( 8 ) in which ϕ i a = ∂ D i a ∂ f i a is a function of the conservation of each amino acid at each position ( see S1 Text , section D for derivation ) [10] . That is , SCA produces a weighted covariance matrix , with the weighting function ϕ controlling the degree of emphasis on conservation . This definition of ϕ has the property of rising steeply as the frequencies of amino acids f i a approach one . As a consequence , these weights damp correlations in C i j a b arising from weakly conserved amino acids ( the gradient of D i a approaches zero as f i a → q a ) , and emphasize conserved correlations . Another way to understand these weights comes from considering their role in determining similarities between sequences comprising the alignment . The mathematical principles are described below , but in essence positional weights ϕ i a redefine the distance between sequences in a manner that emphasizes variation at more conserved positions in the alignment ( see S1 Text , section I ) . It is logical that such a “conservation-biased” distance metric between sequences will provide a better representation of the functional differences ( as opposed to historical differences ) between sequences . The weighting by ϕ i a in Eq ( 8 ) implements the same principle applied to the correlations between positions instead of the correlations between sequences . In principle , the specific form of ϕ should vary depending on the evolutionary history of the protein properties that are under consideration; the more conserved the properties of interest are , the more the weights should emphasize conservation [40] . Indeed , different weighting functions are possible if mathematical formalisms other than the KL entropy are proposed for defining positional conservation , or if other approaches than the first-order perturbation analysis described here are developed . For example , early versions of the SCA method [3 , 6] involved slightly different weights whose technical origins are given in S1 Text , section E . C ˜ i j a b is a four-dimensional array of L positions × L positions × 20 amino acids × 20 amino acids , but we can compress it into a L × L matrix of positional correlations by taking a magnitude ( the Frobenius norm ) of each 20 × 20 amino acid coevolution matrix for each pair of positions ( i , j ) : C ˜ i j = ∑ a , b ( C ˜ i j a b ) 2 ( 9 ) See S1 Text , section F and S1 Fig for additional arguments about compressibility of C ˜ i j a b . Fig 3C and 3D shows the C ˜ i j matrix for the G protein family . As previously reported , the matrix is heterogeneous , with a small number of positions engaged in significantly higher correlations than most positions ( Fig 3C , [6 , 10] ) . Hierarchical clustering makes this heterogeneity more apparent , and reveals the existence of nested clusters of correlated positions ( Fig 3D ) . These findings are qualitatively consistent with a sparse , hierarchical , and cooperative pattern of evolutionary constraints . As we show below , there is also modularity [10] , with quasi-independent groups of positions emerging from the correlations ( the sectors and their subdivisions ) . Unlike the interpretation of first-order conservation ( Fig 3A and 3B ) , none of these properties is obvious in either current analyses of protein structures . How can we understand the pattern of coevolution in the C ˜ i j matrix ? The existence of correlations ( Fig 3C ) indicates that treating the amino acid positions as the basic units of proteins is not the most relevant representation . Instead , we should seek a transformation that re-parameterizes the protein into groups of correlated positions that are maximally independent from each other—a more natural representation that defines the units of evolutionary selection . The first step in this process is spectral ( or eigenvalue ) decomposition . Per this decomposition , the C ˜ i j matrix is written as a product of three matrices: C ˜ = V ˜ Δ ˜ V ⊤ ˜ , ( 10 ) where Δ ˜ is an L × L diagonal matrix of eigenvalues ( ranked by magnitude ) and V ˜ is an L × L matrix whose columns contain the associated eigenvectors . Each eigenvalue gives the quantity of information ( variance ) in C ˜ i j captured , and each associated eigenvector in V ˜ gives the weights for combining sequence positions into transformed variables ( or eigenmodes ) . For both G protein and S1A alignments , the histogram of eigenvalues—the spectrum of C ˜ i j—reveals a few large eigenvalues extending from a majority of small values ( Fig 4A and 4B , black ) . To estimate the number of significant eigenvalues , we compare the actual spectrum with that for many trials of randomized alignments in which the amino acids at each position are scrambled independently [10] ( Fig 4A and 4B , red line ) . This randomization removes true positional correlations , leaving behind the spurious correlations expected due to finite sampling in the alignment . As is the case for all practical alignments in which the number of effective sequences is not large compared to the number of amino acids , these spurious correlations account for the bulk of the spectrum . Indeed , for both alignments this analysis indicates that just the top few eigenmodes are statistically significant ( k* = 4 , G protein , and k* = 7 , S1A; see S2 Fig for an analysis of robustness ) . Thus , the k* associated eigenvectors define a low dimensional space in which patterns of positional coevolution can be studied ( e . g . Fig 4C and 4D ) . It is important to note that the precise value of k* is not a fundamental property of a protein family; it depends on protein size and the number of effective sequences . Nevertheless , with adequate sampling ( M′ > 100 ) the analysis of sectors seems largely robust to its precise value ( see DHFR tutorial , S3 Text ) . Fig 4C and 4D show structure of the space spanned by the top three eigenvectors for the G protein and S1A families , respectively . In these graphs , the ( Euclidean ) distance of a position from the origin reports its overall contribution to the correlations , and the distance between two positions indicate their degree of correlation: strongly correlated positions appear near-by , while weakly correlated positions are far apart , or , for the majority that do not make any substantial contributions to the correlations , clustered near the origin . As a consequence , independent sets of correlated positions are expected to cluster into groups of positions at distance from the origin . When the correlations within these groups are organized hierarchically , these clusters extend radially with positions at extremity representing the core of the hierarchy , and successive layers at decreasing distance from the origin representing progressively weaker levels of the hierarchy . For both protein families , this analysis suggests a few distinct groups of positions that seem to emerge radially from the origin ( Fig 4C and 4D , different colors ) . The spectral decomposition is effective for dimension reduction , but the eigenmodes generally do not provide an optimal representation of groups of coevolving positions . For example , distinct groups of positions emerge along combinations of the k* top eigenvectors [10] . The reason is that just decorrelation of the positions by diagonalizing the C ˜ i j matrix—the essence of eigenvalue decomposition—is a weaker criterion than achieving statistical independence , which demands absence of not only pairwise correlations , but lack of any higher order statistical couplings . In prior work , we managed this problem heuristically , finding combinations of eigenmodes , excluding the first , that happen to represent quasi-independent groups [10] . Here , we introduce the use of independent components analysis ( ICA [41 , 42] ) —an extension of spectral decomposition—that computationally addresses this problem . ICA uses numerical optimization to deduce a matrix W that transforms the k* top eigenmodes of a correlation matrix into k* maximally independent components ( ICs , S1 Text , section G ) , V ˜ 1 ⋯ k * p = W V ˜ 1 ⋯ k * . ( 11 ) The bottom line is that the k* ICs ( in columns of V ˜ p ) should now represent a more appropriate organization of positional coevolution . In both G proteins ( Fig 4E ) and S1A proteases ( Fig 4F ) , ICA produces a representation in which the majority of positions are weakly correlated and cluster near the origin and a relatively small subset of positions comprise quasi-independent groups of amino acids emerging along separate orthonormal axes ( the ICs ) . The ICs need not be strictly independent , a key issue in defining sectors that we discuss in detail below . Nevertheless , spectral decomposition with ICA provides the sort of transformation of protein sequences that we seek—based on their evolutionary correlations , amino acid positions are regrouped and transformed into new effective variables ( the ICs ) that represent collectively evolving modes of the protein under study . How can we examine the relevance of the IC-based decomposition of proteins ? A approach comes from understanding a fundamental mathematical relationship between the pattern of positional correlations ( which defines ICs ) and the structure of the sequence space spanned by the alignment ( which defines sequence subfamilies ) [43 , 44] . The concepts underlying this mapping between positions and sequences were presented either heuristically [10] or partially [17] in prior work; here , we provide a full explanation with new mathematical methods . Consider the two-dimensional binary matrix representation of an alignment Xsn comprised of M sequences by 20L amino acids ( Figs 1C and 5A ) . From Xsn , we can compute two kinds of correlations: ( 1 ) a correlation matrix over rows S r s = 1 L ∑ n X r n X s n , which represents the similarity ( fraction identity ) of each pair of sequences r and s ( Fig 5B ) and ( 2 ) a correlation matrix over columns F n m = 1 M ∑ s X s n X s m , which represents the joint frequency of amino acids at each pair of positions ( Fig 5C ) . F and S are intimately related to each other by a mathematical property known as the singular value decomposition ( SVD ) . Specifically , if U represents the eigenvectors of the sequence correlation matrix S and V represents the eigenvectors of the amino acid correlation matrix F , then X = U Λ 1 / 2 V ⊤ , ( 12 ) where Λ is a diagonal matrix whose entries are ( up to a scaling factor ) eigenvalues of both S and F . The key conceptual point is that by the SVD , the eigenvectors of S are a mapping from the eigenvectors of F , where the “map” is the alignment X itself , U = X V Λ - 1 / 2 . ( 13 ) This introduces the principle of sequence-position mapping , using the full alignment matrix X to relate patterns of amino acid correlations ( in V ) to patterns of sequence divergence ( in U ) . But , to study the pattern of sequence divergence associated with sectors , we need to make a similar mapping using the conservation-weighted dimension-reduced coevolution matrix C ˜ i j ( rather than the unweighted amino acid correlation matrix F ) . Since C ˜ i j is a L × L positional correlation matrix , a sequence-space mapping analogous to Eq ( 13 ) requires a dimension-reduced alignment matrix in which the 20 amino acids at each position are compressed into a single value . The Supplementary Information describes a new approach for this step , effectively reducing the alignment x s i a from a M × L × 20 array to an M × L matrix xsi by projecting the amino acid dimension down to a single scalar value ( S1 Text , section H and S4 Fig ) . By analogy with Eq ( 13 ) , the reduced alignment matrix xsi then defines a mapping between the space of positional coevolution ( in the top ICs of the C ˜ i j matrix ) and the corresponding sequence space . Specifically , if Δ ˜ and V ˜ are the eigenvalues and eigenvectors , respectively , of the SCA positional coevolution matrix C ˜ i j , then U ˜ = x V ˜ Δ ˜ - 1 2 ( 14 ) represents the structure of the sequence space corresponding to the patterns of positional coevolution in V ˜ . Furthermore , if W is the transformation matrix derived from ICA of V ˜ 1 … k * , Eq ( 11 ) , then U ˜ p = W U ˜ ( 15 ) represents the sequence space corresponding to V ˜ p , the ICs of the C ˜ i j matrix . Eqs ( 14 ) and ( 15 ) give us the necessary tools for interpreting the IC-based decomposition of proteins . For the S1A family , Fig 6 shows a mapping between the top six ICs and the corresponding sequence space . Sequences are colored by enzymatic function ( Fig 6A–6C , the haptoglobins are non-catalytic homologs of the S1A family ) , by catalytic specificity ( Fig 6D–6F ) , or by phylogenetic origin ( Fig 6G–6I ) . The data show that ICs 1–3 correspond to essentially orthogonal divergences in the S1A protein family . IC1 ( but not any of the other ICs ) separates the catalytic from non-catalytic S1A proteins ( Fig 6A ) , IC2 uniquely separates S1A proteins by their annotated primary ( P1 site ) catalytic specificity [28] ( Fig 6D ) , and IC3 uniquely separates vertebrate and invertebrate sequences ( Fig 6H ) . ICs4–6 show more subtle inhomogeneities with regard to catalytic specificity ( Fig 6E and 6F ) , indicating finer subdivisions of the annotated sequences—well-defined predictions for further study . Thus , the ICs of the C ˜ i j matrix contain independently evolving functional units within the S1A protease [10] . Fig 7A–7D shows the mapping between the top ICs of the G-protein family and the corresponding sequence space , colored either by functional sub-type ( Fig 7A and 7B ) or by taxonomic origin ( Fig 7C and 7D ) . The data show that IC1 and IC2 separate different sub-classes of the G protein family , suggesting that like in S1A proteases , amino acid motifs in different ICs can control different functional properties ( Fig 7A ) . In contrast , IC3 and IC4 are associated with a near homogeneous distribution of functional subtypes , suggesting either neutral or more fine variations with regard to the broad functional annotations available in this protein family ( Fig 7B ) . With the exception of IC3 in the S1A family ( Fig 6H ) , none of the ICs are obviously associated with the divergence of the main taxonomic groups in the alignment; indeed , all taxa seem nearly homogeneously distributed over the sequence modes ( Up ) corresponding to most of the ICs . Many paralogs of the different functional classes of G proteins and S1A proteases are found in each type of organism and thus functional divergence might therefore not be expected to follow the divergence of species . In contrast , ICs are more associated with taxonomic classification for the DHFR protein family ( S4 Fig and S3 Text ) , consistent with the fact that this core metabolic enzyme is encoded by a single ortholog in each genome . In summary , the sequence-position mapping provides evidence that the ICs of the C ˜ i j matrix represent conserved , differentially evolving functional units in proteins . The ICs are not distinguished by the magnitude of positional conservation ( Fig 8 ) , showing that this decomposition of proteins is fundamentally a property of correlations—the second order terms in conservation . This finding makes an important statement about the “value added” by studying coevolution , as opposed to just the first-order conservation of positions . Indeed , it is difficult to experimentally test the unique value of statistical coevolution by conventional single mutation experiments , even when conducted on a massive scale [21 , 45 , 46] . Coevolution implies the need for higher-order mutational studies , which are difficult to perform quantitatively and only now starting to become feasible [47] . In this regard , the functionally meaningful , quasi-independent divergence of proteins along ICs demonstrates the necessity of coevolution in providing a proper decomposition of protein structure . But , does the existence of k* significant ICs imply k* independent functional units ( and therefore k* sectors [10] ) ? Not necessarily . Sectors typically have an organization in which the constituent positions can be further broken up into subsets of coevolving positions . One generative mechanism for this architecture comes from the tree-like structure of the alignment in which sequences are partitioned into functional subfamilies along which portions of one sector can diverge [43 , 44] . Thus , each IC could have one of two interpretations: ( 1 ) a truly independent sector associated with a distinct function , or ( 2 ) the decomposition of a single sector ( representing one functional property ) into separately diverging sub-parts . In this sense , the term “independent component” is something of a misnomer , but we retain the language here for consistency with the ICA method . How can we systematically distinguish these possibilities to deduce the number and composition of sectors ? We follow a simple procedure ( see tutorials in S3 Text ) . First , we fit each IC to an empirical statistical distribution and identify the positions contributing to the top five percent of the corresponding cumulative density function ( CDF , Figs 4E , 4F and S2 ) . The t-distribution appears to generally fit the ICs well in all cases studied to date ( S5 Fig ) , and IC composition is robust to alignment size when diversity is maintained ( S6 Fig ) . The CDF cutoff is an adjustable parameter , but 5% seems to agree well with experimental significance in the model systems studied [16 , 21] . We then construct a sub-matrix of C ˜ i j that contains only the selected top-scoring positions for the k* ICs , ordered by their degree of contribution to each IC . For the G protein family , this corresponds to a matrix of 54 positions that contribute to the top four significant ICs ( Fig 9A ) . This sub-matrix describes both the pattern of “internal” correlations between positions that make up each IC ( the diagonal blocks ) , and the pattern of “external” correlations between ICs ( the off-diagonal blocks ) . This representation shows that ICs 1 , 2 , and 3 display a set of transitive inter-IC correlations , with IC1 correlated to IC2 and IC2 correlated to IC3 , indicating that IC1–3 together comprise the hierarchically decomposed parts of a single sector ( sector 1 , Fig 9A and 9B ) . In contrast , IC4 shows near-independence from the other ICs , suggesting that it defines a distinct sector ( sector 2 , Fig 9A and 9B ) . These sector definitions are made exclusively from analysis of the IC-based submatrix of C ˜ i j , but correspond to a meaningful spatial architecture in the G protein . These proteins are binary switches that display different conformations depending on the identity of their bound guanine nucleotide [25 , 26] . The exchange of GTP for GDP triggers two specific conformational changes: clamping of the so-called switch I loop closer to the nucleotide binding pocket , and transit of a disordered and weakly interacting surface loop ( switch II ) to an ordered helix that is well-packed against the core domain ( Fig 9C ) [25] . Sector 1 comprises a physically contiguous group of amino acid residues that shows excellent agreement with the nucleotide-dependent allosteric mechanism [50] . The sector is compact in the GTP-bound state but partially disrupted in the GDP-state ( Fig 9B and 9C ) , a finding consistent with the state-dependent connectivity between the nucleotide-binding pocket and the switch loops . Furthermore , the hierarchical breakdown of sector 1 into its constituent ICs 1 , 2 and 3 reveals a meaningful structural organization: IC3 ( cyan ) defines a physically contiguous network that comprises the nucleotide binding pocket , IC1 ( light blue ) defines the packing interactions between switch II and the core domain , and IC2 ( dark blue ) represents a set of surface accessible positions ( including switch I ) that link to the buried core of sector 1 ( Fig 9C ) . Nucleotide exchange substantially reorganizes the structure and connectivity of IC1 and 2 , but is largely inconsequential for IC3 ( Fig 9C ) . Consistent with assignment as an independent sector , sector 2 ( IC4 , red ) also comprises a mostly physically contiguous group within the core of the G protein ( Fig 9B ) ; like IC3 of sector 1 ( cyan ) it shows no nucleotide-dependent conformational plasticity . These results are interesting since IC1 and IC2 ( but not IC3 or IC4 ) are associated with the divergence of functional sub-classes of G protein ( Fig 7A ) . The data suggest that IC3 ( cyan subset , sector 1 ) and IC4 ( sector 2 ) are global functional modes shared by all members of the G protein family , while ICs 1 and 2 correspond to subsets of sector 1 that are specialized for tuning allosteric or effector-binding properties within sub-classes of G proteins . These observations represent new hypotheses for further study . For the S1A family , the IC-based submatrix shows little evidence of inter-IC correlations , and thus we conservatively treat all ICs as separate sectors ( S7 Fig ) . Each sector corresponds to a largely contiguous network of amino acids in the protease tertiary structure , a decomposition consistent with the orthogonal sequence divergences and with previous reports ( S7 Fig , and [10] ) . Examples of sector analysis for two other protein families—the dihydrofolate reductases [51] and the class A beta-lactamases [52]—are provided in S8 Fig and in tutorials ( S3 Text ) . The process of sector identification presented here is heuristic , requiring the judgement of the practicing scientist to determine the grouping of ICs to form sectors . This reflects that fact that various degrees of independence between ICs are possible depending on the statistical nature of selective pressures operating on a protein family . Thus , an automated approach to interpreting the C ˜ i j matrix awaits more broad experience with sector analysis in many protein families . Given the importance of interpreting hierarchical correlation matrices in general ( e . g [53–55] ) , it seems reasonable that such automation might be achieved with further work . Previous work has introduced the concept of sectors as quasi-independent units of protein structure that are associated with distinct functional properties [10] , but has largely ignored their internal architecture . This work presents a more refined description in which a sector may itself be decomposed into a physically contiguous core element ( e . g . IC3 , Fig 9C ) , surrounded by peripheral elements ( e . g . ICs 1 and 2 , Fig 9C ) that have the property of differential variation along functional branches of a protein family ( Fig 7A ) . Thus , we propose a model that sectors are structural units of function and the ICs define patterns of variation within these units . These observations also highlight the practical value of the mapping between positional correlations and sequence subfamilies . When functional divergences between subfamilies are annotated , the mapping can identify the positions responsible for this divergence . For example , in the Hsp70 family of chaperones , the existence of subfamilies with known differences in allosteric function led to the identification of positions involved in the underlying mechanism [17] . Turned around , when the role of specific positions in a protein is known , the mapping can help annotate sequences according to the associated functional property . For example , sequence divergence within sector positions with known function in the S1A family permitted classification of the sequence space according to that functional property [10] . In principle , high-throughput methods for functional annotation of members of a protein family should permit even more refined mappings between amino-acid variation and phylogenetic or functional divergence , a step towards relating genotype-to-phenotype at the molecular level . It is valuable to explain the similarities and distinctions of SCA with other analyses of coevolution in multiple sequence alignments . The direct coupling analysis ( DCA [4] ) and its various extensions [9 , 62–64] are focused on using coevolution to determine physical contacts between amino acids within or between protein tertiary structures . As different as this problem may seem from discovering the pattern of functionally coevolving amino acids , there is a deep relationship . Recent work shows that the two approaches focus on two extremes of the same hierarchical architecture of coevolution [7 , 8] . SCA focuses on the global modes of coevolution ( the top eigenmodes of a conservation-weighted correlation matrix ) , and DCA on the minimal units of coevolution ( the bottom eigenmodes of an unweighted correlation matrix ) . Thus , coevolving direct contacts are at one end of the hierarchy and sectors at the other . Consistent with this , coevolving direct contacts are found within sectors and outside of sectors , but not bridging two independent sectors [8] . Another approach for analyzing coevolution in protein alignments is mutual information , which has been successful at predicting the amino acids responsible for specificity in some protein-protein interactions [5] . The distinction between this method and SCA lies in the nature of the weighting function ϕ; in essence , the mutual information method uses flat positional weights ( ϕ = 1 ) , which has the effect of emphasizing more unconserved correlations and may therefore be more appropriate when studying rapidly diverging functional properties [40] . Taken together , these observations begin to clarify the relationship of the different approaches , and poses the question of the nature of physical information held at various levels of the hierarchy of coevolution , a matter for future experimentation . From a theoretical point of view , the observations highlight the need for a better , more unified framework representing the full hierarchy in amino acid correlations in proteins , a key next goal in advancing the statistical approach to the biology of proteins . Sector analysis provides a representation of proteins that is distinct from the first-order analysis of positional conservation and that ( so far ) is not obtained from structure determination or functional mutagenesis . Thus , it provides a valuable tool for directing experimental studies of protein folding and function , and ultimately , for formulating a physical and evolutionary theory consistent with the design of natural proteins . Multiple sequence alignments were obtained from previous work [10] or from the PFAM database ( release 27 . 0 , accession codes PF00071 ( G proteins ) , PF00186 ( DHFR ) , and PF13354 ( class A β-lactamases ) ) , and were subject to pre-processing with default parameter values as described in Box 1 . Reference sequences/structures selected for each family were rat trypsin ( PDB 3TGI ) , human Ras ( PDBs 5P21 and 4Q21 ) , E . coli DHFR ( PDB 1RX2 ) , and E . coli TEM-1 β-lactamase ( PDB 1FQG ) , and with sub-sampling to the number of effective sequences , yielded the following final alignment statistics: S1A serine proteases ( 928 effective sequences by 205 positions ) , G proteins ( 3366 effective sequences by 158 positions ) , DHFR ( 1157 effective sequences by 151 positions ) , β-lactamase ( 497 effective sequences by 200 positions ) . All calculations were carried out using a new python implementation of the statistical coupling analysis ( pySCA v6 . 2 ) , following the algorithms described in Box 1 and in the main text . Step-by-step tutorials for executing the analysis for the four protein families are provided in the S3 Text and accompany the toolbox distribution . The pySCA toolbox is available for download through GitHub ( https://github . com/reynoldsk/pySCA ) , and with online instructions at http://reynoldsk . github . io/pySCA .	Proteins display the ability to fold , to carry out complex biochemical reactions , and to be adaptive to changing conditions of selection—the essential characteristics contributing to organismal fitness . A major goal is to understand how these properties emerge from the global pattern of interactions between amino acid residues . Here , we describe the principles and implementation of the statistical coupling analysis ( SCA ) , a method to reveal this pattern through analysis of coevolution between amino acids in an ensemble of homologous sequences . The basic result is a decomposition of protein structures into groups of contiguous amino acids called “sectors” which have been linked to conserved functional properties . This work provides conceptual and practical tools for sector analysis in any sufficiently well-represented protein family , and represents a necessary basis for broadly testing the concept of protein sectors .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "sequencing", "techniques", "split-decomposition", "method", "multiple", "alignment", "calculation", "mathematics", "coevolution", "protein", "sequencing", "algebra", "protein", "structure", "molecular", "biology", "techniques", "research", "and", "analysis", "methods", "amino", "acid", "analysis", "sequence", "analysis", "sequence", "alignment", "proteins", "biological", "databases", "molecular", "biology", "molecular", "biology", "assays", "and", "analysis", "techniques", "protein", "structure", "comparison", "biochemistry", "sequence", "databases", "linear", "algebra", "computational", "techniques", "database", "and", "informatics", "methods", "biology", "and", "life", "sciences", "physical", "sciences", "evolutionary", "biology", "eigenvalues", "evolutionary", "processes", "macromolecular", "structure", "analysis" ]	2016	Evolution-Based Functional Decomposition of Proteins
We show that two host-encoded primary RNAs ( pri-miRs ) and the corresponding microRNA ( miR ) clusters – widely reported to have cell transformation-associated activity – are regulated by EBNA3A and EBNA3C . Utilising a variety of EBV-transformed lymphoblastoid cell lines ( LCLs ) carrying knockout- , revertant- or conditional-EBV recombinants , it was possible to demonstrate unambiguously that EBNA3A and EBNA3C are both required for transactivation of the oncogenic miR-221/miR-222 cluster that is expressed at high levels in multiple human tumours – including lymphoma/leukemia . ChIP , ChIP-seq , and chromosome conformation capture analyses indicate that this activation results from direct targeting of both EBV proteins to chromatin at the miR-221/miR-222 genomic locus and activation via a long-range interaction between enhancer elements and the transcription start site of a long non-coding pri-miR located 28kb upstream of the miR sequences . Reduced levels of miR-221/miR-222 produced by inactivation or deletion of EBNA3A or EBNA3C resulted in increased expression of the cyclin-dependent kinase inhibitor p57KIP2 , a well-established target of miR-221/miR-222 . MiR blocking experiments confirmed that miR-221/miR-222 target p57KIP2 expression in LCLs . In contrast , EBNA3A and EBNA3C are necessary to silence the tumour suppressor cluster miR-143/miR-145 , but here ChIP-seq suggests that repression is probably indirect . This miR cluster is frequently down-regulated or deleted in human cancer , however , the targets in B cells are unknown . Together these data indicate that EBNA3A and EBNA3C contribute to B cell transformation by inhibiting multiple tumour suppressor proteins , not only by direct repression of protein-encoding genes , but also by the manipulation of host long non-coding pri-miRs and miRs . Epstein-Barr virus ( EBV ) is a gamma-herpesvirus etiologically linked to several B cell malignancies in humans , including Burkitt lymphoma ( BL ) , Hodgkin lymphoma ( HL ) and diffuse large B cell lymphoma ( DLBCL ) . Primary infection with EBV is usually asymptomatic in early childhood , but if delayed until adolescence it may manifest as a benign lymphoproliferative syndrome known as infectious mononucleosis ( IM ) [1] . After primary infection the virus persists in a latent state in a memory B cell population for the lifetime of infected individuals [2 , 3] . Approximately 90% of the adult human population is latently infected with EBV . Moreover , in vitro , EBV has the unique capacity to infect , activate and induce the continuous proliferation ( also known as “transformation” or “immortalisation” ) of quiescent B cells leading to the establishment of lymphoblastoid cell lines ( LCLs ) [1 , 4] . These cells carry the viral genome as extrachromosomal episomes from which only nine latency-associated proteins are expressed [the latency III program—six nuclear proteins ( EBNAs 1 , 2 , 3A , 3B , 3C , LP ) and three membrane proteins ( LMP1 , LMP2A , LMP2B ) ] along with several RNA species . The latter include Bam H1 A rightward transcript ( BART ) RNAs that are processed to produce ~20 miRs . These latency-associated gene products act together to activate the quiescent B cells into the cell cycle and maintain their proliferation [1 , 4] . EBNA3A , EBNA3B and EBNA3C are three viral proteins encoded by genes that probably arose from gene duplication events during the evolution of EBV since they are adjacent in the viral genome and all have a similar gene structure—short 5’ exon and long 3’ exon [5 , 6] . However , despite this assumed common origin , these three proteins share only limited amino acid sequence homology and have distinct functions . Genetic studies initially indicated EBNA3A and EBNA3C , but not EBNA3B , are required and essential for B cell transformation [7 , 8] . Subsequently it was shown that LCLs expressing a conditionally active form of EBNA3A or EBNA3C proteins fail to proliferate when either EBNA3A or EBNA3C is non-functional [9 , 10] . However , more recently many LCLs have been generated using recombinant EBV from which EBNA3A has been deleted—EBNA3A-knockout ( KO ) viruses [11 , 12] . Nevertheless , EBNA3A is still believed to play an important role in B cell transformation , since cell lines deficient in EBNA3A –at least in the early stages of outgrowth—tend to exhibit reduced rates of proliferation and undergo changes in host gene expression as they become established [11 , 12] . In contrast to EBNA3A and EBNA3C , not only is EBNA3B unnecessary for B cell transformation in culture , in vivo it behaves as a tumour suppressor—apparently attenuating the oncogenic potential of EBV [13] . The EBNA3s are well established as regulators of transcription ( reviewed in [14] . It seems that none of the EBNA3s bind directly to DNA and that they exert their effects on transcription through association with cellular transcription factors such as RBP-JK/CBF1 , PU . 1 , SPI1 , BATF and IRF4 [15 , 16 , 17 , 18 , 19 , 20 , 21] . EBNA3A and EBNA3C also interact with and recruit cellular factors associated with the covalent modification of histones such as histone deacetylases ( HDACs ) , histone acetyltransferases ( eg p300 ) , CtBP and components of the polycomb group protein repressor complexes [12 , 22 , 23 , 24 , 25 , 26 , 27] . It has also recently been shown that they can regulate gene expression through the modulation of chromatin looping between distal regulatory elements and gene transcription start sites ( TSS ) [20] . Chromatin immunoprecipitation coupled to high throughput DNA sequence ( ChIP-seq ) analyses have identified many thousands of specific genomic loci where the EBNA3s can be detected—many of these sites overlap for EBNA3A and EBNA3C binding [19 , 20 , 28 , 29] . Probably related to this co-localisation on chromatin , independent microarray and follow-up studies revealed that EBNA3A and EBNA3C extensively cooperate in the regulation of many cellular genes [14 , 20 , 21 , 29] . Well characterised target genes include those encoding important survival and cell cycle regulators such as the pro-apoptotic , BH3-only protein BIM and the cyclin-dependent kinase inhibitors ( CDKIs ) p16INK4a and p15INK4b [12 , 27 , 30 , 31 , 32] . These are repressed by the combined action of EBNA3A and EBNA3C and this is probably necessary to enhance survival , prevent cell cycle arrest and inhibit cell senescence early after the infection of primary B cells by EBV ( reviewed in [31] ) . Repression of p16INK4A expression appears to be a particularly important function of EBNA3C early during the infection and transformation of B cells [30 , 31] . In addition to their well-established role in regulating the expression of cellular protein-encoding transcripts , we wanted to investigate whether EBNA3A and EBNA3C could also modify the expression of non-coding RNAs , particularly miRs that could also contribute to the B cell transformation process and EBV latency . MiRs are a class of endogenous , short ( ~22 nucleotides ) , non-coding RNAs that play important roles in regulating many physiological processes including apoptosis , cell proliferation , differentiation and oncogenesis , by controlling gene expression at post-transcriptional levels . Most mammalian miRs are initially transcribed by RNA polymerase II ( Pol II ) that generates primary miRNA transcripts ( pri-miRs ) , which are then cleaved by the nuclease , Drosha , into ~79-nt precursor miRNAs ( pre-miRs ) and exported into the cytoplasm . Once in the cytoplasm , these pre-miRNAs are processed into mature miRNAs by the Dicer nuclease and incorporated into the RNA-induced silencing complex ( RISC ) to target specific messenger RNAs ( mRNAs ) leading to either repression of translation or degradation of mRNA or often both ( reviewed in [33 , 34] ) . Each miR species can generally target a large number of different mRNAs [35] and more than one species of miR can target mRNA from a single gene . There is accumulating evidence indicating that miRs are major regulators in the initiation and progression of human cancer by acting as either tumor suppressor or oncogenic miRs ( oncomiRs , [36 , 37] ) . Moreover , various studies indicate that growth-transforming viruses , including EBV , can encode mimics of , and/or modulate the expression of host cell miRs ( for example [4 , 38 , 39 , 40 , 41 , 42] ) . MiR-221 and miR-222 are highly conserved , co-expressed miRs encoded as a cluster located on chromosome X and have been reported to be overexpressed in many types of cancer [43] , including thyroid carcinoma [44] , glioblastoma [45] , prostate carcinoma [46 , 47] , bladder cancer [48] , pancreatic cancer [49] , hepatocellular carcinoma [50] , acute myeloid leukemia [51] and diffuse large B cell lymphoma [52 , 53 , 54] . Meta-analysis performed on over 1000 assorted human tumours , suggests that elevated expression of miR-221 and miR-222 is associated with poor overall survival of many cancer patients [55] . This well characterised oncogenic activity is likely to be related to the ability of miR-221/miR-222 to regulate cell cycle progression by directly targeting mRNA corresponding to CDKIs p57KIP2 ( CDKN1C ) and p27KIP1 ( CDKN1B ) [50 , 56 , 57 , 58 , 59] . In contrast to miR-221/miR-222 , miR-143 and miR-145 are tumour suppressor miRs that have been reported to inhibit the proliferation of many cancer-and non-cancer-derived cell lines . It has also been suggested that they might play roles in cell senescence [60 , 61 , 62] . MiR-143/miR-145 coding sequences are located in a cluster on chromosome 5 and are co-transcribed as a single pri-miR transcript [61 , 63] . Their reduced expression has been observed in a wide range of tumours , including gastric cancer [64] , colorectal cancer [65] , cervical cancer [66] , lung cancer [67] , breast cancer [68] , nasopharyngeal carcinoma [69] , bladder cancer [70] , prostate cancer [71] ovarian cancer [72] , hepatocellular carcinoma [73] and some B cell malignancies [74] . However , although various target mRNAs have been described in these reports , none have been particularly well characterised and to our knowledge no B cell-specific targets have been described . Here—following a relatively unbiased array screen for miRs regulated by EBNA3A and/or EBNA3C in the context of latent infection with EBV—we identified the oncogenic miR-221/miR-222 cluster as being activated and the tumor suppressor miR-143/miR-145 cluster as being repressed by EBNA3A together with EBNA3C . Further characterisation revealed that up-regulation of miR-221/miR-222 –resulting from the transactivation of a 28kb long non-coding pri-miR—was associated with almost complete ablation of p57KIP2 expression in EBV-infected B cells . In order to determine whether EBNA3A and EBNA3C regulate host cell miR levels in the context of latently infected B cells , the expression of 377 human , biologically active , mature miRs was examined using Taqman real-time PCR low density arrays ( TLDA ) to analyse two LCLs conditional for EBNA3C function ( 3CHT lines ) cultured for 28 days with or without the activating ligand 4HT and two EBNA3A-KO LCLs and lines established with the respective revertant virus ( REV ) . Several cellular miRs appeared to be regulated by either EBNA3A or EBNA3C or both . The total set of data acquired was screened for leads to be followed-up by quantitative real-time PCR ( qPCR ) measurements , but was not subjected to statistical analysis . Positive leads that were of particular interest—because they have been reported in the literature to have either oncogenic activity ( the miR-221/miR-222 cluster ) or tumour suppressor activity ( the miR-143/miR-145 cluster ) –were chosen for more detailed analysis . Both of these clusters are well conserved in vertebrate evolution [43 , 75 , 76] . Fig 1A shows the results of qPCR assays for miR-221/miR-222 in extracts from four independent EBNA3A-KO LCLs and four LCLs established with revertant viruses ( and therefore expressing all the latency-associated EBV proteins ) . Consistently , failure to express EBNA3A resulted in a large reduction in miR-221 and miR-222 expression ( Fig 1A and S1 Fig ) . Similarly using two independent LCLs conditional for EBNA3C function ( 3CHT , established in a p16-null B cell background in order to allow the cells to proliferate in the absence of EBNA3C , as described in [30] ) , it was shown that removal of the activating ligand ( 4HT ) resulted in a less substantial , but clearly significant reduction in both miR-221 and miR-222 expression ( Fig 1A and S1 Fig ) . Analysis of the same lines for expression of miR-143 and miR-145 confirmed the TLDA result showing that in the absence of EBNA3A or functional EBNA3C ( by washing out 4HT ) there was an increase in the expression of miR-143 and miR-145 ( Fig 1B ) . When EBNA3A was deleted there was particularly robust expression of both miR-143 and miR-145 . When conditional EBNA3C was inactivated there was a rather modest , but still significant and reproducible increase in both miRs . Consistent with this , when 4HT was added to a p16-null EBNA3C-conditional LCL that had been established in its absence ( never HT ) , there was a substantial repression of the miR-143/miR-145 cluster ( Fig 2A and S1 Fig ) . The differential expression of miRs are not due to the 4HT treatment since no significant change in miR expression was detected in two wild-type LCLs ( D11 and D13 LCL WT ) treated with 4HT for 30 days ( S2 Fig ) . Control RNAs RNU48 and ALAS1 were unaffected by the EBNA3A or EBNA3C status of the LCLs ( S3 Fig ) . Expression of protein-encoding gene clusters previously reported to be regulated by EBNA3A and EBNA3C ( eg CXCL9/CXCL10 and ADAM28/ADAMDEC1 ) was as expected from previous reports ( [20 , 21 , 28] , Fig 2B and S3 Fig ) . Although in some knockout and revertant LCL pairs , EBNA3B expression appeared to influence the levels of these miR clusters , the changes were very slight and/or inconsistent ( S1 and S4 Figs ) . We , and others , have found that when EBNA3A-KO LCLs are produced there is a tendency for the selection of changes in gene expression as the lines become more clonal ( for example loss of/reduced retinoblastoma ( Rb ) expression has been reported in independent studies [11 , 12] ) . In order to establish that the changes in miR expression highlighted by the TLDA ( and subsequently confirmed by qPCR ) were due to direct regulation of transcription by EBNA3A –rather than the result of selection during clone development—it was necessary to construct and validate an EBV recombinant that is conditional for EBNA3A ( EBNA3A-ERT2 ) and use this to produce new LCLs . EBNA3A-ERT2 is very similar to the EBNA3C conditional virus used in the initial assays ( Fig 1 and [12 , 30] ) , but has a fusion of the C-terminus of EBNA3A with a slightly more modified estrogen-receptor that responds to 4HT but not estrogen ( see Material and Methods ) . LCLs established with these viruses were validated for the expression of EBNA3A and other EBV latency-associated proteins by western blotting; in the absence of 4HT the EBNA3A-ERT2 fusion protein is almost completely degraded , but the expression of other latency-associated EBV proteins was unaltered ( Fig 3 ) . Several EBNA3A-ERT2 LCLs produced from two different individual B cell donors ( D11 for LCLs number 1–2 and D13 for cell lines number 3–4 ) and a mixed donor population of B cells ( LCL number 5 ) were established in the presence of 4HT and then analyzed ~30 days after removal ( -4HT ) or leaving in 4HT ( +4HT ) ( Fig 4A ) . Consistently on removal of 4HT ( washed ) , miR-221 and miR-222 were expressed at a lower level ( Fig 4B ) , whereas miR-143 and miR-145 were modestly induced ( Fig 4C ) . As with the experiments using conditional EBNA3C , when EBNA3A-conditional cells that had been grown into LCLs in the absence of 4HT ( never HT ) , there was a substantial repression of miR-143 and miR-145 when 4HT was added to the culture medium ( Fig 5A and S1 Fig ) . All these experiments showed that—like EBNA3C –active EBNA3A is necessary for the regulation of both miR clusters , ruling out the possibility of clonal selection as an explanation for the changes in expression seen in the EBNA3A-KO lines . EBNA3A-ERT2 function was further validated by qPCR for expression of previously characterised EBNA3A target genes ( CXCL9 and CXCL10 [21] , Fig 5B and S5 Fig ) . MiR-221 and miR-222 , which together form a cluster , are thought to both be processed from a common pri-miR . Interestingly , three different species of pri-miR-221/222 , approximately 2kb , 28kb and 108kb , have been described [51] . The expression of these three pri-miR-221/222 differs in different cell lines , however , publically available RNA-seq data from ENCODE revealed that in GM12878 ( an EBV-immortalised LCL ) the major pri-miR-221/222 to be expressed is the 28kb species ( Fig 6A ) . Using LCLs described above ( Figs 1 and 4 ) it was established that both EBNA3A and EBNA3C are necessary to up-regulate the 28kb pri-miR-221/222 ( Fig 6B and 6C ) . The level of pri-miR-221/222 in those cells echoes the level of the mature miR-221/miR-222 detected ( compare Figs 1 and 4 with Fig 6 ) . Consistent with this , when 4HT was added to EBNA3A-ERT2 LCLs or EBNA3C-HT LCLs never HT , there was a significant up-regulation of the pri-miR-221/222 ( Fig 6D and 6E ) . The activation of both EBNA3A and EBNA3C through addition of 4HT not only up-regulates the pri-miR-221/222 , but also increases the expression of the mature miR-221 and miR-222 in these cells ( S6 Fig ) . Furthermore , it was possible to show that EBNA3A and EBNA3C repress the well-characterised pri-miR-143/145 in the same LCLs ( S7 Fig ) . EBNA3A and EBNA3C are viral transcription factors that can be targeted to host genes at sites proximal to transcription start sites ( TSS ) and/or distal regulatory elements and sometimes modulate looping of chromatin between these sites to modify gene expression [19 , 20 , 21 , 29] . Therefore , in order to determine whether the regulation of miR-221/miR-222 and/or miR-143/miR-145 might result from direct binding of either EBNA3A or EBNA3C –or both—to chromatin at the genomic locus of each miR cluster , genome-wide chromatin immunoprecipitation ( ChIP ) data sets were interrogated . ChIP-seq was performed using D11 LCLs expressing either an epitope-tagged EBNA3A ( 3A-TAP ) or an epitope-tagged EBNA3C ( 3C-TAP ) and the immunoprecipitation ( IP ) was performed using an anti-FLAG antibody; this was followed by high throughput DNA sequencing ( S8 Fig and K . Paschos et al , manuscript in preparation; [27] ) . Analysis of the genomic locus including miR-221/miR-222 ( chromosome Xq11 . 3 ) revealed a region located approximately 9kb downstream of the TSS for the 28kb pri-miR-221/222 that includes three binding sites for EBNA3C ( sites BS2a , BS2b and BS3 in Fig 7A ) . One of these sites precisely overlapped an EBNA3A-binding site ( BS2b ) and one partially overlapped ( BS3 ) . Sites BS2a and BS2b are spaced only 1kb apart at a location previously reported to be a cis-acting enhancer element involved in the regulation of both miR-221 and miR-222 [47] . An additional EBNA3C-only binding site ( BS1 ) was located about 60kb downstream of the pri-miR-221/222 TSS ( Fig 7A ) . Robust binding of EBNA3C-TAP to sites BS1 , BS2a , BS2b , and BS3 was confirmed by ChIP-qPCR ( Fig 7B ) , but no binding was observed using control primers corresponding to another region previously described as an enhancer ( EnhA [47] , Fig 7A ) Under similar ChIP-qPCR conditions , and again using the anti-FLAG antibody , significant binding of EBNA3A-TAP could also be detected in BS2a , BS2b and BS3 , whereas no binding was detected for site BS1 or EnhA ( Fig 7C ) . Taken together , these results identified an ‘intragenic’ region where both EBNA3A and EBNA3C bind to chromatin ( BS2a , BS2b and BS3 ) . ENCODE data ( displayed on the UCSC genome browser ) shows this region has high levels of the activation associated histone modification H3K27ac—providing further evidence that it probably acts as an enhancer of transcription in LCLs ( Fig 7A ) . In contrast , interrogation of ChIP-seq data corresponding to the miR-143/miR-145 locus ( chromosome 5q . 32 ) –over a region of more than 1Mb either side of the putative TSS of the pri-miR-143/145 –failed to reveal any EBNA3A- or EBNA3C-binding sites ( compare S9 Fig with Fig 7A ) . Our interpretation of these data is that transcriptional regulation of pri-miR-143/145 ( and hence mature miR-143/miR-145 ) is unlikely to be due to EBNA3A/EBNA3C binding to cis-regulatory elements and is therefore probably a secondary , trans-acting effect of the regulation of an unknown gene ( s ) . However , we cannot rule out binding to extremely long-range regulatory elements . Next , in order to determine whether the up-regulation of pri-miR-221/222 by EBNA3A and EBNA3C correlates with histone modification and chromatin remodeling , ChIP analyses were performed on EBNA3A-KO and EBNA3A-REV LCLs as well as EBNA3C-HT LCLs ( never HT ) or treated with 4HT ( Fig 8 ) . Initially the phosphorylation of the RNA polymerase II at serine-5 ( Ser5 ) –that indicates transcriptional initiation and activation [77]–was investigated ( Fig 8B and 8C ) . This revealed that the level of phospho-Ser5 Pol II is elevated around the TSS of 28kb pri-miR-221/222 only when both EBNA3A and EBNA3C are expressed and functional ( in EBNA3A-REV and EBNA3C-HT cultured with 4HT ) . No binding was seen at the putative 2kb TSS . Primers that amplify CXCL10 TSS and ADAM28 TSS were used as controls for EBNA3A and EBNA3C repressed genes respectively and found , as expected , a higher level of phospho-Ser5 Pol II only when EBNA3A or EBNA3C were absent/non-functional . We then performed ChIP analysis for marks of active chromatin ( H3K4me3 , H3K9ac and H3K27ac ) across the miR-221/miR-222 cluster locus ( Fig 8B and 8C ) . As expected , a higher level of activation marks was found around the 28kb pri-miR-221/222 TSS only when functional EBNA3A and EBNA3C were expressed . Again there were no changes around the putative 2kb TSS that suggest it is regulated . Interestingly , histone activation marks where higher ( in particular H3K9ac and H3K27ac ) at BS2a and BS2b sites , that is the region previously described as an enhancer for miR-221/miR-222 , only when both EBV proteins are functional . Taken together these data are consistent with increased miR-221/miR-222 expression occurring when EBNA3A and EBNA3C are expressed in an active form , bind chromatin at specific sites and alter the epigenetic profile of the locus . It has recently been shown that the EBNA3 viral proteins can regulate transcription by modulating enhancer-promoter loop formation [20] . So in order to determine whether the EBNA3s could either promote or disrupt the formation of looping between ‘intragenic’ enhancer elements ( BS2 and BS3 –where both EBNA3A and EBNA3C bind ) and the promoter of the 28kb pri-miR-221/222 , chromosome conformation capture ( CCC ) analysis was performed . A schematic map of the miR-221/miR-222 genomic locus with the location of the HindIII restriction sites and PCR primers is shown in Fig 9A . The CCC results showed looping interactions between regions BS2 and BS3 and the promoter , only in EBNA3A-REV LCL and EBNA3C-HT LCL treated with 4HT ( Fig 9B and 9C ) ; that is in cells in which both EBNA3A and EBNA3C are active and the 28kb pri-miR-221/222 is up-regulated . No looping was found between a control region ( NC ) and the promoter for 28kb pri-miR-221/222 , but a control PCR product L1/L2 was found in all samples showing that equal amounts of DNA were present in all reactions ( Fig 9D ) . These results demonstrate that EBNA3A and EBNA3C are both required for the formation of looping between two sites within an enhancer region and the 28kb pri-miR-221/222 promoter , leading to increased transcription of the pri-miR ( Fig 9E ) . MiR-221 and miR-222 have been described as oncogenic miRs ( oncomirs ) because they are often expressed at high levels in cancer ( see Introduction ) . Furthermore it has been demonstrated in various types of non-B cell that they can target mRNAs corresponding to several tumour suppressor genes and promote their translational inhibition and/or degradation . There are multiple reports that the CDKIs p57KIP2 and p27KIP1 are targets and a single report of the pro-apoptotic p53-response protein PUMA in epithelial cells [50 , 56 , 57 , 58 , 59 , 76] . We first determined whether p57KIP2 and p27KIP1 were miR-221/miR-222 targets in LCLs . To do this EBNA3A-REV cells were electroporated with LNA anti-miR-221 , anti-miR-222 , both anti-miRs , or a negative control . The electroporation of anti-miR-221 , anti-miR-222 or both was accompanied by an increase in p57KIP2 protein level , and p27KIP1 increased only when miR-221 was inhibited ( Fig 10 ) . For comparison we also analysed expression of the related CDKI , p21CIP1 ( not known to be a miR-221/miR-222 target ) and found no change when the miRs were inhibited . The level of PUMA was also unaltered in this B cell context . These depletion experiments relied upon poor transfection efficiencies common to all LCLs , therefore in most cells of the population the specific miRs were not inactivated . Nevertheless taken together , the data established to our satisfaction that p57KIP2 and ( to a lesser extent ) p27KIP1 are targets of the miR-221/miR-222 cluster in EBV-transformed human B cells . The expression of p57KIP2 was therefore subjected to more detailed and stringent analyses . The results , ( compiled in Fig 11 ) , unambiguously demonstrated across multiple LCLs carrying either EBNA3A-KO or-revertant EBV , or LCLs conditional for EBNA3A or EBNA3C , that p57KIP2 protein expression is almost completely ablated when EBNA3A and EBNA3C are both active . However , if either of these EBNA proteins is absent or inactivated , substantial amounts of p57KIP2 can be detected . Analysis by qPCR also showed significant increases of mRNA corresponding to p57KIP2 in the absence of functional EBNA3A or EBNA3C , but the degree of regulation was rather more variable between cell lines than was the protein expression ( Fig 11 ) . These results suggest that miR-221 and miR-222 not only block translation , but might also enhance the degradation of p57KIP2 mRNA ( both mechanisms of action have been described [33 , 34] . MiR-221 and miR-222 have also been reported to regulate p27KIP1 in non-B cells . In our miR inhibition assay ( Fig 10 ) we showed that p27KIP1 increased after miR-221 inhibition but not miR-222 . Although we see regulation of p27KIP1 protein levels by EBNA3A in LCLs , it is neither as robust nor quite as consistent as the regulation of p57KIP2 ( for examples see S10 Fig ) —currently we do not know the reasons why p27KIP1 and p57KIP2 are differentially regulated in these lines . In order to determine the consequences of up-regulating p57KIP2 in LCL cells that have grown out after infection of normal primary B cells with the minimum of selection , an early passage ( <2 months post-infection ) EBNA3A-ERT2 line produced from a mixed donor population of B cells ( LCL 5 ) was used . This line was established in the presence of 4HT and the cells express little or no p57KIP2 , but on removal of 4HT from the culture medium , they soon produce substantial amounts of p57KIP2 protein ( Figs 11B and 12A ) . Consistent with the increase in p57KIP2 there was a pronounced reduction in the phosphorylation of the tumour suppressor Rb and this was associated with a gradual reduction in proliferation as revealed by reduced DNA synthesis ( EdU incorporation ) and reduced cell population growth ( Fig 12 ) . Similar analysis was repeated on several other independent EBNA3A-conditional LCLs and produced essentially identical results ( S11 Fig ) . An important enzyme necessary for the phosphorylation of Rb is the cyclin-dependent kinase CDK6 . Therefore the most likely explanation for the inhibition of Rb phosphorylation is the binding of p57KIP2 to CDK6 resulting in inactivation of the latter ( see co-immunoprecipitations in S12 Fig ) . The effect of low levels of miR-221/miR-222 and high levels of p57KIP2 on the proliferation of these cells was surprisingly modest , rather less than has been reported previously in EBNA3A-conditional LCLs when EBNA3A was inactivated [10 , 78]; we do not know the reason for this , but as we have indicated above , different lines can have different properties because of clonal variation . The roles and interactions of the various CDKIs regulated by EBV are discussed in more detail below . Using a reverse genetics approach , made possible by the use of LCLs carrying knockout , revertant or conditional-EBV recombinants , we have explored the roles of the EBNA3 proteins in the regulation of cell miRs in B cells . This has revealed that both EBNA3A and EBNA3C –but not EBNA3B –are required for the transactivation of the oncomiRs miR-221 and miR-222 , while concurrently silencing the expression of the tumour suppressor miR-143/miR-145 cluster . Both EBNA3A and EBNA3C were shown by ChIP experiments to associate with multiple sites in a genomic region about 19kb upstream of the miR-221/miR-222 coding sequences and about 9kb downstream from the pri-miR TSS that is dominant in LCLs . These EBNA3 binding sites correspond to a region previously identified functionally and by histone modifications as an enhancer of transcription . So the data are consistent with the miRs being directly transactivated by the combined action of EBNA3A and EBNA3C . This cooperation between these two EBNA3 proteins to modulate transcription is reminiscent of the regulation of many host protein-encoding genes in EBV-infected LCLs [14 , 20 , 28 , 29] . At this stage it is not possible to say what factors are responsible for recruiting EBNA3A and or EBNA3C to BS2a , BS2b and BS3 , nevertheless ENCODE data indicate >20 transcription factors that could bind at these sites in the GM12878 LCL ( S13 Fig ) . This list includes many factors previously reported to be involved in the recruitment of EBNAs to sites across the human genome ( eg ATF2 , BATF , PAX5 , RUNX3 and SPI1 [19 , 20 , 29] ) . Since EBNA3A and EBNA3C were found to bind to multiple sites in a previously characterised enhancer element—we assumed , and then confirmed , that transactivation involves modulation of the local three-dimensional architecture of chromatin ( long-range ‘looping’ ) that brings the enhancer elements into contact with the TSS of the 28kb pri-miR only when functional EBNA3A and EBNA3C are present ( see Fig 9 ) . Similar topological changes have been reported in the regulation of protein encoding genes such as the ADAM28/ADAMDEC1 locus [20 , 21] . However , at most of the EBNA3A/EBNA3C regulated genes that have been well characterised [for example BIM ( BCL2L11 ) , p16INK4a ( CDKN2A ) and the ADAM28/ADAMDEC1 locus] transcription is repressed; and this is probably because these two viral proteins can recruit cellular co-repressors such as HDACs , CtBP and components of polycomb protein complexes ( see Introduction ) . On some genes they can also displace the EBV transactivator EBNA2 , resulting in substantially reduced transcription [20 , 21] . Here , for the first time , we describe a long-range enhancer—promoter interaction mediated by EBNA3A and EBNA3C resulting in increased transcription , ie activation . But very little is known about how EBNA3A and EBNA3C together might activate transcription , although it is probably related to the their capacity to physically interact with each other [27] and perhaps recruit co-activators such as the histone acetyltransferase p300 [22] . This mechanism would be consistent with the increase in acetylation seen on histone H3 lysine-27 ( H3K27ac ) at the enhancer binding sites and around the promoter when the miR-221/miR-222 locus is activated ( Figs 7A and 8 ) . The consensus of opinion is that in this type of gene regulation—that involves chromatin looping—cellular repressors or activators are recruited in a context-specific manner , but what cofactors , features of genomic sequence and chromatin topology determine whether the outcome is repression or activation , remain largely unknown ( reviewed in [79] ) . It has been previously reported that EBV can induce expression of miR-221/miR-222 [80 , 81] and that the latency-associated protein LMP1 can activate miR-221/miR-222 expression after single gene transfer into BL-derived cells [82] . Moreover , the cluster can also be activated by NF-kB [47] . Since LMP1 is expressed in all the LCLs used in this study ( see for example Fig 3 ) and activates NF-kB signaling , we cannot rule out the possibility that signal transduction from this viral membrane protein also contributes to the activation of the 28kb pri-miR . It is possible that EBNA3A and EBNA3C acting together play a role in reorganizing local chromatin in order to potentiate LMP1 and NF-kB-mediated transactivation of miR-221/miR-222 . Silencing of the miR-143/miR-145 locus by the combined action of EBNA3A and EBNA3C remains poorly understood . Since no EBNA3 binding sites were detected within more than a million DNA base pairs either side of the pri-miR-143/145 TSS ( S9 Fig ) , it is likely that this repression of transcription is a secondary , downstream event triggered by altered expression of another EBNA3A/EBNA3C-target gene . However , we are currently unable to formally test this . The two miR clusters described here are deregulated in multiple human cancers , including B cell lymphomas ( see Introduction ) and therefore probably have the potential to influence B cell proliferation , transformation , and EBV-associated lymphomagenesis . The details of how the tumour suppressors miR-143 and miR-145 might inhibit cell proliferation are poorly understood . They co-operatively promote differentiation and repress proliferation in several cancer and primary cell lines and are both up-regulated during senescence in human fibroblasts [60 , 61 , 62] . It may be significant that miR-143 and miR-145 are also repressed by the E7 oncoprotein in epithelial cells infected with human papillomavirus ( HPV ) -31 [83] . Although various target mRNAs have been proposed , there appears to be a lot of cell-type specificity and a general lack of consensus on precisely how miR-143/miR-145 act as tumour suppressors . There are very few data available on the activities of miR-143/miR-145 in B cells , although there has been one report of their down-regulation in EBV-transformed , but not in mitogen-stimulated B cells [84] . In contrast to miR-143/miR-145 , functions of the miR-221/miR-222 cluster is relatively well characterised , with wide agreement that two of the major targets are mRNAs for CIP/KIP CDKIs p57KIP2 and p27KIP1 , the translation of which are robustly inhibited by these miRs in various types of cell ( see Introduction ) . We , and others , had previously shown that EBNA3A and EBNA3C can block transcription of two members of the INK4 CDKI family , p16INK4a and p15INK4b [12 , 30 , 32 , 78] . Furthermore it has been reported that these same two EBV proteins might also repress expression of a third member of the CIP/KIP family , p21CIP1 [85 , 86] . Since intrinsic cell cycle inhibitors are emerging as important targets of EBNA3A and EBNA3C , we focused on the consequences of miR-221/miR-222 induction in LCLs . Utilizing multiple cell lines carrying either knockout mutant viruses or viruses conditional for EBNA3A or EBNA3C expression , we have clearly established that in B cells , transformed by and latently infected with EBV , both EBNA3A and EBNA3C are necessary to inhibit expression of p57KIP2 and ( less robustly and reproducibly ) p27KIP1 . Both of these proteins can act as tumour suppressors by reducing cell proliferation because they target and inhibit CDKs 2 , 4 , 6 ( see schematic in Fig 13; reviewed in [87] ) . Reducing the expression of these CIP/KIP CDKIs via miR-mediated inhibition is therefore likely to enhance the proliferation of LCLs and might play a role in establishing EBV persistence in B cells in vivo . It could also contribute to the development of EBV-associated B cell lymphomas such as those in the immunocompromised , in DLBCL [52 , 53 , 54] and in the sub-group of BL that express the EBNA3s ( known as Wp-restricted BL [88] ) . It should also be noted that in many of the LCLs used in this study—particularly those passaged for prolonged periods—induction of p57KIP2 had little or no effect on proliferation . We suspect in some cases this is because genetic or epigenetic impairment of the Rb tumour suppressor hub is readily selected in the expansion of these rapidly proliferating cell populations ( see Introduction ) . Alternatively , p57KIP2 may not always have a significant anti-proliferative effect in activated B cells and may sometimes produce more subtle phenotypes . It is important to remember that EBNA3A and EBNA3C together repress multiple cell cycle inhibitory pathways that culminate in the inactivation—by phosphorylation—of the tumour suppressor protein Rb ( Fig 13 ) ; that is , there appears to be considerable redundancy . The currently available data suggest that p16INK4a is the major CDKI to be targeted in B cell transformation [30 , 31] , but it now appears that p15INK4b , p57KIP2 , p27KIP1 and p21CIP1 might all have to be controlled in the molecular balance required for the establishment of EBV latency . It will require painstaking analysis using EBV with specific mutations in EBNA3A and EBNA3C and shRNA and/or gene-editing technologies to dissect and resolve these effects on B cell proliferation , senescence and perhaps differentiation . In summary , it appears that during the co-evolution of EBV and its host , two cooperating factors ( EBNA3A and EBNA3C ) have emerged to control transcription of not only host cell genes , but also long non-coding pri-miRs and miRs . It is remarkable that two distinct families of cell cycle inhibitory factors—the INK4 and CIP/KIP CDKIs—are a specific focus in this double-routed regulation of the host proteome . The implication is that extending the proliferation of B cells via the activities of EBNA3A and EBNA3C is a very important feature of EBV biology in vitro and therefore in vivo . RNA was isolated using mirVana miRNA isolation Kit ( Ambion ) including the optional step for enrichment of small RNA ( <200bp ) , which enables more sensitive detection of low-level small RNAs . Taqman MicroRNA RT kit and Megaplex Primer Pool A ( ABI ) were used to reverse transcribe up to 381 microRNAs in a single reaction according to the manufacturer’s instructions . Generally 300ng of RNA was reverse transcribed per reaction and the cDNA product was used in qPCR without pre-amplification . Three hundred and seventy seven human microRNAs were profiled by real-time qPCR using the Taqman MicroRNA A Card v 2 . 0 ( ABI ) . cDNA was diluted , mixed with Taqman Universal PCR Master Mix II ( ABI ) and loaded into the pre-configured micro-fluidic card . Real-time reaction was run on a 7900HT Real-Time PCR System ( ABI ) and data analyzed using the SDS RQ manager software ( ABI ) . An EBNA3A-ERT2 fusion protein ( 3A-ERT2 ) was constructed in the B95-8 EBV background using the 4-hydroxytamoxifen-sensitive human estrogen receptor ERT2 containing the G400V/M543A/L544A triple mutation [89] . The connection between EBNA3A and ERT2 is a small linking sequence of 9 amino acids ( GTGGVGQD ) between the last amino acid of EBNA3A and amino acid 281 of ERT2 . This fusion was recombined into the B95-8 bacterial artificial chromosome ( BAC ) using previously described methods [12 , 90 , 91] to produce BAC containing 3A-ERT2 . Established LCLs were cultured in RPMI-1640 medium ( Invitrogen ) supplemented with 10% fetal calf serum , penicillin and streptomycin . LCL 3A-ERT2 and 3CHT were cultured with addition of 400nM of 4-hydroxytamoxifen ( 4HT , Sigma ) where stated . After the infection of primary B cells , LCLs were grown to a volume and density suitable for freezing multiple aliquots ( typically about 60ml at a density of 3×105 cells/ml or greater ) . This took 4–6 weeks for 3A-ERT2 , 3CHT , revertant LCLs and 6–12 weeks for the 3A-ERT2 grown without 4HT and EBNA3A mutant LCLs . LCLs 3CHT established in a p16-null background are described in [30] . Cells recovered from liquid nitrogen were cultured for at least 10 days ( with 4HT if necessary ) before the start of any experiment . At the end of an experiment the cells were discarded . Twenty-four hours before any experimental treatment , cells were seeded at a density of 3×105 cells/ml . Recombinant viruses were constructed and produced as described previously [12 , 91] . Primary B cells for the generation of LCL 3A-ERT2 and EBNA3-knockout and revertant LCLs were isolated from anonymous buffy coat residues ( UK Blood Transfusion Service ) by centrifugation over Ficoll . EBNA3A-knockout and revertant LCLs were made by infection of CD19+ B cells from four independent donors ( D1 , D2 , D3 , D4 ) . LCL 3A-ERT2 line 5 was made by infection of CD19+ B cells from mixed donors . D11 ( lines 1 and 2 ) and D13 ( lines 3 and 4 ) 3A-ERT2 LCLs were made by infection of PBLs isolated from donors D11 and D13 with EBNA3A-ERT2 virus . To produce all LCLs , between 50μl and 1ml of virus was added to 106 PBLs or 3×106 CD19+ purified B cells in a well of a 24 well plate , and cultured initially in RPMI , supplemented with 15% FCS and with Cyclosporine A ( 500 ng/ml ) for the first 2 weeks . Once LCLs had grown out into large culture volumes , the FCS level in the medium was reduced to 10% . Cell proliferation analyses were performed as described previously [30] by measuring the incorporation into DNA of nucleotide analogue EdU during a 2h pulse ( Life Technologies ) . Cell fluorescence was measured on LSR II ( Becton Dickinson ) flow cytometer . Single cells were gated based on FxCycle Far Red fluorescence—Life Technologies ( comparing fluorescence area to width or height at 633/690 ) . The LIVE/DEAD Fixable Violet stain ( Life Technologies ) was used to determine the viability of cells ( Fluorescence measured by 405/450 filters indicated live/dead status ) , and only live cells were included for the assessment of proliferation by EdU ( 488/530 ) . SDS polyacrylamide gel electrophoresis and Western blotting was performed essentially as described previously [12 , 27 , 91] . In some cases the membrane used for Western blotting was cut horizontally after protein transfer in order to facilitate multiple antibody probes and a single loading control for each blot . LCLs were harvested and lysed in immunoprecipitation ( IP ) buffer ( 50mM Tris-HCl pH 7 . 5 , 150mM NaCl , 1mM DTT and 0 . 5% Nonidet P-40 ) plus protease inhibitors ( Roche Molecular Biochemicals ) . Protein concentration was estimated colorimetrically using the Bio-Rad detergent-compatible assay and 250μg were used per IP . Cell extracts were then pre-cleared with 30μl of Protein G-Sepharose beads ( GE Healthcare ) at 4°C for 1h . Complexes were precipitated with specific antibodies ( S1 Table ) and the mixture was incubated at 4°C overnight . Then , 30μl of protein G-Sepharose beads were added for 1h at 4°C , washed four times in IP buffer and the immunopurified proteins were resolved by SDS-PAGE and detected by western blot . For qPCR , total RNA ( including microRNA ) was extracted from approximately 5×106 cells for each cell line using the miRNeasy mini kit from Qiagen and following the manufacturer's instructions . Expression of miR-221 , miR-222 , miR-143 , miR-145 and two snRNAs , RNU6B and RNU48 were quantified by qPCR using the TaqMan MicroRNA Assay listed in S2 Table ( Applied Biosystem ) . Briefly , cDNA was synthesised from 10ng of total RNA using Taqman miRNA primers and the TaqMan MicroRNA Reverse Transcription Kit . qPCR was then performed using the TaqMan Universal PCR Master Mix . The cycling conditions were 95°C for 10min , followed by 45 cycles of 15sec at 95°C and 60sec at 60°C . For mRNA analysis , one microgram of each RNA sample was reverse-transcribed using SuperScript III First-Strand Synthesis Supermix for qPCR ( Invitrogen ) . 10 ng of cDNA product was then used per qPCR reaction ( except for the detection of pri-miR-143/145 where 100ng were used ) using Platinum Sybr Green qPCR SuperMix UDG kit ( Invitrogen ) . Dissociation curve analysis was performed during each run to confirm absence of non-specific products . Sequences of the primers used are listed in S3 Table . All qPCR were performed on an ABI 7900HT real-time PCR machine . GNB2L1 and RNU6B were used as endogenous controls for mRNA and miR respectively . Relative mRNA or miR expression was calculated using the comparative Ct ( ΔΔCT ) method . The calculated errors in the graphs are the standard errors from three replicate qPCR reactions for each mRNA or miR . ChIP assay and qPCR analysis were performed essentially as described previously [27] . Antibodies and sequences of the primers used in these assays are listed in S1 and S4 Tables respectively . Chromosome conformation capture assay was performed as previously described [92 , 93] with minor modification . Briefly , ten millions LCL were filtered through a 70μm cell strainer to obtain a single-cell preparation and fixed in 1% formaldehyde for 30 minutes at room temperature . The fixation reaction was stopped by quenching with 0 . 125M glycine , cells were washed twice with cold PBS containing protease inhibitors , re-suspended in 500μL of lysis buffer and lysed for 10 minutes on ice . Nuclei were collected and digested with 400 units of HindIII ( 20 , 000 U/mL , New England Biolabs ) overnight at 37°C . The restriction digest reaction was stopped by addition of SDS ( 1 . 6% final concentration ) and incubation at 65° for 30 minutes . The intramolecular ligation was performed by adding 100 Weiss units of T4 DNA ligase to the 10-fold diluted sample for 4h at 16°C followed by a 45 minutes incubation at room temperature . Protein digestion and reverse cross-linking was performed with overnight incubation at 65°C with 300μg proteinase K . RNA was then degraded with 300μg RNase for 1h at 37°C . Finally the DNA was twice phenol/chloroform extracted and ethanol-precipitated . Purified DNA was analysed by conventional PCR . The generation of control template for ligation products was performed as previously described [20] . In brief , DNA regions covering the restriction sites of interest were PCR amplified , purified , mixed in equimolar amount and subjected to digestion with HindIII for 2 hours . The digested PCR products were ligated with 10 Weiss unit of T4 DNA ligase overnight at 16°C , purified and analysed by conventional PCR . LCLs 3A-REV were electroporated with 50 nM of LNA anti-miR-221 oligonucleotide ( hsa-miR-221 miRCURY LNA , Exiqon ) , LNA anti-miR-222 oligonucleotide ( hsa-miR-222 miRCURY LNA , Exiqon ) or scrambled oligonucleotide ( miRCURY LNA microRNA inhibitor control , Exiqon ) using a Bio-Rad Gene Pulser I ( 270V , 960μF ) . After 48 h , dead cells and debris were removed by layering the cells over 3ml Ficoll-plaque ( GE Healthcare ) . Live cells were then collected washed in PBS and extraction was performed as previously described for Western blotting . The primary human B cells used in this study were isolated from buffy-coat residues purchased from the UK Blood Transfusion Service; these were derived from the blood of anonymous volunteer blood donors . No ethical approval is required .	A relatively unbiased screen of human microRNAs ( miRs ) revealed that in EBV-transformed B cells , a miR cluster , miR-221/miR-222 , that is frequently up-regulated in cancer , is induced by the latent EBV only if the viral nuclear proteins EBNA3A and EBNA3C are both expressed . The same two EBV proteins silence a tumour-suppressor miR cluster miR-143/miR-145 . The induction of miR-221/miR-222 results from the activation of a long non-coding primary RNA ( pri-miR ) via long-range chromatin looping between enhancer elements that bind EBNA3A and EBNA3C and the transcription start site of the pri-miR . A well-established target of miR-221/miR-222 is the cyclin-dependent kinase ( CDK ) inhibitor p57KIP2 , which , because it can inactivate various CDKs , can inhibit cell proliferation—but might have additional functions in B cells . Since EBNA3A and EBNA3C also cooperate to repress the expression of at least two other inhibitors of CDKs ( p16INK4a and p15INK4b ) , this implies a degree of functional redundancy in the deregulation of cell cycle checkpoints by latent EBV . This study has shown for the first time that this capacity to reduce expression of multiple cell cycle inhibitors results not only from direct repression of protein-encoding genes , but also the activation of a long non-coding RNA and cluster of oncogenic miRs .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	Epstein-Barr Virus Proteins EBNA3A and EBNA3C Together Induce Expression of the Oncogenic MicroRNA Cluster miR-221/miR-222 and Ablate Expression of Its Target p57KIP2
It has been proposed that rotavirus infection promotes the progression of genetically-predisposed children to type 1 diabetes , a chronic autoimmune disease marked by infiltration of activated lymphocytes into pancreatic islets . Non-obese diabetic ( NOD ) mice provide a model for the human disease . Infection of adult NOD mice with rhesus monkey rotavirus ( RRV ) accelerates diabetes onset , without evidence of pancreatic infection . Rather , RRV spreads to the pancreatic and mesenteric lymph nodes where its association with antigen-presenting cells , including dendritic cells , induces cellular maturation . RRV infection increases levels of the class I major histocompatibility complex on B cells and proinflammatory cytokine expression by T cells at these sites . In autoimmunity-resistant mice and human mononuclear cells from blood , rotavirus-exposed plasmacytoid dendritic cells contribute to bystander polyclonal B cell activation through type I interferon expression . Here we tested the hypothesis that rotavirus induces bystander activation of lymphocytes from NOD mice by provoking dendritic cell activation and proinflammatory cytokine secretion . NOD mouse splenocytes were stimulated with rotavirus and assessed for activation by flow cytometry . This stimulation activated antigen-presenting cells and B cells independently of virus strain and replicative ability . Instead , activation depended on virus dose and was prevented by blockade of virus decapsidation , inhibition of endosomal acidification and interference with signaling through Toll-like receptor 7 and the type I interferon receptor . Plasmacytoid dendritic cells were more efficiently activated than conventional dendritic cells by RRV , and contributed to the activation of B and T cells , including islet-autoreactive CD8+ T cells . Thus , a double-stranded RNA virus can induce Toll-like receptor 7 signaling , resulting in lymphocyte activation . Our findings suggest that bystander activation mediated by type I interferon contributes to the lymphocyte activation observed following RRV infection of NOD mice , and may play a role in diabetes acceleration by rotavirus . Type 1 diabetes is a chronic autoimmune disease marked by infiltration of immune cells into pancreatic islets and destruction of insulin-secreting β cells [1] . Diabetes development is associated with specific high-risk human leukocyte antigen haplotypes [2] . However , genetic susceptibility cannot explain the discordance between monozygotic twins , seasonality of disease , rising incidence and trend towards a younger age of onset [3] . Environmental factors such as dietary proteins , intestinal microbiota and virus infections are implicated in diabetes development [4] , [5] . Widely studied virus modulators of diabetes include enteroviruses [6] , particularly coxsackieviruses [7] . In addition , rotavirus infection in children genetically at-risk of type 1 diabetes is associated with increased islet autoantibody levels and has been proposed to accelerate progression to diabetes [8] , [9] . Virus-mediated acceleration of diabetes development is proposed to occur by three distinct but not mutually-exclusive mechanisms: direct pancreatic infection , T cell molecular mimicry and bystander activation [10] . In the absence of direct β cell infection and lysis , pancreatic infection and molecular mimicry would lead to T cell activation via antigen presentation on the major histocompatibility complex ( MHC ) . However , bystander activation would involve polyclonal lymphocyte activation by cytokine-secreting antigen-presenting cells ( APCs ) . Thus , bystander activation is not antigen-specific , and depends on the presence of autoreactive B and T cells . The host sites targeted by virus and the timing of infection in relation to the degree of islet autoimmunity would influence the likelihood of bystander activation . Rhesus monkey rotavirus ( RRV ) infection in adult non-obese diabetic ( NOD ) mice induces early diabetes onset , but this does not involve pancreatic infection [11] . Instead , this diabetes acceleration is associated with a Th1-biased antibody and cytokine response [12] . As rotavirus infection accelerates diabetes only in mice with established insulitis , the presence of autoreactive cells is required for this process [11] , [13] . Infectious RRV is present in the mesenteric lymph nodes ( MLN ) of NOD mice , where its APC association correlates with increased MHC expression [12] . The pancreatic lymph nodes ( PLN ) , where islet-autoreactive cells accumulate , also contain infectious RRV [12] . Dendritic cells ( DC ) and T cells in the PLN of RRV-infected NOD mice express increased pro-inflammatory cytokine levels , and B cells show increased MHC I expression [14] . Rotavirus does not associate with B or T cells in MLN or PLN of NOD mice [12] . We have proposed that RRV-activated APC induce bystander activation of lymphocytes , which contributes to diabetes acceleration following RRV infection of NOD mice [14] . Infection with murine rotavirus , but not porcine rotavirus CRW-8 , also accelerates diabetes development in NOD mice [11] , [12] , [14] . The potential for bystander activation following exposure to rotavirus has been demonstrated previously in non-autoimmune systems . Polyclonal B cell activation occurs after exposure of isolated murine splenocytes and human peripheral blood mononuclear cells ( PBMC ) to RRV , and is independent of virus replication [15] , [16] . B cell activation is independent of the rotavirus or mouse strain , and prevented by rotavirus outer capsid removal or blockade with neutralising antibodies to outer capsid protein VP7 [15] . Recently , RRV stimulation of human plasmacytoid DC ( pDC ) , and their secretion of type I interferon ( IFN ) , was shown to be required for human B cell activation [17] . Type I IFN signaling and the presence of pDC also contributed to B cell activation in C57BL/6 mice infected with murine rotavirus [17] . RRV-exposed but not productively infected human pDC secrete the type I IFN , IFNα [18] . This pDC activation requires rotavirus structural proteins and its double-stranded ( ds ) RNA genome . Depletion of human pDC in PBMC cultures reduces the frequency of rotavirus-specific T cells expressing IFNγ after rotavirus exposure , so pDC also are important for T cell activation [19] . As IFNα stimulation is not T cell receptor-specific , it is likely that pDC depletion also reduces the activation of T cells specific for antigens other than rotavirus . Therefore , IFNα production by RRV-exposed pDC contributes to non-specific B and T cell activation in non-autoimmune human and mouse model systems . In naive NOD mice , the numbers of IFNα-producing pDC increase in PLN at 3 to 4 weeks of age , and antibody blockade of the type I IFN receptor ( IFNAR ) prior to this age delays and significantly reduces the incidence of diabetes [20] , [21] . Blockade of IFNα expression by pDC also reduces the frequency and activation of islet-specific CD8+ T cells in PLN [22] . Furthermore , pDC depletion significantly reduces diabetes incidence in NOD mice [22] . As IFNα expression by pDC appears to play an important role in diabetes development , it is reasonable to propose that augmentation of pDC responses following rotavirus infection might accelerate diabetes development . Supporting this , the detection of IFNα and coxsackievirus mRNA in the blood of diabetic children is correlated [23] . In mice , both RRV-induced diabetes acceleration and type I IFN responses are associated with the production of a Th1-biased antibody response [12] , [24] . The aim of this study was to determine if exposure of immune cells from NOD mice to rotavirus induces polyclonal bystander activation of lymphocytes , and whether this occurs through virus activation of pDC and type I IFN production . We found that rotavirus-stimulated splenocytes exhibited dose-dependent APC and B cell activation that was independent of virus replication or strain . This activation was associated with IFNα secretion and prevented by rotavirus treatment with VP7 neutralising antibody , inhibition of endosomal acidification or TLR7 signaling and blockade of signaling through the IFNAR . Importantly , pDC ( and to a lesser extent , conventional DC ( cDC ) ) were shown to contribute to B and T lymphocyte activation following RRV exposure . It was further demonstrated that rotavirus induces the activation of islet autoreactive T cells . These data provide evidence that bystander activation may be an important mechanism for lymphocyte activation during RRV-mediated diabetes acceleration in NOD mice . The activation status of NOD mouse splenocytes cultured in the presence of rotavirus was analysed . Splenocytes were stimulated with RRV , I-RRV and CRW-8 , of which only RRV accelerates diabetes onset in diabetes-prone mice [11] , [12] . As expected , the proportion of activated ( CD69+ ) APCs and B cells increased following control stimulation with bacterial lipopolysaccharide ( LPS ) for 12 h or 24 h ( Figure 1A , 1B ) . The proportion of APCs ( Figure 1A ) and B cells ( Figure 1B ) expressing CD69 was significantly increased over unstimulated controls after stimulation with RRV , I-RRV or CRW-8 for 12 h ( p≤0 . 021 and p≤0 . 029 , respectively ) and 24 h ( p≤0 . 023 and p≤0 . 011 , respectively ) . No APC or B cell activation was detected at 1 h after rotavirus exposure ( p>0 . 05 ) . The proportion of activated APCs and B cells increased with the duration of rotavirus exposure . T cell ( CD3+ ) activation was not observed at any time ( data not shown ) . B cell MHC I expression also increased after RRV , I-RRV or CRW-8 stimulation for 24 h compared to unstimulated B cells ( Fig . 1C , p≤0 . 022 ) . MHC I levels were unaltered at 1 h and 12 h after stimulation ( Figure 1C , p>0 . 05 ) . Following 24 h of stimulation with RRV , I-RRV or CRW-8 , B cell expression of CD86 ( Figure S1A; p≤0 . 010 ) and MHC II ( Figure S1B; p≤0 . 028 ) also was upregulated . On APCs , CD86 expression was increased ( Figure S1A; p≤0 . 0017 ) but MHC I and MHC II levels were unaltered ( data not shown ) . CD80 expression on APCs and B cells was unaltered by rotavirus exposure ( data not shown ) . Therefore , exposure of NOD mouse splenocytes to rotavirus induced the activation of APCs and B cells , but not T cells . Activation of these cells was neither virus strain-specific nor replication-dependent . To determine whether RRV exposure also activated CD11c+ DC , the proportion of CD11c+ and CD11c− APCs ( CD3−CD19−MHCII+ ) activated after 24 h of virus exposure was assessed . Control LPS stimulation induced CD11c+ DC and CD11c− APC activation ( Figure 1D; p = 0 . 0001 and p = 0 . 0012 , respectively ) . Stimulation with RRV , I-RRV or CRW-8 significantly increased CD69 expression on CD11c+ DC and CD11c− APCs over unstimulated controls ( Figure 1D; p≤0 . 0036 and p≤0 . 0034 , respectively ) . Therefore , rotavirus induced the activation of DC and other APC subtypes . The concentration dependence of APC and B cell activation by rotavirus was analysed using NOD mouse splenocytes stimulated with serial dilutions of RRV or I-RRV for 24 h . Stimulation with 1 or 10 ng/ml of rotavirus did not significantly activate APCs ( Figure 2A , 2B; p>0 . 05 ) . APC activation was increased at 100 ng/ml of RRV or I-RRV ( p = 0 . 0047 and p = 0 . 0021 , respectively ) and 1000 ng/ml of RRV or I-RRV ( p = 0 . 024 and p = 0 . 0031 , respectively ) . In contrast , 10 ng/ml of RRV or I-RRV was sufficient to induce B cell activation ( Figure 2B , p≤0 . 048 ) . Overall , APC and B cell activation was dose-dependent , with 100 ng/ml being the optimum dose ( as used for the studies described above and in Figure 1 ) . The extent of APC and B cell activation by RRV over a large dose range of NOD mouse splenocytes was compared to RRV activation of splenocytes from C57BL/6 mice to determine if NOD mouse cells show greater sensitivity to rotavirus stimulation than those of mice not prone to autoimmunity . RRV stimulation for 24 h induced substantially more APC ( Figure S2A , p<0 . 0001 ) and B cell ( Figure S2B , p<0 . 0001 ) activation in NOD mouse cells over C57BL/6 mouse cells . This indicates that NOD mouse cells are more sensitive to ex vivo RRV stimulation than those of C57BL/6 mice . To determine if APC and B cell activation could be induced by short-term exposure to rotavirus , splenocytes cultured with LPS , RRV , I-RRV or CRW-8 for 1 h were washed , resuspended in fresh medium and cultured for a further 23 h . Rotavirus concentrations in supernatant fluids collected at 23 h after virus removal were <0 . 2 ng/ml ( <1×102 FCFU/well ) . LPS stimulation for 1 h increased the proportion of activated B cells but not APCs ( Figure 2C; p = 0 . 0071 and p>0 . 05 , respectively ) . Activation by LPS was substantially less than in the earlier experiments due to the requirement for constant ligation of surface Toll-like receptor ( TLR ) 4 by LPS to induce optimal cellular activation . Exposure to RRV , I-RRV or CRW-8 for 1 h was sufficient to raise CD69 expression on APCs and B cells ( Figure 2C; p≤0 . 011and p≤0 . 0030 , respectively ) after 24 h of culture . Thus , 1 h of rotavirus exposure was sufficient for APC and B cell activation . As 1 h would be sufficient for rotavirus-cell adsorption , these data suggested that rotavirus might have stimulated APCs and B cells through an intracellular mechanism . Assays of intracellular Ki67 expression and 3H-Thymidine incorporation were employed to detect B cell proliferation in LPS- and rotavirus-stimulated splenocytes . Compared with unstimulated B cells , LPS stimulation for 48 h or 72 h increased the proportion of activated B cells undergoing proliferation ( Figure 2D; p = 0 . 0005 and p<0 . 0001 , respectively ) . In contrast , RRV , I-RRV or CRW-8 did not induce B cell proliferation ( p>0 . 05 ) . Similar results were observed at 44 h and 68 h when 3H-Thymidine incorporation was measured . LPS stimulation produced 3H-Thymidine uptake ( mean±SEM ) of 2934±445 cpm . In the absence and presence of rotavirus ( RRV , I-RRV or CRW-8 ) , 687±73 cpm and 455±40 cpm , respectively , of 3H-Thymidine was incorporated ( p>0 . 05 ) . From these findings , B cell development into antibody-secreting cells following this rotavirus stimulation ex vivo seemed unlikely , as typically B cell development is proliferation-dependent [25] . Previous studies identified rotavirus outer capsid protein VP7 and rotavirus RNA as potential contributors to B cell and pDC activation [15] , [18] . To investigate the importance of these rotavirus factors for NOD mouse APC and B cell activation , virions were treated an anti-VP7 antibody ( RV-3:1 ) that neutralises rotavirus infectivity by preventing virion decapsidation and release of the dsRNA genome [26] . An isotype-matched antibody ( RV-5:2 ) that binds human rotavirus RV-5 but not RRV or CRW-8 was reacted with virions as a control . As before , LPS activated APCs and B cells ( Figure 3A , 3B; p = 0 . 0023 and p<0 . 0001 , respectively ) , and RRV , I-RRV and CRW-8 induced CD69 expression on APCs ( Fig . 3A; p = 0 . 0014 , p = 0 . 0007 and p = 0 . 0086 , respectively ) and B cells ( Fig . 3B; p<0 . 0001 , p = 0 . 0013 and p<0 . 0001 , respectively ) . Stimulation with control antibody-treated rotavirus induced similar levels of APC and B cell activation to untreated rotavirus ( p>0 . 05 ) . APC ( Figure 3A ) and B cell ( Figure 3B ) activation was significantly reduced following stimulation with anti-VP7-treated rotaviruses over control antibody-treated rotaviruses ( p≤0 . 041 and p≤0 . 0040 , respectively ) . The proportion of APCs and B cells expressing CD69 following stimulation with anti-VP7-treated rotavirus was similar to that in the unstimulated control ( Figure 3A ) . Overall , antibody blockade of rotavirus VP7 prevented almost all APC and B cell activation by rotavirus . This implies that VP7 itself and/or exposure to viral RNA following virion decapsidation was required for activation of APCs and B cells . Blockade of endosomal acidification prevents IFNα expression by pDC following RRV exposure [18] . To ascertain if endosomal acidification was necessary for NOD mouse APC and B cell activation by rotavirus , splenocytes were treated with chloroquine prior to stimulation with virus , LPS or polyinosinic:polycytidylic acid ( poly IC ) . The proportion of activated APC and B cells in RRV-stimulated cell cultures , adjusted for the proportion activated in the absence of RRV , is shown in Figure 4 . As LPS does not signal through endosomal TLRs , only a small reduction in APC and B cell activation was observed following LPS stimulation of chloroquine-treated splenocytes ( Figure 4A , 4B; p = 0 . 0021 and p = 0 . 0059 , respectively ) . In contrast , APC and B cell activation by poly IC , which signals through TLR3 , was reduced when endosomal acidification was blocked ( Figure 4A , 4B; p≤0 . 0001 ) . Incomplete blockade of endosomal acidification or cytoplasmic receptor activation is likely to explain the residual low-level activation detected compared with unstimulated cells , which was particularly evident in B cells . Similarly , stimulation with RRV , I-RRV and CRW-8 induced CD69 expression on fewer APCs ( Figure 4A; p = 0 . 0005 , p = 0 . 0037 and p<0 . 0001 , respectively ) and B cells ( Figure 4B; p = 0 . 019 , p = 0 . 0028 and p = 0 . 0002 , respectively ) following chloroquine treatment compared to PBS-treated cells . Rotavirus and poly IC activated equivalent proportions of chloroquine-treated APCs and B cells ( Figure 4A , 4B , p>0 . 05 ) . Thus , blockade of endosomal acidification prevented efficient APC and B cell activation by rotavirus , indicating the likely importance of signaling through an endosomal TLR , such as TLR7 , TLR9 or TLR3 . As RRV-stimulated pDC contribute to B cell activation in a non-autoimmune mouse model [17] and rotavirus is a dsRNA virus , TLR7 was considered to be the most likely candidate . To determine if signaling through TLR7 was required for APC and B cell activation by rotavirus , splenocytes were pretreated with the TLR7 antagonist , IRS661 , or the control oligonucleotide , Ctrl ODN . Neither the APC nor the B cell activation following stimulation with LPS was affected by TLR7 blockade ( Figure 4C , 4D , p>0 . 05 ) . Conversely , treatment with IRS661 reduced APC ( Figure 4C ) and B cell ( Figure 4D ) activation following stimulation with TLR7 agonist , Imiquimod , compared to Ctrl ODN-treated cells ( p = 0 . 0032 and p<0 . 0001 , respectively ) . Similarly , fewer APC ( Figure 4C ) and B cells ( Figure 4D ) were activated by RRV ( p = 0 . 0002 and p = 0 . 01 , respectively ) , I-RRV ( p = 0 . 016 and p<0 . 0001 , respectively ) or CRW-8 ( p = 0 . 0003 and p = 0 . 0001 , respectively ) following IRS661 treatment compared to Ctrl-ODN treatment . The extent of rotavirus activation of IRS661-treated cells was equivalent to or less than the activation of IRS661-treated cells stimulated with Imiquimod , showing that TLR7 inhibition completely prevented APC and B cell activation by rotavirus . Overall , these data show that TLR7 recognition of rotavirus RNA is required for activation of NOD mouse APC and B cells by rotavirus . To confirm the importance of TLR7 signalling for APC and B cell activation by rotavirus , splenocytes from TLR7 ( -/- ) mice ( on a C57BL/6 genetic background ) and C57BL/6 mice were stimulated for 24 h with Imiquimod , Poly IC , rotavirus or left unstimulated . As expected , APC and B cell activation in TLR7 ( -/- ) splenocytes following stimulation with Imquimod was reduced compared to C57BL/6 cells ( Figure 4E , 4F; p = 0 . 0043 and p = 0 . 0013 , respectively ) . However , no reduction in APC and B cell activation following stimulation with poly IC was observed . The proportion of APC activated by RRV , I-RRV and CRW-8 was significantly reduced in splenocytes from TLR7 ( -/- ) mice compared with C57BL/6 cells ( Figure 4E; p = 0 . 0013 , p = 0 . 010 and p = 0 . 019 , respectively ) . B cell activation was similarly reduced ( Figure 4F; p = 0 . 0004 , p = 0 . 0068 and p = 0 . 0126 , respectively ) . Interestingly , in the TLR7 ( -/- ) cells , the proportion of activated APC ( p = 0 . 0036 , p = 0 . 0009 and p = 0 . 01 , respectively ) and B cells ( p = 0 . 0071 , p>0 . 05 and p<0 . 0001 , respectively ) following RRV , I-RRV and CRW-8 exposure remained higher than the proportion following Imiquimod treatment . Little or no reduction in APC and B cell activation was observed in TLR3 ( -/- ) cells compared with C57BL/6 cells , suggesting that much of this activation does not occur through TLR3 ( Figure S3 ) . These findings indicate that TLR7 signaling is important for APC and B cell activation in non-autoimmune cells . However , in contrast to NOD mouse cells , this activation may not completely depend on TLR7 signaling . Exposure of human immune cells , particularly DC , to rotavirus induces the expression of multiple cytokines , including IFNα , which is capable of bystander activation [16] , [17] , [18] , [27] . To assess the requirement for soluble factors in APC and B cell activation , supernatant fluids were collected from NOD mouse splenocyte cultures that had been treated with rotavirus for 1 h and incubated for a further 23 h . These fluids , which contained <0 . 2 ng/ml of rotavirus protein , were added to cultures of naive NOD mouse splenocytes for 24 h . This supernatant fluid exposure induced CD69 expression on APCs and B cells ( Figure 5A; p≤0 . 0008 and p≤0 . 001 , respectively ) . Thus , rotavirus exposure induced splenocyte production of soluble factor/s capable of activating APC and B cells . As type I IFN has previously been shown to be important for B cell activation by rotavirus [17] the importance of IFNAR signaling for rotavirus activation of NOD-derived APCs and B cells was determined by anti-IFNAR antibody blockade . Treatment with negative control MOPC21 antibody did not significantly affect APC activation after stimulation with I-RRV or CRW-8 ( Figure 5B , p>0 . 05 ) . The proportion of activated APCs following MOPC21 treatment and RRV stimulation was somewhat reduced ( p = 0 . 0091 , respectively ) . However , the proportion of activated cells remained substantially greater after RRV stimulation than in unstimulated controls ( p = 0 . 0072 ) . B cell activation by rotavirus was not affected by MOPC21 treatment ( Figure 5C ) . Blockade with IFNAR antibody prior to RRV , I-RRV or CRW-8 stimulation significantly reduced the proportion of APCs and B cells expressing CD69 compared to MOPC21-treated cells ( Figure 5B , 5C; p≤0 . 0047and p≤0 . 0008 , respectively ) . The proportions of activated APCs and B cells following stimulation of anti-IFNAR treated cells with I-RRV and CRW-8 were equivalent to unstimulated cells ( Figure 5B , 5C; p>0 . 05 ) . However , a small increase in B cell activation was observed following stimulation of anti-IFNAR treated cells with RRV ( p = 0 . 0010 ) . Therefore , signaling through the IFNAR was important for APC and B cell activation following exposure to rotavirus . Splenocytes from IFNAR ( -/- ) mice ( on a C57BL/6 genetic background ) and C57BL/6 mice were stimulated with 100 ng/ml of RRV , I-RRV or CRW-8 for 24 h to analyse the dependence of cellular activation on signaling through IFNAR . PMA/Ionomycin C treatment as a positive control induced CD69 expression on a mean±SEM of 70±0 . 5% of APCs and 67±0 . 5% of B cells from C57BL/6 mice and 62±0 . 5% of APC and 41±4 . 7% of B cells from IFNAR ( -/- ) mice . Stimulation of C57BL/6 splenocytes with RRV , I-RRV or CRW-8 increased the activation of APCs and B cells relative to unstimulated cells ( Fig . 5D; p≤0 . 0160 and p≤0 . 0061 , respectively ) . As expected , the degree of activation was less than that observed in NOD cells . IFNAR ( -/- ) mice showed variable APC activation following stimulation with rotavirus ( Figure 5D ) . CRW-8 stimulation did not activate APCs ( p>0 . 05 ) , whereas a trend for increased APC activation was seen after RRV stimulation ( p = 0 . 0575 ) and I-RRV increased APC activation ( p = 0 . 0090 ) . The relative proportion of activated APCs following RRV stimulation was significantly reduced in IFNAR ( -/- ) compared to C57BL/6-derived cells ( p = 0 . 0017 ) . Stimulation with I-RRV or CRW-8 did not alter the activated APC proportion ( p>0 . 05 ) . This suggests that APC activation may not be completely dependent on IFNAR signaling . However , as the overall degree of APC activation in C57BL/6 splenocytes was not as robust as in NOD splenocytes , this interpretation is somewhat tentative . In contrast to APC , no relative increase in CD69 expression on B cells from IFNAR ( -/- ) mice following rotavirus exposure was observed compared to unstimulated controls ( Figure 5D; p>0 . 05 ) . Furthermore , the relative proportion of activated B cells following stimulation with RRV , I-RRV or CRW-8 was significantly reduced in IFNAR ( -/- ) compared to C57BL/6-derived splenocytes ( Figure 5D; p≤0 . 0084 ) . Overall , signaling through the IFNAR following rotavirus stimulation was strongly implicated in APC activation and essential for B cell activation . To confirm that type I IFN was produced following rotavirus stimulation , IFNα levels in supernatant fluids from NOD-derived splenocytes after rotavirus exposure were determined . As expected , IFNα was not detected in unstimulated supernatant fluids ( Figure 5E ) . In contrast , stimulation with RRV , I-RRV or CRW-8 induced IFNα secretion , at mean levels of 449 pg/ml , 451 pg/ml and 336 pg/ml , respectively . Virus strains and preparations did not differ the level of IFNα secretion induced ( p>0 . 05 ) . Thus , rotavirus induced type I IFN secretion in unsorted NOD splenocyte cultures . RRV and LPS stimulation of T cells ( CD3+CD19− ) , B cells ( CD3−CD19+ ) and non-T and non-B cells ( double negative ( DN ) ; CD3−CD19− ) sorted from NOD splenocytes were assessed . The purity of these sorted DN , B cell and T cells is demonstrated in Figure S4A . LPS induced APC activation in unsorted but not DN cells ( Figure 6A; p = 0 . 0001 and p>0 . 05 , respectively ) . However , APCs in both these populations were strongly activated by RRV exposure ( p = 0 . 0037 and p = 0 . 0001 , respectively ) . Thus , RRV activated APCs in the absence of B and T cells . Additionally , IFNα was detected in supernatant fluids from DN cells stimulated with RRV , but not unstimulated DN cells ( Figure S4B ) . LPS induced B cell activation in unsorted cells and sorted B cells ( Figure 6B; p = 0 . 0012 and p<0 . 0001 , respectively ) . Although the proportion of sorted B cells activated by RRV was increased over unstimulated controls ( Figure 6B; p = 0 . 0002 ) , this proportion was significantly less than that in unsorted splenocytes ( p = 0 . 0003 ) . This indicated that RRV was a poor activator of B cells in the absence of DN and T cells . As before , RRV stimulation of unsorted splenocytes did not induce T cell activation ( Figure 6C , 6D; p>0 . 05 ) . However , CD4+ and CD8+ T cells in unsorted splenocytes were activated by PMA/Ionomycin C ( p = 0 . 0024 and p = 0 . 0013 , respectively ) . Similarly , sorted T cells were activated by PMA/Ionomycin C ( p<0 . 0001 ) but not RRV ( p>0 . 05 ) . Thus , RRV did not directly activate T cells . To determine their contribution to lymphocyte activation , sorted DN cells were cultured with sorted B or T cells in the presence of RRV . The addition of DN cells in the absence of RRV induced a small increase in T cell activation , but no change in B cell activation ( Figure S5A ) . As shown in Figure 7A , inclusion of DN cells in the presence of RRV significantly increased the proportion of activated B cells compared to sorted B cells alone ( p<0 . 0001 ) , and led to the activation of CD4+ and CD8+ T cells ( p = 0 . 0011 and p<0 . 0001 , respectively ) . Co-culture of sorted B and T cells activated a mean±SEM of 6 . 0±1 . 6% of B cells , equivalent to the degree of B cell activation induced by RRV stimulation of sorted B cells alone ( p>0 . 05 ) . Thus , rotavirus-exposed DN cells contributed to B and T cell activation . In a single experiment ( data not shown ) , the activation of anti-IFNAR treated T cells and B cells , cultured with DN cells in the presence of RRV , was assessed . IFNAR blockade induced a similar extent of CD4+ T cell ( 0 . 05±0 . 03% ) , CD8+ T cell ( 0 . 35±0 . 1% ) and B cell ( 4 . 6±0 . 7% ) activation to sorted cells alone ( p>0 . 05 ) . Additionally , IFNα secretion was detected when T cells were cultured with DN cells and RRV , but not in cultures of T cells alone in the presence or absence of RRV ( Figure S5B ) . This further confirms that type I IFN expression by DN cells contributes to T and B cell activation . The ability of RRV-exposed DN cells to contribute to the activation of islet autoantigen-specific CD8+ T cells was assessed using T cells isolated from the splenocytes of T cell receptor-transgenic NOD8 . 3 mice . Most CD8+ T cells from these mice recognise an autoimmunity-related epitope ( amino acids 206-214 ) from islet-specific glucose-6-phosphatase catalytic subunit related protein ( IGRP ) . The addition of DN cells induced a small increase in T cell activation in the absence of RRV , as observed previously ( Figure S5C ) . Culture of DN cells with T cells from these mice significantly increased the proportion of activated of CD4+ T cells and CD8+ T cells compared to sorted T cells alone in the presence of RRV ( Figure 7B; p = 0 . 0093 and p = 0 . 0052 , respectively ) . Importantly , the presence of RRV also led to the activation of IGRP-specific CD8+ T cells , which were detected by IGRP-specific tetramer staining ( p = 0 . 0052 ) . This confirms that this form of bystander activation induced by RRV results in the activation of islet-autoreactive T cells . To determine if CD11c+ DC were required for B cell activation , the effect of splenocyte depletion of CD11c+ DC , and B cell co-culture with CD11c+ DC , on RRV-stimulated B cell activation was determined . The purity of the sorted CD11c+ DC is demonstrated in Figure S6A . As expected , the DC comprised two populations expressing variable levels of CD11c , which indicates the presence of cDC and pDC . Neither the depletion nor addition of CD11c+ DC increased the proportion of activated B cells in the absence of RRV ( Figure S6B ) . However , increased background B cell activation was observed in sorted cells ( Figure S6B ) . As expected , in the presence of RRV purified B cells alone showed less activation than B cells in unsorted splenocytes ( Figure 7C; p = 0 . 0064 ) . The proportion of activated B cells in splenocytes depleted of CD11c+ DC was similar to that in sorted B cells alone , and less than in unsorted splenocytes ( p>0 . 05 and p = 0 . 035 , respectively ) . Furthermore , CD11c+ DC addition to purified B cells increased B cell activation over sorted B cells alone ( Figure 7C; p = 0 . 016 ) . This confirmed that CD11c+ DC were necessary and sufficient for B cell activation following RRV exposure . As signaling through the IFNAR and the presence of DC were shown above to be important for lymphocyte activation by rotavirus , the specific role of pDC was investigated . Sorted T and B cells were cultured alone , with cDC or with pDC , in the presence or absence of RRV for 24 h . The purity of sorted cDC and pDC , distinguished by their levels of MHC II expression , is demonstrated in Figure S7A . In all combinations , B cells were activated following stimulation with LPS and T cells were activated following stimulation with PMA/Ionomycin C , as previously shown ( Figure 6 ) . Culture of B cells with pDC but not cDC induced a small degree of activation in the absence of RRV ( Figure S7B ) . In the absence of RRV , CD4+ and CD8+ T cells showed a small increase in activation following culture with cDC ( Figure S7B ) . B cells exposed to RRV were activated in the presence of both cDC and pDC ( Figure 8 , p = 0 . 0001 ) . However , the equivalent number of pDC induced CD69 expression in a significantly larger proportion of B cells than did cDC ( p = 0 . 0011 ) . Thus , while both cDC and pDC contributed to B cell activation during RRV stimulation , pDC showed an enhanced ability to activate B cells . In cultures of T cells and cDC , RRV stimulation did not activate CD4+ or CD8+ T cells ( Figure 8; p>0 . 05 ) . In contrast , an increased proportion of CD4+ and CD8+ T cells exposed to RRV were activated in the presence of pDC ( Figure 8; p<0 . 0001 ) . Therefore , pDC , but not cDC , contributed to T cell activation by RRV . Using combinations of sorted cell populations , analysis of pDC following RRV stimulation in the presence of either T or B cells showed that RRV induced activation of a mean±SEM of 15 . 2±2 . 5% pDC . In contrast , RRV stimulation of T or B cell cultures in the presence of cDC induced activation of a mean±SEM of only 2 . 6±1 . 0% of cDC . In a single experiment where cDC or pDC were cultured with T cells , RRV stimulation induced IFNα secretion in T cell cultures containing pDC but not cDC ( Figure S7C ) . Thus , RRV activated pDC more efficiently that cDC . Overall , RRV-activated pDC contributed to the activation of B and T cells through the action of type I IFN . Bystander activation is a candidate mechanism for the acceleration of type 1 diabetes development by rotavirus . Here , rotavirus stimulation of splenocytes from diabetes-prone NOD mice was shown to induce APC and B cell activation , which was prevented by VP7 blockade , inhibition of endosomal acidification and interference with TLR7 or IFNAR signaling . RRV rotavirus stimulation directly activated APCs in the absence of lymphocytes , but induced little if any activation of B and T cells in the absence of APCs . Efficient B cell activation was shown to require the presence of CD11c+ DC . Importantly , pDC activated both T and B cells following RRV stimulation . Exposure to RRV also induced the activation of islet-autoreactive T cells and secretion of IFNα . These findings provide strong evidence that this lymphocyte activation occurs through type I IFN expression by RRV-activated DC , mediated by recognition of rotavirus RNA . As described previously , we found that APC and B cell activation by rotavirus was independent of virus strain and replication [15] , [16] . This suggests that other rotaviruses , including human and murine strains , have the potential to similarly induce activation of NOD mouse lymphocytes . Rotavirus is known to induce IFNα expression by pDC , leading to B cell activation [15] , [16] , [17] , [18] . Similarly , depletion of pDC in human PBMC cultures reduces IFNγ expression by T cells [19] . However , to our knowledge this is the first time that type I IFN expression by pDC following rotavirus exposure has been shown to contribute to murine T cell activation , and cDC activation by rotavirus has been linked to B cell activation . In addition , TLR7 signaling is conclusively identified here for the first time as an important pathway for immune activation by rotavirus on both an autoimmune ( NOD ) and non-autoimmune ( C57BL/6 ) genetic background . In C57BL/6 mice , TLR7 signaling appears to be more important for APC and B cell activation than TLR3 signaling . Our data indicate that the process of RRV-stimulated type I IFN expression by pDC leading to B cell activation is similar between NOD mice , non-diabetes prone mice and human PBMC . However , NOD mouse cells are more sensitive to activation following RRV exposure than C57BL/6 mouse cells . Secretion of IFNα by human pDC requires rotavirus outer capsid proteins and dsRNA , which led to the proposal that virus entry ( not phagocytosis ) and signaling through TLR7 or TLR9 are required for this process [18] . Here we showed that VP7 decapsidation blockade and inhibition of TLR7 signaling each ablate the ability of rotavirus to induce activation of APC and B cells from NOD mice . Murine DC activation by in vitro-generated viral dsRNA fragments is TLR7-independent [28] . However , short interfering RNA has previously been shown to trigger TLR7 responses in pDC through the recognition of specific single-stranded RNA motifs [29] . By analogy with the better-studied process of rotavirus entry into epithelial cells [30] , rotavirus entry into APC may involve outer capsid protein permeabilisation of the early endosomal membrane , mediated by loss of stabilising calcium ions from VP7 trimers in the low pH environment . Antibody cross-linking of VP7 inhibits this VP7 disassembly , preventing exposure of the viral dsRNA genome and genome transcription [26] , [31] . Inactivated RRV retains the ability to bind VP7 antibodies and enter cells [32] , [33] . As inactivated RRV also induces cellular activation , it is likely that TLR7 is triggered following dsRNA exposure in the endosome and not by replication intermediates . To our knowledge , this is the first study to demonstrate the induction of TLR7-mediated signaling by a dsRNA virus . Previously , another dsRNA virus , bluetongue , was shown to induce type I IFN expression by pDC independently of TLR7 [34] . The rotavirus activation of APC and B cells from NOD mice seems completely dependent on TLR7 signaling , whereas partial blockade of activation occurs in TLR7 ( -/- ) cells from C57BL/6 mice . Thus , TLR7 signaling plays a more important role in NOD than C57BL/6 cells , which may help explain the increased sensitivity of NOD cells over C57BL/6 cells to RRV stimulation . In our study , VP7 antibody blocked rotavirus infectivity through inhibition of decapsidation , so whether rotavirus binds and infects these DC or is taken up by phagocytosis cannot be determined . However , as rotavirus entry likely induces human pDC activation [18] , we propose that rotavirus signaling through TLR7 in NOD mouse DC also occurs following virus entry . Studies with human pDC suggest that RRV infects a minor proportion of these cells . However , productive infection prevents IFNα production [18] . In our NOD mouse studies here , intracellular rotavirus antigen was below the limit of detection by flow cytometry ( J . A . Pane , N . L . Webster and B . S . Coulson , unpublished data ) . This implies that productive rotavirus infection of murine splenocytes is extremely rare , and shows that very little cell-associated rotavirus is required to trigger these responses . It is likely that a small DC population is directly activated by rotavirus exposure through TLR7 signaling , while a larger population is activated by exposure to secreted type I IFN ( Figure 9 ) . This is supported by the incomplete dependence of APC activation on type I IFN signaling . Importantly , these data provide evidence that the small population of rotavirus-positive APCs detected in MLN and PLN following RRV infection of NOD mice probably would be sufficient to induce the lymphocyte activation observed in RRV-infected NOD mice [12] . RRV stimulation of unsorted NOD mouse splenocytes induced the activation of both CD11c+ and CD11c− APCs , as expected from the RRV propensity to associate with and activate multiple small populations of NOD mouse APC subsets in vivo [12] . However , following ex vivo RRV stimulation , a trend towards increased activation of pDC over cDC activation was observed . Following RRV infection of adult NOD mice , TNF expression is increased on CD11c+CD8α+ DC , CD11c+ DC and CD11c+CD11b+ DC in the PLN [14] . In addition to CD11c , pDC can express CD8α and upregulate CD8α expression upon activation [35] . This suggests that the activated CD11c+ DC population seen in vivo after RRV infection of adult NOD mice may consist of pDC . As rotavirus exposure of DC subsets , including pDC , induces the expression of multiple pro-inflammatory cytokines [18] , [19] , [27] , it is likely that the DC subsets secreting TNF in the PLN following RRV infection of NOD mice also express IFNα . While rotavirus exposure induced minimal cDC activation ex vivo , this was sufficient to induce B cell but not T cell activation , implying that B cells from NOD mice are highly sensitive to type I IFN-induced activation . In addition , larger proportions of B cells than T cells were activated by rotavirus , and unlike T cells , B cells were activated in unsorted splenocytes . This is further supported by previous rotavirus studies where approximately 1 pDC per 200 B cells was sufficient to induce human B cell activation [17] . Therefore , it is likely that the CD11c+ DC subsets other than pDC that are activated following RRV infection of NOD mice also contribute to B cell activation . As well as inducing B and T cell activation , type I IFN also directly increases the ability of DC and B cells to present antigen to T cells [36] . Here , we showed that B cells also upregulate MHC I and CD86 expression , and APC upregulate CD86 expression , following rotavirus stimulation . This supports our findings in RRV-infected NOD mice , where PLN and MLN B cells were increased in MHC I levels and ability to induce autoreactive T cell proliferation [14] . Presentation of autoantigen by B cells on MHC I is required for diabetes development in NOD mice [37] . Additionally , B cell antigen presentation is important for T cell activation and duct injury in RRV-induced biliary atresia , which has been proposed to involve autoreactive T cell-mediated inflammation [38] . Therefore , type I IFN-mediated B cell activation may be responsible for the increased autoantigen presentation by B cells following RRV infection and contribute to diabetes acceleration . From the studies presented here , we hypothesise that upregulated type I IFN expression by DC in the lymph nodes following RRV infection of NOD mice contributes to the direct activation of autoreactive B and T cells , increased antigen presentation by B cells and induction of β cell death by cytokine-secreting autoreactive T cells ( Figure 9 ) . Therefore , type I IFN expression within the lymph nodes of RRV-infected NOD mice should be analysed in future experiments . In addition , blocking this signaling pathway in vivo may prevent the activation of B and T cells in NOD mice following RRV infection , as has been shown for B cell activation in IFNAR knockout mice [17] . We demonstrate a lack of B cell proliferation following rotavirus stimulation of NOD mouse splenocytes , so rotavirus exposure may not induce differentiation of these cells into antibody-secreting cells . This contrasts with human PBMC and sorted human B cells cultured with pDC , where rotavirus exposure induces B cell differentiation [16] , [17] . Type I IFN-induced activation without proliferation previously has been demonstrated for murine B cells [39] . In this scenario , B cell exposure to type I IFN enhances its ability to respond to B cell receptor ligation . Human B cell proliferation following exposure to TLR7 agonists can be induced by co-culture with purified pDC , suggesting that B cell differentiation requires a greater ratio of pDC to B cells than does B cell activation [40] . Therefore , it remains possible that co-culture of purified NOD-derived pDC and B cells could lead to proliferation . However , as pDC comprise <0 . 5% of cells in PLN or spleen [20] , it is unlikely that B cell differentiation occurs following RRV infection of NOD mice . In support of this , we have been unable to detect altered B cell proliferation in the MLN or PLN at day 7 or 14 after infection of NOD mice with RRV ( J . A . Pane , N . L . Webster and B . S . Coulson , unpublished data ) . Although IFNγ production is primarily associated with Th1-biased responses and contributes to IgG2a production , type I IFN also contributes to isotype switching [24] . In the absence of IFNAR signaling , influenza A virus infection induces higher local IgG1 responses , and reduced IgG2a responses [41] . Similarly , pDC depletion reduces IgA and IgG and increases IgM responses following murine rotavirus infection [17] . Therefore , the development of IgG2a-biased antibody responses in NOD mice given RRV [12] may be mediated in part by the expression of type I IFN . In addition to increasing the ability of B cells to induce T cell activation , our data suggest that type I IFN-secreting pDC also directly induce T cell activation , including autoreactive T cells ( Figure 9 ) . Although it is likely that bystander T cell activation also occurs in non diabetes-prone mice , our data suggest that this may be at a reduced level and would be unlikely to produce autoimmune sequela . Our demonstration of autoreactive CD8+ T cell activation by RRV stimulation suggests that autoreactive T cells in the PLN and MLN of NOD mice would participate in this bystander activation , providing a possible explanation for the accelerated diabetes onset following RRV infection . The absence of a distinct autoreactive T cell population in RRV-infected infant NOD mice would prevent this process from occurring , which may help to clarify why they do not develop increased autoimmunity [13] . Although CRW-8 does not modulate diabetes development in NOD mice , this rotavirus induced lymphocyte activation ex vivo , showing that these properties are not necessarily associated . The lesser replicative ability of CRW-8 , and the absence of its infectious form from the MLN and PLN , provides a possible explanation for its inability to modulate diabetes or induce B cell activation following infection of NOD mice [12] , [14] . In contrast , homologous murine rotavirus is expected to replicate efficiently and spread extraintestinally in NOD mice , as occurs in other mouse strains [42] . Murine rotavirus accelerates NOD mouse diabetes development [11] , and induces B cell activation ex vivo [15] and type I IFN-dependent B cell activation in vivo [17] in non-autoimmune mice . It is likely that murine rotavirus infection of NOD mice would activate lymphocytes similarly to RRV . Type I IFN signaling also contributes to B cell activation after influenza A virus infection [41] . In this case , local lung B cell responses are induced by direct signaling through the IFNAR , probably by IFNβ expression . These responses are localised to the lung and draining lymph nodes . It is conceivable that other viruses able to induce type I IFN-mediated bystander activation could augment the NOD mouse autoimmune response , providing that the bystander activation takes place in the PLN ( and possibly the MLN ) where autoreactive cells accumulate . In support of this hypothesis , the presentation of islet antigens in the intestine , MLN and PLN is linked [43] , [44] . Gastrointestinal pathogens may have an enhanced ability to spread to the MLN and PLN from the intestine and induce type I IFN-mediated bystander activation at these sites . Antibody blockade of the IFNAR prior to 3 weeks of age , or depletion of pDC , delays and significantly reduces the incidence of diabetes in NOD mice [20] , [21] . No consistent change in pDC numbers is observed in patients with diabetes , although they show increased IFNα mRNA expression in the pancreas [45] , [46] , [47] . Administration of IFNα can delay diabetes development in NOD mice [48] , [49] . However , IFNα expression during type 1 diabetes development is more likely to be tissue-specific and involve concomitant expression of other cytokines . Therefore , while IFNα administration may mediate diabetes protection , local type I IFN expression probably contributes to diabetes development . Our data suggest that type I IFN expression may be induced following RRV infection of NOD mice , leading to diabetes acceleration . Thus , the importance of type I IFN signaling in diabetes acceleration by RRV requires further analysis . IFNα expression following rotavirus infection of children at-risk of diabetes has not been studied . However , IFNα production appears to correlate with increased severity of initial gastrointestinal symptoms in rotavirus-infected children [50] . IFNα expression following rotavirus infection of children should be investigated in relation to their predisposition to type 1 diabetes . Overall , these studies show that RRV induces B and T lymphocyte activation by triggering endosomal TLR7 responses in pDC and the secretion of type I IFN . Although pDC-mediated activation of bystander lymphocytes following rotavirus exposure is conserved between autoimmune and non-autoimmune mouse models , the presence of autoreactive T cells in adult NOD mice is likely to skew the outcome of this response towards increased autoimmunity . It is now important to determine if type I IFN signaling is required for diabetes acceleration by RRV . The possible use of IFNα/β levels and Th1-biased antibody responses [12] as markers of rotavirus infections likely to exacerbate islet autoimmunity in children at-risk of type 1 diabetes also should be evaluated . NOD/Lt ( NOD ) and C57BL/6 mice were obtained from the Animal Resources Centre ( Canning Vale , Western Australia ) . NOD8 . 3 TCR ( NOD8 . 3 ) mice , expressing the TCRαβ rearrangements of the H-2Kd-restricted , islet β cell-reactive CD8+ T cell clone NY8 . 3 on a NOD genetic background [51] were provided by P . Santamaria , University of Calgary , Calgary , Alberta , Canada . Mice were bred and housed in micro-isolator cages under specific pathogen-free conditions in the Biological Research Facility of the Department of Microbiology and Immunology at the University of Melbourne , as before [11] , [13] . Principles of laboratory animal care' ( NIH publication no . 85–23 ) and the ‘Australian Code of Practice for the Care and Use of Animals for Scientific Purposes ( 2004 ) ’ were followed . All procedures were conducted in accordance with protocols approved by the Animal Ethics Committee of The University of Melbourne ( ID 0911434 ) . Rotaviruses RRV and CRW-8 were amplified , purified by glycerol gradient ultracentrifugation and infectious titers in fluorescent cell-forming units ( FCFU ) /ml determined as described previously [52] , [53] . Psoralen/UV inactivation of RRV was performed as before [32] , [54] . The protein concentration of purified rotavirus was determined using the Bio-Rad Protein Assay . For analysis of unsorted cell populations , spleens were passed through 70 μm mesh , treated with red cell lysis buffer and resuspended in RPMI supplemented with 10% ( vol/vol ) fetal calf serum , 2 mM L-glutamine , 50 units penicillin and streptomycin ( 50 μg/ml ) ( RF10 ) . For cell sorting , spleens were digested using 1 mg/ml collagenase A and 40 μg/ml DNase 1 for 30 min at room temperature prior to processing as above . Isolated cells were then reacted with 5 mM EDTA diluted in RPMI for 5 min . For sorting of cDC and pDC , splenocytes were further enriched by density-gradient centrifugation using Nycodenz ( 1 . 077 g/ml; PROGEN Biotechnik GmbH , Germany ) . The number of isolated cells was determined by trypan blue staining . Splenocytes isolated from IFNAR ( -/- ) mice on a C57BL/6 background [55] were provided by Dr Sammy Bedoui , Department of Microbiology and Immunology , The University of Melbourne . Splenocytes from TLR7 ( -/- ) and TLR3 ( -/- ) mice on a C57BL/6 background [56] were provided by Weisan Chen , Department of Biochemistry , School of Molecular Science , La Trobe University , Bundoora , Victoria , Australia . For isolation of T cells ( CD3+CD19− ) , B cells ( CD3−CD19+ ) and DN cells ( CD3−CD19− ) , splenocytes were stained with anti-CD3 ( 145-2C11 ) -phycoerythrin ( PE ) and anti-CD19 ( ID3 ) -fluorescein isothiocyanate ( FITC ) diluted in RF10 for 30 min at 4°C and sorted using a FACS Aria ( BD Biosciences ) . For depletion of CD3−CD11c+ DC from splenocytes and isolation of CD3−CD11c+ DC , splenocytes were stained with anti-CD3-PE and anti-CD11c ( HL3 ) -Allophycocyanin . For isolation of pDC ( CD3−CD19−MHCII+CD11c+CD45RA+ ) and cDC ( CD3−CD19−MHCII++CD11c++CD45RA− ) , splenocytes were stained with antibodies to anti-CD3 ( 145-2C11 ) -PerCPCy5 . 5 , anti-CD19 ( 145-2C11 ) -PerCPCy5 . 5 , anti-MHCII ( OX-6 ) -FITC , anti-CD11c-Allophycocyanin and anti-CD45RA ( 14 . 8 ) -PE . Population purity was >95% . Purified anti-mouse IFNα/β Receptor 1 ( BD Bioscience; IgG1 , anti-IFNAR ) and protein-A purified neutralising anti-VP7 monoclonal antibody RV-3:1 ( IgG2b , [57] ) were matched for protein concentration with purified isotype control antibodies MOPC21 ( ICN Pharmaceuticals ) and RV-5:2 ( neutralising antibody specific for an irrelevant rotavirus strain [58] ) , respectively . To assess the role of VP7 , rotavirus ( 10 ng ) was incubated with 7 . 5 μg of RV-3:1 or RV-5:2 for 2 h at room temperature prior to culture with splenocytes . To block IFNAR signaling , splenocytes ( 1×106 ) were incubated with 1 μg anti-IFNAR antibody or MOPC21 for 30 min at 4°C . Excess antibody was removed by washing prior to rotavirus exposure . For blockade of endosomal acidification , splenocytes ( 5×105 ) were incubated at 37°C for 1 h prior to exposure to rotavirus with 10 μM chloroquine ( Sigma , MO ) diluted in PBS . For blockade of TLR7 signaling , splenocytes were treated for 30 min prior to rotavirus exposure with the synthetic oligonucleotide IRS 661 ( 5′-TGCTTGCAAGCTTGCAAGCA-3′ ) or control oligonucleotide ( 5′-TCCTGCAGGTTAAGT-3′ ) at 6 μM ( Geneworks , Australia ) , as previously described [59] . Unsorted splenocytes ( 5×105 cells/well in U-bottomed 96-well trays ) were cultured in the presence of 100 ng/ml rotavirus ( unless otherwise indicated ) for 1 h , 12 h or 24 h at 37°C and 5% CO2 ( 200 μl total volume ) . For RRV and CRW-8 rotaviruses , 100 ng/ml corresponded to 3 . 9 × 105 FCFU/ml and 8 . 1 × 104 FCFU/ml , respectively , equivalent to a multiplicity of infection of 0 . 8 and 0 . 2 , respectively . For analysis of cellular activation following short-term rotavirus exposure , splenocytes were cultured with rotavirus for 1 h followed by washing , supernatant fluid collection and replacement with fresh RF10 . The splenocytes were cultured for a further 23 h in the absence of rotavirus . Rotavirus infectious titres in these supernatant fluids , as determined by titration in MA104 cells , were <100 particles/well . For analysis of cell activation by secreted factors , the supernatant fluid ( 150 μl ) of splenocyte cultures that had been stimulated with rotavirus for 1 h and cultured for a further 23 h was applied with 50 μl of fresh RF10 to naive , unstimulated NOD splenocytes for 24 h . Purified B , T and DN cells were stimulated with 100 ng/ml RRV alone ( 5×105 cells/well ) or in combination ( at 4 B/T cells:1 DN; total 5×105 cells/well ) for 24 h at 37°C and 5% CO2 . DC ( pDC or cDC; 1×105 ) cultured in the presence of B or T cells ( 4×105 ) were stimulated with RRV as above . Splenocytes depleted of CD11c+ DC ( 5×105 cells ) , and cultures of DC ( 2 . 5×105 cells ) with B cells ( 2 . 5×105 cells ) , also were stimulated with RRV . As controls , cells were stimulated with Escherichia coli serotype 0111:B4 LPS ( Sigma ) at 100 ng/ml unless otherwise indicated; 50 ng/ml PMA and 500 ng/ml Ionomycin C ( Sigma ) ; 50 μg/ml poly IC ( high molecular weight , InvivoGen , CA ) ; 1 μg/ml Imiquimod ( InvivoGen ) ; or diluent . As indicated , cells or virus were pre-treated with various antibodies prior to culture , as detailed above . All cultures were performed in triplicate , unless indicated otherwise . Cells were stained with the following conjugated antibodies from BD Biosciences as appropriate: anti-CD19 ( ID3 ) -Pacific Blue , anti-CD19-FITC , anti-CD3 ( 500A2 ) -AlexaFluor700 , anti-CD3-PE , anti-CD8α ( 53-6 . 7 ) -PerCP , anti-CD8α ( 53-6 . 7 ) -FITC , anti-CD4 ( RM4-5 ) -Pacific Blue , anti-MHC II-FITC . The activation of B cells , CD8+ T cells ( CD3+CD8α+ ) , CD4+ T cells ( CD3+CD4+ ) and APCs ( CD3−CD19−MHC II+ ) was analysed with anti-CD69 ( H1 . 2F3 ) -PECy7 . In some experiments , APCs were further classified by CD11c expression . APC and B cells also were stained with anti-H2-Kd ( SF1-1 . 1 ) -Biotin ) , anti-MHC II ( OX-6 ) -FITC , anti-CD80 ( 16-10A1 ) -PE and anti-CD86 ( GL1 ) -Allophycocyanin . Cells from mice with a C57BL/6 genetic background were stained with H2-Db ( 27-14-8 ) -Biotin ( BD ) for detection of APC . Biotin was detected with Streptavadin-Allophycocyanin AlexaFluor750 ( Invitrogen ) . For detection of IGRP-specific CD8+ T cells , splenocytes were stained with the IGRP206-214 tetramer or the negative control TUM H-2Kd tetramer ( ImmunoID , Melbourne , Australia ) as described previously [14] . Cells were stained with 7-amino-actinomycin D ( Invitrogen ) to exclude dead cells , as indicated . At least 100 , 000 cells were analysed for each sample . For total proliferation measurement , unsorted splenocytes ( 5×105 cell/well ) in U-bottomed 96-well trays were stimulated with 100 ng/ml rotavirus for 44 h and 68 h at 37°C and 5% CO2 . As controls , cells were stimulated with 100 ng/ml LPS or left unstimulated . Controls lacking cells also were included . 3H-Thymidine ( MP Biomedicals ) was added at 1 μCi/well for the final 20 h of incubation . Cells were collected onto glass fibre filters using an automated cell harvester ( Skatron Instruments ) and β-emission recorded by a liquid scintillation counter ( Packard Bell ) in counts/min . To analyse B cell proliferation , splenocytes stimulated as above for 48 h and 72 h were stained with anti-CD3 , anti-CD19 , anti-CD69 and anti-Ki67 ( B56 ) -PE or mouse IgG1-PE using the eBioscience FoxP3 staining buffer kit according to the manufacturer's instructions . Supernatant fluids pooled from 3 replicate samples were collected from either unsorted or sorted splenocyte cultures during one or more independent experiments and stored at −80°C . The concentration of IFNα present in these pooled samples was measured with the FlowCytomix Mouse IFN-α detection kit ( eBioscience ) . The Student's t-test , with or without Welch's correction was used . Data are derived from one experiment that was representative of the two independent experiments generally conducted . Where a single experiment only was conducted , this is indicated in the Figure legend . On graphs , error bars indicate the standard error of the mean ( SEM ) for replicates within a single representative experiment , unless otherwise indicated in the Figure legend . Significant differences are shown in Figures as follows: * p<0 . 05 , p<0 . 01 , * p<0 . 001 .	Understanding how viruses contribute to type 1 diabetes development is vital for disease prevention . Infection of children at-risk of diabetes with the gastrointestinal pathogen rotavirus is associated with increased immune responses to pancreatic islets , leading to the proposal that rotavirus infection may accelerate progression to diabetes . In a mouse model , we showed previously that rotavirus accelerates diabetes onset , in conjunction with virus spread to the lymph nodes , draining the intestine and pancreas . At these sites , rotavirus associates with antigen-presenting cells of the immune system , including dendritic cells , leading to their maturation , and induces the activation of B and T cells . Here we use this mouse model to define the contribution of rotavirus-exposed antigen-presenting cells to the activation of neighboring B and T cells . We found that rotavirus-exposed dendritic cells induce B and T cell activation through secretion of type I interferon . Activation of these dendritic cells depends on recognition of viral RNA by Toll-like receptor 7 . Our studies suggest that this mechanism of B and T cell activation may occur in RRV-infected mice and contribute to their accelerated diabetes development . A similar mechanism may be involved in the enhanced islet autoantibody responses of children following rotavirus infection .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "medicine", "infectious", "diseases", "immune", "cells", "cytokines", "b", "cells", "clinical", "immunology", "immunity", "innate", "immunity", "autoimmune", "diseases", "antigen-presenting", "cells", "diabetes", "mellitus", "type", "1", "t", "cells", "viral", "diseases", "rotavirus", "infection", "immune", "system" ]	2014	Rotavirus Activates Lymphocytes from Non-Obese Diabetic Mice by Triggering Toll-Like Receptor 7 Signaling and Interferon Production in Plasmacytoid Dendritic Cells
Most cellular processes depend on intracellular locations and random collisions of individual protein molecules . To model these processes , we developed algorithms to simulate the diffusion , membrane interactions , and reactions of individual molecules , and implemented these in the Smoldyn program . Compared to the popular MCell and ChemCell simulators , we found that Smoldyn was in many cases more accurate , more computationally efficient , and easier to use . Using Smoldyn , we modeled pheromone response system signaling among yeast cells of opposite mating type . This model showed that secreted Bar1 protease might help a cell identify the fittest mating partner by sharpening the pheromone concentration gradient . This model involved about 200 , 000 protein molecules , about 7000 cubic microns of volume , and about 75 minutes of simulated time; it took about 10 hours to run . Over the next several years , as faster computers become available , Smoldyn will allow researchers to model and explore systems the size of entire bacterial and smaller eukaryotic cells . One hurdle to the computational modeling of cellular systems is the lack of adequate tools . If one assumes that molecules inside cells are well-mixed , and that they behave deterministically , then one can model the chemical reactions that cells use to operate with differential equations ( recently reviewed by Alves and coworkers [1] ) . However , these assumptions are frequently inadequate . Firstly , most cellular processes depend at least to some extent on intracellular spatial organization . For example , cell signaling systems transmit signals across significant distances within subcellular compartments and across intracellular membranes , such as the nuclear envelope . Also , cell division systems segregate one cell into two and regulate the partition of molecular components . Secondly , many cellular outputs exhibit substantial random variation [2] , which must arise from random differences in molecular collisions . Examples range from the random switching of swimming Escherichia coli bacteria between so-called running and tumbling states [3] to cell-to-cell variation in the operation of cell signaling systems [4] , [5] . More generally , stochastic behavior is likely to affect the outcomes of essentially all cellular processes . Representation of this complexity requires algorithms and programs that model cellular processes with spatial accuracy [6] , and that model the chemical reactions by which they operate with stochastic detail [7] . Computational biologists have pursued four main approaches to simulating biochemical systems with spatial and stochastic detail . These differ in how they represent space , time , and molecules ( Table 1 ) , which in turn affects the classes of biological systems that they can simulate appropriately . ( i ) The spatial Gillespie method [8] is based on Gillespie's stochastic simulation algorithms [9] . It divides the simulation volume into a coarse lattice of subvolumes , each of which contains many molecules of interest . This method can be computationally efficient because it tracks the total number of individual classes of molecules per subvolume , rather than individual molecules [10] . However , the lattice structure it uses to divide space into subvolumes does not work well for realistic membrane shapes , which require special treatment [11] . ( ii ) The microscopic lattice method subdivides space into a much finer lattice , so that each volume can contain zero or one molecule . In this method , molecules diffuse by hopping between sites and can react with molecules in neighboring sites . It naturally lends itself to studies of oligomerization and complex formation [12] , and of the effects of macromolecular crowding on reactions [13] . It has not found wide use for studying cell-sized processes due to the facts that it has high computational demands and specific lattice structures affect simulated reaction rates differently [14] , although recent techniques may circumvent these challenges [15] . ( iii ) Particle-based methods , the primary focus of this article , are the most widely used spatial stochastic methods [7] . These represent individual molecules with point-like particles that diffuse in continuous space over fixed time steps; molecules can react when they collide . The fact that these models use continuous space makes realistic membrane geometries relatively easy to represent [16] , avoids lattice-based artifacts , and offers high spatial resolution . The need to track individual molecules , however , imposes high computational demands , so particle-based methods are about a factor of two slower than spatial Gillespie methods [10] . Finally , ( iv ) Green's function reaction dynamics ( GFRD ) methods [17] enable especially accurate particle-based simulation . GFRD methods step the simulation from the exact time of one individual reaction to the exact time of the next . This makes these methods ideal for systems that can have long delays between individual reactions , but very computationally intensive for most cellular processes [10] . The dominant particle based simulators are ChemCell [18] , MCell [19] , [20] , and Smoldyn [21] , [22] ( see also Table 2 ) . These programs have many common features , but differ in other features and in the quantitative accuracy of their simulations . Of the three , ChemCell has the fewest features , but is particularly easy to use and is the only simulator that supports both spatial and non-spatial simulations . MCell , the oldest program [23] , has been used the most , produces the highest quality graphics [16] , and has a number of features that make it particularly well suited to simulating cellular processes involved in synaptic transmission [19] ( for example , using MCell , it is easy to release agonist and antagonist ligands onto a cell using pulse trains ) . Smoldyn is a relative newcomer , but yields the most accurate results and runs the fastest ( see Text S1 ) . Smoldyn also has a number of attributes , listed in Table 2 and below , which make it well suited to modeling a wide range of cellular processes . This article focuses on the latest version of Smoldyn , Smoldyn 2 . 1 . Smoldyn 1 . 0 embodied several algorithms that were based on Smoluchowski reaction dynamics [21] . It and subsequent versions were used to investigate a spatial version of the classic Lotka-Volterra chemical oscillator [24] , diffusion on hair cell membranes [25] , protein sequestration in dendritic spines [26] , diffusion in obstructed spaces , and intracellular signaling in E . coli chemotaxis [27]–[29] ( Figure 1 ) . Smoldyn 2 . 1 preserves the original focuses on accuracy and efficiency but offers significantly improved functionality . In particular , it can represent realistic membrane geometries , simulate diffusion of membrane-bound molecules , and accurately simulate a wide variety of molecule-membrane interactions [30] . To make it as general a simulator as possible , Smoldyn 2 . 1 also supports spatial compartments , rule-based reaction network generation [31] , [32] , molecules with excluded volume , conformational spread interactions , and over fifty run-time commands for system manipulation and observation . We anticipate that Smoldyn will be particularly useful for ( i ) investigating cellular systems , such as signaling , division , and metabolic systems , ( ii ) studying basic biophysical phenomena , such as the effects of macromolecular crowding on molecular diffusion , and ( iii ) helping to quantify microscopy data [33] , such as diffusion rates investigated by FRAP ( fluorescence recovery after photobleaching , which is based on the time it takes fresh fluorophores to diffuse into the bleached volume ) . The algorithms that Smoldyn uses , and the program's name , derive from a biophysical description of space and chemical reactions that von Smoluchowski defined in 1917 [34] . In his description , each molecule of interest moves by mathematically ideal Brownian motion [35] , which implicitly arises from its collisions with solvent molecules and other non-reacting molecules . Bimolecular chemical reactions occur when two reactants diffuse to within a distance called the encounter radius [36] , or binding radius [21] . With the Smoldyn algorithms , all simulated dynamics approach those of the Smoluchowski model ( within computer round-off error ) as the simulation time step is reduced towards zero . Smoldyn reads a configuration file that describes the system or cellular process under study . This file lists the system dimensionality , initial numbers of molecules , membrane locations , chemical reactions , and the rules for molecule-surface interactions . It also contains directives for a virtual experimenter , a software agent under the direction of the human researcher , which can measure and manipulate the system during its simulation . For example , at any given point during the simulation run , the virtual experimenter can count the number of particular molecules or add a new surface to represent intrusion of a membrane vesicle into the simulated space . After calculating simulation parameters , Smoldyn performs the simulation with fixed-length time steps , typically set by the researcher to be shorter than characteristic reaction or diffusion timescales ( 0 . 1 ms time steps often work well [27]–[29] , but see [22] for a more thorough discussion ) . At each time step , Smoldyn diffuses simulated molecules , performs chemical reactions , processes molecule-membrane interactions , and outputs quantitative data to one or more text files . Smoldyn can display the simulated system to a graphics window as it is computed , or it can work in a text-only mode for more efficient operation . Smoldyn represents molecules and surfaces . Molecules can be in free solution , such as cytoplasmic proteins , or bound to surfaces , such as ion channels or peripheral membrane proteins . Surface-bound molecules can bind to the front or back of the surface , or can be transmembrane with an “up” or “down” orientation . The latter two states can differentiate the two orientations of transmembrane proteins , such as whether the ligand-binding portion of a receptor faces the intra- or extracellular medium . Surfaces , which might represent biological membranes or the sides of a reaction vessel , are modeled as being infinitely thin and locally smooth . Each surface is composed of panels . Panels can be rectangular , triangular , spherical , hemispherical , cylindrical , or disk-shaped . Of course , one only needs triangles to model arbitrarily complex surfaces [37] and for that reason triangle-based meshes are widely used for experimentally-derived membrane geometries ( reviewed by O'Rourke [38] ) . However , the other panel shape options can simplify membrane definitions and improve computational efficiency . For example , Smoldyn can represent a nuclear membrane as a sphere rather than as an icosahedron . Two utility programs , distributed with Smoldyn , simplify surface definitions: one ( wrl2smol ) converts triangle data from the Virtual Reality Modeling Language format , which is widely used for microscopy , to Smoldyn format , and the other ( SmolCrowd ) generates fields of random non-overlapping circles or spheres for investigating macromolecular crowding . We summarize the core algorithms here and in Figure 2 . For more detail , see [21] , [22] , [30] or the Smoldyn user's manual ( supplied with the program ) . ( i ) To simulate diffusion , Smoldyn moves each molecule at each time step using Gaussian-distributed random displacements along each x , y , and z coordinate [21] . If the molecule was surface-bound , Smoldyn then deposits it back upon the surface along the local normal vector ( an exact method for planar surfaces because orthogonal projections of 3-dimensional Gaussian probability densities are 2-dimensional Gaussian probability densities; for this to be accurate for curved surfaces , the radius of curvature needs to be much larger than the length of the average diffusive step ) . Smoldyn can also simulate molecular drift , for example arising from a flow of solvent molecules , and anisotropic diffusion . ( ii ) To simulate interactions between solution-phase molecules and membranes , Smoldyn simulates impermeable membranes with ballistic reflections [21] , partially permeable and adsorbing membranes [39] using interaction probabilities that are exact at steady-state and very accurate away from steady-state [30] , and periodic ( or toroidal ) boundaries with so-called jump surfaces , with which molecules that diffuse out of one side of the system immediately diffuse into the opposite side . These jump surfaces can also be used to add holes to otherwise impermeable surfaces . In addition , “unbounded-emitter” surfaces cause molecular concentrations to equal those that would be observed if the system were unbounded , by absorbing molecules with probabilities based on the emitter positions [30] . ( iii ) To handle transitions from surface-bound states , Smoldyn assigns these reactions a probability computed as described below for first order reactions . Smoldyn then displaces desorbed molecules away from the surface using probability densities that account for diffusion that occurs between the time of desorption and the end of the time step in which desorption occurred ( the density is an error function if desorption is reversible and combines a Gaussian and an error function if desorption is irreversible [30] ) . ( iv ) To handle reaction products without the respective reactants , Smoldyn simulates “zeroth” order chemical reactions . This allows Smoldyn to represent , for example , protein expression , as the de novo appearance of protein in the reaction volume . Smoldyn adds a Poisson-distributed random number of product molecules to random locations in the simulation volume , or within a smaller compartment ( such as a cell nucleus ) , at each time step [21] . ( v ) To handle first order reactions , such as protein conformational changes , Smoldyn performs each reaction with probability 1–exp ( –kΔt ) , where k is the reaction rate constant and Δt is the time step , at each time step . Smoldyn uses an expanded version of this equation if a single reactant can undergo any of several reactions [19] , [21] . ( vi ) Smoldyn performs second order reactions , which have the form A + B → C , when two reactants diffuse closer together than their “binding radius . ” Smoldyn computes these distances from reactant diffusion coefficients , reaction rate constants , and the simulation time step [21] , with results that are typically similar to the sums of the physical radii of the reactants ( e . g . reactants with 10 µm2s−1 diffusion coefficients and a 106 M−1s−1 reaction rate constant have a binding radius of 3 . 4 nm when a 0 . 1 ms time step is used ) . Tournier et al . [40] showed that this method is indistinguishable from a more accurate one in which the reaction probability varies as a function of the inter-molecular separation . Although Erban and Chapman recently extended our method to include a fixed reaction probability , so that binding radii would be larger [41] , Smoldyn does not support their extension because it is less computationally efficient and it does not account for reversible reactions . Neither Smoldyn nor Smoluchowski theory directly simulate reactions with orders that are greater than two or non-integer , but rather decompose these into individual bimolecular reactions . ( vii ) Smoldyn typically deposits reaction products at the reaction location . However , it places them elsewhere in three important cases . ( a ) If two reaction products can react with each other , Smoldyn separates them by a distance called the unbinding radius [21] ( typically similar to the binding radius for these molecules ) , which produces accurate reaction rates by reducing the probability of product recombination [42] . ( b ) If the reaction simulates conformational spread , in which allosteric changes in protein activity are transmitted by direct contact between neighboring proteins [43] , then Smoldyn places the products at the same locations previously occupied by the reactants . For conformational spread , the typical reaction notation is A + B → A' + B , with the interpretation that B converts A to A' through direct contact . Smoldyn simulates conformational spread reactions with first order reaction rates if the reactants are within a pre-defined interaction distance . ( c ) If the reaction is used to give molecules excluded volume , then Smoldyn separates the products by a distance that is slightly larger than the binding radius and places them on the line that connected the reactants . While this method is not as accurate as ballistic reflection methods , it is still useful; for example , it can assure that molecules cannot pass each other in pores and it can be used to separate molecules by realistic distances . ( viii ) So that the user does not need to enumerate the “combinatorial explosion” [44] of individual species and reactions that result when proteins form multimeric complexes or are post-translationally modified ( e . g . phosphorylation ) , Smoldyn generates species and reactions automatically from lists of interaction rules . These rules include molecule binding sites , modification sites , and allosteric interactions . Smoldyn performs this reaction network expansion as new species and reactions arise in a simulation [45] using the libmoleculizer software module [32] ( derived from Moleculizer [31] ) . Taken together , this set of algorithms allows Smoldyn to represent most biochemical processes that take place among proteins and small molecules , in 1 , 2 , or 3 dimensional space and on surfaces and membranes . For example , this capability will bring most cell signaling systems within reach of particle-based modeling . Currently absent , however , from Smoldyn and nearly all comparable simulators are algorithms that specifically address moving or distorting membranes , and the dynamics of biopolymers ( including microtubules , actin filaments , and most conspicuously , DNA ) . Figure 3 shows that simulation data for several of the core Smoldyn algorithms agree well with analytical theory for wide ranges of rate constants , and that this holds for both cumulative results and fluctuations . Using Pearson's χ2 test , we found no statistically significant differences between data and theory ( see Text S2 ) . These agreements , along with additional tests presented in Text S1 , the Smoldyn user's manual and theoretical work in the algorithm derivations [21] , [30] , show that individual Smoldyn algorithms are quite accurate both at and away from steady state . In addition , Smoldyn simulations that combine multiple algorithms become exact as time steps are reduced towards zero . On the other hand , Smoldyn simulations with multiple processes and finite time steps are necessarily approximate , in part because individual simulated molecules can only take part in a single reaction or adsorption events during single time steps . Smoldyn simulation run times scale linearly with the number of simulated molecules ( Text S3 ) . However , the number of chemical reactions minimally affects Smoldyn run times because the Smoldyn algorithms iterate over molecules rather than possible reactions . In contrast , spatial Gillespie methods scale linearly with the number of reactions [10] , or logarithmically if they use the Gibson-Bruck algorithm [46] . We compared the run times for ChemCell , MCell , and Smoldyn using identical models of a Michaelis-Menten reaction . These models comprised 10 , 000 molecules ( 10% enzyme , 90% substrate initially ) and ran for 10 s of simulated time in 1 ms time steps . As Table 2 shows , Smoldyn took 47 s to run this test ( on a MacBook Pro laptop with a 2 . 33 GHz Intel Core 2 Duo processor and OS 10 . 4 . 11 , and with single-threaded operation ) , which was more than a factor of 2 faster than either ChemCell or MCell . All simulated results agreed well with mass action theory ( Text S1 ) . To give a sense of scale , there are 11 , 200 molecules in Figure 1A [29] , about 40 , 000 protein molecules that regulate chemotaxis in a single E . coli [47] , and about 40 , 000 proteins in the core pheromone response system in a single Saccharomyces cerevisiae . This comparison is likely to be representative of typical simulations because bimolecular reaction simulation comprises most of the run time for Smoldyn , and likely for the other simulators as well . Smoldyn also supports multiple program threads although these do not provide substantial speed improvements at present . We used Smoldyn's capabilities to explore a simple model . When haploid Saccharomyces cerevisiae of opposite mating types ( MATa and MATα ) are in proximity , they can mate and form a diploid [48] . MATa cells detect a pheromone ( α-factor ) , secreted by MATα cells , and use the concentration and gradient of the pheromone to grow toward and discriminate among potential MATα partners [49] , [50] . Exposure to pheromone also increases the rate at which MATa cells secrete a pheromone-degrading protease , Bar1 [51] , [52] . Barkai et al . showed that a uniform concentration of Bar1 would attenuate α-factor signals coming from distant MATα cells more than from close cells , thus helping MATa cells locate the closest potential mating partners [53] . However , this result does not apply to low densities of MATa cells . Here , the Bar1 concentration should be highest near MATa cells , both because Bar1 binds to cell walls [54] and because Bar1 dissipates as it diffuses away . This is precisely the experimental setup in partner discrimination assays , such as that shown in Figure 4A . In these assays , MATa cells are surrounded by MATα cells that secrete different amounts of α-factor . Jackson and Hartwell found that MATa cells usually choose the strongest pheromone emitter , and make this choice most accurately when the MATa cells express Bar1 [50] , [55] . We modeled this experiment with Smoldyn . Figure 4A shows a snapshot of the Smoldyn simulation . There are two kinds of MATα cells: one cell secretes a normal level of α-factor ( the “target” cell ) and other cells secrete 5% of the normal α-factor level due to mutations in mfα1 MFα2 [50] . These cells surround a central MATa cell . Onto the surface of the MATa cell we introduced 6622 stationary Ste2 receptor molecules [56] which bound α-factor according to experimentally measured rates [57] . In some simulations , the MATa cell secreted Bar1 . We chose the Bar1 secretion rate , and its catalytic proteolysis reaction rate , so that Bar1 would increase the receptor occupancy EC50 by about a factor of 5 ( Figure 4B ) , which is a conservative estimate for the experimental shift [57] . The simulations shown here did not permit Bar1 to bind to cell membranes for simplicity , but agree well with others ( not shown ) that permitted binding . Finally , we estimated diffusion coefficients with the Stokes-Einstein equation , using the assumption that the extracellular viscosity is similar to that of mammalian cytoplasms [58] . ( See Text S4 for further details . ) We simulated systems at each of nine α-factor secretion rates . During the first 100 s of each secretion rate , the systems equilibrated to a nearly steady state , which we assessed using time-dependent concentrations and concentration gradients of all simulated molecular species . Every 2 s for the next 400 s , Smoldyn recorded the number and the mean position of receptor molecules bound to α-factor . We defined the vector that pointed from the MATa cell center to one of these mean positions , which we label r , to be the “position signal” that the MATa cell received . Each vector represented the cell's instantaneous measurement of the local 3-dimensional α-factor gradient . Next , we defined the “GPCR-α gradient” as the component of r that points towards the target MATα target cell , divided by the radius of the MATa cell . This created a simple metric that could range from –1 to 1; it is 0 if α-factor binds randomly to receptors , and is 1 if α-factor only binds the receptors closest to the target MATα cell . Figure 4C shows GPCR-α gradient values , averaged over all time points of each α-factor secretion rate . It shows that Bar1 substantially increased the measured gradient of receptor-bound pheromone over a wide range of α-factor secretion rates . This occurred because Bar1 degraded α-factor molecules as they diffused past the MATa cell , which steepened the local α-factor concentration gradient . Figure 4D shows a related but likely more physiologically relevant quantity: it shows the average absolute angle between the r vectors and the direction to the target cell . Again , the average is over all time points of each α-factor secretion rate . This figure shows that Bar1 decreased the MATa cell's angular measurement error , again over a very wide range of α-factor secretion rates . We believe that this improvement arose from the steeper concentration gradients . These effects are consistent with Jackson and Hartwell's finding that MATa cells choose the MATα cell that produces the highest level of pheromone from among potential mating partners [50] . In further simulations , we surrounded a MATa cell with 3 target and 3 low-secreting MATα cells and found that Bar1 had the same discrimination enhancing effect . Again , this paralleled experimental results [50] . Many features of Smoldyn facilitated the above investigation . The simulations used nearly diffusion-limited reaction rates for the Bar1 protease reaction , which Smoldyn handles accurately; they tracked up to 190 , 000 molecules at a time , which required high computational efficiency; and they used system boundaries that absorbed molecules so that the distribution of molecules in the bounded system roughly matched the distribution if space extended indefinitely [30] . In addition , Smoldyn's real-time graphics helped us design the simulation and the ability to use spherical cells simplified model building and accelerated simulation run time . We could have used rule-based modeling to automatically generate the receptor-ligand complex species and its reactions , but declared them explicitly instead to make the model simpler . Many aspects of the biochemical reactions that animate cellular processes are inherently spatial . These include diffusion in complex spaces , sub-cellular localization , and transient membrane associations . Additionally , important cellular processes often rely on molecular species present in low numbers . Computer models that ignore these spatial and stochastic aspects of biological function clearly cannot offer insights into them . For example , non-spatial , non-stochastic models of the E . coli chemotaxis signaling network were invaluable for elucidating the basic system architecture [59] but could not aid investigation in the variation in signals received by different individual flagellar motors [28] , the localization of molecules of the CheZ phosphatase [27] , [60] , or the formation of intracellular concentration gradients of the CheY signal transmitting protein [29] . Moreover , predictions about the behavior of systems where space and stochasticity are factors are approximate at best and may be qualitatively incorrect [61] . For example , a simple chemical oscillating system executes nearly sinusoidal oscillation when simulated without spatial or stochastic detail but exhibits intermittent boom and bust cycles when space and stochasticity are included [24]; also , spatial correlations can cause distributive multiple phosphorylation mechanisms , in which kinases release their substrates between phosphorylations , to lose ultrasensitivity to the substrate concentration [62] . We devised Smoldyn 2 . 1 to help address the need for accurate and efficient spatial stochastic simulation software . We designed it to simulate molecules and membranes , and events including diffusion , chemical reactions , adsorption , and desorption . We demonstrated the capabilities of Smoldyn with a model of signaling between yeast cells through a diffusible pheromone . The model suggested that , by degrading pheromone , the protease secreted by MATa type yeast cells steepened the local pheromone concentration gradient , which helps cells locate and choose among potential mating partners . On a contemporary laptop computer ( 2006 MacBook Pro ) Smoldyn can perform useful simulations involving assemblages of more 100 , 000 molecules with relative ease . This power is sufficient to investigate many biochemical systems , including the E . coli chemotaxis signaling system and signaling between neurons . Extrapolating computer power with Moore's law , Smoldyn should be able to simulate all 2 . 6 million proteins in an E . coli cell [63] ( or a mitochondrion or yeast nucleus , which are roughly the same size ) within 5 years , still on a single laptop computer . Using more powerful computers , such as Beowulf clusters [64] or the current NVIDIA Tesla [65] , Smoldyn should be able to simulate entire populations of complete cells over many generations , within a decade . However , many challenges to simulations of entire cells and populations remain . First , neither Smoluchowski dynamics nor Green's Function Reaction Dynamics are wholly adequate . For that reason , researchers will need to develop new physical theories for reactions and diffusion in crowded cytoplasms , the mechanical interactions between cytoskeletal filaments and cell membranes , and the functions of extended macromolecular complexes . These theories , which may be partially empirical , need to isolate the essential behaviors of these processes so that they can be modeled . Second , these theories will need to be embodied in algorithms so that modelers can account for the corresponding processes in their cellular models . Third , not all aspects of cellular processes require attention to spatial and stochastic detail ( for example , there are likely to be over a million ATP molecules in an E . coli cell [63] , probably with weak spatial gradients ) , so multi-level algorithms , such as those that combine stochastic and deterministic methods [66] , will be valuable . Fourth , although it is possible that there might be increasingly standardized and quantitative experiments that assist modeling of some cellular processes , almost by definition , the experiments on newly explored cellular processes will be diverse and incomplete . Any modeling effort that expects to contribute to the understanding of newly articulated cellular phenomena will require simulation software that can work with diverse and incomplete experimental results . We wrote the core portion of Smoldyn in the C programming language . This core is linked to the OpenGL library for graphics , the libtiff library for saving tiff format images , the libmoleculizer library for rule-based reaction network generation [32] , the POSIX library for threaded operation , and the SIMD-oriented fast Mersenne Twister library [67] for random number generation . The combined program compiles and links on Macintosh OS X or Linux systems with the gcc compiler . Windows versions are cross-compiled from Macintosh using the mingw compiler . We used Valgrind to check for memory leaks and gprof for code profiling . All source code , makefiles , executable applications , example configuration files , utility programs , and documentation can be downloaded for free from www . smoldyn . org . The code is licensed under the Gnu General Public License .	We developed a general-purpose biochemical simulation program , called Smoldyn . It represents proteins and other molecules of interest with point-like particles that diffuse , interact with surfaces , and react , all in continuous space . This high level of detail allows users to investigate spatial organization within cells and natural stochastic variability . Although similar to the MCell and ChemCell programs , Smoldyn is more accurate and runs faster . Smoldyn also supports many unique features , such as commands that a “virtual experimenter” can execute during simulations and automatic reaction network expansion for simulating protein complexes . We illustrate Smoldyn's capabilities with a model of signaling between yeast cells of opposite mating type . It investigates the role of the secreted protease Bar1 , which inactivates mating pheromone . Intuitively , it might seem that inactivating most of the pheromone would make a cell less able to detect the local pheromone concentration gradient . In contrast , we found that Bar1 secretion improves pheromone gradient detectability: the local gradient is sharpened because pheromone is progressively inactivated as it diffuses through a cloud of Bar1 . This result helps interpret experiments that showed that Bar1 secretion helped cells distinguish between potential mates , and suggests that Bar1 helps yeast cells identify the fittest mating partners .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "computational", "biology/synthetic", "biology", "cell", "biology/cell", "signaling", "biophysics/theory", "and", "simulation", "computational", "biology/computational", "neuroscience", "biochemistry/theory", "and", "simulation", "computational", "biology/systems", "biology" ]	2010	Detailed Simulations of Cell Biology with Smoldyn 2.1
Spike timing-dependent plasticity ( STDP ) modifies synaptic strengths based on timing information available locally at each synapse . Despite this , it induces global structures within a recurrently connected network . We study such structures both through simulations and by analyzing the effects of STDP on pair-wise interactions of neurons . We show how conventional STDP acts as a loop-eliminating mechanism and organizes neurons into in- and out-hubs . Loop-elimination increases when depression dominates and turns into loop-generation when potentiation dominates . STDP with a shifted temporal window such that coincident spikes cause depression enhances recurrent connections and functions as a strict buffering mechanism that maintains a roughly constant average firing rate . STDP with the opposite temporal shift functions as a loop eliminator at low rates and as a potent loop generator at higher rates . In general , studying pairwise interactions of neurons provides important insights about the structures that STDP can produce in large networks . Spike timing-dependent plasticity ( STDP ) is a widespread mechanism that modifies synapses on the basis of the intervals between ensembles of pre- and postsynaptic spikes [1] , [2] . The most prevalent form of STDP involves potentiation of the synapse when presynaptic spikes precede postsynaptic spikes , and depression for the reverse ordering [3] . STDP is inherently a local synaptic modification rule because the determinant of synaptic modification is the timing of pre- and postsynaptic spikes . Neurons , on the other hand , are typically embedded in interconnected networks in which each neuron receives thousands of synapses from other neurons [4] , [5] . A number of studies have explored how STDP shapes the distribution of synaptic weights for a population of synapses converging onto a single neuron [6]–[10] . Here , we consider the more difficult problem of bridging the gap between the locality of STDP and the global structures that it generates in a recurrent network of spiking neurons . The problem of STDP in a recurrent network has been addressed before in a number of studies [11]–[22] . The generally antisymmetric shape of the STDP window , in which reversing the ordering of pre- and postsynaptic spikes reverses the direction of synaptic change , led to the proposal that this synaptic modification rule should eliminate strong recurrent connections between neurons [11] , [23] . This idea has recently been expanded by Kozloski and Cecchi [21] to larger polysynaptic loops in the case of “balanced” STDP in which the magnitudes of potentiation and depression are equal . These authors also showed that balanced STDP organizes network neurons into in- and out-hubs . The possibility of enhancing recurrent connections through pair-based STDP was also proposed by Song and Abbott [11] and is further explored by Clopath and colleagues [22] in a more complex model . An excessively active group of neurons has been shown to decouple from the rest of the network through STDP [13] , and in presence of axonal delays , STDP enhances recurrent connections when the neurons fire in a tonic irregular mode [14] . Here , we show that , surprisingly , all of these network properties can be explained through an understanding of the effect of STDP on pairwise interactions of neurons . This provides an analytically tractable way of relating the structures arising in a network to properties of the STDP model being used to modify synapses . The simplest form of STDP that we consider is balanced with equal potentiation and depression domains , i . e . with . In this case , the coefficient vanishes because the baseline potentiation and depression cancel each other . In addition , the coefficients and are equal . These conditions greatly simplify the system of equations ( 2 ) . We visualize the dynamics of the synaptic pair by phase planes ( figure 2 ) . Note that all phase planes throughout this paper are snapshots of the dynamics for a given network firing rate , and as we will see below , they should be recomputed if the baseline firing rates change appreciably . When the baseline firing rates are equal ( ) , the values of the synaptic weights do not change when and are equal , i . e . the synaptic drift is zero on the line ( figure 2A , solid line ) . However , this equilibrium is unstable . If the two synapses have unequal strengths , the stronger synapse grows even stronger and weakens the other synapse until they reach the boundary of their allowed range ( figure 2A , arrows ) . Then , the synapses continue their dynamics along the boundary edge until they reach the upper-left ( ) or lower-right ( ) corner of the state space ( figure 2A , filled circles ) , depending on which synapse was stronger to begin with . These “attractors” of the synaptic dynamics indicate that , at steady-state , loops between pairs of neurons are eliminated by this form of STDP . A linear system of differential equations cannot have more than one attractor . The existence of two attractors here is a consequence of restricting the dynamics to a limited range , and it suggests that STDP favors unidirectional connections and eliminates loops in a network ( figure 3A ) , in agreement with the results of [21] . The attractors also imply that , for each neuron , the strengthening of an incoming synapse is accompanied by the weakening of an outgoing synapse and vice versa . If we consider the effect of this interplay at the level of a network , it is expected that for each neuron the number of above-threshold incoming and outgoing synapses will be linearly related with a coefficient of -1 , such that their sum remains constant . Our simulation results and previous work [21] also confirm this prediction ( figure 3B ) . When the baseline rates of the two neurons are not equal , the line of equilibrium is tilted to ( figure 2B , see Text S1 ) . As a result , the size of the basins of the two attractors differ , and the outgoing synapses of the neuron with the higher firing rate are more likely to strengthen , while its incoming synapse are likely to weaken ( figure 2B , top-left corner ) . Conversely , outgoing synapses from the neuron with the lower firing rate are more likely to weaken and its incoming synapses strengthen . If we generalize this behavior to the context of a network , an important prediction can be made: neurons with low initial firing rates should attract strong excitatory synapses onto themselves but project weaker synapse to other neurons . Neurons with high firing rates should experience the opposite trend; they lose incoming synaptic input through synaptic weakening , while their outgoing synapses strengthen . Therefore , if the external input is biased to give a sub-population of excitatory neurons an initially higher ( lower ) firing rate than the rest of the network , these neurons will become out-hubs ( in-hubs ) through STDP . We tested this by setting the mean of the external input to the neurons such that the initial firing rate of the first hundred excitatory neurons ( 1–100 ) was , the initial firing rate of the next hundred excitatory neurons ( 101–200 ) was , and the initial firing rate for the rest of the excitatory neurons ( 201–500 ) together with that of the inhibitory neurons was . The results show that the sub-population with high initial rate indeed turns into out-hubs , while the sub-population with low initial rate turns into in-hubs once the synaptic weights reach steady state ( figure 3C ) . Another related prediction is that the firing rates of neurons in the network tend to equalize through STDP . This is because neurons with high initial firing rate become out-hubs , thereby receiving less input from the rest of the network , which lowers their final , equilibrium firing rates . At the same time , they share their initial high firing rate with the other neurons of the network through the strengthening of their outgoing synapses . The opposite happens to neurons with low initial firing rates; they turn into in-hubs . As a result the final firing rates of the neurons become homogenized across the network at steady-state . To test this prediction , we tracked the evolution of the average firing rates of the above three sub-populations throughout the simulation . The results confirm that the final firing rates of all three sub-populations equalize once the synaptic weights reach steady-state ( figure 3D ) . As another prediction , the steady-state mean of the synaptic weights is expected to converge to the midpoint of its allowed range , regardless of the initial distribution of weights and the initial firing rate of the network . This is due to the precise balance between the potentiation and depression domains of STDP in this case . If the initial mean is already at the midpoint of the allowed range , the potentiation and depression events have equal probabilities across the network due to this balance . If the initial mean is smaller than the midpoint , a number of depression events would not be fully realized because they are likely to push the synaptic weights to , so equality of potentiation and depression is disrupted in favor the former , and the mean tends to increases . Similarly , if the initial mean is larger than the midpoint , some of the potentiation events push the corresponding weight above the maximum value and are truncated . This decreases the mean . Therefore , the mean tends to the midpoint in both cases , and the baseline firing rate ( which already has the tendency to equalize ) does not change this scenario . The simulation results confirm this prediction ( figure 3F ) . If the initial value of the weights are drawn from a uniform distribution , the mean will be at the midpoint from the very beginning , and it remains there throughout the simulation , making the final network firing rate equal to the initial rate ( figure 3E ) . Note that in figure 3E , and in similar figures to follow , we plot quantities as a function of the initial firing rate of the network , as opposed , for example , the external input used to modify this rate . We do this to make apparent changes in the network firing rate caused by STDP . In studies of a single neuron receiving Poisson input through synapses that are modified by STDP , it has been shown that stability requires depression to dominate over potentiation [6]–[9] . If potentiation dominates in this case , all the synaptic weights get potentiated to the maximum allowed value . Interestingly , for STDP within a network of neurons , potentiation dominated STDP can be stable because this instability is counteracted by the depression induced on reciprocal pairs of synapses . In other words , if one synapse between a reciprocally connected pair of neurons grows , the other synapse is likely to be weakened , preventing the outcome in which all the synapses are maximally potentiated . When the potentiation/depression balance is tipped in favor of potentiation ( and in our examples ) , the coefficient in equation ( 2 ) becomes larger than ( see Text S1 ) . In addition , the baseline parameter is positive . By setting the right-hand-sides of equations ( 2 ) to zero , the fixed point values of the two synaptic weights are found to be and . Both of these values are negative , so the fixed point lies outside the allowed range of synaptic strengths . Furthermore , this fixed point is unstable ( in both directions ) , which means that the weights tend to drift away from it ( figure 4; see Text S1 ) . We now examine the influence of the outlying , unstable fixed point on the dynamics within the allowed region of synaptic values when the baseline firing rates of the two neurons are equal . If the initial weights are fairly close to each other ( figure 4A , red area ) , they eventually end up at the attractor in the upper-right corner of the phase space due to repulsion from the outlying , unstable fixed point . This attractor corresponds to strong recurrent connections . Trajectories of weights that hit the upper boundary ( ) perpendicularly , form another fixed point that is unstable ( figure 4 , open circle on top ) . Trajectories to the left of this critical line are eventually absorbed by the attractor at the top-left corner ( corresponding to a unidirectional connection ) , while trajectories to its right are absorbed by the top-right attractor ( corresponding to recurrent connections ) . A similar unstable fixed point exists on the rightmost boundary ( ; figure 4 , open circle on the right ) . As a result , the state-space of the weights is partitioned into three basins of attraction: one leading to the attractor corresponding to recurrent connections ( figure 4 , red shading ) and the others to attractors that produce unidirectional connections ( yellow and green shadings ) . The appearance of the attractor corresponding to recurrent connections leads to a prediction about networks: STDP with dominant potentiation can generate loops in a network in contrast to balanced STDP . This prediction is confirmed by our numerical simulations showing that there are more loops induced in the steady-state weight matrix of a network in this case ( figures 5A ) . As the baseline firing rates of the neurons increase , the basin for the attractor with recurrent connections expands ( figure 4B , red area ) . This leads to the prediction that when a network is driven by stronger external input and consequently has a higher initial average firing rate , it will have more loops . Numerical simulation confirms this observation ( figure 5B ) . To quantify the degree of recurrence in a network , we define a “recurrence index” as the sum of the number of loops with less than synapses divided by the sum of similar loops in a shuffled version of the network ( see Methods ) . Simulation results show that the recurrence index increases as a function of the initial firing rate of the network and rises rather abruptly when the initial rate exceeds , and slightly decreases when the initial rate exceeds ( figure 5B ) . The eventual decrease of the recurrence index is due to the effect of correlations in the baseline firing , which appear in high rates and are not included in our analysis ( see figure S3 ) . The baseline correlations originate from the shared input that the neurons receive from the embedding network and is not related to their pairwise connectivity . Therefore it induces modifications to the reciprocal synapses regardless of the attractor structure explained here . The existence of the attractor corresponding to the recurrent connections also leads to the prediction that a network modified by potentiation dominant STDP settles into higher steady-state firing rates than a network with balanced STDP , starting from the same initial conditions . The simulations confirm this prediction as well ( figure 5C ) . The steady-state mean synaptic weight is expected to increase as a function of the initial firing rate because the basin of attraction corresponding to recurrent connections expands at high firing rates . The simulation results agree with this expectation up to the initial firing firing rate of , after which the steady-state mean decreases slightly ( figure 5D ) . As in the case of the recurrence index ( figure 5B ) this decrease is due to the baseline correlations that appear at high firing rates . If depression dominates over potentiation in STDP ( and in our examples ) , the coefficient in equations ( 2 ) is larger than ( see Text S1 ) , and the baseline parameter is negative . For these conditions , both elements of the fixed point of the weights , and , are negative , which is once again outside of the allowed range of synaptic values . In this case , however , the fixed point is a saddle node , which attracts trajectories from one direction and repels them from the other ( see Text S1 ) . As before , we consider two neurons with equal baseline firing rates . The weight trajectories tend to move toward the outlying fixed point in the direction that passes through the origin ( ; see figure 6A , arrows ) . This tendency makes the origin an attractor of the dynamics within the allowed range of synaptic weights . This attractor correspond to completely disconnected neurons . Because the outlying fixed point is a saddle node , the trajectories also tend to drift away from it in the direction perpendicular to the positive-slope diagonal . This tendency produces attractors corresponding to unidirectional connections ( figure 6 , top-left and bottom-right ) . Once again , trajectories that hit the borders perpendicularly partition the weight space into three basins of attractions corresponding to each attractor ( figure 6 ) . The dynamics of the synaptic pair we have considered suggests that some pairs of neurons in a network should become disconnected when depression dominates over potentiation . This is a more potent mechanism for eliminating loops than the previous cases , so we expect that STDP with dominant depression eliminates more loops in a large network than the other forms we have considered . Numerical simulations confirm that , indeed , there are fewer loops in the steady-state of a network with depression-dominated STDP compared to the previous cases ( compare figure 7A to figures 3A and 5A ) . In addition , the number of disconnected pairs is large , as predicted ( figure 7B ) . When the baseline rates of the two neurons increase , the basin of the attractor corresponding to disconnected pair becomes larger ( figure 6B ) . In a newtork , when neurons become excessively active , more connections should thus be eliminated , and the average rate should return to a lower value . Thus , the steady-state firing rate of a network with depression-dominated STDP should be lower than that of a network with balanced STDP starting from the same initial conditions . Simulation results corroborate this observation by showing that the steady-state firing rate of the network increases moderately as a function of the initial firing rate ( figure 7C , compare with figure 3E ) , so depression dominant STDP implements a partial buffering of steady-state firing rates . Finally , the mean synaptic weight is a decreasing function of the initial firing rate in this case ( figure 7D ) . The rightward shifted STDP model , in which nearly synchronous pre- and postsynaptic action potentials induce depression , has been shown to stabilize the distribution of the synaptic weights converging onto a single neuron . The rightward shift can arise from the finite rise time of activation of NMDA receptors [10] . Here , we study this model in the context of a network . The restriction of spike pairings that induce plasticity to those between nearest neighbor pre- and postsynaptic spikes , which is necessary in this case [10] , makes the dynamics of the pair of weights more complicated than in the previous cases , because the coefficients , and in equations ( 2 ) depend on the baseline firing rates ( see Text S1 ) . Furthermore , the coefficient can become negative at high firing rates , which makes the behavior of the system even more complicated . However , if we divide the analysis into three different rate regimes , we can elucidate the full range of behaviors . If depression dominates over potentiation in this model , the synaptic dynamics will be tantamount to the depression-dominant unshifted STDP described above , and the shift only makes depression even more dominant . Novel properties of this model only arise when potentiation dominates over depression , thus we assume that the potentiation domain is larger than the depression domain ( and as in [10] ) , and we set the amount of the shift to be . When the initial baseline firing rates of the two neurons are low , the coefficients , and are all positive . This is because the pairing intervals are not typically short enough to fall into the depression domain caused by the shift . In addition , the coefficient is slightly smaller than . This makes the fixed point for the weights positive and large , meaning that once again it falls out of the putative range of allowed synaptic weights , but this time on the positive not the negative side ( figure 8A ) . We use the term “putative” here because , as we will see , the upper limits on the synaptic weights are not actually required in this case . The fixed point is a saddle node ( see Text S1 ) and attracts the trajectories of weights along the direction toward the top-right corner of the state space ( figure 8A , arrows ) , which corresponds to recurrent connections . This case is qualitatively similar to what we found for STDP with dominant potentiation ( compare figures 8A and 4A , B ) , so the baseline firing rate tends to become higher than its initial value and eventually the dynamics of the system falls into the regime described by figure 8B . At higher baseline firing rates , the coefficient becomes negative . This occurs because the pairing intervals between presynaptic spikes and their causally induced postsynaptic spikes become short enough to fall into the depression domain caused by the shift . This creates a single stable fixed point for the two weights that lies within the putative allowed range of synaptic weights . Both weights are attracted to this fixed point , forming a recurrent connection ( figure 8B , arrows; see Text S1 ) , though not of maximal strength . If the two neurons start with even higher baseline rates , the coefficients and are both negative . This follows because at very high firing rates , even the intervals between randomly paired spikes of the baseline activity are short enough to fall into the depression domain caused by the shift . This pushes the fixed point of the weights out of the allowed range ( figure 8C ) but , in this case , on the negative side . Because this fixed point is stable , the weights tend to approach it , creating an attractor at the origin that eliminates both weights and disconnects the neurons . This mechanism prunes the weights in the embedding network until the baseline firing rate decreases enough to make the parameter positive . Then , the regime with a stable fixed point within the allowed range ( figure 8B ) is restored . This is why no upper bounds on the synaptic weights are required in this case . Combining these effects , we find that , if the shift is larger than a critical value ( in this case , see figure S1 ) , a network will settle into a regime with a single stable fixed point within a narrow range of steady-state firing rates , regardless of the initial firing rate or the strength of the external input . The condition for this scenario to occur is that the fixed point of the weights becomes stable before it grows negative , as the initial firing rate increases . This happens when the potentiation domain is larger than depression domain and the shift is sufficiently large . The calculations show that a shift of fulfills this condition for our chosen values of potentiation and depression magnitudes ( see figure S1 ) . By generalizing from the dynamics of a pair of synapses , two predictions can be made . First , the steady-state matrix of synaptic weights should have many recurrent connections because there is no mechanism to eliminate loops , and reciprocal connections should tend to be strengthened . This prediction is confirmed by numerical simulations that show a highly recurrent steady-state connectivity ( figure 9A ) . Second , because the pairwise connections settle into a regime with a single stable fixed point regardless of the initial baseline rate , the steady-state firing rate of the network should be resilient to changes in the external input or in the initial firing rate . Numerical simulations show that the steady-state firing rate of the network varies very slightly as a function of the initial firing rate ( figure 9B ) . Interestingly , the narrow range of the steady-state firing rates agrees precisely with the prediction of the pairwise analysis ( dashed lines in figures 9B and figure S1 ) . Thus , rightward shifted STDP implements a homeostatic mechanism that strongly buffers the steady-state firing rates from external influences . Finally , the mean synaptic weight decreases with increased initial firing rate in this case ( figure 9C ) . A leftward shifted STDP model , in which synchronous pre- and postsynaptic spikes cause potentiation as a result of axonal conduction delays , has been shown to have a desynchronizing effect on population bursts and a synchronizing effect on random spiking in a recurrent network [14] . Here , we study this model within the framework of pairwise analysis . As in the previous section , we consider the interactions of nearest-neighboring spikes . If potentiation dominates over depression in this model , the synaptic dynamics will be tantamount to the potentiation-dominant unshifted STDP described above and the shift only makes potentiation further dominant . Therefore , in order to observe novel behaviors of this model , we assume that the depression domain is larger than potentiation domain ( and ) , and we set the amount of the shift to be , i . e . the parameters are chose to be the flipped versions of those in the rightward shifted model above . When the initial baseline firing rate is low , the coefficients and are positive ( ) and is negative . This is because the pairing intervals are not typically short enough to fall into the potentiation domain caused by the shift . As a result the fixed point is positive , unstable in both directions , and out of the allowed range of weights ( figure 10A ) . The weight trajectories tend to drift away from the fixed point in the direction that passes through the origin , so this behavior is qualitatively similar to what we found for STDP with dominant depression . The attractor at the origin corresponds to completely disconnected neurons , therefore the baseline firing rate tends to become less than its initial value . For higher initial baseline firing rates , coefficient becomes negative , because the pairing intervals between pre- and postsynaptic spikes become short enough to fall into the potentiation domain caused by the shift . This turns the fixed point into a saddle node and pushes it into the allowed range of weights ( figure 10B ) . The weights drift away from the fixed point in the directions that passes through the origin and the top-right corner , and are attracted to it in the perpendicular direction . As a result , both the origin and top-right corner turn into attractors , corresponding to disconnected and recurrently connected neurons respectively ( figure 10B , closed circles ) . Because these two points are the only attractors of the system , the network is expected to become highly recurrent in this case and the neurons to become either recurrently connected or disconnected . This regime happens for a narrow range of initial firing rates . As the initial firing rate increases , the basin of the top-right attractor becomes larger ( figure 10C ) . As a result , more recurrent connections form and hence the baseline firing rate increases , which eventually pushes the system into the regime described in the following paragraph . For even higher initial baseline firing rates , not only coefficient becomes negative , but also turns positive and the fixed point is pushed out of the allowed range on the negative side ( figure 10D ) . It remains a saddle node , so the weights are repelled from it in the direction that passes through the top-right corner , which becomes the only attractor of the system corresponding to recurrent connection . Therefore , it is expected that all the synapse in the network potentiate up to the upper limit of the weights in this case . In summary , the above description shows that as the initial baseline firing rate increases , the networks undergoes three different phase: 1 ) for low initial rates it behaves similarly to depression-dominant STDP , i . e . recurrent connections are eliminated and the steady-state firing rate is partially buffered; 2 ) for higher initial rates the network becomes highly recurrent and the steady-state rate increases; 3 ) for even higher initial rates , all the weights become potentiated up to the maximum , and the firing rate is pathologically high . The simulation results confirm these predictions . When the initial rate is less than , the steady-state rate increases modestly ( figure 11B , left ) and the the mean of synaptic weights decreases ( figure 11C , left ) as a function of initial rate . The number of loops also decrease in this regime ( figure 11A , blue ) . For higher initial rates , the mean synaptic weight and the steady-state rate increase rapidly ( figures 11B-C , middle ) and the network is highly recurrent ( figure 11A , red ) . For initial rates higher than , the mean synaptic weight equals the maximum allowed value , implying that all the weights are maximally potentiated ( figure 11C , right ) , and the steady-state rate is pathologically high . Although the simulation results qualitatively show the full range of behaviors predicted by pairwise analysis , the initial firing rate at which the transitions occur in simulations is lower than that predicted from calculations ( see Text S1 ) . This discrepancy is due to baseline correlations that appear at high rates ( see figure S3 ) . In presence of baseline correlations , the neurons tend to fire synchronously regardless of their pairwise connections , and hence the synapses get potentiated indiscriminately due to leftward shift of the STDP . By analyzing pairwise interactions of neurons affected by STDP , we clarified how conventional pair-based STDP functions as a loop-eliminating mechanism in a network of spiking neurons and organizes neurons into in- and out-hubs , as reported in [21] . Loop-elimination increases when depression dominates , and turns to loop generation when potentiation dominates . STDP with dominant depression implements a partial buffering mechanism for network firing rates . Rightward shifted STDP can generate recurrent connections in a network and functions as a strict buffering mechanism to maintain a roughly constant network firing rate . STDP with leftward shift functions as a partial buffer of firing rates and a loop eliminator at low rates , and as a potent loop generator at higher rates . All of our analytical results were obtained by considering the effect of imposing weight constraints on a linear system describing pairwise interactions of neurons in the presence of STDP . The effect of constraints on Hebbian plasticity has been explored before to explain the formation of visual receptive fields [26] . Our work can be viewed as an extension of this approach to a specific form of Hebbian plasticity that involves the timing of spikes , namely STDP . In the context of a recurrent network , this method can predict the outcome of STDP in shaping the connectivity of the network and qualitatively captures the direction of change of firing rates in the network . However , the steady-state firing rate of the network cannot be quantitatively calculated by this approach , since the analysis is focused on the snapshots of the weight dynamics given the current firing rates . The network used in our numerical simulations was densely connected so that every neuron could potentially form a synaptic connection to every other one . However , our analytical results does not rely on any particular assumption about the density or sparsity of network connectivity . Instead , the results indicate that STDP can organize patterns of connectivity in particular ways within the framework provided by anatomical constraints , developmental hard-wiring and other physiological mechanisms , such as other forms of plasticity . In a series of articles , Gilson and colleagues studied the structures that arise from STDP in a recurrent network in response to the patterns of correlations in the external input [15]–[20] . Here , we took a different approach . We focused on the network structures that arise in the absence of correlations either imposed by external input or originated from common inputs within the network , inspired by the observation that these are dramatically reduced by fast and strong recurrent inhibition [25] . Instead , we systematically studied the effect of the shape of the STDP window on the structures that arise in this decorrelated state . Our results can be viewed as a basis over which any structures induced by external correlations will be mounted . A prominent feature of STDP is its ability to organize neurons into in- and out-hubs . The dependence of hub-formation on baseline firing rate shows how heterogeneity at the level of external inputs can influence the internal structure of a neural network . Moreover , this property of STDP can play an important protective role in pathological cases in which a sub-population of excitatory neurons fires at atypically high rates . Through STDP , most of the incoming synapses to this sub-population will be weakened to mitigate the excessive high firing rate . Decoupling of a highly active sub-population from an embedding network through STDP has been observed previously in networks with an excitation-inhibition balance [13] . A related study based on simulations of a small network showed that depending on the external input , an STDP rule that is phenomenologically similar to the triplet model [27] can either induce feedforward structures or recurrent connections , which was argued to be incompatible with simple pair-based STDP [22] . Although our study only addressed structures arising from pair-based STDP , our results show that recurrent connections can arise if potentiation dominates depression or the plasticity window is shifted . Interestingly , the dependence of the structures on external input in the case of leftward shifted STDP is similar to that of the more elaborate model studied by Clopath and colleagues [22] . A number of studies indicate that , apart from the timing of spikes , several other factors including firing rates , inhibitory inputs , dendritic spikes and neuromodulation influence plasticity induction [27]–[30] . Various STDP rules ( including the multi-spike STDP models reviewed in [31] ) have been proposed to incorporate some of these factors . The method we have developed can be used with these other STDP models , but we did not include an analysis of multi-spike STDP or more complex models because we did not want an excessive number of examples nor complexity in the STDP rule to obscure the basic approach and the insights that it provides . The ability of pair-based STDP to generate recurrent connections has been shown previously [11] . Although in that case the depression domain was elongated , but the magnitude of potentiation domain was larger such that overall potentiation was dominant over depression , which agrees with our results on loop generation through STDP . Lubenov and colleagues have shown that STDP with leftward shifted window , arising from axonal conduction delays , can generate recurrent connections and thereby synchronize neurons when the network is initialized with a tonic irregular firing mode . In the bursting mode , leftward shifted STDP has the opposite effect , i . e . it eliminates loops and desynchronizes the neurons [14] . Because the networks we studied were in excitatory/inhibitory balanced state in which the firing patterns are irregular and asynchronous [28]–[30] , our findings about loop generation through leftward shifted STDP agree with the results of [14] , even though the same model can function as a loop eliminating mechanism at low initial firing rates . A combination of axonal , synaptic and dendritic propagation delays can induce a leftward shift in STDP window [14] . On the other hand , the finite rise time of the NMDA receptor activation can give rise to a rightward shift in the window [10] . Thus the exact magnitude and direction of the shift depends on the relative contribution of these opposing factors . For instance , because the back-propagating postsynaptic spikes arrive at distal synapses with a longer delay than at proximal ones , leftward shifted STDP is expected to be observed more in the distal dendrites and rightward shift is expected at proximal sites . Moreover , the relative magnitude of potentiation and depression varies considerably along the dendritic tree [31]–[34] . Therefore , each of the different versions of STDP window analyzed in our study may be relevant in a particular region of the dendritic tree . A general prediction of our study is then that different regions of the dendritic tree may participate in different network structures as a result of differences in their STDP windows . In a number of studies , clusters of three or four synaptically connected neurons have been observed in cortical slices at a higher frequency than expected from a random or distance-based connectivity pattern [35] , [36] . We doubt that STDP can account for these clusters unless network synapses were unrealistically strong , so strong that the causal effect of single spikes from one neuron can pass through two or more synapses and transiently increase the firing rate of another neuron . Otherwise the effects of STDP would be restricted to mono-synaptically connected neurons , even in larger ensembles . In fact , our results show that loops of length 3 are usually the loops least affected by STDP . This can be explained by the direct effect of STDP being confined to loops of length 2 . In loops of length 3 , unlike longer loops , there is no contribution from reciprocally connected pairs of neurons ( loops of length 2 ) . In conclusion , studying pairwise interactions of neurons through STDP provides a number of important insights about the structures that arise from this plasticity in large networks . This approach can be extended to networks with more complex STDP models and more structured external input . We used leaky integrate-and-fire ( LIF ) model neuron in our simulations . The membrane potential of the LIF neuron obeys ( 3 ) where is the membrane time constant , is the resting potential , and is the synaptic input ( see below ) . Although the input appears as a current , it is actually measured in units of the membrane potential ( mV ) because a factor of the membrane resistance has been absorbed into its definition . When the membrane potential reaches the firing threshold , the neuron fires an action potential and the membrane potential resets to the resting value . A network of excitatory and inhibitory LIF neurons was simulated . Each neuron receives excitatory and inhibitory inputs from all the other neurons in the network . The strengths of the excitatory-to-inhibitory , inhibitory-to-excitatory and inhibitory-to-inhibitory synapses are fixed . At the beginning of each simulation , their strengths were drawn from uniform distributions ranging from 0 to , , and respectively . The excitatory-to-excitatory connections are modified by pair-based STDP as described below . They are also initialized at the beginning of each run to random values from a uniform distribution ranging between and . Although the inhibitory connections are stronger than excitatory connections ( but inhibitory-to-excitatory and inhibitory-to-inhibitory connections are equally strong ) , the network settles into an excitation/inhibition balanced state with these initial conditions ( see figure S3 ) . In this state , individual neurons fire irregularly and asynchronously [28]–[30] and the strong recurrent inhibition causes the firing correlations due to shared input to be very week [25] . The connections are all to all and self connections are prohibited for all neurons . Each presynaptic action potential arriving at an excitatory or inhibitory synapse induces an instantaneous jump or fall in synaptic input respectively , by an amount proportional to the appropriate synaptic weight . The input decays exponentially between presynaptic action potentials . In addition to synaptic inputs originating from the neurons within the network , the input to each neuron includes an external constant bias term and independent white noise . Taken together , the input to the excitatory or inhibitory neuron is described by ( 4 ) Here , the synaptic time constant is , denotes the full matrix of connections ( , , and ) and the first sum runs over all neurons ( and for excitatory and inhibitory populations , respectively ) . The second sum runs over all the times of spikes produced by neuron prior to time , indexed by . The parameters and determine the mean and the variability of the input ( has not subscript because it is the same for all neurons ) , and satisfies and , with the brackets denoting averages . The parameter was set to to provide an average initial baseline firing rate of for the neurons in the network when is zero . In the simulations , the value of was changed systematically to modify initial firing rates . Each simulation is run until the excitatory-to-excitatory connections reached a steady-state in which the average firing rate , and the mean and variance of the synaptic weights remained constant . To count the number of closed loops implied by the matrix of excitatory-to-excitatory synaptic weights ( ) , we first turn the network into a directed graph [21] . This is done by comparing each synaptic weight to a threshold value , and assigning the value 1 to the synapse if its weight is greater than or equal to , and assigning a zero otherwise . This defines the adjacency matrix of the resultant directed graph , which can be written formally as ( 5 ) where is the Heaviside step function . The number of closed loops of length in the adjacency matrix is then ( 6 ) where denotes the matrix trace ( the sum of the diagonal elements ) . To evaluate the degree of recurrence in a network , we compare the number of closed loops obtained from the above equation with the number in a randomly permuted ( shuffled ) version of the same matrix . This distinguishes recurrent connections formed by chance from those that arise from plasticity . In the following sections , whenever we mention the number of loops in a network , we are in fact referring to the number of loops in the adjacency matrix formed by turning the network into a directed graph as described above . In addition , when we refer to a “number” of synapses , we really refer to the number of synapses with strengths greater than the threshold . To obtain a loop count that is not biased by the overall strengths of the weights , we chose to be equal to the mean of the excitatory synaptic weights . For figures 3 , 5 , 7 , 9 , 11 ( initial rate ) and 11 ( initial rate ) , respectively , was set to .	The connectivity structure in neural networks reflects , at least in part , the long-term effects of synaptic plasticity mechanisms that underlie learning and memory . In one of the most widespread such mechanisms , spike-timing dependent plasticity ( STDP ) , the temporal order of pre- and postsynaptic spiking across a synapse determines whether it is strengthened or weakened . Therefore , the synapses are modified solely based on local information through STDP . However , STDP can give rise to a variety of global connectivity structures in an interconnected neural network . Here , we provide an analytical framework that can predict the global structures that arise from STDP in such a network . The analytical technique we develop is actually quite simple , and involves the study of two interconnected neurons receiving inputs from their surrounding network . Following analytical calculations for a variety of different STDP models , we test and verify all our predictions through full network simulations . More importantly , the developed analytical tool will allow other researchers to figure out what arises from any other type of STDP in a network .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "circuit", "models", "computational", "neuroscience", "biology", "neuroscience" ]	2013	Pairwise Analysis Can Account for Network Structures Arising from Spike-Timing Dependent Plasticity
The inositol trisphosphate receptor ( ) is one of the most important cellular components responsible for oscillations in the cytoplasmic calcium concentration . Over the past decade , two major questions about the have arisen . Firstly , how best should the be modeled ? In other words , what fundamental properties of the allow it to perform its function , and what are their quantitative properties ? Secondly , although calcium oscillations are caused by the stochastic opening and closing of small numbers of , is it possible for a deterministic model to be a reliable predictor of calcium behavior ? Here , we answer these two questions , using airway smooth muscle cells ( ASMC ) as a specific example . Firstly , we show that periodic calcium waves in ASMC , as well as the statistics of calcium puffs in other cell types , can be quantitatively reproduced by a two-state model of the , and thus the behavior of the is essentially determined by its modal structure . The structure within each mode is irrelevant for function . Secondly , we show that , although calcium waves in ASMC are generated by a stochastic mechanism , stochasticity is not essential for a qualitative prediction of how oscillation frequency depends on model parameters , and thus deterministic models demonstrate the same level of predictive capability as do stochastic models . We conclude that , firstly , calcium dynamics can be accurately modeled using simplified models , and , secondly , to obtain qualitative predictions of how oscillation frequency depends on parameters it is sufficient to use a deterministic model . Oscillations in cytoplasmic calcium concentration ( ) , mediated by inositol trisphosphate receptors ( ; a calcium channel that releases calcium ions ( ) from the endoplasmic or sarcoplasmic reticulum ( ER or SR ) in the presence of inositol trisphosphate ( ) ) play an important role in cellular function in many cell types . Hence , a thorough knowledge of the behavior of the is a necessary prerequisite for an understanding of intracellular oscillations and waves . Mathematical and computational models of the play a vital role in studies of dynamics . However , over the past decade , two major questions about models have arisen . Firstly , how best should the be modeled ? Models of the have a long history , beginning with the heuristic models of [1]–[3] . With the recent appearance of single-channel data from in vivo [4] , [5] , a new generation of Markov models has recently appeared [6] , [7] . These models show that exist in different modes with different open probabilities . Within each mode there are multiple states , some open , some closed . Importantly , it was found [8] that time-dependent transitions between different modes are crucial for reproducing puff data from [9] . However , it is not yet clear whether transitions between states within each mode are important , or whether all the important behaviors are captured simply by inter-mode transitions . Secondly , why do deterministic models of the perform so well as predictive models ? Deterministic models of the have proven to be useful predictive models in a range of cell types . For example , -based models have been developed to study oscillations in airway smooth muscle cells ( ASMC ) [10]–[13] , and these models have made predictions which have been confirmed experimentally . This shows the usefulness of such models in advancing our understanding of how intracellular oscillations and waves are initiated and controlled in ASMC . However , these models are deterministic models which assume infinitely many per unit cell volume , an assumption that contradicts experimental findings in many cell types showing that puffs and spikes occur stochastically , and that intracellular waves and oscillations arise as an emergent property of fundamental stochastic events [9] , [14] , [15] . Here , we answer these two fundamental modeling questions using data and models from ASMC . Firstly , we show that a simple model of the , involving only two states with time-dependent transitions , suffices to generate correct dynamics of puffs and oscillations . Secondly , we show that , although oscillations in ASMC are generated by a stochastic mechanism , a deterministic model can make the same qualitative predictions as the analogous stochastic model , indicating that deterministic models , that require much less computational time and complexity , can be used to make reliable predictions . Although we work in the specific context of ASMC , our results are applicable to other cell types that exhibit similar oscillations and waves . We have previously shown [8] that the statistics of puffs in SH-SY5Y cells can be reproduced by a Markov model of the based on the steady-state data of [5] and the time-dependent data of [4] . In this model the can exist in 6 different states , grouped into two modes , which we call Drive and Park ( see Fig . 1 ) . The Drive mode ( which contains 4 states; 1 open and 3 closed ) has an average open probability of around 0 . 7 , while the Park mode ( which contains the remaining two states; 1 open and 1 closed ) has an open probability close to zero . Transitions between states within each mode are independent of and ; only the transitions between modes are ligand-dependent . In our previous study on calcium puffs [8] , we showed that , to reproduce the experimentally observed non-exponential interspike interval ( ISI ) distribution and coefficient of variation ( CV ) of ISI smaller than 1 , the time-dependent intermodal transitions are crucial . Lack of time dependencies in the Siekmann model leads to exponential ISI distributions and CV = 1 , which is not the case for calcium spikes in ASMC . Fig . 2A shows an example of oscillations generated by 50 nM methacholine ( MCh , an agonist that can induce the production of by binding to a G protein-coupled receptor in the cell membrane ) in ASMC . By gathering data from 14 cells in 5 mouse lung slices , we found that the standard deviation of the interspike interval ( ISI ) is approximately a linear function of the ISI mean , with a slope clearly between 0 and 1 ( i . e . ) , indicating that the spikes are generated by an inhomogeneous Poisson process ( a slope of 1 would denote a pure Poisson process ) ( see Fig . 2B ) . This shows the necessity of inclusion of time-dependent transitions for mode-switching . Using a quasi-steady-state approximation , and ignoring states with very low dwell times , it is possible to construct a simplified two-state version of the full six-state model ( see Materials and Methods ) . In the simplified model the intramodal structure is ignored , and only the intermodal transitions have an effect on behavior . In Fig . 3 we compared the simplified model to the full six-state model . Both models have the same distribution of interspike interval , spike amplitude and spike duration . Moreover , by looking at a more detailed comparison between the two model results ( Figs . 4A , C and E ) and experimental data ( Figs . 4B , D and F ) , we found the 2-state model not only can reproduce the behaviour of the 6-state model , but can also qualitatively reproduce experimental data . The average experimental ISI shows a clear decreasing trend as MCh concentration increases ( although a saturation occurs in the data for high MCh ) , a trend that is mirrored by the model results as the concentration increases . Unfortunately , since the exact relationship between MCh concentration and concentration is uncertain , a quantitative comparison is not possible . In both model and experimental results , the average peak and duration of the oscillations are nearly independent of agonist concentration . The quantitative difference in spike duration between the model results and the data in Figs . 4E and F are most likely due to choice of calcium buffering parameters . For example , adding fast buffer ( see Materials and Methods ) increases the average spike duration to 0 . 54 s or 0 . 7 s respectively , which are close to the levels shown in the data . Thus , the intramodal structure of the six-state model is essentially unimportant , as the model behavior ( in terms of the statistics of puffs and oscillations ) is governed almost entirely by the time dependence of the intermode transitions , particularly the time dependence of the rapid inhibition of the by high , and the slow recovery from inhibition by . The multiple states within each mode are necessary to obtain an acceptable quantitative fit to single-channel data , but are nevertheless of limited importance for function . Hence , even when simulating microscopic events such as puffs it is sufficient to use a simpler , faster , two-state model , rather than a more complex six-state model . In the following , we will use the 2-state model to generate all the simulation results . Although the data ( Fig . 2 ) show that oscillations in ASMC are generated by a stochastic process , not a deterministic one , we wish to know to what extent a deterministic model can be used to make qualitative ( and experimentally testable ) predictions . Our simplified 2-state Markov model of the can be converted to a deterministic model ( see Materials and Methods ) . The result is a system of ordinary differential equations ( ODEs ) with four variables , which takes into account the increased at an open pore , as well as the increased within a cluster of ; the four variables are the outside the cluster ( ) , the within the cluster ( ) , the total intracellular concentration ( ) and an gating variable ( ) . We refer to the reduced 4D model as the deterministic model for all the results and analyses . Note that there is no physical or geometric constraint enforcing a high local ; in this case the spatial heterogeneity arises solely from the low diffusion coefficient of . Our use of is merely a highly simplified way of introducing spatial heterogeneity of the concentration . Since the can only “see” ( as well as the concentration right at the mouth of an open channel , which we denote by ) , but cannot be influenced directly by ( the experimentally observed signal ) , our approach allows for the functional differentiation of the rapid local oscillatory in the cluster , from the slower signal in the cytoplasm , without the need for computationally intensive simulations of a partial differential equation model . Quantitative accuracy is thus sacrificed for computational convenience . Calcium oscillations in the stochastic and deterministic models are shown in Fig . 5A . According to our previous results [8] , the average value of over the cluster of primarily regulates the termination and regeneration of individual spikes . This can be seen in the stochastic model by projecting the solution on the phase plane ( Fig . 5B ) . Upon an initial release from one or more , a large spike is generated by Ca2+-induced release ( via the ) during which time a decreasing gradually decreases the average open probability of the clustered . The spike is terminated when is too small to allow further release . This phenomenon is qualitatively reproduced by the deterministic model ( Fig . 5D ) . In both the stochastic and deterministic models the decrease in average open probability of a cluster of caused by inhibition is the main reason for the termination of each spike . According to Figs . 5B and D , regeneration of each spike requires a return of back to a relatively high value ( i . e . , recovery of the from inhibition by ) . The deterministic model sets a clear threshold for the regeneration , as can be seen in Fig . 5C , where an upstroke in occurs when the trajectory creeps beyond the sharp “knee” of the white curve . When the trajectory reaches the knees of the white curve it is forced to jump across to the other stable branch of the critical manifold , resulting in a fast increase in followed by a relatively fast increase in ( seen by combining Figs . 5C and D ) . In contrast , the stochastic model enlarges the contributions of individual so that the generation of each spike is also effectively driven by random release through the , which can be seen in the inset of Fig . 5B where the site of spike initiation ( blue bar ) exhibits significantly greater variation than that of spike termination ( green bar ) . In spite of this , the essential similarities in phase plane behavior result in both deterministic and stochastic models making the same qualitative predictions in response to perturbations , such as changes in concentration ( ) , influx or efflux . In the following , we illustrate this by investigating a number of experimentally testable predictions . Due to the extensive importance of frequency encoding in many -dependent processes , we focus particularly on the change of oscillation frequency in response to parameter perturbations . As a side issue we also investigate how the oscillation baseline depends on physiologically important parameters . In many cell types a moderate increase in increases the oscillation frequency ( see Fig . 2A in [11] , Fig . 4E in [16] and Fig . 6B in [17] ) , a result that is reproduced by both model types ( Fig . 6A ) . As increases , the stochastic model increases the probability of the initial release through the first open and of the following release , thus shortening the average ISI . Although the oscillatory region of the deterministic model is strictly confined by bifurcations which do not apply to the stochastic model , the deterministic model can successfully replicate an increasing frequency by lowering the “knee” of the red curve in Fig . 5D and shortening the time spent from the termination point c to the initiation point a ( thus shortening the ISI ) . Hence , although the deterministic model cannot be used to predict the exact values of at which the oscillations begin and end , as stochastic effects predominate in these regions , it can be used to predict the correct qualitative trend in oscillation frequency . In many cell types , including ASMC , transmembrane fluxes modulate the total intracellular load ( ) on a slow time scale [16] , [18] , and thereby modulate the oscillation frequency [19] . Experimental data can be seen in Fig . 8 in [16] and Fig . 2 in [18] . Figs . 6B and C show that both stochastic and deterministic models predict the same qualitative changes in oscillation frequency in response to changes in membrane fluxes ( through membrane ATPase pumps and/or influx channels such as receptor-operated channels or store-operated channels ) . The level of sarco/endoplasmic reticulum calcium ATPase ( SERCA ) expression ( or capacity ) is important for airway remodeling in asthma [20] and ASMC oscillations [21] . We thus investigated the predictions of the two models in response to changes in SERCA expression ( ) . As decreases , the deterministic model exhibits a decreasing frequency , in agreement with experimental data ( see Figs . 3 and 4 in [21] ) . The same trend is seen in the stochastic model with only 20 ( see Fig . 6D ) . Calcium buffers have been shown to be able to change the ISI and spike duration , which in turn change the oscillation frequency [15] , [22] . We compared the effects on the two models of varying total buffer concentration ( ) by adding one buffer with relatively fast kinetics to the models ( see Materials and Methods for details ) . In both models the frequency decreases as increases ( see Fig . 6E ) , which is consistent with experimental data ( Fig . 2B in [18] ) . This is not surprising , because increasing can decrease the effective rates of SR release and reuptake . Sustained elevations of baseline during agonist-induced oscillations or transients have been observed experimentally , and are believed to be a result of an increase in influx caused by opening of membrane channels [13] , [16] . Furthermore , there is evidence showing that decreased SERCA expression could also increase the baseline ( Fig . 4 in [21] ) . Those phenomena are successfully reproduced by both models ( see Fig . 7 ) . In this paper we address two current major questions in the field of modeling . Firstly , we show that puffs and stochastic oscillations can be reproduced quantitatively by an extremely simple model , consisting only of two states ( one open , one closed ) , with time-dependent transitions between them . This model is obtained by removing the intramodal structure of a more complex model that was determined by fitting a Markov model to single-channel data [7] . We thus show that the internal structure of each mode is irrelevant for function and mode switching is the key mechanism for the control of calcium release . The necessity for time-dependent mode switching is shown not only by the dynamic single-channel data of [4] ) , but also by the puff data of [9] and our ASMC data . Secondly , we investigate the role of stochasticity of in modeling oscillations in ASMC by comparing a stochastic IP3R-based model and its associated deterministic version , for parameters such that both of the models exhibit spikes but the stochastic model cannot necessarily be replaced by a mean-field model . We find that a four-variable deterministic model has the same predictive power as the stochastic model , in that it correctly reproduces the process of spike termination and predicts the same qualitative changes in oscillation frequency and baseline in response to a variety of perturbations that are commonly used experimentally . The mechanism for termination of individual spikes is fundamentally a deterministic process controlled by a rapid inhibition induced by the high local in the cluster , whereas spike initiation is significantly affected by stochastic opening of . Hence , repetitive cycling is primarily induced by the time-dependent gating variables governing transitions of the from one mode to another . Our simplified two-state model of the is identical in structure ( although not in parameter values ) to the well-known model of [23] . It is somewhat ironic that after 20 years of detailed studies of the and the construction of a plethora of models of varying complexity , the single-channel data have led us around full circle , back to these original formulations . Excitability is arising via a fast activation followed by a slower inactivation , a combination often seen in physiological processes [24] . Encoding of this fundamental combination results directly from the two-mode structure of the . Although similar single-channel data have been used to construct three-mode models [6] , [25] , neither of these models has yet been used in detailed studies of puffs and waves , and it remains unclear whether or not they have a similar underlying structure . In contrast to previous deterministic ODE models , our four-variable model includes a more accurate model , as well as local control of clustered by two distinct microdomains; one at the mouth of an open , the other inside a cluster of . Neglect of either of these microdomains leads to models that either exhibit unphysiological cytoplasmic concentrations or fail to reproduce reasonable oscillations . This underlines the importance of taking microdomains into consideration when constructing any model . Our microdomain model is highly simplified , with the microdomain being treated simply as a well-mixed compartment . More detailed modeling of spatially-dependent microdomains is possible , and not difficult in principle , but requires far greater computational resources . It is undeniable that a more detailed model , incorporating the full spatial complexity – and possibly stochastic aspects as well – would make , overall , a better predictive tool . However , our goal is to find the simplest models that can be used as predictive tools . An important similar study is that of Shuai and Jung [26] . They compared the use of Markov and Langevin approaches to the computation of puff amplitude distributions , compared their results with the deterministic limit , and showed that stochasticity does not qualitatively change the type of puff amplitude distribution except for when there are fewer than 10 . Here , we significantly extend the scope of their study by exploring the effects of stochasticity on the dynamics of spikes , and we do this in the context of an model that has been fitted to single-channel data . Although this is true in a general sense for the Li-Rinzel model , which is based on the DeYoung-Keizer model , which did take into account the opening time distributions of in lipid bilayers , neither model can reproduce the more recent data obtained from on-nuclei patch clamping . When these recent data are taken into account one obtains a model with the same structure , but quite different parameters and behavior . We find that , in spite of a relatively large variation in spike amplitude which is partially caused by a large variation in ISI ( Fig . 5B ) , the mechanism governing individual spike terminations is the same for both a few or infinitely many , which explains why the one-peak type of amplitude distribution is independent of the choice of number ( see Fig . 6A in [26] ) . Another important relevant study was done by Dupont et al . [27] , who compared the regularity of stochastic oscillations in hepatocytes for different numbers of clusters . They found that the impact of stochasticity on global oscillations ( in terms of CV ) increases as the total cluster number decreases . Our study here extends these results , and demonstrates how well stochastic oscillations can be qualitatively described by a deterministic system , even when there is only a small number of ( which appears to be the case for ASMC , in which the wave initiation site is only in diameter ) . Indeed , as we have shown , for the purposes of predictive modeling a simple deterministic model does as well as more complex stochastic simulations . Ryanodine receptors ( RyR ) are another important component modulating ASMC oscillations [16] , [28] , [29] but are not included in our model . This is because the role of RyR is not fully understood and may be species-dependent; for example , in mouse or human ASMC , RyR play very little role in -induced continuing oscillations [17] , [30] , but this appears not to be true for pigs [28] . Our study focuses on the calcium oscillations in mouse and human ( as we did in our experiments ) where inclusion of a deterministic model of RyR should have little effect . An understanding of the role of RyR stochasticity and how the and the RyR interact needs a reliable RyR Markov model , exclusive to ASMC , which is not currently available . Multiple Markov models of the RyR have been developed for use in cardiac cells [31] , but these are based on single-channel data from lipid bilayers , and are adapted for the specific context of cardiac cells . Their applicability to ASMC remains unclear . Although we have not shown that the deterministic model for ASMC has the same predictive power as the stochastic model in all possible cases ( which would hardly be possible in the absence of an analytical proof ) the underlying similarity in phase plane structure indicates that such similarity is plausible at least . Certainly , we have not found any counterexample to this claim . However , whether or not this claim is true for all cell types is unclear . Some cell types exhibit both local puffs and global spikes ( usually propagating throughout the cells in the form of traveling waves ) , showing that initiation of such spikes requires a synchronization of release from more than one cluster of [14] . This type of spiking relies on the hierarchical organization of signal pathways , in particular the stochastic recruitment of both individual and puffs at different levels [32] , and therefore cannot be simply reproduced by deterministic models containing only a few ODEs . However , oscillations in ASMC , as observed in lung slices , may not be of this type , as IP3R-dependent puffs have not been seen in these ASMC . It thus appears that , in ASMC in lung slices , every “puff” initiates a wave , resulting in periodic waves with ISI that are governed by the dynamics of individual puffs . Animal experimentations carried out were approved by the Animal Care and Use Committee of the University of Massachusetts Medical School under approval number A-836-12 . BALB/c mice ( 7–10 weeks old , Charles River Breeding Labs , Needham , MA ) were euthanized via intraperitoneal injection of 0 . 3 ml sodium pentabarbitone ( Oak Pharmaceuticals , Lake Forest , IL ) . After removal of the chest wall , lungs were inflated with of 1 . 8% warm agarose in sHBSS via an intratracheal catheter . Subsequently , air ( ) was injected to push the agarose within the airways into the alveoli . The agarose was polymerized by cooling to . A vibratome ( VF-300 , Precisionary Instruments , San Jose , CA ) was used to make thick slices which were maintained in Dulbecco's Modified Eagle's Media ( DMEM , Invitrogen , Carlsbad , CA ) at in /air . All experiments were conducted at in a custom-made temperature-controlled Plexiglas chamber as described in [17] . Lung slices were incubated in sHBSS containing Oregon Green 488 BAPTA-1-AM ( Invitrogen ) , a Ca2+-indicator dye , 0 . 1% Pluronic F-127 ( Invitrogen ) and sulfobromophthalein ( Sigma Aldrich , St Louis , MO ) in the dark at for 1 hour . Subsequently , the slices were incubated in sulfobromophthalein for 30 minutes . Slices were mounted on a cover-glass and held down with mesh . A smaller cover-glass was placed on top of the mesh and sealed at the sides with silicone grease to facilitate solution exchange . Slices were examined with a custom-built 2-photon scanning laser microscope with a oil immersion objective lens and images recorded at 30 images per second using Videosavant 4 . 0 software ( IO Industries , Montreal , Canada ) . Changes in fluorescence intensity ( which represent changes in ) were analyzed in an ASMC of interest by averaging the grey value of a pixel region using custom written software . Relative fluorescence intensity ( ) was expressed as a ratio of the fluorescence intensity at a particular time ( F ) normalized to the initial fluorescence intensity ( ) . Inhomogeneity of cytoplasmic concentration not only exists around individual channel pores of the , where a nearly instantaneous high concentration at the pore ( denoted by ) leads to a very sharp concentration profile , but is also seen inside an cluster where the average cluster concentration ( ) is apparently higher than that of the surrounding cytoplasm ( ) [33] . This indicates that during oscillations each is controlled by either the pore concentration ( when it is open ) or the cluster concentration ( when it is closed ) . Neither of these local concentrations influence cell membrane fluxes or the majority of SERCAs , which we assume to be distributed outside the cluster . The scale separation between the pore concentration and the cluster concentration allows to treat as a parameter , providing a simpler way of modeling local events ( like puffs ) that has been used in several previous studies [8] , [34] , [35] . However , evolution of the cluster concentration and wide-field cytoplasm concentration are not always separable , so an additional differential equation for the cluster is necessary . A schematic diagram of the model is shown in Fig . 8 . The corresponding ODEs are ( 1 ) ( 2 ) ( 3 ) where representing total intracellular concentration , and thus SR concentration , is given by . and are the volume ratios given in Table 1 . is the flux through the , is a background leak out of the SR , and is the uptake of into the SR by SERCA pumps . is the flux through plasma pump , and represents a sum of main influxes including ( receptor-operated channel ) , ( store-operated channel ) and ( leak into the cell ) . coarsely models the diffusion flux from cluster microdomain to the cytoplasm . Details of the fluxes are Calcium concentration at open channel pore ( ) does not explicitly appear in the equations but is used in the model introduced later . is assumed to be proportional to SR concentration ( ) and is therefore simply modeled by where is the value corresponding to . Alternatively , can also be assumed to be a large constant ( say greater than ) without fundamentally altering the model dynamics . The choice of is not critical as long as it is sufficiently large to play a role in inactivating the open channels . All the parameter values are given in Table 1 . The model used in our ASMC calcium model is an improved version of the Siekmann model which is a 6-state Markov model derived by fitting to the stationary single channel data using Markov chain Monte Carlo ( MCMC ) [5] , [7] , [8] . Fig . 1 has shown the structure of the model which is comprised of two modes; the drive mode , containing three closed states , , and one open state , and the park mode , containing one closed state and one open state . The transition rates in each mode are constants ( shown in Table 2 ) , but and which connect the two modes are -/-dependent and are formulated as ( 4 ) ( 5 ) where , , and are -/-modulated gating variables . , , and are either functions of or constants and are given later . We assume the gating variables obey the following differential equation , ( 6 ) where is the equilibrium and is the rate at which the equilibrium is approached . Those equilibria are functions of concentration at the cytoplasmic side of the , denoted by in the equations , equal to either or depending on the state of the channel ) . They are assumed to be ( 7 ) ( 8 ) ( 9 ) ( 10 ) Hence , we have stationary expressions of and , ( 11 ) ( 12 ) The expressions of s , s , s and s are chosen as follows so that Eq . 11 and Eq . 12 capture the correct trends of experimental values of and ( see Fig . 9 ) and generate relatively smooth open probability curves ( see Fig . 10 ) , Note that the above formulas are different from the relatively complicated formulas used in [8] . The rates , , and , are constants estimated by using dynamic single channel data [4] and given in Table 2 , whereas is not clearly revealed by experimental data . However we have shown that it should be relatively large for high but relatively small for low for reproducing experimental puff data [8] . By introducing two concentrations , and , and the state of the channel become highly correlated , so that we can assume is a relatively large value if the channel is open and is a relatively small value if the channel is closed . Hence , is modeled by the logic function Values of and are chosen so that simulated oscillations in ASMC are comparable to experimental observations . Here we reduce the 6-state model to a 2-state open/closed model . The reduction takes the following steps: Hence , the reduced two-state model contains one “open” state and one closed state with the opening transition rate of and the closing transition rate of . Based on the stochastic calcium model and the reduced 2-state model , we construct a deterministic model . We need to modify three things that are used in the stochastic model but inapplicable to fast simulations of the deterministic model . The first is the discrete number of open channels; the second is state-dependent use of and in calculating and ; the last is the logic expression of . Details of the modifications are as follows , Based on the above changes , the full deterministic model containing 8 ODEs is presented as follows , ( 13 ) ( 14 ) ( 15 ) ( 16 ) ( 17 ) ( 18 ) ( 19 ) ( 20 ) where and are functions of the gating variables given by Eqs . 4 and 5 . All the fluxes are the same as those of the stochastic model except . All the parameter values of the deterministic model are the same as those of the stochastic model and are therefore given in Tables 1 and 2 . The full deterministic model contains 8 variables which make the model difficult to implement and analyze . Thus , we reduce the full model to a minimal model that still captures the crucial features of the full model . First of all , , and are sufficiently large so that we can assume they instantaneously follow their equilibrium functions . Therefore , by taking quasi-steady state approximation to , and , we remove the three time-dependent variables from the full model . By now , the full model has been reduced to a 5D model , ( 21 ) ( 22 ) ( 23 ) ( 24 ) ( 25 ) Second , the rate of change of approaching its equilibrium , ( calculated from Eq . 24 ) , is at least one order larger than those of , and , indicating that taking the quasi-steady state approximation to Eq . 24 could not significantly affect the evolutions of , and . That is , ( 26 ) We emphasize here that the theory of the quasi-steady state approximation has not yet been well established , particularly about the rigorous conditions under which such a reduction is valid . Thus , our criterion of judging the validity of the reduction is checking whether the solutions of the reduced model are capable of qualitatively reproducing that of its original model . For this model , we find the reduction works . Hence , the full model is eventually reduced to a 4D model summarized as follows , ( 27 ) ( 28 ) ( 29 ) ( 30 ) where is given by Eq . 26 . To check the effect of calcium buffers on oscillation frequency , we introduce a stationary buffer ( no buffer diffusion ) , as mobile buffers are too complicated to be included in the current deterministic model . Since we have two different cytoplasmic concentrations , and , two pools of buffer with the same kinetics should be considered . Hence , the inclusion of a stationary calcium buffer is modeled by the following system , ( 31 ) ( 32 ) ( 33 ) ( 34 ) ( 35 ) where ( and ) and represent the concentrations of -bound buffer and total buffer respectively . and are the rates of -binding and -dissociation , indicating how fast the time scale of the buffer dynamics is . Fast buffer refers to the buffer with relatively large . In the simulations , we use a fast buffer with and and vary to test if the stochastic model and the deterministic model have a qualitatively similar -dependency . Results are given in Fig . 6E . For the stochastic model , Eqs . 1–3 and ODEs of the four gating variables in the model are solved by the fourth-order Runge-Kutta method ( RK4 ) and the stochastic states of determined by the model are solved by using a hybrid Gillespie method with adaptive timing [37] . The maximum time step size is set to be either ( for the 6-state model ) or ( for the reduced 2-state model ) . All the computations are done with MATLAB ( The MathWorks , Natick , MA ) and the codes are provided in Supporting information ( Text S1–S2 ) . For the deterministic model , we use ode15s , an ODE solver in MATLAB . Accuracy is controlled by setting an absolute tolerance of applied to all the variables . Data analysis is performed on the traces with relatively stable baselines and less noise . A moving average of every 3 data points is used to improve the data by smoothing out short-term fluctuations ( Fig . 2A is an improved result ) . Due to large variations in baseline , amplitude , and level of noise in data , we used two thresholds to get samples: a low threshold , 20% of the amplitude of the largest spike above the baseline , to initially filter baseline noise out; and a relatively high threshold , 50% of the amplitude of the largest spike above the baseline , to further remove small spikes that cannot initiate waves . For simulated stochastic traces of variable , we first convert it to fluorescence ratio ( ) by using where the dissociation constant of Oregon Green and resting . We then used the same sampling procedure mentioned above to obtain samples . After samples are chosen , ISIs and spike durations are calculated based on the low threshold . Simulated traces used to calculate average frequency are about 200–400 seconds long . All the samplings and linear least-squares fittings are implemented using MATLAB ( see Text S3–S4 for Matlab codes ) .	The inositol trisphosphate receptor ( ) is one of the most important cellular components responsible for calcium oscillations . Over the past decade , two major questions about the have arisen . Firstly , what fundamental properties of the allow it to perform its function ? Secondly , although calcium oscillations are caused by the stochastic properties of small numbers of is it possible for a deterministic model to be a reliable predictor of calcium dynamics ? Using airway smooth muscle cells as an example , we show that calcium dynamics can be accurately modeled using simplified models , and , secondly , that deterministic models are qualitatively accurate predictors of calcium dynamics . These results are important for the study of calcium dynamics in many cell types .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "biophysics", "biology", "and", "life", "sciences", "computational", "biology" ]	2014	A Deterministic Model Predicts the Properties of Stochastic Calcium Oscillations in Airway Smooth Muscle Cells
The ribosomal RNA genes ( rDNA ) comprise a highly repetitive gene cluster . The copy number of genes at this locus can readily change and is therefore one of the most unstable regions of the genome . DNA damage in rDNA occurs after binding of the replication fork blocking protein Fob1 in S phase , which triggers unequal sister chromatid recombination . However , the precise mechanisms by which such DNA double-strand breaks ( DSBs ) are repaired is not well understood . Here , we demonstrate that the conserved protein kinase Tel1 maintains rDNA stability after replication fork arrest . We show that rDNA associates with nuclear pores , which is dependent on DNA damage checkpoint kinases Mec1/Tel1 and replisome component Tof1 . These findings suggest that rDNA-nuclear pore association is due to a replication fork block and subsequent DSB . Indeed , quantitative microscopy revealed that rDNA is relocated to the nuclear periphery upon induction of a DSB . Finally , rDNA stability was reduced in strains where this association with the nuclear envelope was prevented , which suggests its importance for avoiding improper recombination repair that could induce repeat instability . DNA damage can lead to deletion , translocation and amplification of DNA in the genome , which may result in cell death , cancer and cellular senescence [1] . The most hazardous forms of genomic damage is the DNA double-strand break ( DSB ) that can occur randomly in the chromosome during replication , mainly in the S phase of the cell cycle , when the replication fork is arrested by DNA damage , torsional stress , modified nucleotides , or colliding transcription complexes . Stalled replication forks are thought to be targets of endonucleases that induce a DSB [2] . Downstream events of a DSB , such as DNA damage checkpoint control and DSB repair , have been analyzed [3] . Nonetheless , the mechanism of DSB repair in repetitive sequences without rearrangement is not well understood . Insights into the cellular mechanisms that prevent these rearrangements while allowing the broken genome to be repaired will contribute to the development of novel cancer treatments and broaden our understanding of the aging process . Here , we focus on the ribosomal RNA gene repeat ( rDNA ) to investigate the mechanism by which genome rearrangement is prevented after a DSB at a site with a stalled replication fork . In eukaryotic cells the rDNA forms a huge , conserved , tandem repeating structure ( > 100 copies ) on the chromosome . Transcription at this locus generates ribosomal RNA ( rRNA ) that , together with the ribosomal proteins , is assembled into ribosomes . A large number of ribosomes are needed to sustain cell-growth . Indeed , rRNA comprises approximately 80% of the total RNA in a cell [4] and , in the case of budding yeast Saccharomyces cerevisiae , ~ 150 rDNA copies are present on chromosome XII . Each repeating unit contains 35S and 5S rRNA genes , which are transcribed by RNA polymerases I and III , respectively ( Fig 1A ) . The transcript of the 35S rRNA gene is subsequently processed into mature 5 . 8S , 18S and 25S rRNA . The stability of rDNA is affected by recombination among the repeats , which can be easily detected by pulsed field gel electrophoresis [5] . For the upkeep of repeat number , cells can use a gene amplification mechanism that helps to maintain copy number by recombination [6] . In this system , replication is arrested at the replication fork barrier ( RFB ) site , located near the 3’ termination site of the 35S ribosomal RNA gene ( Fig 1A and S1 Fig ) . A complex formed by the binding of Fob1 to the RFB site inhibits replication against the direction of rDNA transcription [7] . A DSB is subsequently induced at the RFB site ( ~6% of arrested forks at the RFB site result in a DSB ) and repaired by recombination with the sister-chromatid [5 , 8 , 9] . When the broken end recombines unequally with a homologous site on the sister chromatid and replication restarts , some copies are replicated twice resulting in an increased copy number ( S1B-1 Fig ) . Thus , cells can use the rearrangement for copy number maintenance . This mechanism is regulated by the interplay between Sir2 , a histone deacetylase , and transcription from the nearby bidirectional promoter E-pro ( S1 Fig ) . In a cell with a wild-type rDNA copy number ( ~150 ) , E-pro transcription is repressed by Sir2 , but this repression does not occur in cells with a low rDNA copy number [10] . Non-coding transcription from E-pro , which prevents sister-chromatid cohesion , stimulates unequal sister-chromatid recombination [8] . When the copy number reaches the wild-type level , amplification stops . Alternatively , a DSB in the rDNA of a strain with a normal copy number can be repaired by a mechanism that does not involve homologous recombination , which reduces the risk of rearrangement ( and thus copy number instability ) . In this mechanism , as we have shown recently , a replisome component Ctf4 protects arrested forks from breakage and end resection . Although this pathway needs to be elucidated in more detail , it appears that DSB repair at arrested forks is regulated differently from replication-independent DSBs [9] . By using the unstable nature of rDNA as a measure , we screened a yeast library of ~4 , 800 deletion mutants of non-essential genes and identified ~700 ribosomal RNA gene unstable mutants ( RiUMs ) [11 , 12] ( http://lafula-com . info/kobayashiken/geldata/index . php ) . Among the RiUMs there was a deletion in TEL1 , which is an orthologue of the human ataxia-telangiectasia mutated ( ATM ) gene that responds to DNA damage and functions in telomere maintenance , damage checkpoint control and DSB repair [13] . Ataxia-telangiectasia or Louis–Bar syndrome is a rare , neurodegenerative , autosomal recessive disease that causes severe disability . In budding yeast , Tel1 regulates telomere length through phosphorylation of proteins involved in DSB repair and promotes elongation of telomere repeats [14] . Although Tel1 functions redundantly with the ATR orthologue Mec1 as S phase checkpoint kinases ( reviewed in [15] ) , the function of these proteins in rDNA maintenance has not been determined . Certain types of DNA repair appear to arise through recruitment of damage to specific subnuclear sites ( reviewed in [16] ) . TEL1 is involved in the relocation of DNA to the nuclear pores after inducing DSBs by means of endonuclease HO during the G1 and S/G2-phases of the cell cycle [17] . This irreparably damaged DNA also binds to the essential Sad1/UNC-84 ( SUN ) domain protein Mps3 in the inner nuclear membrane , but only when DSBs are induced during the S/G2-phase [18–20] . The rDNA instability in tel1Δ observed in our screen prompted us to investigate whether naturally occurring DSBs formed after replication arrest cause rDNA to translocate to the nuclear envelope . Using chromatin immunoprecipitation ( ChIP ) assays , we detected binding of rDNA to the nuclear pores , which required Tel1 and Mec1 , indicating this localization is DNA-damage dependent . In addition , Tof1 , a component of the replisome , which is necessary for fork arrest at the RFB , together with condensin recruiting factors were also found to be required for localization of rDNA to the nuclear pores . Defective association to nuclear pores reduced rDNA stability , suggesting that this association helps to maintain repeat stability . Recently , we screened a yeast deletion library for factors involved in the maintenance of rDNA stability and identified ~700 ribosomal RNA unstable mutants ( RiUM ) [11 , 12] . Among these , there were genes related to DNA repair for which the molecular mechanism with respect to rDNA was not known . In this category , we focused on a protein kinase Tel1 that regulates telomere length through phosphorylation of proteins mediating DSB repair and that enhance elongation of telomere repeats [14] . We first introduced the tel1 deletion to our laboratory strain to confirm the generality of the phenotype . We performed PFGE assays three times and one of the trials was followed by Southern blotting with an rDNA probe ( Fig 1B , S2A and S2B Fig ) . Although the effect was relatively modest as that of the library strain , quantitative analysis revealed that the bands of rDNA-containing chromosome XII were broader in the tel1Δ compared to wild-type ( Fig 1C . See S2B Table and Materials and methods for about the quantification ) . Such variable copy numbers are a hallmark of unstable rDNA [5] . In this assay , the bands of chromosome XII in fob1Δ were not shaper compared to wild-type . The similar observation was made in a previous study illustrating the inherent difficulty of the detection of a more stable band than that of the wild-type strain [12] . To test whether rDNA instability in the tel1Δ is related to replication fork barrier activity that induces a DSB , we made a double mutant , tel1Δ fob1Δ . In the double mutant , the bands of chromosome XII became as sharp as that of the fob1Δ ( Fig 1B and 1C ) , indicating that rDNA instability in the tel1Δ is caused downstream of Fob1 . Thus , Tel1 functions after replication fork arrest mediated by Fob1 and before involvement in rDNA maintenance . We reasoned tel1Δ might have an effect on replication fork blocking activity and therefore DSB frequency at the RFB site . Thus , we examined this possibility by two dimensional gel electrophoresis ( 2D gel assay ) in which the amount of replication fork arrest can be determined from the signal intensity of the “RFB-spot” corresponding to the number of Y-shaped replication intermediates accumulating at the RFB site [21 , 22] . In the tel1Δ , the “Y-arc , Double-Y and RFB-spot” signals , corresponding to replication intermediates , was slightly weaker than that in the wild-type cells , probably because of the reduced number of S-phase cells in the mutant ( Fig 2A ) . To compare these strains , RFB-spot intensity was normalized to the replication intermediates . No significant difference in stalling of the replication forks was observed ( RFB-spot , Fig 2A and 2B and S2B Table ) . The 2D gel-assay also gave insight into the frequency with which a DSB is formed after replication-fork arrest by means of the “DSB-spot” i . e . a signal that corresponds to broken fragments at the RFB site . The signal of the spot ( ~2 . 3 kb ) disappeared in the fob1Δ because there was no arrest of the replication fork [8 , 23] . Relative to the RFB spot , the intensity of the DSB spot was not affected in the tel1Δ ( Fig 2A and 2C and S2B Table ) . Thus , the increased levels of replication fork blocking activity and resulting increased DSBs are unlikely to be the cause of rDNA instability in the tel1Δ . Although the frequency of DSB was not increased in tel1Δ compared to wild-type , the mutant exhibited Fob1-dependent rDNA instability ( Figs 2 and 1 , respectively ) . A previous study demonstrated that Tel1 is required for translocation of HO-induced persistent DSBs to the nuclear pore and pore-binding is implicated in alternative recombination-mediated repair pathways [17] . Therefore , we hypothesized that replication-dependent DNA damage in rDNA might be associated with nuclear pores in a Tel1-dependent manner . To test this hypothesis , we performed chromatin immunoprecipitation ( ChIP ) assays with mAB414 , which is an anti-nucleoporin antibody [20] . Five PCR primer sets in an rDNA unit were designed to detect precipitated rDNA , while two primer sets in SMC2 and CUP1 were used to detect control loci ( Fig 3A ) . The precipitated rDNA was assessed by quantitative real-time PCR ( qPCR ) and relative enrichment was normalized against CUP1 . Our results show that rDNA is enriched at the nucleoporins , which constitute nuclear pores , by 4 . 4- to 8 . 1-fold relative to the CUP1 locus . By contrast , the control SMC2 locus did not display any enrichment ( Fig 3A ) . Intriguingly , enrichment immediately adjacent to the RFB was relatively weak by comparison to the surrounding regions ( Fig 3A and S2A Table ) . Similar results were observed for the HO induced-DSB [17] . Although the underlying mechanism remains unclear , it may involve phosphorylation of histone H2A , recruitment of DNA repair proteins and/or DSB end resection around the DSB . To evaluate the differences between wild-type and mutant strains , we calculated the relative enrichment of mutant strains to wild-type in each ChIP assay and compared the means of three independent assays ( Fig 3B and S2A Table ) . The rDNA association with nuclear pores was significantly reduced both in tel1Δ and mec1Δ sml1Δ , suggesting that association of rDNA with the nuclear pores is dependent on DNA damage checkpoint kinases Tel1 and Mec1 . Tof1 is a component of the replisome and , like Fob1 , is required for the arrest of the replication fork at the RFB and the formation of a DSB [24 , 25] . To test whether the nuclear-pore association depends on the replication block in the rDNA , we performed the ChIP assay with the fob1Δ and tof1Δ , both of which do not exhibit the replication fork block at the RFB [8 , 24 , 26] . In the absence of Tof1 , rDNA association with the nuclear pores was significantly reduced ( Fig 3B and S2A Table ) . In contrast , the reduction was smaller for the fob1Δ and was not statistically significant . The reason for the observed differences between these two mutants is unclear . One possible explanation is that Fob1 is responsible for RFB only , while Tof1 might be related to replication fork arrest at any sites in rDNA as it travels with the replication fork . Indeed , there was no difference in binding to the nuclear pore at RFB between the fob1Δ and tof1Δ mutants ( P-value = 0 . 303477 . S2A Table ) . For tof1Δ , however , nuclear pore-binding was significantly decreased at non-RFB loci in rDNA ( P-value < 0 . 05 ) , except at the 3' end of 35S rDNA ( P-value = 0 . 050003 ) . This observation suggests , unlike Fob1 , the role of Tof1 in nuclear pore binding is not limited to RFB sites ( see Discussion section ) . Because the replication fork block induces DNA damage only in S-phase , the association was expected to occur in this phase of the cell cycle . To confirm that , we synchronized cells in G1 phase and tested the association . Contrary to our expectation , the nuclear-pore association was detected even in G1 phase ( S3 Fig ) . This raises the possibility that the association may be maintained throughout mitosis ( see Discussion ) . In budding yeast , persistent DNA damage is recruited to the nuclear periphery and is associated with nuclear pores through the Nup84 subcomplex [17] , which contains Nup133 , Nup120 , Nup145C , Nup85 , Nup84 , Seh1 , and Sec13 [27–29] . The nuclear pore association of rDNA compromised both the nup84Δ and nup120Δ and the effect was more pronounced in the deletion of NUP120 , suggesting that rDNA association with nuclear pores requires intact Nup84 complex ( Fig 4 and S2A Table ) . The rDNA gives rise to the nucleolus , which is a membrane-less organelle that appears to assemble through phase separation . Importantly , recombination foci are excluded from the nucleolus indicating that rDNA repair occurs in a specific environment distinct from the nucleolus [30] . Although Mec1/Tel1 have been implicated in nuclear pore association of DSB , there may be rDNA-specific factors that are involved in the nuclear pore association . We speculated that putative candidates would interact both with rDNA and with the nuclear pores or the surrounding nuclear membrane proteins . This holds for condensin recruiters Tof2 , Csm1 and Lrs4 , which have been identified as synthetic lethal mutants with a condensin conditional mutant ( smc2-157 ) and that interact with Fob1 and recruit condensin to the rDNA [31] . Csm1 and Lrs4 are also known as cohibin that associates with CLIP ( chromosome linkage inner nuclear membrane proteins , Src1 and Nur1 ) and localizes the rDNA to the CLIP to maintain rDNA stability , even though it has not been shown whether the binding is damage-dependent [32 , 33] . To test the contribution of these proteins to the association of rDNA with nuclear pores , we performed a ChIP-qPCR assay with deletion mutants for the factors . The rDNA association with the nuclear pores in all these mutants was reduced compared to wild-type , indicating that condensin recruiters are required for rDNA relocation to the nuclear pores ( Fig 5 and S2A Table ) . Sir2 also acts as a bridge between rDNA and the nuclear pores as is the case for CLIP ( Fig 5 and [32] ) . For sir2Δ , the association of rDNA with the nuclear pores was also reduced ( Fig 5 and S2A Table ) . To determine the subnuclear localization of spontaneously damaged rDNA , we used a strain in which each copy of the rDNA repeat has a lacO array that associates with LacI-GFP [34] . We scored DSBs on the rDNA by monitoring the foci of Rad52-CFP , a factor essential for homologous recombination that accumulates at DSBs ( S4A and S4B Fig ) . The Rad52 focus was barely detected under normal physiological conditions ( 4 cells scored from 875 asynchronous cells; 1 . 26% ) and colocalization of Rad52-CFP and LacI-GFP / rDNA-lacO was even less frequent ( 0 . 46% ) . Note that Rad52 foci are formed only when the DSBs are excluded from the nucleolus [30] and we estimate that less than 21% of DSBs are marked by discrete Rad52 foci in the rDNA ( see legend to Fig 6C ) . This may result in a loss of data for a large fraction of DSBs if we use Rad52 as a marker of DSB in the rDNA . Instead , we used I-SceI endonuclease to induce DSB in the rDNA [30] . In this assay , I-SceI cleaves the recognition sequence inserted in the rDNA and the location of DSB is detected by TetI fused with mRFP ( monomeric red fluorescent protein ) that associates with the adjacently located tetO array [30] ( Fig 6A ) . The I-SceI induced DSB is known to shift away from the nucleolus to complete homologous recombinational repair [30] . Using this system , we scanned the position of the TetI-mRFP focus and classified them into three zones compared with mRFP-fused nuclear pore proteins [35] ( Fig 6B ) . Before induction of I-SceI , the TetI-mRFP locus was preferentially positioned in the nuclear center . Strikingly , the locus was relocated to the nuclear periphery both in the G1 and S phases within 2 hours of DSB induction ( Fig 6C and 6D ) . No enrichment was observed in the strain lacking the I-SceI endonuclease , confirming the association is damage-specific ( Fig 6E ) . These results indicate that DSB in the rDNA is localized in the nuclear periphery . To test whether rDNA association with the nuclear pores has a biological role in maintaining rDNA stability , we analyzed the migration of chromosome XII in mutants that fail to relocate rDNA to the nuclear pores ( sir2Δ , tel1Δ , nup84Δ , nup120Δ , tof2Δ , csm1Δ , and lrs4Δ ) by pulsed field gel electrophoresis ( PFGE , Fig 7A and 7B ) . The fob1Δ and sir2Δ were used as the negative and the positive control , respectively . All mutants except for nup84Δ exhibited an unstable chromosome XII compared to the wild-type ( Fig 7A and 7B ) . Nup84 and Nup120 belong to the same heptameric Nup84 complex of nuclear pore complex [28 , 29 , 36] . However , the nuclear pore association and the stability of rDNA were differentially affected in these mutants ( Figs 4 and 7 ) . These findings are consistent with the fact that DNA damage sensitivity in the nup120Δ is stronger than that in the nup84Δ [37] . Taken together , these data suggest that Nup120 plays a more prominent role than Nup84 in DNA repair through an unknown mechanism . Mps3 acts as an alternative anchoring site of HO-induced DSBs on the nuclear membrane [18 , 19 , 38] . A mutant form of the essential Mps3 ( mps3Δ65–145 ) , truncated at the N-terminal acidic domain , did not affect rDNA stability according to PFGE analysis ( Fig 7A and 7B , [39] ) . Furthermore , nup120Δ mps3Δ65–145 double mutations did not show any additive effect in terms of rDNA-stability compared to the corresponding single mutations , suggesting that Mps3 does not make a significant contribution to rDNA stability . Given that rDNA instability in tel1Δ was dependent on Fob1 ( Fig 1B and 1C ) , the replication-dependent DNA damage in rDNA appears to bind to the nuclear pores for its maintenance . rDNA is one of the most unstable regions in the genome due to its repetitive nature . Recombination among the repeats would result in deletions ( loss of copies ) leading to copy number instability . Nonetheless , cells appear to have evolved mechanisms to avoid such instability , which would be deleterious . Association of rDNA to the nuclear pores seems to be one such mechanism . By this change in location , the broken rDNA unit is isolated from intact copies and the risk of hazardous recombination thereby reduced . Moreover , alternative repair pathways at the nuclear pore might be facilitated [17 , 40] . In Fig 8 , we summarize how the damaged rDNA is repaired . Recently , we found that the ends of a DSB formed after stalling of a replication fork at the RFB are not resected in a strain with a normal rDNA copy number , and that the DSB is repaired through a pathway that does not involve homologous recombination [9] . In this pathway , the DSB can be repaired without alteration of rDNA copy number . Therefore , we proposed that this homologous recombination-independent repair is the default mechanism used for rDNA repair ( 1st stage , Fig 8 ) . In contrast , when the rDNA copy number is reduced in a strain , resection of the DSB is induced , which triggers unequal sister-chromatid recombination that may amplify the number of rDNA copies [9] . For this reaction , the DSB together with the surrounding region needs to be moved from the nucleolus to the nucleoplasm where the homologous recombination enzymes , including Rad52 , form distinct foci ( 2nd stage ) [30] . Previously , we found that E-pro transcription is activated and cohesin dissociates from the rDNA in the absence of Sir2 . As a result , unequal sister-chromatid recombination was increased and the copy number changed with a high frequency [10] ( S1 Fig ) . The E-pro regulated recombination may occur at this stage just outside of the nucleolus . Finally , if the DSB cannot properly be repaired at the 2nd stage , the DSB with the surrounding region relocates to the nuclear envelope where it is trapped by the nuclear pores ( 3rd stage ) . In the presence of a repair template , no binding of the DSB to the nuclear periphery was observed in a previous HO-induced DSB assay [17 , 19] . Although there are abundant repair templates in the case of damaged rDNA , the locus is relocated to the nuclear pores presumably because it is isolated from the majority of templates at the 2nd and 3rd stages . The 3rd stage may work as a back-up system for the 1st and the 2nd stages and could prevent aberrant genomic changes such as the generation of a large deletion . The isolated broken ends around the nuclear pores may be repaired by homologous recombination with chromosomal rDNA or an ERC . Otherwise , repair of the broken ends may occur via the single strand annealing ( SSA ) pathway that connects repetitive sequences using the homologous sequence without introducing mutations [41] . In this study , proteins involving replication fork bock , DNA damage checkpoint and condensing loading were implicated in the rDNA-nuclear pore binding . Unraveling the hierarchy of these factors is an exciting challenge for future studies . In the tof1Δ , defects in the association to the nuclear pores were more obvious than in the fob1Δ ( Fig 3B ) . The reason for the difference in dissociation between these mutants is unclear . One possible explanation is that Fob1 is specifically responsible for the RFB , while Tof1 might be associated with replication fork arrest at any site in rDNA given that it travels with the replication fork . In the fob1 mutant with a low rDNA copy number , collision between 35S transcription and replication machineries causes inhibition of the replication fork and induces rDNA instability [42] . This damage to the DNA may occur to some extent in a normal copy strain and trigger the relocation . By contrast , in the tof1 mutant , such RFB independent damage might also be reduced , resulting in a lower level of nuclear pore binding . The binding of rDNA to nuclear pores was detected even in the G1 phase ( S3 Fig ) . Because no replication-dependent DSB is induced in G1 phase , the data does not easily fit the DSB dependent-binding model ( Fig 8 ) . Nonetheless , there are several possible explanations for the cell cycle independent association of rDNA to nuclear pores . The first interpretation is that the binding is caused by extra-chromosomal rDNA circles ( ERCs ) that are produced by unequal sister chromatid recombination . However , the ChIP data in sir2Δ does not support this hypothesis ( Fig 5 ) . Because sir2Δ leads to instability of rDNA and produces vast amounts of ERCs , the strains should show an accumulation of rDNA-nuclear pore binding if ERCs bind to the nuclear pores . However , no such accumulation was observed . An alternative interpretation is that a DSB in rDNA that is not repaired in S/G2 phases might be carried into the next cell cycle . It is known that damage in the rDNA does not induce checkpoint control [43] . Once a DSB in rDNA is carried over to the next cell cycle , it can be recruited to or maintained at the nuclear periphery in G1 phase as seen in endonuclease-induced damage ( Fig 6 ) . A third interpretation of cell-cycle independent interaction of rDNA to nuclear pores is that the rDNA binds to the nuclear pore and is maintained at the site even after repair is completed . The replication-dependent rDNA damage occurs in S-phase and rDNA is relocated to the nuclear periphery . The DSB in rDNA is repaired in S/G2 phases and the locus might be kept at the nuclear periphery until the next G1 phase . In either case , we hypothesize that a small portion of damaged rDNA remains in the mother cell with the nuclear envelope , which may be carried into the next cell cycle . Indeed , we detected stacked rDNA in the wells during pulse-field gel electrophoresis specifically of mother-cells in G1 phase ( three or four budded age ) . This observation suggests an accumulation of unstable rDNA in the G1 phase of mother cells [44] . We propose that this accumulation of broken ends could be a cause for senescence of the mother cell . Several recent papers highlight the importance of perinuclear anchoring for continuing damage repair . It has been shown that replication damage associated with expanded triplet repeats and eroded telomeres shift transiently to the nuclear pores [45 , 46] . Su et al . showed that an artificially inserted CAG repeat is localized to the nuclear pores in a replication-dependent manner and this localization was important for CAG repeat stability [45] . As the repeat may form a secondary structure and arrest replication , the CAG repeats and rDNA are expected to share a common mechanism that localizes them to the nuclear periphery , at least partially . Churikov et al . showed that shortened telomeres in a telomerase-deficient yeast strain are relocated to the nuclear pores and this localization was required for type II survivors in which the short terminal TG-tract is elongated by recombination ( ALT in mammals ) [46] . Although the relationship between the shortened telomere recombination and rDNA stability is not known , localization at the nuclear pore seems to be important for many aspects of genome maintenance . In this study , we identified a mechanism that protects damaged repetitive rDNA sequences from undergoing rearrangement ( copy number variation ) by association with the nuclear pores . In this way rDNA stability is maintained probably via the SSA pathway , which cannot be applied to DSBs in non-repetitive sequences . Likewise , in Drosophila cells , a DSB in heterochromatin that mostly comprises repetitive sequences relocates to the nuclear pores for repair in a SUMOylation-dependent manner [47] . SUMOylation also mediates relocation of the DSB in the rDNA to outside of the nucleolus and the eroded telomere to the nuclear periphery in Saccharomyces cerevisiae [30 , 46] . It has been reported that damaged rDNA is relocated to specific loci around the nucleolus of mammalian cells and most of the factors required for this relocation , which were identified in yeast , are well conserved [48] . Because mammalian genomes contain large stretches of repetitive sequences , such as retrotransposons and Alu-repeats , a similar mechanism may operate to maintain genome integrity in higher eukaryotes . Future studies will shed light on the involvement of human homologues in the repair of damaged repetitive DNA . Yeast strains used in this study were derived from NOY408-1b ( a W303 derivative ) . Strains were grown at 30°C in YPD ( YPDA for Figs 1 , 3 , 4 , 5 , 7 and S3 Fig ) medium . YPD ( yeast extract-peptone-dextrose ) and YPDA ( YPD with 0 . 4% adenine ) are rich media used for normal culture . Synthetic complete ( SC ) medium lacking the appropriate amino acids [49] was used for gene marker selection . Yeast strains used in this study are listed in S1 Table . If necessary , G418 ( Sigma ) or clonNAT ( WERNER ) was added to the medium at the following concentration , 500 μg/ml ( G418 ) or 100 μg/ml ( clonNAT ) . Yeast genetic transformation was performed by using Frozen-EZ Yeast Transformation II Kit ( Zymo Research Corporation ) according to the manufacturer’s instructions . To test rDNA stability by pulsed field gel electrophoresis , we used cells that had divided ~45 times after transformation . For the DSB localization assay , yeast cells were grown at 30 °C for 2 days on selective synthetic medium containing 2% glucose ( SD ) . The cells were inoculated in synthetic medium containing 2% raffinose ( SR ) and grown overnight . The culture was diluted to SR next morning and grown for about 4 hours . When the exponentially growing cell population reached around 2 . 5 × 106 cells ml−1 , we added 20% galactose ( final 2% ) to the medium to induce I-SceI . The living cells were directly subjected to microscopy on an SR agarose pad . We used SD/SR-lacking tryptophan and uracil for YCH-252 or lacking tryptophan , uracil and histidine for YCH-244 in these experiments . Samples for pulsed-field gel electrophoresis ( PFGE ) were prepared as described previously [50] . Electrophoresis was performed in a 1% ( 0 . 8% for S2B Fig ) agarose gel with 0 . 5×Tris-borate-EDTA ( TBE ) buffer , using CHEF-MAPPER ( Bio-Rad ) . The conditions were a 300–900 sec pulse time and 100 V for 68 hours at 14 °C . For S2B Fig , after electrophoresis , the rDNA was detected by Southern blot analysis with an rDNA specific probe . To quantify instability of rDNA in PFGE ( Figs 1C and 7B ) , the signal intensities of Chr . XII and Chr . IV were measured by Image J ( Fiji ) using the image of an EtBr stained gel . The signal intensities of Chr . XII were divided by that of Chr . IV , which was expected to be constant between mutants . Broader unstable bands reduce signal intensities in the area . Moreover , chromosomes with an unusual structure cannot enter the gel and thereby reduce signal intensity . Normalization of the Chr . XII band intensity in the mutants to that of Chr . IV , yielded values reflecting their rDNA stability . In the tof2 , csm1 and lrs4 mutants , several minor bands were observed . This suggests some of the cells contained multiple copies of chromosome XII because of chromosome missegregation caused by condensation defects in these mutants [31] . In such cases , the major band was measured . 2D gel electrophoresis was performed as previously described [51] . DNA from early log phase cells ( ~3x106 cells/ml in YPD medium ) were digested in agarose plugs ( 5x107 cells/plug ) using BglII for 4 h at 37 °C . The reaction was carried out in 200 μl reaction buffer with 150 units of BglII . After electrophoresis , the rDNA was detected by Southern analysis with an rDNA specific probe . RFB and DSB signals were quantified by ImageQuant ( GE ) . The signal intensity of the RFB spot was divided by the signal intensity of total replication intermediates signal for normalization . The signal intensity of the DSB spot was normalized to the RFB signal to show the relationship between the DSB and the arrested fork it was derived from . ChIP was carried out as previously described [52] with minor modifications described below . Yeast cells cultured in 45 ml medium were cross-linked with 1% formaldehyde at 30 °C for 20 min . Cell pellets were resuspended in 600 μl of lysis buffer ( 50 mM HEPES-KOH at pH 7 . 5 , 500 mM NaCl , 1 mM EDTA at pH 8 . 0 , 1% Triton X-100 , 0 . 1% sodium deoxycholate and protease inhibitors ) and disrupted with zirconia beads using a Multi-bead shocker ( Yasui Kikai ) . The recovered chromatin fraction was subjected to sonication using a Bioruptor ( Cosmo Bio ) to obtain fragmented chromatin < 500 bp in length . An anti-nuclear pore FG-repeat antibody ( mAB414 , Abcam ) combined with Dynabeads Protein G ( Thermo Fisher ) , was used for IP . Beads were washed twice in lysis buffer , once with wash buffer ( 10 mM Tris-HCl at pH 8 . 0 , 250 mM LiCl , 0 . 5% Nonidet P40 ( IGEPAL ) , 0 . 5% sodium deoxycholate , 1 mM EDTA at pH 8 . 0 and protease inhibitors ) , and once with TE ( 10 mM Tris-HCl at pH 8 . 0 and 1 mM EDTA at pH 8 . 0 ) at 4 °C . ChIP DNA was purified and analyzed by quantitative real-time PCR using primers amplifying various regions of the rDNA , the SMC2 ( condensin complex ) locus on Chr . VI or the CUP1 locus on Chr . VIII ( primer sequences are listed in S3 Table ) . Enrichment was normalized to that from the genomic CUP1 locus in IP and Input DNA samples and were calculated as [rDNA or SMC2 ( IP ) / CUP1 ( IP ) ] / [rDNA or SMC2 ( Input ) / CUP1 ( Input ) ] . Details of the formula used for these calculations is given below: Relativeenrichment=%Input ( Testlocus ) /%Input ( Controllocus ) %Input ( Testlocus ) =100×2∧ ( Ct ( AdjustedInput ) −Ct ( IP ) ) %Input ( Controllocus ) =100×2∧ ( Ct ( AdjustedInput ) −Ct ( IP ) ) Ct ( AdjustedInput ) =Ct ( Input ) −LOG ( 10 , 2 ) The “Relative enrichment over CUP1” is shown in Fig 3A . To compare wild-type and mutant cells , we divided the values corresponding to the mutants ( or G1-phase wild-type cells in S3 Fig ) by that of the wild-type cells ( or asynchronous wild-type cells in S3 Fig ) in each ChIP assay . The mean values of three ( Figs 3 , 4 and 5 ) and five ( S3 Fig ) independent assays are shown . Fluorescence microscopy and quantification was performed according to published methods [35 , 53] using an ECLIPSE Ti microscope ( Nikon ) fitted with a Zyla 4 . 2P sCMOS ( Andor Technology ) camera . TetI-mRFP position was determined with a through-focus stack of 12 0 . 3 μm steps and was measured by ImageJ ( Fiji ) and the plug-in software PointPicker [53] . The numbers of nuclei scored are shown in S2C Table . The efficiency of DSB induction was determined by real-time PCR with SYBR Green as previously described [54] . To determine zone enrichment , we applied a χ2 test comparing zone 1 or zone 3 to a random distribution ( degree of freedom = 2 , confidence limit = 95% ) . p-values are indicated in S2C Table .	Ribosomal RNA genes ( rDNA ) comprise an unstable region of the genome due to their highly repetitive structure and elevated levels of transcription . Collision between transcription and replication machineries of rDNA , which may lead to DNA damage in the form of a double-stranded break , is avoided by the replication fork barrier . When such a break is repaired by homologous recombination with a repeat on the sister chromatid , the abundance of homologous sequences may lead to a change in copy number . In most organisms , however , only small variations in copy number are observed , indicating that the rDNA is stably maintained . Our results suggest that some parts of rDNA become localized to the nuclear pore complex in a DNA double-strand break-dependent manner . This localization requires the protein kinase Tel1 , which is involved in the DNA damage response pathway , and factors that recruit condensin , which facilitates condensation and segregation of rDNA during mitosis . We found that the rDNA becomes unstable when association with the nuclear envelope was prevented . Thus , the localization represents a unique strategy for maintaining repeat integrity after DNA damage .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "cell", "cycle", "and", "cell", "division", "cell", "processes", "dna", "damage", "fungi", "model", "organisms", "dna", "replication", "experimental", "organism", "systems", "dna", "recombination", "dna", "cell", "nucleus", "cellular", "structures", "and", "organelles", "nuclear", "pores", "homologous", "recombination", "research", "and", "analysis", "methods", "saccharomyces", "animal", "studies", "ribosomes", "yeast", "biochemistry", "rna", "eukaryota", "ribosomal", "rna", "cell", "biology", "nucleic", "acids", "genetics", "biology", "and", "life", "sciences", "dna", "repair", "saccharomyces", "cerevisiae", "yeast", "and", "fungal", "models", "non-coding", "rna", "organisms" ]	2019	Ribosomal RNA gene repeats associate with the nuclear pore complex for maintenance after DNA damage
Plant nucleotide-binding leucine-rich repeat ( NB-LRR ) disease resistance ( R ) proteins recognize specific “avirulent” pathogen effectors and activate immune responses . NB-LRR proteins structurally and functionally resemble mammalian Nod-like receptors ( NLRs ) . How NB-LRR and NLR proteins activate defense is poorly understood . The divergently transcribed Arabidopsis R genes , RPS4 ( resistance to Pseudomonas syringae 4 ) and RRS1 ( resistance to Ralstonia solanacearum 1 ) , function together to confer recognition of Pseudomonas AvrRps4 and Ralstonia PopP2 . RRS1 is the only known recessive NB-LRR R gene and encodes a WRKY DNA binding domain , prompting suggestions that it acts downstream of RPS4 for transcriptional activation of defense genes . We define here the early RRS1-dependent transcriptional changes upon delivery of PopP2 via Pseudomonas type III secretion . The Arabidopsis slh1 ( sensitive to low humidity 1 ) mutant encodes an RRS1 allele ( RRS1SLH1 ) with a single amino acid ( leucine ) insertion in the WRKY DNA-binding domain . Its poor growth due to constitutive defense activation is rescued at higher temperature . Transcription profiling data indicate that RRS1SLH1-mediated defense activation overlaps substantially with AvrRps4- and PopP2-regulated responses . To better understand the genetic basis of RPS4/RRS1-dependent immunity , we performed a genetic screen to identify suppressor of slh1 immunity ( sushi ) mutants . We show that many sushi mutants carry mutations in RPS4 , suggesting that RPS4 acts downstream or in a complex with RRS1 . Interestingly , several mutations were identified in a domain C-terminal to the RPS4 LRR domain . Using an Agrobacterium-mediated transient assay system , we demonstrate that the P-loop motif of RPS4 but not of RRS1SLH1 is required for RRS1SLH1 function . We also recapitulate the dominant suppression of RRS1SLH1 defense activation by wild type RRS1 and show this suppression requires an intact RRS1 P-loop . These analyses of RRS1SLH1 shed new light on mechanisms by which NB-LRR protein pairs activate defense signaling , or are held inactive in the absence of a pathogen effector . Plant innate immunity relies on two layers of pathogen detection . Cell surface-localized pattern recognition receptors detect pathogen-associated molecular patterns ( PAMPs ) of invading microorganisms and activate PAMP-triggered immunity ( PTI ) [1] . Successful pathogens must circumvent PTI to colonize plants , and many bacterial pathogens use type III secretion ( T3S ) to deliver effectors that suppress PTI into plant cells [1] . Effectors can be detected directly or indirectly by plant disease resistance ( R ) proteins , which then activate effector-triggered immunity ( ETI ) generally together with a hypersensitive response ( HR ) of the infected tissue [2] . Most intracellular R proteins are modular , with an amino-terminal coiled coil ( CC ) or Toll/interleukin-1 receptor/R protein ( TIR ) domain , a nucleotide binding ( NB ) domain and a leucine-rich repeat ( LRR ) domain [3] . Some NB-LRR proteins also carry an additional carboxyl-terminal extension , the function of which is unknown [3] . In addition , NB-LRR protein function generally requires an intact P-loop motif ( GxxxxGKT/S ) in the NB domain , presumably for ATP binding and energy-dependent conformational changes [3] , [4] . Plant NB-LRR proteins and mammalian Nod-like receptors ( NLRs ) exhibit both structural and functional similarities [5] . Signaling following TIR-NB-LRR protein activation requires other key regulators such as Enhanced Disease Susceptibility 1 ( EDS1 ) , the EDS1-related proteins PAD4 and SAG101 , and biosynthesis of the plant hormone salicylic acid ( SA ) for full immunity [6] . EDS1 was recently reported to interact with several NB-LRR proteins [7] , [8] . Mis-regulation of R protein accumulation , localization or activation can cause constitutive defense responses , which are usually deleterious or lethal . For instance , the dwarf suppressor of npr1-1 , constitutive 1 ( snc1 ) mutant carries a point mutation between NB and LRR domains of the TIR-NB-LRR protein SNC1 , which results in constitutive defense signaling [9] , [10] . Suppression of the stunted snc1 phenotype in mos ( modifier of snc1 ) mutants allowed the identification of several genes required for nuclear defense signaling [11]–[14] . Although most R proteins function to recognize a corresponding avirulent effector ( Avr ) , some NB-LRR proteins appear to act downstream of R protein activation . The tobacco and tomato CC-NB-LRR proteins , “N-required gene 1” ( NRG1 ) , and “NB-LRR protein required for HR-associated cell death 1” ( NRC1 ) , are required for TIR-NB-LRR protein N-mediated resistance to tobacco mosaic virus and receptor-like protein Cf-4-mediated resistance to tomato leaf mold pathogen , respectively [15] , [16] . Arabidopsis CC-NB-LRR Activated Disease Resistance 1 ( ADR1 ) family proteins are required for SA-dependent ETI [17] . The Arabidopsis accession Col-0 downy mildew resistance locus RPP2 comprises two distinct closely linked NB-LRR genes RPP2A and RPP2B , both of which are required for resistance [18] . The rice Pia locus for blast ( Magnaporthe ) resistance comprises two divergently transcribed CC-NB-LRR genes , RGA4 and RGA5 , again both required for resistance [19] . In mammals , the NLR NAIP2 confers specific recognition of PrgJ , whereas NLRs NAIP5 and NAIP6 confer responses to flagellin . However , the NLR NLRC4 is required for defense responses to both PrgJ and flagellin [20] , [21] . NLRC4 association with either NAIP2 or NAIP5/6 , upon provision of PrgJ or flagellin respectively , is required for defense activation [20] , [21] . The T3S effectors AvrRps4 and PopP2 from Pseudomonas syringae and Ralstonia solanacearum respectively , are recognized by paired TIR-NB-LRR proteins RPS4 ( resistance to P . syringae 4 ) and RRS1-R ( resistance to R . solanacearum 1 ) , and activate ETI in Arabidopsis [22]–[24] . RRS1-R alleles , found in accessions Ws-2 , No-0 and Nd-1 , confer recognition of PopP2; the RRS1-S allele of Col-0 does not recognize PopP2 , but does recognize AvrRps4 [22]–[24] . Lack of AvrRps4 recognition in accession RLD is due to non-synonymous mutations in RPS4 , and RRS1-S in Col-0 is truncated compared to RRS1-R because of an early stop codon [24]–[26] . RPS4 and RRS1-R genetically function together , as plants lacking RPS4 , RRS1-R or both show similar enhanced susceptibility to bacterial strains expressing AvrRps4 or PopP2 [25] , [26] . RRS1 ( also annotated as WRKY52 ) is an atypical NB-LRR protein that also carries a C-terminal WRKY DNA-binding domain [22] . In this study , we delivered PopP2 using Pseudomonas T3S by fusing it with the N-terminal region of AvrRps4 ( AvrRps4N ) . Pseudomonas-delivered AvrRps4N:PopP2 triggers RPS4- and RRS1-dependent HR and immunity in resistant Arabidopsis genotypes when tagged with a nuclear localization signal ( NLS ) but not when tagged with a nuclear exclusion signal ( NES ) . We show that the delivery of PopP2 , or an inactive PopP2C321A variant , from a Pseudomonas fluorescens strain ( Pf0-1 ) that lacks other effectors [27] , results in the induction of ETI-specific genes that overlaps substantially with previously reported AvrRps4-regulated genes [28] , [29] . The presence of a single amino acid ( Leu ) insertion in the WRKY domain of RRS1-R ( RRS1SLH1 hereafter ) causes the recessive lethal phenotype of the sensitive to low humidity 1 ( slh1 ) mutant in No-0 [30] . RRS1SLH1-induced lethality is associated with enhanced defense gene expression and high SA accumulation . Similarly to other mutants displaying spontaneous cell death , slh1 mutant growth can be restored to wild type phenotype at 28°C [30]–[32] . In contrast to snc1 , the slh1 mutant allele is recessive and heterozygotes show no constitutive defense activation [30] . RRS1 is also recessive and an RRS1-R/RRS1-S heterozygote is unable to recognize PopP2 [22] , [23] . Here , we used the conditional RRS1SLH1-mediated lethal phenotype to gain insights into RPS4/RRS1 gene pair function . Transcriptional profiling of the slh1 mutant shows that genes induced during RRS1SLH1-mediated defense activation in the absence of Avr overlap with those induced by AvrRps4- or PopP2-triggered immunity . Genetic screening for mutations that suppress slh1-triggered aberrant immunity reveals the critical role of RPS4 in RRS1SLH1-mediated activation of defense signaling . Transient expression of RPS4 and RRS1SLH1 in tobacco results in HR in the absence of AvrRps4 or PopP2 , which can be suppressed by co-expression of wild type RRS1-R , consistent with the recessive nature of RRS1SLH1 . Our study sheds new light on how paired R proteins work cooperatively and illustrates the similarities between auto-active and Avr-dependent defense signaling . To compare AvrRps4- or PopP2-triggered HR and immunity , we established the delivery of PopP2 via the Pseudomonas T3S . We engineered pEDV6 , a Gateway-compatible version of pEDV3 [33] , to carry full-length or N-terminally truncated PopP2 variants ( Figures 1A and S1A–B ) . pEDV6 enables expression of a translational fusion between the N-terminal part of AvrRps4 ( 137 first amino acids; hereafter , AvrRps4N ) and an effector of interest . We used a non-pathogenic Pseudomonas fluorescens Pf0-1 engineered to carry a functional T3S system ( hereafter , Pf0-1 ( T3S ) ) in HR assays because unlike Pseudomonas syringae pv . tomato ( Pto ) DC3000 , Pf0-1 ( T3S ) does not elicit non-specific tissue collapse . When delivered from Pf0-1 ( T3S ) or Pto DC3000 , PopP21–488 ( full-length ) or PopP2149–488 triggered HR and immunity in Arabidopsis accession Ws-2 , whereas the PopP2 variants that were further truncated did not ( Figure S1C–D ) . Interestingly , the N-terminal 148 amino acids of PopP2 that include a nuclear localization signal ( NLS ) are dispensable in our assay . Based on this finding , we used the PopP2149–488 ( hereafter , PopP2 ) variant for the rest of our experiments . To verify that Pseudomonas-delivered PopP2 confers genotype-specific avirulence , we investigated the responses of Arabidopsis natural variants to PopP2 . When delivered from Pf0-1 ( T3S ) , PopP2 and AvrRps4 triggered HR in accessions Nd-0 and Ws-2 whereas Col-0 and RLD showed no symptoms at 24 hours post-infection ( hpi ) ( Figure 1B ) . Col-0 RRS1-S confers HR-deficient disease resistance to Pst DC3000 delivered AvrRps4 but not to PopP2 [22] , [34] . In addition , transgenic expression of Ws-2 RRS1-R in Col-0 confers strong HR in response to Pseudomonas-delivered AvrRps4 [35] . HopA1 was used as an additional control; it triggers HR in Nd-0 , Ws-2 and RLD , but not in Col-0 , as expected . Next , we tested if Pf0-1 ( T3S ) -delivered PopP2 triggers RPS4- and RRS1-dependent HR in Arabidopsis . Pf0-1 ( T3S ) -delivered PopP2 triggered strong HR in wild type Ws-2 whereas Ws-2 rrs1-1 , rps4-21 , rrs1-1/rps4-21 or eds1-1 mutants did not show any response ( Figure 1C ) . In contrast , Pf0-1 ( T3S ) -delivered AvrRps4 triggered weak but robust HR even in the absence of RPS4 or RRS1 in Ws-2 ( Figure 1C ) . When delivered from Pto DC3000 , AvrRps4 triggered immunity in wild type Ws-2 , rrs1-1 , rps4-21 or rrs1-1/rps4-21 mutants because AvrRps4 recognition leads to RPS4/RRS1-dependent and -independent immunity ( Figure 1D ) [26] . To test if Pseudomonas-delivered PopP2 can trigger RPS4/RRS1-dependent immunity in Arabidopsis , we engineered a virulent Pto DC3000 to deliver PopP2 . Pto DC3000 ( PopP2 ) showed reduced virulence in wild type Ws-2 but not in rrs1-1 , rps4-21 or rrs1-1/rps4-21 mutants compared to Pto DC3000 ( pEDV5 ) indicating that Pseudomonas-delivered PopP2 triggers only RPS4/RRS1-dependent immunity ( Figure 1D ) , consistent with previously reported Ralstonia-delivery assay results [26] . By contrast , HopA1-triggered immunity was not affected in rrs1-1 , rps4-21 or rrs1-1/rps4-21 mutants compared with wild type Ws-2 ( Figure 1D ) . All tested Pto DC3000 strains showed unrestricted growth in the eds1-1 mutant compared to other genotypes . Taken together , these data indicate that AvrRps4N-mediated delivery of PopP2 from Pseudomonas can trigger RPS4/RRS1-dependent HR and immunity in Arabidopsis . We further tested if Pseudomonas-delivered PopP2 recognition requires a specific subcellular localization , as reported for AvrRps4 [8] . We engineered a PopP2149–488 variant lacking the native NLS , to carry a NLS or a nuclear export signal ( NES ) tag at the C-terminus . The avirulence activity of these PopP2 variants was tested in two resistant transgenic Arabidopsis lines , RLD ( RPS4Ler ) and Col-0 ( RRS1Ws-2 ) . Pf0-1 ( T3S ) -delivered PopP2NES , failed to trigger HR in both transgenic lines and in wild type Ws-2 , despite being expressed during plant infection , indicating that nuclear localization of PopP2 is required to trigger RPS4/RRS1-dependent HR ( Figure S2A , S2E and S3 ) . The PopP2NES variant induced a response comparable to PopP2C321A , an enzymatically inactive variant that does not trigger RPS4/RRS1-R-dependent immunity [36] in wild type Ws-2 when HR-inducing activity was quantified by ion leakage measurements ( Figure S2B ) . We could also show that PopP2NES , in contrast to PopP2NLS , could not restrict the virulence of bacteria when delivered from Pto DC3000 , nor trigger expression of defense genes when delivered from Pf0-1 ( T3S ) ( Figures S2C and S2D ) . As these data suggest that PopP2 triggers HR and immunity in the nucleus , we independently assessed previously reported AvrRps4 variants [8] . Unexpectedly , both AvrRps4NLS and AvrRps4NES variants triggered HR and elevated ion leakage in the Ws-2 accession when delivered from Pf0-1 ( T3S ) ( Figure S2B and S2E ) . RRS1 is a TIR-NB-LRR protein with a WRKY DNA binding domain , which belongs to Group III of the WRKY superfamily [37] . RRS1SLH1 , which carries a leucine insertion near the WRKY motif , shows strongly reduced DNA binding by its WRKY domain [30] . This reduced DNA binding correlates with auto-immunity of the slh1 mutant , suggesting a critical role of RRS1 in transcriptional regulation of defense genes . Delivery of PopP2 from Pseudomonas via T3S , combined with the RPS4/RRS1-R dependence of this PopP2-triggered HR , enables direct assessment of RRS1-R-dependent transcriptional regulation . To identify PopP2-triggered and RPS4/RRS1-dependent early transcriptional responses , genome-wide expression profiling was carried out using EXPRSS , an Illumina sequencing based method [38] . Wild type Ws-2 and rrs1-1 plants were infiltrated with Pf0-1 ( T3S ) delivering PopP2WT or PopP2C321A . The infiltrated leaf tissues were collected at 2 , 4 , 6 and 8 hpi for total RNA extraction , as onset of HR began at 8 hours after bacterial infiltration in an incompatible interaction ( PopP2WT/Ws-2 ) . For differential expression analysis , PopP2WT-infiltrated Ws-2 samples were compared either to PopP2C321A mutant on Ws-2 or PopP2WT on rrs1-1 . Essentially complete overlap was observed between the differentially regulated genes in the two comparisons ( Figure 2A ) , consistent with our results showing that Pf0-1 ( T3S ) -delivered PopP2 triggers RRS1- and acetyltransferase activity-dependent immunity ( Figures 1 and S2 ) . In total , 719 genes were differentially expressed in an RRS1- and acetyltransferase activity-dependent manner in at least one of the time points surveyed ( Table S1 ) . Gene ontology enrichment analysis using ATCOECIS [39] showed that most of the up-regulated genes are involved in defense , while most of the down-regulated genes are involved in photosynthesis and enriched in chloroplast-localized genes ( Table S2 ) . Interestingly , the majority of genes differentially expressed at 4 and 6 hpi were up-regulated , while many down-regulated genes were observed at 8 hpi ( Figure 2A ) . The early ( 4 and 6 hpi ) up-regulated genes , such as SID2 , FMO1 , NudT7 , PBS3 and PAD4 , have previously been implicated in plant defense ( Table S3 ) . Further analysis of mean expression of genes induced at 4 and 6 hpi ( Table S3 ) showed that there was greater gene induction in Ws-2 infiltrated with PopP2WT ( ∼20–100 fold ) than in Ws-2 infiltrated with PopP2C321A or in rrs1-1 infiltrated with PopP2WT ( ∼2–10 fold ) . For simplicity , we interpret genes induced by PopP2C321A as induced by the repertoire of PAMPs in Pf0 ( thus , PTI-induced ) , and by PopP2WT as PTI+ETI-induced . To validate our transcriptional expression profiling results , we performed quantitative RT-PCR ( qRT-PCR ) verification of EDS5 , NudT6 , WRKY18 and WRKY40 with the cDNA used for Illumina libraries . Expression of EDS5 and NudT6 but not WRKY18 and WRKY40 was specifically regulated by ETI in our expression profiling data . In qRT-PCR experiments , PopP2 but not PopP2C321A variant delivered from Pf0-1 ( T3S ) induced EDS5 and NudT6 in an RRS1-dependent manner , while expression of WRKY18 and WRKY40 was induced in the absence of ETI ( Figure S4 ) . AvrRps4- and PopP2-dependent transcriptional changes in resistant plants have been investigated previously [28] , [29] , [40] . We compared these available micro-array and RNA-seq data with our results . To minimize the effects of experimental and technical differences from the AvrRps4/Ws-2 data [28] , genes altered in expression at 6 hpi due to mock treatment were subtracted from the comparison; similarly , only the GMI1000/GMI1000ΔPopP2-infected Nd-1 data were used from the Hu et al . [40] study . For comparative analysis the differential expression from PTI , PTI+ETI and ETI responses were combined for data presented in this study ( Table S1 ) and the data from Howard et al . [29] . A summary of these comparisons is presented in Figure S5 and details of genes from comparative datasets are presented in Table S4 . Transcriptional changes upon AvrRps4 infection on Col-0 and Ws-2 [28] , [29] considerably overlapped with PopP2-regulated genes identified both in our study and the GMI1000/GMI1000ΔPopP2 study [40] ( Figure S5 ) . The majority of early PTI+ETI-induced genes detected in our study were also found to be AvrRps4-responsive [28] , [29] ( Figure S5 and Table S4 ) . We next tested the expression of four PopP2-responsive genes PBS3 , SARD1 , FMO1 and PR1 by qRT-PCR in Ws-2 , rps4-21 and eds1-1 . At 8 hpi , Pf0-1 ( T3S ) -delivered AvrRps4WT , HopA1 or PopP2WT triggered similar levels of induction of the four genes in Ws-2 ( Figure 2B ) . Induction of all four genes was strictly dependent on EDS1 and abolished when non-functional variants of the effectors ( AvrRps4KRVY-AAAA , AvrRps4E187A and PopP2C321A ) were delivered . PTI+ETI-induction of all four genes in response to PopP2 was reduced to PTI-induced levels in both rps4-21 and in rrs1-1 mutants , confirming RPS4/RRS1-R-dependence of PopP2-induced transcriptional changes . AvrRps4-triggered induction of all four genes was reduced but not abolished in the rps4-21 mutant , likely due to RPS4-independent recognition of AvrRps4 in Ws-2 [26] , [41] . These expression profiling data thus reveal the genes specifically regulated at very early stages of PopP2-triggered , RPS4/RRS1-dependent immunity in Arabidopsis . Moreover , these ETI transcriptional changes are very similar after AvrRps4 or PopP2 recognition . To compare slh1 aberrant defense responses to effector-triggered RPS4/RRS1-mediated immunity , we conducted transcription profiling of the slh1 mutant over a time course after shifting plants from 28°C to 19°C , using Illumina tag sequencing [38] . A total of 1821 genes showed temperature-dependent differential expression in RRS1SLH1 after 24 hours ( h ) compared to wild type No-0 ( Figure 3A ) . We confirmed the temperature-dependent regulation of 3 genes with differential induction in slh1 by qRT-PCR . PR1 , PBS3 and CBP60g transcript accumulation was induced in slh1 plants between 9 and 24 h after the shift from 28°C to 19°C whereas it was unaltered in temperature-shifted No-0 plants ( Figure 3B ) . We compared the slh1/No-0 temperature-shift transcriptional dataset to the PopP2/RRS1-time course dataset by analyzing the pairwise overlap of genes differentially expressed in both experiments ( Figure 4 ) . Each time course response was categorized according to the mode of elicitation as PTI , ETI , temperature shift , auto-immunity , or corresponding combinations ( e . g . PTI+ETI ) . We found that most ( ∼83% ) of the PopP2/RRS1 ETI genes were differentially expressed in slh1 auto-immune and temperature shift responses , while up to 54% of ETI genes were also differentially expressed in the auto-immune response but not by temperature shift ( Figure 4 , black box ) . Similarly , more than 55% of auto-immune genes were also differentially expressed in PTI and PTI+ETI ( Figure 4 , dotted block box ) . Most ETI genes were also differentially expressed in PTI+ETI ( more than 85% ) and in PTI ( up to 70% ) . However , less than 10% of the PTI genes were differentially expressed during ETI ( Figure 4 , blue box ) . This strongly suggests that many ETI responses involve potentiation of a subset of PTI responses , with few genes solely regulated by effector recognition . The ETI-specific genes that are regulated in PopP2 acetyltransferase activity- and RRS1-dependent manner include nucleotide/ATP-binding protein encoding genes such as NB-LRRs ( Table S1 ) . Similarly , we found that most temperature shift-regulated genes ( up to 83% ) ( Table S5 ) were also differentially expressed by PTI or PTI+ETI , but only 25% were specifically affected by ETI , and less than 5% of the PTI-responsive genes were differentially expressed by temperature shift ( Figure 4 , green box ) . Up to 50% of PTI or PTI+ETI genes were also differentially expressed by temperature shift and auto-immune response , while about 25% of PTI or PTI+ETI genes were differentially expressed by auto-immune response ( Figure 4 , green box ) . These results indicate that PTI broadly activates genes responsive to heat , auto immunity and ETI . These analyses indicate that slh1 auto-immunity overlaps strongly with PopP2- and RPS4/RRS1-R-dependent ETI . Thus , RRS1SLH1-induced transcriptional reprogramming results in similar gene expression changes to those observed in AvrRps4- or PopP2-triggered immunity , indicating that the slh1 lethal phenotype mimics RPS4/RRS1-dependent ETI at the transcriptional level . Lethality of slh1 at 21°C is correlated with constitutive activation of defense responses including high expression of Pathogenesis Related ( PR ) genes and SA accumulation [30] . We hypothesized that mutations that affect RRS1SLH1-mediated signaling components or RRS1SLH1 expression would suppress slh1 lethality . To identify genetic components required for RRS1SLH1-dependent immunity , we conducted a suppressor screen . slh1 seeds were incubated with ethyl methanesulfonate ( EMS ) , ∼7 , 000 M1 plants were grown at 28°C and M2 seeds were harvested . By screening ∼500 , 000 M2 mutant plants at 21°C , we identified 83 families with a suppressor of slh1 immunity ( sushi ) mutant phenotype . Among them , 69 and 14 could rescue the slh1 lethal phenotype to a wild type-like and an improved morphology , respectively . We further analyzed the progeny of 7 selected fully rescued sushi mutants for morphological development and defense marker gene expression in the M3 generation ( Figure 5 ) . Growth of sushi mutants at 21°C was similar to wild type No-0 , whereas slh1 plants did not develop beyond the first true leaf stage ( Figure 5A ) . PR1 , PBS3 and FMO1 expression was elevated in slh1 mutants grown constantly at 21°C or 24 h after shift from 28°C to 21°C , but not in fully rescued sushi mutants ( Figures 5B and S6 ) . To exclude any contamination with wild type seeds , we confirmed the presence of the slh1 mutation in 72 of the 83 M3 individual sushi mutants identified using a cleaved amplified polymorphic sequences ( CAPS ) marker [30] . Next , we carried out Sanger sequencing of RRS1 and RPS4 coding regions in these mutants . As expected from the complete suppression of the slh1 phenotype , we identified 6 sushi intragenic suppressor mutants that carry an early stop codon in RRS1SLH1 and 8 other non-synonymous mutations ( Table S6 ) . Surprisingly , non-synonymous mutations were also identified throughout the RPS4 coding region in 34 rescued sushi mutants ( Table S6 ) . Most of the altered amino acid residues have not previously been shown to be required for RPS4 function [24] . However , sushi52 and sushi22 harbour non-synonymous mutations at positions R28 and E88 that are important for RPS4TIR+80-triggered HR in tobacco [42] , further verifying the crucial role of the RPS4 TIR domain function in RRS1SLH1-mediated defense activation . It was previously reported that mutations in SID2/ICS1/EDS16 or SID1/EDS5 result in suppression of the RRS1SLH1 mutant phenotype [30] . We sequenced the coding region of these genes in the non-RRS1 , non-RPS4 mutants , and found one sushi mutant that carried a mutation in SID2/ICS1/EDS16 ( sushi70 , Table S6 ) , and no mutants that carry mutations in SID1/EDS5 . Similarly to Arabidopsis accession Col-0 , wild type No-0 carries two copies of EDS1 . Therefore , EDS1 coding sequence was not verified in the sushi lines . The 23 remaining unassigned SUSHI mutations are now subjected to further analysis to identify new signaling components of RRS1SLH1-mediated immunity . Homo- or hemizygous , but not heterozygous , No-0 plants carrying RRS1SLH1 display a stunted phenotype at 21°C due to elevated immunity [30] . To verify that RPS4 is required for RRS1SLH1 function , we crossed 7 sushi lines carrying mutations in RPS4 ( sushi17 , 64 , 24 , 12 , 41 , 58 and 32 ) to rrs1-1 and rrs1-1 rps4-21 knockout mutants [26] . The resulting F1 individuals from both crosses were hemizygous RRS1SLH1/rrs1 for RRS1 locus ( Figure S7 ) and either RPS4sushi/RPS4WT or RPS4sushi/rps4 at the RPS4 locus . As expected , the F1 plants derived from a cross between the sushi and rrs1-1 were stunted and showed elevated PR1 expression level ( Figure 6A–C ) . These phenotypes were both completely suppressed in the F1 plants derived from a cross between sushi mutants in RPS4 and rrs1-1 rps4-21 double mutant . This result confirms that RPS4 is required for RRS1SLH1-mediated activation of immunity . To further verify the functional requirement for RPS4 in RRS1SLH1-mediated immunity , we recapitulated RRS1SLH1-mediated defense activation in Nicotiana tabacum . As shown recently [39] , Agrobacterium-mediated transient co-transformation ( hereafter , agroinfiltration ) of RPS4-HA , RRS1-His-Flag and wild type AvrRps4-GFP or PopP2-GFP induced strong HR within 3 dpi ( Figure S8A ) . The specificity of recognition was further verified by comparing functionally characterized mutant variants of AvrRps4 or PopP2 to wild type . As expected , AvrRps4E187A , AvrRps4KRVY-AAAA and PopP2C321A variants did not induce RPS4/RRS1-dependent HR in tobacco ( Figure S8A ) . We have also verified that AvrRps4 and PopP2 recognition in tobacco activate defense genes orthologous to those that are regulated by RRS1 in Arabidopsis . The transcripts of the defense genes NtWRKY51 and NtNudT7 were highly up regulated when PopP2-GFP was co-expressed with RPS4-HA and RRS1-His-Flag in tobacco ( Figure S8B ) . Agroinfiltration of GFP or PopP2C321A-GFP with RPS4-HA and RRS1-His-Flag induced significantly lower accumulation of defense gene transcripts compared to wild type PopP2 ( Figure S8B ) . Taken together , these results further demonstrate that our transient agroinfiltration assay can also be used to investigate RPS4/RRS1 regulated immunity . Agroinfiltration of epitope-tagged RRS1SLH1-His-Flag and RPS4WT-HA triggered HR in tobacco leaf cells , whereas RRS1SLH1 co-expressed with GFP or RPS4K242A ( P-loop mutant ) did not ( Figures 6D and 8B ) . Consistent with our Arabidopsis genetic data ( Figure 6B ) , agroinfiltration of RRS1SLH1 with each RPS4SUSHI variant did not trigger HR in tobacco ( Figure 6D ) . Protein accumulation of the 7 tested RPS4SUSHI variants was comparable to that of RPS4WT , indicating that the lack of HR was not due to low protein expression levels ( Figure S9 ) . Moreover , as expected from our genetic analysis , RPS4SUSHI variants did not have a dominant negative effect on RPS4WT function , when both were co-expressed with RRS1SLH1 ( Figure S10 ) . We then tested whether SUSHI mutant alleles of RPS4 confer RRS1-dependent recognition of AvrRps4 or PopP2 . Agroinfiltration of RRS1WT , RPS4WT and either AvrRps4 or PopP2 , triggered RPS4 P-loop-dependent HR in infiltrated tobacco leaf sectors [43] ( Figure 6D ) . Importantly , agroinfiltration of the 7 RPS4SUSHI variants did not confer responsiveness to AvrRps4 or PopP2 ( Figure 6D ) . Taken together , these data show that RPS4 is required for RRS1SLH1-mediated and Avr-triggered/RRS1-dependent defense signaling activation . Recently , we showed the physical interaction of RRS1 and RPS4 [43] . We hypothesized that RPS4SUSHI variants may have lost their ability to interact with RRS1SLH1 . However , RPS4SUSHI-HA variants and RPS4WT-HA were co-immunoprecipitated by RRS1SLH1-Flag or RRS1WT-Flag ( Figure S9A-B ) . This result suggests that RPS4-RRS1 interaction is insufficient for signaling activation . We identified six additional sushi mutants that carry mutations in the TIR domain of RPS4 , the structure of which is known [43] . The stunted growth and elevated defense transcript accumulation of slh1 at 21°C were considerably suppressed in sushi52 ( R28H ) , 14 ( A38V ) , 22 ( E88K ) , 71 ( L101F ) , 89 ( P105L ) and 29 ( G120R ) ( Figure S11 ) . The RPS4 TIR domain structure suggests that side-chains from R28 and A38 are surface exposed , while the side-chains of the other mutated residues are buried ( Figure 7A ) . RPS4TIR expression is sufficient to trigger HR in tobacco after agroinfiltration ( Figure 7B ) [42] . Therefore we introduced these six SUSHI mutations into an RPS4TIR construct ( amino acids 1 to 250 ) to test their individual effect on RPS4 TIR domain signaling . Strikingly , all six mutations suppressed this response , suggesting that each of these residues is important for RPS4 TIR domain defense activation either through interaction with downstream partners or by maintaining the correct signalling-competent structural conformation , as the protein stability/accumulation was not significantly altered when expressed as GFP fusions in tobacco ( Figure S12D ) . Intriguingly , when SUSHI mutations were tested in the RPS4 full-length context by co-expression in tobacco with RRS1 and the effectors , A38V and L101F did not suppress RRS1SLH1- nor AvrRps4- and PopP2-triggered HR ( Figure 7C ) . This discrepancy was not due to inconsistent level of protein accumulation ( Figure S12E ) but might illustrate a limitation of the transient expression system in tobacco , or subtle differences between defense activation by RPS4TIR , and by the activated RPS4/RRS1 complex . As nuclear localization of RPS4 is necessary for AvrRps4-triggered immunity [41] , we investigated the role of RPS4 nuclear localization in RRS1SLH1-mediated cell death . Co-expression of RRS1SLH1 with RPS4WT or RPS4NLS induced HR ( Figure 8A ) . However , RPS4NES did not induce HR when co-expressed with RRS1SLH1 , indicating the importance of RPS4 nuclear localization for RRS1SLH1 function , consistent with a previous report [41] . Nucleotide binding to the invariant Lys residue of the P-loop motif in the NB domain of R proteins is critical for conformational change and immunity activation [4] , [44] , [45] . Agroinfiltration of RPS4WT , but not the P-loop mutant RPS4K242A , triggered HR when co-expressed with RRS1SLH1 ( Figure 8B ) . However , RPS4K242A does interact with RRS1SLH1 and RRS1WT ( Figure S9C ) . Therefore , a functional RPS4 P-loop motif is required for activation of RRS1SLH1-induced defense but is not an absolute requirement for RPS4-RRS1 interaction . Surprisingly , introduction of the P-loop mutation ( K185A ) in the RRS1SLH1 protein sequence did not affect HR-inducing activity when co-expressed with RPS4WT ( Figure 8B ) . Thus , P-loop motif-dependent conformational change may not be required for defense activation by RRS1SLH1 , consistent with the functionality of an RRS1 P-loop mutant in AvrRps4 or PopP2 recognition [43] . Structural analysis of RPS4 and RRS1 TIR domains revealed an “SH motif” in regions that mediate heterodimerization between RPS4 ( S33 and H34 ) and RRS1 ( S25 and H26 ) [43] . Moreover , RPS4 or RRS1 variants carrying a mutated SH motif ( SH-AA ) cannot recognize AvrRps4 or PopP2 in tobacco agroinfiltration [43] . To investigate if TIR-TIR domain heterodimerization is also required for RRS1SLH1 function , SH-AA mutations were introduced in RPS4WT and RRS1SLH1 variants . Agroinfiltration of RRS1SLH1 and RPS4SH-AA , or RRS1SLH1/SH-AA and RPS4WT did not induce HR in tobacco suggesting that TIR-TIR domain heterodimerization between RRS1 and RPS4 is required for RRS1SLH1-dependent defense activation ( Figure 8B ) . However , in the context of the full-length proteins the RRS1SLH1/SH-AA variant could still interact with RPS4WT ( Figure S9C ) . RRS1SLH1-dependent lethality is recessive [30] . In agreement , agroinfiltration of RRS1WT but not of GFP interfered with HR induced by co-expression of RRS1SLH1 and RPS4WT in tobacco ( Figure 8C–D ) . Interestingly , the RRS1K185A variant did not interfere with RRS1SLH1-induced HR whereas the RRS1SH-AA variant did ( Figure 8C ) , indicating that nucleotide-binding function but not RPS4/RRS1 TIR-TIR domain interaction is required for RRS1-mediated interference with RRS1SLH1-induced HR . These agroinfiltration data are consistent with our transcriptomic and genetic analyses and demonstrate the striking similarity of RRS1SLH1 and Avr-triggered/RRS1-dependent defense activation . As RRS1SLH1/RPS4-dependent constitutive HR is prevented by co-expression of RRS1WT , we tested if RRS1SLH1 interferes with RRS1WT recognition of AvrRps4 or PopP2 . Interestingly , in the presence of both RRS1 variants and RPS4 , AvrRps4- or PopP2-triggered HR is still observed suggesting that RRS1SLH1 did not completely abolish RRS1WT function ( Figure 8D ) . However , AvrRps4-triggered HR was attenuated considerably compared to PopP2-triggered HR under the same conditions ( Figure 8D ) . Although both AvrRps4 and PopP2 are recognized by RPS4 and RRS1 , a thorough comparison of immune responses , particularly of early transcriptional changes , has been difficult due to the distinct infection modes of the bacterial pathogens from which AvrRps4 ( Pseudomonas syringae ) and PopP2 ( Ralstonia solanacearum ) originate . Root infection of Arabidopsis plants with R . solanacearum causes wilting within 2 weeks , whereas Pseudomonas-delivered AvrRps4 triggers HR in Arabidopsis Ws-2 leaf cells within 24 hours . PopP2 delivery from Pf0-1 ( T3S ) allowed us to compare the transcriptional reprogramming caused by recognition of AvrRps4 or PopP2 at the earliest stages and has resulted in the identification of a set of similarly regulated ETI-specific genes . It is interesting that the NLS is dispensable for the avirulence activity of PopP2 in our assays . It was shown that removal of the N-terminal NLS renders localization of PopP2 and co-expressed RRS1-S/R variants nuclear-cytoplasmic [46] . However , the significance of this PopP2 NLS-dependent relocalization of RRS1 is not known , as there have been no studies showing ETI phenotypes triggered by PopP2 lacking the NLS . As shown in Figure S2 , a PopP2 variant lacking an N-terminal NLS shows similar levels of avirulence compared to wild type . Thus , PopP2 NLS-dependent relocalization of RRS1 may not be significant in PopP2-triggered immunity . Alternatively , the portion of RRS1 that is localized in the nucleus with the NLS lacking PopP2 might be sufficient to activate ETI . It is intriguing to find that AvrRps4NES and AvrRps4NLS are comparable in their ability to elicit HR in Arabidopsis Ws-2 ( Figure S2E ) . AvrRps4NES triggers a slightly lower ion leakage level than AvrRps4NLS ( Figure S2C ) . We conclude that regardless of AvrRps4 contribution to defense activation in the cytoplasm , its major role is in the nucleus via interactions with the RPS4/RRS1 complex . Pseudomonas T3S delivery of PopP2 provides a useful tool to investigate RPS4/RRS1-dependent transcriptional regulation at an early stage of ETI . In addition , by comparing non-functional variants of AvrRps4 and PopP2 to wild type proteins , we could identify the genes whose transcriptional changes were specific to Avr function . As Pf0-1 ( T3S ) carries a mutated HopA1 gene which is unable to trigger RPS6-dependent immunity in Arabidopsis , the gene expression change in rrs1-1 infiltrated with PopP2WT or in Ws-2 infiltrated with PopP2C321A can be considered as PTI resulting from perception of the Pf0-1 PAMP repertoire . We thus report defense gene expression changes as PTI vs . PTI+ETI ( Table S3 ) . Gene ontology enrichment has shown that the majority of early up-regulated genes are involved in plant defense . Comparative analysis with previously published microarray data shows that many PopP2-triggered early gene expression changes overlap substantially with AvrRps4-triggered transcriptional regulation [28] , [29] . It is interesting to note that PopP2-regulated genes also overlap substantially with previously reported PopP2-induced genes at a later stage of infection when delivered from R . solanacearum [40] . Our discovery of early responding genes will allow us to test if they are directly regulated by RPS4/RRS1 . It has been recently shown that WRKY18 and WRKY40 positively contribute towards AvrRps4-triggered immunity [47] . Consistent with this , WRKY18 and WRKY40 were highly induced at 3 and 6 hpi by AvrRps4 ( Table S4 ) . However , our experimental design enabled us to show that both WRKY18 and WRKY40 are primarily induced due to PTI ( Figure S4 ) . PTI+ETI and PTI induction of WRKY40 expression are indistinguishable . There is slightly higher PTI+ETI-induced expression of WRKY18 in response to PopP2WT in Ws-2 at later time points ( 6 and 8 hpi ) compared to PTI elicited by PopP2C321A in Ws-2 or PopP2WT in rrs1-1 ( Figure S4 ) , but this could be due to elevated SA levels that we presume are responsible for strong PR1 induction at 8 hpi . It is interesting to note that AvrRps4-induced regulation of ETI genes only partially requires RPS4 . This is consistent with AvrRps4 recognition being conferred by both RPS4/RRS1-dependent and -independent mechanisms . Identification of an R gene ( s ) that confer RPS4/RRS1-independent immunity will enable comparative analysis of how AvrRps4-induced ETI genes are transcriptionally regulated by different R genes . It was remarkable to observe that AvrRps4 , PopP2 and HopA1 induced common genes at early stage of defense activation , suggesting a possible EDS1-dependent conserved gene activation mechanism in ETI . Several auto-active alleles of NB-LRR genes have been found [9] , [10] , [30] , [48] , [49] , though unlike the recessive slh1 , all others are dominant or semi-dominant . Plants carrying an auto-active R gene typically show temperature-dependent lethality and enhanced resistance to virulent pathogens [30]–[32] . However , in many cases the overlap between elevated disease resistance that is conferred by an auto-active R gene allele and by Avr-triggered immunity is poorly defined . Unlike most other auto-active R gene alleles , RRS1SLH1 carries a single amino acid insertion in the WRKY-DNA binding domain that reduces its DNA-binding affinity [30] . To address the role of RRS1 in transcriptional activation or repression , we tested whether RRS1SLH1-induced transcription changes overlap with AvrRps4- or PopP2-triggered transcription changes . Based on previously reported expression profiling data and the present study , we propose that the genes whose transcripts were differentially regulated by RRS1SLH1 , and by AvrRps4 and PopP2 are directly regulated by RRS1 upon Avr detection . As exons 6 and 7 of RRS1SLH1 show reduced binding to a W-box in vitro , RRS1 may act as a transcriptional repressor of plant immunity , or at least as a repressor of RPS4 function , and this repression may be relieved upon Avr perception [30] . However , RRS1 could act both as repressor and activator of defense gene transcription , as has been found for other plant transcription factors [50] . Loss of RRS1-DNA binding may be part of the activation of defense transcription , but paradoxically , rrs1 knockout lines do not show enhanced immunity . Identification of RPS4 mutant alleles among the SUSHI mutations was unexpected , as we had anticipated that RRS1 might act downstream of RPS4 to regulate defense gene transcription directly . Notably , it would have been difficult to recover recombinants between RRS1SLH1 and an adjacent mutant allele of RPS4 , so without a genetic screen , this discovery might not have been made . Based on the genetic requirement of RPS4 for RRS1SLH1-induced defense gene transcription , we now hypothesize that RPS4 is required to form a functional immune receptor complex with RRS1 . This hypothesis is further supported by the fact that RPS4 and RRS1 interact with each other , in part but not solely by forming a TIR-TIR domain heterodimer [43] . In addition , the requirement of a functional P-loop motif for RPS4 but not for RRS1 function suggests that RPS4 contributes to defense activation by providing ATP-dependent conversion of a repressive immune receptor complex to an activated state . PopP2 interacts with RRS1 [46] , as does AvrRps4 [43] . We hypothesize that RPS4 activates defense upon recognition of perturbations in RRS1 by effectors , and that RRS1SLH1 mimics the results of effector action upon RRS1 . Can this be reconciled with the observation that a 35S:RPS4 constitutive defense phenotype partially requires RRS1 [51] ? Conceivably , RRS1 might also play a chaperone-like role in facilitating conversion of RPS4 from an inactive to an active form , and RRS1SLH1 has enhanced activity in facilitating this conversion . The TIR domain of RPS4 induces cell death when transiently overexpressed in tobacco . Several amino acid residues were shown to be required for RPS4 TIR domain auto-activity [42] . Among the 33 single amino acid polymorphisms of RPS4 that we identified in sushi mutants , two residues , R28 and E88 , were previously implicated as being required for RPS4 TIR domain-induced auto-activity in tobacco . R28H and E88K mutations are unlikely to alter the overall structure of RPS4 TIR domain , judging from the crystal structure of RPS4/RRS1 TIR domain heterodimer [43] . A study on RPS4 natural variants identified Y950 as an important residue for function as a susceptible RLD allele of RPS4 carries a Y950H mutation , and a Y950H substitution in the functional Ler allele of RPS4 abolishes its AvrRps4-recognition capability [24] . Interestingly , we identified several mutations ( S914F , G952E and G997E ) in this C-terminal domain ( CTD ) of RPS4 . Although the function of the RPS4 CTD remains unclear , it appears to be important for immune signaling . Conceivably , the sushi-mutated residues found in the TIR domain ( R28 , E88 , P105L and G120R ) and in the CTD ( S914F , G952E , and G997E ) are involved in the interaction with RRS1 or other yet unknown partner ( s ) . AvrRps4 and PopP2 interact directly with RRS1 [43] , [46] . Conceivably , after interaction of AvrRps4 or PopP2 with RRS1 , dissociation of the activated RPS4/RRS1 immune complex from target DNA induces RPS4 P-loop-dependent de-repression/activation of defense gene transcription , perhaps via WRKY18 and WRKY40 [47] . There may be multiple WRKY transcription factors that can replace the transcriptional repression function of RRS1 , but not its Avr-recognition function . However , the Ws-2 RRS1SLH1 allele may make additional contributions to assembling a defense-activating complex beyond vacating W-boxes . An intriguing feature of RRS1 is that it is the only known recessive NB-LRR-encoding R gene . Consistent with this observation , the slh1 mutation is also recessive . We were able to recapitulate this feature by transiently co-expressing RRS1 with RPS4 and RRS1SLH1 and suppressing RPS4/RRS1SLH1-triggered HR . This suppression is abolished if the RRS1-R carries a mutation in its P-loop motif . Intriguingly , this result suggests that the RRS1-R P-loop is not required for RPS4-dependent HR activation , but potentiates assembly of an inactive , poised complex . Thus , we suggest that the recessive nature of RRS1 in the Col-0 ( S ) /Nd-0 ( R ) or Col-0 ( S ) /Ws-2 ( R ) cross is the result of the Col-0 allele encoding a protein that can interfere in trans with PopP2 responsiveness and thus acts as a “poison subunit” . There are nine TIR-NB-LRR gene pairs reported in the Arabidopsis Col-0 genome [26] . It is important to better understand how paired R proteins have evolved and recognize effectors . It is interesting to note that all three TIR-NB-LRR-WRKY encoding genes ( At5g45260 , At5g45050 and At4g12020 ) found in Arabidopsis are paired with TIR-NB-LRR genes [26] . This suggests that at least some other paired R proteins may function cooperatively in the nucleus by directly regulating transcriptional processes . In conclusion , the deployment of a Pseudomonas T3S delivery of PopP2 allowed a detailed comparison of AvrRps4- and PopP2-triggered RPS4- and RRS1-dependent transcriptional regulation . We found that an auto-active allele of the TIR-NB-LRR-WRKY protein RRS1 , RRS1SLH1 , induces immune responses comparable to Avr-triggered immunity . The suppressor of slh1 immunity screening enabled us to uncover the critical role of RPS4 in RRS1SLH1-mediated defense activation . Furthermore , we defined additional properties of RPS4 and RRS1 that are essential for function , and these results significantly enhance our understanding of NB-LRR protein function in plants . Arabidopsis plants were grown in short day conditions ( 10 h light/14 h dark ) at 21°C or 28°C . Nicotiana benthamiana and Nicotiana tabacum ‘Petit Gerard’ plants were grown in long day conditions ( 16 h light/8 h dark ) at 24°C . No-0 and slh1 are described in [30]; Ws rrs1-1 and Ws rrs1-1 rps4-21 are described in [26] . To create pEDV6 ( gateway destination variant of pEDV3 ) , the nucleotide sequence encoding the HA epitope tag was inserted at SalI site of pEDV3 [33] that resulted in AvrRps4N ( 1-137aa ) :HA:ClaI:BamHI ( pEDV5 ) . Subsequently , pEDV5 was digested with ClaI and BamHI , treated with T4 DNA-polymerase to generate blunt ends and ligated with EcoRV digested Gateway reading frame cassette B ( RFB ) ( Invitrogen ) to create pEDV6 . Construction of pBBR1MCS-5:avrRps4 was described previously [35] . The NES- or NLS-tagged avrRps4 variants were kindly provided by Jane Parker laboratory and the cloning procedure was described previously [8] . To generate pEDV6:popP2 variants , full-length or truncated popP2 fragments were amplified from Ralstonia solanacearum genomic DNA by polymerase chain reaction and cloned in the Gateway entry vector , pCR8 ( Invitrogen ) . Introduction of popP2 fragments in pEDV6 was performed according to manufacturer's instructions by using LR recombinase II ( Invitrogen ) . The pBin19:RPS4:HA construct was described previously [52] . To obtain C-terminally GFP tagged AvrRps4 or PopP2 variants , avrRps4 or popP2 coding regions were PCR amplified and cloned at ClaI and BamHI sites of EpiGreenB5:GFP . Construction of 35S:RRS1:His-Flag is described in [43] . Wild type and mutant variants of AvrRps4 and PopP2 were PCR amplified from previously reported plasmid constructs [35] , [53] . The resulting PCR fragments were cloned in pCR8 ( Invitrogen ) and correct sequences were confirmed . These pCR8 constructs were used for LR reaction with the Gateway destination vector pK7FWG2 ( 35S promoter and C-GFP ) to generate C-terminally GFP-tagged AvrRps4 and PopP2 variants . Wild type and SH-AA mutant variants of RPS4-HA and RRS1-His-Flag are described in [43] . Introduction of SLH1 and SUSHI mutations in RRS1 and RPS4 , respectively , was achieved by using Quikchange II XL site-directed mutagenesis kit ( Agilent ) . The C-terminally GFP-tagged RPS4 constructs were generated by inserting ClaI/BamHI digested RPS4 in EpiGreenB5-GFP-WT/NES/NLS . Escherichia coli DH5α was used for maintaining plasmid constructs and bacterial conjugation . For hypersensitive response assay and in planta bacterial growth assay , Pseudomonas fluorescens Pf0-1 ( T3S ) and Pseudomonas syringae pv . tomato ( Pto ) DC3000 strains were used , respectively . To introduce various constructs carrying avrRps4 , popP2 or hopA1 in Pf0-1 ( T3S ) and Pto DC3000 , standard triparental mating method was used by using E . coli HB101 ( pRK2013 ) as a helper strain as previously described [33] . For transformation of Agrobacterium tumefaciens strain AGL1 , standard electroporation method was used . For hypersensitive response assay , freshly grown Pf0-1 ( T3S ) strains on King's B agar plates containing appropriate antibiotics were harvested in 10 mM MgCl2 . The final concentration of bacterial suspensions was adjusted to A600 = 0 . 2 . Leaves of five week-old Arabidopsis plants were hand-infiltrated by using 1 mL needless syringes and kept 20–24 h further for symptom development . For ion leakage assays , leaf discs were sampled at 0 . 5 hpi , floated on water for 30 minutes ( with gentle shaking at room temperature ) and transferred to fresh water ( 1 hpi sample ) . Ion leakage was measured at 24 hpi using a conductivity meter . For in planta bacterial growth assays , Pto DC3000 strains were grown and harvested as for Pf0-1 ( T3S ) . Leaves of five week-old Arabidopsis plants were hand-infiltrated with bacterial suspensions ( A600 = 0 . 001 ) by using 1 mL needless syringes and kept 3–4 days further before sampling . Infected leaf samples were ground in 10 mM MgCl2 , serially diluted , spotted on L agar plates containing appropriate antibiotics and kept at 28°C for 2 days before counting colonies to estimate bacterial population in infected leaves . Agrobacterium tumefaciens AGL1 strains carrying the different constructs were grown in liquid L-medium supplemented with adequate antibiotics for 24 h . Cells were harvested by centrifugation and re-suspended at OD600 0 . 5 in infiltration medium ( 10 mM MgCl2 , 10 mM MES pH 5 . 6 ) . For co-expression , bacterial suspensions were mixed in 1∶1 ratio before infiltration with needleless syringes in 5 week-old N . benthamiana or N . tabacum leaves . Tobacco hypersensitive response was generally observed and photographed 2 to 3 days after infiltration . EXPRSS tag-seq cDNA library construction and data analysis was carried out as described previously [38] . The sequence data presented in this publication have been deposited in NCBI's Gene Expression Omnibus [54] and are accessible through GEO Series accession number GSE48247 and GSE51116 . Tag to gene associations were carried out using uniquely mapped reads , with the considerations described previously [38] . Bowtie v0 . 12 . 8 [55] was used to map short reads to TAIR10 genome and Novoalign v2 . 08 . 03 ( http://www . novocraft . com/ ) was used to align remaining reads to TAIR10 cDNA sequences . Differential gene expression analysis was performed using the R statistical language ( v2 . 11 . 1 ) with the Bioconductor package [56] , edgeR v1 . 6 . 15 [57] with the exact negative binomial test using tagwise dispersion and selected genes with false discovery rate ( FDR ) <0 . 01 . From RNA-seq data for avrRps4 on Col-0 [29] , uniquely mapped read counts to genes were used for reanalysis using edgeR and selected gene with FDR <0 . 05 . Microarray data files from Pto DC3000 ( AvrRps4 ) infiltration ( Array Express E-MEXP-546 , [28] ) and Interaction of Arabidopsis thaliana and Ralstonia solanacearum ( NASCARRAYS-447 , [40] ) were used . Data analysis was performed using the R statistical language as described previously [38] , [58] . Differentially expressed genes were identified using the rank products method with FDR <0 . 05 [59] . As Pto DC3000 ( AvrRps4 ) data has only one replicate , differential expression analysis was carried out with untreated and MgCl2 infiltrated 3 hpi samples as controls and compared against 3 and 6 hpi of avrRps4 and 6 hpi of MgCl2 . For GMI1000/GMI1000ΔPopP2 data , only Nd-1 samples were used . Total RNAs were extracted from 4 to 5 week-old Arabidopsis plants using the TRI reagent ( Invitrogen ) according to the manufacturer's instructions . First-strand cDNA was synthesized from 5 µg RNA using SuperScriptII Reverse Transcriptase ( Invitrogen ) and an oligo ( dT ) primer , according to the manufacturer's instructions . cDNA was amplified in triplicate by quantitative PCR using SYBR Green JumpStart Taq ReadyMix ( Sigma ) and the CFX96 Thermal Cycler ( Bio-Rad ) . The relative expression values were determined using the comparative Ct method and Ef1α ( At5g60390 ) as reference . Primers used for quantitative PCR are described in Table S7 . The presence of the slh1 mutation in sushi M3 generation and F1 individuals resulting from the genetic cross with rrs1-1 or rrs1-1 rps4-21 was assessed using the CAPS marker described in [30] . For sequencing of candidate genes on sushi mutants genomic DNA , 10 , 6 , 4 and 4 couples of primers respectively were used to amplify regions of RRS1 , RPS4 , EDS16 and EDS5 coding sequence ( see Table S7 ) . PCR products were purified on Sepharose and sequences were analyzed using the Vector NTI assembly software ( Invitrogen ) . Protein samples were prepared from N . benthamiana 48 h after Agrobacterium-mediated transformation . One infiltrated leaf was harvested and ground in liquid nitrogen . Total proteins were extracted in GTEN buffer ( 10% glycerol , 100 mM Tris-HCl pH 7 . 5 , 1 mM EDTA , 150 mM NaCl ) supplemented extemporaneously with 5 mM DTT , 1% ( vol/vol ) plant protease inhibitor cocktail ( Sigma ) and 0 . 2% ( vol/vol ) Nonidet P-40 . Lysates were centrifuged for 15 min at 5 , 000 g at 4°C and aliquots of filtered supernatants were used as input samples . Immunoprecipitations were conducted on 1 . 5 mL of filtered extract incubated for 2 h at 4°C under gentle agitation in presence of 20 µL anti-FLAG M2 or EZview anti-HA affinity gel ( Sigma ) . Antibodies-coupled agarose beads were collected and washed three times in GTEN buffer , re-suspended in SDS-loading buffer and denatured 10 min at 96°C . Proteins were separated by SDS-PAGE and analyzed by immunoblotting using anti-FLAG M2-HRP , anti-GFP-HRP or anti-HA-HRP conjugated antibodies ( Sigma , Santa Cruz and Roche respectively ) .	How plant NB-LRR resistance proteins and the related mammalian Nod-like receptors ( NLRs ) activate defense is poorly understood . Plant and animal immune receptors can function in pairs . Two Arabidopsis nuclear immune receptors , RPS4 and RRS1 , confer recognition of the unrelated bacterial effectors , AvrRps4 and PopP2 , and activate defense . Using delivery of PopP2 into Arabidopsis leaf cells via Pseudomonas type III secretion , we define early transcriptional changes upon RPS4/RRS1-dependent PopP2 recognition . We show an auto-active allele of RRS1 , RRS1SLH1 , triggers transcriptional reprogramming of defense genes that are also reprogrammed by AvrRps4 or PopP2 in an RPS4/RRS1-dependent manner . To discover genetic requirements for RRS1SLH1 auto-activation , we conducted a suppressor screen . Many suppressor of slh1 immunity ( sushi ) mutants that are impaired in RRS1SLH1-mediated auto-activation carry loss-of-function mutations in RPS4 . This suggests that RPS4 functions as a signaling component together with or downstream of RRS1-activated immunity , in contrast to earlier hypotheses , significantly advancing our understanding of how immune receptors activate defense in plants .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "mutation", "plant", "science", "genetic", "screens", "gene", "identification", "and", "analysis", "gene", "expression", "genetics", "plant", "genetics", "biology", "and", "life", "sciences", "molecular", "genetics", "gene", "function" ]	2014	The Nuclear Immune Receptor RPS4 Is Required for RRS1SLH1-Dependent Constitutive Defense Activation in Arabidopsis thaliana
Little is currently known about dynamic brain networks involved in high-level cognition and their ontological basis . Here we develop a novel Variational Bayesian Hidden Markov Model ( VB-HMM ) to investigate dynamic temporal properties of interactions between salience ( SN ) , default mode ( DMN ) , and central executive ( CEN ) networks—three brain systems that play a critical role in human cognition . In contrast to conventional models , VB-HMM revealed multiple short-lived states characterized by rapid switching and transient connectivity between SN , CEN , and DMN . Furthermore , the three “static” networks occurred in a segregated state only intermittently . Findings were replicated in two adult cohorts from the Human Connectome Project . VB-HMM further revealed immature dynamic interactions between SN , CEN , and DMN in children , characterized by higher mean lifetimes in individual states , reduced switching probability between states and less differentiated connectivity across states . Our computational techniques provide new insights into human brain network dynamics and its maturation with development . Our ability to adapt to a constantly changing environment is thought to depend on the dynamic and flexible organization of intrinsic brain networks [1 , 2] . Characterizing the temporal dynamics of interactions between distributed brain regions is fundamental to our understanding of human brain organization and its development [2–8] . However , most of our current knowledge of functional brain organization in adults and children is based on investigations of time-independent functional coupling . Progress in the field has been impeded by both a lack of appropriate computational techniques to investigate brain dynamics as well as an inadequate focus on core brain systems involved in higher-order cognition [3 , 4 , 9 , 10] . In particular , progress has been limited by weak analytical models for identifying time-varying brain states , and their occurrence rates and mean lifetimes , for quantifying transition probabilities between brain states , and for characterizing the dynamic evolution of functional connectivity patterns over time [9–11] . Here we overcome limitations of extant methods by developing and applying novel computational techniques for characterizing dynamic functional interactions between distributed brain regions and address two key neuroscientific goals . The first scientific goal of our study was to investigate the dynamic functional connectivity of the salience network ( SN ) , the central-executive network ( CEN ) and the default mode network ( DMN ) , three core neurocognitive systems that play a central role in cognitive and affective information processing [1 , 12] . Our second scientific goal was to characterize the maturation of the dynamic functional connectivity of the SN , CEN and DMN between childhood and adulthood in order to address important gaps in the literature regarding the nature of dynamic cross-network interactions over development and the question of how brain systems become more flexible during the period between childhood and adulthood . The SN is a limbic-paralimbic network anchored in the anterior insula and dorsal anterior cingulate cortex with prominent subcortical nodes in affective and reward processing regions including the amygdala and ventral striatum [13 , 14] . The SN plays an important role in orienting attention to behaviorally and emotionally salient and rewarding stimuli and facilitating goal-directed behavior [12 , 14–16] . The fronto-parietal CEN is anchored in the dorsolateral prefrontal cortex and supramarginal gyrus and is critical for actively maintaining and manipulating information in working memory [17 , 18] . The DMN is anchored in the posterior cingulate cortex , medial prefrontal cortex , medial temporal lobe , and angular gyrus [19–21] and is involved in self-referential mental activity and autobiographical memory [22] . In adults , task-based fMRI studies have consistently demonstrated that SN , CEN and DMN nodes are involved in a wide range of cognitive tasks , with the strength of their responses increasing or decreasing proportionately with task demands [12 , 23 , 24] . Analysis of causal interactions between these networks has also shown that high-level attention and cognitive control processes rely on dynamic interactions between these three core neurocognitive networks [16 , 25–27] . Thus , far from operating independently , these three brain networks , which have only been probed using static time-invariant connectivity analysis , must form transient dynamic functional networks ( DFNs ) allowing for flexible within- and cross-network interactions . While the SN , CEN and DMN can be reliably identified in most individuals using static network analysis of rs-fMRI data [26 , 28] , progress in characterizing their dynamic temporal properties has been limited by currently available computational tools and procedures . Most current studies of dynamic brain connectivity use a sliding window approach [29 , 30] , which is problematic because of arbitrary parameters such as window size , which can lead to erroneous estimates of dynamic connectivity [7 , 9 , 11] . Furthermore , extant methods do not provide information about the occurrence and lifetimes of individual dynamic brain states , transition probabilities between network states or unique dynamic network configurations associated with each brain connectivity state . To address these weaknesses , we developed a novel variational Bayesian hidden Markov model ( VB-HMM ) [31] to uncover time-varying functional connectivity . HMM uses a state-space approach to model multivariate non-stationary time series data [32 , 33] and cluster them into distinct states , each with a different covariance matrix reflecting the functional connectivity between specific brain regions . Importantly , VB-HMM automatically prunes redundant states , retaining only those that significantly contribute to the underlying dynamics of the fMRI data , and provides the posterior distribution of parameters rather than point estimates of maximum likelihood-based methods . We then used VB-HMM to characterize dynamic functional interactions between the SN , CEN and DMN to address our two neuroscientific goals . VB-HMM allowed us to examine for the first time several important metrics of brain dynamics: the number of distinct brain states , their occupancy rates and mean lifetimes , and switching probabilities between brain states and DFNs . Crucially , VB-HMM enabled us to investigate the temporal dynamics and evolution of states where the SN , DMN and CEN are fully segregated from each other , and states where they interact with each other . We hypothesized that segregation of the SN , DMN and CEN would constitute a dominant state with high occupancy rates and mean lifetimes . We further hypothesized that states with high occupancy rates would be temporally stable and marked by a higher probability of switching within the state compared to switching across states . We use sub-second resting-state fMRI ( rs-fMRI ) datasets acquired as part of the Human Connectome Project ( HCP ) ( http://www . humanconnectome . org ) and demonstrate the robustness of our findings across two independent cohorts of healthy adults . Next , we used VB-HMM and insights from our analyses of the adult brain to characterize the maturation of dynamic functional networks and connectivity associated with the SN , DMN and CEN between childhood and adulthood . Flexible and dynamic cross-network functional interactions are essential for mature brain function [5 , 34] , yet little is known about the nature of dynamic organization and time-varying connectivity in children relative to adults . Studies using static connectivity analyses suggest that functional brain networks undergo significant reconfiguration from childhood to adulthood , with analysis of time-averaged whole-brain connectivity patterns suggesting prominent increases as well as decreases in connectivity between childhood and adulthood . In a previous study we showed that time-averaged connectivity within key nodes of the SN and DMN as well as their inter-network interactions is weaker in children relative to adults [28] . Recent reports suggest that time-varying connectivity between distributed brain areas changes significantly with age , with greater temporal variability of connection strengths in children compared to adults[34] . Based on these observations , we hypothesized that compared to adults , children would show immature and less flexible patterns of dynamic connectivity between the SN , CEN and DMN . Crucially , VB-HMM allowed us to , for the first time , probe developmental changes in dynamic networks properties including the occurrence rates and mean lifetimes of distinct brain states , such as those in which the SN , CEN and DMN are fully segregated from each other with decreased switching probabilities . This study was approved by the Stanford University Institutional Review Board . Written informed consent was obtained from all the subjects . We first describe a novel VB-HMM framework we developed for characterizing dynamic brain networks in human fMRI data . In the following sections , we represent matrices by using uppercase letters while scalars and vectors are represented using lowercase letters . Let Y={{yts}t=1T}s=1S be the observed voxel time series , where T is the number of time samples and S is the number of subjects . yts is an M dimensional time sample at time t for subject s , where M is the number of brain regions or nodes of the dynamic functional network under investigation . Let Z={{zts}t=1T}s=1S be the underlying hidden/latent discrete states , where zts is the state label at time t for subject s . Let Z be a first order Markov chain , with stationary transition ( A ) and initial distributions ( π ) defined as: p ( zts=k\|zt−1s=j ) =Ajk ( 1 ) p ( z1s=k ) =πk ( 2 ) where 0≤Ajk≤1 , ∑k=1KAjk=1 , and πk≥0 , ∑k=1Kπk=1 . We assume the probability of the observation yts given its state zts=k to be a multivariate normal distribution with parameters mean μk and covariance Σk: p ( yts\|zts=k ) =N ( μk , Σk ) ( 3 ) Here we assume that the number of possible states K is not known a priori . Each state k has M μk and an M x M Σk . Let Φ = {π , A , Θ} ( where Θ={μk , Σk}k=1K ) be the unknown parameters of the HMM model . Using the factorization property [35] of the Bayesian network shown in Fig 1A , the joint probability distribution of the observations , hidden states , and parameters can be written as p ( Y , Z , Φ ) =∏s=1Sp ( z1s\|π ) ∏t=2Tsp ( zts\|zt−1s , A ) ∏t=1Tsp ( ( yts\|zts , Θ ) P ( Φ ) ( 4 ) In maximum likelihood methods , the parameters Φ of the model are assumed to be unknown deterministic quantities , whereas in the Bayesian approach they are treated as random variables with prior probability distributions . Here we assume that conjugate priors [35] for Φ and Z are defined as in [31] with the goal of estimating the joint posterior distribution p ( Z , Φ\|Y ) of the hidden states and parameters . Estimating this posterior distribution is analytically intractable but inference methods , such as sampling or variational methods , can instead be used [31 , 35] . Here , to estimate p ( Z , Φ\|Y ) , we use a variational Bayesian ( VB ) method [31] , which not only provides an elegant analytical approximation to the required posterior distribution but is also computationally faster than sampling approaches . Let q ( Z , Φ\|Y ) be any arbitrary probability distribution and p ( Z , Φ\|Y ) be the true posterior probability distribution . Then the log of the marginal distribution of observations Y can be written as log⁡P ( Y ) =F ( q ) +KL ( q\|\|p ) ( 5 ) where F ( q ) is known as the negative free energy and KL ( q\|\|p ) is the Kullback-Leibler ( KL ) divergence between the approximate and true posterior . These quantities are given by F ( q ) =∫dZdΦq ( Z , Φ\|Y ) log⁡p ( Y , Z , Φ ) q ( Z , Φ\|Y ) ( 6 ) KL ( q\|\|p ) =−∫dZdΦq ( Z , Φ\|Y ) log⁡p ( Z , Φ\|Y ) q ( Z , Φ\|Y ) ( 7 ) Since KL ( q\|\|p ) is nonnegative , F ( q ) serves as the strict lower bound on log P ( Y ) . F ( q ) and log P ( Y ) are equal if and only if the approximate posterior q ( Z , Φ\|Y ) is equal to the true posterior p ( Z , Φ\|Y ) for which KL ( q\|\|p ) = 0 . The goal of VB approximation is to find the approximate posterior for which the lower bound F ( q ) is maximized . We make a mean field approximation on the approximate posterior [31] wherein it factorizes as q ( Z , Φ\|Y ) =q ( Z , A , Θ , π\|Y ) =q ( Z\|Y ) q ( π\|Y ) q ( A\|Y ) q ( Θ\|Y ) ( 8 ) The functional forms of these factors are defined by the priors on the parameters and the likelihood of the data . We assume conjugate priors for the priors , which results in elegant analytical approximations to the required posterior distributions of the Eq ( 8 ) . Accordingly , the conjugate prior for π and rows of A is the Dirichlet ( Dir ) distribution , while the prior over the parameters of the Gaussian distribution Θ is the Normal-Wishart ( NW ) distribution . We further assume that the prior distribution over Φ factorizes as P ( Φ ) =p ( π ) p ( A ) p ( Θ ) ( 9 ) The forms of the Dirichlet and Normal-Wishart distributions are defined in [31] . We provide the values of the hyper-parameters of these distributions in the Appendix . Since we define conjugate priors on the model parameters , q ( π\|Y ) and q ( A\|Y ) follow multinomial distributions and q ( Θ\|Y ) follows the Normal-Wishart distribution [31] . The update equations for the posterior parameters are provided in the Supplementary Material . The posterior distribution of the hidden states can be estimated using an efficient forward-backward method similar to the Baum-Welch algorithm for ML-HMM [33 , 35] . Furthermore , our VB-HMM estimates the parameters of Normal-Wishart distribution for each state . VB-HMM therefore discovers states for which the parameters of the Normal-Wishart distributions are distinct for each state . A new state will be discovered if either mean or covariance or both are different in that state with respect to other states . In task-based fMRI studies it is important to discover states with both mean and covariance differences . However , in resting-state fMRI studies , as in the current study , differences in absolute signal levels are not relevant and states are based solely on changes in covariance over time . This can be accomplished elegantly in our Bayesian framework using the hyperparameter λk in the joint Normal-Wishart distribution . A non-informative prior value ( say , λk = 0 . 001 ) allows the data to determine the joint posterior distributions for the mean and covariance . However , setting it to a very high value ( λk = 1000 ) biases the posterior to the prior mean which is 0 in our case ( equation S . 10 ) . This ensures that our states are discovered only by the changes in covariance/inverse covariance in each state . Similar to the expectation maximization algorithm for ML-HMM , the posterior distributions for the latent and model parameters are iteratively updated in VB-HMM as follows: We iterate steps ( b ) and ( c ) until the fractional lower bound F ( q ) between two consecutive thresholds is below a set threshold value of tol = 10−3 . We initialize the states using the K-means algorithm with the number of clusters/states K set to a high value ( K = 25 ) . The sparsity property of VB-HMM prunes away unwanted clusters/states in the model . Like ML-HMM , VB-HMM provides suboptimal estimates of the posterior distributions , and these estimates are sensitive to the initial estimates of states using K-means initialization . To account for this , we repeat VB-HMM with 100 different random initializations and choose the solution for which the lower bound F ( q ) is maximum . We validated VB-HMM using three different simulation models; the details of each are provided in the Supplementary Materials . Briefly , in Simulation-1 , we created datasets with two nodes and two hidden states . The hidden states were constructed using a typical block design with two conditions ( or states ) : “OFF” and “ON” as shown in S2A Fig . We simulated observations with two nodes where the nodes are negatively correlated in the “OFF” state and uncorrelated in the “ON” state . In Simulation-2 , we simulated data with six nodes and two hidden states using the HMM generative model given by Eqs 1–3 . In this case , the two hidden states were constructed using a specified state transition matrix A and six nodes/ROIs with observations drawn from a zero-mean multivariate Gaussian distribution and state specific covariance matrices ( S3A Fig ) . Simulation-3 also consisted of six nodes and two hidden states . Here , however , the first three nodes/ROIs were correlated in the first half ( 116 samples ) of the experiment ( state 1 ) while the other three ROIs were correlated in the second half ( 116 samples ) of the experiment ( state 2 ) ( S4A Fig ) . Five datasets were simulated ( akin to a group size of five subjects in fMRI studies ) for each simulation type . We first validated VB-HMM using computer-simulated datasets generated from three different simulation models . Here we briefly summarize the results from these simulations; details are in the Supplementary Materials . For all three simulations , we applied VB-HMM with the number of hidden states ( K ) initialized to 25 and used VB-HMM to automatically determine the optimal number of states from the data . S2 Fig shows the actual states , the estimated posterior probabilities and the Viterbi decoded states for Simulation-1 . Among the 25 states , the occupancy rates of 18 states are zero suggesting that VB-HMM penalizes redundant states . Further analysis suggests that among the seven with non-zero occupancy rates , four states together constitute 98% of the total occupancy rate and these states match the underlying true states in terms of their associated estimated Pearson correlation matrices and their occurrences with respect to their respective true states . Similarly , 21 out of the 25 states in Simulation-2 had zero occupancy rates ( S3 Fig ) . The top two most dominant states comprise 98% of the total occupancy rate and are well matched with the temporal occurrence of the underlying actual states . Lastly , Simulation-3 yielded 21 out of 25 states with an occupancy rate of zero ( S4 Fig ) . Of the four states with non-zero occupancy rates , the top two account for 99 . 2% of the total occupancy rate and match the true states used to generate the data . These simulations demonstrate that VB-HMM can accurately discover the optimal number of states and the underlying dynamic connectivity across different models of simulated data . We applied VB-HMM on rs-fMRI data to uncover dynamic functional interactions between the SN , CEN and DMN in two cohorts of HCP data . Our first goal here was to identify dynamic brain states and their associated functional networks . We computed the occupancy rates and mean lifetimes of each state as well as the switching probabilities between states . A particular theoretical focus was on the occurrence of brain states in which the three networks were disconnected from each other . We conducted separate analyses on Cohorts 1 and 2 and investigated the robustness and consistency of our key findings across the two cohorts . To characterize the connectivity patterns associated with each functional state , we used a community detection algorithm on the estimated partial correlations in each state and examined the functional connectivity between ROIs . Below we describe the salient features of the dynamic functional network structure in each cohort . Given our focus on the temporal properties of the state in which the SN , CEN , and DMN were disconnected from each other , we combined states with a similar community structure into distinct DFNs ( see S1 Text ) . We then examined the occupancy rates , mean lifetimes and switching probabilities of these DFNs . Based on our primary goal of characterizing the network structure associated with segregated SN , CEN and DMN as encapsulated by DFN-1 ( Fig 3B ) and the common patterns of network structure involving DFN-1 and DFN-2 in both cohorts ( see previous sections ) , we next examined state transitions between these networks . In each cohort , network structures corresponding to all other functional states were combined together into a mixed DFN-M . As in previous sections , these analyses were conducted separately in the two cohorts with the aim of elucidating replicable findings . We next used VB-HMM to characterize the maturation of dynamic functional interactions between the SN , CEN and DMN in a Stanford cohort of IQ- and gender-matched adults and children . We used the same analytic procedures as described above on data from adults and children and then compared dynamic network properties between the two groups . To investigate whether DFN occupancy rates and mean lifetimes differ between children and adults , we focused on DFN-1 and DFN-2 , the two dominant DFNs with identical community structures in adults and children that together account for about 77% occupancy rates in both groups . Network configurations corresponding to all other functional states were combined into DFN-M . The mean lifetimes , but not the occupancy rates , of all three DFNs were significantly greater in children compared to adults ( p < 0 . 05 , FDR corrected ) ( Fig 6A and 6B ) . These findings indicate that children tend to persist longer in the same DFN than adults , as illustrated by the time evolution of the three DFNs ( Fig 5A and 5F ) . Below we further investigate this pattern of developmental differences in terms of transition probabilities between DFNs . To further investigate whether children tend to stay in one DFN configuration longer than adults , we computed transition probabilities in children and adults and compared them between the groups . The probability of within-DFN transitions was not significantly different between the two groups ( p > 0 . 05 , FDR corrected ) . However , transition probabilities to the fully disconnected SN-CEN-DMN configuration ( DFN-1 ) from both connected network configurations ( DFN-2 and DFN-M ) were significantly higher in adults compared to children ( p < 0 . 05 , FDR corrected ) ( Fig 6D ) . In contrast , children showed a higher probability of switching between the two connected network configurations ( p < 0 . 05 , FDR corrected ) . These findings demonstrate that , compared to children , adults switch back more frequently to DFN-1 , in which the SN , DMN and CEN are completely segregated from each other . Finally , to investigate how dynamic functional connectivity matures with age we compared the strength of DFN connectivity assessed using within- and cross-network links as described above . In this analysis , we further excluded participants with DFN connectivity beyond 3 standard deviations from their specific group or for whom both DFNs were not present . After exclusion , our sample consisted of 22 adults and 16 children . We found a significant three-way interaction between DFN ( DFN-1 vs . DFN-2 ) , link type ( within- vs . cross-network ) , and participant groups ( children vs . adults ) ( F1 , 36 = 10 . 99 , p = 0 . 002 ) ( Fig 6C ) , such that DFN-1 and DFN-2 configurations differed in connection strength by link type in adults ( F1 , 21 = 119 . 5 , p < 0 . 001 ) but not in children ( F1 , 15 = 0 . 491 , p = 0 . 494 ) . These results demonstrate that DFN connectivity is weaker and less differentiated in children relative to adults . The main scientific aims of our study were to ( 1 ) investigate the temporal properties of dynamic functional connectivity between the SN , CEN and DMN , three core neurocognitive networks implicated in a wide range of goal directed behaviors [12 , 15 , 16 , 26 , 48 , 49] , and ( 2 ) investigate how the temporal properties of dynamic functional connectivity between these core networks change from childhood to adulthood . To accomplish this , we first developed a novel Bayesian HMM ( VB-HMM ) model for quantifying dynamic changes in functional connectivity . A variational Bayes approach for estimating latent states and unknown HMM model parameters allowed us to overcome weaknesses associated with conventional methods and to investigate dynamic changes in intrinsic functional connectivity between three networks , which have previously only been investigated using static network analysis . VB-HMM allowed us to quantify the temporal evolution of distinct brain states and probe the dynamic functional organization of the SN , CEN and DMN in an analytically rigorous manner . Contrary to previous observations based on static time-averaged connectivity analysis [20 , 50] , we found that temporal coupling between the SN , CEN and DMN varies considerably over time and that these networks exist in a completely segregated state only intermittently with relatively short mean lifetimes . VB-HMM also revealed immature and inflexible dynamic interactions between the SN , CEN and DMN characterized by higher mean lifetimes in individual states and reduced transition probability between states , in children relative to adults . VB-HMM is a novel machine learning approach for identifying dynamic changes in functional brain connectivity . VB-HMM has several advantages over existing methods [6 , 9 , 29 , 51 , 52]: ( i ) the automated estimation of latent states and their temporal evolution; ( ii ) estimation of posterior probabilities of latent states and model parameters; ( iii ) selection of models based on a trade-off between the model complexity and fit of the data , thereby reducing overfitting; ( iv ) use of sparsity constraints resulting in pruning of weak states without having to specify the number of states a priori; and ( v ) a generative model that has the potential to provide a more mechanistic understanding of human brain dynamics . Our approach also overcomes weaknesses of existing HMM methods that are based on a maximum likelihood estimation approach and require a priori specification of the number of hidden states . Furthermore , in contrast to conventional HMM methods , VB-HMM can discover dynamic changes in states based on signal mean or covariance or both . This flexibility can be useful in uncovering latent brain dynamics during cognitive task processing , where states typically differ in both signal mean and covariance , as well as rs-fMRI , where states are better characterized by changes in covariance rather than mean signal levels . In applications to rs-fMRI , as in the present study , this is accomplished in VB-HMM by setting the prior hyperparameter value λk = 1000 for each state k . This choice forces the posterior mean values for each state ( μk ) close to prior mean ( which is zero ) ( Equation S . 10 ) and ensure that states are characterized by differences in the covariance matrices ( Σk ) , but not the mean ( μk ) . Another advantage of our Bayesian approach is that the covariance ( or inverse covariance ) estimates are regularized and the extent of regularization is determined by the data ( Eqs S11–S . 13 ) . This regularization ensures that the covariance matrices are full rank and therefore invertible to estimate partial correlations . Such regularized estimation is not possible with maximum likelihood approaches . Our simulations using three different simulation models demonstrate that VB-HMM can accurately discover the number of states , their temporal evolution , the transition probabilities between states and dynamic connectivity patterns associated with each state ( see S1 Text for details ) . We next used VB-HMM to characterize the temporal evolution of dynamic brain states in two independent cohorts of adult participants from the HCP . VB-HMM identified multiple stable states in both cohorts of participants . The observation that the number of states is strictly greater than one is consistent with previous results demonstrating that the rs-fMRI time series is not stationary[29 , 53] . Importantly , VB-HMM identified similar patterns of stable brain states in both cohorts and provided reliable and replicable estimates of occupancy rates , mean lifetimes , and state transition probabilities associated with each brain state . Although VB-HMM identified 16–19 states in both adult cohorts , only three states had occupancy rates greater than 10% ( Fig 2C and 2G ) , and these states demonstrated the highest mean lifetimes . However , even these dominant states had short mean lifetimes ranging from 7–10 s , demonstrating that brain states are temporally persistent over durations far shorter than the length of a typical rs-fMRI scan session . These features were observed in both adult cohorts , demonstrating the robustness of our findings . Furthermore , analysis of the state transition probability indicated that each state had the highest probability of transitioning to itself rather than other states ( Fig 2D and 2H ) , suggesting that temporal stability of individual states does occur . Taken together , these results demonstrate the existence of dynamic , yet stable , brain states in rs-fMRI and identify distinct connectivity patterns associated with each state . We suggest that this balance of temporal stability and dynamic connectivity is a fundamental principle of brain organization . By construction , VB-HMM states are characterized by distinct patterns of inter-node connectivity ( Figs 2 and 3 ) . To test specific hypotheses related to the dynamic interactions between the SN , CEN and DMN and interpret the neurobiological relevance of connectivity profiles , we identified dynamic functional connectivity profiles associated with the three previously known static networks . To accomplish this we applied modularity-based community detection algorithms [36] on the functional connectivity matrix estimated by VB-HMM for each state ( Fig 1B ) . This analysis revealed that , in some cases , states with non-identical connectivity matrices had similar overall community structures ( S5 , S6 , S8 and S9 Figs ) . For example , multiple states ( S5 and S6 Figs ) demonstrated a pattern in which the SN , CEN and DMN formed separate , segregated communities , reminiscent of the static functional networks previously identified by independent components analysis [50] . We next combined states with identical community structures into dynamic functional networks ( DFNs ) and examined the temporal properties of segregated and non-segregated DFNs as well as the dynamic interactions between key nodes of the SN , CEN and DMN . The SN , CEN and DMN formed separate communities and were segregated from each other ( DFN-1 in Fig 3 ) approximately 31% of the time ( 31% and 27% in Cohorts 1 and 2 , respectively ) . In this case , all three networks maintained their within-network connectivity structure–AI and ACC nodes of the SN were connected with each other , PMC and VMPFC nodes of the DMN were connected with each other , and DLPFC and PPC nodes of the CEN were connected with each other . Crucially , VB-HMM also revealed that this DFN had a mean lifetime of about 7–10 s ( 8 . 3 s and 8 . 8 s in Cohorts 1 and 2 , respectively ) ( Fig 3C and 3G ) . These findings suggest that although this particular DFN configuration is a prominent feature of SN , CEN and DMN organization , it has a relatively short lifetime . The second dominant DFN identified by VB-HMM had a community structure in which the CEN and DMN were interconnected in one community , while the SN nodes remained segregated from the CEN and DMN , forming an independent network ( DFN-2 in Fig 3 ) . This DFN configuration had occurrence rates of 36% and 18% in Cohorts 1 and 2 , respectively ( Fig 3 ) . The remaining states had distinct DFN configurations ( S5 and S6 Figs ) , with varying levels of cross-network interactions , but their occurrence rates were lower and not consistent across the two cohorts . Previous work from our lab [12] [39] and recent work by other labs [54 , 55] has indicated that the SN plays a critical role in switching between the DMN and the CEN . Our results suggest that this switching is transient ( i . e . doesn’t persist for a long time ) and may occur not very frequently . Finally , analysis of the switching probability between DFNs revealed that each DFN had a high probability ( 0 . 91 in Cohort 1 and 0 . 93 in Cohort 2 ) of making self-transitions ( Fig 3D and 3H ) . Thus , as with individual brain states , the two dominant DFN configurations ( DFN-1 and DFN-2 in Fig 3 ) were stable over time but persistent only for short time intervals . Taken together , these findings identify key features of dynamic functional interactions associated with the SN , CEN and DMN and confirm that the static segregated networks previously identified using independent component analysis occur only about 30% of the time . The organization of brain networks in adults is shaped by years of development , learning and brain plasticity [5] . Previous studies using static connectivity analysis have pointed to changing topological organization of connections with age [56–61] . More specifically , it has been suggested that interactions between the SN , CEN and DMN are immature in children [26 , 28] , but their dynamical temporal properties are not known because previous analyses have assumed that brain networks are static over time . To address this gap in knowledge , we used VB-HMM to investigate how dynamic functional interactions between the SN , CEN and DMN mature from childhood to adulthood . VB-HMM revealed significant differences in key temporal properties , such as mean lifetime and state transition probabilities , between children and adults and provides a new level of detail regarding immature brain dynamics in childhood . To test the hypothesis that dynamic functional interactions between the SN , CEN and DMN are different in children , we first identified two common dominant DFNs with identical network structure in both children and adults . We used this commonality to probe the maturation of dynamic brain networks using measures of occupancy rate , mean lifetime and switching probabilities derived using VB-HMM . VB-HMM identified two DFN configurations with the same community structure in both groups: DFN-1 , in which the SN , CEN and DMN were segregated from each other and DFN-2 , in which the AI and ACC nodes of the SN were decoupled from each other and showed significant cross-network interactions with the DMN and CEN , respectively ( Fig 5B and 5F ) . Critically , the network structures of the DFN-2 were different between the HCP Adult cohorts ( Fig 3B and 3F ) and Stanford Adult cohort ( Fig 5B ) . The differences may have arisen from the slower sampling rate in the Stanford data which used a standard TR = 2 seconds compared to the faster TR = 0 . 73 seconds used in the HCP data . Critically , patterns were consistent within scanners–the first and second DFNs were identical in the two HCP cohorts and in the two Stanford cohorts . Analysis of connectivity profiles across nodes of the SN , CEN and DMN showed that the two DFNs were less differentiated in children relative to adults ( Fig 6C ) . Critically , the mean lifetimes of the two common DFN configurations ( DFN-1 and DFN-2 ) were significantly greater in children compared to adults ( Fig 5G ) suggesting that immature brain network organization is characterized by greater dwelling time in specific network configurations . Analysis of transition probabilities further revealed that the likelihood of transitions into the configuration in which the SN , CEN and DMN were completely segregated from each other was significantly lower in children compared to adults ( Fig 6D ) . In contrast , children showed a higher likelihood of switching between non-segregated network configurations ( Fig 6D ) . These findings support the notion that relative to those of adults , children’s brains are less flexible and less likely to switch to the segregated DFN configuration from other network configurations . Taken together , these findings demonstrate that children have less flexible dynamic cross-network interactions , characterized by reduced switching between distinct brain states and longer persistence in specific network configurations . In summary , we developed a novel Bayesian HMM ( VB-HMM ) approach for estimating the temporal properties of dynamic functional networks in fMRI data and applied it to characterize time-varying connectivity of the SN , DMN , and CEN , three neurocognitive networks that play a crucial role in human cognition . VB-HMM uncovered latent states , dynamic functional connectivity and state transition probabilities associated with these three networks , thereby revealing transient dynamic functional networks ( DFNs ) that allow for flexible within and cross-network interactions . In adults , VB-HMM revealed that the SN , CEN and DMN–systems that were previously characterized only by static network analysis–were in a segregated , disconnected state , only about 30% of the time with mean lifetimes of 7–10 s . VB-HMM also revealed that dynamic functional interactions between the SN , CEN and DMN are weaker and immature in children . Critically , the uncovered brain dynamics were not related to individual differences in age and in-scanner micro-movements . Our computational techniques provide new insights into the dynamic functional organization of the SN , DMN and CEN and their maturation with development . More generally , our computational approach may be useful for investigating the dynamic aspects of functional brain organization in neurodevelopmental and psychiatric disorders , including autism , schizophrenia and mood disorders[27] .	Characterizing the temporal dynamics of functional interactions between distributed brain regions is of fundamental importance for understanding human brain organization and its development . Progress in the field has been hampered both by a lack of strong computational techniques to investigate brain dynamics and an inadequate focus on core brain systems involved in higher-order cognition . Here we address these gaps by developing a novel variational Bayesian Hidden Markov Model ( VB-HMM ) that uncovers non-stationary dynamical functional networks in human fMRI data . In two cohorts of adults , VB-HMM revealed multiple short-lived states characterized by rapid switching and transient connectivity between the salience ( SN ) , default mode ( DMN ) , and central executive ( CEN ) networks—three brain systems critical for higher-order cognition . In children , relative to adults , VB-HMM revealed immature dynamic interactions between SN , CEN , and DMN , characterized by higher mean lifetimes in individual states , reduced switching probability between states and less differentiated connectivity across states . Our findings suggest that the flexibility of switching between distinct brain states is weaker in childhood , and they provide a novel framework for modeling immature brain network organization in children . More generally , the approach used here may prove useful to the investigation of dynamic brain organization in neurodevelopmental and psychiatric disorders .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "children", "medicine", "and", "health", "sciences", "diagnostic", "radiology", "functional", "magnetic", "resonance", "imaging", "markov", "models", "neural", "networks", "applied", "mathematics", "random", "variables", "neuroscience", "covariance", "magnetic", "resonance", "imaging", "algorithms", "simulation", "and", "modeling", "age", "groups", "adults", "probability", "distribution", "mathematics", "brain", "mapping", "neuroimaging", "families", "research", "and", "analysis", "methods", "computer", "and", "information", "sciences", "imaging", "techniques", "hidden", "markov", "models", "probability", "theory", "people", "and", "places", "radiology", "and", "imaging", "diagnostic", "medicine", "population", "groupings", "biology", "and", "life", "sciences", "physical", "sciences" ]	2016	Temporal Dynamics and Developmental Maturation of Salience, Default and Central-Executive Network Interactions Revealed by Variational Bayes Hidden Markov Modeling
Hypersensitive response programmed cell death ( HR-PCD ) is a critical feature in plant immunity required for pathogen restriction and prevention of disease development . The precise control of this process is paramount to cell survival and an effective immune response . The discovery of new components that function to suppress HR-PCD will be instrumental in understanding the regulation of this fundamental mechanism . Here we report the identification and characterisation of a BTB domain E3 ligase protein , POB1 , that functions to suppress HR-PCD triggered by evolutionarily diverse pathogens . Nicotiana benthamiana and tobacco plants with reduced POB1 activity show accelerated HR-PCD whilst those with increased POB1 levels show attenuated HR-PCD . We demonstrate that POB1 dimerization and nuclear localization are vital for its function in HR-PCD suppression . Using protein-protein interaction assays , we identify the Plant U-Box E3 ligase PUB17 , a well established positive regulator of plant innate immunity , as a target for POB1-mediated proteasomal degradation . Using confocal imaging and in planta immunoprecipitation assays we show that POB1 interacts with PUB17 in the nucleus and stimulates its degradation . Mutated versions of POB1 that show reduced interaction with PUB17 fail to suppress HR-PCD , indicating that POB1-mediated degradation of PUB17 U-box E3 ligase is an important step for negative regulation of specific immune pathways in plants . Our data reveals a new mechanism for BTB domain proteins in suppressing HR-PCD in plant innate immune responses . The capacity of plants to protect themselves against pathogens depends on detection mechanisms that recognize pathogen-derived molecules and then activate host defence responses . Plants have evolved an armoury of defence mechanisms that allow them to counter infection . These encompass both basal responses , triggered by recognition of conserved pathogen-associated molecular patterns ( PAMPs ) , and pathogen-specific responses , mediated via pathogen- and plant-specific gene-for-gene recognition events . Recognition leads to activation of immune responses [1] . One feature of the immune response in plants is the hypersensitive response ( HR ) . This involves a highly localized programmed cell death ( PCD ) of the infected region that helps to contain pathogen spread [2] . Control over HR-PCD is thus central to determining susceptibility or resistance to disease . Understanding how infected cells bring about downstream molecular signalling to establish and regulate cell death is a fundamental challenge in plant biology . PCD also plays a critical role in development in plants and animals [3] . Therefore , understanding the mechanisms that control cell death has a significance that extends beyond plant-pathogen interactions . It is well established in many organisms that ubiquitin ( Ub ) is a key modifier of signalling . The Ub-conjugation pathway involves the activity of 3 enzymes called E1 ( Ub activase ) , E2 ( Ub conjugase ) and E3 ( Ub ligase ) . The E3 facilitates the formation of an isopeptide linkage between Ub and the target protein . A polyUb chain is often formed by the addition of multiple Ub monomers . Polyubiquitination of a given substrate serves not only as a signal for degradation but also for targeting and re-profiling [4 , 5] . Even the addition of a single ubiquitin moiety plays an important role in determining the fate of the substrate [6] . In this context , the critical role in substrate specificity resides mainly with the E3 ligase and these enzymes form the largest group of proteins within the Ub-enzyme cascade , indicating the presence of a large number of targets in the proteome . The Arabidopsis genome project identified more than a thousand genes encoding putative E3 ligases and these can be divided into seven classes which can be subdivided into two basic groups , dependant on the occurrence of either a HECT ( Homology to E6-AP C-Terminus ) or RING ( Really Interesting New Gene ) / U-box domain [7] . RING containing proteins can either ubiquitinate substrates independently or function as part of a multi-subunit complex which , in plants , includes: SCF ( Skp1-Cullin1-F-box ) , CUL3 ( Cullin 3 ) -BTB/POZ ( Bric a brac , Tramtrack and Broad complex/Pox virus and Zinc finger ) , CUL4-DDB1 ( UV-Damaged DNA-Binding protein 1 ) and APC ( Anaphase Promoting Complex ) . Examples of E3 ligases from nearly all classes have been documented to be involved in immune responses against a range of pathogens in distantly related plant species , indicating plants have evolved to utilize the ubiquitin system for immunity against invading pathogens . A number of E3 ligases have been implicated in regulation of pathogen-associated molecular pattern ( PAMP ) -Triggered Immunity ( PTI ) , activated in response to perception of conserved microbial molecules , and HR-PCD , which can be a component of PTI , but is often activated by perception of pathogen effector molecules ( effector-triggered immunity , or ETI ) . Examples include the F-box protein COI1 that is required to initiate the defence response mediated by the phytohormone JA [8] . Similarly , another F-box protein , SON1 or RPD4 , regulates defence responses independent of salicylic acid and systemic acquired resistance ( SAR ) . Interestingly it is also targeted for destruction through ubiquitin-mediated and non-ubiquitin-mediated pathways [9] . F-box protein ACRE189 ( or ACIF1 ) is required for Cf-4 and Cf-9 HR-PCD in tomato and tobacco . Arabidopsis homologs of ACIF1 act upon jasmonate and abscisic acid responsive gene pathways to promote defence [10] . Plant U-box ( PUB ) E3 ligases PUB12 and PUB13 work in concert to attenuate PTI by ubiquitinating the flg22 receptor FLS2 , facilitating its degradation [11] . PUB13 regulates a number of processes , including flowering and senescence , in addition to immunity . Indeed , recent yeast-2-hybrid ( Y2H ) screens have revealed several potential substrates and binding partners for PUB13 , including: phosphatidylinositol-4 kinase and RABA4B , with which it complexes to negatively regulate salicylic acid ( SA ) -mediated defences [12]; the transcription factor HFR1 , involved in flowering [13]; and the ABA regulator ABI1 , a protein phosphatase 2C family member , which is a PUB13 substrate for ubiquitination and degradation [14] . In addition to PUB12 and PUB13 , PUBs 22 , 23 , and 24 also act in combination to negatively regulate plant immunity [15] . Single , double and triple pub22 pub23 pub24 mutants exhibited progressively attenuated suppression of the flagellin-induced ROI burst , MPK3 activation and downstream PTI marker gene expression [15] . PUB22 attenuates PTI by targeting the exocyst component exo70B2 for ubiquitination and degradation . However there is very little information on the regulators or the substrates of U-box E3 ligases that positively regulate immunity . PUB17 and its tobacco homolog ACRE276 ( NtPUB17 ) were originally identified as positive regulators of defence against the fungal pathogen Cladosporium fulvum [16] . PUB17 functions in the plant nucleus to activate Cf-mediated HR-PCD [17] . A further member of this family , PUB20/CMPG1 , is required for HR-PCD mediated by Avr9/Cf-9 , Pto/AvrPto interactions , and recognition of the PAMP INF1 [18] . The RXLR effector AVR3a from Phytophthora infestans suppresses these HR-PCD events by stabilizing CMPG1 to prevent its normal activity [19 , 20] . Here we report the identification and characterisation of a tobacco BTB domain E3 ligase protein , POB1 , that functions to suppress HR-PCD during the plant immune response by targeting tobacco PUB17 for ubiquitin-mediated proteasomal degradation . Tobacco and Nicotiana benthamiana plants with reduced POB1 activity show accelerated HR-PCD mediated by a broad range of elicitors whilst those with increased POB1 levels suppress HR-PCD . We demonstrate that POB1 dimerization and nuclear localization are required for its function in HR-PCD suppression . Mutated versions of POB1 that show reduced interaction with PUB17 fail to suppress HR-PCD indicating that POB1-mediated degradation of a U-box E3 ligase is a critical step for regulating plant innate immunity . Recent yeast-2-hybrid ( Y2H ) screens with plant U-box proteins have identified a number of substrates for these E3 ubiquitin ligases [12 , 13] . Therefore to identify potential substrates of the U-box protein NtPUB17 ( Tobacco homolog of AtPUB17 and previously known as NtACRE276 ) we initiated a Y2H screen using the ARM repeat region of NtPUB17 ( assumed to be the substrate-interacting domain ) and identified a BTB domain protein which bore significant sequence homology to the Arabidopsis protein AtPOB1 ( Fig 1A ) . We thus named this interactor NtPOB1 . Co-immunoprecipitation studies using Agrobacterium-mediated transient assays ( agroinfiltration ) in Nicotiana benthamiana demonstrated that NtPUB17 interacts with NtPOB1 in planta ( Fig 1B ) . The BTB domain is an evolutionarily conserved , versatile protein-protein interaction motif that participates in a wide range of critical cellular functions in eukaryotes , including transcriptional regulation , cytoskeleton dynamics , ion channel assembly and gating , and targeting proteins for ubiquitination ( reviewed in [21] ) . The Arabidopsis genome encodes 80 putative BTB domain genes , of which only a handful have been assigned function . These include the BTB ankyrin repeat domain protein NPR1 and the SA receptors NPR3 and NPR4 which regulate systemic acquired resistance ( SAR ) in plants [22] . Phylogenetic relationships between NtPOB1 and several representative BTB/POZ proteins are shown in S1A Fig Close analysis of the BTB/POZ protein superfamily indicated that BTB proteins could be classified into subfamilies based on additional motifs associated with the BTB/POZ domain . These motifs include the MATH , TRP , armadillo and ankyrin repeats , which are often involved in mediating protein-protein interactions . The predicted protein sequence of NtPOB1 identified an additional BTB-associated domain , called the BACK domain ( S1B Fig ) . In other eukaryotes , proteins containing the BACK domain are always associated with a C-terminal KELCH domain [23] . However , interestingly , the BACK domain in NtPOB1 is not associated with any obvious KELCH-repeat and is therefore referred to as a BTB protein containing just an additional BACK domain . Along with AtPOB1 and AtPOB2 there are another 9 BTB-BACK proteins in Arabidopsis , of which At4G01160 is the closest homolog to AtPOB1 and AtPOB2 ( S1C Fig ) . BLAST searches using the NtPOB1 cDNA sequence against the N . benthamiana genome sequence [24] identified several cDNA species with high similarity to NtPOB1 . The NtPOB1-related N . benthamiana cDNA sequence fragments were used to isolate the full-length NbPOB1 by RACE experiments [25] . The predicted protein sequence of NbPOB1 is 92% similar to NtPOB1 . The phylogenetic tree in S1A Fig shows that NbPOB1 groups with AtPOB1 , AtPOB2 and At4G01160 , raising the possibility of functional conservation of these proteins between plants , similar to that demonstrated for AtPUB17 and NtPUB17 , which were functionally exchangeable between tobacco and Arabidopsis [16] . Previously we demonstrated that NtPUB17 acts as a positive regulator of Cf9-Avr9 mediated defence against C . fulvum [16] . Tobacco plants and cell cultures carrying a Cf-9 transgene respond to Avr9 peptide with rapid induction of HR-PCD and associated responses in a strict gene-for-gene manner , indicating that all components required for efficient execution of the Avr9 effector triggered HR are present in this heterologous model system [26] . To assess whether tobacco POB1 ( NtPOB1 ) plays a role in Avr9-mediated HR , hairpin ( hp ) RNA-mediated silencing was used to specifically knockdown NtPOB1 gene expression in transgenic Cf-9 tobacco plants [27] . Both HG: NtPOB1 ( NtPOB1 specific hpRNA ) and the control empty vector constructs were delivered into Cf-9 tobacco leaves by agroinfiltration to transiently silence NtPOB1 gene expression . NtPOB1 transcript levels were analysed by RT-PCR three days following agroinfiltrations using NtPOB1-specific primers downstream of the expressed hpRNA fragment . A significant decrease in NtPOB1 mRNA levels was observed in the leaves infiltrated with HG: NtPOB1 compared to the control leaves expressing the HG:00 empty vector construct ( Fig 2A ) . The HG: NtPOB1 and HG:00 constructs were expressed in the same leaf , opposite the main vein . Diluted Avr9 peptide stock solutions were infiltrated into the Cf-9 tobacco leaf patches required to trigger limited HR in the control leaf segments . Silencing of NtPOB1 triggered a more rapid Cf-9/Avr9-dependant HR while no visible HR was detectable in the control leaf segments at the same time point , suggesting that NtPOB1 functions as a negative regulator of Cf-9 mediated resistance ( Fig 2B ) . Virus-induced gene silencing ( VIGS ) [28 , 29] was used to similarly study the function of NbPOB1 . Gene silencing vectors based on the Tobacco Rattle Virus ( TRV ) [30–32] carrying 2 different and non-overlapping DNA fragments from the NbPOB1 sequence were used to induce sequence-specific degradation of NbPOB1 . BLAST analyses confirmed that both cDNA fragments ( TRV: NbPOB1_A and TRV: NbPOB1_B ) were unique to NbPOB1 , ruling out potential silencing of closely related ‘off target’ genes . The efficiency and specificity of NbPOB1 silencing was confirmed by RT-PCR on mRNA from emerging leaves of plants 21 days after inoculation with the TRV: NbPOB1_A and TRV: NbPOB1_B constructs ( S2 and S3 Figs ) . NbPOB1-silenced plants had no obvious morphological differences to vector only control plants . To ascertain the potential involvement of NbPOB1 in immunity in N . benthamiana , pathogen infection studies were carried out with Pseudomonas syringae pv . tabaci ( Ps . tabaci ) , a virulent pathogen of tobacco [33] , on NbPOB1-silenced plants . A low titre of Ps . tabaci ( 1x104 c . f . u . ml-1 ) was used to infect N . benthamiana plants that were inoculated with the NbPOB1 silencing vectors TRV: NbPOB1_A and TRV: NbPOB1_B . Bacterial colony counts were performed 3 days and 5 days following Ps . tabaci infiltrations . No significant difference in bacterial growth was detected at 3 days post infection ( dpi ) between the control and the NbPOB1-silenced plants . However , the growth of the pathogen was significantly reduced in NbPOB1-silenced plants at 5 dpi , when compared to bacterial growth in the control plants ( Fig 3A and 3B ) . In addition , POB1-silenced N . benthamiana plants were inoculated with the late blight pathogen Phytophthora infestans . P . infestans leaf colonisation was also significantly attenuated in POB1 VIGS plants , measured as reduced lesion diameter and reduced development of sporangia ( Fig 3C and 3D ) . These data indicate that plants down-regulated for NbPOB1 transcript accumulation have enhanced capability of restricting both Ps . tabaci growth and P . infestans colonisation . To further investigate the potential involvement of POB1 in negatively regulating immunity , a range of elicitors and gene-for-gene interactions that trigger HR-PCD were studied . Significantly , at an early stage of HR-PCD development in control plants ( 4 days post-inoculation ) , silencing NbPOB1 by VIGS revealed accelerated HR-PCD induced by the P . infestans PAMP INF1 , and by co-expression of tomato Cf4 and C . fulvum Avr4 , tomato Pto and P . syringae AvrPto , and potato Rx with Potato Virus X coat protein ( CP ) ( Fig 3E and 3F ) . These data indicate that POB1 negatively regulates multiple HR-PCD events . Loss-of-function studies established that NbPOB1 potentially negatively regulates HR-PCD triggered by the PAMP INF1 and by multiple gene-for-gene interactions . We examined the effect of over-expressing NbPOB1 on disease resistance signalling by Agrobacterium-mediated transient over-expression in Cf-9 tobacco . Cf-9 tobacco leaf sections were inoculated with Agrobacterium containing a construct expressing NbPOB1 fused to the Green Fluorescent Protein ( GFP ) at its N-terminus . As a control , Agrobacterium containing GFP alone was used to inoculate the corresponding leaf sections on the other side of the main vein . Three days after Agroinfiltrations , the leaf segments were infiltrated with Avr9 peptide to induce Cf-9-dependant HR-PCD . Overexpression of NbPOB1 suppressed the Cf-9/Avr9-mediated HR-PCD in 70% of leaves tested compared to the corresponding GFP control ( Fig 4A ) . Previously , we established that AtPUB17 was able to complement the pro-cell death function of NtPUB17 in tobacco lines where NtPUB17 was silenced , establishing functional conservation between Solanaceae and Brassicacea species [16] . Given the high sequence similarity between AtPOB1 and NbPOB1 we tested whether AtPOB1 was also functionally interchangeable with NbPOB1 and was able to suppress Cf-9 dependant HR . Agrobacterium containing the GFP-AtPOB1 construct was infiltrated into one side of Cf-9 tobacco leaves . Three days after Agroinfiltrations the leaf segments were then infiltrated with Avr9 peptide to elicit HR . Leaf segments over-expressing the GFP-AtPOB1 fusion protein suppressed the formation of the HR cell death in 70% of the leaves tested , compared to the leaf segments expressing the GFP control ( Fig 4B ) . Immunoblots indicated that both GFP-NbPOB1 and GFP-AtPOB1 were stable as intact fusion proteins in planta ( Fig 4C ) . This indicates that AtPOB1 and NbPOB1 are functional orthologs ( hereafter called POB1 ) acting as negative regulators of Avr9-elicited HR-PCD . We tested whether over-expression of GFP-POB1 would enhance susceptibility to P . infestans on N . benthamiana leaves . Compared to the GFP control , GFP-POB1 expression resulted in significantly enhanced P . infestans leaf colonisation ( Fig 4D ) . Finally , we demonstrated the central role of POB1 by extending this HR-PCD suppression to different classes of cell death inducers in N . benthamiana . INF1 , Cf4 with Avr4 , Rx with PVX coat protein , and Pto with AvrPto , were each co-expressed with either GFP alone or GFP-POB1 . In each case , HR-PCD was attenuated specifically when the HR-PCD elicitors were co-expressed with GFP-POB1 ( Fig 4E and 4F ) . These data underline the importance of POB1 as a regulator of diverse defence pathways . To ascertain the biological significance of POB1-PUB17 interaction in defence suppression we investigated the levels of PUB17 and POB1 protein in N . benthamiana when they are co-expressed . We observed that the abundance of PUB17 protein was decreased when co-expressed with POB1 and this effect was mitigated when co-infiltrated with the proteasome inhibitor MG132 ( Fig 5A ) . In contrast , silencing of POB1 by VIGS resulted in enhanced accumulation of PUB17 protein ( S4 Fig ) . Taken together with the yeast-two-hybrid and co-immunoprecipitation data ( Fig 1 ) this indicated that PUB17 is a substrate for ubiquitination by POB1 , leading to its proteasome-mediated degradation . Transient expression assays in N . benthamiana indicated that GFP-POB1 was predominantly located in the nucleoplasm ( Fig 5B ) , whilst PUB17 was shown previously to be located in the nucleoplasm and nucleolus [17] . We performed bimolecular fluorescence complementation ( BiFC ) , fusing one half of YFP ( YC ) to the N-terminus of PUB17 and the other half ( YN ) to POB1 . When expressed alone in planta , YC-PUB17 was detectable in immunoblots . However , when co-expressed with YN-POB1 , YC-PUB17 was only detected in the presence of MG132 ( Fig 5C ) . Confocal microscopy of N . benthamiana cells co-expressing YC-PUB17 and YN-POB1 in the presence of MG132 revealed reconstituted YFP fluorescence in the nucleoplasm , implicating this as the site for interaction between these proteins ( Fig 5D ) . To further investigate the importance of the nuclear localisation of POB1 for its function , a nuclear export signal ( NES ) was added to the N-terminus of GFP-POB1 to form NES-GFP-POB1 . Transient expression of NES-GFP-POB1 in N . benthamiana leaves revealed that the fusion protein was stable and that GFP fluorescence was mainly located in the cytoplasm of the cells examined ( Fig 6A and 6B ) . Crucially , in contrast to GFP-POB1 , NES-GFP-POB1 expression failed to enhance P . infestans colonisation ( Fig 6C and 6D ) and failed also to suppress Cf4-Avr4-mediated cell death ( Fig 6E and 6F ) . Our data indicates that the nuclear localisation of POB1 is required for its HR-PCD suppression function and this involves the degradation of PUB17 . BTB proteins have been shown to form dimers in other eukaryotes [23] . However there is little in vivo evidence for the functional significance of dimerization of BTB domain proteins in plants . Alignment of several BTB/POZ domains from different proteins with the POB1 BTB/POZ domain identified a conserved Aspartate residue ( D146 ) . Mutation of the corresponding residue in PLZF ( D35 ) exhibited reduced dimerization efficiency [34] . The D146 residue in POB1 was mutated to an alanine residue . The effect of this mutation on POB1 dimerization was tested using the yeast-two-hybrid system . These experiments demonstrate that the dimerization efficiency of POB1D146A with itself or with POB1 is markedly reduced in yeast , compared to dimerization of POB1 with itself ( Fig 7A ) . To test whether this was the case also in planta , Agrobacterium containing HA-POB1 and GFP-POB1 and separately the D146A mutant versions ( HA-POB1D146A and GFP-POB1D146A ) were co-expressed in tobacco leaves . Three days after Agroinfiltration , total protein was extracted and the expression of all protein was confirmed by immunoblotting with anti-HA and anti-GFP antibodies ( Fig 7B ) . Both wild-type and mutant POB1 were expressed to the same level ( Fig 7B ) , indicating that the D146A mutation did not affect the stability of POB1 . Co-immunoprecipitations were performed using anti-HA antibodies to immunoprecipitate HA-POB1 and HA-POB1D146A . Whereas GFP-POB1 co-immunoprecipitated with HA-POB1 , HA-POB1D146A did not co-immunoprecipitate GFP-POB1D146A ( Fig 7B ) , confirming the effect of the D146A mutation in disrupting BTB/POZ-mediated dimerization in planta . To examine the effect of the D146A mutation on the ability of POB1 to suppress Cf-9/Avr9-dependant HR-PCD , GFP-POB1D146A and GFP-POB1 were transiently expressed in Cf-9 tobacco leaves . Three days following Agroinfiltrations , Avr9 peptide was infiltrated into the leaf segments and suppression of HR-PCD was observed only in the segments expressing the wildtype GFP-POB1 and not in GFP-POB1D146A segments ( Fig 7C ) . Moreover , whereas overexpression of GFP-POB1 suppressed Cf4/Avr4-induced cell death , the GFP-POB1D146A dimerization mutant was unable to suppress this cell death , when compared to the GFP control ( Fig 7D ) , indicating that the dimerization of POB1 is essential for its function in suppressing HR-PCD . Dimerization and formation of high order complexes are a characteristic of BTB/POZ proteins and are essential for their function in acting as E3 ligases [35] . As expected for BTB domain proteins , N . benthamiana transient assays followed by co-immunoprecipitation experiments revealed that HA-POB1 and HA-POB1D146A were able to interact with GFP-tagged-CUL3 indicating that POB1 is able to form ubiquitin E3 ligase complexes in planta ( S5 Fig ) . However the dimerization mutant GFP-POB1D146A that is unable to suppress HR-PCD had significantly reduced capacity to co-immunoprecipitate PUB17 . Moreover , PUB17 was more stable when co-expressed with GFP-POB1D146A than with GFP-POB1 ( Fig 7E ) . This indicates that suppression of cell death by POB1 is in part due to targeting PUB17 for proteasomal degradation . Previous biochemical characterization of the tobacco PUB17 provided clear evidence for its role in positively regulating HR-PCD mediated by Cf9-avr9 [16] . However , PUB17 does not play a role in regulating INF1-triggered HR-PCD [17] . Thus , the suppression of INF1-triggered HR-PCD by POB1 indicates that it negatively regulates immune responses in addition to those controlled by PUB17 . PUB17 is thus unlikely to be the only positive regulator of immunity that is a target for POB1-mediated degradation , and further work is needed to identify additional candidate substrates for ubiquitination by POB1 . Strong evidence supporting our claim that POB1 is a conserved regulatory hub controlling HR-PCD comes from transient overexpression data . We show that , when overexpressed , AtPOB1 and NbPOB1 are functional homologs capable of efficiently suppressing HR-PCD mediated by PAMPs and effectors from evolutionarily diverse pathogens . These data suggest that POB1 is a rate-limiting factor for efficient signaling flux to activate HR-PCD . Most negative regulators of cell death have proved to be indirect , as the lesions result from the perturbation of metabolic pathways [35 , 36] . However RIN4 , BON1 , and BAP1 are some examples of the negative regulators than directly activate specific gene function [16 , 37–39] . We have demonstrated that POB1 is a newly identified negative regulator that may be comparable to RIN4 , BON1 and BAP1 . However , more recently , a BTB protein called NRL1 has been shown to act as a negative regulator of immunity , suppressing INF1-triggered cell death . NRL1 acts as a so-called susceptibility ( S ) factor , as it is required for infection by the late blight pathogen P . infestans [40] . Unlike POB1 , however , it does not negatively regulate such a wide range of immune pathways . In addition , whereas POB1 functions in the nucleus , NRL1 apparently acts at the plant plasma membrane [40] . Interestingly , an RXLR effector from P . infestans targets NRL1 , presumably supporting or enhancing its role as a negative regulator of immunity [40] . It will be interesting to see whether POB1 is similarly targeted by pathogen effectors to support its role in suppressing immunity . Dimerization and formation of high order complexes are a characteristic of BTB/POZ proteins and are essential for their function [34] . Deletion of the BTB domain of PLZF affected its transcriptional repression activity , causing cell cycle arrest [41] . The importance and conserved function of the BTB domain was also demonstrated by replacing the BTB domain of the human BCL-6 BTB domain fused to the Drosophila Tramtrack69 ( ttk69 ) and therefore replacing its own BTB domain without disrupting its activity in neuronal photoreceptor cell differentiation [42] . To test the importance of POB1 dimerization in suppressing the Avr9-triggered HR , a point mutation was generated based on the 3D structural model of the PLZF BTB domain [41] . The conserved Aspartate residue in POB1 ( D146; corresponding to D35 of PLZF ) located in the charged pocket of the BTB fold was mutated to an Alanine residue , which therefore neutralized the negative charge of Aspartate . The mutation not only reduced the dimerization ability of POB1 , but this mutant also failed to suppress the Avr9-triggered HR-PCD in Cf-9 tobacco , indicating that the charged pocket of the BTB is important not only in mediating dimerization of POB1 but also for its function . It is remarkable that the point mutation in the BTB domain and the BTB deletion mutant were both unable to suppress HR-PCD , emphasizing the importance of an intact BTB domain for the activity of POB1 . The BTB fold is structurally similar to the linker proteins Skp1 and Elongin C of the SCF and ECS ubiquitin ligase complexes , respectively [43] . Skp1 links the scaffold CUL1 to substrate-specific adaptors ( F-box ) , while ElonginC bridges CUL2 with SCOS-box proteins [44] . Several reports identified BTB-proteins as CUL3-interactors , where the BTB domain directly interacts with CUL3 . In agreement with this we found that AtPOB1 interacts with AtCUL3A in yeast and in planta , indicating that it can form BTB-CUL3 ubiquitin ligases in vivo ( S3A and S3B Fig ) . However the POB1 POB1D146A mutant that was unable to suppress Avr-9 mediated HR-PCD was also found to interact with CUL3 , indicating that dimerization was not required for CUL3 interaction . In mammals , the MATH-BTB protein SPOP has been linked to ubiquitination of MacroH2A , to regulate its deposition on the inactive X chromosome [45] , and DAXX , to regulate transcriptional repression of proapoptotic proteins such as p53 [46] . It was shown that dimerization mutants of SPOP also co-purify with CUL3 . However , the dimerization-defective SPOP is significantly impaired for ubiquitination activity on its substrate protein . To support a similar effect on POB1 we found that POB1D146A mutant was less effective in binding to PUB17 and concomitantly less able to suppress HR-PCD indicating that PUB17 is indeed the substrate for POB1 . We demonstrated nuclear localization of POB1 to be important for its function by fusing the N-terminal nuclear exclusion signal ( NES ) to the full-length POB1 . The NES-AtPOB1 fusion protein was successfully excluded from the nucleus and it subsequently could not suppress HR-PCD elicited by Avr4 in Cf-4 N . benthamiana cells and fail to enhance P . infestans colonisation ( Fig 6C and 6E ) . Previous work in plant innate immunity has indicated that subsets of BTB domain proteins are located within the nucleus . The best understood nuclear localized BTB protein in plant immunity is NPR1 , the master coactivator of SAR in plants [47 , 48] . Unlike POB1 , NPR1 contains recognizable ankyrin repeats at its C-terminus and evidence suggests that , in the absence of salicylic acid ( SA ) , NPR1 is continually degraded in the nucleus in a proteasome-dependent manner , probably via CUL3 binding [49] . However a portion of NPR1 is also located in the cytoplasm and during SAR this portion is recruited to the nucleus and interacts with CUL3 to possibly ubiquitinate and degrade transcriptional repressors to fully activate SA marker genes . Although we did not find any evidence for POB1 localization in the cytoplasm excluding POB1 prevented its interaction with PUB17 and also attenuated HR-PCD suppression indicating that the main function of POB1 occurs within the nucleus . Tobacco plants were grown in environmentally controlled cabinets at 24°C with 16hr light and 8 hr dark cycles . The Cf-9 transgenic N . tabacum cv Petite Havana line 34 . 1B were previously described [50] . N . benthamiana was grown in individual pots in a glasshouse at 22°C ( 16/8 h light/dark cycle ) as described in Bos et al . [19] . Total RNA from two 2 cm diameter leaf discs was isolated using the Trizol Reagent ( Sigma-Aldrich ) method following the manufacturer’s instructions . First-strand cDNAs were synthesized from 2 ug of total RNA using Superscript II RNAse H_ Reverse Transcriptase ( Invitrogen ) . cDNAs for tobacco and Nicotiana benthamiana POB1 were amplified by PCR using gene- specific primers ( S1 Table ) . RACE experiments were performed using a GeneRacer RACE Ready cDNA Kit ( Invitrogen , USA ) and according to the manufacturer’s instructions but with gene-specific primers . The virulent strain Pseudomonas syrinage pv . tabaci EV ( Ps pv . tabaci ) [51] was used to infect N . benthamiana plants . The Bacterial strain was grown at 28°C on King’s B agar plates containing the appropriate antibiotics for 2 days . The bacteria were then grown overnight in liquid King’s B medium at 28°C with constant shaking ( 200 rpm ) , washed once , and resuspended in 10 mM MgCl2 to a final concentration adjusted to 1x105 cfu/ml . The culture was then infiltrated into the leaves using a 1mL syringe . The trays were put under a humidifier and bacterial counts were performed after 3 and 5 days . Phytophthora infestans strain 88069 was used for N . benthamiana infection and was cultured on rye agar at 19°C for 2 weeks . Plates were flooded with 3 mL of water and scraped with a glass rod to release sporangia . The resulting solution was collected in a falcon tube , Sporangia numbers were counted using a hemocytometer and adjusted to 60 , 000 per mL . The suspension was incubated in the refrigerator at 4°C for one to four hours to let the sporangial cytoplasm cleave and release zoospores . 10 mL droplets were inoculated onto the abaxial side of leaves of detached or intact N . benthamiana plants stored on moist tissue in sealed boxes . The boxes containing the infected plants were gently wrapped with plastic cling-film , covered with black plastic , and incubated overnight at 19 degrees Celsius after inoculation . The black plastic covers were removed the following morning , and disease incubation continued under 16/8 hours light/dark conditions . The disease phenotype was recorded by measuring the lesion diameter of the leaves from intact plants at 5 days post inoculation and 7 days for the detached leaves . Sporangia counts were performed on 7 dpi leaves from VIGSed plants which had been immersed in 2ml H2O and vortexed to release sporangia . A Haemocytometer was used to count the number of sporangia recovered from each leaf and was expressed as sporangia/ml . Transgenic Avr9 N . tabacum plants express the Cladosporium fulvum Avr9 peptide in the intercellular fluids , which was extracted as described in [52] . Leaves of 7–8 week-old Avr9 tobacco plants were cut into thin strips , immersed into distilled water and subjected to a vacuum . Following release of the vacuum , the leaf strips were put in 20mL syringes attached to 50mL falcon tubes and centrifuged to discharge the intercellular fluid containing the Avr9 peptide . Several dilutions of the Avr9 peptide were used to trigger cell death in 7-week-old Cf-9 tobacco [53] . Confocal microscopy The target proteins were expressed in N . benthamiana leaves by Agrobacterium-mediated transient expression . A . tumefacien containing target protein fusions were pressure infiltrated into leaves of 4-week old N . benthamiana plants . N . benthamiana leaf cells expressing fluorescent protein fusions were imaged no later than 2 days after agroinfiltration by using a Leica TCS-SP2 AOBS confocal microscope . GFP was excited with 488 nm from an argon laser and its emissions were detected between 500 and 530 nm . Split-YFP was excited with 514 nm from an argon laser with emissions collected between 530–575 nm . Images were only collected from leaf cells expressing low levels of the protein fusions to minimise possible artefacts of ectopic protein expression . MG132 ( 100 μM ) treatments were infiltrated 5 hours before imaging under the confocal microscope . Transient assays in N . benthamiana and N . tabacum . Tobacco transient assays were performed as described in [53] . Agrobacteria containing the desired construct were grown on LB plates at 28°C for 2 days . A few colonies were then picked and grown overnight at 28°C with constant shaking ( 200 rpm ) in 10mL LB culture vials containing the appropriate antibiotics . The bacteria was pelleted and washed once with 10mM MgCl2 . The cells were then resuspended in 10mM MgCl2 containing 200μM acetosyringone and the OD600 was adjusted to 0 . 2 . The cultures were left for 2–3 hours at room temperature before infiltration . The Agrobacteria cultures were infiltrated into N . benthamina leaves by pressure-infiltration on the lower side of the leaves using a 1mL syringe . As for Nicotiana tabacum , infiltrations were carried out through small incisions on the lower side of the leaves ( using a clean razor blade ) . The infiltrated plants were then put in a controlled growth cabinet at 22°C under constant light for 2–3 days to allow gene expression . Loss-of-function studies in N . benthamiana were carried out through the virus-induced gene silencing ( VIGS ) system [54] based on the Tobacco Rattle Virus ( TRV ) strain PPK20 [55] . The TRV genome consists of 2 RNAs . RNA1 ( involved in movement ) and RNA2 ( involved in transmissibility ) . RNA1 and RNA2 are encoded by the pBINTRA6 and pTV00 respectively . pBINTRA6 was transformed into the Agrobacterium strain LB440 , whereas pTV00 ( including the target sequence ) was transformed into the Agrobacterium strain GV3101 containing the helper plasmid pSA-rep . Agrobacteria containing pBINTRA6 and pTV00 constructs were grown overnight in 10mL YEB cultures at 28°C with constant shaking ( 200 rpm ) . The cultures were then spun at 4000rpm for 10 minutes , washed once , and resuspended in 10mM MgCl2 containing 200μM acetosyringone to a final OD600 of 0 . 5 . The Agrobacteria containing the respective pTV00 constructs were mixed with the Agrobacteria containing the pBINTRA6 vector in a 1:1 ratio and incubated at room temperature in the dark for 2 hours . NbPOB1 gene was amplified by PCR using gene- specific primers ( S1 Table ) . A 426bp POB1_A fragment and a 732bp POB1_B fragment were amplified using primers described in S1 Table from N . benthamiana and cloned into pTV00 VIGS vector . The sequences used to silence NbPOB1_A and NbPOB1_B do not share homology with any other sequence annotated in the N . benthamiana genome . A pTV00 construct expressing GFP , was used as a control The mixed cultures were then used to inoculate 3–4 weeks old N . benthamiana plants by pressure infiltration at the lower side of the leaves . Inoculated plants were put in the growth rooms and grown for ~3 weeks until the onset of gene silencing . The silencing level of NbPOB1 was analysed by RT-PCR two to three weeks post infiltration . The silenced leaves at 3 weeks post-infiltration were used for pathogenicity assays . The gene-silencing vector pHELLSGATE12 [27] induces gene silencing transiently in N . tabacum , 3–4 days after Agrobacterium infiltrations through hairpin RNA ( hpRNA ) -mediated silencing . The pHELLSGATE12 vector uses the Gateway technology ( Invitrogen ) to facilitate cloning of the insert in sense and anti-sense directions , separated by the PDK and Catalase-1 introns . An entry clone contain 732bp of NtPOB1 gene sequence ( amplified by gene specific primers indicated in S1 Table ) were mixed with the pHELLSGATE12 vector in the presence of LR clonase enzyme mix ( as described in manufacturers instructions ) and incubated at 25°C for 2h to overnight . The pHELLSGATE12 vector with the target sequence was next transformed into the Agrobacterium strain GV3101 by electroporation . Transformed Agrobacteria were grown overnight in 10mL LB cultures with the appropriate antibiotics ( Rifampicin 100μg/ml and Spectinomycin 50μg/ml ) at 28°C with constant shaking ( 220rpm ) . The cultures were next spun ( 2500 rpm , 10minutes ) , washed once and resuspended in 10mM MgCl2 . The OD600 was adjusted to 0 . 5 and the cultures were incubated at room temperature before infiltration into Cf-9 N . tabacum leaves , allowing 3–4 days for hpRNA constructs to be generated , and consequently silencing of the target gene . The silencing level of NtPOB1 was analysed by RT-PCR using gene specific primers ( S1 Table ) 4 days post Agroinfiltration . The sequence used to silence NtPOB1 recognizes both NtPOB1 isoforms in tobacco since they share 99% homology . POB1D146A Site-directed mutagenesis was carried out using the QuickChange Site-Directed Mutagenesis Kit ( Stratagene , U . S . A . ) by using PfuTurbo DNA polymerase and pTOP-POB1 as a template . Primers containing the desired mutation were designed accordingly that are shown in S1 Table . Parental ( i . e . , the nonmutated ) supercoiled dsDNA was removed by DpnI treatment . The resultant Ile-146 was verified by sequencing . One conserved amino acids of POB1 , Aspartate at positions 146 , were substituted with alanine , resulting in the mutant POB1D146A . The mutant POB1D146A was recombined , using LR clonase , into pB7WGF2 or HA for in planta assays . Plant tissue from Arabidopsis and tobacco was ground in ice-cold homogenization buffer [25mM Tris-HCl , 150mM NaCl , 1mM EDTA , 10% ( v/v ) glycerol , 5mM DTT , 0 . 1% Triton X-100 , Complete EDTA-free protease inhibitor tablet cocktail] using a mortar and pestle and clarified by centrifugation ( 12000g , 4°C , 15minutes ) . The supernatant was transferred to a clean Eppendorf tube and protein concentrations were determined using the Bradford assay ( BioRad ) . The total protein fraction was then mixed with 1x SDS loading buffer [25mM Tris-HCl pH6 . 8 , 10% ( v/v ) glycerol , 2% ( w/v ) SDS , 5% ( v/v ) β-mercaptoethanol , 0 . 001% ( w/v ) Bromophenol blue] and boiled for 5 minutes before being separated by SDS-PAGE . 2μl of protein extract were mixed with 18μl sterile water and 900μl diluted ( 5-fold ) Bradford reagent ( BioRad ) . The mixture was then transferred to a cuvette and incubated at room temperature for 5min . The absorbance of the samples was read at a wavelength of 595 , against a blank sample ( no protein extract added ) . The concentration of the samples was calculated based on a standard curve . Protein interactions were tested by expressing the proteins of interest transiently in N . benthamiana through Agroinfiltration . Three days after infiltration , 3–4 leaves were ground in 1ml of ice-cold extraction buffer using a pestle and mortar . The homogenate was spun at 12 , 000 rpm at 4°C and the supernatant was applied to a Sephadex-G25 ( Sigma ) column that had been pre-equilibrated in lysis buffer . Fifty μl of epitope antibody agarose conjugates ( also pre-equilibrated in lysis buffer ) were added to the protein samples and were left to rotate for 3–4 hours at 4°C . The agarose beads were then pelleted by centrifugation at 1 , 000g for 10sec at 4°C and washed 3 times with 1mL lysis buffer . After the final wash , 50μl of 1x SDS-loading buffer was added to the beads , boiled for 5min and the supernatant analyzed by SDS-PAGE . After the SDS-PAGE run , proteins were transferred to a PVDF membrane ( BioRad ) using the Mini-PROTEAN Trans-Blot transfer cassette ( BioRad ) filled with cold transfer buffer [25mM Tris-HCl , 190mM glycine , 20% ( v/v ) methanol] . Electro-transfer of proteins to nitrocellulose membranes was carried out at 100V for 1h or at 30V overnight . The membrane was then blocked in 5% ( w/v ) non-fat dried milk dissolved in TBS-T [15mM Tris-HCl pH7 . 6 , 150mM NaCl , 0 . 1% ( v/v ) Tween 20] with gentle agitation to reduce non-specific binding . Then primary antibodies diluted in TBS-T containing 5% ( w/v ) non-fat dried milk were added to membrane and incubated for 1–3 hours . Membranes were then washed several times with TBS-T for a total of 20min before incubation for 45 min with secondary HRP-conjugated antibodies . Following incubation , the membranes were washed five times with TBS-T for a total of 25min . Immunodetection was carried out using Immobilon ( Millipore ) for chemiluminescent detection in accordance with the manufacturer’s instructions . The membrane , washed from any unbound HRP-conjugated secondary antibody , was incubated in 1:1 Luminol Reagent and Peroxide Solution for 5 minutes at room temperature . Excess substrate was drained and the membrane was covered in cling film and placed in an X-ray cassette . The membrane was then exposed to X-ray film ( Kodak ) under safe red light conditions . The film was developed using the X-OMAT developing system ( FUJI ) . Small-scale transformation of yeast was performed as described in the HybriZAP-2 . 1 Two-Hybrid Libraries instruction manual ( Stratagene ) . A few colonies of AH109 cells growing on YPD agar ( BD Biosciences ) plates were inoculated in 10mL YPD medium ( BD Biosciences ) and grown overnight at 30°C with constant shaking ( 200rpm ) . The culture were then added to a fresh 100mL YPD medium and incubated for an additional 4–5 hours until the OD600 reached 0 . 8 . At this stage , the yeast cells were spun at 1 , 000g for 5min , washed once in sterile water , and resuspended in 1mL LiAc/TE ( 0 . 1M LiAc and 10mM TE pH8 . 0 ) solution . Approximately 200ng of bait and prey plasmids were mixed with 100μg of pre-heated salmon testes DNA in an eppendorf tube . Hundred μl of the yeast cell suspension and 600μl of transformation solution [40% ( w/v ) PEG , 0 . 1M LiAc , 10mM TE pH8 . 0] were next added to the mixture and mixed by vortexing , followed by an incubation period of 30 min at 30°C with constant shaking ( 200 rpm ) . Seventy μl of DMSO was added to the transformation mix , mixed gently by inverting , heat-shocked at 42°C for 15min and then placed on ice for 10min . The cells were pelleted by centrifugation at maximum speed for 10sec and resuspended in 200μl sterile water . The cells were plated on Minimal SD agar Base medium ( BD Biosciences ) containing Leu-/Trp- DO Supplement ( BD Biosciences ) as a selective medium for plasmid DNA co-transformation or Leu-/Trp-/His- SO Supplement ( BD Biosciences ) with 20mM X-α-Gal ( Melford ) as a selective medium for testing interacting partners . The plates were incubated at 30°C for 3 to 5 days until colonies developed . Full-length protein sequences of Arabidopsis POB and BTB-containing genes , Mouse KEAP1 and NbPOB1 were aligned using MAFFT [56] . Up to 1000 iterations were allowed , and post-alignment local pairwise refinement was employed . Amino acid sequences of BTB and BACK domains ( where present ) from the above genes were also aligned separately using the same method , and manually refined in the case of the BACK domain alignment . Arabidopsis gene At2G30600 contains two BTB domains so both were included separately in the BTB alignment . Phylogenies were estimated separately from the BTB and BACK multiple alignments using the maximum likelihood ( ML ) method in MEGA5 [57] . Phylogenies were bootstrapped using 1000 replicates , Jones-Taylor-Thornton evolutionary model was used , and 5 gamma-distributed variation rates among sites were allowed . Gaps were treated by deletion . Nearest-neighbour interchange and automatic initial tree inference were used . Phylogenies were rooted using the well-established relevant domain from Mouse KEAP1 as outgroup . For comparison , phylogenies were also estimated from both sets of domain alignments in MEGA5 using maximum parsimony ( MP ) , minimum evolution ( ME ) , neighbor joining ( NJ ) , and UPGMA methods with default parameters . The domains on proteins were identified based on models from Prosite , Panther , Pfam , Smart , Superfam and Gene3D [58; 59 , 60] and manually refined . Sequence data for the Nicotiana Benthamiana POB1 has been deposited in the GenBank data library under the accession number 1399145 .	Control over PCD in plants , like in animals , is central to determining susceptibility or resistance to disease . Yet there is a real paucity in understanding of the biochemical processes that are crucial in determining how PCD is co-ordinated during plant immune responses . Here we demonstrate that a BTB domain protein , POB1 , is a conserved novel negative regulator of plant immune responses triggered by evolutionarily diverse pathogens . BTB domain proteins have been shown to associate with Cullin-3 proteins to form ubiquitin E3 ligases . We reveal that the U-Box E3 ligase PUB17 , a well-established positive regulator of multiple immune pathways in plants , is a target for degradation by the ubiquitin E3 ligase POB1 . We also demonstrate that this targeted proteolysis occurs in the nuclei of plant cells . In this report we provide clear evidence that distinguishes between BTB domain dependant dimerization and Cullin-3 ubiquitin ligase assembly and link this to target degradation in plant immune signalling .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "cell", "death", "plant", "anatomy", "enzymes", "cell", "processes", "enzymology", "microbiology", "plant", "science", "plants", "flowering", "plants", "ligases", "bacteria", "agrobacteria", "physical", "chemistry", "chemical", "properties", "nicotiana", "dimerization", "plant", "microbiology", "proteins", "chemistry", "leaves", "biochemistry", "plant", "biochemistry", "cell", "biology", "protein", "domains", "biology", "and", "life", "sciences", "physical", "sciences", "organisms" ]	2017	BTB-BACK Domain Protein POB1 Suppresses Immune Cell Death by Targeting Ubiquitin E3 ligase PUB17 for Degradation
The mainstay of toxoplasmosis treatment targets the folate biosynthetic pathways and has not changed for the last 50 years . The activity of these chemotherapeutic agents is restricted to one lifecycle stage of Toxoplasma gondii , they have significant toxicity , and the impending threat of emerging resistance to these agents makes the discovery of new therapies a priority . We now demonstrate that auranofin , an orally administered gold containing compound that was FDA approved for treatment of rheumatoid arthritis , has activity against Toxoplasma gondii in vitro ( IC50 = 0 . 28 µM ) and in vivo ( 1 mg/kg ) . Replication within human foreskin fibroblasts of RH tachyzoites was inhibited by auranofin . At 0 . 4 µM , auranofin inhibited replication , as measured by percent infected fibroblasts at 24 hrs , ( 10 . 94% vs . 24 . 66% of controls; p = 0 . 0003 ) with no effect on parasite invasion ( 16 . 95% vs . 12 . 91% p = 0 . 4331 ) . After 18 hrs , 62% of extracellular parasites treated with auranofin were non-viable compared to control using an ATP viability assay ( p = 0 . 0003 ) . In vivo , a previously standardized chicken embryo model of acute toxoplasmosis was used . Fourteen day old chicken embryos were injected through the chorioallantoic vein with 1×104 tachyzoites of the virulent RH strain . The treatment group received one dose of auranofin at the time of inoculation ( 1 mg/kg estimated body weight ) . On day 5 , auranofin-treated chicken embryos were 100% protected against death ( p = 0 . 0002 ) and had a significantly reduced parasite load as determined by histopathology , immunohistochemistry and by the number of parasites quantified by real-time PCR . These results reveal in vitro and in vivo activity of auranofin against T . gondii , suggesting that it may be an effective alternative treatment for toxoplasmosis . Toxoplasma gondii is the second leading cause of hospitalizations ( 8% ) and deaths ( 24% ) among foodborne pathogens in the US . People typically become infected by three principal routes of transmission: foodborne , animal-to-human ( zoonotic ) and mother-to-child ( congenital ) , and rarely as post-solid organ transplant , blood transfusion or work related injuries . The number of primarily infected individuals varies widely worldwide: 22 . 5% of the American population is infected with this parasite [1] , while in other parts of the world , the infection prevalence can be as high as 95% . These individuals are at risk of developing disease which usually follows after congenital transmission or reactivation of T . gondii latent forms ( bradyzoites ) in immunocompromised hosts [2] . Unfortunately , current available therapies have significant toxicity and are only active against one lifecycle stage of the parasite , the tachyzoite , and have no effect over the bradyzoite form [3] , [4] . Furthermore , the impending threat of emergence of resistance to these therapies makes the discovery of new therapeutic targets a priority . One promising re-profiled drug , auranofin , a gold containing compound that is FDA approved for the treatment of rheumatoid arthritis , has recently shown broad antiparasitic activity against Plasmodium falciparum [5] , Leishmania infantum [6] , Schistosoma mansoni [7] and Entamoeba histolytica [8] among others . Auranofin's anti-parasitic activity seems to stem from its gold molecule that readily dissociates and targets thioredoxin reductase , which we have recently demonstrated in our work with Entamoeba histolytica trophozoites [8] and cysts of Entamoeba invadens ( manuscript in preparation ) . Given that thioredoxin reductase is a highly conserved enzyme in protozoan parasites [9] and based on our preliminary data , we hypothesized that auranofin has activity against T . gondii . Per Public Health Services ( PHS ) Policy , the Institutional Animal Care and Use Committee ( IACUC ) oversight is not required for egg model of toxoplasmosis using unhatched eggs . PHS Policy is applicable to proposed activities that involve live vertebrate animals . While embryonal stages of avian species develop vertebrae at a stage in their development prior to hatching , Office for Protection from Research Risks ( OPRR ) has interpreted "live vertebrate animal" to apply to avians ( e . g . , chick embryos ) only after hatching ( http://www . gpo . gov/fdsys/pkg/CFR-2009-title9-vol1/xml/CFR-2009-title9-vol1-chapI-subchapA . xml ) . Primary human foreskin fibroblasts ( HFF ) were initially cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum ( Gibco , Life Technologies , Carlsbad , Calif . ) , penicillin and streptomycin ( 50 µg/ml each ) and maintained subsequently in the same medium with 2% fetal bovine serum ( FBS ) . T . gondii RH tachyzoites ( National Institutes of Health AIDS Reference and Reagent Repository , Bethesda , MD ) and RH tachyzoites expressing cytoplasmic yellow fluorescent protein ( YFP , kindly provided by M . J . Gubbels , Boston College , Boston , Massachusetts ) [10] were maintained by serial passage in HFF monolayers at 37°C in a humid 5% CO2 atmosphere . Auranofin ( Enzo Life Sciences ) , was dissolved in 100% ethanol as a stock solution ( 4 mg/mL ) and then diluted in complete tissue culture medium ( DMEM +2% FBS ) for final concentrations of 0 . 1 to 19 µM . For in vivo experiments , the auranofin concentration used was 1 mg/kg of estimated body weight . Pyrimethamine ( Sigma Aldrich ) was dissolved in 100% ethanol in a stock concentration of 5 mg/mL . Three final dilutions in complete tissue culture medium were examined: 0 . 02 , 0 . 1 and 0 . 2 µM . Sulfadiazine ( Sigma Aldrich ) was dissolved in complete medium at a final stock concentration of 5 mg/mL . Three doses in complete medium were evaluated: 0 . 2 , 1 and 2 µM . Testing of pyrimethamine/sulfadiazine combinations by checkerboard method ( using abovementioned concentrations ) was carried out in triplicates and in three independent experiments . Black , 96-well , tissue culture-treated plates with optical clear bottoms were purchased from Greiner bioOne ( Germany ) . HFF cells were added to wells in a final volume of 200 µl and grown to confluence . Freshly syringed , lysed parasites were filtered through a 5- µm polycarbonate filter ( sterile Millex SV low binding durapore PVDF syringe filters ) , centrifuged , and resuspended in parasite culture medium without phenol red ( Gibco , Life Technologies , Carlsbad , Calif . ) . HFF host cells were infected with 0 . 5×103 YFP- expressing RH tachyzoites in the presence of different dilutions of auranofin , sulfadiazine , pyrimethamine , or control ( ethanol alone ) in triplicate . Plates were kept in a humidified incubator with 5% CO2 at 37°C for five days . After 5 days of incubation , plates were washed , fixed with 4% paraformaldehyde and read with a Synergy Mx BioTek ( Vermont , US ) multimode microplate reader Gen5 Software ( excitation 510 nm; emission 540 nm ) . For excitation , a single flash from a UV Xenon lamp was used for each well , and emission signals were recorded with a sensitivity setting of 100 . The values are presented as percentages of growth inhibition relative to the untreated controls ( defined as 100% survival ) . YFP-expressing RH tachyzoites , grown in DMEM 2% FBS without phenol red , syringed , lysed and filtered as described above , were used to infect 6-well plates containing fresh , confluent HFF monolayers . Each well was inoculated with 0 . 5×103 YFP-expressing RH tachyzoites . The treatment group was treated with auranofin ( 0 . 4 µM ) , and the parasites were allowed to grow for seven days before fixation with 4% paraformaldehyde . Plaques were defined as independent foci of green fluorescence that correspond to a cluster of YFP tachyzoites infecting multiple adjacent HFFs . These plaques were visualized and counted per low power field ( 20x ) with an inverted fluorescence microscope . CellTiter 96 Non-Radioactive Cell Proliferation Assay was performed according to the manufacturer's instructions ( Promega ) . Confluent monolayers of HFF host cells ( approximately 1×104 ) plated in clear bottom 96-well plates were treated with multiple dilutions of auranofin ( 0 . 1–19 . 2 µM ) . After 5 days incubation , 15 µL of dye solution ( tetrazolium salt ) was added to each well and the plates were incubated at 37°C for 4 hr . The Solubilization Solution/Stop Mix was then added to the culture wells to solubilize the formazan product , and the absorbance at 570 nm was recorded using a 96-well plate reader ( Synergy Mx BioTek ( Vermont , US ) multimode microplate reader Gen5 Software ) . The 570 nm absorbance reading is directly proportional to the number of cells normally used in proliferation assays . The values are presented as percentages relative to the untreated controls ( defined as 100% survival ) . In preparation for invasion and replication assays , RH wild type tachyzoites were grown for 72 hrs . Intracellular parasites were collected as described above . Monolayers of confluent HFF cells were grown in 8-well chamber slides . Three independent experiments were conducted with triplicates , and at least 100 cells were counted . The number of infected cells and the number of tachyzoites per vacuole was determined per each high power field by light microscopy . The effect of auranofin on tachyzoites of RH wild type T . gondii was assessed by ATP assays . Extracellular parasites ( 0 . 25×106 per experimental group ) collected as previously described , were kept in suspension with complete medium ( DMEM +2%FBS ) with or without auranofin ( 0 . 4 µM ) for 0 , 2 , 4 , 6 , 8 hrs and 18 hrs ( overnight ) . After incubation , experimental groups were sonicated , aliquots were spun down at 4000 rpm for 5 min , and supernatants extracted for ATP assays . CellTiter-Glo Luminescent Cell Viability Assay was performed according to manufacturer's instructions ( Promega ) . Fifty microliters of supernatant from each experimental group were aliquoted into wells in a 96 opaque-wells plate ( Nunc ) in triplicate . Subsequently , an equal volume of CellTiter-Glo Reagent was added to each well . Stabilization of the luminescence was accomplished by incubating the plate for 10 mins at room temperature . Readings were performed with a GloMax Luminometer ( Promega ) . The readings are presented as relative luminescence units ( RLU ) . We have previously standardized the chicken embryo model [11] . Briefly , twelve day old pathogen-free fertilized chicken eggs ( McIntyre Farms , Hemet , CA ) were incubated at 37°C in a humid incubator . At 14-days old , a small window was cut through the shell with a hand drill directly over the blood vessel in each egg , and the vein was visualized with 1 drop of sterile mineral oil on the exposed membrane . Tachyzoites ( 1×104 ) in Dulbecco's modified Eagle's medium with or without auranofin ( 1 mg/kg of predicted weight for age ) were injected directly into the chorioallantoic vein [11] with a 28-gauge needle without pre-incubation . The windows were sealed with tape , and the embryos were incubated in a 37°C incubator . The eggs were candled once a day daily to assess viability . Livers and brains were harvested from the embryos either at the time of their death or 5 ( or 8 ) days post-infection , whichever occurred first . One half of each collected organ was fixed in 4% paraformaldehyde for histopathology ( hematoxylin and eosin staining and for immunohistochemistry staining for T . gondii with anti-T . gondii HRP antibody ) . The second half of each organ was frozen at −70°C for subsequent quantitative PCR analysis . To quantify tachyzoites in vivo , a standard curve was constructed by adding 107 tachyzoites to brain or liver samples ( 100 mg ) from 19-day-old chicken embryos and homogenizing the preparation with a cordless homogenizer ( VWR ) in cell lysis solution . Total genomic DNA was extracted from 25 mg tissue aliquots with the Qiagen DNeasy Blood & Tissue Kit per manufacturer's instructions ( Qiagen , Alameda , CA ) . DNA was eluted in 200 µl of DNA elution buffer , and then serially diluted to create a standard curve [11] . Tissue from experimental embryos was similarly harvested and genomic DNA extracted as above . Using 2 µl aliquots of eluted genomic DNA as template , quantitative PCR amplification was performed to determine the relative amount of T . gondii surface antigen ( SAG1 ) gene , a constitutively produced gene . Quantitative PCR was performed in duplicate using primers 5′-GTC ATT GTA GTG GGT CCT TCC-3′ and 5′-GCC TCA TCG GTC GTC AAT AA-3′ and PrimeTime probe 5′-TCC TAC GGT GCA AAC AGC ACT CTT-3′ ( IDT ) , and the cycling conditions were 95°C for 10 min , followed by 40 cycles of 95°C for 15 s , and 60°C for 1 min . The relative amount of product generated was measured by determining the threshold cycle when the level of specific PCR product as measured by probe fluorescence that exponentially increased and crossed the threshold of a passive reference dye ( ROX ) in each sample . The standard curves ( with known numbers of tachyzoites added to uninfected liver or brain ) were used to extrapolate the numbers of tachyzoites present in unknown samples . Results are presented as relative log10 of relative numbers of tachyzoites per organ . Results were analyzed using GraphPad Prism software 6 . 0 . All the in vitro experiments and the qPCR quantification results were analyzed with two-tailed , non paired , non-parametric tests to determine statistically significant differences ( p<0 . 05; CI 95% ) between control and treatment groups . A Kaplan Meier survival curve was calculated comparing control vs . chicken embryos treated with a single dose of auranofin . Response data measurements were fit to a sigmoid Emax model using the computer program NONMEM ver 7 . 2 ( ICON , Dublin , Ireland ) . A naïve-pooled approach was employed incorporating all individual experiment results in the analysis . For IC50 determination experiments , 96-well plates ( clear bottoms ) with confluent monolayers of HFF cells were infected with 0 . 5×103 YFP-RH tachyzoites for 4 hrs . At the end of infection period , extracellular parasites were removed and complete medium ( DMEM +2% FBS without phenol ) was added back with twofold serial dilutions of auranofin yielding a concentration range of 0 . 15–4 . 8 µM . Fluorescence measurements at day 5 post infection showed that auranofin inhibited growth by 50% at a concentration of 0 . 28 µM ( IC50 ) ( Figure 1A , Hill coefficient = 1 . 94; maximum response or Emax: 82% ) . In comparison , all different combinations of pyrimethamine and sulfadiazine generated a maximum response of 80% ( Emax: 80%; Figure 1C ) . For host cell toxicity assays , 96-well plates with confluent monolayers of HFFs host cells were treated with twofold serial dilutions of auranofin yielding a concentration range of 0 . 3–19 . 2 µM . Triplicates per experimental group were read at 120 hrs ( Day 5 ) . By measuring absorbance at day 5 post-infection , auranofin caused cell cytotoxicity in 50% of the cells at a concentration of 8 . 21 µM ( TD50 ) ( Figure 1B , Hill coefficient 3 . 89 ) . Fourteen day old chicken embryos were injected through the chorioallantoic vein with 1×104 tachyzoites of the virulent RH strain . The treatment group received auranofin at the time of inoculation at a dose of 1 mg/kg ( estimated body weight ) . While all control embryos died by day 4 , auranofin-treated chicken embryos were 100% protected against death by day 5 ( p = 0 . 0002 ) ( Figure 3A ) and had a significantly reduced parasite load as determined by histopathology and by the number of parasites quantified by real-time PCR ( expressed as a log10 ) from their brains ( 5 . 27 vs 2 . 98; p = 0 . 002 ) and livers ( 6 . 705 vs 3 . 11; p = 0 . 0003 ) ( Figure 3B ) and histopathology and immunohistochemistry ( Figure 4 ) . Of note , the amount of tissue decomposition found in the control chicken embryos suggested that they died one day before their documented death ( on day 3 ) . We demonstrate for the first time , that auranofin has significant activity against T . gondii . In vitro , auranofin reduced parasite replication , while in vivo , it reduced the parasite load and most remarkably , only one dose of auranofin prevented death in a model of acute toxoplasmosis . Our study of the anti- T . gondii effects of auranofin in vitro showed that it affects parasite viability , reducing its replication ability without affecting its capacity to invade the host cell in the absence of host cell toxicity . These results are compelling since auranofin is a FDA approved drug with a known safety profile , which can expedite its use in clinical trials . Auranofin is active against T . gondii similarly to the activity described for other protozoans of great public health importance such as Plasmodium falciparum [5] and Leishmania infantum [6] . In our study , our Emax modeling of the treatment response to auranofin demonstrated a Hill coefficient greater than 1 , suggesting positive co-operativity or independent binding of auranofin to its target . Additionally , auranofin's maximum inhibition of T . gondii growth was 82% , which is equivalent to the effect observed with current standard of therapy for toxoplasmosis ( all sulfadiazine and pyrimethamine combinations , across the board , generated a calculated maximum effect of 80% ) ( Figure 1C ) . Auranofin had a TD50 for HFF host cells that was 29 fold higher than the IC50 of 0 . 28 µM suggesting a high therapeutic index . This therapeutic index supports its safety as a potential alternative treatment for acute T . gondii infection . The independent IC50 of sulfadiazine ( 26 . 05 µM ) and pyrimethamine IC50 ( 0 . 402 µM ) , as demonstrated by Meneceur et al [12] are higher than that of auranofin ( 0 . 28 µM ) . However , the combination of lower doses of pyrimethamine and sulfadiazine demonstrated a maximum inhibitory effect over the growth of T . gondii of 80% consistently throughout multiple combinations ( Figure 1C ) . Auranofin's maximum inhibitory effect was equivalent to that of sulfadiazine-pyrimethamine ( Figure 1A and 1C ) . This suggests that auranofin could be an alternative candidate for the treatment of acute toxoplasmosis , although further studies are needed before consideration of clinical trials . Auranofin's anti-toxoplasmic mechanism of action is not known . However , from our in vitro studies we can surmise that auranofin affects replication while it does not exert any effect during the T . gondii dynamic invasion process . We challenged HFFs with T . gondii tachyzoites for 5 , 15 and 30 min and found no differences in invasion ( data not shown ) . Even when the invasion time was prolonged to 1 hr , we did not detect any difference in the rate of invasion whether cells were treated or not with auranofin . Contrary to these findings , we observed statistically significant differences in the percentage of infected cells after overnight incubation post-infection in the presence of auranofin as compared to controls . This latter observation strongly suggests that auranofin affects replication of the parasite by inhibiting its growth . On the other hand , the molecular target of auranofin , as an antiparasitic agent , is strongly suggested by the current literature . Angelucci et al [7] , demonstrated that auranofin inhibited Schistosoma mansoni glutathione-thioredoxin reductase , which the parasite solely relies on for antioxidant protection . Similarly , we recently reported that Entamoeba histolytica was rendered vulnerable to auranofin's antiparasitic effect because it inhibits its sole thioredoxin reductase [8] . As an intracellular parasite , T . gondii needs to circumvent host cell-mediated oxidant attacks during its invasion and replication; therefore , it is conceivable that T . gondii thioredoxin reductase might be the target for auranofin . However , T . gondii possesses multiple anti-oxidant enzymes that might be directly or indirectly affected by auranofin: thioredoxin reductase , glutathione reductase and thioredoxin-dependent peroxidases [13] . Other targets within the dynamic parasitophorous vacuole ( which results from the direct interaction between the parasite and the host cell ) are also unknown . We performed assays to differentiate the load of reactive oxygen species ( ROS ) with dichlorodihydrofluorescein between auranofin-treated and control cells . Although we observed differences between control and auranofin-treated infected cells per fluorescence microscopy , we failed to demonstrate quantifiable differences ( data not shown ) . Further studies are underway to determine the exact molecular target and mechanism of action involved in auranofin's anti-Toxoplasma activity . The most striking results came from our in vivo chicken embryo model of acute toxoplasmosis . All chicken embryos treated with a single dose of auranofin survived to the end of the experiment ( after 5 days post infection ) , while all the control subjects succumbed to overwhelming infection no later than day 3 post infection ( Figure 3A ) . Similar effects were observed if the embryos were injected on day 12 ( instead of day 14 ) and were allowed to incubate for 8 days post infection ( data included in the survival curve ) . Hence , a single dose of auranofin provided 100% protection from death in all acutely infected embryos , while it reduced the parasite load in organs such as the brain and the liver with almost no inflammatory reaction associated with the lower parasite load ( see hematoxylin & eosin histology in Figure 4 ) . Although , parasites were not completely absent in the auranofin-treated group , the observed 100% in vivo survival was achieved with only one dose of auranofin . We cannot demonstrate the viability of the parasites detected in the treatment group since we used DNA detection by qPCR . Further studies with mouse animal models with a standard daily treatment regimen are part of our immediate future studies . One of the limitations of this chicken embryo model is its short course . The chicken embryo is not allowed to hatch , hence we are not able to prolong incubation beyond 21 days ( 5–8 days post infection ) . Chicken embryos inoculations with T . gondii tachyzoites at stages earlier than 12 days old are technically challenging given the fragility of their blood vessels , which is also the reason why repeated doses of auranofin are not possible . In contrast , the standard mouse animal model of acute toxoplasmosis requires at least 10 days of daily therapy and subsequent post-treatment follow up in order to determine the efficacy of the study drugs to eradicate parasite load and ensure survival of the mice . We are planning further experiments in this standard model in the future . Given its effect on both the parasite and the host , auranofin stands out as a unique anti-parasitic agent: it can protect sanctuary organs such as the brain , where the host‘s own protective inflammatory responses might cause further organ damage . Additional pharmacokinetic studies for auranofin in the CNS are necessary in order to establish its bioavailability in the setting of an abnormally permeable blood brain barrier during a CNS infection . This is particularly important , since most pharmacokinetic studies on auranofin were performed in the early ‘80 s [14]–[16] in uninfected animals with normal blood brain barriers . In summary , these results reveal significant in vitro and in vivo activity of auranofin against T . gondii , suggesting that it may be an effective alternative treatment for acute toxoplasmosis in the future .	Toxoplasma gondii is a protozoan parasite that infects at least two thirds of the world human population . Once it infects the human host , it has great predilection for the brain and the retina of the eye . It remains latent until the host's immune system weakens , and then causes organ tissue damage . There are very few treatments available that are active against this parasite , and they all fail to eradicate it from the human body . Hence , there is always a risk for recurrence and/or disabling long-term complications such as blindness or neurological abnormalities . Despite this fact , it has been over fifty years since most anti-Toxoplasma agents were initially described . Most recently , in an attempt to expedite the process of drug discovery , older drugs are making a comeback by being re-purposed for new diseases . Auranofin , which was originally designed to treat rheumatoid arthritis , has consistently shown antiparasitic activity against multiple organisms , including parasites of great public health importance such as Plasmodium , Schistosoma and Leishmania , although most of these reports are based on in vitro assays . Herein , we present our studies that demonstrate that auranofin is active against Toxoplasma gondii in vitro and in an animal model of acute Toxoplasma infection , suggesting that auranofin has great potential to become a new anti-Toxoplasma agent .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "infectious", "diseases", "infectious", "diseases", "of", "the", "nervous", "system", "medicine", "and", "health", "sciences", "neglected", "tropical", "diseases", "tropical", "diseases", "parasitic", "diseases" ]	2014	Auranofin Is Highly Efficacious against Toxoplasma gondii In Vitro and in an In Vivo Experimental Model of Acute Toxoplasmosis
Clostridium difficile is a Gram-positive spore-former bacterium and the leading cause of nosocomial antibiotic-associated diarrhea that can culminate in fatal colitis . During the infection , C . difficile produces metabolically dormant spores , which persist in the host and can cause recurrence of the infection . The surface of C . difficile spores seems to be the key in spore-host interactions and persistence . The proteome of the outermost exosporium layer of C . difficile spores has been determined , identifying two cysteine-rich exosporium proteins , CdeC and CdeM . In this work , we explore the contribution of both cysteine-rich proteins in exosporium integrity , spore biology and pathogenesis . Using targeted mutagenesis coupled with transmission electron microscopy we demonstrate that both cysteine rich proteins , CdeC and CdeM , are morphogenetic factors of the exosporium layer of C . difficile spores . Notably , cdeC , but not cdeM spores , exhibited defective spore coat , and were more sensitive to ethanol , heat and phagocytic cells . In a healthy colonic mucosa ( mouse ileal loop assay ) , cdeC and cdeM spore adherence was lower than that of wild-type spores; while in a mouse model of recurrence of the disease , cdeC mutant exhibited an increased infection and persistence during recurrence . In a competitive infection mouse model , cdeC mutant had increased fitness over wild-type . Through complementation analysis with FLAG fusion of known exosporium and coat proteins , we demonstrate that CdeC and CdeM are required for the recruitment of several exosporium proteins to the surface of C . difficile spores . CdeC appears to be conserved exclusively in related Peptostreptococcaeace family members , while CdeM is unique to C . difficile . Our results sheds light on how CdeC and CdeM affect the biology of C . difficile spores and the assembly of the exosporium layer and , demonstrate that CdeC affect C . difficile pathogenesis . Clostridium difficile [1] , first reclassified as Peptoclostridium difficile [1] and more recently re-classified as Clostridioides difficile [2] , is a Gram-positive , sporogenic anaerobic bacterium that is the most common cause of antibiotic-associated diarrhea within healthcare systems of the developed world [3 , 4] . The clinical manifestation of the infection is diarrhea and in severe cases can produce pseudomembranous colitis , toxic megacolon and death [5] . Mortality of C . difficile infections ( CDI ) may reach up to 5% of CDI cases , but in several outbreaks , it has increased up to 20% [3] . Conventional metronidazole and/or vancomycin treatment ( depending on the severity of the symptoms ) although resolve single episodes of CDI , exhibit high rates of recurrence of the infection after a first episode . The rate of recurrence of CDI of a first , second and third episode may reach up to 20% , 40% and 60% , respectively [6 , 7] . During the infection , C . difficile colonization leads to secretion of large toxins ( TcdA and TcdB ) that glycosylated intestinal epithelial cell proteins , induce massive inflammation of the gut epithelium , causing disease symptoms ranging from mild diarrhea to pseudomembranous colitis , toxic megacolon and even death [8] . However , before C . difficile can colonize a susceptible host , the highly resistant and metabolically dormant spore must germinate in response to secondary bile salts present in high levels in the gastrointestinal tract of antibiotic-treated host [9 , 10] . In addition to toxin-production during C . difficile colonization of the host , a subset of C . difficile vegetative cells initiates a sporulation program that culminates with the formation of metabolically dormant spores [11 , 12] . These spores have intrinsic resistance properties enabling their survival to enzymatic degradation [13 , 14] , phagocytic cells [15] and chemicals normally found in the host´s gastrointestinal ( GI ) environment [16] , enabling their persistence in the host´s GI tract . To persist in the host , C . difficile spores must interact with the host´s colonic mucosa through specific interactions mediated by spore-ligand ( s ) molecules and host cellular receptor ( s ) [17] . In this context , as demonstrated in other spore-former species [18] , the surface of C . difficile spores is likely to be the primary site of spore-host interactions that contributes to spore persistence . Consequently , there is keen interest to understand fundamental aspects of the outermost exosporium layer of C . difficile spores [19] . Notably , the exosporium layer of C . difficile spores differs from previously described outermost layers [19–21] . For example , in contrast with the exosporium layer of spores of the Bacillus cereus group , where an interspace gap separates the exosporium from the spore coat [18 , 19 , 22] , the exosporium of C . difficile spores is in direct contact with the spore coat layers in a similar fashion as the outer crust of Bacillus subtilis spores [13 , 18 , 19] . Despite these differences with the outer layer of spores of other bacterial endospore formers , the exosporium layer of most C . difficile strains have hair-like extensions similarly as those observed in spores of the B . cereus group [19 , 22] . However , in striking difference from other endospore formers , during the sporulation program , C . difficile forms spores with two distinctive exosporium morphotypes that arise from the same clonal sporulating culture , during either standard sporulation conditions ( i . e . , agar plates ) , or during biofilm development conditions [20 , 21] . These exosporium morphotypes include: i ) spores with a thick-exosporium layer , defined by an electron dense material surrounding the spore coats; and ii ) a thin-exosporium layer , where the electron-dense material that surrounds the spore coat is notably thinner [20 , 21] . Recently , the composition of the outermost exosporium layer of C . difficile spores of the laboratory 630erm strain has been determined with several interesting features [23] . Orthologs of the BclA family of proteins have been identified , yet the structural proteins known to be involved in the exosporium assembly of the exosporium layer of the B . cereus group , are absent in the C . difficile exosporium proteome [23] . Moreover , CdeC , CdeM , CdeA and CdeB were shown to be uniquely localized in the exosporium layer of C . difficile 630erm spores and accessible to antibodies [23] , suggesting exposure to the spore-surface; of these , CdeC and CdeM exhibited an unusually high content of cysteine residues [23] . Cysteine-rich proteins have been reported to be essential for the assembly of the exosporium in B . anthracis spores ( i . e . , ExsY ) [24 , 25] and of the outer crust layer in B . subtilis spores ( i . e . , CotY and CotZ ) [26] . In B . subtilis , the cysteine-rich proteins of the spore crust , CotY and CotZ , are capable of cooperatively self-assembling into thermally stable structures favored by strong disulfide cross-linking [27] . Studies on the outer spore layer of C . difficile have shown that 630erm strain forms spores that albeit have both exosporium morphotypes , they lack the hair-like projections observed in most epidemic strains [19–21] , suggesting that the mechanisms underlying exosporium assembly might exhibit slight difference between both strains . For example , the cysteine-rich protein , CdeC , shown to be required for the morphogenesis of the coat and exosporium layer of spores of the epidemically relevant R20291 strain [13 , 19] , is present at 100-fold higher levels in 630erm spores compared to R20291 spores [23] and exhibits a deletion in the N-terminal domain ( S3 Fig ) . The only known functional role of CdeM is that inactivation of cdeM leads to a loss of competitive fitness during infection of germ free mice [11] . Consequently , it is likely that both cysteine-rich proteins , CdeC and CdeM , might be involved in the differences observed between the exosporium layer of 630erm and R20291 spores . In this context , in our systematic approach to gain more insight into the mechanisms of assembly of the exosporium layer of C . difficile spores , the aim of this work was to address the functional roles of CdeC ( i . e . , CD1067 in 630erm strain ) and a novel morphogenetic factor , CdeM ( CD1581 in strain 630erm ) . Using a series of microscopic , genetic , molecular biology and cellular biology assays , we have characterized the cdeC and cdeM phenotypes and demonstrate their implications in the assembly of the exosporium layer and C . difficile spore biology . We also demonstrate that the absence of CdeC and CdeM differentially affect in vivo spore adherence , infection recurrence , and fitness in a series of mouse models , contributing to understand their implications in C . difficile pathogenesis . A recent proteomic study [23] identified two cysteine rich proteins ( i . e . , CdeC [CD1067] and CdeM [CD1581] ) which were uniquely located in the outermost exosporium layer of 630erm spores [23] . Functional analysis of CdeC in the epidemic R20291 strain demonstrated that this protein is required for the correct assembly of the exosporium layer of R20291 spores [13] . However , the higher levels of CdeC observed in 630erm spores , suggests that CdeC might have a more predominant role in the assembly of the exosporium layer in 630erm spores , while the role of CdeM remains unclear . Both proteins are encoded by monocistronic genes whose promoters are controlled by the late-mother cells specific sigma factor , σK ( Fig 1A ) , which the late-mother cells specific [28 , 29] . cdeC in 630erm is flanked by genes encoding uncharacterized proteins transcribed by σE-regulated promoters; by contrast , cdeM , located 570 , 775 bp downstream of cdeC , is flanked by genes encoding enzymes involved in amino acid biosynthesis ( Fig 1A ) . The 1218-bp cdeC gene encodes a 405-amino acid protein with a predicted molecular weight of 44 . 7-kDa , and a high content of cysteine residues ( 9% of the amino acid content ) , suggesting that it might be prone to disulfide bridge formation and therefore , play a role in the crosslinking of other exosporium proteins [27] [26] . Analysis of the amino acid sequence revealed no conserved domains , but several noteworthy sequence repeats conserved in all sequenced genomes of C . difficile: i ) in the N-terminal domain ( NTD ) two motifs of unknown function were identified ( i . e . , KKNKRR and three consecutive histidine residues ) ; ii ) a 3xHistidine repeat near the NTD; iii ) in the central region , a 6 NPC repeat followed by two CCRQGKGK repeat; and iv ) cysteine rich sequence CNECC at the C-terminal domain ( CTD ) of CdeC ( Fig 1A ) . The 483-bp cdeM gene encodes a 161-amino acid encoded protein with a predicted molecular weight of 19 . 1-kDa , and a high content of cysteine residues ( 8 . 7% of the amino acid sequence ) , suggesting that CdeM , similarly as CdeC , might also be prone to disulfide bridge formation contributing to the crosslinking of exosporium proteins . Analysis of the primary sequence of CdeM gave no conserved domains , but some interesting features: i ) three RREA repeats near the NTD of CdeM; ii ) two NGNNGGNNNNC and three CHK repeats in the central region of CdeM; and iii ) two CNCCNCCRK repeats at the CTD ( Fig 1A ) . Since we observed unique sequences in these two proteins , we wondered how conserved the CdeC and CdeM was among other C . difficile and related Peptostreptococcaceae family members , due to a recent reclassification of C . difficile into the Clostridioides genus , a member of the Peptostreptococcaceae family rather than in the Clostridiaceae family [1] . To assess the conservation of the cysteine rich proteins , CdeC and CdeM , in other Clostridial organisms , we searched for protein homologues to the C . difficile CdeC and CdeM in a blastp search ( Fig 1B ) . This analysis was performed in a chosen subset of strains of a wide variety of ribotypes and C . difficile genome groups ( S2 Table , S1 Fig ) ; both , cdeC and cdeM , were found to be conserved in all C . difficile isolates tested ( Fig 1B . Interestingly both , CdeC and CdeM , and their unique repeats were present in all C . difficile strains analyzed ( S2 and S3 Figs ) . Three out of 15 strains encoded a CdeC with a truncated NTD ( S2 Fig ) , while five out of 15 C . difficile strains had an insertion in the NTD of CdeM ( S3 Fig ) . Taken together , CdeC and CdeM are highly conserved in C . difficile representative strains . When a blastp against C . difficile CdeC and CdeM was expanded to include additional members of the Peptostreptococcaceae , we observed that CdeC was conserved in all 8 Peptostreptococcaceae family members analyzed ( Figs 1B and S2 , S2 , S3 and S4 Tables ) . By contrast , CdeM was unique to C . difficile ( Figs 1B , S2 , S3 and S4 ) . Notably , despite the absence of CdeM , the genomes of Clostridioides mangenotii , Paraclostridium bifermentas , Paraclostridium sordellii , Peptostreptococcaceae bacterium , Terrisporobacter othiniensis , Paraclostridium benzolyticum had different CdeC variants with most of the sequence motifs conserved ( Figs 1B , S2 , S3 and S4 ) . These results collectively suggest that , while CdeM is specific for C . difficile , CdeC is a conserved exosporium protein in members of the Peptostreptococcaceae family . We sought to apply a similar analysis to a subset of Clostridiaceae and Lachnospiraceae family members to evaluate whether C . difficile CdeC and CdeM were present ( S4 and S5 Tables ) . Strikingly , only CdeC but not CdeM , was found in members of the Clostridiaceae family , specifically in Clostridium dakarense and 5 Clostridium sp . ( Fig 1B and S4 and S5 Tables ) . Despite the phylogenetic divergence ( S5 Fig ) , the cysteine residues in the conserved motifs of CdeC are highly conserved in members of the Peptostreptococcaceae and Clostridiaceae families ( S6 and S7 Figs ) . CdeC and CdeM were not present in members of the Lachnospiraceae family . Collectively , these results indicate that although CdeC is present in a few members of the Clostridiaceae family , the amino acid sequence is highly conserved in them . To evaluate the functional role of CdeC and CdeM in C . difficile 630erm strain , we used ClosTron mutagenesis by redirecting the group II L1 . ltrB intron into the antisense strands of the N-terminal domain of both genes at positions 30 and 123 to inactivate cdeC and cdeM , respectively ( S8A , S8B and S8C Fig ) . After many attempts to inactivate each individual gene , we were able to obtain several independent mutant clones of cdeC and cdeM as shown by PCR screening for insertions ( S8A and S8B Fig ) [33] . Mutants were confirmed by PCR using flanking primers and sequencing of the PCR amplicons ( S8A and S8B Fig ) . Clones C2 , C4 and C8 for cdeC mutant strain and C2 , C3 and C4 for the cdeM mutant strain . These clones were used for further phenotypic characterization . Unlike the exosporium layer of most epidemic strains , 630erm spores have an exosporium layer that does not exhibit bumps and the typical hair-like extensions [19 , 20] , and also have higher levels of CdeC in the spore surface layers than R20291 spores [23] . Given these differences , we hypothesized that CdeC would have a greater impact in exosporium and spore coat assembly than previously observed in epidemic R20291 strain [13] . Insertional inactivation of cdeC lead to the formation of cdeC spores with an outermost exosporium layer ( i . e . , 29 . 6 nm ) that was 50% thinner than wild-type spores ( i . e . , 55 nm ) ( Fig 2B ) . We observed that inactivation of cdeC in 630erm spores affected the thickness of the spore coats ( Fig 2B ) to a greater extent than in our previous observations in previous observation in C . difficile R20291 epidemic strain [13] . A significant decrease of 32% ( wild-type 32 . 8 nm and cdeC 22 . 1 nm ) in the thickness of the external spore coat was evidenced in cdeC spores compared to wild-type spores , while an increase of 35% in the thickness of the inner spore coat was observed in cdeC spores compared to wild-type spores ( i . e . , wild-type , 22 . 5 nm; cdeC , 30 . 6 nm ) ( Fig 2B ) . Despite these differences , the overall thickness of the spore coat ( i . e . , inner coat plus outer coat ) remained similar between wild-type ( i . e . , 55 . 3 nm ) and cdeC ( i . e . , 52 . 7 nm ) spores ( Fig 2B ) . Collectively , these observations indicate that: i ) CdeC affects the exosporium assembly and the thickness of the inner and external spore coat of 630erm spores; ii ) the impact of insertional inactivation of cdeC in the thickness of the inner and external spore coat is greater in 630erm spores than in epidemic R20291 spores [13] . To explore the impact of insertional inactivation of cdeM in the assembly of the exosporium layer of C . difficile spores , cdeM spores were also analyzed by transmission electron microscopy . Strikingly , analysis of more than 50 individual cdeM spores revealed that inactivation of cdeM yielded spores with almost complete absence of the exosporium layer ( Fig 2A ) . Upon comparison of the thickness of the exosporium layer of wild-type and cdeM spores ( Fig 2A ) , we evidenced a striking decrease of 85% in the thickness of the exosporium layer of cdeM spores ( i . e . , 8 . 1 nm ) compared to that of wild-type spores ( i . e . , 55 nm ) ( Fig 2B ) . In contrast to the effect of inactivation of cdeC on the spore coat , inactivation of cdeM led to a slight but significant increase in the thickness of the external spore coat layer , from 32 . 8 nm ( i . e . , wild-type spores ) to 36 . 4 nm ( i . e . , cdeM spores ) ( Fig 2B ) . Conversely , a significant decrease in the thickness of the inner spore coat from 22 . 5 nm ( i . e . , wild-type spores ) to 16 . 5 nm ( i . e . , cdeM spores ) was observed ( Fig 2B ) . Despite these differences , the overall thickness of the spore coat varied slightly from 55 . 3 nm in wild-type spores to 52 . 9 nm in cdeM spores . Collectively , these observations clearly indicate that CdeM is essential for the morphogenesis of the exosporium layer and , affects to some degree the assembly of the spore coat layer of 630erm spores . The morphological defects observed as described above suggest that CdeC and CdeM are surface proteins . Indeed , previous work has demonstrated that CdeC and CdeM are located mainly in the exosporium layer [23] . To evaluate whether CdeC is surface-located , immunofluorescence of wild-type and cdeC spores; significant immunofluorescence signal was detectable in wild-type spores , while no detectable fluorescence signal was evidenced in cdeC mutant spore ( Fig 2C ) . Similarly , immunofluorescence assay with anti-CdeM detected immunofluorescence signal in wild-type but not in cdeM spores ( Fig 2D ) . These results indicate that both cysteine-rich proteins are accecible to antibodies . The fact that cdeC and cdeM spores had defective exosporium layers suggested that the protein profile of cdeC and cdeM spores might differ from that of wild-type spores . Reasoning that the protein profile would differ due to the mutations , we standardized the amounts of spores loaded by optical density , ensuring that the same number of spores were loaded in each lane . Our first observation from the SDS-PAGE analysis of the Laemmli buffer-extracted spore coat and exosporium proteins from wild-type spores was that the protein profile of 630erm spores differed from the previously reported one from R20291 strain [13] . Analysis of the spore coat and exosporium extracts of cdeC spores revealed the appearance of 6 major protein species of molecular weights estimated in 150- , 58- , 53- , 50- , 18- and 16-kDa , levels of which decreased to 34 , 12 , 16 , 34 , 63 and 28% relative to wild-type levels ( Fig 3A and 3B ) . Strikingly , complementation of cdeC mutation , albeit had no effect on the levels of the 18- and 16-kDa protein species , and increased the levels of 150- , 50-kDa proteins but not to wild-type levels ( Fig 3B ) . A similar protein profile was observed in Laemmli-extracts of the spore coat and exosporium ( remnants ) extracts of cdeM spores; the levels of the protein species of 150- , 58- , 53- , 50- , 18- and 16-kDa were decreased to 78 , 88 , 77 , 66 , 56 and 6% relative to levels in wild-type spores ( Fig 3A ) . Complementation of the cdeM mutation increased the levels of most of the dominant protein species to levels near or higher than those in wild-type spores ( Fig 3B ) . These results indicate that the absence of both cysteine rich proteins , CdeC and CdeM , affect the relative abundance of the major protein species in the spore coat and exosporium extracts . Previous work , using an anti-630erm spore goat antiserum [13 , 34] , demonstrated that the immunodominant proteins are located in both , the spore coat and exosporium layer [13 , 34] . Therefore , since inactivation of cdeC and cdeM affected the assembly of the exosporium layer of 630erm spores , we evaluated how their inactivation affects the presence of immunodominant proteins in the spore coat and exosporium extracts analyzed by western blots with anti-spore goat serum . Several loading controls of C . difficile spores have been applied recently to normalize immunoreactive intensities . Given the defects observed in the spore coat and exosporium in cdeC and cdeM spores , we first sought to evaluate whether mutations in cdeC and cdeM would affect the abundance of a loading control protein , SpoIVA , which has been used as a loading control in several studies [35 , 36] . Notably , inactivation of cdeC caused a ~7-fold increase on the levels of SpoIVA , complementation of cdeC with wild-type cdeC restored SpoIVA levels to near wild-type level ( Fig 3C ) . By contrast , inactivation of cdeM had no effect on SpoIVA levels , and complementation of cdeM with wild-type cdeM did not affect SpoIVA levels ( Fig 3C ) . Therefore , to analyze the relative amounts of immunoreactive proteins we loaded similar amounts of spores based on optical density measurements . Analysis of the spore coat/exosporium extracts of cdeC spores revealed that the levels of the 180- and 107-kDa immunoreactive protein species significantly decreased 35 and 50% relative to that of wild-type spores , respectively ( Fig 3D ) . Levels of the 103-kDa immunoreactive protein species increased by ~2-fold relative to wild-type spores ( Fig 3D ) . Complementation of cdeC with wild-type cdeC had no effect on the levels of the immunoreactive proteins of 180- and 107-kDa; however , the levels of the 103-kDa immunoreactive protein species were restored to wild-type levels ( Fig 3D ) . Analysis of the spore coat/exosporium extracts of cdeM spores revealed that the levels of the 180- and 107-kDa , but not 103-kDa , immunoreactive protein species significantly increased by 9- and 2 . 5-fold relative to wild-type levels ( Fig 3D ) . Complementation of cdeM lead to spores with wild-type levels of all three immunoreactive protein species ( Fig 3D ) . Collectively , these results indicate that: i ) CdeC is required for the normal levels of immunoreactive protein species of the outer layers of C . difficile spores; ii ) absence of CdeM leads to spores with increased levels of immunoreactive proteins . The spore coat of C . difficile spores acts as an impermeable barrier to enzymes with molecular masses higher than 14 kDa , such as lysozyme , proteinase K and trypsin [14] . The impact of insertional inactivation of cdeC and cdeM in the protein profile of spore coat/exosporium extracts raised the question of whether absence of CdeC and/or CdeM would impact the permeability of the spore coat to lysozyme triggered-germination . Hence , to answer this question , we explored a lysozyme permeability assay of cdeC and cdeM mutant spores . After treatment of wild-type spores with 1 mg/mL of lysozyme for 5 h at 37°C , only a small fraction of spores ( 1% ) changed to phase dark ( Fig 4A and 4B ) . Contrastingly , under similar treatment conditions , ~90% of cdeC spores changed to phase dark ( Fig 4A and 4B ) . However , less than 1% of cdeM spores changed to phase dark upon lysozyme treatment ( Fig 4A and 4B ) . cdeC complementation partially restored the resistance of the spore coat to lysozyme , where only 34% of the spores became phase dark ( Fig 4A and 4B ) . Despite the negligible effect of a cdeM mutation in lysozyme resistance , complementation of cdeM strain with wild-type cdeM caused 38% of the spores to become phase dark after lysozyme incubation ( Fig 4B ) . Altogether , these results indicate that , despite the impact of both cysteine-rich proteins ( i . e . , CdeC and CdeM ) on the spore coat and exosporium proteins , only the absence CdeC increases the permeability barrier of the spore coat to lysozyme , which is consistent with those results previously reported for a insertional inactivation of cdeC in epidemic R20291 spores [13] . The previous work in spores of the epidemic strain R20291 demonstrated that inactivation of cdeC led to spores with an increased sensitivity to ethanol and heat resistance [13] . First , we evaluated whether absence of CdeC and/or CdeM affected ethanol resistance of C . difficile 630erm spores . Hence , when wild-type spores were treated with ethanol for 1 h at 37°C , spore viability decreased by 0 . 2 log reduction ( Fig 5A ) . When cdeC spores were treated with ethanol under similar conditions , a significant decrease of 2 log cycles was observed ( Fig 5A ) . By contrast , no significant difference in loss of spore viability was observed between wild-type and cdeM spores after ethanol-treatment ( Fig 5A ) . These results indicate that CdeC increases ethanol-killing , presumably via an increase in the permeability of the spore inner membrane . To gain more insight of the effects of CdeC and CdeM on resistance of C . difficile spores , heat resistance of wild-type , cdeC and cdeM spores at 75°C was assessed . Heat treatment ( 75°C ) of wild-type spores progressively decreased spore viability ( Fig 5B ) ; after 60 min of treatment , only 4 . 5% of wild-type spores remained viable ( Fig 5B ) . Upon heat treatment of cdeC spores , higher levels of inactivation became evident as early as 5 min after treatment ( Fig 5B ) ; after 60 min at 75°C only 0 . 06% of cdeC spores remained viable ( Fig 5B ) . When cdeM spores were subjected to similar heat treatment conditions , a significantly higher extent of inactivation than wild-type was observed after 5 min at 75°C ( Fig 5B ) . After 60 min at 75°C , only 0 . 5% of cdeM spores remained viable , amount that was 10-fold lower than wild-type spores but 10-fold higher than cdeC spores ( Fig 5B ) . To address whether the decreased heat resistance of cdeC and cdeM spores was attributed to the levels of dipicolinic acid ( DPA ) , spores of all strains were assayed for spore-core DPA content , yet no significant difference was observed in spore-core DPA content between the strains ( Fig 5C ) . These results indicate that the absence of both exosporium morphogenetic proteins affect the resistance of C . difficile spores to heat . C . difficile spores are resistant to phagocytic cells , and capable of surviving for more than 48 h without significant macrophage-mediated inactivation of C . difficile spores [15] . Therefore , we assessed whether the inactivation of cdeC and cdeM affected the viability of C . difficile spores during infection of Raw 264 . 7 macrophage-like cells . As expected , infection of Raw 264 . 7 cells with wild-type spores led to no significant spore-inactivation after 24 h of infection . Notably , a slight but significant increase in spore colony formation was observed after 48 h of infection ( Fig 5D ) , suggesting that macrophage factors activated C . difficile spores to germinate in BHIS plates supplemented with taurocholate . Strikingly , while no significant inactivation of cdeC spores was observed after 24 h of infection of Raw 264 . 7 murine macrophage-like cells , ~1 log reductions in spore viability were observed after 48 of infection , respectively ( Fig 5D ) . By contrast , no inactivation of cdeM spores was evidenced upon infection of Raw 264 . 7 macrophage-like cells after 48 of infection ( Fig 5D ) . Collectively , these results indicate that the absence of CdeC , but not CdeM , leads to C . difficile spores susceptible to macrophage-killing . Previous work demonstrated that inactivation of cdeC in R20291 epidemic strains lead to an increased adherence to components of the intestinal mucosa ( i . e . , mucin , fibronectin and adherence to intestinal epithelial Caco-2 cells ) [17] , suggesting that CdeC contributes to decrease the persistence of C . difficile spores in the intestinal tract . To begin answering this question , we used a colonic loop mouse model to evaluate the impact of an insertional inactivation of cdeC and cdeM in C . difficile spore adherence to healthy intestinal mucosa by confocal fluorescence microscopy ( S9 Fig ) . In contrast to our expected results , data shown in Fig 6 demonstrates that cdeC mutant spores have significantly reduced adherence in comparison to wild-type spores ( Kluskal Wallis test P = 0 . 036 ) ( Fig 6A , 6B and 6D ) , while cdeM mutant spores seemed to adhere lower than wild-type to the colonic mucosa ( Kluskal Wallis test P = 0 . 101 ) ( Fig 6A , 6C and 6D ) . These data indicate that , in a healthy colonic mucosa , CdeC , and perhaps CdeM , contribute to reduce the adherence of C . difficile spores to the colonic mucosa , contrasting with the proposed observations from in vitro studies [17] . As mentioned , the absence of a correctly assembled exosporium layer affects spore adherence to healthy colonic mucosa . Therefore , to investigate the implication of CdeC and CdeM in an infectious context , we used a mouse model of infection and recurrent infection of C . difficile . Antibiotic-treated mice were infected with C . difficile spores of wild-type ( n = 6 ) , cdeC ( n = 6 ) , and cdeM ( n = 5 ) , and at day 3 of infection , mice were treated with vancomycin for 5 days and subsequently monitored to evaluate the recurrence of the infection ( Fig 7A ) . Mice infected with wild-type and cdeC spores yielded more animals developed significantly higher diarrhea scores than those infected with cdeM spores ( Fig 7B ) . Mice infected with cdeC spores also had higher weight lost than those infected with wild-type and cdeM spores ( S12A Fig ) . Recurrence was observed after vancomycin treatment as described in Fig 7A . Diarrhea became evident at day 4 after vancomycin treatment , and 6 of 6 ( 100% ) of the mice infected with cdeC developed recurrent diarrhea , whereas only 3 of 6 ( 50% ) and 3 of 5 ( 60% ) of the mice infected with wild-type and cdeM spores developed recurrent diarrhea ( Fig 7C ) . Mice infected with cdeC spores also had higher diarrhea score than those infected with wild-type and cdeM spores ( Fig 7C ) . The higher recurrence rate in mice infected with cdeC spores correlated with higher toxin titer ( Fig 7D ) and CFU ( Fig 7E ) recovered post-mortem from cecum contents . To further evaluate whether the increased colonization of cdeC spores could be attributed due to differences in spore germination , we evaluated whether inactivation of cdeC and cdeM affected spore germination . A reduced extent of germination in cdeC spores versus wild-type spores was evidenced in the presence of taurocholate after 60 min of incubation ( S10A Fig ) . By contrast , no significant germination defect was evidenced in cdeM spores , which germinated similarly as wild-type spores ( S10B Fig ) . It is also noteworthy that the colony formation efficiency of cdeC and cdeM spores in BHI agar plates with taurocholate was 25±5 and 50±5% relative to that of wild-type spores , respectively . Note that cytotoxic assay of culture supernatant on Vero cells showed no difference between strain ( S11 Fig ) and therefore , these parameters were not responsible for the differences observed in the in vivo severity and cytotoxic between strains . We also found no differences in the levels of fecal C . difficile spore loads and anti-vegetative and -spore antibodies raised during the infection ( S12 Fig ) . Taken together , these data indicate that during the infection , insertional inactivation of cdeC , but not cdeM , leads to increased colonization and recurrence of the diarrhea after vancomycin treatment . To gain more insight on how the absence of CdeC affected C . difficile colonization , we performed a competitive assay where healthy C57BL/6 mice ( n = 10 per group ) were orally infected after antibiotic cocktail treatment with an equal number of viable wild-type and cdeC or wild-type and cdeM spores ( 1 x 107 spores of each strain ) , and the numbers of fecal-shedded spores were monitored for 8 days after the challenge . cdeC spores were detected at significantly higher levels than wild-type spores at days 1 , 2 and 4 post-challenge ( Fig 8A and 8C ) . Interestingly , the persistence dynamics of cdeM strain differed from that of cdeC strain; cdeM spores were present at significantly lower levels than 630erm spores in fecal sampled only at day 4 post infection ( Fig 8B and 8D ) . These results suggest that absence of CdeC , but not CdeM , increases the fitness of C . difficile during infection . To gain a better understanding on how these cysteine-rich proteins affect the assembly of the exosporium layer , we sought to evaluate the abundance of known proteins of the exosporium layer ( i . e . , BclA1 , BclA2 , BclA3 , CdeA , CdeB , and CdeM ) and of the spore coat ( i . e . , CotA and CotB ) proteins [23] , by using wild-type and cdeC mutant spores containing plasmids expressing FLAG fusion proteins ( S1 Table ) . First , we evaluated whether the absence of CdeC and/or CdeM affected the abundance of the collagen-like BclA glycoproteins . All three BclA proteins were detectable in wild-type spores; BclA1 and BclA3 were detected forming high molecular mass complex of 110-kDa as well as a low molecular mass species of 48-kDa , while BclA2 was detectable as a 48-kDa species ( S13A , S13B and S13C Fig and S14A , S14B and S14C Fig ) . In the absence of CdeC or CdeM , a significant reduction in the high molecular mass complex of both , BclA1 and BclA3 , was evidenced ( Table 1 , S13A and S13C Fig , S14A and S14C Fig ) . By contrast , absence of CdeC leads to an increase in low molecular mass complex of all three BclA orthologues , whereas absence of CdeM leads to a decrease in the low molecular mass complex of all three BclA proteins ( Table 1 , S13A , S13B and S13C Fig , S14A , S14B and S14C Fig ) . Note that further dilution of the amount of anti-flag used provides similar results in the case of BclA1-FLAG ( S15 Fig ) . These results demonstrate that: i ) CdeC is essential for the presence of the high , but not low , molecular mass complexes of all three BclA proteins , while CdeM is essential for the presence of high and low molecular mass complexes of all three BclA proteins . As previously described [23] , the cysteine-rich protein , CdeA , was found in the spore surface as a 19- and 47-kDa immunorreactive species ( Table 1 , S13A , S13B , S13C Fig and S14A , S14B and S14C Fig ) . Absence of CdeC or CdeM lead to a significant increase of 47-kDa CdeA species , and a significant decrease of the 19-kDa CdeA species ( Table 1 , S13D Fig and S14D Fig ) . Another exosporium protein previously identified is CdeB , which was found to be present in wild-type spores as a 48-kDa immunoreactive species as previously described [23] . Notably , while the absence of CdeC lead to a significant increase of CdeB , the abundance of CdeB in absence of CdeM lead to lower levels of CdeB compared to wild-type spores ( Table 1 , S13E Fig and S14E Fig ) . Note that further dilution of the amount of anti-flag used provides similar results in the case of CdeA-FLAG ( S15 Fig ) . These data indicate that the levels of CdeA and CdeB are affected by CdeC and CdeM . The aforementioned results suggest that levels of CdeC depend on the presence of CdeM or vice versa . To explore this hypothesis , levels of CdeC in cdeM spores relative to wild-type and levels of CdeM in cdeC spores relative to wild-type spores were assessed . Results evidenced that while a significant increase of CdeM was observed relative to wild-type spores ( Table 1 , S13F Fig ) . By contrast , a significant decrease in high ( 120-kDa ) and low ( 44 kDa ) molecular mass CdeC species was evidenced in cdeM spores relative to wild-type spores ( Table 1 , S14F Fig ) , indicating that spore levels of CdeC depend on CdeM . The altered thickness of cdeC spores evidenced by transmission electron micrographs suggest that the absence of CdeC might affect the levels of spore coat proteins . To address this question , we evaluated the levels of two spore coat proteins ( i . e . , CotA and CotB ) [37] . CotA and CotB were present as 47-kDa immunoreactive protein species , as reported previously in wild-type spores [23] . CotA was found at similar levels in cdeC spores relative to wild-type , but significantly lower levels of CotB were observed in cdeC spores compared to wild-type spores ( Table 1 , S13G and S13H Fig ) . Next , we addressed whether the absence of CdeM affected CotA and CotB levels . As shown in Table 1 ( S14G and S14H Fig ) , cdeM spores had significantly lower levels of both CotA and CotB than wild-type spores ( Table 1 , S14G and S14H Fig ) . These results indicate that only CdeM affects CotA , but that CdeC and CdeM affect CotB . C . difficile spores exhibit an outermost exosporium layer that provides the first site of interaction with the host . Recent studies on the outermost exosporium layer of C . difficile spores have uncovered the ultrastructural variability , composition and functional properties of this layer [14 , 20 , 23 , 38–40] . Extensive studies have demonstrated that cysteine-rich proteins have been involved in the assembly of the exosporium layer of spores of members of the B . cereus group and in the outer crust layer of B . subtilis spores [18 , 24–26] . In C . difficile , there are three cysteine-rich proteins identified in the exosporium layer of C . difficile spores , CdeC , CdeM and CdeA [23] . Previously , we demonstrated that CdeC is an exosporium morphogenetic protein in epidemic C . difficile strain R20291 by performing functional analysis of a cdeC mutant strain [13] . In this work , we have used the laboratory strain 630erm due to its genetic ease , to investigate how two exosporium cysteine-rich proteins , CdeC and CdeM , contribute differentially to the spore biology and pathogenesis of C . difficile: CdeC and CdeM are both required for the correct formation of the exosporium layer . Whereas cdeC mutant exhibit defective spore coat assembly ( Fig 2A and 2B ) and permeability to lysozyme ( Fig 4A and 4B ) , increased susceptibility to ethanol , heat- and macrophage-inactivation ( Fig 5A , 5B and 5C ) , cdeM spores behaved as wild-type spores . Notably , CdeC is specific to C . difficile and related Peptostreptococcaceae family members , while CdeM is specific to C . difficile ( Fig 1 ) . In a healthy colonic mucosa , spore adherence of cdeC and cdeM spores was lower than wild-type spores ( Fig 6 ) ; while during infection cdeC mutant , but not cdeM , exhibited higher diarrhea score , and persistence during recurrence of infection ( Fig 7 ) . In concordance , cdeC mutant , but not cdeM mutant , exhibited increased fitness in a competitive infection mouse model . Thus , this work contributes to our understanding on the mechanisms underlying exosporium assembly , and how this impacts C . difficile spore biology and pathogenesis . It was surprising to observe that despite the fact that both , CdeC and CdeM , are cysteine rich proteins , they have cause differential impacts in the integrity of the exosporium layer and spore coat . C . difficile spores . Altogether , the results provided in Table 1 and S13 and S14 Figs allow the elaboration of a first interaction map and exosporium model ( Fig 9A ) . Reasoning that we observed that the presence of CdeC was CdeM-dependent and not vice-versa ( Table 1 , S13F Fig and S14F Fig ) , CdeC-dependent proteins were defined as those with reduced levels in a cdeC genetic background; consequently , CdeM-dependent proteins were defined as those whose abundance were reduced in a cdeM but not cdeC genetic background . In this context , suggested CdeC-dependent proteins include CdeA , CotB and the high molecular complex BclA1 and BclA3 ( Fig 9A ) . By contrast , CdeM-dependent proteins include CotA , CdeB , and the low molecular mass complex formed by BclA1 , BclA2 , BclA3 and CdeB ( Fig 9A , S13 Fig and S14 Fig ) . It is noteworthy , that the high molecular , and to some extent , the low molecular mass complex formed by CdeC , are CdeM-dependent ( Fig 9 , S14F Fig and Table 1 ) . Coupling these findings with previous localization studies [23] , we propose putative locations of these proteins in the spore outer surface ( Fig 9B ) . CotA and CotB were previously shown to be located in the spore coat layers [23] , while the BclA and Cde proteins are located in the exosporium; however , the fact that the absence of CdeC affects the abundance of CotB and causes a permeable spore coat , suggests that the location of monomeric CdeC might be on the interface of the spore coat and exosporium layers , while the high molecular complex CdeC forms might be more exosporium oriented; CdeM , by contrast seems to be located uniquely on the exosporium layer . The recruitment of CotA might be related to additional unidentified proteins . Since these experiments were performed with plasmid-based complementation , we were unable to evaluate how restoring the wild-type gene into the mutant strain affected the relative abundance of FLAG-tagged proteins . A major difference between CdeC and CdeM , was that CdeC had profound implications in the assembly and permeability of the spore coat and spore resistance . It was somewhat surprising that cdeM spores had an impermeable spore coat to lysozyme , while the majority of cdeC spores germinated in the presence of lysozyme ( Fig 4 ) . A plausible explanation could be attributed to the lower levels of the CotB , additional key spore-coat constituents or to the absence of CdeC in cdeC spores . It is likely that the presence of monomeric CdeC in cdeM spores , might sufficient to be implicated in the spore coat resistance to lysozyme , or that it might be recruiting additional constituents . However , a major question that remains unanswered , is how is CdeC , but not CdeM , implicated in spore resistance ? the increased permeability of the spore coat to enzymes and of the spore inner membrane to chemicals is consistent with the elevated levels of killing of C . difficile spores to Raw 264 . 7 cells , where spores are subjected to low pH and a variety of stressors ( i . e . , release of hydrogen peroxide , lysozyme and proteases ) [41] , suggesting that CdeC is essential for C . difficile spores ability to survive host´s phagocytic cells . Dipicolinic acid is a known factor that contribute to heat resistance of C . difficile spores; thus , it was interesting to find that the levels of this molecule in the spore core was unaffected by the inactivation of cdeC and cdeM ( Fig 5 ) . Another major question raised by this work is how can the absence of CdeC , but not CdeM , contribute to a decreased spore adherence to healthy intestinal mucosa , but during infection to an increased colonization , fitness and severity of the infection and recurrence ? Our finding that cdeC spores adhere to lower levels than wild-type spores to healthy colonic mucosa in the colonic loop mouse model ( Fig 6 ) , suggests that CdeC , and/or additional exosporium proteins with reduced levels in cdeC spores , play a role in spore adherence to the colonic mucosa during health . By contrast , we observed an increased severity of the infection in mice infected with cdeC spores and increased recurrence of the infection ( Fig 7C and 7E ) as well as fitness ( Fig 8A and 8C ) . A possible explanation for these contrasting observations could be attributed to the differences between a healthy and damaged colonic mucosa . For example , during infection experiments ( Fig 7 and Fig 8 ) , C . difficile toxins TcdA and TcdB cause significant remodeling of the colonic environment , including disruption of tight junctions , mucosal ulcerations and epithelial erosion [8] . These toxin-mediated epithelium damage will in turn , expose new spore-binding sites rich in extracellular matrix components to which C . difficile spores have already been shown to bind , and that include vitronectin and fibronectin [17] . Therefore , as previously shown for cdeC mutant spores in C . difficile R20291 genetic background , which have higher affinity against components of the intestinal mucosa such as adherence to intestinal epithelial cells , fibronectin and vitronectin [17] , suggests that it is conceivable that the absence of CdeC , and/or lower additional exosporium proteins , contribute to a greater persistence of C . difficile in the host during infection , indicating that CdeC negatively contributes to C . difficile pathogenesis . In this context , the fact that 630erm spores have ~ 100-fold higher levels of CdeC in the spore surface than R20291 spores [23] , might explain why strain R20291 is able to cause more episodes of recurrent infection than 630erm strain in a mouse model [42] . An increased amount of low molecular mass immunoreactive species of BclA1 , BclA2 and BclA3 was observed in cdeC spores ( Table 1 , S13A , S13B and S13C Fig ) that might also contribute to disease . Further studies to address how CdeC , and/or BclA proteins , contribute to interactions of C . difficile spores with components of the colonic mucosa could identify mechanism through which CdeC and/or BclA proteins modulate C . difficile spore-host interactions and may also provide insight into the mechanisms underlying the reduced adherence to healthy colonic mucosa ( Fig 6 ) , increased severity of infection and recurrence ( Fig 7 ) and fitness during infection ( Fig 8 ) . In summary , in identifying two cysteine rich proteins , where one is conserved ( i . e , CdeM ) in C . difficile and the other ( i . e . , CdeC ) conserved in other Peptostreptococcaceae family members , our study provides insight into the mechanism of assembly of the exosporium layer of C . difficile spores and in the implications of these proteins during C . difficile infection . While many unanswered questions remain , the correct assembly of the exosporium layer is subjected to CdeC and CdeM , where CdeC seems to have a pleiotropic role in the assembly of C . difficile spores , contributing to spore resistance and persistence as well . By contrast , given that CdeM is unique to C . difficile , it can be considered as a potential target for spore-targeted therapies given the limited conservation of CdeM in other spore-forming organisms . All experiments using mice were conducted in agreement with the ethical standards and according to the local animal protection law . All experimental protocols were conducted in strict accordance with , and under the formal approval of the Institutional Animal Ethics Committee of the Universidad Andrés Bello ( Protocol number 020/2010 and 026/2018 ) in strict accordance to the Chilean national Law 20 . 380 . Once experiments finalized , animals were sacrificed by euthanasia by 4 times the anesthetic doses of ketamine/xylazine combinations were administered intraperitoneally . The name of the Universidad Andrés Bello Institutional Animal Care and Use Committee is: “Comité de Bioética de la Vicerrectoría de Investigación y Doctorados” . The "Comité de Bioética" provided ethical approved in the Acta # 014/2015 . Genome assemblies for selected strains ( shown in Fig 1 ) were obtained via ftp from NCBI Assembly which included genomes of 336 Peptostreptococcaceae ( taxid:186804 ) , 214 Lachnospiraceae ( taxid:186803 ) and 338 Clostridiaceae ( taxid:31979 ) . Many of these genomes were incomplete and were not annotated completely , therefore they were reannotated using Prodigal v2 2 . 6 . 3 [43] . A database of predicted proteins was created and searched locally using makeblastdb tool from the BLAST+ 2 . 3 . 0 package [44] using the C . difficile 630erm CdeC and CdeM proteins as queries ( UniProt id: Q18AS2 and Q186D6 , respectively ) . Since CdeC and CdeM have no protein motives , in order to reduce the number of false positive hits , we used blastp instead of delta- and psi-blast . Matching proteins with a threshold < 50 bits [45] . Multiple sequence alignment was performed using localpair FLAG of MAFFT v7 . 294b [46] . The inference of phylogenetic trees was calculated using distance-based UPGMA model of Segotron [47] . The logo was created using Seq2Logo V2 . 0 [48] with a minimum stack width of 0 . 1 and probability weighted Kullback-Leibler Logo . The C . difficile and Escherichia coli strains used in this study are described in S1 Table . C . difficile was routinely grown under anaerobic conditions using a gas mixture containing 90% N2 , 5% CO2 , 5% H2 . Culture medium was 3 . 7% Brain Heart Infusion supplemented with 0 . 5% yeast extract and 1% cysteine ( BHIS ) broth or on 1 . 5% agar BHIS plates . Caco-2 cells were grown in Dulbecco's modified Eagle's minimal essential medium ( DMEM ) ( Hyclone , U . S . A ) . All media were supplemented with 10% ( v/v ) fetal-calf serum ( Hyclone , U . S . A . ) , penicillin ( 100 IU mL-1 ) and streptomycin ( 100 μg mL-1 ) . Two derivatives of C . difficile strain 630erm with an intron inserted in cdeC or cdeM genes , respectively , were constructed as follows . To target the L1 . ltrB intron to each gene cdeC or cdeM , we used plasmid pDP306 and pDP370 ( S1 Table ) . Three short sequence elements from the intron RNA involved in base pairing with the DNA target sites were modified by PCR , using cdeC-specific primers P68 , P69 , P70 and universal primer IBS described elsewhere [13]; and cdeM specific primers P85 ( 5'-AAAAAAGCTTATAATTATCCTTACAGTTCGAACCTGTGCGCCCAGATAGGGTG-3' ) , P86 ( 5'- CAGATTGTACAAATGTGGTGATAACAGATAAGTCGAACCTCTTAACTTACCTTTCTTTGT-3' ) and P87 ( 5'-TGAACGCAAGTTTCTAATTTCGGTTAACTGTCGATAGAGGAAAGTGTCT-3' ) . The clostron plasmids pDP306 or pDP370 were transformed into E . coli HB101 ( pRK24 ) and subsequently transferred through conjugation to C . difficile strain 630erm . Thiamphenicol resistant clones were selected and re-grown on BHIS plates containing thiamphenicol and FeSO4 to induce expression of the Targetron system . Erythromycin-resistant clones were selected and then isolation streaked on BHIS plates supplemented with erythromycin ( 5 μg/mL ) . Positive clones were screened by colony PCR for a 2 . 8-kb insertion in cdeC with pair primer P62 ( 5'-GAATTTACTTAGCCACCGGTGTTTCGGG-3' ) , P63 ( 5'-TTTCTTCCTACTATATCTCCTAATGGGTCTAAATCG-3' ) , and cdeM with pair primer P83 ( 5'-GACCATATGGAAAATAAAAAATGTTATTCAGAAGATTGGTATGAAAG-3' ) , P84 ( 5'-GACGGATCCGATTTCCATTTCTTCTAGCTTCACATTCC-3' ) , ( S8 Fig ) . Three independent clones were phenotypically characterized . To evaluate whether the observed cdeC and cdeM phenotypes were attributed to inactivation of cdeC and cdeM , these strains were complemented with cdeC- and cdeM-FLAG fusions using plasmids pDP345 and pDP360 ( S1 Table ) . Briefly , C . difficile 630erm cdeC and cdeM mutants were complemented by conjugating with E . coli HB101 containing plasmids pDP345 , pDP360 , pPCR3 and pPCR4 respectively ( S1 Table ) . Trans conjugants were selected in BHIS agar plates containing erythromycin and thiamphenicol . Spore suspensions were prepared by plating a 1:100 dilution of an overnight culture onto a 70:30 medium ( 63 g Bacto peptone ( BD Difco ) , 3 . 5 g proteose peptone ( BD Difco ) , 0 . 7 g ammonium sulfate ( NH4 ) 2SO4 , 1 . 06 g Tris base , 11 . 1 g brain heart infusion extract ( BD Difco ) and 1 . 5 g yeast extract ( BD Difco ) for 1L ) and incubating it for 7 days at 37°C under anaerobic conditions . After incubation , plates were removed from the chamber and the surface was scraped up with ice-cold sterile water . Next , the spores were washed five times gently with ice-cold sterile water in micro centrifuge at 14 , 000 rpm for 5 min . Spores were loaded onto a 50% Nycodenz solution , centrifuged ( 14 , 000 rpm , 40 min ) . After centrifugation , the spores pellet was washed five times ( 14 , 000 rpm , 5 min ) with ice-cold sterile water to remove Nycodenz remnants . The spores were counted in Neubauer chamber and volume adjust at 5x109 spores per mL . To analyze the ultrastructure of spores of C . difficile 630ermB wild-type , cdeC and cdeM mutant spores ( ~2x108 ) were fixed with 3% glutaraldehyde and 0 . 1 M cacodylate buffer ( pH 7 . 2 ) overnight at 4°C , and stained for 30 min with 1% tannic acid . Samples were further processed and embedded in spurs resin as previously described [38] . Thin sections obtained with a microtome were placed on glow discharge carbon-coated grids and double-lead stained with 2% uranyl acetate and lead citrate . Grids were analyzed with a Phillips Tecnai 12 Bio Twin at the Electron Microscopy facility of the Pontificia Universidad Católica de Chile . C . difficile wild-type , cdeC and cdeM mutant spores were fixed with 3% paraformaldehyde ( pH 7 . 4 ) for 20 min in poly-L-lysine-coated glass cover slides . Fixed spores were rinsed three times with PBS and blocked with 1% bovine serum albumin ( BSA ) for 30 min and further incubated for 2 h at room temperature with primary antibodies 1:50 of rat antiserum raised against CdeC [13] or with 1:100 of rabbit antiserum raised against CdeM ( kindly provided by Dr . Adriano Henriques , Universidade Nova Lisboa ) . Next , covers containing fixed spores were incubated for 1 h at room temperature with 1:500 anti-rat IgG-Alexa488 conjugate ( Thermo Fisher ) or with 1:500 anti-rabbit IgG-Alexa488 conjugate ( Thermo Fisher ) in PBS-1% BSA and washed three times with PBS and once with distilled water . Dried samples ( 30 min at room temperature ) were mounted with Dako fluorescence mounting medium ( Dako North America ) and sealed with nail polish . Samples were analyzed with a BX53 Olympus fluorescence microscope . Samples ( 10 μl ) of coat and exosporium extracts of 5x107 spores of C . difficile 630erm wild-type and cdeC or cdeM mutant strains were treated twice at 100°C for 5 min in SDS-PAGE loading buffer and run on SDS-PAGE gels ( 12% acrylamide ) . Proteins were transferred to a nitrocellulose membrane ( Bio-Rad ) and blocked for overnight at 4°C with 2% bovine serum albumin ( BSA ) in TBS ( pH 7 . 4 ) . These western blots were probed with a 1:1 , 000 dilution of anti-FLAG for 1 h at room temperature and then with 1:10 , 000 dilution of anti-mouse-horseradish peroxidase ( HRP ) conjugate ( Promega ) for 1 h at room temperature in PBS 1X with 1% BSA and 0 . 05% Tween20 . In the western blot with goat antiserum raised against C . difficile 630erm spore [30] and anti-SpoIVA ( kindly provided by Dr . Shen Tufts University , U . S . A . ) , after the transference , the nitrocellulose membrane was blocked for 1 h at room temperature with 10% milk–Tris-buffered saline ( TBS ) ( pH 7 . 4 ) . These western blots were probed with a 1:500 goat antiserum raised against spores of C . difficile 630erm , 1:2500 rabbit antiserum raised against SpoIVA [32] for 1 h and then with a 1:10 , 000 dilution of anti-goat and anti-rabbit horseradish peroxidase ( HRP ) conjugate ( Promega ) for 1 h at room temperature in PBS–1X BSA–0 . 1% Tween 20 . In both cases , HRP activity was detected with a chemoluminescence detection system ( Fotodyne Imaging system ) by using PicoMax sensitive chemiluminescent detection system HRP substrate ( RockLand Immunochemicals ) . Each western blot also included 1 μl of PageRuler Plus prestained Protein Ladder ( Fermentas ) . Each western blot was repeated at least 3 independent times , and analyzed by densitometry to quantify the relative amounts of protein by ImageJ as previously described [13] . Antibodies against SpoIVA were a gift from Dr . Aimee Shen [36] . To quantify the effect of a cdeC and cdeM mutation on spore forming efficiency , aliquots of C . difficile 630erm wild-type and cdeC and cdeM spores ( 1x107 spores/mL ) were plated with or without a heat activation ( 65°C , 20 min ) onto BHIS-ST agar plates and incubated anaerobically for 36 h at 37°C . Spore viability was calculated using the following formula: [ ( c . f . u . mL-1 ) / ( spore particles mL-1 ) ] x 100 and expressed relative to wild-type strain . Ethanol resistance of C . difficile 630ermB wild-type , cdeC and cdeM spores was measured by resuspending 3x106 spores in 30 μl of 50% ethanol in PBS 1X . Spores were incubated with ethanol for 30 min at 37°C and shaking ( 200 rpm ) . Aliquots were plated onto BHIS-ST agar plates and incubated anaerobically for 36 h at 37°C . Heat resistance of C . difficile spores was determined as previously described [13] . Briefly , 3x106 spores of strains C . difficile 630erm wild-type , cdeC and cdeM were resuspended in 30 μl of PBS 1X pH 7 . 4 and heat treated at 75°C for 60 min . Aliquots at appropriate dilutions were plated onto BHIS-ST agar plates and incubated anaerobically for 36 h at 37°C . As a control of non-heat-treated spores , an aliquot was plated onto BHIS-ST agar plate prior to the experiment and colonies counted as described above . C . difficile spore-lysozyme resistance was measured by resuspending 3x106 spores in 30 μl of PBS 1X with 1 mg/mL of lysozyme and incubated for up to 5 h at 37°C with shaking ( 200 rpm ) . Germinated spores were analyzed by phase contrast microscopy . Spore viability was measured by plating aliquots onto BHIS-ST agar plates and incubated anaerobically at 37°C for 36 h and colonies counted . In some experiments , lysozyme-treated C . difficile 630erm wild-type , cdeC and cdeM spores were subsequently treated with 50% ethanol for 30 min at 37°C with shaking ( 200 rpm ) and aliquots plated onto BHIS-ST agar plates and colonies counted after 36 h of incubation under anaerobic conditions . To quantify spore-core DPA content , 200 μl of 5x109 spores/ml were boiled 60 min , cooled on ice for 2 min , centrifuged at 14 , 000 rpm x 5 min , and 190 μl of the supernatant was mixed with 10 μl 800 μM TbCl3 in a 96-well plate , and DPA release was monitored with an excitation of 270 nm and emission of 545 nm in a Synergy H1 Hybrid Multi-Mode Reader ( BioTek ) as described [49 , 50] . To measure the adherence of C . difficile 630erm wild-type cdeC and cdeM mutant spores to Raw 264 . 7 cells ( ATCC , U . S . A . ) , a 96-wells plate was seeded ( 5x105 cells per well ) and incubated at 37°C in 5% CO2 atmosphere . Confluent Raw 264 . 7 monolayers were infected with 40 μl of RPMI containing C . difficile 630erm wild-type , cdeC and cdeM spores at an MOI of 10 . After 30 min of incubation at 37°C , macrophages were washed three times with PBS 1X to rinse out unbound spores . Infected macrophages were lysed with 0 . 01% Triton X-100 , and adhered spores were counted by plating appropriate aliquots onto BHIS-ST agar plates and incubated for 36 h anaerobically at 37°C . Colonies were counted and expressed as c . f . u . mL-1 for colony counts , no additional colonies appeared upon further incubation periods . Total spores were counted by lysing the infected macrophages prior to rinsing off the unbound spores and plating appropriate dilutions onto BHIS-ST agar plates and colonies counted after 36 h of incubation at 37°C under anaerobic conditions . To evaluate C . difficile spore survival during infection of macrophages , after monolayer of Raw 264 . 7 cells were washed three times with PBS , macrophages were infected at an MOI of 10 as described above and unbound spores were rinsed off with three washes with PBS and macrophages were resuspended in 80 μl of RPMI with FBS 1% ( to avoids macrophage replication ) . Viability of C . difficile spores was determined at 0 . 5 , 24 , 48 and 72 after infection by lysing infected macrophages with 0 . 01% Triton X-100 , and serial dilutions plated onto BHIS-ST agar plates . The purified spores were heat activated for 30 min at 60°C . Next , were diluted in BHIS only or BHIS supplemented with 10 mM sodium taurocholate ( Sigma-Aldrich ) . Heat-activates spores in BHIS only was used as control . The OD600 was monitored immediately ( zero time ) and various times for 1h at 37°C . To determine citotoxicity of C . difficile strains an aliquot from a C . difficile was inoculated into BHIS broth and incubated for 24 h at 37°C under anaerobic conditions . Next , 1 mL of a 24-h BHIS culture was centrifuged and filtered and diluted 1:100 in Dulbecco Minimum Eagles Medium ( Lonza , USA ) supplemented with 10% filtered fetal bovine serum and 100 μL to each well of a 96-well plate containing Vero cells . The cells were incubated at 24 h under 5% CO2 . The circularity of the cells was recorded ( more than 50% of the cells ) . The cytotoxicity was measured with the following formula: Log10 ( ( percentage of rounded cells ) x 100 ) . 6-8 weeks old C57BL/6 ( male or female ) were obtained from breeding colony at the Facultad de Odontología de la Universidad de Chile ( Santiago , Chile ) that was originally established using animals purchased from Jackson Laboratories . All mice used in the experiments were housed individually cages and were acclimated for 1 week at the Animal Infection Facility of the Microbiota-Host Interactions and Clostridia Research Group at the Universidad Andrés Bello before the experiment . Water , bedding and cages were autoclaved , and mice has a 12-hour cycle of light and darkness . The C . difficile murine model of infection was used to perform competitive index ( CI ) experiments . For each competitive assay , wild-type C57BL/6 mice ( n=5 ) were challenged with 107 spores via gavage in 0 . 2 mL PBS . Equal amounts of spores ( 5x106 ) from the parental wild-type 630erm , cdeC and cdeM mutant were used . Fecal samples were collected and enumerated by plating on TCCFA agar , with and without erythromycin , and incubated for 48 h . Agar supplemented with erythromycin selected for the knockout containing the ermB cassette . The CI number was determined using the following ratio: [ ( 630 cdeC or cdeM/630 wild-type ) output] / [ ( 630 cdeC or cdeM/630 wild-type ) input] . Statistical testing was performed using the Mann Whitney test applied to Log10 values of the CI ratios . To induce C . difficile susceptibility in mice , prior the infection mice were administrated with a wide spectrum antibiotic , cefoperazone ( 0 , 5mg/mL ) ( Sigma ) in drinking water for 5 days , following 2 days of normal water as has been previously described [51 , 52] . Then animals were orogastrically infected with 3x107 C . difficile spores strain 630erm ( n = 6 ) ; cdeC ( n = 6 ) or cdeM ( n = 5 ) . All procedures and mouse handling were performed aseptically in a biosafety cabinet to contain spore-mediated transmission . To evaluate recurrence of CDI , from days 3 to 9 , all groups of mice were orogastrically administered 100 μl of PBS containing vancomycin ( 50 mg/kg; Sigma-Aldrich ) . During all the experiment , mice were daily monitored , and weight loss and diarrhea score and C . difficile spore shed . Sickness behaviors monitored daily , and fecal samples , and at the end of the assay , animals were sacrificed with a lethal dose of ketamine/xylazine and cecum content and colonic tissue were collected . The clinical condition of mice was monitored daily with a scoring system ( CDI ) . The presence of diarrhea was classified according to severity as follows: ( i ) normal stool ( score = 0 ) ; ( ii ) color change/consistency ( score = 1 ) ; ( iii ) presence of wet tail or mucosa ( score = 2 ) ; ( iv ) liquid stools ( score = 3 ) . A score higher than 0 was considered as diarrhea [52] . Collected fecal samples were stored at -20°C until spore quantification . Feces were hydrated with 500 μL sterile MilliQ water ON at 4°C and then added 500 μL of absolute ethanol ( Merck ) and at RT incubated for 60 min . Serially diluted of sample were plated on onto selective medium supplemented with taurocholate ( 0 . 1% w/v ) , Cefoxitin ( 16 μg/mL ) , L-cycloserine ( 250 μg/mL ) ( TCCFA plates ) . The plates were incubated anaerobically at 37°C for 48 h , colonies counted , and results expressed as the Log10 [CFU/g of feces] [52] . Colonic tissue was collected from mice , washed three times with PBS with a syringe . The spore load in the colon was determined in two sections: proximal colonic tissue , medium colonic tissue and distal colonic tissue and cecum tissue . First proximal colonic tussue was collected in three sections ( proximal , medium , distal ) and the first cm of each section ( from the cecum ) was obtained . For cecum tissue 1 cm from the base was obtained . After , tissue was weighted , and PBS: Absolute ethanol ( 1:1 ) was added ( 10 μl/mg of tissue ) , homogenized and incubated by 1 hour . The amounts of spores were quantified plating the tissue homogenization onto TCCFA plates . The plates were incubated anaerobically at 37°C for 48 h . Finally , the colony count was expressed as the Log10 [CFU/gram of colon] . Vero cell cytotoxicity was performed as described previously [51] . Briefly , 96-well flat bottom microtiter plates were seeded with Vero cells at a density of 105 cells/well . Mice cecum contents were suspended in PBS at a ratio of 1:10 ( 10 μL of PBS per mg of cecum content ) , vortexed and centrifuged ( 14 , 000 rpm , 5 min ) . Filter-sterilized supernatant was serially diluted in DMEM supplemented with 10% FBS and 1% penicillium streptomycin; 100 μL of each dilution was added to wells containing Vero cells . Plates were screened for cell rounding 16 h after incubation at 37°C . The cytotoxic titer was defined as the reciprocal of the highest dilution that produced rounding in at least 80% of Vero cells per gram of luminal samples under X200 magnification . Serum from infected animals were tested against or 630erm vegetative cells by ELISA . 1 . 6x107 spores or 3 . 0 x 106 vegetative cells prefixed in PFA 4% per well were incubated in 96-wells plate by 16 hrs at 4°C . Plates were washed with PBS- Tween20 0 . 05% , 3 times and blocked with 2% BSA by 1 hr at 37°C . After 3 washes , wells were incubated with serum dilutions 1:200 , and incubated for 2 hr at 37°C . After 5 washes , secondary anti-mouse HRP antibody was added at 1:10 , 000 and incubated at 30°C for 1 hour and finally washed 5 times . Colorimetric reaction was initiated upon addition of 50 μL of reaction buffer ( 0 . 05 M citric acid , 0 . 1 M disodium hydrogen phosphate ) containing 2mg/mL of o-phenlyendiamine ( Sigma-Aldrich , U . S . A . ) and 0 . 015% of H2O2 ( Merck , Germany ) . Reaction was stopped after 20 min with 25 μL of 4 . 5 N of H2SO4 and absorbance was measured at 492 nm . Background reactivity was performed using IgY from eggs obtained prior immunization . Before to surgery mice were deeply anesthetized in a general way with Small Animal Anesthesia Machine for which the mice were induced in a chamber with 5% isoflurane ( RWD ) , then the mice were maintained with 1 . 5% isoflurane during the surgery administered by air . Briefly , after a midline laparotomy , 1 . 5 cm ileal and proximal colon were ligated and injected with 3 . 3x108 spore/cm in 0 . 1 mL of PBS ( pH 7 . 2 ) for intestinal loops ( n = 6 for wild-type and cdeC; n=5 for cdeM ) . The abdomen was closed with superglue , and the animals were allowed to regain consciousness . The mouse was kept for 5 h at which time the animal was euthanized , and the ligated loops were removed and washed gently in PBS and fixed in 4% paraformaldehyde 30% sucrose during 16h , washed and subjected to indirect immunofluorescence . Tissue were made permeable by incubation with 0 . 2% Triton X-100 in PBS 1X and blocked with 3% BSA in PBS for 3h . Tissue were made permeable by incubation with 0 . 2% Triton X-100 in PBS 1X and blocked with 3% BSA in PBS for 3h . The same buffer was used for subsequent incubation with antibodies . Intestine fragments were incubated with a primary polyclonal IgY anti-C . difficile spore and fluorescently labelled phalloidin ( Alexa Fluor 568 ) for 12-16h at 4°C . Following PBS washed , samples were reacted with goat anti-chicken IgY secondary antibodies ( Alexa Fluor 488 ) and Hoechst . For mounting was applied a drop of DAKO fluorescent mounting medium onto the tissue segment and mount cover glass over it and sandwich the tissue section . The ends of the cover glass should be fixed to the glass slide with a vinyl tape to hold the tissue sections in place . To acquiring images Leica TCS LSI microscope was used , with 5X ( optical zoom 20X ) , numerical aperture 0 . 5 . Confocal Imaging 405 nm , 488 nm and 532 nm excitation wavelengths were used for nuclei staining ( Hoechst ) , Alexa Fluor 488-llabeled bacteria and Alexa Fluor 568-labelled phalloidin , and signals were detected with an ultra-high dynamic PMT spectral detector ( 430-750nm ) . Emitted fluorescence was split with four dichroic mirrors ( QD405nm , 488nm 561nm and 635nm ) . Images ( 1024x1024 ) acquired with a 0 . 7-μm Z step were smoothed by median filtering at kernel size 3x3 pixels . Z projection of intestinal epithelium were performed using ImageJ software ( NIH ) . Villi and crypt were visualized by Hoechst and phalloidin signals . For quantification of tissue associated bacterial signals stacks Z step were smoothed by median filtering at kernel size 3x3 pixels . Nuin PBSmber of positive spots/1 , 000 μm2 from ileal and proximal colon and area occupied by individual spots were analyzed . Data were not normally distributed and were analyzed by non-parametric tests . Student’s t-test was used for pairwise comparison in most experiment . Where stated , non-parametric test were used .	We discovered a mechanism of assembly of the outer most layer of Clostridium difficile spores , the exosporium . While CdeC is conserved in several Peptostreptococcaeace family members , CdeM is unique to C . difficile . We show that two proteins that are rich in cysteine amino acid residues , CdeC and CdeM , are essential for the recruitment of additional spore coat and exosporium proteins . The absence of CdeC , had profound implications in the correct spore coat assembly which were related to decreased spore resistant properties that are relevant for in vivo infection such as lysozyme resistance , macrophage infection . Notably , the absence of either cysteine rich proteins leads to a decrease in spore adherence of C . difficile spores to healthy colonic mucosa; but only the absence of CdeC affected in vivo competitive fitness in a mouse model , recurrence of the disease in a mouse model of recurrent infection . Considering the importance of the outer layers of C . difficile spores in spore-host interactions , our findings have broad implications on the biology of C . difficile spores and to C . difficile pathogenesis .	[ "Abstract", "Introduction", "Results", "Discussion", "Material", "and", "methods" ]	[ "bacteriology", "peptostreptococcus", "medicine", "and", "health", "sciences", "gut", "bacteria", "chemical", "compounds", "animal", "models", "of", "disease", "microbiology", "organic", "compounds", "animal", "models", "model", "organisms", "experimental", "organism", "systems", "amino", "acids", "sequence", "motif", "analysis", "bacteria", "digestive", "system", "research", "and", "analysis", "methods", "clostridium", "difficile", "cysteine", "animal", "models", "of", "infection", "sequence", "analysis", "microbial", "physiology", "animal", "studies", "bioinformatics", "proteins", "bacterial", "spores", "mouse", "models", "chemistry", "gastrointestinal", "tract", "sulfur", "containing", "amino", "acids", "biochemistry", "bacterial", "physiology", "anatomy", "organic", "chemistry", "database", "and", "informatics", "methods", "biology", "and", "life", "sciences", "physical", "sciences", "colon", "organisms" ]	2018	Clostridium difficile exosporium cysteine-rich proteins are essential for the morphogenesis of the exosporium layer, spore resistance, and affect C. difficile pathogenesis
Living tissues undergo deformation during morphogenesis . In this process , cells generate mechanical forces that drive the coordinated cell motion and shape changes . Recent advances in experimental and theoretical techniques have enabled in situ measurement of the mechanical forces , but the characterization of mechanical properties that determine how these forces quantitatively affect tissue deformation remains challenging , and this represents a major obstacle for the complete understanding of morphogenesis . Here , we proposed a non-invasive reverse-engineering approach for the estimation of the mechanical properties , by combining tissue mechanics modeling and statistical machine learning . Our strategy is to model the tissue as a continuum mechanical system and to use passive observations of spontaneous tissue deformation and force fields to statistically estimate the model parameters . This method was applied to the analysis of the collective migration of Madin-Darby canine kidney cells , and the tissue flow and force were simultaneously observed by the phase contrast imaging and traction force microscopy . We found that our monolayer elastic model , whose elastic moduli were reverse-engineered , enabled a long-term forecast of the traction force fields when given the tissue flow fields , indicating that the elasticity contributes to the evolution of the tissue stress . Furthermore , we investigated the tissues in which myosin was inhibited by blebbistatin treatment , and observed a several-fold reduction in the elastic moduli . The obtained results validate our framework , which paves the way to the estimation of mechanical properties of living tissues during morphogenesis . The body of multicellular organisms must be properly shaped in order to exert its functions , and this proper formation is based on the orchestration of cellular behaviors , such as cell division , differentiation , migration , and other . One of the key processes in morphogenesis is the coordinated change in cell shapes and positions . The coordination depends on cell-generated mechanical forces that introduce stress , which induces multicellular deformation and flow [1] . Therefore , the research on how the molecular components responsible for force generation and propagation , such as motor proteins and cell-cell adhesion molecules , are regulated in space and time during the morphogenesis has attracted a lot of attention recently [2] . In parallel , remarkable progress has been made in the development of the technologies allowing the measurements of the generated forces and stress in the living tissues [3] , which represents a crucial step towards linking the underlying molecular activities with the morphogenesis . Epithelial tissues represent important model systems for the understanding of the force dynamics during morphogenesis , because their two-dimensional sheet structure facilitates the observation of the processes that occur in these tissues and analysis . In particular , many valuable insights have been obtained using the cultured cell monolayer , i . e . , one-cell-thick sheet of tightly-connected epithelial cells [4–6] . The cells belonging to a monolayer collectively migrate in order to fill a cell-free surface , which replicates in vivo tissue remodeling , such as wound repair , which occurs during regeneration , and epiboly , during embryonic development . When migrating , the cells exert forces on the underlying substrate to propel themselves forward , and in the unicellular motion , this force , known as the cell traction force , can be visualized by the displacement of fluorescent beads embedded into the substrate [7] . The simple flat-sheet structure of the monolayer allows us to apply the same technique to observe a spatio-temporal profile of the cell traction force in a wide field of view [8] , and to determine where and how the force and stress are generated [9 , 10] . In order to achieve the quantitative understandings of the resultant tissue morphogenesis , however , we need to elucidate the other mechanical factors as well , i . e . , the mechanical properties that describe the relation between the deformation and forces . Although several pioneering works exist [11–14] , our access to the mechanical properties is still limited . The characterization of these properties often requires exogenous manipulation of the tissue to induce deformation , but the procedure itself perturbs cell physiology and interferes with the tissue morphogenesis . Here , force measurement in a non-invasive manner gives a way to bypass this issue , and we can infer mechanical properties by associating spontaneous tissue deformation with the observed force dynamics . In this study , we propose a reverse-engineering method to identify the mechanical properties , which is based on the combination of tissue mechanics modeling and statistical machine learning . Our strategy is to represent a cell monolayer as a continuum-mechanical system [15] , and to use the passive and simultaneous observations of the deformation and traction force in order to compute the maximum likelihood estimate of the mechanical parameters . We formulated the inference as an inverse of the forward processes in which the mechanical properties and reaction force to the traction cause the tissue deformation . Our inference algorithm is based on the sequential updates of estimates; using the current model state and parameters , the mechanical model predicts the traction force field , and then the error feedback based on the observation is used to update the model state and parameters . Here , we applied our method to a cultured monolayer system to infer the elastic moduli from the collected tissue deformation and traction force data . To characterize the tissue deformation , we used velocity field of tissue motion , hereafter called tissue flow field . MDCK cells ( strain II ) were maintained in minimal essential medium ( MEM; Invitrogen ) supplemented with 10% fetal bovine serum ( FBS; Equitech-Bio ) , GlutaMAX ( Invitrogen ) , and 1 mM sodium pyruvate , in a 5% CO2 humidified incubator at 37 C° . According to a previously published protocol [9] , 48 h before the image acquisition , 3 μl drop of dense cell suspension ( 8 × 106 cells/ml ) was added to each dish containing the gel and 3 ml medium . Afterward , 3 h before the image acquisition , the medium was replaced by 3 ml CO2-independent medium supplemented with 10% FBS and GlutaMAX . For the myosin II inhibition , we added blebbistatin ( Sigma Aldrich ) at a final concentration of 25 μM following the replacement of the medium . Polyacrylamide gel substrates were prepared according to the previously published protocols [8 , 9] . Briefly , the gel solution was prepared with 3% acrylamide , 0 . 25% bisacrylamide , 0 . 8% ammonium persulfate , 0 . 08% TEMED ( Bio-Rad products ) , and 0 . 01% red fluorescent carboxylate-modified beads ( 0 . 5μm diameter , Invitrogen ) . 20 μl of this mixture was added to each dish and the samples were covered with glass cover slips with 18 mm diameter ( Matsunami ) . After the polymerization , the surface was coated with type I collagen ( Purecol , Advanced BioMatrix ) using 4 μM sulphosuccinimidyl-6- ( 4-azido-2-nitrophenylamino ) hexanoate ( Sulfo-SANPAH; Pierce ) . Young’s modulus of the gel was characterized by the conventional method using the Hertz equation [16] , obtaining E = 2500±600 Pa . The Fourier-transform traction microscopy [8] was used to estimate traction force fields from bead displacement fields . Confocal imaging was conducted at 48 h after the seeding of the cells . We used FV10i-LIV ( Olympus ) to simultaneously acquire phase contrast images of the cells and fluorescent images of the beads . The trial period lasted for 6-10 h and the sampling rate was one frame per 5 min . After each trial , we removed the cells by the trypsinization and imaged the strain-free pattern of the fluorescent beads . To increase the field of view , we stitched tiled images by the Grid/Collection stitching plugin in Fiji [17] . Following this , the images at different time points were aligned to match the bead configurations in a cell-free region [18] . To obtain velocity fields in the phase contrast image and bead displacement fields , we adopted an advanced optical flow technique , which tracks changes between two images by matching the patterns of intensity and its gradient [19] . S1 Movie shows a representative result of the image analysis . For the tissue flow , we used images from subsequent time points , while for the bead displacement , we compared the stress-free image with each fluorescent image . The image resolution was 0 . 61μm/pixel and the grid spacing of the vector fields was 14 . 7 μm . Finally , the flow and force fields were down-sampled in space and time into Δx = 29 . 4μm and Δt = 10 min . We adopted a continuum modeling of the monolayer mechanics , in which the deformation of the tissue , or strain , determines the stress [20 , 21] . The cell monolayer was represented as a two-dimensional sheet , and therefore , we represented the stress as a symmetric matrix σ ( x , y , t ) ≡ ( σ x x σ x y σ x y σ y y ) = π I + σ ˜ . ( 1 ) In the second line , we applied the deviatoric decomposition where the stress tensor is given as the summation of isotropic ( the first term ) and distortional ( the second term ) components . The strain tensor was also represented by a two-by-two matrix as ϵ i j ≡ 1 2 ( ∂ j u i + ∂ i u j ) , ( i , j = x , y ) , ( 2 ) where ( ux , uy ) represents the displacement vector of the tissue from the stress-free state . In a linearly elastic material , the relationship between the stress and strain tensors becomes simply linear , meaning that the stress accumulates in response to the strain . However , the stress-free state of the living tissue can vary in time due to cell growth and death . Therefore , we adopted an alternative formulation using the strain rate tensor e ( x , y , t ) ≡ ϵ ˙ → e i j = 1 2 ( ∂ j v i + ∂ i v j ) → e = 1 2 ( ∇ · v ) I + e ˜ , ( 3 ) where v = ( vx , vy ) is the flow velocity vector in the tissue . In the last line , we applied the deviatoric decomposition . Although previous works suggested that anisotropic cell division can contribute the tissue mechanics [22 , 23] , we modeled the cell growth simply as isotropic and homogeneous expansion with the rate Dg , which is partially supported by a previous report that cell division in the monolayer shows no particular orientation [24] . Since the observed total expansion of the tissue is the summation of the growth and the deformation-originated expansion , i . e . , Dtotal = Dg + Dmaterial , the subtraction Dtotal − Dg should appear in the stress-strain relation [15] . Taken together , our elastic model was written as π ˙ ( x , y , t ) = K ( ∇ · v ( x , y , t ) - D g ) + ξ ( x , y , t ) σ ˜ ˙ x x ( x , y , t ) = 2 G e ˜ x x ( x , y , t ) + ξ x x ( x , y , t ) σ ˜ ˙ x y ( x , y , t ) = 2 G e ˜ x y ( x , y , t ) + ξ x y ( x , y , t ) , ( 4 ) where K and G are the in-plane bulk and shear elastic moduli , respectively ( in S1 Text , we derived the relation of the in-plane moduli to the conventional three dimensional moduli ) . ξs are the stochastic terms representing random variables with Gaussian distribution that is not space or time-dependent . Dg is associated with the cell division interval tdiv as Dg = ln 2/tdiv , and we adopted tdiv = 1 division per day . We found that essentially the same results are obtained by increasing or decreasing the rate of growth rate two times . On the other hand , the tissue stress tensor and traction force vector were related through the force balance equation [25] - T x ( x , y , t ) = ∂ π ∂ x + ∂ σ ˜ x x ∂ x + ∂ σ ˜ x y ∂ y + η x - T y ( x , y , t ) = ∂ π ∂ y - ∂ σ ˜ x x ∂ y + ∂ σ ˜ x y ∂ x + η y , ( 5 ) where ηs are noises in the force quantification assumed to be normally-distributed , and we call them the observation noises . Here , we briefly describe the inference algorithm ( the details of derivation is given in S1 Text ) . Let Y and Λ represent the collected spatio-temporal fields of traction force and tissue flow , respectively , and X represent the stress tensor field that was discretized in space and time according to Y and Λ . Additionally , let θ represent the model parameters . Then , our aim is to find such θ ^ that maximizes the log-likelihood: ln L ≡ ln p ( Y \| Λ , θ ) = ln ∫ p ( Y \| X , θ ) p ( X \| Λ , θ ) d X . ( 6 ) Note that p ( X\|Λ , θ ) and p ( Y\|X , θ ) are corresponding to the stress evolution and force balance equations , i . e . , Eqs 4 and 5 , respectively . Since the integration w . r . t . X is analytically intractable , we adopted the expectation-maximization ( EM ) algorithm , which maximizes the lower-bound of the log-likelihood by executing the following E and M steps alternately [26] . ( E-step ) Estimate the stress fields by computing p ( X\|Y , Λ , θ* ) with the Rauch-Tung-Striebel smoother [27] , where θ* is a tentative estimate of the parameters . ( M-step ) Compute the expected complete-data log-likelihood: Q ( θ ) = E p ( X \| Y , Λ , θ * ) [ ln p ( X , Y \| Λ , θ ) ] , ( 7 ) and update the parameters through maximizing Q ( θ ) . Repeating the E and M steps , which offers monotonic increase and convergence of the likelihood . After the convergence , we obtain the maximum likelihood estimate of the parameters θ ^ . In order to collect the data on tissue deformation and force , we adopted a model system , Madin-Darby canine kidney ( MDCK ) epithelial cell monolayer , and analyzed it using the colony expansion assay [9] . We performed phase contrast imaging to measure the flow of cells in the monolayer and , simultaneously , traction force microscopy , in order to visualize the generated force using fluorescent beads embedded into the soft substrate . Our inference algorithm used a mechanical model of the cell monolayer , i . e . , a spatio-temporal model of mechanical stress within the tissue . Our mechanical model of the cell monolayer , represented by Eq 4 ( see Materials and methods section ) , included two biophysical factors that are essential for the colony expansion: tissue elasticity and cell growth . Elasticity is a basic property of a material , which resists the influence of an external force and shows a recoverable deformation . According to the previous studies [9 , 14] , we assumed the linear elasticity where the deformation is proportional to the force . Additionally , cellular growth supplies new cells into the tissue , and thereby promotes tissue expansion [28] . Our mechanical model also included stochasticity , which represents other mechanical processes such as viscosity , plasticity , active contractile force , and others . This model , represented by Eq 4 , had two mechanical parameters describing the elastic properties of the monolayer , the in-plane bulk modulus K and shear modulus G . The values of the bulk and shear moduli represent the resistance against area-changing and area-preserving deformation in the monolayer , respectively . Additionally , the elastic model contains the variance parameters for strength of the stochastic effects in the stress dynamics . Our inference algorithm , using the movie data showing tissue flow and traction force , estimated the values of these parameters ( Fig 1A and 1B and “Materials and methods” section ) . We found that the flow speed of the tissue at the periphery was approximately 10-30 μm/h ( Fig 2A ) , the strength of the traction force was distributed around 10-100 Pa ( Fig 2B ) , and both flow speed and force strength decreased monotonically along the distance from the edge ( Fig 2A and 2B ) . These results are consistent with the previous observations [4 , 8 , 29] . We computed maximum likelihood estimates of the parameters in our elastic model from the collected data . For this estimation , the model state , i . e . , the tissue stress field , was corrected by the current traction force data at each time point in the movie sequence; following this , the model was numerically simulated , using the tissue flow data , in order to predict the traction force field at the following time point . As a result , even though the model inference was based on the one-step prediction ( Δt = 10 min ) , we found that the estimated model can provide a long-term forecast ( >1 h ) without corrected by the traction force data . To quantitatively demonstrate this result , we divided each movie data on tissue flow and traction force , into two parts in time: The earlier , training data and the following test data . Using the training data , the inference algorithm was used to estimate the model parameters and the stress field in the monolayer , and then this model was examined in terms of the forecast accuracy for future force fields by using the test data . As a quantitative measure , we employed the correlation between the forecasted and observed force vector fields: R = ⟨ T forecast · T data ⟩ ⟨ \| T \| forecast · \| T \| data ⟩ , ( 8 ) where 〈⋅〉 represents an average over all spatial grid points . The correlation plotted against time is represented in Fig 3A . As shown , the forecast provided by the elastic model was highly correlated even 3 h after the initiation of the test part . For comparison , we adopted a null-hypothetical , zero-elasticity model , where K = G = 0 ( Fig 3A ) . In Fig 3B , the correlation in both models at the last time-point in the test part , the long-term forecast accuracy , is shown . These results demonstrate that our data-driven elastic model showed better forecast which is clearer especially in longer time forecast . We also computed the difference of correlation between the models in a sample-wise manner ( Fig 3C ) , and confirmed statistically-significant superiority of the elastic model compared with the zero-elasticity model ( p < 10−4 , Wilcoxon signed-rank test ) . Representative forecasted and observed traction forces at the last time point are shown in Fig 3D . Therefore , despite of its simplicity , our data-driven elastic model captured the stress evolution in the tissue expansion by the estimated bulk ( K ) and shear ( G ) moduli . Next , we examined if the elastic moduli are different in the tissues treated with blebbistatin , a myosin inhibitor . Previous studies showed that the inhibition of the molecular motors considerably reduces the traction force strength , which is expected . However , this inhibition does not slow down the tissue expansion rate [6] , indicating the alterations in tissue mechanical properties . By comparing the elastic moduli estimated from a different dataset from blebbistatin-treated tissues with those estimated under the standard conditions ( Fig 4A , 4B and 4C ) , we obtained the results consistent with those obtained in previous experiments . We observed a significant decrease in the elastic moduli associated with the treatment ( Fig 4D and 4E ) . In the standard experimental setting , both moduli were within the order of the magnitude of ∼ 103Pa · μm , while myosin inhibition induced several-fold reduction in the elastic modulus values , i . e . , softening ( p < 0 . 01 , U-test ) . Note that the estimated moduli are not guaranteed to have positive values because unmodeled monolayer mechanics , such as viscosity and anisotropic tissue growth , might affect the stress dynamics . In fact , the estimated moduli from myosin-inhibited tissues have frequently shown negative values , suggesting that the elasticity effect was no longer dominant over unmodeled effects due to the softening . Finally , we assessed the spatial distribution of the elastic moduli . When considering difference in the flow speed and force strength dependent on the distance from the tissue edge ( Fig 2 ) , we expected the mechanical properties that would correlate with this distance . However , as shown in Fig 5 , we found that the estimated moduli are homogenous along with the distance . The framework we presented in this study would benefit from the advances in force measurement . For example , although the traction force microscopy is applicable only to in vitro tissues , in vivo measurement techniques are being actively developed [38 , 39] . We can apply our reverse-engineering method to the in vivo measuring by modifying the model of the observation process , Eq 5 in our case . Additionally , another interesting direction would be to use a more elaborate model of tissue mechanics , in particular , by directly including cellular processes such as cell division [40 , 41] . We hope that , with the advancements in the technology of force/stress measurement , our method may assist further understanding of the mechanics underling tissue development and maintenance .	In order to shape the body of a multicellular organism , cells generate mechanical forces and undergo deformation . Although these forces are being increasingly determined , quantitative characterization of the relation between the deformation and forces at the tissue level remains challenging . To estimate these properties , we developed a reverse-engineering method by combining tissue mechanics modeling and statistical machine learning , and then tested this method on a common model system , the expansion of cultured cell monolayer . This statistically sound framework uses the passive observations of spontaneous deformation and force dynamics in tissues , and enables us to elucidate unperturbed mechanical processes underlying morphogenesis .	[ "Abstract", "Introduction", "Materials", "and", "methods", "Results", "Discussion" ]	[ "tissue", "mechanics", "fluorescence", "imaging", "mechanical", "properties", "classical", "mechanics", "cell", "cycle", "and", "cell", "division", "cell", "processes", "biomechanics", "developmental", "biology", "molecular", "motors", "actin", "motors", "materials", "science", "damage", "mechanics", "morphogenesis", "motor", "proteins", "research", "and", "analysis", "methods", "contractile", "proteins", "imaging", "techniques", "proteins", "deformation", "biophysics", "physics", "biochemistry", "cytoskeletal", "proteins", "cell", "biology", "myosins", "biology", "and", "life", "sciences", "physical", "sciences", "material", "properties" ]	2018	Inverse tissue mechanics of cell monolayer expansion
Pulmonary Francisella tularensis and Burkholderia pseudomallei infections are highly lethal in untreated patients , and current antibiotic regimens are not always effective . Activating the innate immune system provides an alternative means of treating infection and can also complement antibiotic therapies . Several natural agonists were screened for their ability to enhance host resistance to infection , and polysaccharides derived from the Acai berry ( Acai PS ) were found to have potent abilities as an immunotherapeutic to treat F . tularensis and B . pseudomallei infections . In vitro , Acai PS impaired replication of Francisella in primary human macrophages co-cultured with autologous NK cells via augmentation of NK cell IFN-γ . Furthermore , Acai PS administered nasally before or after infection protected mice against type A F . tularensis aerosol challenge with survival rates up to 80% , and protection was still observed , albeit reduced , when mice were treated two days post-infection . Nasal Acai PS administration augmented intracellular expression of IFN-γ by NK cells in the lungs of F . tularensis-infected mice , and neutralization of IFN-γ ablated the protective effect of Acai PS . Likewise , nasal Acai PS treatment conferred protection against pulmonary infection with B . pseudomallei strain 1026b . Acai PS dramatically reduced the replication of B . pseudomallei in the lung and blocked bacterial dissemination to the spleen and liver . Nasal administration of Acai PS enhanced IFN-γ responses by NK and γδ T cells in the lungs , while neutralization of IFN-γ totally abrogated the protective effect of Acai PS against pulmonary B . pseudomallei infection . Collectively , these results demonstrate Acai PS is a potent innate immune agonist that can resolve F . tularensis and B . pseudomallei infections , suggesting this innate immune agonist has broad-spectrum activity against virulent intracellular pathogens . Francisella tularensis is a highly infectious , Gram-negative facultative intracellular bacterium that causes the zoonotic infection tularemia . F . tularensis infections can occur via insect bites , cutaneous contact with infected animal carcasses , ingestion of contaminated food and water , or inhalation of viable organisms [1] . The type and severity of tularemia depends on the strain , dose , and route of infection [2] . F . tularensis subspecies tularensis ( type A ) and holarctica ( type B ) cause the majority of human cases , with subspecies tularensis being more virulent [2] . Cutaneous tularemia is the most common form of human disease , but is rarely fatal [3] . Inhalation of F . tularensis results in respiratory or pneumonic tularemia and is most common in people in endemic areas who perform tasks that predispose them to infectious aerosols [2] . Untreated respiratory forms of disease have mortality rates of >30% [4] , while antibiotic treatment can decrease this number to approximately 2% [5] . Pulmonary tularemia can present from a mild pneumonia to an acute infection with high fever , malaise , chills , cough , delirium , and pulse-temperature dissociation [2] . The high infectivity ( 10–50 microorganisms ) [3] and mortality of F . tularensis infections have led to the weaponization of the organism , including the introduction of antibiotic resistance , by several nations [5] . Due to these concerns , F . tularensis has been determined to be a Category A Bioterrorism agent by CDC . No vaccines are currently licensed to prevent tularemia . Although a live vaccine strain ( LVS ) derived from F . tularensis subspecies holarctica was created over 50 years ago , questions remain regarding its efficacy and possible reversion to virulence , and it is not licensed for human use [2] . LVS is attenuated in humans , but remains virulent for mice , although it is not as virulent as wild-type A and B strains . As LVS causes a disease in mice that mimics tularemia in humans , it has been studied extensively as a model intracellular pathogen [6] and is utilized here as model to assay the efficacy of agonists to enhance resistance to Francisella in vitro , while our in vivo studies employ the fully virulent SchuS4 strain of type A F . tularensis . Burkholderia pseudomallei and B . mallei are gram-negative facultative intracellular bacterial pathogens . B . pseudomallei is the etiologic agent of melioidosis and is endemic in parts of southeast Asia and northern Australia [7] . The clinical manifestations of melioidosis are protean and may vary from acute sepsis to chronic focal pathology and latent infection , which can reactivate decades later from an , as yet , unknown tissue reservoir [8] . Melioidosis can also mimic other infections such as glanders , typhoid fever , bacterial sepsis , and TB , depending on whether the disease is acute or chronic [8]–[10] . Community-acquired infection with melioidosis is most likely due to exposure to bacteria in soil or water through cuts or skin abrasions or via inhalation or ingestion [8] . No licensed prophylactic or therapeutic vaccine exists for Burkholderia infections , and B . pseudomallei is intrinsically resistant to a wide range of antimicrobial agents . In addition , prolonged antibiotic therapy ( up to 6 months ) is required to treat Burkholderia infections , and 10–15% of patients may relapse when antibiotic therapy is withdrawn [8] , [11] . Due to the lack of efficacious vaccines and concerns about F . tularensis acquiring resistance to antibiotics via natural or illicit means and the intrinsic antimicrobial resistance of B . pseudomallei , we hypothesized that alternative immune or natural therapeutic-based intervention strategies could prove beneficial to augment current treatment regimens . Activation of the innate immune system can enhance resistance to a variety of bacterial and viral infections [6] , 11–14 . Immunotherapeutics may be particularly beneficial against diseases caused by intracellular pathogens since the antibiotics often recommended for treatment of these diseases , such as gentamicin , poorly penetrate host cells and therefore fail to reach the etiological agent of disease [14] . In situations where the etiological agent of disease is unknown , stimulation of innate immunity may also be useful since these immune responses are often capable of providing protection against a broad range of pathogens [6] , [14] . To achieve this goal several natural agonists , including apple polyphenols ( APP ) , amphotericin B ( AmpB ) , securinine , Yamoa PS , and Acai PS were tested for their ability to enhance immunity to F . tularensis since each of these agonists has been previously shown to exhibit proinflammatory properties [12] , [15]–[19] . Herein , we showed polysaccharides isolated from the Acai berry ( Acai PS ) enhanced clearance of F . tularensis from human macrophages upon co-culture with autologous natural killer ( NK ) cells . Mucosal administration of Acai PS also conferred both prophylactic and therapeutic protection against pulmonary F . tularensis and B . pseudomallei infections . The immunological basis for Acai PS-mediated protection both in vitro and in vivo is elucidated in this study . An initial screen of natural agonists for their ability to enhance macrophage resistance to F . tularensis infection was conducted . RAW264 . 7 cells , a murine macrophage-like cell line [20] , were treated overnight prior to F . tularensis LVS infection . LPS ( E . coli 0∶55 , B5 ) was also included in our screen as a positive control for macrophage activation . Both intracellular bacterial burden and NO2 accumulation were measured ( Figure 1 ) . While amphotericin B ( Amp B ) , Apple Polyphenol ( APP; [21] ) , LPS , Acai PS , and Yamoa polysaccharides ( Yamoa PS [15] ) all enhanced nitric oxide ( NO ) production by RAW264 . 7 cells ( Figure 1B ) , only LPS , Acai PS , and Yamoa PS significantly enhanced macrophage resistance to F . tularensis LVS ( Figure 1A ) at the indicated doses . Preliminary in vivo experiments indicated that only Acai PS was able to provide protection against pulmonary LVS challenge ( data not shown ) . Yamoa PS previously was shown to induce strong reactivity to the Limulus Amebocyte Lysate ( LAL ) assay [15] and therefore was eliminated from further study . However , Acai PS has low amounts of endotoxin reactivity as measured by LAL assay , and its immunomodulatory effects are resistant to polymyxin B treatment [19]; therefore , it was selected for further evaluation . To assess the immunomodulatory effects upon surface activation molecule expression by Acai PS treatment , RAW264 . 7 cells ( originally derived from BALB/c mice ) were treated with varying doses of Acai PS overnight prior to mock- or LVS- ( Multiplicity of Infection [MOI]∼300 ) infection . RAW264 . 7 macrophages were then cultured for an additional 20 h prior to assessment of changes in surface activation molecule expression by flow cytometry , and cytokine and NO production were also measured in cell culture supernatants . Acai PS alone markedly stimulated CD40 , CD80 , and CD86 ( Table S1 ) . Subsequent LVS infection , Acai PS enhanced surface expression of CD11b , CD40 , CD80 , CD86 , TLR2 , and MHC class II in a dose-dependent fashion , while TLR4 expression was downregulated in both mock- and LVS- infected macrophages ( Table S1 ) . Acai PS also enhanced generation of NO , TNF-α , and IL-6 in a dose-dependent manner by RAW264 . 7 cells ( Table S2 ) and also induced trace amounts of IL-1β and IL-12p40 ( <300 pg/ml , data not shown ) . To investigate the mechanism by which Acai PS enhances RAW264 . 7 cell resistance to F . tularensis infection , RAW264 . 7 cells and BMDMs derived from BALB/c mice were treated with varying amounts of Acai PS before infection with F . tularensis SchuS4 . Pretreatment of RAW264 . 7 cells with as little as 10 µg/ml of Acai PS reduced SchuS4 replication , while the greatest protection was obtained using a 100 µg/ml dose ( Table 1 ) . Although the addition of 400 µM NG-Methyl-L-arginine ( L-NMA ) , an inhibitor of NO production [22] , did not totally abrogate NO production by RAW264 . 7 cells prestimulated with 100 µg/ml Acai PS ( Table 2 ) , L-NMA treatment did significantly diminish Acai PS-mediated resistance to F . tularensis SchuS4 , while having no effect on unstimulated cells ( Table 1 ) ; similar results were obtained using F . tularensis LVS ( Figure S1 ) . While Acai PS reduced intracellular replication of F . tularensis SchuS4 in RAW264 . 7 cells in an NO-dependent manner , Acai PS did not induce NO or enhance the clearance of F . tularensis SchuS4 from murine BMDMs ( Table 1 ) , which highlights the limitations of using cell lines as surrogates for primary cells . However , pretreatment of BMDMs with Acai PS did enhance phagocytosis of F . tularensis SchuS4 ( Table 1 ) . In addition , while infection of macrophages with strains of Francisella that do not cause disease in humans , such as F . novicida , results in rapid activation of the inflammasome and cell death [23] , we did not find type A F . tularensis infection , or Acai PS to induce robust cytotoxicity of murine BMDMs or primary human macrophages at 20 h post-infection under the conditions tested ( Table S3 ) . This is in concordance with other studies that show type A F . tularensis does not vigorously activate the inflammasome in human dendritic cells [24] . While Acai PS was unable to restrict the replication of F . tularensis in primary BMDMs , Acai PS previously was found to activate a variety of human leukocytes [19] . Therefore , we adopted a co-culture system in which primary human macrophages were infected with F . tularensis and co-cultured with autologous NK cells . Briefly , CD14+ cells were sorted and differentiated prior to Acai PS overnight treatment . Macrophages were infected with F . tularensis LVS and then cultured with or without purified autologous NK cells , some of which were also prestimulated with varying amounts of Acai PS overnight . CFU determinations were performed 20 h after infection , and total RNA was isolated from the NK cells at the same time . As little as 1 µg/ml of Acai PS was able to reduce LVS replication in macrophages co-cultured with autologous NK cells ( Figure 2A ) . When Acai PS-treated macrophages were cultured without autologous NK cells , Acai PS-mediated protection occurred only at elevated concentrations ( ≥100 µg/ml ) and varied from donor to donor ( data not shown ) . While Acai PS was not found to augment IFN-γ mRNA expression by NK cells in the absence of infected macrophages ( data not shown ) , Acai PS did enhance IFN-γ mRNA expression by NK cells co-cultured with F . tularensis LVS-infected macrophages ( Figure 2B ) in a manner inversely correlated with intracellular replication of LVS . Acai PS also augmented TNF-α mRNA by NK cells co-cultured with F . tularensis LVS-infected macrophages ( Figure 2B ) ; however , this effect was not consistent amongst all donors ( data not shown ) . Acai PS was not found to consistently enhance mRNA's characteristic of cytotoxic activity ( granzyme B , perforin , TRAIL ) or the expression of IL-17 and IL-21 by NK cells co-cultured with F . tularensis LVS-infected macrophages ( Figure 2B ) . Since Acai PS enhanced the resistance of human primary macrophages co-cultured with NK cells to F . tularensis infection , subsequent studies addressed the relevance of IFN-γ to this protection . Macrophages were prestimulated with Acai PS ( 100 µg/ml ) overnight , infected with wild-type F . tularensis SchuS4 ( MOI∼30 ) , and then cultured with or without purified , autologous NK cells , some of which had been prestimulated with Acai PS ( 100 µg/ml ) overnight . CFU determinations were performed at 20 h after macrophage infection . Similar to what was observed with LVS , Acai PS treatment of human macrophages alone had no effect on intracellular bacterial burden , while Acai PS treatment of macrophage/NK cell co-cultures reduced intracellular bacterial burden >100 fold without affecting phagocytosis ( Figure 3A ) . Neutralization of IFN-γ completely ablated the protective capacity of Acai PS , while neutralization of IFN-γ in the absence of Acai PS or NK cells had no effect on intracellular bacterial replication ( Figure 3B ) . The addition of 400 µM L-NMA to co-cultures treated with Acai PS had no effect upon bacterial replication , and NO was not detected via the Griess Reaction , indicating that the protective effect of IFN-γ induced by Acai PS is independent of NO production ( data not shown ) . We previously found Acai PS to induce immunomodulatory effects when instilled into the lungs of naïve mice [19] . In particular , Acai PS was shown to induce IL-12 , which is protective against F . tularensis LVS infection [25] . To assess whether Acai PS could confer protection against pulmonary infection with virulent F . tularensis SchuS4 , C57BL/6 mice were treated nasally with 10 , 100 , or 1000 µg of Acai PS 24 h prior to aerosol infection with F . tularensis SchuS4 , and changes in body weight and morbidity were recorded over time for up to 28 days after infection . Treatment of mice with 100 µg of Acai PS led to 80% survival , while 10 or 1000 µg Acai PS doses exhibited less potency ( Figure 4A ) . Importantly , mice treated with Acai PS that survived infection showed negligible weight loss ( Figure 4B ) and clinical symptoms ( data not shown ) ; indicating Acai PS confers protection against both morbidity and mortality induced by virulent F . tularensis infection . Since the 100 µg dose of Acai PS was found to be optimal against aerosol challenge , in subsequent experiments mice were treated with 100 µg Acai PS at various time points after infection with F . tularensis SchuS4 . When delivered by the intranasal ( i . n . ) route immediately after aerosol infection , Acai PS conferred 70–80% survival upon treated mice ( Figure 4C ) , while all vehicle-treated animals succumbed to infection . Sixty percent of mice treated i . n . with Acai PS 24 h after aerosol challenge with F . tularensis SchuS4 survived , and even when Acai PS was given 48 h after infection , 33% of animals still survived ( Figure 4C ) . As described above for prophylactic therapy , animals treated with Acai PS after aerosol infection that survived challenge displayed negligible weight loss and clinical symptoms ( data not shown ) . Oral treatment of animals with Acai PS also conferred some level of protection against type A F . tularensis infection; however , this effect was variable ( Table S4 ) . To determine the mechanism by which Acai PS confers protection against F . tularensis infection , expression of intracellular IFN-γ by pulmonary leukocytes was assayed by flow cytometry . These studies utilized a 1000 µg Acai PS pretreatment , which we found to be optimal to protect against intranasal F . tularensis SchuS4 challenge ( data not shown ) . The finding that a 1000 µg Acai PS dose was optimal against i . n . F . tularensis SchuS4 infection , while a 100 µg Acai PS dose was optimal against aerosol F . tularensis SchuS4 infection may reflect variations in the aerosol versus i . n . challenge models used in this study . We found i . n . pretreatment of mice enhanced intracellular expression of IFN-γ by NK T cells within two days after F . tularensis SchuS4-infection ( Figure 5A ) . In addition , while Acai PS reduced bacterial burdens in the lungs and spleens of F . tularensis SchuS4 , neutralization of IFN-γ abrogated this effect ( Figure 5B–C ) . Stimulation of innate immunity with an immunotherapeutic such as Acai PS would be particularly valuable in situations where the etiological agent of disease is unknown , such as a bioterrorist attack , as induced innate immune responses are often capable of providing protection against a broad range of organisms . In addition , immunotherapy could be of particular benefit to counter infections due to bacteria that are intrinsically resistant to antibiotics , such as B . pseudomallei , a CDC Category B Bioterrorism agent . As Acai PS augmented immunity to F . tularensis infection , along with enhancing the expression of IFN-γ , which is crucial for protection from B . pseudomallei infection [26] , we tested the effects of Acai PS against B . pseudomallei infection to assay whether Acai PS has potential as a broad spectrum therapeutic to combat pulmonary infections . C57BL/6 mice were treated i . n . with Acai PS prior to , or immediately after , i . n . infection with 3×103 CFUs of B . pseudomallei 1026b . Body weights and clinical scores were recorded . I . n . treatment of mice with 100 or 1000 µg Acai PS 24 h prior to , or immediately after , B . pseudomallei infection resulted in significantly diminished weight loss and clinical scores ( Figure 6A–B ) . Treatment of mice with ≤10 µg of Acai PS or treatment of mice with Acai PS ( 10–1000 µg ) ≥24 h after B . pseudomallei infection did not result in significant protection ( data not shown ) . Next , to determine the effects of Acai PS on bacterial colonization and dissemination , mice were treated i . n . with 100 or 1000 µg of Acai PS 24 h prior to i . n . infection with 3×103 CFUs of B . pseudomallei 1026b . Bacterial burdens were determined in the lungs , spleens , and livers 72 h after infection . Treatment of mice with either dose of Acai PS reduced B . pseudomallei replication in the lungs by ∼10 , 000-fold ( Figure 6C ) . Treatment of mice with Acai PS also reduced dissemination into peripheral tissues . B . pseudomallei CFUs were below the limit of detection ( ∼33 CFUs ) in the spleens of 80% of animals treated with either dose of Acai PS , while no bacteria were recovered from the livers of any animals treated with either dose of Acai PS ( Figure 6C ) . In addition , all mice treated prophylactically with 100 or 1000 µg Acai PS ( n = 20 ) survived nasal infection with 3×103 CFUs of B . pseudomallei 1026b; however , the lethality of this dose in control animals varied from 60–100% in different experiments ( 8/10 control animals succumbed to infection ) . While 100 and 1000 µg Acai PS doses conferred similar protection against challenge with 3×103 CFUs of B . pseudomallei , a 1000 µg Acai PS provided the best protection against high dose i . n . challenge ( 1×104 CFUs ) with B . pseudomallei ( Figure S2A–B ) . These results indicate that an elevated dose of Acai PS may be required against a high dose bacterial challenge in order to protect the host against a more acute disease . To assess the mechanism of protection mediated by Acai PS on innate lymphocytes during pulmonary infection , the B . thailandensis ( BSL-1 strain ) model of Burkholderia infection [27] was used . C57BL/6 mice were given Acai PS i . n . 24 h prior to i . n . infection with 5×105 CFUs of B . thailandensis E264 . Pulmonary NK and γδ T cells were then assayed 24 h after infection by flow cytometry for the intracellular expression of IFN-γ . Acai PS enhanced IFN-γ expression by both NK and γδ T cells in B . thailandensis-infected mice ( Figure 7 ) ; indicating Acai PS can augment the IFN-γ responses of innate lymphocytes during pulmonary Burkholderia infection . As Acai PS was found to enhance the pulmonary Th1-type response , which is critical for control of Burkholderia infections [11] , 26 , 28 , the role of Th1-type responses in Acai PS-mediated protection against B . pseudomallei infections was further investigated . For these studies , mice were treated i . n . with 1000 µg of Acai PS 24 h prior to infection . Some mice were also depleted of IFN-γ or NK cells via neutralizing antibody 24 h prior to Acai PS treatment ( control animals received rat IgG ) . While the survival conferred by Acai PS in control animals was suboptimal against a high-dose challenge , Acai PS-mediated survival was totally ablated in IFN-γ-depleted mice and partially reduced in mice depleted of NK cells ( Figure 8 ) . In addition , while Acai PS mitigated clinical symptoms in B . pseudomallei-infected mice , this effect was abrogated in the absence of IFN-γ ( data not shown ) . These results indicate that , similar to what was observed in vitro and in vivo with F . tularensis; Acai PS requires IFN-γ and possibly NK cells for protection against pulmonary infection with B . pseudomallei . Enhancing innate immunity by agonist therapy could potentially augment resistance to infection and could also complement traditional vaccination and antibiotic strategies for treating infectious diseases [6] , [11] , [12] , [14] , [15] . In this study , the abilities of several natural agonists with immunomodulatory capabilities: APP , AmpB , securinine , Yamoa PS , Acai PS , and LPS [12] , [15]–[18] were assayed for their ability to potentiate macrophage resistance to F . tularensis LVS infection . While APP , AmpB , Yamoa PS , and Acai PS each enhanced NO production by LVS-infected RAW264 . 7 macrophages , only Yamoa PS and Acai PS conferred significant resistance to LVS replication at the doses tested . LPS also enhanced NO production and LVS clearance , which was not surprising , as TLR4 agonists have been shown to increase resistance to infection with F . novicida , a strain of Francisella that is virulent for mice , but rarely causes disease in humans [29] . Yamoa PS was not further examined due to concerns about possible endotoxin contamination , presumably due to endophytic bacteria residing in bark [15] , the source of this polysaccharide . In contrast , Acai PS contains low amounts of endotoxin ( <0 . 01 EU/µg ) , has MyD88-independent effects , and has immunomodulatory effects resistant to polymyxin B treatment [19] . In addition , Acai PS is non-toxic to lymphocytes at concentrations up to 500 µg/ml [19] and is not found to have direct antibacterial ( cytotoxic ) effects against Francisella in PBS or cell culture media ( data not shown ) . Acai PS is derived from the berry of the palm tree Euterpe oleracea Mart . indigenous to the Amazon River basin in South America . This fruit is commonly used to make beverages and food additives and is used as a herbal medicine [30]–[34] . Biochemical studies reveal Acai contains numerous compounds , particularly anthocyanins , proanthocyanidins , and other flavonoids [34] . While many studies have focused on the antioxidant properties of Acai [33] , [35]–[38] presumably attributable to its polyphenols and related classes of compounds , here we concentrated on the activities of Acai PS as the polysaccharide fraction , rather than the polyphenol fraction of Acai , induces a proinflammatory response [19] . We previously demonstrated that Acai PS stimulates both γδ T cells and myeloid cells in vitro and incites the recruitment of neutrophils and activates DCs/macrophages to the lung in vivo [19] . Therefore , since Acai PS has potent immunomodulatory activities and is effective at restricting the replication of F . tularensis LVS in RAW264 . 7 cells , it was investigated for its potential as an innate immune agonist . In addition to augmenting the clearance of F . tularensis LVS and SchuS4 in RAW264 . 7 cells via NO , Acai PS also enhanced cell surface expression of CD11b , CD40 , CD80 , CD86 , MHC class II , and TLR2 in a dose-dependent manner in both mock- and F . tularensis LVS infected-macrophages; however , TLR4 expression was downregulated . TLR4 expression has been shown to be downregulated following LPS stimulation [39] , and while Acai PS is low in endotoxin ( <0 . 01 EU/µg ) , is resistant to polymyxin B neutralization , and has MyD88-independent effects [19] , it is possible Acai PS may still signal through TLR4 via an alternative mechanism such as TRIF [40] . While Acai PS was able to reduce the intracellular replication of F . tularensis in RAW264 . 7 cells , Acai PS was not found to induce NO or restrict the replication of F . tularensis in primary human macrophages or murine BMDMs . This finding is presumably due to the fact that primary cells and , in particular human macrophages , do not produce NO as readily as do macrophage cell lines [41] , and such findings stress that cell lines are not always a suitable surrogate for primary cells . While Acai PS did not enhance the clearance of F . tularensis in macrophages alone , Acai PS can also activate innate lymphocytes in addition to macrophages [19] . Therefore , we adapted a co-culture system in which we tested the effect of Acai PS treatment on human monocyte-derived macrophages ( cultured with or without autologous NK cells ) infected with F . tularensis . While Acai PS was not able to directly stimulate human primary macrophages for clearance , a ∼100–1000-fold reduction in replication occurred when macrophages were co-cultured with autologous NK cells . Others have shown murine NK cells stimulated in vivo could impair intracellular replication of F . tularensis LVS in vitro [42] , and depletion of NK cells reduces the time to lethality during pulmonary infection [43] . Of particular interest is that NK cells are a major source of IFN-γ in pulmonary tularemia [44] . RT-PCR analysis revealed Acai PS stimulated human NK cells co-cultured with LVS-infected macrophages possessed elevated levels of IFN-γ mRNA , while neutralization of IFN-γ in vitro diminished the protective effect of Acai PS in macrophages infected with F . tularensis LVS or SchuS4 . NO was not detected in cell culture supernatants from our human macrophage/NK cell co-cultures , and iNOS inhibition had no effect on replication , indicating the protection conferred by Acai PS-induced IFN-γ in the human co-culture model is NO-independent , similar to what others have described for IFN-γ treated macrophages infected with F . tularensis SchuS4 [45] . Treatment of NK cells in the absence of Francisella-infected macrophages did not result in robust induction of IFN-γ mRNA , indicating there may be a synergetic effect of Acai PS and infection upon NK cells . Our previous finding that Acai PS induces IL-12 in vivo may indicate the macrophage is responsible for IL-12 production , which in turn induces IFN-γ mRNA by the NK cell . Indeed , neutralization of IL-12 in vitro did reduce IFN-γ mRNA by NK cells in some co-cultures treated with Acai PS ( data not shown ) . While it is known human NK cells can enhance the clearance of intracellular organisms , such as Brucella in autologous macrophages via contact-dependent , cytotoxic mechanisms [46] , the effect of Acai PS on NK cell mediated-cytoxicity may be minimal , since marked differences in the mRNA expression for perforin , granzyme B , or TRAIL were not observed by Acai PS-treated NK cells co-cultured with autologous LVS-infected macrophages . As IFN-γ was induced by Acai PS in vitro , and is essential for protection against experimental tularemia [43] , we assessed whether Acai PS could confer protection against in vivo challenge by employing an aerosol model of type A F . tularensis infection thought to most mimic human disease [47] . We utilized F . tularensis SchuS4 for all our in vivo infections , because , while F . tularensis LVS is widely used as a model organism to study immunological responses [48] , emerging evidence suggests the in vivo immune response differs between SchuS4 and LVS [24] , [47] , [49] , and immunotherapeutic strategies that confer potent protection against pulmonary LVS infection only confer partial or negligible protection against pulmonary infection with SchuS4 [6] , [50] , [51] . Since Acai PS enhanced the clearance of Francisella in murine macrophages and in human macrophages co-cultured with NK cells , and because Acai PS had potent immunomodulatory effects in the lung [19] , Acai PS was tested as a mucosal immunotherapeutic to treat pulmonary type A F . tularensis infections . It was found that i . n . pretreatment of mice with Acai PS conferred up to 80% protection against F . tularensis-induced mortality , which , to our knowledge , is the highest degree of protection demonstrated by an immunotherapeutic and also represents the first mucosal immunotherapeutic to confer significant survival against pulmonary type A F . tularensis infection . Importantly , Acai PS provided significant protection when administered i . n . within 48 h after pulmonary infection and thus is the first immunotherapeutic demonstrating post-exposure protection of any kind against pulmonary type A F . tularensis infection . Acai PS was also able to reduce bacterial burdens in the lungs and spleens of mice infected with F . tularensis SchuS4 . In addition , similar to what was observed in vitro in human cells , Acai PS augmented IFN-γ expression by NK cells in the lungs of treated animals infected with F . tularensis SchuS4 , while neutralization of IFN-γ abrogated the protective effect of Acai PS . The finding that Acai PS is able to protect against infection even when administered one or two days after infection , at which time SchuS4 is already present in the spleen and liver , is intriguing . Mucosal administration of therapeutics can have systemic effects , and compounds delivered nasally enter the bloodstream . In preliminary studies , we have found that mucosal administration of Acai PS enhances serum levels of TNF-α ( Holderness et al , manuscript in preparation ) . As TNF-α is protective against tularemia , it is possible that nasal administration of Acai PS also has an effect against systemic replication of Francisella , which may account for the post-exposure protection conferred by Acai PS observed here . Acai PS is also heat-resistant , and we have found it to have potent protective effects after shipment at ambient temperature , as demonstrated by the protection observed in Figure 6A–C . Therefore , since Acai PS does not require refrigeration ( cold-chain management ) and adapts a needle-free mucosal method of administration , it offers a practical strategy during emergencies , such as pandemics or bioterrorist attacks , when expeditious treatments of the affected populace would be required [52] . The downregulation of TLR4 by Acai PS observed by flow cytometry indicated Acai PS may signal at least partially through TLR4 . However , work by others indicates TLR4 stimulation alone is an insufficient method to protect against experimental tularemia , particularly , when administered after infection . Nasal administration of a TLR4 agonist prior to , but not after , pulmonary infection with F . novicida could confer protection [29] , while intraperitoneal administration of a TLR4 agonist could confer some level of protection when given 48 h before pulmonary infection with type A F . tularensis [51]; however , this effect is diminished when the TLR4 agonist was given only at the time of infection . In addition , others have demonstrated that pulmonary administration of LPS has minimal effects upon the immune response when given 24 h after type A F . tularensis infection [53] , indicating that type A F . tularensis infection actively suppresses TLR4 signaling . Since we found Acai PS has potent protective effects when given ≥24 h after infection , it would appear that Acai PS also signals through a receptor in addition to TLR4 , and the low levels of LPS present in Acai PS are not responsible for the observed protection . Botanical polysaccharides are known to signal through a variety of receptors , including TLRs and carbohydrate receptors [54] . Work on the receptors utilized by Acai PS is ongoing in a separate study , but Acai PS appears to require both TLR4/TRIF along with carbohydrate receptors ( Holderness et al , manuscript in preparation ) to mediate its effects . Therefore , future studies of the receptors required for Acai PS-mediated signaling and protection could reveal receptors to be targeted for immunotherapy against F . tularensis and other diseases . Stimulation of innate immunity with an immunotherapeutic such as Acai PS would be particularly valuable in situations where the etiological agent of disease is unknown , as induced innate immune responses are often capable of providing protection against a broad range of organisms [6] . In addition , immunotherapy could be of particular benefit to counter bacterial infections intrinsically resistant to antibiotics . To this end , we tested Acai PS against pulmonary infection with B . pseudomallei , an organism intrinsically resistant to antibiotics , to determine if Acai PS has potential as a broad spectrum immunotherapeutic . We found Acai PS enhanced immunity to B . pseudomallei when given prior to , or immediately after , infection . Acai PS also potently restricted B . pseudomallei replication within the lungs and dissemination to peripheral tissues . To assess the effects of Acai PS on innate leukocytes during infection , we assayed the expression of IFN-γ in leukocytes from the lungs of mice infected with B . thailandensis . We found Acai PS augmented the expression of IFN-γ in both NK and γδ T cells from B . thailandensis-infected animals , indicating an enhanced Th1-type response of these cell types . Since Acai PS also enhanced the IFN-γ response of human and murine NK cells during in vitro and in vivo models of F . tularensis infection , the role of NK cells and IFN-γ in Acai PS-mediated protection against B . pseudomallei was assessed . As a result , IFN-γ was entirely responsible for Acai PS-mediated protection , while NK cells were also required to some extent . These results demonstrate Acai PS mediates protection against infection in human cells in vitro and in in vivo murine models in a similar manner as NK cells and IFN-γ are required for protection in both systems , indicating our protective effects in vivo with mice have relevance to humans . As neutralization of NK cells did not entirely ablate protection against B . pseudomallei in vivo , it is possible the effects of Acai PS on other cells are also required for protection . We have previously found Acai PS to stimulate human γδ T cells in vitro [19] , and here we show Acai PS augments the Th1-type responses of γδ T cells in infected lungs; therefore , as γδ T cells are known to confer protection against a number of intracellular pathogens such as Brucella and Listeria [55] , [56] , future studies will investigate the role of γδ T cells in Acai PS-mediated protection . In a clinical setting , an immunotherapeutic such as Acai PS would most often be used in conjunction with antibiotics . Recent studies have demonstrated immunotherapy can synergize with antibiotic therapy of bacterial infections , including Burkholderia [26]; therefore , additional studies will assess the effects of Acai PS in combination with antibiotic therapy . In summary , we show immunotherapy with natural agonists such as Acai PS is an effective means to confer protection against bacterial infection . In fact , Acai PS appears to be the most potent immunotherapeutic reported to date to combat pulmonary type A F . tularensis infections and is also the first one demonstrated to confer significant survival when given mucosally or after infection . Of particular interest is Acai PS was also able to confer protection against pulmonary infection with both F . tularensis and B . pseudomallei , as previous studies demonstrated immunotherapeutics that induce potent protection against B . pseudomallei may only confer partial or negligible protection against type A . F . tularensis [6] , [11] , [50] , indicating Acai PS has broad spectrum immunotherapeutic potential to combat intracellular bacterial infections . Acai PS also enhanced the Th1 cell response of innate leukocytes during infection both in vivo and in human cells . As optimal Th1 cell immunity is required for protection against a broad range of infections , Acai PS should be investigated as a possible immunotherapy that could augment or complement traditional antibiotic and vaccination strategies against a range of pathogens . All animal care and procedures were in accordance with the recommendations in the Guide for the Care Use of Laboratory Animals of the National Institutes of Health . All animal protocols were approved by Institutional Animal Care and Use Committees at Montana State University ( protocol approval: 2009-27 , 2011-25 ) or Colorado State University ( protocol approval 09-001 ) and all efforts were made to minimize suffering . Human subjects testing was performed in accord with the Institutional Review Board of Montana State University ( protocol approval: JS072809 ) , and written , informed consent was obtained from all individuals . F . tularensis SchuS4 or LVS was cultured in modified Mueller–Hinton ( MMH ) broth ( 0 . 025% ferric pyrophosphate , 2% IsoVitaleX and 0 . 1% glucose ) at 37°C with constant shaking overnight , aliquotted into 1 ml samples , frozen at −80°C , and thawed just before use , as previously described [57] . Frozen stocks were titrated by enumerating viable bacteria from serial dilutions plated on MMH agar ( 0 . 025% ferric pyrophosphate , 2% IsoVitaleX , 0 . 1% glucose , and 0 . 025% fetal bovine serum ) . The numbers of viable bacteria in frozen stock vials varied by less than 5% over a 10 month period . These stocks were used to generate cultures for F . tularensis SchuS4 or LVS infection studies . Frozen stocks of B . pseudomallei of known titers were prepared from cultures grown in Luria-Bertani ( LB ) broth ( BD Biosciences , San Jose , CA ) by freezing the cultures in LB medium containing 20% glycerol . Inocula for in vivo infection with B . pseudomallei were prepared by thawing and diluting frozen stocks in sterile PBS [11] . All experiments with F . tularensis SchuS4 or B . pseudomallei 1026b were performed in biosafety level 3 facilities at Montana State University or Colorado State University . Burkholderia thailandensis E264 was acquired from ATCC ( Manassas , VA ) . Prior to infection , B . thailandensis were grown from frozen glycerol stock in LB at 37°C overnight and freshly diluted 1∶100 into 100 ml of LB . The bacteria were grown to an optical density ( OD600 ) of 1 . 9 ( ∼1×109 cfu/ml ) and diluted in PBS prior to infection [27] . Six-week-old female C57BL/6 or BALB/c mice were purchased from Charles River Laboratories . All mice were housed in sterile microisolater cages in the laboratory animal resources facility at Montana State University or the Biohazard Research Building BSL-3 facility at Colorado State University and were provided with sterile water and food ad libitum . Acai fruit pulp was obtained from Acai Berry Pure ( Acai Berry Pure Bulk; Carlsbad , CA ) . Polysaccharides were isolated from this powdered Acai , as described previously [15] , [19] . Briefly , 1500 g of Acai powder was extracted with 8 liters boiling distilled H2O for 1 h . The aqueous extract was then centrifuged at 2 , 000× g for 15 min , and a 4-fold volume of ethanol was added to the supernatant to precipitate polysaccharides overnight at 4°C . The precipitate was pelleted by centrifugation , re-dissolved in distilled H2O and centrifuged at 2 , 000× g for 15 min . The supernatant fluid ( crude polysaccharide extract ) was fractionated using ion-exchange chromatography on a DEAE-cellulose column equilibrated with 0 . 05M Tris-HCl buffer ( pH 8 . 0 ) . Bound material was sequentially eluted with 0 . 05M Tris-HCl buffer and 2M NaCl . The presence of polysaccharides in the unbound fraction , eluted with 0 . 05M Tris-HCl buffer , was minimal ( <0 . 1% of total bound fraction ) . The Acai-PS fraction was generated from this preparation after concentration in an Amicon concentrator with a 10 kDa Amicon PM10 membrane ( Millipore; Billerica , MA ) . This preparation yields a fraction that is >99% carbohydrate and >92% polysaccharides . Monosaccharide analysis reveals that Acai PS consists primarily of arabinose , galacturonic acid , and galactose [19] . Endotoxin levels were determined using the LAL assay , as described [19] . Endotoxin levels for the Acai PS used in this study were <0 . 01 EU/µg . Bone marrow-derived macrophages ( BMDM ) were generated by flushing the bone marrow from the femurs of BALB/c mice with RPMI 1640 media . Freshly collected bone marrow cells were cultured overnight in complete media ( CM; RPMI 1640 , 10% fetal bovine serum [Atlanta Biologicals , GA] , 10 mM HEPES buffer , 10 mM nonessential amino acids , 10 mM sodium pyruvate ) containing 5 ng/ml recombinant murine M-CSF ( Peprotech , Rocky Hill , NJ ) . The non-adherent cells were then collected and cultured for an additional six days in CM with 30 ng/ml M-CSF to generate macrophages . Murine BMDMs or RAW264 . 7 macrophages were seeded at 1×106 cells/well in CM without antibiotics in 24-well microtiter plates ( BD Labware , Franklin Lakes , N . J . ) at 37°C/5% CO2 prior to infection . Macrophages were infected with Francisella tularensis LVS at an MOI of ∼300 or F . tularensis SchuS4 at an MOI of ∼30 for two h at 37°C . Cells were then washed once with PBS , and then fresh CM containing 50 µg/ml gentamicin were added to each well , and cells were incubated for 30 min at 37°C to kill extracellular bacteria . Cells were then washed twice with PBS , and then fresh complete media without antibiotics were added to the wells for the remainder of the experiment ( this is considered the “0 hour” time point ) . For time points of >8 h , gentamicin was added to the wells for the last 45 min of incubation . To enumerate intracellular bacteria , cells were washed three times with PBS and then lysed with sterile deionized water . Serial logarithmic dilutions of macrophage lysates were then performed and plated in triplicate onto MMH agar for incubation at 37°C/5% CO2 for 2–3 days . In some cases , macrophages were stimulated at various time points before or after infection with varying concentrations of agonist . In addition , L-NMA ( Sigma-Aldrich , St . Louis , MO ) was added to selected wells to inhibit NO production . Supernatants were collected and frozen until analysis by cytokine ELISA or the Griess reaction . Supernatants from Francisella-infected RAW264 . 7 and human macrophages were collected at various time points and measured for cell death , production of cytokines and , the oxidized product of NO . Cell death was determined by measuring lactate dehydrogenase LDH release using a cytotoxicity detection kit according to manufacturer's instructions ( Roche , Indianapolis , IN ) . Cytokine-specific ELISAs were performed , as described previously [58] , [59] . All NO2− detection chemicals were obtained from Sigma-Aldrich . Aliquots of 50 µl of cell culture supernatant were reacted with equal volumes of Griess reagent ( 1% sulfanilamide , 0 . 1% naphthylenediamine dihydrochloride , 2 . 5% H3PO4 ) at room temperature ( RT ) for 10 min . Sodium nitrite was used to generate a standard curve for NO2− production , and peak absorbance was measured at 550 nm with a Thermomax microplate reader ( Molecular Devices , Sunnyvale , CA ) . Cell-free medium contained <1 . 5 µM NO2− . RAW264 . 7 cells were detached from 24-well culture plates , resuspended , and washed . Immunofluorescent staining for cell surface molecule expression was performed using the following fluorochrome-labeled mAbs from eBioscience ( San Diego , CA ) , Biolegend ( San Diego , CA ) , or BD Biosciences: CD11b ( clone M1/70 ) , CD80 ( clone 16-10A1 ) , CD40 ( clone 3/23 ) , TLR4 ( clone MT5510 ) , CD86 ( clone GL1 ) , TLR2 ( clone T2 . 5 ) and MHC-II ( clone AMS-32 . 1 ) . Fluorescence was acquired on FACSCaliber , LSRII , or Canto ( BD Biosciences ) . FlowJo ( Tree Star , Ashland , OR ) software was used for analysis . Heparinized human peripheral blood was subjected to Histopaque 1077 ( Sigma-Aldrich ) density gradient centrifugation . The collected mononuclear cell fraction was collected , and monocytes were isolated with CD14 microbeads ( Miltenyi Biotec , Auburn , CA ) according to manufacturer's instructions . Monocytes ( >95% purity , 104–105/well ) were then seeded into 48-well microtiter plates in CM without antibiotics , supplemented with 10 ng/ml GM-CSF ( Peprotech , Rocky Hill , NJ ) for 4–5 days at 37 C°/5% CO2 to generate macrophages . Human macrophages were infected with F . tularensis LVS ( MOI∼300 ) or SchuS4 ( MOI∼30 ) in the same manner as described above for murine macrophages . One day prior to macrophage infection , autologous “untouched” NK cells were isolated from human PBMCs using an NK cell isolation kit from Miltenyi Biotec according to manufactures instructions . Isolated NK cells ( >95% purity ) were cultured overnight in complete media at 37 C°/5%CO2 with or without agonist stimulation . NK cells were washed with fresh CM prior to being added to wells containing infected autologous macrophages ( ∼2–20 NK cell/macrophage ) . To inhibit the effects of IFN–γ in vitro , a neutralizing mAb ( IFN–γ [clone B27 , 1 µg/ml] in a no azide/low endotoxin ( NA/LE ) format was purchased from BD Biosciences and added to selected wells containing Francisella-infected macrophages with or without NK cells [60] . Human NK cells cultured with or without LVS-infected macrophages and/or Acai PS were centrifuged and resuspended in RNAlater reagent ( Qiagen , Valencia , CA ) until RNA extraction . Cells were then centrifuged and resuspended in Qiagen RLT buffer prior to lysis on a Qiashredder Column ( Qiagen ) and RNA extraction with an RNeasy Mini Kit ( Qiagen ) . cDNA was generated using the Superscript III First Strand Synthesis System ( Invitrogen ) . Primers for immune-related genes ( TNF-α , IFN-γ , IL-17A , IL-21 , IL-22 , granzyme B , perforin , and TRAIL ) , along with β-actin ( endogenous control ) , were designed using the PrimerQuest application from IDTDNA . com . The reference sequences used to generate these primers are listed below ( paragraph “Accession numbers” ) . Amplicons were visualized under UV illumination on a 2% agarose gel containing GelRed ( Biotium , Hayward , CA ) . Mice were infected with F . tularensis SchuS4 at Colorado State University via a whole-body low-dose aerosol , as previously described [53] , [61] . Conscious mice within a stainless steel basket were exposed to the SchuS4 strain of F . tularensis by aerosol exposure in a Glascol Inhalation Exposure System ( Glas-Col , Inc . , Terre Haute , IN , USA ) . Prior to exposure , the nebulizer was loaded with bacteria diluted in PBS to a concentration of ∼5×106 CFU/ml . Mice were exposed to a total of ∼4×107 bacteria , aerosolized into a volume of 5 cubic feet over a period of 30 min , followed by a 20 min period of cloud decay in which airflow was maintained without introducing additional bacteria . This inoculum method generally delivers ∼50 CFUs of F . tularensis to the lungs of exposed mice and routinely results in 100% mortality and a mean time to death of 5–6 days following infection [53] . Mice infected with F . tularensis SchuS4 at Montana State University were infected with a 20 µl nasal volume ( 50 CFUs ) placed onto the anterior nares following anesthesia induced by intraperitoneal ( i . p . ) injection of 100 µl of ketamine ( 12 . 5 mg/ml ) +xylazine ( 3 . 8 mg/ml ) . For survival experiments , mice were monitored for morbidity and mortality twice daily for up to 28 days , at which time survivors were euthanized . Mice were treated with varying doses of Acai PS ( in PBS ) before or after infection . Mice were treated nasally under anesthesia ( 10 µl/nare induced by i . p . injection with ketamine/xylazine cocktail . For oral treatments , mice received 200 µl volume via gavage . Control mice were inoculated with PBS . For in vivo neutralization studies , mice were treated with 500 µg of mAb i . p . to neutralize IFN-γ ( clone XMG 1 . 2 ) on day −2 , while control mice received rat IgG [62] . In some experiments , mice were sacrificed 2 days post post-infection for CFU determination in lungs and spleens . Mouse organs were homogenized in sterile PBS , and homogenates were serially diluted and plated on MMH plates , which were then incubated at 37°C for 48 h , at which time CFUs were enumerated . For B . pseudomallei infection , mice under ketamine/xylazine-induced anesthesia were infected with i . n . ( 10 µl/nare ) with 3×103 or 1×104 CFUs of B . pseudomallei 1026b . Clinical scores were graded as 0 = normal; 1 = slightly ruffled; 2 = ruffled , sick looking; 3 = hunched posture and obviously ill; 4 = moribund; 5 = euthanized . For in vivo neutralization studies , mice received 500 µg of mAb i . p . to neutralize IFN-γ ( clone XMG 1 . 2 ) or NK cells ( clone PK136 ) on day −2 , while control mice received rat IgG [62] , [63] . Mice were sacrificed at 3 days post-infection for CFU determination in lungs , livers and spleens . Mouse organs were homogenized in sterile PBS , and homogenates were serial diluted and plated on Tryptic Soy Agar ( BD Biosciences ) plates , which were then incubated at 37°C for 48 h , at which time CFUs were enumerated . For Francisella studies , C57BL/6 mice were nasally treated with Acai PS 24 h prior to i . n . infection with 50 CFUs of F . tularensis SchuS4 . Forty-eight h after infection , lung tissue was minced followed by digestion for 1 h at 37°C in CM containing 200 U/ml collagenase , ( Sigma ) and 0 . 08 U/ml DNAse ( Promega , Madison , WI ) . The resulting cell suspensions were filtered through 35 mm NitexH nylon mesh ( Sefar America; Depew , NY ) to remove tissue debris , washed in CM , resuspended in 30% Percoll ( Pharmacia , Uppsala , Sweden ) and layered onto 70% Percoll , and subjected to density gradient centrifugation . Mononuclear cells were removed from the interface layer , washed , resuspended in CM , and cultured for 4 h in the presence of 12-myristate 13-acetate ( PMA; 50 ng/ml ) , 500 ng/ml ionomycin , and 10 µg/ml brefeldin A . Cells were then analyzed by FACS analysis using conventional methods [64] , [65] . Cells were stained for extracellular markers with fluorochrome-conjugated mAbs ( Becton Dickinson or eBioscience , San Diego , CA ) : anti-NK1 . 1 ( clone PK136 ) ; prior to fixation with 2% paraformaldehyde . Cells were then permeabilized with 0 . 2% saponin and stained for intracellular expression of IFN-γ ( clone XMG1 . 2 ) . Stained leukocytes were analyzed using an LSRII flow cytometer ( BD Biosciences ) and analyzed using FlowJo software ( Tree Star Inc . , Ashland , OR ) . For Burkholderia studies , C57BL/6 mice were i . n . treated with Acai PS 24 h prior to i . n . infection with 5×105 CFUs of B . thailandensis E264 . Twenty-four h after infection , lung tissue was processed , and cells were cultured and stained as described above . Statistical differences between two groups were determined using a Student's t test with the significance set at P<0 . 05 . For comparison between three or more groups , analysis was done by one-way ANOVA followed by Tukey's multiple comparisons test with significance determined at P<0 . 05 . For in vivo studies , significance in survival was assessed using log-rank analysis with significance set at P<0 . 05 . The GenBank ( http://www . ncbi . nlm . nih . gov ) accession numbers for DNA sequences utilized to generate primers are as follows: NM000594 ( TNF-α ) , NM000619 ( IFN-γ ) , NM002190 ( IL-17A ) , NM021803 ( IL-21 ) , NM020525 ( IL-22 ) , NM004131 ( granzyme B ) , FJ555237 ( perforin ) , BC032722 ( TRAIL ) , and NM001101 ( β-actin ) .	Activation of the innate immune system offers an alternative and effective means to counter infection , particularly , in cases when the etiologic agent is unknown , such as a potential bioterrorism attack or when the agent is resistant to antibiotics . Here we report that a natural polysaccharide extract derived from the Acai berry ( Acai PS ) has potent abilities to counter infection when applied as a mucosal immunotherapeutic . Acai PS diminishes the replication of F . tularensis in human macrophages co-cultured with NK cells in vitro . In addition , nasal treatment of mice , prophylactically or therapeutically , with Acai PS results in significant protection against morbidity and mortality against pulmonary infection with virulent F . tularensis or B . pseudomallei . Of particular interest is that Acai PS utilizes the same mechanism of protection by enhancing Th1 cell immunity by both human and murine cells . Since an optimal Th1-type response is required for protection against a wide variety of infectious diseases , Acai PS represents a novel immunotherapeutic that could augment antibiotic therapy against a broad range of pathogens .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "immunology", "biology", "microbiology" ]	2012	Nasal Acai Polysaccharides Potentiate Innate Immunity to Protect against Pulmonary Francisella tularensis and Burkholderia pseudomallei Infections
Plasmodium vivax , one of the five species of Plasmodium parasites that cause human malaria , is responsible for 25–40% of malaria cases worldwide . Malaria global elimination efforts will benefit from accurate and effective genotyping tools that will provide insight into the population genetics and diversity of this parasite . The recent sequencing of P . vivax isolates from South America , Africa , and Asia presents a new opportunity by uncovering thousands of novel single nucleotide polymorphisms ( SNPs ) . Genotyping a selection of these SNPs provides a robust , low-cost method of identifying parasite infections through their unique genetic signature or barcode . Based on our experience in generating a SNP barcode for P . falciparum using High Resolution Melting ( HRM ) , we have developed a similar tool for P . vivax . We selected globally polymorphic SNPs from available P . vivax genome sequence data that were located in putatively selectively neutral sites ( i . e . , intergenic , intronic , or 4-fold degenerate coding ) . From these candidate SNPs we defined a barcode consisting of 42 SNPs . We analyzed the performance of the 42-SNP barcode on 87 P . vivax clinical samples from parasite populations in South America ( Brazil , French Guiana ) , Africa ( Ethiopia ) and Asia ( Sri Lanka ) . We found that the P . vivax barcode is robust , as it requires only a small quantity of DNA ( limit of detection 0 . 3 ng/μl ) to yield reproducible genotype calls , and detects polymorphic genotypes with high sensitivity . The markers are informative across all clinical samples evaluated ( average minor allele frequency > 0 . 1 ) . Population genetic and statistical analyses show the barcode captures high degrees of population diversity and differentiates geographically distinct populations . Our 42-SNP barcode provides a robust , informative , and standardized genetic marker set that accurately identifies a genomic signature for P . vivax infections . P . vivax is a significant disease threat and the most widely distributed human malaria parasite [1] . Its disease burden is imposed predominantly on Asia , Central and South America , the Middle East , Oceania , and East Africa , where nearly 2 . 5 billion people are at risk for infection [1] and approximately 132–391 million clinical infections are reported annually [2] . Historically , P . vivax malaria has been considered relatively benign as it has produced lower mortality than P . falciparum [2] , the most virulent parasite and the predominant parasite in Africa . This perceived lack of severity , combined with an inability to maintain continuous in vitro cell culture , has hampered research on P . vivax for decades [3 , 4] . Nevertheless , reports of drug resistance [5] and increased virulence [6] make P . vivax elimination efforts urgent . Genotyping assays with high sensitivity , simplicity , and low cost are powerful tools to identify parasite infections and may provide insight into the population genetics and diversity of P . vivax infections over time [7] . Highly polymorphic microsatellites have been the mainstay of this kind of analysis , revealing high levels of genetic diversity and multiple-clone infections [8–11] and proving useful for mapping focal outbreaks within countries [7 , 12–14] . However , the interpretation of microsatellite assays can be difficult to standardize across laboratories , and microsatellites are less amenable than SNPs to high-throughput genotyping [15] . The recent whole genome sequencing of numerous P . vivax genomes has been an important advance , uncovering a large number of SNPs [16–20] that could potentially be adapted to genotyping applications . Through these sequencing efforts , researchers have discovered that P . vivax is nearly twice as diverse as P . falciparum at the global population level [20] . This parasite genomic diversity can be explored efficiently in larger collections of samples with a SNP molecular barcode . A SNP barcode contains a combination of SNPs that together express the unique pattern of variation for the parasite sequence . Several successful SNP barcodes have been developed and deployed for P . falciparum [21–24] using high resolution melting ( HRM ) analysis . Their utility has been validated as genotyping tools to monitor malaria transmission , evaluate interventions , and pursue local and regional malaria elimination [25] . HRM is a simple , rapid and low-cost SNP genotyping method based on amplicon melting . Its advantages are that it requires only a saturating double stranded DNA dye and a post-PCR short melting step [26] . The SNP allele is identified by the amplicon melting curve , which depends on its sequence length , percent GC content , and heterozygosity . By melting the amplicon products of a PCR reaction through a gradual increase in temperature , slight genetic differences such as SNPs can be detected by monitoring the dissociation rate of the DNA with a dye and plotting the progress as a melt curve [27] . This approach provides a robust , cost-efficient , closed-tubed system that allows for rapid genotyping of DNA sequence variants without costly post-PCR methods such as direct sequencing [28] . In this study we used HRM analysis to design 42 SNP genotyping assays using whole amplicon melting . Together these assays create a 42-SNP barcode for P . vivax that can uniquely identify parasite infections and differentiate their geographic origins , and may ultimately provide insight into changes in parasite population dynamics and transmission . We identified candidate SNPs for our barcode by performing SNP calling with the Genome Analysis Toolkit ( GATK ) Unified Genotyper [29] using available sequence data ( with quality scores of > 30 ) on isolates from Central America ( Salvador 1 [16] ) , South America ( Brazil I [20] and Acre4 ( E . Winzeler , University of California San Diego , personal communication ) ) , Peru ( IQ07[30] , PQSJ62 ( E . Winzeler , University of California San Diego , personal communication ) , and PQSJ96 ( E . Winzeler , University of California San Diego , personal communication ) and Asia ( North Korea , and India VII ) [20] . Additionally , we used available SNP annotation data from sequenced isolates from Cambodia ( C08 , C15 and C127 ) and Madagascar ( M08 and M19 ) [19] . We designed primer pairs using the LightScanner primer design software version 2 . 0 ( BioFire Defense , U . S . A . ) . The software uses HRM design parameters including: primer length of 18–28 nucleotides , melting temperature ( Tm ) of 58–60°C , GC content of 40–60% , and amplicon size of < 60 base pairs ( bp ) . We then checked the primer pairs for the potential to form primer-dimers or alternative amplicons using Genious software version 6 . 1 ( Biomatters Ltd , New Zealand ) . If the reaction specificity was acceptable , we screened the primers for species-specificity to P . vivax using a BLAST genomic database search ( National Library of Medicine ( NLM ) U . S . A ) . We used the P . vivax strains ( North Korea , India VII , and Brazil I ) that were sequenced at the Broad Institute [20] as internal controls to identify both the reference and alternate alleles for each assay based on the Sal 1 strain . All three strains are clonal infections that were adapted for growth in monkeys and are publicly available via the Malaria Research and Reference Reagent Resource Center ( MR4 ) . The genomic DNA of all three strains was derived from leukocyte depleted monkey blood . Animals were obtained from the National Institutes of Health ( NIH ) -approved sources , housed in agreement with the Animal Welfare Act and the Guide for the Care and Use of Laboratory Animals ( ILAR , 1996 ) ; procedures were performed following the NIH Guidelines under protocols approved by the Animal Care and Use Committee of the National Institute of Allergy and Infectious Diseases ( Animal Study Protocol LMVR 15 ) . We used 87 clinical samples for validation of our barcode assays . The samples were derived from blood specimens identified by microscopy to contain P . vivax and were collected at study sites in Brazil ( 31 samples ) , Sri Lanka ( 19 samples ) , Ethiopia ( 15 samples ) and French Guiana ( 22 samples ) with informed consent under human subject guidelines approved by the relevant Institutional Review Boards . The clinical samples from French Guiana were collected by venous blood draw between 2006 and 2010 at study sites located in Cayenne ( 4 ) , Inland ( 3 ) , Lower Maroni ( 4 ) , Upper Maroni ( 4 ) , Lower Oyapock ( 4 ) and Upper Oyapock ( 3 ) . All 31 clinical samples from Brazil were collected by venous blood draw between 2005 and 2007 in the rural community of Granada [31] . All 19 samples from Sri Lanka were collected between 2003 and 2008 by venous blood draw from patients attending medical health clinics in Trincomalee [32] . All 15 samples from Ethiopia were collected by whole blood spotted on Flinders Technology Associates ( FTA ) membranes between 2006–2008 from patients with fever attending health clinics in Assendabo [32] . For all clinical samples from Brazil the Institutional Review Board ( IRB ) approval of the study protocol was obtained from the Ethical Review Board of the Institute of Biomedical Sciences of the University of São Paulo , Brazil ( 773/2007 ) . Written informed consent was obtained from all study participants or their parents/guardians . For all clinical samples from Sri Lanka and Ethiopia the IRB approval was obtained from the Human Subjects Committee of the Harvard School of Public Health ( #P10299–106/0209GENE ) and the Ethical Review Committee of the Faculty of Medicine , University of Colombo ( EC/08/092 ) . Written informed consent was obtained from study participants or their parents/guardians . Illiterate native-speaking study participants who could understand the native language , but were physically unable to talk or write , were entered into the study by oral consent using consent forms that were approved by both the Human Subjects Committee of the Harvard School of Public Health ( #P10299–106/0209GENE ) and the Ethical Review Committee of the Faculty of Medicine , University of Colombo ( EC/08/092 ) . An impartial third party was present to witness the entire consent process and signing of the consent documents . After informed consent was obtained , all samples were de-linked from patient identifiers and provided as discarded samples from the clinic . For all clinical samples from French Guiana the analyzed samples were all obtained by blood collections required by the standard medical care for any patient presenting fever on hospital admission in French Guiana . According to the French legislation ( article L . 1211–2 and related of the French Public Health Code ) , biobanking and secondary use for scientific purpose of human clinical remaining samples are possible as long as the corresponding patients are informed and have not given any objection to them . In the present research , this requirement is fulfilled: information is given to every patient through the Hospital brochure entitled ‘‘Information for patients” , and no immediate or delayed patient opposition was reported by the hospital clinicians to the Malaria NRC . Moreover , in application of French legislation ( article L . 1243–3 and related of the French Public Health Code ) , samples received at the Malaria NRC had been registered for research purpose in the NRC biobank declared to both the French Ministry for Research and a French Ethics Committee , which both approved and registered this thematic biobank ( declaration number DC-2010–1223; collection Nu2 ) . No institutional review board approval is required according to the French legislation . Genomic DNA was extracted from clinical samples from using the QIAmp DNA Blood Mini Kit ( Qiagen , Germany ) [31 , 32] . We performed whole genome amplification ( WGA ) on the clinical samples using the Illustra GenomiPhi V2 DNA Amplification Kit ( GE Healthcare Bio-Sciences , Piscataway , NJ , USA ) according to the manufacturer’s instructions . For each WGA reaction , 1 μl of each clinical sample was used , yielding 40 ul of amplified material . Following WGA , we purified the DNA using Agencourt AMPure XP system ( Beckman Coulter , Inc . , Beverley , MA , USA ) according to manufacturer’s instructions . Following genome amplification , we quantified the concentration of total DNA in clinical samples based on OD260 using a NanoDrop 3300 Fluorospectrometer ( Thermo Scientific , Waltham , MA , USA ) . Herein , we refer to the DNA concentration as total DNA , for clinical samples since it contains the presence of both human and P . vivax material . All DNA solutions were diluted in 1X Tris-EDTA ( TE ) Buffer ( VWR , Radnor , PA , USA ) . We identified the optimal PCR profile in clinical samples as a two-step protocol . The protocol includes: 120 sec at 95°C; 40 cycles of 94°C , 30 sec and 64°C , 60 sec and a final HRM cycle of 95°C , 15 sec; 55°C , 15 sec; and 95°C , 15 sec . The optimized master mix for the PCR reaction contains: 3 μl of DNA sample containing 1 ng/μL DNA in 1X TE Buffer , 1 μL of PCR grade water ( VWR , Radnor , PA , USA ) 4 μL of 2 . 5X LightScanner master mix ( BioFire Diagnostics Inc . , Salt Lake City , Utah , USA ) , and 2 μL of primer solution containing 0 . 1 to 0 . 5 μM of forward and 0 . 1 to 0 . 5 μM reverse primers diluted in 1X TE Buffer depending on the individual assay ( Integrated DNA Technologies , Inc . , Coralville , Iowa , USA ) for a total reaction volume of 10 μL . Our studies show that the presence of human DNA did not interfere with the performance of the assays . To improve the sensitivity as well as the signal-to-noise ratio of the assays in clinical samples , we identified the optimal primer pair concentration for each assay by performing an optimization matrix . We evaluated five different primer concentrations ( 0 . 1 μM , 0 . 2 μM , 0 . 3 μM , 0 . 4 μM , and 0 . 5 μM ) and their combinations , leaving all other reaction conditions unchanged . Following amplification , we evaluated the amplicons for the designed product size as described above . Next , we verified genotyping calls for each assay by melt curve analysis using the Eco Real-Time PCR System software version 4 . 1 ( PCRmax , Staffordshire , UK ) or the Applied Biosystems ViiA 7 Real-Time PCR System ( Life Technologies , Grand Island , NY , USA ) . We considered a primer concentration to be optimal when the amplification: ( 1 ) resulted in an amplicon of the correct size , ( 2 ) performed at a cycle threshold ( Ct ) less than 30 ( as anything greater than 30 could generate unreliable melting profiles [33] ) , and ( 3 ) produced a single peak in the melt profile with no primer-dimer artifacts identified by Eco Real-Time PCR System software version 4 . 1 or the Applied Biosystems ViiA 7 Real-Time PCR System . To perform HRM SNP barcoding , we first quantified the concentration of DNA for all clinical samples , as described above using a Nanodrop . We then diluted the samples in 1X TE buffer to a concentration of total DNA ( human and parasite DNA ) at 1 ng/μl based on based on OD260 . For all assays , we included sequenced control samples in each assay plate to identify the reference and alternate SNP temperature melt ( Tm ) curves . Next , we prepared a master mix ( described above ) for each assay . We calculated the amount of master mix by multiplying the volume of each component by the number of PCR wells or test samples . We added 7 μl of the master mix to each well in the PCR plate . We gently vortexed and centrifuged the reaction mixture , then added 3 μl of the DNA dilution at 1 ng/μl ( final assay concentration 3 ng/μl ) to each well for a total reaction volume of 10 μl . We placed a seal on top of the PCR plate with a roller and then centrifuged at 1000 RPM for 1 min . We optimized this protocol to be performed in either a 48-well plate in the Eco Real-Time PCR System or in a 384-well plate in the Applied Biosystems ViiA 7 Real-Time PCR System . We used HRM software on the Eco Real-Time PCR System or the Applied Biosystems ViiA 7 Real-Time PCR System for HRM genotyping analyses . See supplemental materials S1 File for the full barcode assay protocol . To identify a sample genotype we analyzed the derivative Tm curve for each assay . Because P . vivax is haploid in peripheral blood , we defined the detection of two alleles at any assay position as a polymorphic site and the detection of one allele as a monomorphic site . We considered samples with more than one polymorphic SNP position to be polygenomic . We used the control samples to identify the reference and alternate alleles for each SNP . Because the HRM method relies on saturating DNA dyes , the reference and alternate alleles produce single Tm peaks that differ by 0 . 7–1 . 2°C [26] . This allowed us to identify monomorphic genotypes by their single Tm peak and their alignment with a control Tm curve . By contrast , we identified polymorphic genotypes by their skewed or shifted Tm curves . We evaluated assay reproducibility by genotyping samples in duplicate on two different genotyping platforms: the PCRmax Eco ( 48-well format ) and the Applied Biosystems ViiA 7 ( 384-well format ) machines . We screened the 42 assays against three clinical samples and three sequenced controls in duplicate and calculated the difference in Tm values for each assay measured on both platforms . We then calculated the mean and standard deviation ( SD ) of these Tm differences to use as a metric of the reproducibility and robustness of each assay on both platforms . Additionally , we evaluated the reproducibility of the 42 assays in our pilot screen by evaluating the variability of the Tm value across duplicates for all monomorphic genotype calls from the 87 clinical samples and three control samples . Following DNA quantification , we evaluated the robustness of the assays by determining the reaction efficiency and limit of detection using the standard curve method . To perform the standard curve reaction , we prepared a 10-fold serial dilution of DNA from both sequenced control and clinical samples at a starting concentration of 300 ng/μl total DNA ( human and primate material ) . We then used these serial dilutions as the template DNA in the qPCR-HRM assays and measured the percent efficiency and identified the minimal DNA concentration . We identified the limit of detection for all assays by setting the criterion of having a cycle threshold ( Ct ) of 30 , as any Ct greater than 30 could generate shifted or unreliable melting profiles [33] . The slope of the standard curve can be translated into an efficiency value and the acceptable standard for assay efficiency is between 90–100% ( −3 . 6 ≥ slope ≥ −3 . 3 ) . We evaluated the ability of all assays to detect mixed infections in both the sequenced controls and clinical samples . We quantified the DNA of each sample using the PrimerDesign P . vivax probe according to the manufacturer’s protocol and normalized the DNA to 200 copies/μl . We then prepared DNA mixtures that contained the following ratios of reference to alternate allele: 1:10 , 1:4 , 1:2 , 1:1 , 10:1 , 4:1 , and 2:1 . Finally , we performed all 42 assays in duplicate using the mixed DNA templates and analyzed the melt profiles to assess the ratios of reference to alternate allele present . We evaluated the ratio of the reference and alternate alleles present in each mixture with and without WGA . First , we identified 10 assays that represented each SNP type in the barcode ( CT , CA , GT , and GA ) and a pair of monogenomic clinical samples that represented the reference and alternate alleles for each assay . We then prepared mixtures in ratios of 1:1 , 1:2 , 1:4 , 2:1 and 4:1 , following the method described above . Since these samples were identified as monogenomic by our SNP barcoding , the resulting mixture had a complexity of infection of two by design . Next , we amplified 1 μl of each mixture and purified the resultant 40 μl ( as described in DNA sample preparation ) . We performed all 10 assays in duplicate and analyzed the melt profiles to assess the percentage of the reference and alternate allele present in each mixture with and without WGA . We calculated the MAF from allele counts for each SNP in each population . For each polymorphic genotype we counted calls for both the reference and alternate alleles , each with a half contribution compared to monomorphic genotypes . After obtaining the MAF values for each of the populations , we calculated the average MAF ( AMAF ) as the unweighted mean of the MAF values for the four populations for each SNP . Since we do not know the number of parasite types within a polygenomic sample , the MAF calculation assumes a complexity of infection of two , thus this approach may inflate the MAF calculation . We calculated LD using the r2 statistic [34] . We calculated r2 within each of the four collection sites ( Brazil , Sri Lanka , Ethiopia , and French Guiana ) for all combinations of SNP pairs that fall on the same chromosome yielding 66 pairs of barcode SNPs . Although the polygenomic samples were included in this analysis , the r2 calculation was only performed on pairs of genotypes that were monomorphic within such samples . To adjust for background LD induced by small sample size and/or stratification , we computed r2 between all pairs of barcode SNPs on distinct chromosomes within each population . For SNPs on the same chromosome , we considered r2 values significant if they were above a defined percentile in the background levels of LD . This percentile was defined according to a Bonferroni correction at 100* ( 1–0 . 05/66 ) . We assessed sample uniqueness by comparing genotypes across all pairs of samples . We counted both pairs of distinct monomorphic genotypes ( e . g . , A/G ) and pairs of monomorphic and polymorphic genotypes ( e . g . , A/N ) as mismatches . To measure population diversity we calculated the ‘barcode π’ statistic , which is the average of number of pairwise differences at assayed SNPs between all members of a population divided by the number of assayed SNPs . In this calculation , we weighted monomorphic/polymorphic genotype pairs at half the value of a monomorphic/monomorphic mismatch . We assessed the variability of barcode π values with 10 , 000 iterations of nonparametric bootstrapping . Since we do not know the number of parasite types within a polygenomic sample , this calculation assumes a complexity of infection of two , thus this approach may inflate the barcode π values . To measure population divergence we calculated FST for all pairs of populations . We also generated 100 , 000 nonparametric bootstrap replicates of 42-SNP barcodes to obtain a 95% confidence interval . For each replicate we calculated the FST values for each pair for the four populations using the Weir-Hill unbiased estimator [34 , 35] . To determine whether each of the FST values represented statistically significant population divergence ( i . e . , statistical support for rejecting the null hypothesis of a interbreeding ( panmictic ) population ) we performed 100 , 000 permutations of the population assignments for samples within each pair of populations , then calculated FST . We then performed a Wilcoxon rank-sum test to compare the two distributions of FST values . We performed PCA with the program SmartPCA in the Eigensoft package [36] on the four populations ( Brazil , Sri Lanka , Ethiopia , and French Guiana ) together and in pairs . In addition , we performed PCA on the four populations using only the data from a reduced set of SNP assays ( either 14 or 28 ) , where we populated the barcode subsets with the assays with the highest AMAF values . We assessed the significance of PCA results using the percent of variance ( POV ) explained , which is calculated as the sum of the eigenvalues corresponding to the first and second principal components over the sum of all eigenvalues . The genotype data for the studied isolates has been recorded in S1 Table and will be hosted by PlasmoDB [12] ( http://plasmodb . org/ ) . For the P . vivax barcode , we chose SNPs in putatively neutral genomic loci , identified using all P . vivax sequence information made available to us through collaborators or community sequencing efforts . Candidate SNPs were restricted to those that are globally polymorphic , having two alleles present in two continental populations—South America and Asia . In these populations , we identified 16 , 288 SNPs from South America ( Brazil and Peru ) and 37 , 721 SNPs from Asia ( India and North Korea ) . Screening these for globally polymorphic SNPs yielded 2 , 818 candidates . We added an additional 599 SNPs that were polymorphic in Cambodia and in Madagascar P . vivax genome sequences . Screening these candidates for those at putatively neutral sites ( intergenic , intronic or 4-fold degenerate sites ) produced a set of 438 SNPs for HRM assay development . We winnowed the 438 candidates to a final 42-SNP barcode based on HRM primer design guidelines , species-specificity , and the ability to robustly distinguish the reference and alternate alleles in clinical samples . We designed 187 primers for HRM analysis that were specific to P . vivax . We screened these assays for accurate HRM genotyping using P . vivax controls . One hundred and fifty-seven of the 187 had a single amplicon of the expected size and accurately called the target SNP as either the reference or alternate allele among sequenced control DNAs . We tested the 157 assays for genotyping accuracy and robustness in a broad panel of clinical samples from Brazil ( 31 samples ) , Ethiopia ( 15 samples ) , Sri Lanka ( 19 samples ) , and French Guiana ( 22 samples ) . We identified 42 out of the 157 as high-performing assays ( S1 Table ) that could successfully distinguish both alleles by HRM with a 0 . 7–1 . 2°C Tm shift ( S2 Table ) . We measured the efficiency of the PCR reaction for all assays by performing a standard 10-fold dilution series of the DNA . Using this method , all 42 assays were robust , with a dynamic range in clinical samples of 300 ng/μl DNA with a detection limit of 0 . 3 ng/μl DNA in 16 out of 42 assays and 0 . 03 ng/μl DNA in 26 out of 42 assays . The overall efficiency range for all 42 assays was 90% to 100% ( S3 Table ) . We chose 3 ng/μl total DNA as an optimal assay concentration , allowing for a universal PCR method by eliminating the need to rescreen clinical samples with low or poor DNA quantity , which ultimately improves genotyping throughput . By using WGA , we were still able to use a limited quantity of DNA from our P . vivax samples . Here , we carried out all assay performance studies and the pilot screen using only 1 μl of each P . vivax sample that was amplified by WGA , as described in the methods section . We evaluated the sensitivity of all assays to detect polymorphic genotypes by examining the detection limit of both the alternate and reference allele in mixed genome samples in duplicate . All 42 assays were highly sensitive , with a dynamic range in mixed genome samples of P . vivax down to a detection limit of 10 copies/μl ( S4 Table ) . The HRM analysis clearly showed the sensitivity of the assays to distinguish each ratio of reference to alternate alleles ( 1:10 , 1:4 , 1:2 , 1:1 , 10:1 , 4:1 , and 2:1 ) in mixed genome samples ( Fig . 1 ) . We evaluated assay reproducibility of the 42 assays by screening them against three clinical samples and three sequenced controls . The assays were run in duplicate on two different genotyping platforms: the PCRmax Eco ( 48-well format ) and the Applied Biosystems ViiA 7 ( 384-well format ) Real-Time PCR Systems . All genotype calls ( 504 out of 504 SNP calls ) were concordant across duplicates and on both genotyping platforms . The variability of the Tm values for each assay in duplicate was ±0 . 038°C on the Eco , ±0 . 046°C on the ViiA 7 , and ±0 . 038°C between platforms ( S5 Table ) . We then compared manual genotyping to the automatic genotyping feature of the ViiA 7 . The ViiA 7 genotyping software automatically genotyped 98% ( 494 out of 504 SNP calls ) of the SNP calls accurately . We successfully genotyped the 10 missed calls by manual inspection . We evaluated the overall reproducibility of the 42 assays in our pilot screen by evaluating the variability of the Tm values across duplicates for all monogenomic calls from the 87 clinical samples and control samples . The variability of the Tm values across duplicates for monogenomic calls was ±0 . 070°C on the Eco ( S6 Table ) . The assays were highly sensitive where all 87 samples were successfully genotyped ( 3654 out of 3654 SNP calls ) in duplicate by all 42 assays ( S1 Fig ) . Using a series of known allelic mixtures , we found that WGA prior to PCR did not affect the genotyping accuracy of the HRM assays or resulting genotype . We selected assays representing each SNP type present in the barcode ( CT , CA , GT , and GA ) . We then evaluated mixtures of monogenomic clinical samples with varying ratios of reference and alternate alleles ( 1:1 , 1:2 , 1:4 , 2:1 and 4:1 ) . The HRM analysis showed that all assay SNP types detected the ratio of reference and alternate alleles with the same accuracy in duplicate in sample mixtures processed with and without WGA ( Fig . 2 ) . We further compared the 42 assays using 17 clinical samples from French Guiana with and without WGA , and found no significant differences . For all 42 assays and 17 samples , all genotype calls were concordant between sample treatments . The 42 SNPs span all 14 chromosomes of the P . vivax genome , and each SNP is highly informative across four distinct geographic populations: Brazil , Sri Lanka , Ethiopia , and French Guiana . The SNPs captured high degrees of diversity , with the AMAF value for each SNP > 0 . 1 ( Fig . 3 , and S7 Table ) . Moreover , the SNPs were independently informative ( S2 Fig ) . We did not expect SNPs to be in LD since the closest pair of SNPs is 21 , 237 bp apart ( SNPs 5 and 6 ) , which is beyond the map distance over which we typically see significant LD in Plasmodium populations [37 , 38] . Using the r2 statistic to measure LD ( which examines allele correlation from 0 to 1 ) , we found that all SNP pairs had r2 < 0 . 53 in each . None of the r2 values were significantly different from the background LD levels after multiple comparison corrections . The barcode was able to distinguish all samples except for two pairs of Brazil samples ( B13 and B21 and B19 and B20 ) and one trio of Brazil samples ( B1 , B4 , B8 ) ; these samples were identical at all genotypes , including polymorphic loci ( S1 Fig ) . In all populations other than Brazil , each sample pair was separated by at least five differences . Overall , the 42-SNP barcode captured high levels of population diversity , with the median bootstrapping values of barcode π at 0 . 37 in Brazil , 0 . 39 in Sri Lanka , 0 . 36 in Ethiopia , and 0 . 37 in French Guiana . The 42-SNP barcode also uncovered a high prevalence of polygenomic infections in the global sample set , with 45% in Brazil ( 14 of 31 ) , 74% in Sri Lanka ( 14 of 19 ) , 60% in Ethiopia ( 9 of 15 ) , and 41% in French Guiana ( 9 of 22 ) . While we chose our assays primarily to capture population diversity , seeking SNPs that were highly informative in all populations , we still examined whether the barcode could additionally detect population divergence . With PCA we showed that the 42-SNP barcode visually distinguished populations , separating the 87 samples by collection site with the exception of Brazil and French Guiana ( POV = 22 ) ( Fig . 4A ) . Shorter barcodes of 28 SNPs ( POV = 20 ) and 14 SNPs ( POV = 25 ) selected to maximize the capture of population diversity were unable to distinguish populations ( Fig . 4B-C ) . We used the median value obtained from bootstrapping of the FST statistic to quantitatively assess pairwise population divergence , and confirmed the separation visually identified by PCA . We found clear population divergence by both FST and PCA for Brazil and Sri Lanka ( FST = 0 . 18; POV = 26 ) , Ethiopia and Sri Lanka ( FST = 0 . 21; POV = 30 ) , French Guiana and Sri Lanka ( FST = 0 . 21; POV = 26 ) , Ethiopia and French Guiana ( FST = 0 . 27; POV = 30 ) , and Ethiopia and Brazil ( FST = 0 . 31; POV = 33 ) ( Fig . 5A ) . Bootstrapping analysis rejected the null hypothesis of a panmictic population in each case ( p < 10-5 ) , showing that all the FST values capture statistically significant population divergence . Although separation of the samples from the neighboring countries of Brazil and French Guiana was not visually resolved by the first two principal components in PCA ( POV = 21 ) , the FST value was still highly significant ( FST = 0 . 10 , p < 10-5 ) ( Fig . 5B ) . The P . vivax 42-SNP barcode provides an important baseline universal assay set to distinguish parasite infections and their geographic origins , and may provide insight into changes in parasite population dynamics and transmission . We chose HRM as our technology platform because it is a rapid genotyping technique that requires limited training and has minimal subjective data due to its visualization software . Thus , independent research groups should be able to employ this common set of genotyping assays , and their data and findings should be portable across studies , allowing the results from individual sites to be more accurately compared and contextualized . We have thus made our assays and protocols available to the P . vivax community for their implementation in disease elimination and research efforts ( http://broadinstitute/org/infect/malaria/pvivax/ ) . While we present all 42 assays here in our universal set , over time , as the assays are employed by more researchers , some may choose to use shorter regional barcodes or barcodes designed for specific purposes . The barcode may also evolve to include additional variants as more genomic information about P . vivax becomes available . In particular , drug resistance SNPs can be incorporated into barcode extensions as markers for these important phenotypes are identified in the future; this would allow simultaneous tracking of population characteristics and the emergence of drug resistance .	Plasmodium vivax malaria is a major global public health problem , with nearly 2 . 5 billion people at risk for infection and approximately 132–391 million clinical infections annually . It has a wide geographical range , with a high disease burden in Asia , Central and South America , the Middle East , Oceania , and East Africa . Advances in sequencing technology and sample processing have made it possible to characterize the genetic diversity of P . vivax populations . This genetic variation provides a means to identify parasites by unique genetic signatures , or “barcodes . ” We developed such a genetic barcode for P . vivax , composed of 42 robust and informative variants . Here we report its development and validation based on 87 clinical samples identified by microscopy to contain P . vivax from geographically diverse parasite populations from South America ( Brazil , French Guiana ) , Africa ( Ethiopia ) and Asia ( Sri Lanka ) . We show that the SNP barcode provides a genotyping tool that can be performed at low cost , providing a means to uniquely identify parasite infections and distinguish geographic origins , and that barcode data may offer new insights into P . vivax population structure and diversity .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[]	2015	Development of a Single Nucleotide Polymorphism Barcode to Genotype Plasmodium vivax Infections
A fundamental , but unanswered question in host-pathogen interactions is the timing , localization and population distribution of virulence gene expression during infection . Here , microarray and in situ single cell expression methods were used to study Vibrio cholerae growth and virulence gene expression during infection of the rabbit ligated ileal loop model of cholera . Genes encoding the toxin-coregulated pilus ( TCP ) and cholera toxin ( CT ) were powerfully expressed early in the infectious process in bacteria adjacent to epithelial surfaces . Increased growth was found to co-localize with virulence gene expression . Significant heterogeneity in the expression of tcpA , the repeating subunit of TCP , was observed late in the infectious process . The expression of tcpA , studied in single cells in a homogeneous medium , demonstrated unimodal induction of tcpA after addition of bicarbonate , a chemical inducer of virulence gene expression . Striking bifurcation of the population occurred during entry into stationary phase: one subpopulation continued to express tcpA , whereas the expression declined in the other subpopulation . ctxA , encoding the A subunit of CT , and toxT , encoding the proximal master regulator of virulence gene expression also exhibited the bifurcation phenotype . The bifurcation phenotype was found to be reversible , epigenetic and to persist after removal of bicarbonate , features consistent with bistable switches . The bistable switch requires the positive-feedback circuit controlling ToxT expression and formation of the CRP-cAMP complex during entry into stationary phase . Key features of this bistable switch also were demonstrated in vivo , where striking heterogeneity in tcpA expression was observed in luminal fluid in later stages of the infection . When this fluid was diluted into artificial seawater , bacterial aggregates continued to express tcpA for prolonged periods of time . The bistable control of virulence gene expression points to a mechanism that could generate a subpopulation of V . cholerae that continues to produce TCP and CT in the rice water stools of cholera patients . Clinical and pathological studies of diverse bacterial pathogens disclose a common theme: the infectious process evolves as a series of spatial and temporal patterns during migration of the pathogen through different tissues and cellular compartments of the host . At each stage different microenvironments are encountered; in response to these , adjustments of the microbial transcriptome and phenotype are thought to occur: virulence genes are induced or silenced and changes in replication rate take place . Spatial differences at the site of infection are magnified by differences in scale between bacteria ( microns ) and infected host tissues ( millimeters or centimeters ) leading to situations where bacteria in the same organ could experience dramatically different microenvironments . Spatial and temporal sources of microbial heterogeneity can be compounded by stochastic events that cause cell-to-cell transcriptional and phenotypic differences between genetically-identical individuals in the same microenvironment [1] , [2] . Together , these two sources of variation , one deterministic and the other probabilistic , pose significant experimental challenges that impede a deeper understanding of pathogenesis . To address these challenges here we describe results from a study that employed a combination of site-specific and single cell gene expression methods to study Vibrio cholerae infecting the small intestine . V . cholerae is the etiological agent of cholera , a purging diarrheal illness that occurs in rapidly spreading , seasonally-determined epidemics in south Asia . Ingestion of contaminated water or food containing V . cholerae leads to colonization of the small intestine where production of a powerful enterotoxin induces epithelial cells to secrete water and electrolytes into the bowel lumen [3] . A variety of methods have been used to study murine [4]–[6] and rabbit models of cholera [7]–[9]; collectively , these have identified virulence determinants and characterized early , intermediate and late events in the infectious process . One such study employed confocal microscopy and GFP-labeled bacilli to identify different stages in the infection of rabbit ileal loops [10] . Four hours after the inoculation of V . cholerae into a ligated ileal loop , bacteria migrated from the bowel lumen through the mucus gel to epithelial cell surfaces; eight hours post inoculation large numbers of V . cholerae collect on the surfaces of microvillae and fluid accumulation in the ileal loop lumen is evident; then , between 8 and 12 hours post inoculation , near synchronous detachment of bacteria from the mucosal surface occurs heralding the mucosal escape response , a process that is associated with RpoS-dependent down regulation of virulence gene expression . In humans and in the murine and rabbit models of cholera , the infectious process is accompanied by prodigious intra-intestinal replication of the organism leading to a vast expansion of its biomass and to the possibility that the progeny of the infection , shed in feces , will spread to other susceptible hosts or re-enter natural aquatic reservoirs where the organism resides between outbreaks of human disease [11] , [12] . Consistent with the survival of fecal V . cholerae in natural water sources was the identification of sets of genes induced late in the infection of a murine model of cholera that appear to be involved with adaptation of the organism to aquatic environments [13] . Human volunteer studies of V . cholerae mutants show that the clinical hallmarks of cholera require the combined production of two virulence determinants [14]: TCP , which is specified by the tcp operon and promotes small bowel colonization [15] , and CT [16] , which is encoded by the ctx operon and elicits a secretory response by small bowel epithelial cells . Expression of the tcp and ctx operons is coordinately regulated by the ToxR regulon [15] , [17] , a multi-component hierarchy of transcription factors that integrates physical and chemical signals , cell density and the physiological state of the organism [18] . Studies of cholera animal models illustrate the difficulty of discerning where , when and by which subsets of the bacterial population key virulence determinants are produced . Earlier work based on a recombination based reporter system showed induction of tcpA and ctxAB by most bacteria during the early stages of infection of a murine model [5] . However , microarray expression profiles of V . cholerae in the fluid that accumulates in ligated loops of the rabbit ileum failed to identify significant expression of tcpA and ctxAB even though CT is required for fluid secretion in this model [19] . This unexpected observation led to the hypothesis that genes encoding CT may only have been expressed by a minority of bacteria or only for a short period of time [19] . A similar paradox can be found in the study of naturally infected cholera patients . Though TCP genes were expressed in the vomitus of cholera patients early in the infectious process [20] , a series of three studies showed low , variable or absent expression of these genes by organisms in freshly passed cholera stools [20] , [21] , [22] even though TCP production is required for the clinical manifestations of the disease [14] . These discrepancies might in part reflect technical variations in the gene expression methods employed or be due to differences in the microenvironments of the colon and rectum ( from which the stools were collected ) compared to the small bowel ( the principal locus of the infection ) . However , these results also are consistent with the idea that V . cholerae virulence genes are either expressed at a low level by most fecal V . cholerae or at a very high level by a small sub-population of the bacteria [19] , [21] . Studies of fecal V . cholerae also showed that freshly-passed cholera stools from naturally infected humans harbor bacteria that are hyper-infectious when tested in an infant mouse model [22] . In vitro propagation of the same strain abolishes the hyper-infectious phenotype indicating that the hyper-infectious state is transient . The hyper-infectious phenotype has now been corroborated by others [23] , [24] , [25] , [26] , [27] . Passage of V . cholerae through the infant mouse intestine also induces a hyper-infectious state that was shown to be TCP-dependent [28] . The presence of a hyper-infectious subpopulation of V . cholerae in the stools of cholera patients could promote the rapid dissemination of V . cholerae within households and communities during outbreaks . In the study reported below we have used the rabbit ileal loop model of cholera and site specific expression profiling to identify virulence genes that are differentially expressed on epithelial surfaces . Further resolution was obtained through the use of single cell expression methods and confocal microscopy to precisely localize where and when in the intestine the gene encoding the principal repeating subunit of the TCP filament is expressed . Expression analysis of a ribosomal promoter , imaged by confocal microscopy at the single cell level of resolution , showed that TCP gene expression and growth are co-localized in the intestine . The capacity to capture the expression behavior of specific virulence genes by individual bacteria led to the discovery that genes encoding TCP and CT are controlled by an epigenetic bistable switch that bifurcates the population into TCP/CT-expressing and TCP/CT-non-expressing subpopulations . The study of mutants in the ToxR regulatory cascade showed that this switch requires autocatalytic regulation of the gene encoding the ToxT transcription factor and the CRP-cAMP complex . Together these observations provide a mechanistic explanation for the presence of a subpopulation of TCP-expressing , hyper-infectious bacteria in the stools of cholera patients . Confocal microscopy studies of GFP-labeled V . cholerae O1 El Tor ( strain A1552 ) in the rabbit ileal loop model of cholera 4 , 8 and 12 hours after inoculation have previously shown that V . cholerae resides in at least three anatomically distinct sites in the ileal loop at the same time point: the epithelial surface; the mucus gel overlying the epithelial surface; and , in fluid that collects in the lumen of the loop [10] . To determine if virulence gene expression differs as a function of anatomical location , we obtained microarray expression data from V . cholerae collected from two anatomically distinct sites of the same ileal loop at 4 , 8 and 12 hours post inoculation . Because the epithelial surface and overlying mucus gel could not be separated , these contiguous sites were obtained as a single fraction . V . cholerae in luminal fluid were collected as a second fraction by isolating the liquid contents of incised loops . Dramatic up-regulation of TCP biosynthetic genes was evident in the mucus gel/epithelial surface fraction of bacteria ( Fig . 1 ) . Expression of tcpA ( VC0828 ) , the gene encoding the principal repeating subunit of the TCP filament , was 30 . 0- , 17 . 0- and 11 . 5-fold up-regulated in this fraction 4 , 8 and 12 hours post-inoculation , respectively , compared to its expression in mid log phase LB cultures . Also strongly induced in the mucus gel/epithelial cell fraction were the eight downstream genes which together with tcpA compose the operon VC0828 –VC0837 within Vibrio Pathogenicity Island I ( VPI-1 ) ( Fig . 1 ) . One of these , tcpF ( VC0837 ) , which encodes a soluble colonization factor [29] was up-regulated 19 . 9- , 15 . 2- , and 6 . 7-fold in the mucus gel/epithelial cell fraction at 4 , 8 and 12 hours post-inoculation , respectively . By contrast the expression of these genes was markedly lower in the luminal fluid: either no greater than their expression in the mid log phase reference; or , in the case of tcpA , 4 . 9- to 3 . 2-fold lower than its expression in the mucus gel/epithelial cell fraction at the same time point . Even more striking was the localized expression of ctxA and ctxB which encode the A and B subunits of CT: both genes were strongly expressed in the mucus gel/epithelial cell fraction . By contrast , they were not significantly up-regulated in luminal fluid compared to the mid log phase reference ( Fig . 1 ) . Taken together , these results show temporal and anatomical localization of V . cholerae virulence gene expression: the expression of these genes is strongest on or close to epithelial cell surfaces early in the infectious process . ToxT , a member of the AraC/XylS family of transcriptional regulators [30] , directly and positively regulates the expression of tcpA-F and ctxAB [31] . The gene encoding ToxT , which resides in VPI-1 adjacent to the tcpA-F operon , was expressed 2 . 8- , 2 . 5- and 1 . 4-fold greater in the mucus gel/epithelial surface fraction 4 , 8 and 12 hours post inoculation , respectively , compared to its expression in the LB broth mid log phase reference ( Fig . 1 ) . By contrast , toxT was not up-regulated by V . cholerae collected from the ileal loop fluid . Thus , the expression of toxT , which encodes the proximal regulator of tcpA-F and ctxAB , parallels the expression of the genes it controls . These findings , as demonstrated by microarray expression analysis , were corroborated by quantitative RT-PCR ( Table S2 ) , where an even stronger induction of tcpA , ctxA , toxT and tcpP in the mucus fractions was observed , especially during early stages of the infection . The localization studies described above were not able to discriminate between the expression of virulence genes by bacteria directly in contact with the epithelial surfaces and their expression by bacteria embedded in the overlying mucus gel . Yet these two microenvironments , though only microns apart , likely represent distinct biochemical milieus . To determine if mucus-embedded and cell-associated bacteria differ with respect to their expression of virulence determinants , tcpA expression by bacteria in ligated ileal loops was monitored by confocal microscopy at the single cell level of resolution . The tcpA promoter [32] was cloned , fused to the coding sequence of a destabilized variant of the green fluorescent protein ( GFP ) and inserted as a single copy in the neutral intergenic region between VC0487 and VC0488 on the large chromosome ( Fig . 2A ) . The reporter strain thus harbored the native tcpA gene in VPI-1 and the tcpA-gfp ( ASV ) reporter at a separate site on the same chromosome . The GFP protein variant encoded by this reporter was modified by the addition of 11 amino acids to the carboxy terminus which targets it for destruction by the ClpXP protease system [33] . As a result , this destabilized derivative of GFP , denoted GFP-ASV , has a markedly reduced half-life ( 40 minutes in Escherichia coli ) , yielding a reporter capable of monitoring both increased and decreased activity of the promoter to which it is fused [33] , [34] , [35] . Thus , this reporter differs from those which encode the stable variant of GFP or which use the recombinase-based reporter of transcription designated RIVET [5] . These systems report activation of the promoter with which they are associated , but cannot report a subsequent decrease in promoter activity . The V . cholerae tcpA-gfp ( ASV ) reporter strain was phenotypically indistinguishable from the wild type parent with respect to growth and its capacity to colonize the ileal loop and elicit a secretory response . The strain emitted only weak background fluorescence during growth in LB liquid medium . However , significant induction of fluorescence from the tcpA-gfp ( ASV ) reporter was observed in AKI medium , which induces the expression of the ctxAB and tcpA-F operons [36] , [37] . In contrast , fluorescence was not detected when toxR , the global regulator of ctxAB and tcpA-F expression , was deleted from the tcpA-gfp ( ASV ) reporter strain and the ΔtoxR reporter grown in AKI medium ( data not shown ) . Taken together , these results demonstrate that the fluorescence from the tcpA-gfp ( ASV ) construct parallels the expression behavior of tcpA . V . cholerae tcpA-gfp ( ASV ) was inoculated into rabbit ligated ileal loops and expression of the tcpA-gfp ( ASV ) fusion monitored by confocal microscopy as a function of time and site . In addition , all V . cholerae ( i . e . , GFP-positive and GFP-negative bacteria ) were visualized in the same sample using an O1 antigen-specific antibody so that even bacteria not producing GFP could be identified and distinguished from GFP-producing bacteria . Since the tissue samples were washed before microscopy to remove bacteria present in the luminal fluid , only the bacteria attached to epithelial cell surfaces or residing in the mucus gel coating these surfaces were visualized . Four hours after inoculation of rabbit ligated ileal loops , confocal microscopy showed tcpA-gfp ( ASV ) expression especially by bacteria which had reached the epithelial surface ( Fig . 2B–D ) . By contrast , tcpA-gfp ( ASV ) expression was not evident at this time point for most non epithelial surface-associated bacteria that were located in the overlying mucus gel . This distinction was quantified by correlating GFP fluorescence intensity with distance from the nearest epithelial surface . Quantitative image analysis of the ratio between green fluorescence from GFP and red fluorescence from the V . cholerae specific antibody showed significantly stronger expression of tcpA-gfp ( ASV ) by bacteria ≤5 µm from an epithelial cell surface ( Fig . 3D ) . The average expression level of tcpA-gfp ( ASV ) declined rapidly for bacteria at greater distances from the epithelium: bacteria further than 5–10 µm from an epithelial surface showed an average fluorescence intensity five-fold lower than bacteria 0–5 µm from an epithelial surface ( Fig . 3D ) . Using the same model system , we have previously shown that expression of GFP from a constitutive promoter was homogeneous throughout the intestine and that the fluorescence did not increase in close proximity to the epithelial surfaces [10] . Consequently , the differences in fluorescence from the tcpA-reporter strain shown in Fig . 2 are very likely caused by differential gene expression . Increased numbers of bacteria were noted eight hours post inoculation . Many were closely associated with epithelial cell surfaces and strongly expressed the tcpA-gfp ( ASV ) reporter ( Fig . 2 E–G ) . Taken together , single cell tcpA-gfp ( ASV ) expression data from the 4 and 8 hour post-inoculation time points suggest that bacteria encounter tcpA-inducing signals as they approach or contact the epithelial cell surface . Whether these signals emanate from the epithelial cell as a kind of chemical gradient or require physical contact with the cell surface was not investigated . Microarray experiments showed the highest expression of tcpA 4 hours post induction , whereas the tcpA-gfp ( ASV ) reporter showed the strongest induction 8 hours post infection . Thus these two expression methods yielded somewhat different temporal profiles for tcpA expression . However it is not possible to directly compare results from these two expression methods because microarray experiments estimate the average gene expression magnitude of all bacteria in the population , whereas the tcpA-gfp ( ASV ) reporter captures gene expression magnitude at the single cell level . An exception to the increasing gradient of tcpA expression from the mucus gel towards the epithelial surface was observed in some locations of the mucus gel where V . cholerae , located at significant distances from the nearest epithelial cell surface , were found to express tcpA-gfp ( ASV ) . Systematic examination of these non cell-associated , tcpA-gfp ( ASV ) -expressing bacteria showed that they were mainly found in aggregates ( denoted by arrows in Fig . 2E ) compared to non-aggregated bacteria that do not express tcpA-gfp ( ASV ) . This aggregation-associated , tcpA-inducing phenomenon was evident at the 4 , 8 and 12 hour time points and is consistent with the previously-reported TCP-mediated auto-aggregation phenotype [38] , [39] . Twelve hours post-inoculation , confocal images showed that most cell-associated bacteria had detached from the epithelial surface and re-entered the mucus gel as part of the previously-described mucosal escape response [10] . Expression of tcpA-gfp ( ASV ) was markedly reduced at this time point compared to eight hours post-inoculation ( Fig . 2 H–J ) , corroborating microarray expression data which showed that dispersal of bacteria from the villous surface coincides with decreased virulence gene expression ( Fig . 1 and [10] ) . Single cell gene expression analysis of tcpA-gfp ( ASV ) confirmed that RpoS is required for decreased tcpA expression during the mucosal escape response ( Fig . S1 ) as postulated by Nielsen et al . [10] . Passage of V . cholerae through the human intestine vastly amplifies its biomass . It is , however , unknown how growth rate and virulence are orchestrated as a function of time and anatomical site . To address this question , we used the rabbit ligated ileal loop model of cholera , viable plate counts of V . cholerae in ileal loop fluid and expression of a growth rate-dependent promoter to estimate growth by V . cholerae in the mucus gel and on the epithelial surface . Beginning with an inoculum of 106 CFUs injected into the ileal loop , viable plate counts of V . cholerae in fluid from the ileal loop lumen increased rapidly for the first four hours ( Fig . 3B ) . Increase in the number of luminal bacteria slowed between the fourth and eighth hour and by hour 12 the apparent growth rate appeared to have further declined ( Fig . 3B ) . However these values likely do not accurately represent in situ growth rates: ( 1 ) they register changes in bacterial biomass in only one compartment of the ileal loop ( the lumen ) ; ( 2 ) , they may reflect , but do not directly monitor growth rate since viable plate counts are a function of replication rate , death rate , plating efficiency and other factors; and ( 3 ) , they do not have the capacity to co-localize virulence gene expression and growth at the micron scale required to correlate both measurements with distance from the epithelial cell surface . These limitations were addressed by performing single cell expression studies and confocal microscopy using gfp ( ASV ) fused to a growth rate-regulated promoter . The ribosome synthesis rate and thus the concentration of ribosomes in a cell is directly correlated with growth rate for a wide range of bacteria [40] , [41] , [42] , [43] . The growth rate-dependent P1 promoter of the E . coli rrnB ribosomal operon was coupled to gfp ( ASV ) and the resulting rrnBP1-gfp ( ASV ) fusion inserted as a single copy between VC0487 and VC0488 on the large chromosome of the same wild type V . cholerae strain that had been used to study in situ expression of tcpA . We selected this promoter fusion because the fluorescence intensity of a Pseudomonas putida strain harboring an E . coli rrnP1-gfp ( ASV ) reporter accurately reflected both increased and decreased growth and thus could be used in conjunction with confocal microscopy to study bacterial growth rates in complex environments [34] . Growth of the V . cholerae rrnBP1-gfp ( ASV ) reporter strain in LB medium showed strong fluorescence during exponential growth as quantified by flow cytometry ( Fig . 3A ) . During transition into stationary phase ( OD600 = 0 . 9 ) , a reduction in growth rate was correlated with a reduction in fluorescence intensity . Further progression into stationary phase was accompanied by a rapid decline in fluorescence to approximately 10% of the level observed during exponential growth . Assuming that expression of gfp ( ASV ) from the rrnP1 promoter ceased after entry into stationary phase , then the rate of the decline in GFP fluorescence corresponds to a maximal GFP ( ASV ) half life of 40 minutes in V . cholerae . V . cholerae rrnBP1-gfp ( ASV ) was inoculated into rabbit ligated ileal loops and samples obtained four , eight and twelve hours post inoculation . The fluorescence intensity of individual bacteria was then monitored by confocal microscopy as a function of time and anatomical site . Four hours post inoculation , bacteria juxtaposed to the epithelial cell surface were found to express the highest levels of GFP , indicating that these bacteria were replicating at a higher rate or were more metabolically active than bacteria residing in the mucus ( Fig . 4A–C ) . The ratio between green fluorescence intensity ( from GFP ) and red fluorescence intensity ( from the V . cholerae O1-specific antibody ) was quantified in multiple confocal planes from different images of epithelial tissue four hours post inoculation . Bacteria within 5 µm of the nearest epithelial cell surface produced nearly twice the amount of rRNA-associated fluorescence when compared to bacteria residing at distances further away from the epithelial surface ( Fig . 3C ) . In contrast to the 4 hour time point , fluorescence from the ribosomal promoter fusion was markedly reduced 8 and 12 hours post inoculation ( Fig . 4D–I ) . While most of the rapidly growing bacteria were concentrated on or near epithelial cells early in the infectious process , islands of rapid growth were evident in mucus at sites >10 µm from any cell surface ( indicated by arrows in Fig . 4A ) . Most of these islands were found to be associated with extruded epithelial cells ( Fig . S2 ) . Comparison of single cell gene expression images and quantification of induction levels for the tcpA-gfp ( ASV ) and rrnBP1-gfp ( ASV ) promoters suggest that the replication of V . cholerae and the production of TCP are co-localized: both promoters are most active on or near epithelial cell surfaces ( Fig . 3C–D ) . Exceptions to this relationship are the expression of tcpA in non-epithelial surface associated bacterial aggregates and the expression of rrnBP1 within extruded epithelial cells . Examination of confocal images of V . cholerae tcpA-gfp ( ASV ) in rabbit ileal loops showed apparent variation between adjacent bacteria in the expression of tcpA twelve hours post-inoculation ( Fig . 5A–B ) . To explore this observation under a more homogeneous condition of growth , fluorescence microscopy was used to qualitatively characterize tcpA-gfp ( ASV ) expression by V . cholerae in a liquid culture where all bacteria are exposed to the same condition . For this purpose we first used AKI medium , growth in which induces tcpA and ctxAB expression after the bacteria are cultivated four hours in a stationary test tube followed by one hour of growth in a shaken ( and thus aerated ) flask [36] . Fluorescence microscopy of the culture was conducted at the five hour time point . A small number of strongly GFP-positive bacteria were noted within clumps of weakly fluorescent bacteria ( Fig . 5C ) . However , because four hours of growth in an unstirred test tube preceded the microscopic study of cells at the five hour time point , gradients of oxygen and other metabolites may have formed and given rise to heterogeneity in tcpA expression . To address this issue , we studied the expression of tcpA-gfp ( ASV ) under a homogeneous condition of growth that did not allow formation of chemical gradients . Expression from the tcpA promoter was induced by adding bicarbonate to stirred HEPES-buffered LB broth containing early exponential phase cultures ( OD600 = 0 . 2 ) , a variation of previously described methods that used bicarbonate or carbon dioxide to induce the production of CT through the activation of the ToxT transcription factor [36] , [37] , [44] . The distribution of fluorescence intensity was monitored qualitatively by confocal microscopy and quantitatively by flow cytometry . Induction of tcpA-gfp ( ASV ) expression 30 minutes after addition of bicarbonate to a shaken exponential culture was shown by flow cytometry to be an almost linear function of the bicarbonate concentration ( Fig . 6A ) . The strongest induction of tcpA was observed after the addition of 100 mM bicarbonate , a physiologically relevant value as judged by the 44 mM concentration previously measured in the human ileum [45] . The 100 mM concentration was therefore used in subsequent experiments since it did not cause significant changes in the pH of the medium or alter the growth rate of the bacteria ( Fig . 6B ) . Three hours after the addition of bicarbonate to a stirred exponentially growing culture , at which time the culture had transitioned into early stationary phase , fluorescence microscopy showed a significant bifurcation of GFP-fluorescence: approximately 50% of the population fluoresced intensely whereas the rest of the population showed only background levels of fluorescence ( Fig . 5D ) . This finding prompted us to use flow cytometry to quantify the ratio between tcpA-gfp ( ASV ) -expressing and non-expressing bacteria as a function of time after the addition of bicarbonate to an exponentially-growing culture ( Fig . 6C ) . Expression of tcpA-gfp ( ASV ) was significantly induced in all cells 30 minutes after the addition of bicarbonate to an early exponential phase culture ( OD600 ∼0 . 2 ) . Growth of the culture to an OD600 of approximately 0 . 9 , ninety minutes after the addition of bicarbonate , was associated with bifurcation of tcpA-gfp ( ASV ) expression . This time point corresponds to transition of the culture into stationary phase ( Fig . 6B ) . Serial flow cytometry measurements of the same culture showed progressive bifurcation over time ( Fig . 6C ) as the culture progressed into early stationary phase . Examination of the flow cytometry data showed that the bifurcation phenotype came from continued high-level expression of tcpA-gfp ( ASV ) by ∼50% of the population and declining tcpA-gfp ( ASV ) expression in the remaining 50% . Several hours after entry into stationary phase , all cells eventually down-regulated tcpA-gfp ( ASV ) expression ( data not shown ) . The concentration of bicarbonate not only controlled the initial induction of tcpA-gfp ( ASV ) during exponential growth ( Fig . 6A ) , but it also affected the distribution of the bifurcation phenotype during entry into stationary phase . Increasing concentrations of bicarbonate resulted in a greater fraction of induced cells in stationary phase ( Fig . 6D ) . To test if the bifurcation phenotype was limited to the V . cholerae strain A1552 , an identical tcpA-GFP ( ASV ) reporter strain was created in V . cholerae N16961 . This strain also exhibited significant heterogeneity in the expression of the tcpA-gfp ( ASV ) after addition of bicarbonate and progression into stationary phase ( data not shown ) . To further investigate the segregation of tcpA-gfp ( ASV ) expression into two populations during entry into stationary phase , V . cholerae tcpA-gfp ( ASV ) was induced with bicarbonate at OD600 = 0 . 2 and grown to early stationary phase until the bifurcation phenotype was observed . Then , the bacteria were fixed with paraformaldehyde and sorted by fluorescence intensity using a fluorescence activated cell sorter ( FACS ) . Cells that continued to produce GFP ( ASV ) from the tcpA promoter and cells that were GFP ( ASV ) -negative were collected as two separate populations . RT-PCR was performed to measure the abundance of the tcpA transcript in each of the two populations . The RT-PCR assay employed primers and probes corresponding to regions of the native tcpA gene in VPI-1; these regions were not present in the tcpA-gfp ( ASV ) construct . In this way , the expression of wild type tcpA could be monitored ( by RT-PCR ) in parallel with the expression of tcpA-gfp ( ASV ) ( by flow cytometry ) . To determine if the expression of the genes encoding CT , which is co-regulated with tcpA , also segregate into the same two populations , ctxA mRNA abundance was also monitored . In addition , the RT-PCR multiplex assay employed primers and probes corresponding to three genes ( cheA1-3 ) that are not regulated by ToxT and are located at three different sites on the genome , and six components of the ToxR regulatory network that governs the expression of the tcp and ctx operons [18] . These include: ( 1 ) ToxT , which binds and activates the tcpA and ctxAB promoters [46]; ( 2 ) AphA , AphB , ToxR and TcpP , which function at higher levels in the regulatory cascade and positively regulate the expression of tcpA-F and ctxAB [47] , [48] , [49]; and ( 3 ) , HapR , which negatively controls tcpA and ctxAB expression as a function of population density [50] . When transcripts corresponding to these genes were measured , no significant differences were noted in the expression of cheA1-3 , hapR , aphA , aphB or toxR in the two sorted populations ( Fig . 6E ) . The gene encoding TcpP showed less than two-fold greater transcript abundance in the GFP-positive population when compared to GFP-negative cells . By contrast , the toxT transcript was six-fold more abundant in the GFP-positive cells . The accumulation of toxT mRNA in cells expressing the tcpA-gfp ( ASV ) reporter was associated with an 80-fold greater abundance of the tcpA transcript and a 20-fold greater abundance of the ctxA transcript in the GFP-positive population ( Fig . 6E ) . Taken together , these data provide conclusive evidence that fluorescence from the tcpA-gfp ( ASV ) reporter can be used as a valid measure of tcpA expression , thus confirming that tcpA expression bifurcates into two populations . These data show that ctxA , long recognized to be co-regulated with tcpA [51] also exhibits the bifurcated phenotype . Thus , expression of the genes coding for the virulence determinants required for V . cholerae colonization ( TCP ) and virulence ( CT ) in humans exhibits the bifurcation phenotype . Finally , these data demonstrate bifurcation of toxT expression; thus toxT expression , like the ctxA and tcpA promoters it binds and activates , also segregates into two populations after induction with bicarbonate and progression into stationary phase . The difference in the expression of toxT between the two populations was lower than the difference seen for tcpA , which indicate that smaller changes in the expression of the transcriptional regulator may affect expression of tcpA and ctxAB significantly . To test whether the observed bifurcation in tcpA expression was caused by the presence in the culture of two pre-existing expression variants , only one of which sustains tcpA-gfp ( ASV ) expression during entry into stationary phase , a culture was induced with bicarbonate during exponential growth and allowed to undergo bifurcation . The two subpopulations [sustained or transient expression of tcpA-gfp ( ASV ) ] were then separated by FACS and inoculated onto separate agar plates . Single colonies grown from each of the two sorted populations were picked , grown to early exponential phase in separate shaken flasks and then treated with bicarbonate to induce tcpA-gfp ( ASV ) expression . Bacteria from each of the sorted populations were found to exhibit the same bifurcation phenotype as the non-sorted progenitor ( Fig . S3 ) . From these results we conclude the following . ( 1 ) The bifurcation phenotype is not the consequence of two pre-existing populations; rather , all members of the population appear able to exhibit the bifurcation phenotype . ( 2 ) Induction of the bifurcation phenotype does not generate variants durably assigned to one or the other of two populations; the bifurcation phenotype is reversible . To determine if the continued presence of bicarbonate is required to maintain the tcpA-gfp ( ASV ) bifurcation phenotype , bicarbonate was removed once the bicarbonate-treated culture had reached early stationary phase by washing and resuspending the bacteria in filter-sterilized conditioned media from cultures grown in parallel to the same OD without bicarbonate . Then , flow cytometry was used to compare the fraction of GFP-positive cells in the bicarbonate-depleted culture with the fraction in a bicarbonate-containing culture as a function of time . Fig . 6F shows that the proportion of GFP-positive cells in the bicarbonate-depleted and bicarbonate-containing cultures was equivalent for up to 150 minutes after removal of bicarbonate . Thus , while 100 mM bicarbonate is required to fully elicit the bifurcation phenotype , once established it can be sustained even after bicarbonate is removed . This cannot be explained by persistence of the GFP ( ASV ) protein in non-replicating cells since , if no more GFP ( ASV ) were produced , the 40 minute half-life of this reporter would have caused a much more rapid decline in the fraction of GFP-positive cells . Nor can it be explained by small numbers of GFP-positive cells that might persist because of low residual concentrations of bicarbonate since fewer than 5% of cells exhibit the bifurcation phenotype in cultures containing ≤10 mM bicarbonate ( Fig . 6D ) . Instead , this result is more likely explained by intracellular factors that persist in bicarbonate-depleted cultures and that are responsible for sustaining tcpA expression in the fraction of cells that continue to be GFP-positive after progression into stationary phase . Although the mechanism by which bicarbonate enhances the activity of ToxT is not yet known [44] , ToxT is active even in the absence of bicarbonate . It is therefore possible that the concentration of ToxT in the strongly induced population of bacteria is high enough to maintain the positive feedback induction of tcpA and toxT thus causing this sub-population to continue to express tcpA even after transfer of the culture to a bicarbonate-free medium . If so , then the duration of sustained tcpA expression would depend on how long cellular concentrations of ToxT remain above a critical threshold concentration . This kind of biochemical memory is characteristic of systems that exhibit hysteresis: the capacity to sustain an induced phenotype for a period of time after the responsible inducer has been removed or its concentration reduced below the level required to elicit the phenotype [52] . Many of the features of the tcpA bifurcation phenotype described above are consistent with a variety of genetic , DNA modifying and epigenetic mechanisms by which a clonal population can generate , at high frequency and in a homogeneous environment , two or more subpopulations [1] , [2] . These include genetic mechanisms giving rise to reversible switching between two states ( phase variation ) , including site specific recombination , gene conversion and slipped-strand mispairing [1] . Also compatible with some aspects of the bifurcation phenotype is the reversible methylation of DNA at sites affecting gene expression [53] . In contrast to these mechanisms , bistability is an epigenetic process that does not entail rearrangement or chemical modification of DNA . Bistability provides a compelling explanation for the tcpA bifurcation phenotype because it typically results in two distinctive states , is reversible and demonstrates hysteresis [54] , [55] . In addition to these properties bistable switches are typically controlled by auto-regulated , positive-feedback circuits that govern the expression of a master regulator and the genes it controls . The regulation of tcpA by ToxT is such a system: the tcpA promoter reads through to toxT thereby creating a positive feedback induction of toxT expression [32] , [56] that not only drives tcpA expression , but also ctxAB expression [56] . This autocatalytic circuit is depicted in Fig . 2A . Like other members of the AraC family of transcriptional regulators , ToxT has an N-terminal dimerization domain that is required for transcriptional activation of tcpA through binding of the two toxbox domains upstream of the tcpA promoter [46] , [57] . Thus , in addition to the positive autoregulation of toxT described above , the dimerization of ToxT also may be important in the generation of bistability since it could render activation of the tcpA-promoter hypersensitive to the concentration of ToxT . This is supported by experiments using virstatin which blocks ToxT dimerization [58] , [59] . Under in vitro inducing conditions , virstatin reduces the expression of tcpA to a few percent of normal levels , thus reinforcing other findings that ToxT is essential for activation of the tcpA promoter and must dimerize to exert its effect . The ToxT autocatalytic regulatory circuit and the ∼6-fold greater abundance of toxT transcripts in GFP ( ASV ) -positive compared to GFP ( ASV ) -negative sorted cells ( Fig . 6E ) , led us to test if the bifurcation phenotype depends on the positive autoinduction of toxT expression through the tcpA promoter ( Fig . 2A ) . We modified the genetic background of the tcpA-gfp ( ASV ) reporter strain by deleting the indigenous tcpA-promoter in VPI-1 thus interrupting the positive feedback loop . The tcpA-gfp ( ASV ) reporter , which is located at an ectopic site on the same chromosome , was left intact ( Fig . 2A ) . The tcpA promoter deletion version of the tcpA-gfp ( ASV ) reporter strain was then monitored for fluorescence intensity during induction with bicarbonate . Since the tcpA-gfp ( ASV ) reporter is inserted at a different locus ( Fig . 2A ) , it continues to report the effect of ToxT on the ectopic tcpA promoter , but without the effect of the autocatalytic circuit . As illustrated in Fig . 7B , initial induction of the tcpA-gfp ( ASV ) reporter was still observed in the tcpA-promoter deletion mutant 30 minutes after addition of bicarbonate to a mid exponential culture . However , in contrast to the wild type reporter strain ( Fig . 7A ) , all cells of the tcpA promoter mutant showed unimodal decreased expression of the tcpA-gfp ( ASV ) reporter during entry into stationary phase three hours post induction ( Fig . 7B ) . Deletion of the tcpA-promoter therefore completely prevented bifurcation of tcpA-gfp ( ASV ) expression . Thus , in the absence of the indigenous tcpA promoter , ToxT is still capable of responding to bicarbonate induction through its own promoter during exponential phase growth and to induce expression of tcpA-gfp ( ASV ) at an ectopic site , but the tcpA promoter mutant has lost the capacity to sustain expression of tcpA in a fraction of the cells during entry into stationary phase . Therefore , bifurcation of the tcpA-gfp ( ASV ) -expressing phenotype appears to depend on positive feedback induction of toxT through the tcpA promoter . Examination of the flow cytometry data in Fig . 7A and B shows that the average level of tcpA-gfp ( ASV ) expression in the bicarbonate-induced tcpA-promoter deletion mutant is 20% lower when compared to the average level of expression of the tcpA-gfp ( ASV ) reporter in the wild type background . This result suggests that autocatalytic control of toxT expression may be required to increase ToxT concentrations above a critical threshold necessary to sustain tcpA-gfp ( ASV ) expression by a fraction of bicarbonate-induced bacteria during entry into stationary phase . The initial expression of tcpA in response to bicarbonate during exponential growth results in a unimodal population of induced bacteria; bifurcation of the induced population into the two tcpA-expressing populations , depicted in Fig . 6C and 7A , was only observed during entry into stationary phase . This observation indicates that the tcpA bistable phenotype not only comes from positive auto-regulation of ToxT production , but also from other factors that are able to repress tcpA expression in a fraction of induced cells during entry into stationary phase . We reasoned that one such factor might be cAMP and the catabolite regulatory protein CRP with which it interacts . The CRP-cAMP complex is part of a global regulatory network that controls gene expression in response to the availability of carbon and energy sources in the environment . The effects of cAMP on gene expression are caused by an allosteric modification of CRP that occurs when cAMP binds CRP and the CRP-cAMP complex interacts with upstream promoter motifs . Depending on the position of the CRP-cAMP motif relative to other sites on the promoter , transcription of downstream genes is either increased or decreased ( for a review , see [60] ) . CRP-cAMP has been shown to decrease the expression of ctxAB and tcpA [61] . This effect has been attributed to proven or hypothesized affects at three separate sites in the regulatory cascade that controls ctx and tcp expression . First , CRP-cAMP increases the expression of HapR , a repressor of the regulatory cascade [62]–[63] . Second , CRP-cAMP directly competes with the positive regulators AphA and AphB on the tcpPH promoter [64]; this acts to reduce the expression of TcpP and TcpH which , together with ToxR and ToxS , activate toxT gene expression ( Fig . 2A ) [49] , [65] . Third , a putative CRP-cAMP binding site has been identified at the −35 domain ( −50 to −29 ) of the tcpA promoter [66] , [67] that overlaps the ToxT binding site ( −59 to −41 ) [32] . Therefore , occupation of this site by CRP-cAMP could potentially compete with dimeric ToxT and consequently with the positive autoregulation of toxT . Thus at each of the three sites , CRP-cAMP would decrease the expression of toxT and the ToxT-dependent genes , tcpA-F and ctxAB . To test the hypothesis that CRP-cAMP is required for the tcpA bistable phenotype , we constructed a crp deletion mutant in the tcpA-gfp ( ASV ) reporter strain . Results from flow cytometry studies of the tcpA-gfp ( ASV ) Δcrp reporter before the addition of bicarbonate , and then 30 minutes and 3 hours after addition of bicarbonate to an exponential phase culture are depicted in Fig . 7C . Expression of tcpA-gfp ( ASV ) by the wild type parent and the crp mutant was similar 30 minutes after addition of bicarbonate to log phase cultures ( Fig . 7A and C ) : both showed an unimodel induced population . By contrast to tcpA-gfp ( ASV ) expression in the wild type parent ( Fig . 7A ) , all cells of the crp mutant remained strongly induced 4 hours after bicarbonate induction as the cells entered stationary phase ( Fig . 7C ) . Moreover , at this time point , the average fluorescence intensity of the crp mutant ( 7×101 FL-1 FITC ) was 350% greater than the average intensity of the induced fraction of the wild type population ( 2×101 FL1-H FITC ) . This shows that CRP plays an important role in repressing virulence gene expression and that it is required for generation of the tcpA bistable phenotype . These results also demonstrate that CRP mainly acts as a repressor of tcpA expression during entry into stationary phase , since the crp mutant showed near normal levels of tcpA expression during exponential phase growth when compared to the wild type parent . To further test the role of CRP-cAMP in the bistable phenotype , we studied a mutant deficient in the production of adenylate cyclase ( cyaA ) , an enzyme responsible for the synthesis of cAMP . As predicted by the requirement of cAMP for activation of CRP , results from flow cytometry studies of tcpA-gfp ( ASV ) expression by the cyaA mutant were identical to the crp mutant ( data not shown ) . To correlate the results from the crp and cyaA mutants with levels of cAMP during exponential growth and stationary phase , cAMP concentrations in the cytosol were measured during growth in LB medium . The concentration of cAMP was found to increase ∼2 . 5 fold between mid exponential phase and early stationary phase ( OD600 = 0 . 9 ) where the bifurcation phenotype is first observed ( Fig . S4 ) . These findings are compatible with the results of previous studies with other species showing that the concentration of cAMP increases during nutrient limiting conditions of growth ( for a review , see [68] ) . Taken together , these results suggest that the growth-phase dependency of the tcpA bistable phenotype could be due to increasing concentrations of the CRP-cAMP complex during entry into stationary phase and competition between this complex and dimeric ToxT on the tcpA promoter . If so , then within the population of bicarbonate-induced cells , those bacteria with higher concentrations of ToxT would be able to sustain tcpA expression during progression into stationary phase whereas tcpA expression would decline in cells with lower concentrations of ToxT . Because the concentration of cAMP in each cell determines the activity of CRP and thus its capacity to compete with ToxT on the tcpA promoter , it is possible that the observed bifurcation in tcpA expression could be caused by different cAMP levels in the two populations . However , CRP-cAMP is known to induce the expression of hapR , and since the expression of hapR was not different between the induced and repressed populations when investigated by RT-PCR ( Fig . 6E ) , the bifurcation in tcpA expression is not likely a result of difference in cAMP levels between the two populations . CRP-cAMP has been shown to increase the expression of rpoS and hapR , both of which down regulate the expression of the tcp and ctx operons [63] . To investigate the effect of HapR on the bistable phenotype , flow cytometry studies of an hapR mutant were carried out after induction of an exponential phase culture with bicarbonate . In contrast to the crp mutation , deletion of hapR did not completely abolish bistability , but rather increased the proportion of cells that continue to express tcpA-gfp ( ASV ) during progression into stationary phase ( Fig . S5 ) . A similar phenotype was observed for the V . cholerae N16961 strain , which is defective in HapR ( data not shown ) . Thus , the role of CRP-cAMP on bistability is not principally mediated through its regulation of hapR , but rather likely comes from the effects of CRP-cAMP at other sites in the regulatory cascade . As discussed above , one such site , predicted by the presence of a CRP promoter recognition motif , is the tcpA promoter [66] , [67] ( see Fig . S5 for more details ) . The in vitro studies reported above demonstrate that tcpA expression bifurcates into two populations in a growth phase dependent manner and that this phenotype is controlled by a bistable switch that is governed by ToxT and by the CRP-cAMP complex . To determine if a bistable switch also segregates tcpA expression into two populations in vivo , we used single cell expression profiling to study tcpA expression by bacteria in the fluid that collects in the lumen of V . cholerae-infected ligated ileal loops . These fluids were used to address the following questions: ( 1 ) does luminal fluid contain a population of tcpA-expressing and a population of tcpA-non-expressing bacteria; ( 2 ) is the adoption of these two phenotypes a random and reversible process; ( 3 ) do tcpA-expressing bacteria in luminal fluid continue to express tcpA after they are transferred from luminal fluid to a medium that does not contain an inducer of tcpA expression; and ( 4 ) , does disruption of crp abolish the bistable phenotype , yielding a unimodal , tcpA-expressing population of bacteria in the loop lumen . In addition , we sought to determine if the tcpA-expressing population of bacteria in luminal fluid coalesce into aggregates compared to the tcpA-non-expressing population . Twelve hours post inoculation , ileal loop fluid samples containing the V . cholerae tcpA-gfp ( ASV ) reporter strain were examined by fluorescence microscopy and the ratio of tcpA-expressing to tcpA-non-expressing bacteria determined . Striking heterogeneity in the expression of tcpA was observed as shown in Fig . 8A–C , where approximately 10% of the individual bacteria produced high levels of the tcpA-gfp ( ASV ) reporter . Similar heterogeneity was observed in 10–20 micrographs from several ileal loops from two individual rabbits . Thus , tcpA expression bifurcates into two populations in vivo in a manner resembling the bifurcation phenotype revealed by the in vitro experiments depicted in Fig . 5–7 . In vitro experiments had also shown that appearance of the bistable phenotype during entry into stationary phase could be explained by positive feedback induction by ToxT acting on the tcpA promoter and down regulation by CRP-cAMP . To determine if crp and the tcpA promoter affect the bistable phenotype in vivo as demonstrated by the in vitro studies reported above , we monitored tcpA-gfp ( ASV ) expression by the crp and tcpA promoter deletion mutants in ileal loop luminal fluid . As expected , none of the cells of the tcpA promoter deletion mutant were observed to express tcpA in the lumen 12 hours post inoculation ( data not shown ) . By contrast , deletion of crp resulted in strong and homogeneous tcpA expression by all of the bacteria observed in the lumen ( Fig . S6 ) , corroborating the results from the in vitro experiments depicted in Fig . 7C . To determine if the heterogeneous expression of tcpA in ileal loop fluid shown in Fig . 8 A–C is a reversible phenotype , ileal loop fluid containing the tcpA-gfp ( ASV ) reporter strain was obtained 12 hours post inoculation , plated onto rifampicin containing LB media . Twenty colonies were picked and individually tested in vitro for their ability to respond to bicarbonate by the induction of tcpA expression . No bacteria from the 20 tested colonies showed tcpA-gfp ( ASV ) expression during exponential growth in bicarbonate-free LB . With the addition of bicarbonate to the medium , tcpA-gfp ( ASV ) expression by all bacteria occurred; bifurcation of tcpA expression ensued during entry into stationary phase ( data not shown ) . This parallels the in vitro findings depicted in Fig . S3 . The bifid population of bacteria that had been isolated from luminal fluid were found to still exhibit the bifurcation phenotype after in vitro growth and re-induction of tcpA expression with bicarbonate . This observation is consistent with the idea that the bifurcation phenotype observed in vivo also is caused by a reversible epigenetic switch . While most V . cholerae in luminal fluid were found as individual planktonic cells , dense aggregates of bacteria were also observed 12 hours post inoculation . Planktonic populations of individual bacteria were found to contain both tcpA-gfp ( ASV ) -expressing and non-expressing bacteria . By contrast , most bacteria in aggregates were found to strongly express the tcpA reporter; one such aggregate is depicted in Fig . 8 D–F . Some aggregates were associated with exfoliated epithelial cells; others appeared to consist entirely of V . cholerae . By contrast , no aggregates were found that were composed of bacteria not expressing the tcpA-gfp ( ASV ) reporter . Thus , while the planktonic population contains both tcpA-expressing and non-expressing bacteria , aggregates appear to be composed mainly of tcpA-expressing bactera . To test if the expression of tcpA-gfp ( ASV ) by some bacteria in luminal fluid was dependent on the continued presence of a chemical inducer , an aliquot of luminal fluid , obtained 12 hour post inoculation , was diluted 10-fold in artificial seawater . This sample together with undiluted luminal fluid was incubated at 30°C for up to four hours and the two samples then assessed by fluorescence microscopy and compared to images of a sample of the same fluid examined immediately after incision of the loop ( Fig . 8 ) . Significantly fewer of the single cells in luminal fluid and in artificial seawater showed strong fluorescence from the tcpA-gfp ( ASV ) reporter after four hours of incubation . By contrast , aggregates of bacteria continued to show very strong expression of tcpA ( Fig . 8G–I ) . Since any potential inducer of tcpA expression that is present in the luminal fluid would be diluted significantly by addition of artificial seawater , these results indicate that the bacteria in the clumps continue to express tcpA for a period of time even in the absence of inducer , a finding that is consistent with the hysteresis phenomenon of bistable switches . Taken together , these results indicate that the mechanism controlling the bifurcation of tcpA expression in vitro is also responsible for causing heterogeneous expression of tcpA during infection of the small intestine . The presence of clumps of V . cholerae expressing high levels of tcpA , and the prolonged expression tcpA in aggregates in luminal fluid and in artificial seawater may ensure the sustained expression of this virulence determinant in recently passed rice water stools . This study was undertaken with the idea that a purely deterministic gene expression model would explain how the anatomical site and time course of infection governs the expression of V . cholerae virulence genes . Microarray expression profiling of V . cholerae collected from two distinct compartments of the intestine seemed to confirm this model ( Fig . 1 ) . Genes encoding CT , the TCP assembly apparatus and the principal subunit of the TCP filament ( tcpA ) were powerfully up-regulated by bacteria on the epithelial surface or in the overlying mucus compared to their expression in fluid secreted into the lumen of the same ileal loop . Further , their expression was greater early in the time course before the mucosal escape response reduced the average expression magnitude of these genes [10] . To more precisely localize tcpA expression and growth in the intestine , single cell gene expression analysis was performed using confocal microscopy and two reporters: tcpA-gfp ( ASV ) to monitor tcpA expression; and , rrnBP1-gfp ( ASV ) as a measure of growth . The magnitude of tcpA expression and the rate of growth as determined by fluorescence from the rrnBP1 reporter varied directly as a function of a bacterium's proximity to the nearest epithelial surface ( Fig . 2 and 4 ) . The recently published crystal structure of ToxT showed that the binding of a fatty acid to the ToxT protein inactivates its transcriptional function [69] . It was speculated that the relative absence of this fatty acid in the mucus compared to the luminal fluid may prime ToxT for induction of tcpA and ctxAB . Breakage of flagella during penetration through the mucus has also been hypothesized to serve as a signal for V . cholerae to maximize virulence gene expression in the presence of the right inducers [70] . A possible chemical inducer of virulence gene expression in the intestine is bicarbonate , which is actively secreted by the mucosal epithelia in the ileum [71] , the primary site of V . cholerae colonization in the small intestine . Bicarbonate has been demonstrated to induce tcpA-F and ctxAB expression in vitro through the activation of ToxT [44] . Virulence gene expression and growth mainly co-localized to the mucosal surface . While the mechanism of these mucosal-associated affects was not explored here , we are intrigued by the possibility that the secretion of V . cholerae virulence-associated proteins by bacteria adjacent to mucus membranes might liberate nutrients from the host . Of particular interest are: CT; the haemagglutinin ( HA ) /protease of V . cholerae encoded by hap; and V . cholerae cytotoxins . In addition to its function as a secretory enterotoxin , CT also triggers the release of mucin from goblet cells in villus and crypt epithelia [72] . hap , which encodes the mucin degrading enzyme ( HA ) /protease , was found in our study to be 2 . 3-fold more strongly expressed in the mucus/epithelial surface compartment eight hours post inoculation than by mid log phase bacteria ( data not shown ) . Thus , based on these site specific gene expression results , it is possible that CT-mediated mucus release and the degradation of mucus by ( HA ) /protease both occur in the mucus/epithelial compartment and that the hydrolytic products of mucin degradation provide growth promoting nutrients for V . cholerae . Two V . cholerae cytotoxins were also expressed in the mucus/epithelial cell compartment . hlyA , which encodes haemolysin A , a pore forming toxin that causes cell lysis , was expressed 2 . 5- , 4 . 3- and 7 . 4-fold more strongly in the mucus/epithelial cell compartment than by the mid log phase reference 4 , 8 and 12 hours post inoculation ( data not shown ) . Similarly , hlx , which encodes a hemolysin , was found to be expressed 3 . 6- , 3 . 2- and 3 . 0-fold more strongly in the mucus/epithelial cell compartment than by the mid log phase reference 4 , 8 and 12 hour post-inoculation ( data not shown ) . These cytolytic proteins could release intracellular growth promoting nutrients , including iron , from host cells; this effect might explain why islands of growing V . cholerae are found near extruded epithelial cells ( Fig . S2 ) . Review of confocal images of V . cholerae tcpA-gfp ( ASV ) in ileal loops indicated that a deterministic gene expression model could not explain all the results of the single cell expression study: in some locations on mucus membranes individual adjacent bacteria appear to produce dramatically different amounts of the TcpA-GFP ( ASV ) reporter ( Fig . 5A ) . Similar heterogeneity in the expression of tcpA was also observed in bacteria present in the luminal fluid 12 hours post inoculation ( Fig . 8 ) . By contrast , little variation between adjacent bacteria was seen in fluorescence emitted by the rrnBP1-gfp reporter ( data not shown ) . Because the irregular physical features of the intestinal environment made it difficult to confirm , quantify or systematically study the apparent cell-to-cell variation in tcpA expression we turned to in vitro studies . To exclude the possible effects of physical or chemical gradients we studied well-stirred homogenous cultures of V . cholerae tcpA-gfp ( ASV ) containing bicarbonate , an inducer of tcpA and ctxAB expression . When bicarbonate was added to early exponential phase LB broth cultures to a final concentration of 100 mM , all cells in the population were induced as a monomodal peak typical of a normally-distributed , cell-to-cell variance in gene expression ( Fig . 6C and 7A ) . However , during transition into early stationary phase , the population bifurcated into two nearly equal fractions: in one fraction , tcpA-gfp ( ASV ) expression persisted undiminished for at least 4 hours whereas in the other fraction the average level of tcpA-gfp ( ASV ) expression declined to pre-induction levels . Studies using bicarbonate-induced cells sorted by FACS showed that the bifurcation phenotype was entirely reversible ( Fig . S3 ) . Thus , it is very likely not caused by durable genetic changes nor is it a manifestation of two pre-existing populations . Taken together , these results lead to a combined deterministic and probabilistic model of tcpA expression . It is deterministic in that it requires exposure to an inducer and is inducer concentration and growth phase dependent . It is probabilistic in that each cell in the pre-induced population has an equal chance to be in each of the two post-induction populations that characterize the bifurcation phenotype . The toxin-coregulated pilus was so-named because its biosynthesis is governed by the same ToxR hierarchy of transcription factors that regulate the production of CT [73] . FACS and a multiplex RT-PCR assay were used to determine if other components of the ToxR regulon were expressed in a bifid manner by tcpA-gfp ( ASV ) -expressing and non-expressing cells from the same population . In addition to tcpA , ctxA was also found to exhibit the bifurcation phenotype . This result shows that genes encoding the main determinants of virulence in humans , CT and TCP , both exhibit the bifurcation phenotype . Moreover , significantly larger numbers of the toxT transcript are produced in the same population of sorted cells that also contains larger numbers of the tcpA and ctxA transcripts . This result mechanistically implicates ToxT as a key component of the tcpA/ctxA bifurcation phenotype . Expression of toxT is positively autoregulated: ToxT dimerizes to activate the promoter of the tcpA-F operon; read-through to the next transcriptional unit activates the toxT promoter thus resulting in a positive feedback loop ( Fig . 9 ) [56] . To determine if this feedback loop is required for the bifurcation phenotype , the indigenous tcpA promoter in VPI-1 was deleted in the reporter strain that carries the promoter-containing tcpA-gfp ( ASV ) construct in an ectopic location . Exposure of this mutant to bicarbonate during early exponential growth induced tcpA-gfp ( ASV ) expression , but abolished the bifurcation phenotype ( Fig . 7B ) . Instead of segregating into two populations during progression into stationary phase , a monomodal peak of declining fluorescence occurred . This result supports the idea that tcpA expression ( and by inference ctxAB expression as well ) is governed by an autocatalyic process that renders it hypersensitive to the concentration of ToxT dimers in the cell . This hypersensitivity comes in part from the stochasticity of the biochemical processes that underlie transcription and translation and which cause the concentration of a particular protein to vary , in a normally distributed manner , between individual cells [74] . The consequences of deleting the native tcpA promoter , depicted in Fig . 7B , show that the autoinduction of toxT very likely magnifies the cell-to-cell random variation in ToxT concentrations according to the following model . In cells with low ToxT monomer concentrations , few ToxT dimers form and a non-linear increase in ToxT production via the positive feedback loop does not occur . By contrast , in cells containing higher concentrations of ToxT monomers , ToxT dimer concentrations are correspondingly high . This favors an autocatalytic , non-linear increase in ToxT production resulting in yet higher concentrations of ToxT dimers . Cells with a high concentration of ToxT dimers are capable of sustaining tcpA expression after progression into stationary phase ( Fig . 9 ) . The autoinduction of toxT may also explain why cells continue to sustain tcpA expression after removal of the inducer ( bicarbonate ) : the half-life of sustained tcpA expression is a function of the half-life of transcriptionally active ToxT dimers . These scenarios were not directly tested in the work presented here by measurements of ToxT monomers and dimers in each of the two populations . However , they are strongly supported by the demonstration that toxT transcript abundance is 6-fold greater in FACS-sorted cells that sustain tcpA expression compared to those that do not ( Fig . 6E ) . The molecular model provided above does not yet explain the growth phase dependency of the bifurcation phenotype . To address this question we showed that mutants with deletions of the CRP or adenylate cyclase coding sequences do not repress bicarbonate-induced tcpA-gfp ( ASV ) expression during entry into stationary phase ( Fig . 7C ) . Thus , in addition to the ToxT positive feedback loop , the bifurcation phenotype requires CRP and adenylate cyclase . To integrate this observation with the tcpA-dependent toxT autcatalytic circuit discussed above and with previous work identifying predicted CRP-binding motifs in the tcpA promoter [61] , we refine our model by introducing the role of nutrient limitation and its capacity to increase cAMP production ( Fig . 9 ) . We propose that V . cholerae encounters nutrient limitation during entry into stationary phase and during late stages of the infectious process . Consistent with this model is the effect of the adenylate cyclase mutation on the bifurcation phenotype , the increase in cAMP concentrations during entry into stationary phase ( Fig . S4 ) and results from our previous work on the role of RpoS and the mucosal escape response [10] . The resulting increase in the intracellular concentrations of the CRP-cAMP complex would then suppress tcpA expression in the population of cells with lower average ToxT dimer concentrations . The autocatalytic nature of this transcriptional system , bifurcation of the population into two distinct tcpA-expressing populations and persistence of the bifid pattern after the removal of inducer are characteristic of an epigenetic mechanism that gives rise to a bistable switch . During in vitro conditions , the bistable regulation of tcpA expression was found to involve positive feedback induction by ToxT acting through the tcpA promoter in conjunction with CRP-cAMP-mediated repression of tcpA expression . To test if the apparent heterogeneity in the expression of the tcpA-gfp ( ASV ) reporter on epithelial cell surfaces and in the overlying mucus layer ( Fig . 5 ) was caused by the same mechanism , we used luminal fluids from freshly incised V . cholerae-infected ileal loops to monitor expression of the tcpA-gfp ( ASV ) reporter by the tcpA promoter and crp mutants . Three reasons led to our use of luminal fluids and the 12 hour post-inoculum time point to study the bistable control of tcpA expression in vivo: ( 1 ) this condition corresponds in time to the previously described mucosal escape response [10] , when the average levels of tcpA and ctxAB expression decline; ( 2 ) the bacterial growth phase in luminal fluid at this time point , as determined by viable plate counts , correlates with entry into or early stationary phase [10] , a time in the growth cycle of in vitro cultures when the population bifurcates into tcpA-expressing and tcpA-non-expressing subpopulations; and 3 ) , expression of the tcpA-gfp ( ASV ) reporter by bacteria in luminal fluid is more easily studied compared to bacteria embedded in mucus or attached to epithelial surfaces . The tcpA promoter mutant lacking positive feedback induction of toxT from the tcpA promoter showed homogeneous low-level fluorescence from the tcpA reporter fusion after 12 hours of infection . This result likely indicates low and homogeneous ToxT concentrations . As expected , the crp deletion mutant showed strong and homogeneous expression from the tcpA promoter in luminal fluid late in the infectious process ( Fig . S6 ) . The study of bacteria isolated from luminal fluid also showed that the heterogeneous expression of tcpA is reversible and thus not caused by durable genetic changes of the bacteria . In addition , the study of bacteria removed from the ileal loop demonstrated continued ( ≥ four hours ) strong expression of tcpA , particularly by bacteria within aggregates , even after dilution of the luminal fluid into artificial seawater . This finding indicates that tcpA expression by bacteria removed from the ileal loop can be sustained by a fraction of the population for a period of time and thus is likely not dependent on the continued presence of a chemical inducer ( Fig 8G–I ) . Taken together these results provide compelling evidence that the mechanisms responsible for the bistable regulation of virulence gene expression , documented and studied by in vitro experiments , is also responsible for bifurcation of the tcpA-expressing population in ileal loops . Other bacterial systems that employ a bistable switch to generate heterogeneity in the bacterial population include a positive feedback loop that controls the expression of a toxin ( hipA ) and an antitoxin ( hipB ) in E . coli . This system generates slow or non-growing persister cells with increased antibiotic resistance [75] , [76] . Genetic competence in Bacillus subtilis is regulated by a positive feedback loop partly controlled by quorum sensing [77] , [78] . In this system , the master regulator ComK , which controls expression of DNA transport genes , binds its own promoter in a dimerized form . Similar to the role of dimeric ToxT described in this report , the combination of a positive feedback loop and dimerization of the transcription factor causes the expression of B . subtilis competence genes to be hypersensitive to changes in the concentration of ComK . As a consequence , only a fraction of the cells becomes competent for natural transformation . Sporulation in B . subtilis is also controlled by a positive feedback loop acting on Spo0A , which results in a subpopulation of cells that sporulate during entry into stationary phase [79][80] , [81] . Likewise , the expression of genes involved in B . subtilis biofilm formation are subject to bistable expression through regulation by Spo0A [82] . However , to our knowledge , the heterogeneous expression of tcpA and ctxAB in V . cholerae presented here is the first example of bistable regulation of a major virulence pathway in a pathogenic organism . In the present study , the rabbit ileal loop model was chosen since it is a well established model of V . cholerae infection [7] . Unlike the murine model system , infection of the rabbit ileal loop with V . cholerae leads to a diarrheal response mimicking the one observed in humans [9] , [83] . A potential disadvantage of the rabbit model system is the closed nature of the loops that potentially could offset the timing and extent of the observed phenomenon late during the infection . In a landmark study , Lee et al . used a recombinase-based reporter system to demonstrate that 95% of V . cholerae bacteria isolated from the intestine of an infant mouse model of cholera expressed tcpA during the initial stages of the infection [5] . In the present study , only a subset of bacteria close to the epithelial surface was shown to express tcpA . Differences between the two model systems may explain these contradicting findings . However , perhaps a more likely explanation comes from differences in the reporter systems used . Activation of the recombinase reporter results in a permanent , non-reversible change that is monitored by plating and scoring bacteria isolated from the intestine . Thus , if a recombinase-linked promoter is activated at any time during the infectious process , bacteria harboring that reporter will be scored as positive even if the promoter is silenced at later time points . The recombinasae reporter thus provides a cumulative average of promoter activation events up to the sampling point . By contrast , the unstable ( ∼40 min half-life ) GFP ( ASV ) reporter used here reports real time levels of promoter activity . Consequently , activation of the gfp ( ASV ) linked promoter early in the infectious process , followed by silencing of the promoter subsequently , will be scored negative in samples taken at later time points and evaluated by in situ confocal microscopy . Consequently , the recombinase reporter system would not have detected silencing of virulence gene expression by a fraction of the V . cholerae population during the mucosal escape response that occurs late in the infectious process [10] . When considered together , the results of Lee et al . [5] and those reported here suggest a very dynamic regulation of virulence determinants including their induced expression by most bacteria early in the time course and repressed expression of the same genes at later stages of the infectious process . The bistable control of virulence gene expression could potentially contribute to the transmission of cholera . V . cholerae was found to exist in luminal fluid as individual planktonic bacteria and as bacterial aggregates ( Fig . 8A–F ) . Heterogeneity of this kind has been observed by others in luminal fluid from the rabbit ileal loop model [24] and in patient stool samples [25] [23] . In the present study , bacterial aggregates were found to express high levels of tcpA , a finding that is consistent with the autoaggregating properties of TCP . It has been suggested that the enhanced infectivity of V . cholerae shed in human stools is due to the presence of clumps of cells that disperse in vivo , thereby providing a high infectious dose of the pathogen [23] , [24] . Other studies have identified hyperinfectious individual bacteria in feces [22] , [25] , [26] , [27] . Thus , the relative importance of aggregated versus individual bacteria as causes of the hyperinfectious state is at present unresolved . Nonetheless , it seems clear that V . cholerae remain hyperinfectious for at least 5 hours after passage from patients into an aquatic environment [27] . In the present study , continued expression of tcpA by bacteria in aggregates was observed even after dilution of luminal fluid into artificial seawater for at least four hours ( Fig . 8G–I ) . This observation favors the view that sustained production of TCP by a sub-population of autoaggregating bacteria in feces accounts for the hyperinfectious , biofilm-like aggregates of V . cholerae passed in the rice water stools of cholera patients . All animal experiments were performed in accordance to NIH guidelines , the Animal Welfare Act , and US federal law . Such experiments were approved by Stanford University's Administrative Panel on Laboratory Animal Care ( A-PLAC ) , which has been accredited by the Association of Assessment and Accreditation of Laboratory Animal Care International ( AAALAC ) . All animals were housed in a centralized and AAALAC-accredited research animal facility that is fully staffed with trained husbandry , technical , and veterinary personnel . A V . cholerae O1 stool isolate from a cholera patient was used for all experiments . The strain , supplied by the California State Department of Health , was acquired from discarded material and previously de-identified . The isolate is a smooth phase variant of strain A1552 ( wild type , El Tor , Inaba and RifR ) . The construction and properties of the rpoS and hapR deletion mutants of the parental strain were previously described [10] . E . coli strain DH5 was used for standard DNA manipulation experiments , and the E . coli strain S17-1λ pir was used for conjugation with V . cholerae . Bacteria were grown in Luria Bertani ( LB ) broth with 0 . 5% NaCl at 37°C . When appropriate , 100 µg/ml ampicillin , 100 µg/ml rifampicin or 50 µg/ml gentamycin was added to the media . Induction of virulence gene expression and bistability of tcpA expression was studied during bacterial growth in AKI conditions [84] and in LB media supplemented with 100 mM NaHCO3 . Intracellular concentration of cAMP was determined using the Biotrak cAMP competitive enzyme-immunoassay system RPN225 ( GE Healthcare , Piscataway , NJ ) using the non-acetylated method according to the manufacturers instructions . Concentration of cAMP was normalized to the total protein concentration in the sample as determined by a standard Bradford assay [85] . Non-polar deletions were generated essentially as described [86] . Crossover PCR was performed to amplify a fragment ( with primers 1 and 4 ) that brings an upstream gene fragment ( produced by PCR with primers 1 and 2 ) to a downstream gene fragment ( produced by PCR with primers 3 and 4 ) thereby creating an in-frame deletion . The fragment was ligated into the sucrose-based counter selectable plasmid pGP704-Sac28 [10] . The plasmid was introduced into V . cholerae A1552 by biparental mating . Sucrose-based counter selection was done essentially as described [86] . Deletions were confirmed by PCR . Primers used for construction of mutants are listed in Table S1 . The BglI fragment from pBK-mini-Tn7-eyfp-a [87] , [88] containing a mini-Tn7 was cloned into pGP704 [89] between the EcoRI and SalI sites . A NotI fragment internal to the Tn7 was removed and a SacI site outside of the transposon was destroyed by blunting and religating . The NotI fragment from Tn5-rrnBP1-gfp ( ASV ) vector pSM1695 [34] , which encodes the promoter fragment containing nucleotides −70 to +3 relative to the transcription initiation site and the sequence encoding a destabilized GFP [designated GFP ( ASV ) ] , was cloned into the Tn7 NotI site of the Tn7 vector . For the tcpA reporter , the 192 nucleotides preceding the start codon in the tcpA promoter were amplified by PCR and inserted between SacI and SphI , immediately upstream of the gfp ( ASV ) ORF . These transposons were introduced into V . cholerae A1552 by triparental mating using helper plasmid pUX-BF13 ( carrying the transposase genes ) followed by selection on TCBS/gentamycin . Transposition into the chromosome was confirmed by PCR . For fluorescence analysis of the rrnB-GFP reporter strain , the cells were grown in a total volume of 20 ml in a 100 ml shake flask in LB broth with strong agitation ( 250 rpm ) at 37°C . Multiple samples were taken from mid-exponential phase and every half hour during transition into stationary phase . The bacteria were fixed in 2% paraformaldehyde in 100 mM phosphate buffer ( pH 7 . 4 ) for 2 hours , washed twice in PBS buffer and analyzed by flow cytometry ( FACSCaliber , BD biosciences , San Jose , CA ) . A total of 100 , 000 bacteria were analyzed for each sample . For fluorescence analysis of the tcpA-gfp ( ASV ) reporter strain , the cells were grown with strong agitation ( 250 rpm ) at 37°C in a 100 ml shake flask containing 20 ml of LB broth supplemented with 50 mM HEPES buffer . Bicarbonate was used to induce tcpA expression by the tcpA-gfp ( ASV ) reporter strain . At an OD600 of 0 . 4 , NaHCO3 from a freshly prepared 1 . 0 M NaHCO3 stock solution in LB was added to the medium to a final concentration of 100 mM . Samples were subsequently taken during exponential phase and during entry into stationary phase , at which time bistable expression of tcpA-gfp ( ASV ) was first evident . At each time point , bacteria were fixed in 2% paraformaldehyde in 100 mM phosphate buffer ( pH 7 . 4 ) for 2 hours . Then , the cells were washed in PBS buffer twice before flow cytometry ( FACS Caliber , BD biosciences , San Jose , CA ) . A total of 100 , 000 bacteria were analyzed for each sample . Fluorescence activated cell sorting ( FACS ) was used to sort V . cholerae in order to separate and collect cells that either produced or suppressed fluorescence emission from the tcpA-gfp ( ASV ) reporter . A mid log phase culture was induced with 100 mM bicarbonate and grown for two hours at which time the bistable phenotype had developed during entry into stationary phase; bacteria were collected and placed on ice . A fluorescence activated cell sorter ( BD Digital Advantage , BD Biosciences , San Jose , CA ) was used to sort bacteria depending on their GFP fluorescence level . Forward light scatter was used as a second parameter to create a positive identification of bacteria in the solution . A total of 1 , 000 , 000 bacteria showing either high or low GFP fluorescence were collected . For RT-PCR expression analysis , bacteria were induced with 100 mM bicarbonate and grown for approximately two hours until bistability was observed . Then , the bacteria were gently fixed with 0 . 5% paraformaldehyde in 100 mM phosphate buffer , pH 7 . 4 for 15 minutes . The fixed cells were washed three times with PBS buffer and placed on ice . A total of 1 , 600 , 000 bacteria showing either high or low levels of GFP fluorescence were collected using FACS . The cells were harvested by centrifugation and frozen on dry ice until RNA extraction . To extract RNA for RT-PCR analysis , the frozen cells were thawed , lysed with RLT buffer ( Qiagen , Valencia , CA ) and treated with proteinase K ( Qiagen , Valencia , CA ) for 10 minutes at 55°C in order to degrade protein cross linked to the RNA . RNA was isolated from the solution using an RNeasy kit ( Qiagen , Valencia , CA ) combined with DNaseI degradation of DNA ( Applied Biosystems , Austin , TX ) . RNA from FACS-sorted bacteria was recovered as described above , and an equivalent of 20 ng of total RNA was used in a real-time RT-PCR reaction as previously described [89] . Validation and calibration experiments were performed for all TaqMan probe and primer sets and these showed the expected linear relationship between the cycle threshold , Ct , and the logarithm of the template amount , using genomic DNA as template . All probe-primer sets ( Table S1 ) yielded a curve with the same slope demonstrating that the amplification efficiency of the various targets was similar ( data not shown ) . To select an internal reference for normalization , we performed real-time RT-PCR with primer-probe sets for six house keeping genes with the cDNA samples from the different experiments . We then used the program GENORM to identify the most stably expressed control gene in these samples as previously described [89] . Relative expression levels in the different samples were calculated by using the comparative Ct method with VC1186 and VC2233 as internal controls . For quantitative real-time RT-PCR of samples isolated by fluorescence activated cell sorting , 15 cycles of pre-amplification were performed with the appropriate primers to ensure adequate signal due to the limited concentration of RNA in the samples . All animal work was conducted according to national and international guidelines . The animal experimental protocol was reviewed and approved by the institutional animal care and use committee of Stanford University . Ileal loop preparation and inoculation was performed essentially as previously described [10] . GFP-labeled bacteria were grown overnight , diluted 100-fold in fresh LB broth and grown to an optical density of OD600 = 0 . 3 . Bacteria were diluted tenfold to OD600 = 0 . 03 in PBS buffer and kept at room temperature until injected into ileal loops . After appropriate incubation , ileal loops used for scanning confocal microscopy were cut open and stretched gently onto cardboard discs . The tissue was allowed to adhere to the cardboard before it was gently submersed into 2% paraformaldehyde in 100 mM phosphate buffer pH 7 . 4 and allowed to fixate for two hours . The fixative was washed away in three subsequent washes with PBS buffer . Then , blocks of approximately 0 . 3 cm2 were excised and transferred to a 96 well microtiter plate . The tissue samples were then permeabilized and blocked in staining buffer ( PBS with 1% saponin and 3% BSA ) for one hour before overnight staining with a 2 . 5 µg/ml anti-Vibrio cholerae O1 IgG monoclonal mouse antibody ( VCM-5261-5 , Austral biologicals , San Ramon , CA ) in staining buffer . Samples were washed gently three times in PBS buffer and subsequently stained for 5 hours with 20-fold diluted Alexa Fluor 660 phalloidin ( A-22285 , Molecular Probes ) and 10 µg/ml goat anti-mouse Alexa Fluor 594 antibody ( A11005 , Molecular Probes ) in staining buffer . After staining , samples were gently washed twice in PBS buffer and mounted for microscopy in Slowfade Light Antifade Kit ( Molecular probes , Eugene , OR ) . Samples were imaged with a BioRad MRC1000 confocal microscope adjusted to identical settings for all images . The z-stacks were reconstructed onto z-projections using the Imaris-software ( Bitplane , Zurich , Switzerland ) and figures were assembled with Photoshop CS ( Adobe , San Jose , CA ) . For quantification of GFP fluorescence as a function of distance to the nearest epithelial surface , the epithelial surfaces in stacks of images was first outlined in Photoshop CS . The Border function was used to select incremental layers of 5 µm distances from the epithelial surface . The fluorescence in the green ( GFP ) channel versus the red channel ( V . cholerae fluorescent antibody ) was quantified using the program Scion Image ( Scion Corporation , Frederick , MA ) . Luminal fluid was isolated 12 hours post inoculation of the rabbit ileal loop with the tcpA-GFP reporter strain . A sample of luminal fluid was immediately fixed in 2% paraformaldehyde in 100 mM phosphate buffer pH 7 . 4 . Another part of the luminal fluid was incubated at 30°C while an aliquot was diluted 10-fold in defined artificial sea water ( 234 mM NaCl , 27 . 5 mM MgSO4 , 1 . 5 mM NaHCO3 , 4 . 95 mM CaCl2 , 5 . 15 mM KCl , 0 . 07 mM Na2B4O7 , 0 . 05 mM SrCl , 0 . 015 mM NaBr , 0 . 001 mM NaI , 0 . 013 mM LiCl , 18 . 7 mM NH4Cl , 0 . 187 mM K2HPO4 , 50 mM HEPES , pH 7 . 4 ) and incubated at 30°C for up to four hours . Samples were isolated every hour and fixed in paraformaldehyde . Samples from the artificial seawater were harvested by gentle centrifugation and resuspension in a smaller volume of PBS buffer . For staining of V . cholerae , 20uL samples were spread out and allowed to dry on poly-L-lysine coated microscope slides and washed in 96% ethanol . Subsequently , the slides were washed in staining buffer ( PBS with 1% saponin and 3% BSA ) for 10 minutes before staining with a 2 . 5 µg/ml anti-Vibrio cholerae O1 IgG monoclonal mouse antibody ( VCM-5261-5 , Austral biologicals , San Ramon , CA ) in staining buffer for 1 hour . The slides were washed three times with PBS buffer and subsequently stained for 30 minutes with 10 µg/ml goat anti-mouse Alexa Fluor 594 antibody ( A11005 , Molecular Probes ) in staining buffer . The slides were then washed in PBS buffer and analyzed using fluorescence microscopy . V . cholerae , grown to mid-exponential phase in vitro , were used as the source of “reference RNA” for the two-color hybridization microassay assay described below . For this purpose , bacteria were grown to OD600 = 0 . 3 in LB medium at 37°C; then , the bacteria were quickly centrifuged , the bacterial pellet resuspended in Trizol reagent ( GIBCO/BRL ) and the suspension frozen on dry ice . Transcriptional analysis of wild type V . cholerae isolated from the fluid that collects in rabbit ileal loops during infection was performed by quickly pelleting bacteria from the ileal loop fluid after eight and 12 hours of infection . The bacteria were then resuspended in Trizol reagent and frozen on dry ice . The mucus and cell-associated bacteria were isolated as a single fraction from ileal loops four , eight and 12 hours post inoculation by scraping the epithelial surface of the intestine with a disposable plastic cell scraper after the loops had been cut open and gently rinsed in PBS buffer to wash away remaining luminal fluid . The mucus gel/cell associated fraction was suspended in Trizol and frozen on dry ice . Total RNA , which includes RNA from the infecting bacteria and from the host ( see below ) , was isolated from the thawed Trizol suspension , treated with DNaseI ( Applied Biosystems , Austin , TX ) and cleaned by using the RNeasy kit ( Qiagen , Valencia , CA ) . To avoid microarray expression artifacts imparted by the presence of host RNA , RNA was prepared from a healthy rabbit and added to the V . cholerae reference RNA , isolated as described above from bacteria grown to mid exponential phase in LB broth . The amount of host RNA added to the V . cholerae reference RNA was determined as follows . RT-PCR analysis using Taq Man probe specific for V . cholerae 16S rRNA was used to estimate the amount of V . cholerae RNA in the total RNA fraction . This estimate was used to determine the amount of rabbit RNA that was added to the mid log phase reference V . cholerae RNA . The presence of contaminating host RNA in the mucus gel/epithelial RNA extract was further managed as follows . Primers specific for over 85% of the ORFs identified by the V . cholerae genome sequencing project were used to prime reverse transcriptase reactions in order to enrich for V . cholerae cDNAs prepared from the RNA extracted from the mucus gel/epithelial surface fraction . Labeling of cDNA and microarray hybridizations were performed as described [7] . cDNA from the mucus gel/epithelial cell fraction or luminal fluid fraction was labeled with Cy5 whereas cDNA prepared from mid exponential phase LB grown V . cholerae was labeled with Cy3 . RNA from an exponentially growing LB culture was chosen as a reference since virulence gene expression and accumulation of virulence factors is not observed under these conditions . Microarrays were scanned with a GenePix 400A instrument ( Axon Instruments ) , using the GENEPIX 5 . 0 software . To avoid fluctuations in intensity values from genes that are not expressed to a measurable level , we designated a minimum background level for each channel [89] . Statistically significant changes in gene expression were identified by conducting a one-class analysis using the Significance Analysis of Microarrays program [90] with a threshold of 2-fold change and a 0% false discovery rate for all samples . Raw microarray data are available in Tables S3 , S4 , S5 , S6 and S7 and at http://smd . stanford . edu/ . The TIGR-CMR genome database ( http://cmr . tigr . org/tigr-scripts/CMR/GenomePage . cgi ? database=gvc ) accession numbers used in this paper are tcpA ( VC0828 ) , toxT ( VC0838 ) , tcpA-F ( VC0828-37 ) , ctxA ( VC1457 ) , ctxB ( vc1456 ) , crp ( vc2614 ) , toxR ( vc0984 ) , toxR ( VC0984 ) , rpoS ( VC0534 ) , aphA ( VC1049 ) , aphB ( VC2647 ) , tcpP ( VC0826 ) , hapR ( VC0583 ) , tcpH ( VC0827 ) , toxS ( VC0983 ) , cyaA ( VC0122 ) , hapA ( VCA0865 ) , hlyA ( VCA00219 ) , hlx ( VCA0594 ) , cheA1 ( VC1397 ) , cheA2 ( VC02063 ) , cheA3 ( VCA1095 ) and rrnB ( E . coli b3968 ) .	Most pathogenic microorganisms infect in a stepwise manner: colonization of host surfaces is followed by invasion and injury of host tissues and , late in the infectious process , dissemination to other hosts occurs . During its residence in the host , the pathogen produces essential virulence determinants and often replicates rapidly , leading to a vast expansion of its biomass . Although this scenario is well established also for Vibrio cholerae , the cause of a potentially fatal diarrheal illness , it has not previously been possible to identify precisely when or where virulence determinants are produced in the intestine . We addressed this question by investigating the expression of virulence genes by individual V . cholerae during infection of the small intestine . Virulence genes were found to be powerfully expressed early in the infectious process by bacteria in close proximity to epithelial surfaces . Increased replication rates were also localized to epithelial surfaces . During later stages of the infection , the population of V . cholerae bifurcates into two fractions: one subpopulation continues to express virulence genes , whereas these genes are silenced in the other subpopulation . The genetic program controlling the continued production of virulence genes may mediate the persistence of a hyper-infectious subpopulation of bacteria in the stools of cholera patients .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "infectious", "diseases", "genetics", "and", "genomics/gene", "expression", "microbiology/microbial", "physiology", "and", "metabolism", "infectious", "diseases/bacterial", "infections", "genetics", "and", "genomics/epigenetics", "microbiology/microbial", "growth", "and", "development" ]	2010	A Bistable Switch and Anatomical Site Control Vibrio cholerae Virulence Gene Expression in the Intestine
The West Nile virus ( WNV ) , isolated in 1937 , is an arbovirus ( arthropod-borne virus ) that infects thousands of people each year . Despite its burden on global health , little is known about the virus’ biological and evolutionary dynamics . As several lineages are endemic in West Africa , we obtained the complete polyprotein sequence from three isolates from the early 1990s , each representing a different lineage . We then investigated differences in growth behavior and pathogenicity for four distinct West African lineages in arthropod ( Ap61 ) and primate ( Vero ) cell lines , and in mice . We found that genetic differences , as well as viral-host interactions , could play a role in the biological properties in different WNV isolates in vitro , such as: ( i ) genome replication , ( ii ) protein translation , ( iii ) particle release , and ( iv ) virulence . Our findings demonstrate the endemic diversity of West African WNV strains and support future investigations into ( i ) the nature of WNV emergence , ( ii ) neurological tropism , and ( iii ) host adaptation . West Nile virus ( WNV ) is a member of the Japanese Encephalitis virus ( JEV ) serocomplex and is a part of the genus Flavivirus of the family Flaviviridae . The WNV is a single-stranded , positive-sense RNA virus . The genomic RNA is about 11 kilobases ( kb ) , containing one long open reading frame ( ORF ) flanked by 2 non-coding regions . This ORF encodes for a polyprotein , which is processed into three individual structural ( Capsid , pre-Membrane , Envelope ) , and seven non-structural ( NS1 , NS2A , NS2B , NS3 , NS4A , NS4B , and NS5 ) proteins [1–4] . West Nile fever disease ( WN fever ) is caused by the WNV . WN fever in humans can range from asymptomatic infections or mild acute febrile illness , to neurological diseases including meningitis , encephalitis , and acute flaccid paralysis [5–7] . WNV’s host range is extensive: it has been detected in over 65 species of mosquitoes and ticks , 225 species of birds , and 29 different animals [8 , 9] . A human vaccine or specific antiviral treatment for WN fever is currently unavailable . WNV was first discovered and isolated from the blood of a woman suffering from febrile illness in 1937 in Uganda [10] . Cases of WN fever were documented in Israel and Egypt in the early 1950s , France in the 1960s , and South Africa in the 1970s [11] . The global awareness of WN fever increased in the 1990s , as sporadic and major outbreaks occurred , primarily in the Mediterranean Basin and occasionally in Europe [5] . In 1999 , WNV unexpectedly emerged in New York City , signifying the first confirmed incidence of WNV in the Western Hemisphere . Since then , WNV has spread throughout the Americas , causing over 20 , 265 cases of neurological disease and 1 , 783 case fatalities in humans and even higher rates of mortality among birds in the United States [12–16] . Meanwhile , WNV continued to spread and cause WN disease and encephalitis in Europe , Asia , and Oceania [17] . In the 1990s , the largest outbreaks occurred in Romania in 1996 [18] , and Russia in 1999 [19] , with 17 and 40 human fatalities , respectively . In the 21st century , emergences of WN fever and encephalitis have been reported in Europe [20] , with a hallmark human neurological outbreak in Greece in 2010 [21] , and several noteworthy outbreaks in Italy [22–24] , Hungary [25] and Serbia [26] . WNV is biologically diverse; up to nine lineages have been proposed [27–30] . However , most human outbreaks of WN encephalitis have been attributed to lineages 1 and 2 . Lineage 1 is globally spread and exists in distinct clades . Clade 1a comprises of strains isolated from Europe , Africa , and the Americas . Clade 1b , also referred to as Kunjin virus , has been restricted to Oceania . Major outbreaks in Europe , Africa , and the Americas with neurological diseases are caused by strains belonging to lineage 1 , with an exception to clade 1b where neurological disease is rarely reported [12 , 31 , 32] . Lineage 2 was exclusively reported in Africa up until 2004 , until it was isolated from humans and bird populations in Hungary , Greece , and Italy [4 , 21 , 23 , 33] . Lineage 2 was also considered to be less pathogenic than lineage 1 , until it caused severe disease in South Africa and encephalitis among birds and humans in Europe [4 , 21 , 23 , 33 , 34] . Both lineages include strains with varying degrees of neuroinvasiveness in humans [35] . Besides lineages 1 and 2 , there are lineages that are less widespread . Lineage 3 , also referred to as Rabensburg virus , was repeatedly isolated in the Czech Republic [36–38] . Lineage 4 has been isolated and reported from Russia [39] . The 5th lineage was isolated from India , and is often identified as a distinct clade of lineage 1 ( clade 1c ) [40] . A putative 6th lineage , based on a small gene fragment , has been described from Spain [27 , 41] . Koutango virus ( lineage 7 ) was initially classified as a different virus , but is now a distinct lineage of WN virus [31 , 42] . Lineage 7 strains were isolated from ticks ( this study ) and rodents , a rare feature among WN virus lineages [4] . The Koutango strain virus has also been shown to have a higher virulence than the lineage 1a strain “NY99” in mice [43 , 44] . Although there was a report of an accident where a Senegalese lab worker was symptomatically infected with the Koutango strain , a natural human infection has yet to be confirmed [45] . Additionally , a new lineage ( putative lineage 8 ) of WNV was isolated from Culex perfuscus in Kedougou , Senegal in 1992 [4] . Finally , a putative 9th lineage , or sublineage of lineage 4 , was isolated from Uranotaenia unguiculata mosquitoes in Austria [27] . Despite the presence of lineages 1 , 2 , 7 ( Koutango ) and a putative 8th lineage circulating in Africa [4 , 46 , 47] , WNV has had minor impact on human health . Sporadic outbreaks were observed in several African counties [48–50] , with lower frequencies of neurological disease than that reported from outbreaks in the USA [51 , 52] . For example , Senegal has never had a major outbreak of WN fever , but was the source of several endemic genotypes that were identified and sequenced . Moreover , in Senegal , WNV antibody seroprevalence has been around 80% in sampled humans , horses , and birds [53–57] . A recent study on the vector competence of African WNV lineages demonstrated that local mosquito populations lack efficient transmission of WNV [4] . Besides vector competence—i . e . intrinsic genetic variations among lineages—host adaptation , movement of host populations , climate and ecological factors could play a role in viral replication , virulence , and the outcome of infection . The N-linked glycosylation site of the envelope protein may be associated with differences observed in: ( i ) WNV neuroinvasiveness in mice , ( ii ) viral replication , and ( iii ) transmission of WNV in mosquitoes [4 , 58–60] . In this regard , Senegal has been a focal point in the studies of WNV virus , where multiple lineages of WNV are co-circulating endemically , but whose biology remains poorly understood . To address these questions , we analyzed complete coding regions ( polyproteins ) of four different lineages circulating in Senegal and West Africa . Using additional WNV sequences from Genbank , we performed a phylogenetic analysis using the complete polyprotein sequences of the viruses and investigated sites for positive selection . We also analyzed the biological properties of these 4 WNV lineages using in vitro and in vivo models . Ultimately , understanding the relationships among ecological and genetic differences will ameliorate our understanding of WNV emergence , epidemiology , and its maintenance in nature . In this study , three complete polyprotein genes from Senegal isolates were sequenced: ArD76986 , ArD96655 , and ArD94343 ( Table 1 ) . These novel sequences are representative of lineages 1 , 7 ( Koutango ) and 8 ( putative ) , respectively . The lineage 1 and lineage 8 strains were isolated from Culex mosquito species , while the lineage 7 strain was isolated from a tick species . Acknowledging previous works that have reconstructed the evolutionary history and those that have characterized novel isolates and lineages of WNV , we included seven additional complete ORF sequences to compare differences at the gene and protein level ( Fig 1 ) . Among representative sequences , the average nucleotide pairwise identity is 77 . 6% ( s . d . = 4 . 1% ) and the amino acid average pairwise identity is 90 . 1% ( s . d . = 3 . 3% ) . When comparing individual sequences , the NY99 strain ( Accession number: AF196835 , lineage 1a , United States 1999 ) shared a 99 . 5% pairwise identity to ArD76986 ( Accession number: KY703854 , lineage 1a , Senegal 1990 ) at the amino-acid level ( Fig 1A ) . The sequence diversity of endemic WNV lineages in Senegal ( SN ) is notable , as the lineage 1 strain ( ArD76986 ) was 88 . 9% and 90 . 9% identical at the amino-acid level to the ArD96655 ( Accession number: KY703855 , lineage 7 , SN 1993 ) and the ArD94343 ( Accession number: KY703856 , lineage 8 , SN 1992 ) strains respectively . Between the lineage 7 and lineage 8 strains , the amino-acid pairwise identity was 87 . 9% ( Fig 1A ) . The 1937 WNV isolate of strain B956 ( Accession number: AY532665 , lineage 2 , Uganda 1937 ) is of particular interest , as it is the oldest clinical isolate available with a complete ORF sequenced . Amino acid pairwise identity was 93 . 9% to the NY99 strain sequence and to the lineage 1a ( SN ) sequence , 89% to the lineage 7 strain sequence and 91 . 1% to the lineage 8 strain sequence . At the nucleotide level , the lineage 2 strain ( UG ) had a pairwise identity of 79 . 1% to the NY99 sequence , 79 . 3% to the lineage 1a strain ( SN ) sequence , 76 . 9% to the lineage 7 strain sequence and 77 . 7% to the lineage 8 strain sequence . Additionally , B956 contains a 12 base pair deletion at nucleotide position 1 , 331 , corresponding to the WNV envelope glycosylation site . We compared published sequences and published works to identify whether mutations that have been shown to influence WNV virulence and replication were present in the newly sequenced open-reading frames . For each strain , we discovered amino acid changes that were associated to a phenotypical change and many additional mutations with unknown consequences ( Fig 1B ) . For example , the 22nd and 72nd codon sites of the pre-membrane protein ( prM ) have been shown to play a role in enhancing the virulence and particle secretion in WNV [61] . At this site , we found alterations in the lineage 7 strain ( SN_1993_L7 ) , and the lineage 8 strain ( SN_1992_L8 ) . Another example is the glycosylation site found in the 154-156th positions of the envelope ( Env ) protein , which is considered a virulence factor [62] . We found that the lineage 1 strain from Senegal ( SN_1990_L1a ) , the Kunjin strain ( AU_1991_L1b ) , the NY99 strain ( US_1999_L1a ) and the lineage 8 strain ( SN_1992_L8 ) harbored the NYS motif while other strains had variations or deletions in this locus . Next , the 249th codon position of the NS3 protein [63] , the helicase protein , was found to increase viremia and virulence in birds , and could play a role in other hosts . We observed several variations in our data at the 249th codon position ( Fig 1B ) . Additionally , changes in the highly conserved 120P-E-P-E123 region of the NS4A protein can attenuate or even impair virion replication and release [64] , which we found present in the lineage 8 strain . Finally , a mutation in the NS5 protein , serine ( S ) to phenylalanine ( F ) at the 653rd position in the NS5 protein , is associated with an increased resistance to interferon [65] , a mutation that is shared by the lineage 7 strain ( SN_1993_L7 ) ( Fig 1B ) . We also found several synonymous changes in positions corresponding to known virulence motifs , such as variability in the third codon site position ( the wobble base ) during the translation of serine ( S ) at the 156th codon site . We also investigated sites within the NS2A [66] , NS4B [67] , and additional sites within the NS5 region that are known to impact on infectivity and virulence [65] , but no mutations were present in our sequences . The phylogenetic analysis revealed a similar topology to the ones obtained from previous maximum likelihood trees [27 , 40 , 42 , 68 , 69] . Currently , up to 9 distinct lineages have been suggested . A total of 95 sequences , including 3 novel polyprotein sequences from Senegalese isolates ( Table 1 ) , were used to estimate a maximum-likelihood tree with FastTree ( S1 Fig ) and a very similar relaxed clock Bayesian maximum-clade credibility ( MCC ) tree ( Fig 2 ) , summarizing the MCMC runs with BEAST . The MCC tree was scaled to time ( years ) and branch tip-nodes were colored to identify previously classified lineages [27] . Here , the time to the most recent common ancestor ( tMRCA ) with its corresponding 95% highest posterior density ( HPD ) interval for WNV was estimated in the unit of years . The tMRCA of WNV is predicted to have originated in the late 16th/early 17th century ( 95%HPD: 1476–1765 ) , a major split that diverges lineages 1 , 5 and 7 from lineages 2 , 3 , 4 , 8 , and 9 . Both lineage 1 and 2 show multiple introductions into Europe and other New World countries . Additionally , we see that lineages 1 , 2 , 7 , and 8 have been isolated in West Africa , yet only lineages 1 and 2 have emigrated . Infection , viral proliferation , and virulence in each cell type were measured by 4 different tests over a 146 hours post-infection period: quantitative reverse transcriptase PCR ( qRT-PCR ) of the lysed cell fracture to measure genome replication ( Fig 3A and 3B ) , qRT-PCR of the supernatant fraction to detect genome replication dynamics ( i . e . , total number of particle release ) ( Fig 3C and 3D ) , immunofluorescence staining of the cells to visualize the infectivity of cells and estimate protein translation efficiency ( Fig 3E and 3F ) , and plaque assays to determine the amount of infectious viral particles ( PFU/ml ) from the supernatant fraction ( Fig 3G and 3H ) . Using Ap61 and Vero cells , our goal was to replicate the biology of WNV in a mosquito vector and its vertebrate host . We found that African lineages have different growth dynamics in mosquito and mammalian cell lines . In Aedes pseudoscutellaris cells , growth dynamics were similar for all lineages , ( Fig 3 , left column ) where all lineages exhibited successful replication and generation of infectious particles . In Vero cells ( Fig 3 , right column ) , lineages 1 , 2 , and 7 showed exceptional growth , with lineage 2 strain exhibiting the highest replication and particle release capabilities , and lineage 7 strain having exceptional translational dynamics and highest PFU/ml during the infection interval . We observed cell-specific growth differences among different WNV strains . For example , Fig 3A and 3B showed differences in genome replication dynamics in the cells with respect to host cells . Interestingly , lineage 1 strain had higher genome replication in Ap61 cells ( p-value ranging from 2 . 22x10-16 to 0 . 002 ) while the lineage 2 strain had higher genome replication in Vero cells ( statistically comparable ) . Lineage 8 showed a lower significant replication profile in Vero cells ( p-value ranging from 8 . 81x10-13 to 0 . 031 ) . Furthermore , differences in growth at T0 further supports that WNV lineages could have a preference to a specific cellular environment . The rate of viral attachment , entry and replication initiation can all depend on the genetics of the infecting strain [70] . We estimated the total number of released particles at different times post infection by measuring the WNV RNA copy number in the cell supernatant . All tested lineages had comparable genome copy numbers in Ap61 supernatants ( Fig 3C ) . However , we found a significantly higher copy number of total particles released for the lineage 2 strain at 22 , 28 , and 50 hours post-infection ( hpi ) in both in vitro models ( p-value ranging from 2 . 22x10-16 to 0 . 023 ) . Lineage 8 strain showed significantly lower genome copy numbers in Vero supernatants ( p-value ranging from 8 . 81x10-13 to 0 . 031 ) . Next , we approached differences in protein translation efficiency between lineages by detecting viral proteins using an immunofluorescence assay ( IFA ) ( Fig 3E and 3F ) . The lineage 7 strain displayed more efficient protein translation in both cells ( p-value ranging from 3 . 98x10-13 to 0 . 011 ) , while lineage 8 strain had significantly lower levels of protein translation in Vero cells . Nevertheless , the translation rate in lineage 8 increased significantly from T124-146 in Ap61 cells . We also noticed a delay on translation detection in both cells , with no detectable protein production until T99 hours ( Fig 3E ) and T50 hours ( Fig 3F ) respectively . To quantify the infectious particles of different WNV strains , we used plaque assays to estimate the amount of infectious viral particles ( PFU/ml ) in the supernatant fractions . In Ap61 cells , we found a similar profile of infectious particles production for all lineages , with significant higher rates at 124 hpi and 146 hpi for the lineage 7 strain ( p-value ranging from 2 . 22x10-16 to 0 . 028 ) ( Fig 3G ) . In Vero cells , lineage 1 and lineage 7 strains had higher number of PFU/ml , while lineage 2 had an intermediate profile and lineage 8 had the lowest amount of infectious viral particles , with significant differences from 28 to 124 hpi ( p-value ranging from 2 . 22x10-16 to 0 . 049 ) ( Fig 3H ) . Finally , we approximated the replication efficiency by finding the ratio of the number of virions released in the supernatant–particles that completed the infectious cycle–divided by the number of plaque forming units ( PFU ) [71–73] . We estimated the ratio for each strain in each cell and found significant differences in replication efficiency ( p-value ranging from 5 . 49x10-16 to 0 . 0223 ) ( Fig 4 ) . There are some consistencies with Fig 3 , where the lineage 7 strain was the most efficient and the B956 strain was the least efficient in vitro . Lineage 1 and lineage 7 strains seem to be more cell-specific; both replicated less efficiently in Ap61 cells . To determine the virulence of WNV strains ( Table 1 ) , we challenged five- to six-week-old mice with three different viral doses and observed their overall survival for 21 days . Depending on the strain and dose used , several mice developed clinical disease and died ( Table 2 ) . Clinical signs included tremors , reduced activity and reluctance to move , hind leg paralysis and closed eyes . The PBS-inoculated control groups exhibited no signs of disease throughout the experiment . The lineage 7 strain was the most virulent of the strains at all administered doses ( Wilcoxon rank sum test , p-values < 0 . 05 ) . In fact , the lineage 7 strain induced the shortest survival time compared to the other strains and always resulted in 100% mortality in every experiment ( Fig 5 and Table 2 ) . Interestingly , in most cases , mice inoculated with the lineage 7 strain died without showing any clinical signs . Comparatively , mice inoculated with the lineage 1 and lineage 2 strains usually showed signs of disease at least 1 day before dying . However , lineage 8 showed no virulence ( 100% survival ) at 100 and 1000 PFU doses ( Fig 5B and 5C ) . In fact , only one mouse mortality was observed at 10000 PFU ( Fig 5A and Table 2 ) . To determine the evolutionary pressures acting on the WNV ORF , we estimated the ratio of nonsynonymous ( dN ) to synonymous ( dS ) substitutions per codon site ( where dN—dS > 0 , signifies positive selection ) using 95 sequences , which represent all investigated WNV lineages . Our investigation on selection regimens acting on all WNV complete ORF sequences—with the FUBAR method—revealed 3313 well supported ( posterior probability ≥ 0 . 9 and Bayes Factor < 3 . 0 ) sites under purifying selection ( S1 Table and S2 Fig ) . However , we found 95 statistically significant sites ( p-value ≤ 0 . 1 ) under diversifying episodic selection ( S2 Table and S3 Fig ) , using MEME method . Despite the presence of at least four different lineages in West Africa , there has never been a major outbreak , nor a large frequency of encephalitic cases connected with WNV . The lack of a WN disease “burden” within Senegal could suggest that WNV is endemic , which could explain the high seroprevalence , and therefore , few susceptible hosts [54] . However , the threat of WNV emerging to places where the population’s seroprevalence is much lower or even naive is a serious concern . Avian migratory routes could have played a role in the emigration of WNV strains from Africa [53] , and for other African-borne arborviruses such as Usutu virus [74] . The extensive genetic diversity ( Fig 1A ) and broad host range of WNV [8 , 9] could have also contributed to its global dissemination ( Fig 2 ) , as certain mutations have been previously prosecuted with lineage 1’s entrance in the United States [32] and lineage 2’s emergence in Europe [75] . As a consequence , several groups have investigated how specific genetic changes and selective pressures within the WNV ORF can affect the phenotypical behavior of a WNV strain . In our study , the growth kinetics of the different West African WNV lineages were explored in Aedes pseudoscutellaris ( Ap61 ) and African green monkey kidney cells ( Vero ) to reflect infection dynamics in two common classes of WNV hosts ( insect vector and primate ) ( Figs 3 and 4 ) . The virulence of these lineages in mice was also analyzed ( Fig 5 and Table 2 ) . We found that these 4 West African lineages have significant differences in their ability to proliferate in our tested cell lines and their degree of virulence in mice ( Figs 3 , 4 and 5 ) . We also explored how our in vitro and in vivo results could be explained by their evolutionary ( Fig 2 ) and individual genetic variations ( Fig 1 ) . In agreement with other viruses that use alternate hosts ( vertebrate-arthropod-vertebrate ) and cause acute infections , we found that the majority of WNV codon sites are undergoing purifying selection [76] . Nevertheless , some of the significant episodic diversifying sites that we found are related to virulence , like the 444th and 446th codons in polyprotein gene ( 154-156th positions of the Env protein ) that encode the N-glycosylation motif ( NYS ) . This site is present in many lineage 1 strains , some neuroinvasive lineage 2 strains [34 , 77] , and the lineage 1 and 8 strains from this study ( Fig 1B ) . We found significant diversifying selection acting on these codon sites in 3% and 24% in all of the WNV lineages , respectively ( S1 Table ) . These episodic non-conservative changes could have resulted in the loss of the N-linked glycosylated site motif , which is related to less efficient replication in Culex cells [59] and better replication in Aedes albopictuscells [60] . This N-linked site is also associated with neuroinvasiveness in mice [62] . Second , we found that 1754th codon in polyprotein ( 249th in NS3 ) was under diversifying selection ( ω = 33 . 1 ) 13% of the time and under purifying selection ( ω = -0 . 91 ) 87% of the time ( S1 Table ) in our WNV dataset . As this site was discovered to increase viremia and virulence in birds [63] , further experiments in avian cell lines should be explored to see if our discovered substitutions effect replication in avian hosts and transmission dynamics . Although no significant diversifying selection was observed on the cleavage site in the NS4A protein ( 120PEPE123 motif ) , we did discover the P122S substitution in the lineage 8 strain ( Fig 1B ) . Crucially , induced mutations in this motif are related to low rates of replication and protein production in Vero cells [64] , which we expected and observed for the lineage 8 strain in vitro ( Fig 3B , 3D and 3F ) , and could help explain its low virulence in vivo ( Fig 5 ) . In general , we observed little change in viral replication and protein production between West African strains in Ap61 cells ( Fig 3A , 3C and 3E ) . This could suggest that the conservation of PEPE motif may have a lesser role in replication in mosquito cells ( lineage 8 ) or that the strains may have been “pre-adapted” prior to our experiment . The lineage 7 strain has the S653F NS5 mutation that is associated with an increased resistance to interferon [65] , which could help explain its phenotypical virulence in vitro and in vivo ( Figs 1B , 3 , 4 and 5 ) . However , because Vero cells are known to be interferon-deficient , we could not associate this mutation to our in vitro results for lineage 7 in Vero cells ( Fig 3 , right column ) . Nevertheless , all three West African strains contained a non-synonymous change in a locus that was previously explored by site-directed mutagenesis experiments ( Fig 1B ) . Interestingly , we also detected synonymous changes in the “wobble” base position of the codons in “sites of interest” . However , our knowledge of how synonymous changes impact infectivity , virulence , and replication of WNV is still limited . As previously described , lineages 1 and 2 originated in Africa and emerged as a New World pathogen over the last 60 years [68] . Lineage 7 could be following a similar path; besides Senegal , it has been detected in Somalia , Gabon and possibly in Italy [78–80] . Our study supports that there are other lineages besides 1 and 2—such as lineage 7—that can exhibit high virulence in mice and efficient replication in mammalian cells ( Figs 3 , 4 and 5 ) . This high virulence of lineage 7/Koutango strains in mice has been explored in two other studies , where the high virulence is suggested to be a result of delayed viral clearance and a weak neutralizing antibody response [43 , 44] . All three doses tested resulted in 100% mice mortality ( Table 2 ) , which agreed with previous results . The differences in average survival time and mortality rates compared to previous studies , could be explained by differences in the passage history of viral strains , and the age of the infected mice [81] . The Senegalese lineage 1 strain exhibited moderate virulence in mice ( Fig 5 ) and caused comparably less mortality than the NY99 strain when compared to similar studies [44] . Differences in neuroinvasive potential and virulence among lineage 1 strains has been reported and could be explained by genetic differences [35] . Alternatively , lineage 8 showed poor growth capabilities in Vero cells ( Figs 3 and 4 ) and almost no virulence in mice ( Fig 5 and Table 2 ) , suggesting that it may be restricted to vertical transmission or is species restrictive . Lineage 8 was described to have a similar phenotype to Rabensburg virus ( lineage 3 , Czech Republic 1997 ) . Moreover , growth kinetics and vector competence studies revealed poor growth of the Rabensburg virus in mammalian cell lines and low virulence in mice [36 , 82] . This similarity could indicate that both lineage 8 and the Rabensburg strain may be restricted in host range and are also maintained in nature through vertical transmission . Investigating the vector competence of lineage 8 in different arthropod species ( i . e . Culex , Aedes , and tick species ) could lead to a better understand the transmission dynamics and maintenance cycles of WNV in nature . The low virulence phenotype of the lineage 8 strain could also be a factor for its consideration as a potential vaccine candidate for West Nile fever . Further studies could complement our analysis , particularly , on other factors that could explain differences in WNV host and disease dynamics . Exploring variations in codon usage bias could also help explain biological differences [83] , as distinct lineages have shown different degrees of natural selection and mutational bias . Site-directed mutagenesis studies may also help explain how strain-specific mutations , both synonymous and non-synonymous , could explain deviations in replication efficiency and virulence for our in vitro and in vivo results . For example , future studies in cell lines with interferon may help clarify the impact of the S653F NS5 mutation for the lineage 7 strain . Additionally , the flavivirus 5’ and 3’ untranslated regions ( UTR ) can affect replication and translation; certain mutations in these regions can cause complete viral attenuation [84–87] . Unfortunately , we could not investigate their impact in this study , as the majority of WNV UTR’s were publically unavailable . Taking everything into account , especially differences in sequences , growth dynamics and virulence in vivo , the West Nile virus is a pathogen with the capability to cause severe epidemics anywhere in the globe . As complete genome sequences including the 5’ and 3’ UTR regions are currently being generated , this could lead to future studies focused on in vivo transmission and growth dynamics . As additional strains of WNV are characterized , monitoring the global diversity and distribution will aid in threat assessment and epidemiological modeling if future outbreaks are to occur . Two cell lines have been used for virus cultivation and growth kinetics . Ap61 cells ( Aedes pseudocutellaris ) were grown in L15 ( Leibovitz’s 15 ) medium ( 10% heat-inactivated fetal bovine serum [FBS] , 1% penicillin-streptomycin , 0 . 05% amphotericin B [Fungizone] ( GIBCO by life technologies; USA ) and 10% tryptose phosphate ( Becton , Dickinson and Company Sparks , USA ) and incubated at 28°C without CO2 . Vero cells ( African green monkey kidney epithelial cells; Cercopithecus aethiops ) ( obtained from Sigma Aldrich , France ) were grown using the same medium without tryptose phosphate and CO2 . Furthermore , PS ( Porcine Stable kidney cell line , American type Culture Collection , Manassas , USA ) cells were grown in same conditions than Vero cells and have been used for plaque assay . The virus strains used in this study corresponding to lineages 1 , 2 , Koutango ( lineage 7 ) and 8 were described in Table 1 . The virus stocks were prepared by inoculating Aedes pseudoscutellaris ( Ap61 ) continuous cells lines for 4 days . The infection status was tested by immunofluorescence assay ( IFA ) , real-time RT-PCR ( Reverse Transcriptase-Polymerase Chain Reaction ) and plaque assay . The supernatant of infected cells were aliquoted , frozen at -80°C , and used as viral stocks for growth kinetics . A total of 862 complete WNV polyprotein gene sequences with country and year of isolation data were available and initially downloaded from Genbank for this study . A large number of sequences were from the Americas and formed a monophyletic group of lineage 1a comprising 770 sequences . To reduce computer-processing requirements while maintaining the authenticity of our results , we removed all lineage 1a sequences except for a single representative sequence denoted “NY99” ( accession number: AF196835 ) . With the addition of 3 new sequences , a total of 95 sequences were aligned using Muscle v3 . 8 . 31 [88] and manually curated using Se-Al v2 [89] . For Fig 1 , the available complete polyprotein sequences representative of WNV diversity ( excluding lineage 6 , which there is only a partial sequence available ) were included to compare genetic percent identities . Likelihood mapping analyses for estimation of data quality were performed using Tree-Puzzle ( Quartets ranged between 10 , 000 and 40 , 000 ) [90 , 91] . For each alignment we performed recombination screening ( RDP , GeneConv , Chimaera , MaxChi , BootScan and SiScan ) in RDP4 . 61 [92] . The Bayesian phylogenetic analysis was performed using Bayesian Inference ( BI ) using a general time-reversible with gamma-distributed rate variation and invariant sites model ( GTR+Γ+I ) , as selected by Akaike's information criterion ( AICc ) in jModelTest 0 . 1 [93] . The evolutionary analysis was conducted assuming a relaxed Gamma clock and GMRF Bayesian Skyride coalescent tree prior . We then employed a Bayesian MCMC approach using BEAST v1 . 8 . 4 and performed five independent MCMC runs with up to 100 million generations to ensure the convergence of estimates . Trees were summarized in a maximum clade-credibility tree after a 10% burn-in [94] and used Tracer ( http://beast . bio . ed . ac . uk/Tracer ) to ensure convergence during MCMC by reaching effective sample sizes greater than 100 . To reduce the number of sequences from the original 862 downloaded from Genbank , a maximum likelihood tree was estimated using FastTree v2 . 1 . 7 [95] after identical alignment and curating methods . FastTree was run using GTR+Γ+I nucleotide model with 2000 Γ-rate categories , exhaustive search settings , with 5000 bootstrap replications using the Shimodaira-Hasegawa ( SH ) test . The analysis was repeated for the dataset of 95 sequences to compare tree topologies inferred by the Bayesian approach ( S1 Fig ) . All alignments referred to in this manuscript can be found at https://github . com/caiofreire . To perform this study and make it comparable with other studies [60 , 96] , viral stocks were standardized in number of plaque forming units per milliliter ( PFU/mL ) for cell infections rather than copy numbers of genome . The growth kinetics assays were performed in 12-well plates using one plate per virus strain with one uninfected well as a negative control . Each well was seeded with 2 . 4x105 Ap61 or Vero cells in a volume of 400 μl of appropriate medium and infected with 2 . 4x103 PFU ( plaque-forming unit ) of virus in 400 μl of medium , resulting in a multiplicity of infection ( MOI ) of 0 . 01 . After an incubation time of 4 hours , the medium was removed and replaced with 2 ml of new medium to set a zero point for the growth curves ( T0 ) . The harvesting of one well occurred at 22 , 28 , 50 , 75 , 99 , 124 , and 146 h post infection . Each harvest was performed as follows . Supernatants were removed and frozen at -80°C in small aliquots . Cells were washed once with phosphate-buffered saline ( PBS ) and then removed in 500 μl PBS . A volume of 20 μl of cell suspension was dried on a glass slide for a subsequent immunofluorescence assay as previously described [97] to measure viral proteins production . The remaining cell suspensions were frozen at -80°C . RNA was extracted from cell suspensions and supernatants and copy numbers of genome were quantified by real time RT-PCR as previously described [98] . Infectious viral particles were measured in supernatants by plaque assay also as previously described [99] . This study was performed two times on each cell type . The initial titers of lineages 1 , 2 , 7 and 8 were respectively 3x108 , 5x104 , 7 . 5x106 and 1010 PFU/ml . For each lineage , 2 . 4x103 PFU were used for kinetics in mosquito and mammal cells . The ratio of particles per infectious unit in the initial viral stocks ranged from 8 to 600 [98] . Our viral stocks had a similar ratio of particles per infectious unit as that seen produced by fully infectious extracellular WN virus particles [100] and mosquito-derived replicon WN virus particles [101] . Variances in replication efficiency between studies observed during in vitro infection could be explained to differences in the viral strain and to the infection conditions i . e . very low MOIs ( 0 . 01 ) , and distinct cell lines . Mice were produced in the Institut Pasteur de Dakar farm , located in Mbao , approximately 15 kilometers from Dakar , Senegal . After one week of acclimatization , five-to-six-week-old Swiss mice were challenged by intraperitoneal ( IP ) injection with 100 , 1000 and 10000 PFU of WNV lineages diluted in phosphate buffer saline + 0 . 2% endotoxin-free serum albumin ( BSA ) . For each lineage and dose , two independent experiments of infection were made . Each individual experiment had 4 to 8 mice . A group of mice inoculated in parallel with an equivalent volume of phosphate buffer saline + 0 . 2% endotoxin-free serum albumin ( BSA ) was maintained as a control . Mice were kept on clean bedding and given food and water ad libitum . Infected animals were monitored daily for first signs of encephalitis ( hunching , lethargy , eye closure , or hind legs paralysis ) and death throughout the 21 days after infection . All statistical inferences were calculated using the Wilcoxon rank sum test . To evaluate selection patterns on the complete coding sequences , we estimated the ratio of substitution rates ( ω ) per non-synonymous site ( dN ) over synonymous substitutions per synonymous site ( dS ) per codon sites . Briefly , sites with ω>1 are assumed to be under positive ( diversifying ) selection , and sites where ω<1 are undergoing negative ( purifying ) selection . When ω = 0 , the site is undergoing neutral selection . To estimate ω , we applied three maximum likelihood methods: single likelihood ancestor counting ( SLAC ) , fixed-effects likelihood ( FEL ) , and internal fixed-effects likelihood ( IFEL ) . We also investigated the presence of transient ( episodic ) selective pressures , using the mixed-effects model of evolution ( MEME ) [102] and fast , unconstrained Bayesian approximation ( FUBAR ) [103] approaches . For FEL , SLAC , IFEL , and MEME analyses , sites were identified as undergoing significant positive selection when p-value ≤0 . 10 . For FUBAR , sites were identified as undergoing positive selection when there was a posterior probability ≥0 . 90 . All estimations were implemented using HyPhy v2 . 11 [104] . Extraction of viral RNA from supernatants was performed with the QIAamp viral RNA mini kit ( Qiagen , Heiden , Germany ) according to manufacturer’s instructions . For cell fractions , prior to RNA extraction , cells were lysed by serial cycles of freeze/thaw . For the detection and quantification of viral RNA , a consensus WNV real-time RT-PCR assay and corresponding RNA standard were used as previously described [98] . The real-time PCR assays were performed using the Quantitect Probe RT-PCR Kit ( Qiagen , Heiden , Germany ) in a 96-well plate under the following conditions: 50°C for 10 min , 95°C for 15 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 min . Copy numbers of genome were calculated using Ct ( Cycle threshold ) and corresponding RNA standard . Overlapping RT-PCRs were done to recover the complete genome . All primer sequences can be found in the S3 Table . The NS5 , envelope and NS5-partial 3’UTR regions were first amplified using flavivirus consensus or West Nile specific primers [1 , 105 , 106] , followed by amplification of NS3 region using designed WNV primers . The 5’ non-coding region of the genome was obtained using the 5’RACE kit ( Invitrogen , Carlsbad , USA ) and a designed consensus primer in the capsid protein for reverse primer . Finally , specific primers were designed according to the first sequences obtained and a second step of RT-PCR was done to obtain the complete genome . The PCR fragments were obtained using AMV reverse transcription kit ( Promega , Madison , USA ) for reverse transcription and Go-Taq PCR kit ( Promega , Madison , USA ) for amplification . The RT conditions were set according to the manufacturer’s instructions , and the PCR conditions were as follows: 5 minutes at 95°C , 40 cycles of 1 minute at 95°C , 1 minute at 53°C , 1 to 4 minutes ( according the size of the PCR product ) at 72°C , and 10 minutes at 72°C . The PCR products were purified from the agarose gel using the Gel extraction kit ( Qiagen ) and sequenced by Cogenics ( Beckman Coulter Genomics , Essex , UK ) . Infected cells at different time points were dissolved in PBS and dropped on a glass slide . After complete drying , cells were fixed for at least 20 min in cold acetone , dried again , and then stored at -20°C until staining . Staining was done with a WNV-polyclonal mouse immune ascit diluted in PBS and incubated for 30 minutes at 37°C . After washing three times with PBS , cells were incubated with the second antibody ( goat anti-mouse IgG , fluorescein isothiocyanate [FITC] conjugated Biorad ) , diluted 1:40 and blue Evans 1/100 in PBS , for 30 minutes at 37°C in the dark . The cells were washed again three times with PBS , dried , and covered with 50% glycerol in PBS . After dehydration , examination was done by fluorescence microscopy . KJ831223 , FJ159131 , AY277251 , FJ159130 , FJ159129 , AY765264 , KY703856 , DQ176636 , HM147823 , FJ425721 , KT207791 , KJ934710 , KP780840 , KP780839 , KT359349 , KC496016 , KC407673 , KF179639 , KJ883346 , KC496015 , HQ537483 , KF647251 , KT207792 , JN858070 , KP109692 , KF179640 , KM203863 , KP780838 , KP780837 , KM203861 , KM203862 , JN393308 , EF429197 , EF429198 , HM147824 , EF429199 , KM052152 , EF429200 , GQ903680 , HM147822 , GQ851605 , DQ256376 , GQ851604 , JX041632 , GQ851602 , KT934796 , KT934801 , JX123031 , JX123030 , KT934800 , KT934802 , KT934803 , GQ851603 , KT934797 , KT934799 , KT934798 , JX041628 , JX041629 , JX041630 , HM051416 , KT163243 , EU249803 , KC601756 , JX442279 , JX041634 , KU588135 , JQ928175 , JN858069 , KF647253 , KC954092 , JQ928174 , JX556213 , JF707789 , FJ766331 , FJ766332 , JF719069 , FJ483549 , FJ483548 , JF719066 , JF719067 , KF234080 , GU011992 , JF719068 , DQ786573 , AY701413 , HM152775 , AY701412 , GQ851606 , GQ851607 , GQ379161 , AF196835 , KY703855 , EU082200 , KY703854 , AY532665 .	The West Nile virus ( WNV ) can cause severe neurological diseases including meningitis , encephalitis , and acute flaccid paralysis . Differences in WNV genetics could play a role in the frequency of neurological symptoms from an infection . For the first time , we observed how geographically similar but genetically distinct lineages grow in cellular environments that agree with the transmission chain of West Nile virus—vertebrate-arthropod-vertebrate . We were able to connect our in vitro and in vivo results with relevant epidemiological and molecular data . Our findings highlight the existence of West African lineages with higher virulence and replicative efficiency in vitro and in vivo compared to lineages similar to circulating strains in the United States and Europe . Our investigation of four West African lineages of West Nile virus will help us better understand the biology of the virus and assess future epidemiological threats .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "taxonomy", "invertebrates", "vero", "cells", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "pathogens", "biological", "cultures", "microbiology", "vertebrates", "animals", "viruses", "phylogenetics", "data", "management", "rna", "viruses", "phylogenetic", "analysis", "sequence", "motif", "analysis", "mammalian", "genomics", "insect", "vectors", "research", "and", "analysis", "methods", "sequence", "analysis", "infectious", "diseases", "computer", "and", "information", "sciences", "birds", "bioinformatics", "medical", "microbiology", "microbial", "pathogens", "viral", "replication", "evolutionary", "systematics", "cell", "lines", "disease", "vectors", "insects", "animal", "genomics", "arthropoda", "mosquitoes", "eukaryota", "west", "nile", "virus", "flaviviruses", "virology", "viral", "pathogens", "database", "and", "informatics", "methods", "genetics", "biology", "and", "life", "sciences", "species", "interactions", "genomics", "evolutionary", "biology", "amniotes", "organisms" ]	2017	Biological and phylogenetic characteristics of West African lineages of West Nile virus
Opisthorchis viverrini is considered among the most important of the food-borne trematodes due to its strong association with advanced periductal fibrosis and bile duct cancer ( cholangiocarcinoma ) . We investigated the relationship between plasma levels of Interleukin ( IL ) -6 and the risk of developing advanced fibrosis and bile duct cancer from chronic Opisthorchis infection . We show that IL-6 circulates in plasma at concentrations 58 times higher in individuals with advanced fibrosis than age , sex , and nearest-neighbor matched controls and 221 times higher in individuals with bile duct cancer than controls . We also observed a dose-response relationship between increasing levels of plasma IL-6 and increasing risk of advanced fibrosis and bile duct cancer; for example , in age and sex adjusted analyses , individuals with the highest quartiles of plasma IL-6 had a 19 times greater risk of developing advanced periductal fibrosis and a 150 times greater risk of developing of bile duct cancer than individuals with no detectable level of plasma IL-6 . Finally , we show that a single plasma IL-6 measurement has excellent positive predictive value for the detection of both advanced bile duct fibrosis and bile duct cancer in regions with high O . viverrini transmission . These data support our hypothesis that common mechanisms drive bile duct fibrosis and bile duct tumorogenesis from chronic O . viverrini infection . Our study also adds a unique aspect to the literature on circulating levels of IL-6 as an immune marker of hepatobiliary pathology by showing that high levels of circulating IL-6 in plasma are not related to infection with O . viverrini , but to the development of the advanced and often lethal pathologies resulting from chronic O . viverrini infection . Over 750 million people ( 10% of the human population ) are at risk of infection with a food-borne trematode , with more than 40 million people currently infected [1] . Opisthorchis viverrini is considered among the most important of the food-borne trematodes due to its strong association with hepatobiliary pathologies that include advanced bile duct fibrosis ( periductal fibrosis or APF ) [2]–[5] and cholangiocarcinoma ( CCA ) [3] , [6]–[11] . In Thailand , an estimated 10 million people are infected with O . viverrini [12] , where uncooked cyprinoid fish ( the intermediate host for the parasite ) are a staple of the diet [13] . While infection with O . viverrini can be eliminated by chemotherapy ( the anthelmintic praziquantel ) , culinary practices in Thailand result in rapid and prolonged re-infection after treatment . In the Northeastern region of Thailand ( Isaan ) , individuals are often infected with O . viverrini for a lifetime [7] , [14] , [15] . As shown in our community-based ultrasound studies along the Chi River basin in Isaan ( Khon Kaen , Thailand ) [16] , [17] , important pathogenic changes occur early and asymptotically to the bile duct during O . viverrini infection , including fibrosis in the intrahepatic bile duct ( periductal fibrosis ) . As with other forms of hepatic fibrosis , periductal fibrosis from chronic opisthorchiasis is probably the result of repeated injury sustained by the biliary epithelium from a combination of the mechanical , toxic , and immune mechanisms of the fluke in the bile duct [for review see [18]] . As individuals are infected with O . viverrini for many years ( often a lifetime ) , a persistent cycle of tissue damage and repair takes place in the intrahepatic biliary ducts , creating a chronic inflammatory milieu that stimulates periductal fibrogenesis [for review of this process see [18] , [19]] . In both the animal and human models of chronic O . viverrini infection , this fibrotic deposition along the biliary epithelium is a precursor event to CCA . For example , during the final phase of O . viverrini infection in the hamster model ( at 12 weeks after infection ) , the inflamed biliary epithelium manifests fibrosis throughout its length , which is followed rapidly by tumorogenesis [20] , [21] . In human autopsy studies , extensive fibrotic deposition in the intrahepatic bile duct routinely accompanies CCA tumors [10] , [22] , [23] . Although the exact mechanism has yet to be determined , advanced fibrotic lesions may precede tumorogenesis in the bile duct by producing a “smoldering and chronic inflammatory milieu” [24] , [25] . A key factor in the maintenance of this chronic inflammatory milieu is the production of soluble growth factors such as the cytokine IL-6 [24] , [25] . Studies of other hepatic pathogens also strongly implicate IL-6 in progressive pathogenic fibrosis and carcinogenesis , including hepatocellular fibrosis and hepatocellular carcinoma ( HCC ) from Hepatitis B ( HBV ) and Hepatitis C Virus ( HCV ) infection [26] , [27] . Due to its role in systemic inflammation , IL-6 is readily detected in plasma [28] . As such , we sought to determine if the concentration of IL-6 in the plasma of O . viverrini infected individuals with APF and CCA was higher than age , sex , and nearest-neighbor matched controls infected with O . viverrini but without these advanced pathologies . We also sought to determine the sensitivity and specificity of a single measurement of plasma IL-6 to detect APF or CCA in these same individuals . Given the poor prognosis associated with CCA , especially in resource-poor settings such as Thailand , an early marker for the risk of the hepatobiliary pathologies related to O . viverrini infection is urgently needed . The current study presents baseline data collected from a community–based cohort study of the risk factors associated with the development of Advanced Periductal Fibrosis ( APF ) from chronic opisthorchiasis ( Table 1 ) . A detailed description of this population and the methods can be found in [17] . Individuals from seven villages with high O . viverrini transmission , including Nongnangkwan , Nongmuang , Loopka , Nongkham , Nongno , Lawa and Chikokor in the vicinity of Khon Kaen ( Thailand ) were surveyed . From this group of O . viverrini-infected individuals , 184 males and 236 females between the ages of 20 and 60 years ( inclusive ) were enrolled into the study as “cases” or age , sex , and nearest neighbor matched “controls” . Individuals were enrolled as “cases” on the basis of an ultrasound ( US ) in which APF was determined [17] . Individuals who were positive for O . viverrini but negative for APF by US were defined as “controls” and matched with cases by age ( ten year age intervals ) , sex , and residence in the same village ( nearest-neighbor method ) . A positive pregnancy test excluded female volunteers from participation in the ultrasonography and blood draw . As such , 210 individuals were identified as cases and matched 210 individuals as controls . Both cases and control were asked to provide 30 ml of blood for baseline immunology and hematological parameters . Individuals positive for O . viverrini were referred to the local public health outpost for treatment with praziquantel . All subjects provided written informed consent using Informed Consents Forms approved by the Ethics Committee of Khon Kaen University School of Medicine , Khon Kaen , Thailand ( reference number HE480528 ) and the Institutional Review Board of the George Washington University School of Medicine , Washington , D . C ( GWUMC IRB# 020864 ) . Aliquots of plasma from 121 cases of histologically proven , O . viverrini associated CCA cases were pulled from the biological repository of the Liver Fluke and Cholangiocarcinoma Research Center , Faculty of Medicine , Khon Kaen University , Thailand ( Table 1 ) . Of these pulled samples , 83 were male and 38 were female . All samples were from patients who had liver resection surgery as a part of palliative care for O . viverrini-associated CCA at the Khon Kaen University Srinagarind Hospital , Khon Kaen , Thailand . A detailed description of the ultrasonography methods used in this study can be found in the following references [16] , [17] . Briefly , a mobile , high-resolution ultrasound ( US ) machine ( GE model LOGIQ Book XP ) was used . Hepatobiliary abnormalities including portal vein radical echoes , echoes in liver parenchyma , indistinct gallbladder wall , gallbladder size , sludge and suspected CCA were graded and recorded as previously described [16] , [17] . Individuals were classified as “Non-Advanced Periductal Fibrosis” or “controls” if the US grade was 0 or 1 , and “Advanced Periductal Fibrosis” or “case” if the US grade was 2 or 3 as described in detail in our previous study [16] , [17] . Individuals with alcoholic liver disease , which is seen as fatty liver by US exam , were excluded from the analysis component of the study . Also , individuals with marked hepatic fibrosis not related to O . viverrini infection ( e . g . , cirrhosis from HBV or HCV ) were also excluded from the analysis component study . Eggs/parasite identification and egg counts were performed by certified medical technicians using light microscopy at 10× and 40× magnifications as described in detail in our previous study [16] . Of the approximately 30 ml of blood collected from participants , 8 ml of blood were collected in heparinized tubes for the measurement of plasma cytokines . Plasma aliquots were grouped in case-control sets that were handled together throughout the processing and biochemical analysis . After blood draw , plasma was separated and immediately aliquoted into 500 microliter ( mL ) cryogenic tubes and frozen at −80°C until use . Biochemical analysis was done on all samples simultaneously , with the position of plasma samples varying at random during the processing . Plasma samples from CCA patients were taken at the time of liver resection , separated and aliquoted as above , and then stored at −80°C until use . CCA plasma samples were also randomly interspersed among the cases and control sample sets during biochemical analysis . The 21 non-endemic O . viverrini negative control plasma were handled as above and interspersed repeatedly at random among the actual case-control samples during biochemical analysis . Cytokine levels in plasma were examined for IL-1β , IL-2 , IL-4 , IL-5 , IL-6 , IL-8 , IL-10 , IL-12p70 , interferon ( INF ) γ , tumor necrosis factor ( TNFα ) , and TNFβ production . The quantification of plasma cytokines was analyzed using commercial FlowCytomix bead-based multiplexing assays kits ( Beckman-Coulter ) . Values were quantified from standard curves using human recombinant cytokines . A quantile regression model was used to determine the median along with 95% confidence intervals ( 95% CI ) to estimate the differences between cases and controls for plasma cytokine levels . When the data suggested a significant difference between cases and controls , such as the levels of plasma IL-6 , the plasma were analyzed again using a sandwich ELISA as described below . IL-6 cytokine levels in the plasma were measured by sandwich enzyme-linked immunoadsorbent assay ( ELISA ) using a DuoSet ( R & D systems , Inc ) according to the manufacturer's instruction . The level of IL-6 was determined by interpolating the Optical Density of sample duplicates into a 4-parameter logistic-log model of a standard curve of recombinant human IL-6 run in serial dilutions on each ELISA plate . The concentration of IL-6 level is expressed as picograms per microliter . The percent distribution of selected demographic characteristics were calculated for the four study groups: ( 1 ) non-endemic negative controls , ( 2 ) O . viverrini positive ( OV+ ) and APF negative ( controls ) , ( 3 ) OV+ APF positive ( APF cases ) , and ( 4 ) OV+ and CCA+ ( CCA cases ) . Box and whisker plots display the distribution of plasma IL-6 according to these groups . Age and sex adjusted Odds Ratios ( OR ) and 95% Confidence Intervals ( CIs ) for quartiles of plasma IL-6 concentration and their association with APF or CCA status were determined using age and sex adjusted multiple logistic regression analyses . A chi-square test for trend was also used to test the effect of increasing quartile level of plasma IL-6 on increasing risk of APF or CCA . The significance level for all tests was set at 0 . 05 , with a Bonferroni correction for multiple testing . All statistical tests were two-sided . All analyses were performed using Stata version 10 ( College Station , TX ) . Receiver-operating-characteristic ( ROC ) curves were obtained by plotting the sensitivity versus 1–specificity for the full range of IL-6 cut-off points in picograms per milliliter ( pg/mL ) for both cases and controls to estimate the cutoff value that had the highest overall validity . “Sensitivity” was calculated as the number of individuals positive for APF or CCA testing positive for plasma IL-6 at various cutpoints ( by pg/mL ) divided by the total number of cases . “Specificity” was calculated as the number of individuals negative for APF and CCA ( controls ) testing negative for plasma IL-6 at various cutpoints ( by pg/mL ) divided by the total number of controls . The area under the ROC curve was calculated by determining the probability of correctly identifying ( accuracy ) a randomly selected participant as either a case ( APF positive or CCA positive ) or a non-case ( APF negative or CCA negative ) . The 45-degree line in each ROC curve graph subsumes an area equal to 0 . 5 ( 50% ) , which is equivalent to using a coin toss procedure to classify participants as either cases or controls . Using the ROC curves , an optimal cutpoint was determined for the concentration of plasma IL-6 that maximized the sensitivity and specificity in classifying an individual at APF+ or CCA+ . Based on this cutoff-point , the positive predictive value ( PPV ) , and negative predictive value ( NPV ) of plasma IL-6 concentration to detected APF+ or CCA+ status were determined . All analyses were performed using Stata version 10 ( College Station , TX ) . As part of the inclusion criteria of the study , both cases and controls had to be positive for O . viverrini as determined by the presence of at least a single parasite ovum in feces . The baseline characteristics of the enrolled participants are shown in Table 1 , with 210 cases ( O . viverrini+ and APF+ ) and 210 controls ( O . viverrini+ and APF− ) . The controls were matched with cases by age ( ten year age intervals ) , sex , infection ( O . viverrini positive ) and nearest-neighbor status ( same village ) . The study sample included more females ( 56 . 2% ) than males ( 43 . 8% ) ( P<0 . 001 ) . The mean ages for cases ( 46 . 6 years of age ) and controls ( 46 . 6 years of age ) did not differ significantly ( P = 0 . 999 ) . No statistically significant difference ( P>0 . 811 ) was observed in the intensity of O . viverrini infection between cases ( median = 142 epg ) and controls ( median = 130 epg ) . As shown in Table 1 , both case and control groups had a similar distribution of individuals when age was stratified by ten-year intervals . Among the 121 CCA samples in this study ( Table 1 ) , 117 ( 96 . 7% ) were hepatectomies and 4 ( 3 . 3% ) were small biopsy specimens . Of the 117 hepatectomies , 74 ( 63 . 2% ) were of the mass-forming type and 8 ( 6 . 8% ) , 21 ( 18% ) , and 14 ( 12% ) were periductal-infiltrating , invasive intraductal , and mixed types , respectively . Histologically , there were 41 well-differentiated ( 33 . 9% ) , 8 moderately differentiated ( 6 . 6% ) , 9 poorly differentiated adenocarcinomas ( 7 . 4% ) , 61 papillary carcinomas ( 50 . 4% ) , and 2 adenosquamous carcinomas ( 1 . 6% ) of the 121 CCA cases studied . Of the 11 cytokines tested only levels of the inflammatory cytokine IL-6 were significantly elevated in individuals with advanced periductal fibrosis and CCA compared to controls . The frequency distributions of plasma IL-6 concentration for all groups are presented in Figure 1 . The IL-6 concentrations in plasma for both cases and controls ranged between <0 . 01 pg/mL ( undetectable ) and 538 . 6 pg/ml , with a much wider range in APF cases ( <0 . 01 to 538 . 6 pg/ml ) and CCA cases ( <0 . 01 to 536 . 2 pg/ml ) than APF- controls ( <0 . 01 to 173 . 1 pg/ml ) . An even narrower range of plasma IL-6 concentration was recorded for the non-endemic controls ( <0 . 01 to 91 . 4 pg/ml ) . These data indicate that there is far wider variation for the concentration of plasma IL-6 in the advanced forms of O . viverrini induced pathogenesis than in individuals with O . viverrini infection but no O . viverrini associated pathology . Moreover , the median levels of plasma IL-6 concentration were 58 times higher in APF+ cases than in controls ( 58 versus 1 pg/ml; P<0 . 001 ) and 221 times higher in CCA cases than in controls ( 221 versus 1 pg/ml; P<0 . 001 ) . CCA cases also had higher levels of plasma IL-6 than APF cases ( 221 versus 58 pg/ml; P<0 . 001 ) . Non-endemic controls had significantly lower ( mostly undetectable or <0 . 01 pg/mL ) plasma IL-6 concentrations ( P<0 . 0001 ) . Increasing plasma concentrations of IL-6 significantly increased the risk of APF: for every 10 pg/ml increase of IL-6 , there is an increase in the risk of APF by 22% ( OR = 1 . 22; 95% CI 1 . 16 to 1 . 29 , P<0 . 001 ) in a model adjusted for age and sex ( data not shown ) . As shown in Table 2 , the risk of APF increased with increasing quartile concentration of plasma IL-6: OR = 1 . 00 for Quartile 1 ( reference quartile ) ; OR = 3 . 94 ( 95% CI 2 . 21 to 6 . 81 ) for Quartile 2; OR = 7 . 95 ( 95% CI 4 . 73 to 13 . 36 , P = <0 . 001 ) for Quartile 3; and OR = 18 . 94 ( 95%CI 10 . 17 to 35 . 25 , P = <0 . 001 ) for Quartile 4 . A strong and significant trend for increasing quartile concentration of IL-6 and increasing risk of APF was found compared to individuals with no detectable levels of plasma IL-6 ( P<0 . 001 ) . Elevated plasma concentrations of IL-6 significantly increased the risk of CCA: for every 10 pg/ml increase of plasma IL-6 concentration , there was a 26% ( OR = 1 . 26; 95% CI 1 . 19 to 1 . 34 , P<0 . 001 ) increase in the risk of CCA in a model adjusted for age and sex . Table 3 shows that the risk of CCA increased in the 3rd and 4th highest quartile concentrations of plasma IL-6: OR = 4 . 55 ( 95% CI 2 . 05 to 10 . 11 , P = <0 . 001 ) for Quartile 3 and OR = 149 . 11 ( 95% CI 40 . 42 to 550 . 15 , P = <0 . 001 ) for Quartile 4 . A strong and significant trend ( P<0 . 001 ) for increasing quartile concentration of IL-6 and increasing risk of CCA was also found compared to individuals with no detectable level of plasma IL-6 ( <0 . 01 pg/mL ) . Note that the term “control” refers to the age , sex , and nearest-neighbor matched controls for APF group and not for the CCA group and that the logistic model is adjusted for age and sex . Figure 2 shows the ROC curve obtained by plotting the True Positive Probability ( sensitivity ) against the False Negative Probability ( 1–specificity ) for the entire range of IL-6 cut-off points to predict the presence of APF . Using a cutoff of greater than 11 pg/mL of plasma IL-6 , the sensitivity for the detection of APF was 80% and the specificity was 74% ( Table 4 ) . In addition , for the cutoff of greater than 11 pg/mL of plasma IL-6 , the area under the ROC curve , which is an indication of the “accuracy” of the test or proportion of all tests that have given the correct result , was 78% ( 95%CI 74% to 83% ) . Table 4 also shows the Positive Predictive Value ( PPV ) and Negative Predictive Value ( NPV ) of 76% and 79% , respectively , for the detection of APF using plasma IL-6 levels greater than 11 pg/mL . Figure 3 shows the ROC curve obtained by plotting the True Positive Probability ( sensitivity ) against the False Negative Probability ( 1–specificity ) for the entire range of IL-6 cut-off points to predict the presence of CCA . Using a cutoff of greater than 64 pg/mL of plasma IL-6 , the sensitivity for the detection of CCA was 80% and the specificity was 90% ( Table 4 ) . In addition , for the optimal cutoff of >64 pg/mL of plasma IL-6 , the area under the ROC curve , which is an indication of the “accuracy” of the test , was 89% ( 95% CI = 85% to 93% ) . Table 4 also shows the Positive Predictive Value ( PPV ) and Negative Predictive Value ( NPV ) of 82% and 88% , respectively , for the detection of APF using plasma IL-6 levels >64 pg/mL . Note that the term control refers to the age , sex , and nearest neighbor-matched controls for APF group and not for the CCA group . These data show that elevated plasma concentrations of IL-6 are associated with a marked and significant increase in the risk of O . viverrini-associated APF and CCA . As shown in Tables 2 and 3 , increasing levels of plasma IL-6 associate with increasing risk of these advanced pathologies in a dose-dependent manner: for example , individuals with the highest quartiles of plasma IL-6 concentration had a 19 times greater risk of developing APF and a 150 times greater risk of developing CCA than individuals with undetectable levels of plasma IL-6 ( <0 . 01 g/mL ) . As the data were collected by a cross-sectional study design , the complicity of IL-6 in these pathogenic processes remains to be determined: that is , elevated plasma concentrations of IL-6 may simply reflect the presence of these hepatobiliary abnormalities or , as shown in a number of other studies , elevated plasma concentrations of IL-6 may play a key role in these pathogenic processes by creating an inflammatory milieu that favors fibrotic deposition and carcinogenesis in the bile duct [for reviews see [28]–[30]] . Notably , our data show that O . viverrini infection alone does not elevate IL-6 levels circulating in the plasma: e . g . , age and sex matched O . viverrini infected individuals without APF or CCA had negligible levels of IL-6 in their plasma . It is only in the presence of advanced pathology from chronic opisthorchiasis that plasma concentrations of IL-6 are significantly elevated . This study also shows that plasma IL-6 concentration can be used to detect the risk for the advanced pathologies associated with O . viverrini infection , many of which are subclinical . Our data on the predictive values of a single measurement of plasma IL-6 to detect CCA closely resemble recent data from Cheon and co-workers [31] as well as other studies that have also reported a high sensitivity and specificity for IL-6 in serum for non-Opisthorchis-associated CCA [see [32] for review] . Our study adds a unique aspect to the literature on circulating levels of IL-6 as an immune marker of hepatobiliary pathology by showing that high levels of circulating IL-6 in plasma are not related to infection with O . viverrini , but to the development of the advanced and often lethal pathologies resulting from chronic O . viverrini infection: i . e . , age and sex matched controls with O . viverrini infection and no pathology ( APF- and CCA- ) had undetectable levels of this inflammatory cytokine in their plasma . Moreover , the ability of a single IL-6 measurement to detect risk for O . viverrini associated CCA and O . viverrini associated pathogenic fibrosis in the bile duct ( APF ) is especially important in regions where O . viverrini is endemic . In Isaan , Thailand , for example , the prevalence of O . viverrini infection can reach as high as 79%; hence , an easily accessible immune marker which can distinguish between infection with O . viverrini and the advanced pathology induced by this parasite would be particularly useful , where the incidence of intraheptic CCA is among the highest in the world [33] . Currently , Thai individuals in O . viverrini endemic areas are diagnosed only in the most advanced stages of CCA , when treatment is essentially palliative . An easily measurable biomarker for a precursor stage to CCA such as APF would be of great public health importance by identifying those at greatest risk of CCA . IL-6 is produced by numerous cell types , with a broad range of effects , including the promotion of the innate immune response to pathogens as well as the subsequent chronic inflammatory reactions [28] , [30] , [34] . Once known as “hepatocyte-stimulating factor” ( HSF ) , IL-6 has been shown to play a key role in chronic inflammatory conditions of the liver that lead to fibrotic lesions , e . g . , chronic alcohol consumption or viral hepatitis [26] , [35] . Our previous studies have shown that crude antigens extracts antigens from this parasite stimulate host PBMC from O . viverrini infected individuals to produce high levels of IL-6 [16] . Our new data indicate that elevated levels of IL-6 are not confined to the immune response to parasite antigens , but circulate at elevated levels in the plasma of individuals with advanced pathologies associated with chronic O . viverrini infection , a role consistent with the reputation of IL-6 as a key player in systemic inflammation . We speculate that elevated concentrations of IL-6 circulating in the plasma reflect the constant tissue repair response of the biliary epithelia to the chronic mechanical , toxic , and immune injury induced by the fluke [18] . When these tissue repair mechanisms are activated transiently , the normal hepatobiliary structure and function rapidly recovers: i . e . , cholangiocytes would regenerate and replace the necrotic or apoptotic cells , a process associated with a local and transitory IL-6 response . However , as with chronic alcohol consumption or infection with either HBC or HCV infection , chronic opisthorchiasis creates a persistent inflammatory milieu that stimulates fibrotic deposition [25] , [29] , [36] , a process that could be reflected or even partially induced by the elevated concentrations of IL-6 circulating in the plasma . As with other inflammation associated hepatic pathologies [25] , [29] , [36] , [37] , tumors in the biliary tract may have activated the wound-healing programs of the host chronically infected with O . viverrini in an exaggerated and prolonged manner . A parallel scenario is the presence of pathogenic fibrosis before the development of HCC in patients with chronic HCV [for review see [27]] . In the case of HCC , it is hypothesized that the development of fibrosis and/or cirrhosis , plus a microenvironment conducive to genomic instability , promotes neoplastic transformation in areas of hepatic fibrosis . Prominent among these processes is the production of soluble growth factors such as IL-6 [25] , [29] , [36] . In the hamster animal model of O . viverrini-induced CCA , the constant injury to bile duct epithelium by the liver fluke engages a similar process [11] . During the early stage of infection ( 4 weeks ) , fluke injury to the biliary epithelium results in inflammatory cell infiltration , hyperplasia , and in the adenomatous change in the bile duct epithelium . By the chronic stage of the infection ( 12 weeks ) , the immune response has transformed into prolonged inflammation with fibrotic deposition along the bile duct wall and ultimately CCA [18] , [38]–[41] . In autopsy studies [22] , [23] and studies on liver tissue resected from Opisthorchis-induced CCA patients for palliative care [5] , [18] , fibrosis is routinely detected proximal to neoplastic bile duct tissue . In summary , this study identifies a significant relationship between plasma IL-6 concentration and the advanced pathologies associated with chronic O . viverrini infection . While in other settings ( e . g . Western countries ) the causative agent of CCA remains obscure , the single most important risk factor for intrahepatic CCA in Thailand has long been established–infection with the liver fluke O . viverrini [9] . As such , an easily accessible biomarker such as plasma IL-6 would have great utility in predicting those at risk ( APF ) or already with early CCA in a setting , where half of the population are routinely exposed to this class I carcinogen O . viverrini through the daily consumption of raw fish [42] .	O . viverrini is among the few parasites considered a Class 1 carcinogen because of its strong association with bile duct cancer ( cholangiocarcinoma ) . Currently , more than 40 million people are infected with O . viverrini worldwide . Thailand has the highest prevalence of O . viverrini at 10 million people infected and also the highest incidence of Opisthorchis-associated bile duct cancer ( cholangiocarcinoma ) in the world . In the current study , we also show that levels of IL-6 in plasma are associated in a dose-dependent fashion with Opisthorchis-induced bile duct fibrosis ( periductal fibrosis ) and cholangiocarcinoma . More importantly , we show that O . viverrini infection alone does not elevate IL-6 levels in the plasma . It is only in the presence of these advanced pathologies ( advanced fibrosis or cholangiocarcinoma ) from chronic O . viverrini infection that significantly elevates plasma levels of IL-6 are observed . Moreover , we show that plasma IL-6 is an easily accessible biomarker for the detection of advanced periductal fibrosis and cholangiocarcinoma , which would be a critical advance for this region of Thailand and other countries in Southeast Asia , where the prevalence of O . viverrini infection can reach as high as 80% and the incidence of bile duct cancer is the highest in the world .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine", "infectious", "diseases", "public", "health", "and", "epidemiology", "diagnostic", "medicine", "clinical", "immunology", "clinical", "research", "design", "epidemiology", "immunology", "biology", "public", "health" ]	2012	Elevated Plasma IL-6 Associates with Increased Risk of Advanced Fibrosis and Cholangiocarcinoma in Individuals Infected by Opisthorchis viverrini
Dialysis patients with chronic renal failure receiving deferoxamine for treating iron overload are uniquely predisposed for mucormycosis , which is most often caused by Rhizopus oryzae . Although the deferoxamine siderophore is not secreted by Mucorales , previous studies established that Rhizopus species utilize iron from ferrioxamine ( iron-rich form of deferoxamine ) . Here we determined that the CBS domain proteins of Fob1 and Fob2 act as receptors on the cell surface of R . oryzae during iron uptake from ferrioxamine . Fob1 and Fob2 cell surface expression was induced in the presence of ferrioxamine and bound radiolabeled ferrioxamine . A R . oryzae strain with targeted reduced Fob1/Fob2 expression was impaired for iron uptake , germinating , and growing on medium with ferrioxamine as the sole source of iron . This strain also exhibited reduced virulence in a deferoxamine-treated , but not the diabetic ketoacidotic ( DKA ) , mouse model of mucormycosis . The mechanism by which R . oryzae obtains iron from ferrioxamine involves the reductase/permease uptake system since the growth on ferrioxamine supplemented medium is associated with elevated reductase activity and the use of the ferrous chelator bathophenanthroline disulfonate abrogates iron uptake and growth on medium supplemented with ferrioxamine as a sole source of iron . Finally , R . oryzae mutants with reduced copies of the high affinity iron permease ( FTR1 ) or with decreased FTR1 expression had an impaired iron uptake from ferrioxamine in vitro and reduced virulence in the deferoxamine-treated mouse model of mucormycosis . These two receptors appear to be conserved in Mucorales , and can be the subject of future novel therapy to maintain the use of deferoxamine for treating iron-overload . Mucormycoses , caused by fungi in the order Mucorales , are life threatening infections that afflict immunosuppressed patients due to neutropenia , corticosteroids treatment , hyperglycemia , and trauma [1 , 2] . These relatively uncommon infections , mainly caused by Rhizopus spp . , have been steadily increasing in numbers for the last three decades [3 , 4] . Despite the current aggressive treatment options against mucormycosis which constitutes reversal of immunosuppressive predisposing factors , surgical removal of infected foci ( when possible ) and antifungal therapy , overall mortality remains at >50% and approaches 100% for patients with brain involvement , prolonged neutropenia and hematogenously disseminated disease [5 , 6 , 7 , 8] . Clearly , novel strategies to prevent and/or treat the disease are critically needed . Clinical observations strongly link the ability of organisms to obtain iron from the host as an essential virulence factor of Mucorales [9 , 10] . For example , hyperglycemia , diabetic ketoacidosis and other forms of acidosis predispose the host to mucormycosis because of compromised ability of transferrin to chelate iron , thereby making iron available to invading organisms [9 , 11 , 12] . Similarly , dialysis patients with chronic renal failure who suffer from iron overload due to blood transfusion were historically treated with the bacterial siderophore , deferoxamine [13 , 14 , 15] . These patients were uniquely susceptible to lethal form of mucormycosis [13 , 15 , 16] . Although deferoxamine is an iron-chelator from the perspective of the human host , Rhizopus spp . utilize ferrioxamine ( deferoxamine + Fe3+ ) as a xenosiderophore to obtain previously unavailable iron [17 , 18] . It was also reported that Rhizopus obtains iron from ferrioxamine by intracellular transport of the reduced iron without deferoxamine internalization [18] . Recently , we found treatment of Rhizopus-infected mice with the iron chelators deferiprone [19] or deferasirox [20] ( which are not utilized as xenosiderophores by Rhizopus ) markedly improved survival . These results further confirm the unique importance of iron in the pathogenesis of mucormycosis . In the current study , we sought to identify the fungal cell surface protein that binds to ferrioxamine and its role in the pathogenesis of mucormycosis . We provide evidence that the ferrioxamine binding ( Fob ) cell surface proteins ( namely Fob1 and Fob2 ) are the fungal receptors that mediate attachment to ferrioxamine , thereby facilitating fungal iron uptake from this xenosiderophore via the reductase/high affinity iron permease pathway . Importantly , Fob1 and Fob2 are inducible proteins that are required for full virulence of R . oryzae only in the deferoxamine-treated mouse model of mucormycosis . To isolate putative ferrioxamine receptor ( s ) , R . oryzae protoplasts were allowed to regenerate in the presence or absence of ferrioxamine , which is the iron-rich form of the bacterial siderophore , deferoxamine [21] . SDS-PAGE analysis of proteins demonstrated the presence of a major band at ~ 70 kDa in cell-free supernatants concentrated from R . oryzae regenerating protoplasts in the presence , but not in the absence , of ferrioxamine . Other minor bands were also detected around 40kDa in supernatants from medium with or without ferrioxamine ( Fig 1A ) . For a protein to act as a receptor it must bind to ferrioxamine . Therefore , we incubated the protein preparations collected from these supernatants with radiolabeled ferrioxamine ( deferoxamine+55Fe ) prior to running on a non-denaturing PAGE followed by autoradiography . A band from proteins concentrated in the supernatant collected from protoplasts regenerated in the presence of ferrioxamine , but not in the absence of ferrioxamine , bound to radiolabeled ferrioxamine ( Fig 1B ) . Because the 70 kDa band was abundant in the ferrioxamine-containing supernatant and absent in the supernatant concentrated from media lacking ferrioxamine , we sequenced this band by MALDITOF MS/MS . The overwhelming indentified protein was predicted to be encoded by RO3G_11000 ( 352 amino acid ) , which is annotated as a CBS-domain-containing protein ( http://www . broadinstitute . org/ , see Discussion for CBS-domain protein definition ) in the R . oryzae 99-880 database ( also known as R . delemar ) ( Table 1 ) . We also found that the ORF RO3G_11000 shares ~80% identity with ORF RO3G_05087 ( 350 amino acid ) at the DNA or protein levels . The RO3G_05087 and RO3G_11000 ORFs are predicted to encode proteins with ~ 39 kDa in size and were named FOB1 , and FOB2 , respectively . These predicted proteins have 4 putative CBS domains and multiple possible N- and O-glycosylation sites ( Fig 1C ) . Interestingly and despite the presence of Fob2 protein in the supernatant of regenerated protoplasts , we did not observe classic features of secreted proteins such as N-terminal signal peptide or C-terminus GPI-anchor sequence . However , using the MEMSAT program it was predicted that Fob1 and Fob2 proteins have one to two transmembrane domains with an extracellular fragment and a cytoplasmic tail ( S1 Fig ) . Further localization analysis using antibody staining confirmed the surface localization of Fob proteins ( see below ) . We also examined whether this family of genes was present in other Mucorales known to cause mucormycosis . Blast search of FOB1 and FOB2 confirmed the presence of orthologs of FOB1/2 in every Mucorales genome published thus far with percent identity at the amino acid levels ranging from 42% for Mortierella verticillata FOB1 ortholog to 78% for Mucor circinelloides f . circinelloides FOB2 ( Table 2 ) . Orthologs from other fungi were also found with lesser degree of identity ( e . g . ~ 20% for Aspergillus , Saccharomyces cerevisiae , and Candida ) . Because radiolabeled ferrioxamine bound to a band in protein preparations from supernatants collected from regenerated protoplasts in the presence of ferrioxamine and not in the siderophore’s absence ( Fig 1B ) , we reasoned that the ferrioxamine receptor is likely induced by ferrioxamine . To test if FOB2 and its FOB1 homologue fulfill this criterion , we cultured R . oryzae 99-880 in medium containing ferrioxamine , FeCl3 , deferoxamine , rhizoferrin ( a siderophore produced by Rhizopus [18 , 22] ) , or Fe3+ containing rhizoferrin as the sole source of iron prior to analyzing the expression of FOB1 and FOB2 by qRT-PCR . As can be seen in Fig 2A , only the iron-rich ferrioxamine consistently enhanced the expression of FOB1 and FOB2 by ~ 9 fold vs . any other condition . Neither the iron depleted ferrioxamine ( i . e . deferoxamine ) , nor rhizoferrin with or without iron induced the expression of either FOB1 or FOB2 . However , the presence of FeCl3 as a source of iron instead of ferrioxamine resulted in a modest increase in the expression of FOB2 . If FOB2 and/or FOB1 indeed encode a ferrioxamine receptor , the lack of their expression by deferoxamine indicated that iron uptake from radiolabeled ferrioxamine should not be inhibited by deferoxamine but rather by cold ferrioxamine . To test this possibility , iron uptake of 55Fe3+ from ferrioxamine by R . oryzae was carried out at increased concentrations of cold ferrioxamine or deferoxamine . Cold ferrioxamine , but not deferoxamine , inhibited 55Fe3+ uptake by >70% when introduced at 3–10 fold higher concentrations than the radiolabeled ferrioxamine ( Fig 2B ) . Because FeCl3 had a modest effect on the expression of FOB1 and FOB2 when compared to ferrioxamine , while deferoxamine had no effect on their expression , this implicated ferric iron as a critical signal for triggering the expression of these two genes . We studied the expression of FOB1 and FOB2 in response to varying concentrations of FeCl3 . As can be seen in Fig 2C , FeCl3 , modestly enhanced the expression of both genes in a non-linear manner . Collectively , these results show that the expression of FOB1 and FOB2 is maximal only when the siderophore is iron–rich ( i . e . ferrioxamine ) , the iron-depleted form ( i . e . deferoxamine ) does not induce nor it inhibits iron uptake from ferrioxamine , and the native siderophore of rhizoferrin does not induce the expression of these two genes . To study if enhancement in FOB1 and FOB2 gene expression resulted in enhanced protein expression on R . oryzae cell surface , we raised antibodies against an E . coli-produced Fob2 protein ( rFob2p ) by vaccinating mice and used these polyclonal antibodies to detect the expression of Fob proteins on R . oryzae cell surface by FACS analysis . Sera collected from mice vaccinated with rFob2p + adjuvant had an antibody titer of 1:32 , 000 vs . a titer of 1:1600 of sera collected from mice vaccinated with the adjuvant alone . Additionally , sera collected from rFob2p vaccinated mice reacted to the E . coli produced Fob2 protein as well as a protein band isolated from whole cell protein extract of R . oryzae wild-type hyphae only when grown in the presence of ferrioxamine . These two bands were detected at ~40 kDa with slight difference in size which is likely attributed to the presence of His-tag on the rFob2p . Also the antibodies did not detect any bands from whole cell extracts of R . oryzae strain in which both FOB1 and FOB2 were down regulated regardless of the presence or absence of ferrioxamine in the growth medium ( see below for FOB1/FOB2 inhibition by RNA-i ) ( Fig 3A for Western blot and S2A Fig for SDS-PAGE gel picture ) . Analysis of the recognized bands by nano-LC ESI MS/MS confirmed the identity of the proteins to Fob2p ( Table 3 ) . Interestingly , we could not detect the protein in cell-free supernatants collected from media containing R . oryzae wild-type or FOB1/FOB2 inhibition mutant grown in the presence or absence of ferrioxamine ( Fig 3B for Western blot , and S2B Fig for SDS-PAGE gel picture ) . Furthermore , radiolabeled ferrioxamine reacted to crude cell protein extracts or proteins immunoprecipitated by anti-Fob2 antibodies from R . oryzae wild-type whole cell extracts only when grown in a medium containing ferrioxamine ( Fig 3C ) . Finally , the immunoprecipitated band from the whole cell extract of R . oryzae wild-type cells as well as the immunoprecipitated band from rFob2p had sizes of around 40 kDa ( Fig 3D for Western blot and S2C Fig for SDS-PAGE gel picture ) similar to what noticed in Fig 3A . These two immunoprecipitated bands were confirmed to be Fob2 protein by nano-LC ESI MS/MS ( Table 3 ) . Because immunoprecipitation of whole cell extracts of R . oryzae wild-type cells resulted in almost no apparent band on SDS-PAGE gel , we reasoned that the majority of the protein was retained bound to the antibody coated beads . Therefore , we ran the beads on a SDS-PAGE gel after boiling them . Coomassie Brilliant Blue stain showed two major bands at ~40 kDa as well as ~70 kDa ( S2D Fig ) . Because the 70 kDa band corresponded to the size of the identified Fob2 protein from concentrated supernatant of regenerated protoplasts grown in ferrioxamine-containing medium ( Fig 1A ) , we sequenced this band as well as the 40 kDa band by nano-LC ESI MS/MS . The major protein present in these two bands was identified as Fob2p ( Table 3 ) . Collectively , these results demonstrated the specificity of the antibodies to Fob proteins ( the antibodies are likely to recognize Fob1 and Fob2 proteins due to the predicted similar size and the 80% identity ) , strongly indicate that Fob proteins are cell-associated rather than secreted , and likely Fob2 proteins oligomerize under certain conditions to form larger protein size . To discern if the proteins are expressed on the cell surface or intracellularly , we used the anti-Fob2p antibodies to stain R . oryzae germlings without permeabilization and studied the staining patterns by flow cytometry and confocal imaging . We noticed enhanced fluorescence when R . oryzae wild-type cells were incubated with ferrioxamine and to a lesser extent when FeCl3 was used as the sole source of iron compared to fungal cells grown in the absence of iron source ( Fig 2D ) . Confocal images using these antibodies also confirmed the cell surface localization of Fob proteins when the fungal germlings were incubated with ferrioxamine ( S3 Fig ) . Collectively , these results show that Fob proteins are expressed on the cell surface of germinated R . oryzae when ferrioxamine is present . To determine the role of these putative receptors in the uptake of iron from ferrioxamine , we down regulated the expression of either of FOB1 or FOB2 ( Fig 4A ) . As expected RNA-i constructs targeting either of the 2 ORFs resulted in ~90% inhibition of the expression of the intended target ( i . e . , FOB1 or FOB2 ) . However , there was no inhibition in the expression of the other gene ( Fig 4B and 4C ) . In contrast , gene silencing of one FOB appeared to induce the expression of the other gene ( e . g . , a plasmid targeting FOB1 blocked expression of FOB1 expression , while FOB2 expression increased compared to cells transformed with empty plasmid and grown in the presence of ferrioxamine ) ( Fig 4B and 4C ) . These findings were further confirmed by immunostaining and FACS analysis . Cell surface expression of R . oryzae transformed with RNA-i constructs targeting either FOB1 or FOB2 had reduced levels compared to R . oryzae cells transformed with empty plasmid and grown in the presence of ferrioxamine . However , this reduction in Fob protein expression did not reach the low levels of R . oryzae transformed with the empty plasmid and grown in the absence of ferrioxamine ( Fig 4D ) . These results indicate that targeting one FOB gene results in a gene expression compensatory effect of the other . Having generated the two R . oryzae strains with reduced expression in FOB1 or FOB2 , we compared the FOB1 or FOB2 silencing effect on the germination and growth of R . oryzae in a medium supplemented with ferrioxamine . Silencing FOB1 or FOB2 did not have an effect on the ability of R . oryzae to germinate in a medium supplemented with ferrioxamine ( Fig 5A ) . In contrast , R . oryzae transformed with RNA-i constructs targeting FOB1 or FOB2 gene expression resulted in 20–25% inhibition in growth on medium supplemented with ferrioxamine as a sole source of iron when compared to R . oryzae wild-type or R . oryzae strain transformed with the empty plasmid ( Fig 5B ) . This reduction in growth is likely due to the compromised ability of R . oryzae transformed with RNA-i constructs , targeting FOB1 or FOB2 , to take up iron from ferrioxamine since attenuation of either FOB1 or FOB2 reduced the ability of R . oryzae to take up iron from radiolabeled ferrioxamine ( Fig 5C ) . Although initial incubation periods of 1–2 hrs showed 60–80% reduction in 55Fe3+ uptake from ferrioxamine among R . oryzae transformed with either RNA-i constructs targeting FOB1 or FOB2 compared to wild-type or R . oryzae transformed with empty plasmid , only 45% inhibition was detected after 4 hrs . This modest reduction in iron uptake is likely due to the compensatory expression of the other FOB gene . Because gene silencing of one of the two FOB genes resulted in a modest effect on growth and iron uptake from ferrioxamine , we employed a strategy of silencing both genes in a single construct driven by a single pdc promoter ( i . e . , pFOB1/2 ) ( Fig 6A ) . Upon transforming R . oryzae with this construct we were able to reduce the expression of FOB1 and FOB2 by >90% ( Fig 6B ) . Additionally , anti-Fob2p antibodies could not detect any bands from whole cell protein extract of R . oryzae transformed with dual inhibition construct and grown in the presence or absence of ferrioxamine ( Fig 3A ) . Moreover , the inhibition of FOB1/FOB2 translated to near complete blocking of cell surface expression to the level of R . oryzae transformed with the empty plasmid and grown in the absence of ferrioxamine ( Figs 6C and S3 ) . Furthermore , the dual inhibition of FOB1 and FOB2 resulted in severe retardation of R . oryzae germination ( Fig 7A ) and more than 40–50% inhibition in growth on medium supplemented with ferrioxamine ( Fig 7B ) when compared to wild-type strain or R . oryzae transformed with empty plasmid . Finally , this retardation in germination and growth on ferrioxamine supplemented medium was accompanied by up to 85% inhibition of iron uptake from ferrioxamine ( Fig 7C ) . Notably , there was no difference in growth of any of the three strains on FeCl3 supplemented YNB medium ( S4 Fig ) . Collectively , these results clearly implicate Fob1 and Fob2 proteins in the uptake of iron from ferrioxamine . Because FOB1 and FOB2 are involved in iron uptake from ferrioxamine and since iron acquisition from the host is a critical virulence factor for mucormycosis [9 , 10 , 23] , we hypothesized that these two genes are critical determinants of virulence especially in a mouse model of deferoxamine treatment . To test this hypothesis , we compared the virulence of R . oryzae with reduced cell surface expression of Fob1 and Fob2 proteins to wild-type or to R . oryzae transformed with empty plasmid using a deferoxamine-treated mouse model infected with a high inocula of 105 spores or a lower dose of 103 spores . At the higher inoculum , R . oryzae cells harboring the empty plasmid was as virulent as wild-type R . oryzae ( median survival time of 3 days of the wild-type and the empty plasmid strains infected mice P = 0 . 33 ) . In contrast , mice infected with one of two independently generated dual RNA-i transformants had attenuated virulence with > 21 day median survival time and 1/2 of the mice surviving the lethal infection ( P<0 . 001 ) . Further , at the lower inoculum of 103 spores , the two RNA-i transformants were avirulent with 100% of the mice surviving the infection and appeared healthy at the termination of the experiment , while mice infected with R . oryzae transformed with the empty plasmid had 80% mortality as early as 5 days post infection ( P<0 . 001 ) ( Fig 8A ) . In support of these data , mice infected with the RNA-i transformant had ~ 2 log reduction in fungal burden in the brains and kidneys ( primary and secondary target organs ) when compared to same organs recovered from mice infected with wild-type cells or those infected with the empty plasmid transformant ( P <0 . 0001 ) ( Fig 8B ) . To compare the severity of infection , we conducted histopathological examination on mice organs infected with R . oryzae transformed with the empty plasmid vs . those infected with R . oryzae transformed with RNA-i plasmid targeting both FOB1 and FOB2 . Brains and kidneys harvested from mice infected with R . oryzae transformed with RNA-i construct had minimal inflammation , edema and minimal to no presence of fungal elements . In contrast , organs taken from mice infected with R . oryzae transformed with the empty plasmid had abundance of fungal abscesses characterized by phagocyte infiltration and substantial edema ( Fig 8C ) . To confirm that the RNA-i construct inhibited FOB1 and FOB2 , we assessed the pattern of in vivo expression of these genes in fungi recovered from the same mouse organs that were processed for tissue fungal burden . Relative to the fungal actin house keeping gene ( ACT1 ) , FOB1 and FOB2 were expressed in organs collected from mice infected with the wild-type , or the empty plasmid transformant . Importantly , fungal cells recovered from brain and kidneys of mice infected with R . oryzae transformed with the RNA-i construct had ~80% reduction in FOB1 or FOB2 gene expression ( Fig 8D ) . These results indicate that FOB1 and FOB2 are expressed in vivo and the reduced virulence in mice infected with R . oryzae transformed with the RNA-i is due to attenuation of expression of both FOB genes . Finally , R . oryzae transformed with the dual construct of pFOB1/FOB2 had identical virulence to R . oryzae transformed with the empty plasmid in the diabetic ketoacidotic mice ( Fig 8E ) . Collectively , these data show that FOB1 and FOB2 are required for maximal virulence of R . oryzae only in the deferoxamine-treated mice . Having established the importance of FOB1 and FOB2 in the uptake of iron from ferrioxamine in vivo , we wanted to investigate the mechanism by which R . oryzae obtains iron from ferrioxamine . Since one of the most common pathways of obtaining iron in fungi and R . oryzae is the reductase/permease pathway [9] , we sought to investigate if this pathway plays a role in iron uptake from ferrioxamine . R . oryzae wild-type cells growing in medium containing ferrioxamine as a sole source of iron demonstrated ~ a 3 fold increase in reductase activity when compared to cells growing on medium containing iron in the form of FeCl3 ( Fig 9A ) . Importantly , the extracellular , membrane-impermeable ferrous chelator bathophenanthroline disulfonate ( BPS ) inhibited growth of R . oryzae when ferrioxamine is used as a sole source of iron ( Fig 9B ) . This inhibition was due to chelating ferrous since increased concentrations of FeCl3 reversed the inhibitory effect of BPS ( Fig 9C ) and the addition of BPS blocked 55Fe3+ uptake from ferrioxamine ( Fig 9D ) . Similar results were obtained with the membrane-permeable ferrous chelator bipyridyl ( S5 Fig ) . Since ferrioxamine chelates ferric iron , this inhibition implies the reduction of the ferric iron to ferrous prior to transportation to the fungal cell . Previously , we have shown that targeted gene disruption of FTR1 in the multinucleated R . oryzae results in a mutant with reduced copies of the high affinity iron permease ( FTR1 ) rather than a completely ftr1 null mutant [24] . This mutant with reduced copies of FTR1 had a retarded growth on medium containing ferrioxamine B as a sole source of iron which implies a role for Ftr1p in iron transport from ferrioxamine [24] . To determine if Ftr1p is required to transport the released ferrous to the fungal cell wall , we compared the ability of this R . oryzae strain with reduced FTR1 copy number to wild-type or PyrF-complemented strain in their ability to take up 55Fe3+ from ferrioxamine . After 6 hours of incubation with radiolabeled ferrioxamine , R . oryzae with reduced FTR1 copy number ( i . e . , putative ftr1 mutant ) had significant decrease in taking up 55Fe3+ compared to wild-type cells ( P<0 . 01 ) . Moreover , after 8 hours of incubation with ferrioxamine , R . oryzae with reduced FTR1 copy number had significant reduction in 55Fe3+ uptake when compared to wild-type or PyrF-complemented strains ( P<0 . 005 ) ( Fig 10A ) . To confirm these in vitro findings , we compared the virulence of two R . oryzae with attenuated expression of FTR1 to their corresponding control strains in the deferoxamine-treated mouse model . R . oryzae strain with reduced FTR1 copy number ( Fig 10B ) and R . oryzae transformed with RNA-i construct targeting FTR1 expression [24] ( Fig 10C ) had attenuated virulence in the deferoxamine-treated mouse model when compared to PyrF-complemented strain or R . oryzae transformed with the RNA-i empty plasmid , respectively . Specifically , at day 21 post infection , 30% of mice infected with reduced FTR1 copy number strain survived the infection vs . 0% for mice infected with the PyrF-complemented strain ( P = 0 . 001 ) and 40% of mice infected with R . oryzae transformed with RNA-i construct targeting FTR1 survived the infection vs . 0% survival for mice infected with R . oryzae transformed with the empty plasmid ( P = 0 . 013 ) . Collectively , these results clearly show that iron transport from ferrioxamine relies , at least in part , on the reductase/permease system . Because the ftr1 mutants were not completely avirulent in deferoxamine-treated mice , we investigated the possibility of the involvement of other mechanisms in transporting iron from ferrioxamine . All identified eukaryotic siderophore transporters belong to the siderophore–iron transporters ( SIT ) family , a subfamily of the major facilitator superfamily [25 , 26 , 27 , 28] . We have conducted homology searches of the S . cerevisiae SIT genes [25 , 29 , 30] , including ARN1 , ARN2 , ARN3/SIT1 , and ARN4 with the published R . oryzae genome data base . Although the overall amino acid identity was low ( i . e . 20–23% ) , we identified 9 ORFs that share homology with the S . cerevisiae SIT encoding genes and therefore potentially encode Sit proteins in R . oryzae ( S1 Table ) . We compared the pattern of expression of four of these genes ( SIT1 , 4 , 6 , and 9 ) with the highest homology to S . cerevisiae SIT genes to FOB1 , FOB2 , and FTR1 over time in the presence of deferoxamine or ferrioxamine as iron-poor and iron-rich media , respectively . In iron-rich ferrioxamine medium , FOB2 expression was induced as early as 30 min and reached a maximum expression of 9 fold increase by 24 hrs . Interestingly , FOB1 expression was significantly increased only after 24 h incubation ( Fig 11A ) . This enhancement of FOB gene expression was accompanied by gradual reduction in iron starvation genes ( i . e . FTR1 and SIT ) that reached >99% decrease after 24 h ( FTR1 and SIT9 had almost complete reduction in expression after 30 min of incubation ) when compared to 10 min data set ( Fig 11B ) . In contrast and as expected , none of the FOB gene expression was enhanced in the presence of deferoxamine ( Fig 11C ) . Finally , FTR1 and putative SIT gene level of expression did not considerably change in deferoxamine containing medium over time ( level of expression fluctuated between 0 . 3–1 . 4 fold relative to ACT1 ) ( Fig 11D ) . These results confirm the induction of FOB genes in response to ferrioxamine and that this induction changes the conditions of the fungal cell from iron-deplete to iron-replete conditions as shown by the reduction in the iron-starvation gene expression . Further , the similar pattern of SIT genes expression to the FTR1 expression in ferrioxamine containing medium suggests that these genes might play a role in iron uptake from this siderophore . The unique predisposition of dialysis patients receiving deferoxamine for treating iron overload to mucormycosis has been clinically known for decades [13 , 15 , 16] . Previous data demonstrated that Rhizopus species can utilize iron from ferrioxamine as a xenosiderophore to obtain iron from the host [17 , 31 , 32] . This process was found to be energy dependent and possibly doesn’t require internalization of the siderophore [18] . In this study , we have identified two ORFs that encode Fob1 and Fob2 proteins which are required for iron uptake from ferrioxamine . We introduce evidence that these proteins are receptors to which ferrioxamine binds to during the process of obtaining iron by R . oryzae . First , Fob proteins are mainly induced by ferrioxamine and detected in the supernatants of regenerating R . oryzae protoplasts before they are covalently incorporated into the nascent cell surface [33] . Second , these two proteins were found to be expressed on the cell surface of R . oryzae , a criterion required for cell receptors . Third , whole cell extracts from R . oryzae grown in the presence , but not the absence , of ferrioxamine bound to radiolabeled ferrioxamine under non-denaturing conditions . Fourth , Fob2 protein isolated from whole cell extracts using anti-Fob2 antibodies also bound to radiolabeled ferrioxamine . Finally , a mutant of R . oryzae with attenuated FOB1 and FOB2 expression bound much less , and had retardation in , iron uptake from radiolabeled ferrioxamine versus the wild-type strain . Collectively , these features indicate that Fob1 and Fob2 proteins are ferrioxamine induced , surface exposed , and directly interact with the siderophore . The identified Fob proteins were found to belong to the CBS domain family . CBS domain is a conserved protein sequence region , which is found in all kingdoms of life [34] . It was first described in Cystathionine β Synthase , hence the name CBS . Other proteins belonging to this family include inosine monophosphate dehydrogenase , AMP kinase , and chloride channels . Therefore , CBS-containing proteins functions are very diverse and could range from affecting metabolism , multimerization and sorting of proteins , channel gating , to ligand binding . Based on their function , CBS domain proteins can be present in the cytoplasm or associated with the cell membrane [34] . For example , despite the lack of classical features of secreted proteins such as N-terminal signal peptide sequence , CBS domain proteins were reported to be associated with cell membrane in eukaryotic cells functioning as CLC chloride channels with outer membrane sequence , transmembrane domain , and a cytoplasmic tail [35] . Concordant with this report is our finding that Fob1p and Fob2p are predicted to be surface associated by using the MEMSAT program ( S1 Fig ) . In most cases , however , CBS-domain proteins bind to ligands related to adenosine , including S-adenosylmethionine ( SAM ) , AMP , ADP [36] , as well as recently reported DNA and RNA fragments [37 , 38] . They also have been reported to bind to metallic ions such as Mg2+ [39] . To our knowledge , we show for the first time that a ferrioxamine receptor belongs to the CBS domain family of proteins . Given the diverse function of these proteins and the previous report of binding metallic ions , it is possible that Fob1 and Fob2 proteins on R . oryzae cell surface bind ferrioxamine via Fe3+ . This assumption is supported by our results that only ferrioxamine , but not deferoxamine , induces the expression of FOB1 and FOB2 and by the fact that deferoxamine was not able to competitively inhibit the uptake of radiolabeled ferrioxamine by R . oryzae Importantly , ferric-rich siderophores are known to induce a conformational change of the secondary structure of their receptors and likely explain the ability of ferrioxamine and inability of deferoxamine to induce and bind to Fob proteins [40 , 41] . An apparent discrepancy in our results is the actual size of the Fob2p of ~40 kDa as indicated by the predicted sequence , the size of the rFob2p and recognized band from whole cell extracts vs . the originally isolated 70 kDa band . We initially assumed that the presence of multiple possible N- and O-glycosylation sites might account for this increase in size of the protein due to post-translational modification . However , the isolation of Fob proteins from whole R . oryzae extracts at the predicted size of 40 kDa argues against a post-translational modification of the protein . A more likely explanation is the possible oligomerization of the protein when the sample is present in concentrated form mixed with SDS . First , the 70 kDa band was detected in concentrated supernatants from regenerating R . oryzae protoplasts in SDS-PAGE . Second , when we immunoprecipitated whole cell extracts from wild-type cells , the majority of the sample remained bound to the antiFob2p antibody-coated beads and that sample contained Fob proteins at the ~40 and ~70 kDa bands when separated on SDS-PAGE . It is prudent to mention that cell membrane proteins are notorious for oligomerization even in denaturing conditions when mixed with SDS [42 , 43] . In fact , a study found that SDS enhances the dimerization of β-amyloid proteins from human cortical tissues and the level of this unnatural dimerization increases with the increased concentration of SDS in the sample [44] . Alternatively , CBS domain proteins were reported to oligomerize ( e . g . Streptococcus pyogenes IMPDH protein [34] ) , therefore it is possible that Fob proteins are present as homo- or heterodimers . Our sequence data which identified Fob2p in two independent samples detected at ~70 kDa without the identification of Fob1p suggest that Fob2p is the major form expressed . However , it is possible that Fob1p is expressed in low quantities and is required for the receptor activity by forming a heterodimer with Fob1p . Investigations into the possibility of homodimer vs . heterodimer formation of Fob2p and Fob1p or the possibility of induction of unnatural oligomerization during sample processing are currently underway . Deferoxamine is a siderophore belonging to the hydroxamate family that is mainly isolated from bacteria including Gram-positives ( e . g . , Streptomyces and Nocardia ) as well as Gram negatives Enterobacteriaceae [45 , 46 , 47] . Up to now , deferoxamine has not been isolated from fungi but many studies indicated the utilization of this siderophore by fungi in obtaining iron including Mucorales , Aspergillus , Cryptococcus [17] and S . cerevisiae [25 , 27 , 48] . In addition to utilizing deferoxamine as xenosiderophores , Mucorales are known for secreting their own siderophore , rhizoferrin [22 , 49] . However , rhizoferrin belongs to the carboxylate family which is structurally distinct from the hydroxamate siderophores [22 , 50] . Interestingly , rhizoferrin ( with or without iron ) did not induce the expression of FOB1 or FOB2 indicating that these genes encode receptors specific to iron uptake from ferrioxamine . However , the ability of Mucorales to utilize iron from structurally distinct siderophores points to the critical role iron plays in the survival of this group of fungi . Our in vitro data with down regulation of a single FOB gene demonstrated a modest effect on germination , growth and iron uptake from medium supplemented with ferrioxamine . These results can be explained by the fact that the untargeted gene was overexpressed and had a compensatory effect . This compensatory mechanism by related genes is well documented when other genes of the family are nullified [51] . However , this compensatory gene expression effect did not translate into full restoration of cell surface protein to the level of R . oryzae transformed with empty plasmid and grown in ferrioxamine containing medium ( Fig 4D ) . The reduction of fluorescence in a single inhibition strain could be due to the possibility that the enhancement of non-targeted gene expression did not rise to the level that would entirely compensate for inhibition of the other gene . It is also possible that the gene expression of the non-targeted gene does not localize entirely to the cell surface . Our strategy of targeting both genes with a single construct successfully drastically attenuated the expression of FOB1 and FOB2 , resulting in a significant retardation in germination , growth and iron uptake from medium supplemented with ferrioxamine as a sole source of iron . However , the inhibition of growth was not complete . Further , the inhibition of iron uptake from ferrioxamine by R . oryzae with abrogated FOB1/FOB2 expression was marginally reduced at later time points ( Fig 7C , compare 4 h with earlier time points ) . These two phenomena can be explained by the fact that old or heat-damaged ferrioxamine ( due to incubation at 37oC ) is unstable and undergoes an autoreduction step to release ferrous iron [52] , that is taken up by the fungus . This assumption is corroborated by the yielding of a pink-colored complex in the presence of ferrozine in our experiment . It is also probable that secondary mechanisms of obtaining iron from ferrioxamine which are independent from FOB1/FOB2 are operative . Our results with pattern of expression of SIT-like genes in medium supplemented with ferrioxamine or deferoxamine over time supports the model of SIT-like genes being involved in iron uptake from ferrioxamine . Specifically and similar to FTR1 gene , SIT-like genes were transiently expressed as early as 10 min then the expression faded over time which indicates change of the conditions of the fungal cell from iron-deplete to iron-replete conditions . This transient expression and tight regulation is critical for the organism to avoid the toxic carnage of excess iron [53 , 54] . Importantly , a R . oryzae mutant with dual attenuation of FOB1/FOB2 expression had severe virulence retardation in the deferoxamine-treated mice vs . mice infected with wild-type or empty plasmid strains with 50% of the mice infected with R . oryzae transformed with RNA-i targeting FOB1/FOB2 surviving the infection and looking healthy . It is imperative to point out that this difference in virulence was not detected in the DKA mouse model which is consistent with the fact that Fob1 and Fob2 are proteins induced by ferrioxamine . Fob1 and Fob2 are likely dispensable in the DKA model since Fe3+ is released from transferrin by the action of acidosis and hyperglycemia [9 , 11] and transported into the fungal cell via the Ftr1p [24] . In microorganisms , the mechanism of iron uptake from ferrioxamine involves either the uptake of the entire siderophore complex , reduction of the Fe3+ to Fe2+ to release it from the siderophore followed by uptake of iron via oxidase/permease system , or both . For example , in S . cerevisiae it has been reported that both mechanisms are operative [30 , 55] . A previous study demonstrated that iron uptake from ferrioxamine by R . oryzae is likely energy dependent , require a reductive step and doesn’t involve the uptake of the ferrioxamine complex [18] . In contrast , by using a fluorescent derivative of deferoxamine , a recent study showed that the entire siderophore was taken up by the R . oryzae cells [56] . Our results mainly support the model by which R . oryzae obtains iron from ferrioxamine via the reductase/permease system . First , the reductase activity of R . oryzae when incubated with ferrioxamine far exceeds the reductase activity when fungal cells were incubated with FeCl3 . Second , the addition of the ferrous chelator BPS inhibited growth of R . oryzae in medium with ferrioxamine as a sole source of iron and this inhibition was due to attenuation of iron uptake from ferrioxamine . Finally , we previously reported that a R . oryzae mutant with reduced copy number of the high affinity iron permease ( FTR1 ) had reduced growth on medium supplemented with ferrioxamine [24] . In this study we show that this growth inhibition is likely due to the diminished ability of this mutant to take iron from ferrioxamine . More significantly , we demonstrated reduced virulence in the deferoxamine-treated mouse model of mucormycosis , even though the R . oryzae mutant with reduced FTR1 copy number or inhibited expression of FTR1 do not have full abrogation of the function of the permease activity [24] . However , a minor mechanism reliant on internalization of the entire siderophore cannot be completely ruled out because the effect of BPS in vitro and the attenuated virulence with ftr1 mutants were not complete . It is possible that the reductase/permease pathway represent a major uptake pathway for iron acquisition from ferrioxamine , while the uptake of the siderophore by shuttle mechanism[56] represents a secondary mechanism . This secondary mechanism of iron uptake from ferrioxamine is supported by the pattern of transient and tightly regulated SIT-like gene expression in ferrioxamine containing medium ( Fig 11 ) . Also the assumption that ferrioxamine is taken by a Sit-like shuttle transporter is a secondary mechanism is enforced by the fact that the patchy distribution of fluorescent ferrioxamine was observed only at the later time point of 24 hours of incubation and not earlier [56] . Consequently , we propose that R . oryzae overexpresses Fob1 and Fob2 in response to ferrioxamine , which accumulates in a host being treated with deferoxamine to eliminate iron overload toxicity . These receptors bind ferrioxamine and through fungal cell surface reductase activity , Fe3+ is released from the siderophore as Fe2+ prior to being transported across the fungal cell membrane by the oxidase/permease complex ( i . e . , Fet3/Ftr1 ) to support fungal growth and virulence ( Fig 12A ) . In the absence of Fob1/Fob2 or the Ftr1p , this process is compromised and the fungal virulence is reduced ( Fig 12B ) . Also in this model , transport of the entire ferrioxamine siderophore by Sit-like transporter is likely to be operative as a secondary pathway . In summary , we have identified Fob1 and Fob2 proteins as ferrioxamine receptors in R . oryzae . These cell surface proteins are induced by ferrioxamine and are required for mucormycosis pathogenesis in the deferoxamine-treated mouse model . The mechanism of iron uptake from ferrioxamine is reliant on binding of the iron-rich siderophore to its receptor , followed by transport of iron to the fungal cell via the reductase/permease system without internalization of the siderophore-iron complex . These two receptors appear to be conserved in at least another five Mucorales and can be the subject of future novel therapy to maintain the use of deferoxamine for treating iron-overload . All organisms used in this study are listed in S2 Table . R . oryzae 99-880 is a clinical strain isolated from a patient with a brain abscess and obtained from the Fungus Testing Laboratory of University of Texas Health Science Center at San Antonio . The organism was routinely grown on potato dextrose agar ( PDA ) plates ( BD ) for 4-5 days at 37°C . R . oryzae M16 is a pyrF null mutant that is derived from R . oryzae 99-880 and is unable to synthesize its own uracil [57] , was grown on YPD medium ( MP Biomedicals ) supplemented with 100 μg/ml uracil . R . oryzae PyrF-complemented strain , R . oryzae with reduced FTR1 copy number , and R . oryzae transformed with RNA-i targeting FTR1 expression were all derived from strain 99-880 and M16 and previously described in detail [24] . In RNA-i experiments , a chemically-defined synthetic medium containing yeast nitrogen base ( YNB ) supplemented with complete supplemental mixture without uracil ( CSM-URA ) ( MP Biomedicals ) ( i . e . , YNB+CSM-URA ) ( formulation/liter , 17 g yeast nitrogen base without amino acids ( YNB ) ( BD ) , 200 g dextrose , and 7 . 7 g complete supplemental mixture minus uracil ( CSM-URA ) ) was used . To count and isolate single colonies all media above were supplemented with 0 . 1% Triton X-100 ( Sigma ) . Spores were collected in endotoxin free PBS containing 0 . 01% Tween 80 , washed twice with PBS , and counted with a hemacytometer for inoculum preparation . To identify the R . oryzae cell surface protein ( s ) that interacts with ferrioxamine , freshly harvested R . oryzae 99-880 spores ( 2 . 5x108 ) were added to 50 ml YPD and pregerminated with shaking at 37°C . After 4 hours , the spores were pelleted by centrifugation and washed twice with 0 . 5 M sorbitol buffer , and suspended in 50 ml protoplasting solution which consists of 10 mM sodium phosphate , pH 6 . 4 , 0 . 5 M sorbitol , 250 μg/ml lysing enzyme ( Sigma ) , 150 μg/ml chitinase ( Sigma ) and 150 μg/ml chitosanase ( US Biologicals ) . The suspension was incubated for ~ 2 h at 30°C with gentle shaking ( 100 rpm ) until majority of the spores ( > 90% ) formed protoplasts as determined by light microscopy examination and their sensitivity to water . Protoplasts were then collected by centrifugation at low speed ( 100 rpm ) , washed twice with 0 . 5 M sorbitol , and then resuspended in 10 ml YNB with amino acids containing 0 . 5 M sorbitol in the presence or absence of 10 μM ferrioxamine ( Sigma ) . The protoplasts were allowed to regenerate for 2 h by incubating at 37°C with gentle shaking at 100 rpm . During the early stages of this regeneration process , many surface protein precursors are shed into the extracellular medium but not yet covalently incorporated into the nascent cell surface , thereby enabling their easy isolation and identification [33] . Thus , supernatants containing cell wall materials from regenerated protoplasts were isolated from fungal cells by centrifugation at 1000 rpm , followed by filtration through a 0 . 25 μm pore-size membrane after the addition of Halt Protease Inhibitor Cocktails ( Thermo Scientific ) . The filtrate was then concentrated using iConcentrator tube with molecular weight cutoff = 9 . 0 kDa ( Thermo Scientific ) and protein concentration was measured with the Bradford method ( Bio-Rad ) . Proteins from supernatant of regenerated protoplasts in the presence or absence of ferrioxamine were separated on 10% SDS-polyacrylamide gel ( Bio-Rad ) and stained with Coomassie dye R-250 ( Pierce ) . To investigate if the proteins from supernatant of regenerated protoplasts bind to ferrioxamine B , 25 μg of protein extract was mixed with 55Fe-labeled ferrioxamine ( for preparation , see below ) , and incubated for 1 h at room temperature . The mixture was then mixed with native protein sample buffer and separated by running a native gel ( Bio-Rad ) in the absence of SDS and exposed to X-ray film . The candidate band from SDS-PAGE was cut and microsequenced using MALDITOF MS/MS ( UCLA Molecular Instrumentation Center ) . 55Fe3+ labeled ferrioxamine was prepared as previously described [58] . Briefly , 55FeCl3 ( 3 μl of 3 . 5 Ci/mmol , Perkin Elmer ) was mixed with iron-free deferoxamine B ( Sigma ) in a molar ratio of 0 . 9:1 in a volume of 100 μl , and incubated for 2 h at room temperature . Free iron was then removed by adding sodium phosphate ( pH 7 . 4 ) followed by centrifugation at 20 , 000 g for 2 min . Supernatant was diluted to 3 . 0 ml with ultra-pure water with a resistivity of 18 . 2 megohm-cm ( Barnstead International , Dubuque , Iowa ) and treated with 150 mg Chelex 100 resin ( Bio-Rad ) twice to get rid of any residual free iron . To quantify gene expression of putative ferrioxamine receptors and other SIT genes of R . oryzae in vitro , total RNA was isolated from mycelia that have been grown in YNB with amino acids containing 10 μM of siderophore ( ferrioxamine , deferoxamine , rhizoferrin with or without ferric iron [EMC Microcollections GmbH] ) ) or varying concentrations of FeCl3 overnight with shaking at 37°C . The RNA was isolated using RNeasy Plant Mini Kit ( Qiagen ) after grinding the mycelia in liquid nitrogen . cDNA was synthesized from 2 μg of total RNA from each sample after treating with Turbo DNA-free DNase I ( Life Technologies ) to remove any contaminating DNA . After removing DNase I with DNase inactivation Reagent ( Life Technologies ) , cDNA was synthesized using RETROscript kit ( Life technologies ) . qPCR was carried out using the Power SYBR Green method in the StepOne Real-Time PCR System ( Life Technologies ) with a thermal-cycling program as follows: initial denaturing step for 10 min at 95°C , followed by 40 cycles of denaturing at 95°C for 15 s , and annealing/elongation at 60°C for 1 min . R . oryzae actin gene ( ACT1 ) was used as a reference control , putative ferrioxamine receptor gene-specific primers are listed in S3 Table . The comparative Ct method was used for analysis [59] . RNA-i was used to silence FOB1 or FOB2 expression individually or collectively . For individual silencing , two reverse-complemented fragments ( ~450 bp ) were PCR-amplified and cloned into plasmid pRNAi-pdc-intron in an opposite direction and separated by a 100 bp-intron sequence , hence the total insert is ~1 kb [24] . Two constructs , termed pRNAi-FOB1 and pRNAi-FOB2 , targeting FOB1 and FOB2 expression were respectively generated . To silence both genes at the same time , the 1 kb insert from pRNAi-FOB1 was cut and re-cloned downstream of pRNAi-FOB2 , separated by a 50-bp interval of random sequence to yield pFOB1/2 . All constructs were transformed into R . oryzae M16 ( pyrF null mutant ) using a biolistic delivery system ( Bio-Rad ) as previously described [60] and transformants were selected on YNB+CSM-URA plates supplemented with 0 . 1% triton . Primers used for this work are listed in S3 Table . To compare growth rate among R . oryzae isolates on medium with ferrioxamine as a sole source of iron , 105 spores of each strain were plated in the middle of the YNB+CSM-URA agar plates which have been treated with 1 mM of ascorbic acid ( Sigma ) and 1 mM ferrozine ( Sigma ) to chelate iron and then supplemented with 10 μM ferrioxamine . The plates were incubated at 37°C and the diameter of the colony was calculated after 24 or 48 h . To determine the effect of the ferrous chelators BPS or bipyridyl on the growth of R . oryzae in the presence of ferrioxamine , R . oryzae wild-type spores ( 5x105/ml ) were added to YNB+CSM-URA medium supplemented with 10 μM ferrioxamine and one of the two ferrous chelators ( 0 . 2–0 . 8 mM of BPS or 0 . 2–0 . 4 mM bipyridyl ) . In some experiments , FeCl3 was added in increased concentrations ( 10–1000 μM ) to investigate if iron can reverse the effect of the ferrous chelator . Cultures were incubated at 37°C with shaking . At selected time intervals , 1 ml sample was taken from the culture and the optical density measured at 400 nm [17] . Spores from R . oryzae wild-type , transformants transformed with empty plasmid or RNA-i constructs targeting FOB1 , FOB2 , or FOB1/FOB2 were harvested from their plates as above and inoculated into YNB+CSM-URA medium supplemented with 10 μM ferrioxamine as a sole source of iron at 5 x 106 spores/ml . The cultures were incubated at 37°C with shaking ( 200 rpm ) . At selected time periods , a 10 μl sample was collected from each culture , and examined with a phase contrast microscope to determine if there is any effect of FOB gene silencing on germination . Pictures were taken with a Nikon E5400 camera . To determine the ability of different strains of R . oryzae to take up iron from ferrioxamine , we used our previously described methods [24] . Briefly , R . oryzae spores ( 5 x 106/ml ) were pre-germinated for 3 h in YNB medium supplemented with 1 mM ferrozine and 1 mM ascorbic acid at 37°C with shaking . Cells were harvested by centrifugation , washed twice with assay buffer ( YNB + 10 mM 4-morpholinepropanesulfonic acid + 1 mM ferrozine ) , and resuspended in 7 ml of assay buffer supplemented with 10 μM of 55Fe-labeled ferrioxamine at a concentration of 5 x 106 spores/ml , and incubated at 37°C with shaking . In some experiments , to determine if iron uptake from ferrioxamine required the reduction of Fe3+ to Fe2+ , the ferrous chelator BPS ( 0 . 2–0 . 8 mM ) or bipyridyl ( 0 . 2–0 . 4 mM ) were added to the assay buffer . At different time points , 1 ml of sample was collected by Whatman GF/C glass filters ( GE Healthcare ) through a vacuum filtration manifold ( Millipore ) and washed twice with cold water . Cell associated 55Fe was counted in a liquid scintillation counter ( Packard Instrument , Downers Grove , IL ) . For competition studies using cold deferoxamine or ferrioxamine , 55Fe3+ uptake studies were conducted as above while including increasing concentrations of cold deferoxamine or ferrioxamine . Iron uptake was determined after 2 h incubation period and the results expressed as % inhibition relative to samples run without the inclusion of cold siderophores . To determine the reductase activity of R . oryzae wild-type in the presence of ferrioxamine or FeCl3 , we used the method of Dancis et al . [61] . Briefly 5x107 of R . oryzae spores were cultured in 10 ml of YNB+CSM-URA medium supplemented with 100 μM FeCl3 or 10 μM ferrioxamine . The culture was incubated at 37°C with shaking . At time intervals , 1 ml of the culture was collected , centrifuged at 3000 rpm and the spores washed twice with cold distilled water . The spores were suspended in 1 ml of assay buffer ( 0 . 05 M sodium citrate ( pH 6 . 5 ) supplemented with 5% glucose ( w/v ) ) . The culture was incubated at 30°C for 15 min before adding BPS and FeCl3 to a final concentration of 1 mM for both compounds . The mixture is vortexed and then incubated for additional 20 min , after which spores were removed by centrifugation and the optical density of the supernatant determined at 535 nm using a blank processed similarly but without any added spores . The amount of generated Fe2+ was estimated with a standard curve constructed from solutions of known Fe2+ concentrations using FeSO4 . To determine the cell surface localization of the putative ferrioxamine receptors , we sought to raise polyclonal antibodies against one of the two proteins ( i . e . , Fob1p or Fob2p ) . Since Fob1 and Fob2 predicted proteins are ~ 80% identical at the amino acid level , we chose to express FOB2 ( RO3G_11000 ) . To express FOB2 , we first PCR-amplified R . oryzae FOB2 gene ( S3 Table ) , the PCR product was gel-purified and digested with BamHI/PstI before cloning into E . coli expression vector pQE32 ( Qiagen ) to yield pQE32/FOB2 . This resulted in a construct with 6xhis tag on the N-terminus of Fob2 protein . Plasmid pQE32/FOB2 was transformed into E . coli XL-10 gold ( Agilent Technologies ) and transformants isolated on Luria-Bertani ( LB ) agar plates supplemented with ampicillin ( 100 μg/ml ) . Gene cloning was confirmed by DNA sequencing . To purify Fob2 protein , a single colony from the transformed E . coli was grown overnight . On the second day , the culture was diluted 200 times in fresh medium containing ampicillin , and incubated with shaking at 37°C until reaching an OD = 0 . 6 . At this stage , 1 mM IPTG was added to the culture which was then incubated at 37°C with shaking for additional four hours . The bacterial cells were then harvested by centrifugation , and 6xhis tagged Fob2p was purified to >90 homogeneity with HisPur Cobalt Purification Kit according to the instructional manual ( Pierce ) . The purified recombinant protein was extensively dialyzed in phosphate buffered saline ( PBS ) ( Mediatech , VA ) , and the purity and size confirmed by SDS-PAGE analysis followed by MALDI-TOF–mass spectrometry/mass spectrometry analysis . To generate anti-Fob2p antibodies , normal BALB/c mice were immunized by subcutaneous injection of 20 μg recombinant Fob2p mixed in equal amounts with complete freund’s adjuvant ( CFA , Sigma-Aldrich ) . The mice were boosted with a subcutaneous administration of a similar dose of the protein mixed with incomplete freund’s adjuvant ( IFA ) three weeks following the initial immunization . Mice vaccinated with the diluent ( i . e . , PBS ) mixed with CFA/IFA without rFob2p using the same regimen served as control mice . Twelve days after the boost , blood samples were collected from mice , and anti-Fob2p Ab titers were determined by using ELISA plates coated with Fob2p as we previously described [62] . The ELISA titer was taken as the reciprocal of the last serum dilution that gave a positive OD reading ( i . e . , more than the mean OD of negative control samples plus 2 standard deviations ) . To prove Fob proteins are R . oryzae cell surface proteins , freshly harvested spores grown on YNB+CSM-URA were pregerminted for 3–4 hours in the presence or absence of 10 μM deferoxamine . The pregerminated spores were incubated with the anti-Fob2p IgG at 1:100 for 1 hour on ice after a blocking step using 1 . 5% goat serum in PBS for 1 hour . Cells were washed 3 times with Tris-buffered saline ( TBS; 0 . 01 M Tris HCl [pH 7 . 4] , 0 . 15 M NaCl ) containing 0 . 05% Tween 20 , then counterstained with alexa 488-conjugated anti-mouse secondary at 1:100 for 1 hour on ice . The stained cells were imaged with Leica confocal microscope using an excitation wavelength of 488 nm . To quantify the expression of Fob proteins on R . oryzae cell surface , the pregerminated spores were stained as above , then 1 ml samples were analyzed using a FACSCalibur ( Becton Dickinson ) instrument equipped with an argon laser emitting at 488 nm . Fluorescence emission was read with a 515/40 bandpass filter . Fluorescence data were collected with logarithmic amplifiers . The mean fluorescence intensities of 104 events were calculated using CELLQUEST software [20 , 63] . R . oryzae ( 5 x 107 cells ) were grown overnight in YNB+CSM-URA with or without 10 μM ferrioxamine . Mycelia were collected by filtration , washed briefly with PBS , and then ground thoroughly in liquid nitrogen using pestle and mortar for 3 min . The ground powder was immediately transferred to microfuge tube containing 500 μl extraction buffer which consisted of 50 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 10 mM MgCl2 . The extraction buffer was supplemented with 1X Halt Protease Inhibitor Cocktails ( Thermo Scientific ) . The sample was vortexed vigorously for 1 min , then centrifuged for 5 min at 21000 g at 4°C . The supernatant was transferred to a new tube and the protein concentration determined using Bradford method . For detecting protein secretion , supernatants from R . oryzae cells growing in YNB+CSM-URA with or without 10 μM ferrioxamine for overnight , were collected after filtration of the hyphal mate using 0 . 22 μM membrane units ( Millipore ) . The cell-free supernatants were concentrated by 200 x and the total proteins measured as above . For Western blotting , 5 μg of each sample was used to separate proteins on an SDS-PAGE . Separated proteins were transferred to PVDF membranes ( GE Water & Process Technologies ) , and treated with Western blocking reagent ( Roche ) for 1 h at room temperature . PVDF membranes were then probed with mouse anti-Fob2p antibodies ( 1:100 diluted ) . After washing 3 times with TBS-T buffer ( Tris-buffered saline + 0 . 05% Tween 20 ) , the membranes were incubated with HRP-conjugated sheep anti-mouse IgG ( 1:10 , 000 dilution ) ( Sigma-Aldrich ) . Fob bands were visualized by adding the HRP substrate ( SuperSignal West Dura Extended Duration Substrate , Thermo Scientific ) , and the chemiluminescent signal was detected using a Sony CCD camera . In-gel tryptic digest followed by nano-liquid chromatography–tandem mass spectrometry analysis was used to confirm the identity of the band . Dot blots were used to determine the direct binding of radiolabeled ferrioxamine to crude protein extracts or Fob proteins purified by immunoprecipitation using anti-Fob2p antibodies . Briefly , anti-Fob2p antibodies ( 1:10 dilution ) were coupled to protein G column according to the manufacturer’s recommendation ( Thermo Scientific ) . Protein extracts ( 100 μl ) , or bovine serum albumin ( BSA ) were added to the column at room temperature for 1 h after which the flow through is discarded followed by washing the column 3 times with PBS . 55Fe3+ labeled ferrioxamine was then added to the anti-Fob2 antibodies-coupled protein G column and incubated for another 1 h at room temperature . Next , the flow through is discarded and the resin was washed three times with PBS before blotting beads containing antibody/antigen complex on PVDF membrane followed by exposure to X-ray film . The contribution of FOB to R . oryzae virulence was determined using both a deferoxamine-treated and DKA mouse models . In both models , ICR mice ( ≥20 g ) ( Taconic , USA ) were used . For the deferoxamine-treated mouse model we adapted the method of Abe et al [64] . Briefly , mice were injected i . p . with three doses of deferoxamine ( 100 mg/kg/per dose ) given on days -1 , 0 , and +1 relative to infection with R . oryzae . For the DKA mouse model , mice were rendered diabetic in slight ketoacidosis with a single i . p . injection of 210 mg/kg streptozotocin in 0 . 2 ml citrate buffer 10 days prior to fungal challenge as we previously described [20 , 65 , 66 , 67] . Glycosuria and ketonuria were confirmed in all mice 7 days after streptozotocin treatment . In both models , mice were infected with 105 spores by intravenous injection in the tail vein . Additionally , in the deferoxamine-treated model , a lower inoculum of 103 was tested . The primary efficacy endpoint was time to moribundity . In some experiments , as a secondary endpoint , fungal burden in the brains and kidneys ( primary and secondary target organs ) was determined 48 hours after infection by quantitative PCR assay . Briefly , mouse organs were disrupted/homogenized in Lysing Matrix tubes ( MP Biomedicals ) using FastPrep FP120 Homogenizer ( Thermo Electron Corporation ) , total DNA was then isolated according to the instructional manual of DNeasy Tissue Kit ( Qiagen ) . qPCR was then carried out as previously described [68] , using R . oryzae-specific 18S rRNA primers ( S3 Table ) . Values were expressed as log10 spore equivalent/g tissue . Histopathological examination was carried out on sections of the harvested organs after fixing in 10% zinc formalin . The fixed organs were embedded in paraffin , and 5-mm sections were stained with H&E to detect R . oryzae hyphae [20] . For in vivo expression of FOB genes , brains and kidneys were collected from mice 48 hours after wild-type , empty plasmid–transformed , or RNA-i transformed R . oryzae infection were flash frozen in liquid nitrogen , disrupted/homogenized as above , and then processed for RNA extraction using a Tri Reagent solution ( Ambion ) . Reverse transcription was performed with RETROscript ( Ambion ) using primers listed in S3 Table . For quantitative RT-PCR , SYBR green assays were performed . Constitutively expressed ACT1 was used as a control for all reactions . Calculations and statistical analyses were performed using StepOne Real-Time PCR System ( Applied Biosystems ) . Also , the contribution of the reductase/permease pathway to the pathogenesis of R . oryzae in the deferoxamine-treated mouse model was determined by comparing the virulence of R . oryzae with reduced copy number of FTR1 or with reduced expression of FTR1 due to RNA-i targeting FTR1 [24] , to their corresponding control strains of PyrF-complemented or R . oryzae transformed with the empty plasmid , respectively . Infection was carried out as above with a targeted inoculum of 1 x 103 spores . The primary efficacy endpoint was time to moribundity . All procedures involving mice were approved by the IACUC of the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center ( protocol # 11671-08 ) , according to the NIH guidelines for animal housing and care . Differences in FOB expression , growth rates , germination , reductase activity , iron uptake , and tissue fungal burden were compared by the nonparametric Wilcoxon rank-sum test . The nonparametric log-rank test was used to determine differences in survival times . A P value <0 . 05 was considered significant .	Deferoxamine is an iron-chelating agent often used to treat patients with acute iron poisoning , such as seen in dialysis patients with chronic renal failure . These patients are uniquely predisposed to a deadly fungal infection , called mucormycosis , because deferoxamine supplies iron that supports growth of fungi causing this infection . Apart from the important basic knowledge in delineating iron uptake mechanisms in cells , understanding how organisms causing mucormycosis obtain iron from ferrioxamine ( deferoxamine bound with iron ) is likely to develop strategies to treat mucormycosis infections in patients treated with deferoxamine . In this study we identified two cell surface receptors that bind ferrioxamine and facilitate iron uptake in Rhizopus oryzae , the most causative fungus of mucormycosis . These receptors are required for full virulence of R . oryzae in mice treated with deferoxamine . From genetic and biochemical studies it appears that the fungus binds ferrioxamine via these two receptors then liberates iron through a chemical modification step prior to transporting into the fungal cell without the internalization of deferoxamine .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2015	Fob1 and Fob2 Proteins Are Virulence Determinants of Rhizopus oryzae via Facilitating Iron Uptake from Ferrioxamine
Development is often strongly regulated by interactions among close relatives , but the underlying molecular mechanisms are largely unknown . In eusocial insects , interactions between caregiving worker nurses and larvae regulate larval development and resultant adult phenotypes . Here , we begin to characterize the social interactome regulating ant larval development by collecting and sequencing the transcriptomes of interacting nurses and larvae across time . We find that the majority of nurse and larval transcriptomes exhibit parallel expression dynamics across larval development . We leverage this widespread nurse-larva gene co-expression to infer putative social gene regulatory networks acting between nurses and larvae . Genes with the strongest inferred social effects tend to be peripheral elements of within-tissue regulatory networks and are often known to encode secreted proteins . This includes interesting candidates such as the nurse-expressed giant-lens , which may influence larval epidermal growth factor signaling , a pathway known to influence various aspects of insect development . Finally , we find that genes with the strongest signatures of social regulation tend to experience relaxed selective constraint and are evolutionarily young . Overall , our study provides a first glimpse into the molecular and evolutionary features of the social mechanisms that regulate all aspects of social life . Social interactions play a prominent role in the lives of nearly all organisms [1] and strongly affect trait expression as well as fitness [2–4] . Social interactions in the context of development ( e . g . parental care ) often strongly regulate developmental trajectories and resultant adult phenotypes , for example via transferred compounds such as milk in mammals [5 , 6] , milk-like secretions in arthropods [7 , 8] , and other forms of nutritional provisioning [9 , 10] . In many taxa including certain birds , mammals , and insects , care for offspring and the regulation of offspring development has shifted at least in part from parents to adult siblings , who perform alloparental care [11] . In eusocial insect societies , sterile nurse workers regulate the development of their larval siblings by modulating the quantity and quality of nourishment larvae receive [12–14] , as well as through the direct transfer of growth-regulating hormones and proteins [15 , 16] . At the same time , larvae influence nurse provisioning behavior via pheromones [17–20] and begging behavior [21 , 22] . In general , traits such as caregiving behavior that are defined or influenced by social interactions are the property of the genomes of multiple interacting social partners [2 , 14] . This has implications for both the mechanistic ( e . g . , molecular ) underpinnings of development and trait expression as well as the genetic basis of trait variation at the population level—i . e . how allelic variation in the genomes of interacting social partners affects trait variation [2 , 14] . Furthermore , because social traits are expressed in one individual but impact the fitness of other individuals , social behavior and socially-influenced traits experience distinct forms of selection , including kin selection and social selection [23 , 24] . Altogether , these distinct genetic features and patterns of selection are often thought to lead to distinct evolutionary features , such as rapid evolutionary dynamics in comparison to other traits [25–27] . In eusocial insects , previous studies show that variation in larval developmental trajectories and ultimate adult phenotypes ( including reproductive caste , body size , etc . ) depends on the combination of larval and nurse genotypes [28–34] . However , the identity of specific genes and molecular pathways that are functionally involved in the expression of social interactions ( e . g . , genes underlying nurse and larval traits affecting nurse-larva interactions ) and the patterns of molecular evolution for these genes have remained less well studied [15 , 16 , 35 , 36] . Transcriptomic studies are often used to identify sets of genes underlying the expression of particular traits by performing RNA-sequencing on individuals that vary in the expression of such traits . For example , in social insects , recent studies have compared the transcriptomes of workers that perform nursing versus foraging tasks [37–39] , or nurses feeding larvae of different stages or castes [35 , 40] . However , given the phenotypic co-regulation known to occur between interacting social partners ( here , nurses and larvae ) , it is likely that genes expressed in one social partner affect the expression of genes in the other social partner , and vice-versa , such that interacting social partners are connected by “social” gene regulatory networks [14 , 32 , 41 , 42] . Thus , identifying the genes important for social interactions such as nurse-larva interactions is only possible by studying the transcriptomic dynamics of both interacting social partners across a time series of interactions . To understand the transcriptomic basis of host-symbiont interactions , recent studies have reconstructed gene regulatory networks acting between hosts and symbionts by collecting and profiling the transcriptomes of each social partner across a time series of interactions [43–47] . Here , we use analogous methodology to study transcriptomic signatures of nurse-larva interactions in the pharaoh ant , Monomorium pharaonis . We sample a developmental time series of larvae as well as the nurses that feed each larval stage in this series , collecting individuals at the moment of interaction in order to identify genes involved in the expression of nurse-larva interactions , as well as genes affected by these interactions ( i . e . the full “social interactome” [14] ) . Pharaoh ant nurses tend to specialize on feeding young versus old larvae , and nurses feeding young versus old larvae show different transcriptomic profiles [40] . Larval transcriptomic profiles also change over development [48 , 49] . Given these results , we predicted that we would observe concerted changes in broad-scale gene expression in larvae and their nurses across larval development ( Fig 1 ) , reflective of the functional importance of nurse-larva interactions . Based on our dual RNA-seq data , we infer social gene regulatory networks acting between nurses and larvae to identify candidate genes predicted to have important social regulatory effects . Finally , we combine our measures of social regulatory effects with available population genomic data [48] to characterize the patterns of molecular evolution of genes underlying nurse-larva interactions . To elucidate transcriptomic signatures of nurse-larva interactions , we performed RNA-sequencing on worker-destined larvae across five developmental stages and nurses that fed larvae of each developmental stage ( termed “stage-specific” nurses; see S1 Fig for sampling scheme , S1 Table for list of samples ) , building upon a previously published dataset focused on caste development in M . pharaonis [48] . We hypothesized that if genes expressed in larvae regulate the expression of genes in nurse and vice versa , we would observe correlated expression profiles across larval development in larvae and nurses ( Fig 1 ) . As a biological control , we collected “random nurses” that we observed feeding any stage of larvae in the colony , and hence would not be expected to show correlated expression dynamics with larvae across the five larval developmental stages . We also collected reproductive-destined larvae , but unless clearly stated otherwise , all analyses were performed on only worker-destined larvae . We collected ten individuals of each sample type to pool into one sample , and we sequenced whole bodies of larvae but separated nurse heads and abdomens prior to sequencing . We grouped genes into co-expression profiles or “modules” using an algorithm designed to characterize gene co-expression dynamics across a short time series [50] , known as Short Time-Series Expression Mining ( STEM ) [51] . Each module represents a standardized pre-defined expression profile , consisting of five values that each represent the log2 fold-change between the given developmental stage and the initial ( L1 ) stage ( see S2 Fig; this results in a total of 81 possible modules ) . We sorted genes into the module that most closely represented their expression profile by Pearson correlation . We identified modules containing a greater than expected number of genes , where we formed null expectations using permutation tests across developmental stages [50] . We identified such significantly-enriched modules separately for larvae , stage-specific nurse heads , stage-specific nurse abdomens , random nurse heads , and random nurse abdomens . We focused on both parallel ( i . e . positive regulation or activation ) and anti-parallel ( i . e . inhibitory ) correlated expression patterns by identifying significantly-enriched modules that were shared in both larvae and nurses ( parallel ) , as well as significantly-enriched modules for which the inverse of the module was identified as significantly-enriched in the social partner ( anti-parallel ) . Larvae and stage-specific nurses shared many significantly-enriched modules ( S2 Table ) . These shared modules contained the majority of genes expressed in nurses ( 65% of genes in stage-specific nurse heads and 76% in abdomens ) . A substantial proportion of the larval transcriptome was also shared with stage-specific nurse heads ( 22% of larval genes ) and abdomens ( 60% of larval genes ) . Overall there was a widespread signature of correlated transcriptional patterns between stage-specific nurses and larvae across larval development ( Fig 2A–2D ) . These coordinated dynamics were dominated by parallel associations in nurse abdomens ( possibly reflecting shared metabolic pathways ) but anti-parallel associations in nurse heads ( possibly reflecting the social regulation of larval growth ) . In contrast to stage-specific nurses , random nurses ( our biological control ) shared few significantly-enriched modules with larvae ( S2 Table ) , and modules shared between random nurses and larvae contained significantly fewer genes than modules shared between stage-specific nurses and larvae ( Fig 2E; Wilcoxon test , P < 0 . 001 for all comparisons ) . Specifically , 2% of genes expressed in random nurse heads and 13% of genes expressed in random nurse abdomens were in modules shared with larvae; 3% of genes expressed in larvae were in modules shared with random nurse heads , and 2% of genes expressed in larvae were in modules shared with random nurse abdomens . Given that we observed transcriptome-wide patterns consistent with nurse-larva transcriptional co-regulation across larval development , we next identified the genes that might be driving these patterns ( see S3 Fig ) . We performed differential expression analysis to identify genes that varied in larval expression according to larval developmental stage , as well as genes that varied in nurse expression according to the developmental stage of larvae they fed . We identified 8125 differentially expressed genes ( DEGs ) in larvae ( 78% of 10446 total genes ) . We identified 2057 and 1408 DEGs in stage-specific nurse heads and abdomens , respectively , compared to 599 and 520 DEGs in random nurse heads and abdomens , respectively . We removed genes differentially expressed in both stage-specific and random nurses ( N = 272 DEGs in heads , N = 140 DEGs in abdomens ) , which might differ among our colony replicates due to random colony-specific effects that were not consistently associated with social regulation of larval development . After this removal , we retained the top 1000 DEGs , sorted by P-value , for each sample type other than random nurses ( larvae , stage-specific nurse heads , stage-specific nurse abdomens ) for social gene regulatory network reconstruction , reasoning that these genes were the most likely to be involved in the regulation of larval development . To infer putative gene-by-gene social regulatory relationships between nurses and larvae , we reconstructed gene regulatory networks acting within and between nurses and larvae ( S3 Fig ) . The output of regulatory network reconstruction is a matrix of connection strengths , which indicate the regulatory effect ( positive or negative ) one gene has on another , separated according to the tissue the gene is expressed in . To identify the most highly connected ( i . e . centrally located , upstream ) genes of regulatory networks , we calculated within-tissue connectivity and social connectivity by averaging the strength of connections across each connection a gene made , differentiating between within-tissue ( nurse-nurse or larva-larva ) and social connections ( nurse-larva ) ( Fig 1B ) . On average , within-tissue connectivity was higher than social connectivity ( Wilcoxon test; P < 0 . 001 in all tissues ) , and within-tissue connectivity was negatively correlated with social connectivity in each tissue ( S4 Fig ) . The top enriched gene ontology terms based on social connectivity in nurses were entirely dominated by metabolism ( S3 and S4 Tables; see also S5 Table for the top 20 genes by nurse social connectivity ) . While based on our data it is not possible to distinguish between genes that code for protein products that are actually exchanged between nurses and larvae versus genes that affect behavior or physiology within organisms ( Fig 1A ) , proteins that are known to be cellularly secreted represent promising candidates for the social regulation of larval development [40] . We downloaded the list of proteins that are known to be cellularly secreted from FlyBase [52] and used a previously-generated orthology map to identify ant orthologs of secreted proteins [40] . Genes coding for proteins with orthologs that are cellularly secreted in Drosophila melanogaster had higher social connectivity than genes coding for non-secreted orthologs in nurse heads ( Fig 3A; Wilcoxon test; P = 0 . 025 ) , though not for nurse abdomens ( P = 0 . 067 ) . For the most part , we have focused on broad patterns of nurse-larva gene coregulation . In this paragraph , we will highlight the potential social role of one of the genes with the highest social connectivity within nurse heads , giant-lens ( S6 Table; giant-lens is the 7th highest gene coding for secreted proteins by social connectivity in nurse heads ) . Giant-lens is an inhibitor of epidermal growth factor receptor ( EGFR ) signaling [53] , and giant-lens expression in nurse heads was negatively associated with the expression of the homolog of eps8 , human EGFR substrate 8 in larvae , most prominently seen in the spike in nurse giant-lens expression accompanied by a drop in larval eps8 expression at the end of larval development ( Fig 3B ) . Giant-lens was also used in regulatory network reconstruction in larvae ( i . e . it was one of the top 1000 DEGs ) , and giant-lens expression in larvae drops steadily throughout development ( S5 Fig; in contrast to the pattern of giant-lens expression in nurse heads ) . Interestingly , eps8 does not exhibit a similar peak and drop in expression level in reproductive-destined larvae in comparison to worker-destined larvae ( S6 Fig ) . It is important to note that these patterns were not seen for all genes in the EGFR pathway , and the results presented here cannot be taken as concrete evidence of EGFR regulation via social processes . Nonetheless , the mechanism illustrated here represents a tangible example of how nurse-larva interactions could function at the molecular level . To investigate the selective pressures shaping social regulatory networks , we used population genomic data from 22 resequenced M . pharaonis workers , using one sequenced M . chinense worker as an outgroup [48] . Using polymorphism and divergence data , we estimated gene-specific values of selective constraint , which represents the intensity of purifying selection that genes experience [54] . To identify genes disproportionately recruited to the core of social regulatory networks , we calculated “sociality index” as the difference between social connectivity and within-tissue connectivity for each gene . Sociality index was negatively correlated to selective constraint due to a positive correlation between within-tissue connectivity and constraint and a negative correlation between social connectivity and constraint ( Fig 4A–4C ) . Additionally , genes differed in sociality index according to their estimated evolutionary age , with ancient genes exhibiting lower sociality indices than genes in younger age categories ( Fig 4D ) . Finally , while evolutionary age and evolutionary rate appear to be somewhat empirically confounded [55] , selective constraint and evolutionary age were each independently associated with sociality index , based on a model including both variables as well as tissue ( GLM; LRT; evolutionary age: χ2 = 21 . 536 , P < 0 . 001; selective constraint: χ2 = 22 . 191 , P < 0 . 001 ) . In organisms with extended offspring care , developmental programs are controlled in part by socially-acting gene regulatory networks that operate between caregivers and developing offspring [14 , 42] . In this study , we sequenced the transcriptomes of ant nurses and larvae as they interacted across larval development to assess the effects of social interactions on gene expression dynamics . We found that large sets of genes ( i . e . modules ) expressed in ant larvae and their caregiving adult nurses show correlated changes in expression across development ( Fig 2 ) . The majority of nurse and larval transcriptomes was represented in these correlated modules , suggesting that the tight phenotypic co-regulation characterizing nurse-larva interactions over the course of larval development is also reflected at the molecular level . To characterize the overall network and evolutionary patterns of genes involved in nurse-larva interactions , we reverse engineered nurse-larva gene regulatory networks and calculated the “social connectivity” for each gene , defined as the sum of inferred social regulatory effects on all genes expressed in social partners . We found that genes with high social connectivity tended to have low within-individual connectivity ( S4 Fig; where within-individual connectivity is defined as the sum of inferred regulatory effects acting within a given tissue ) . Nurse-expressed genes with higher sociality indices ( i . e disproportionately higher social connectivity than within-individual connectivity ) tended to be evolutionarily young and rapidly evolving due to relaxed selective constraint ( Fig 4 ) . Genes with high social connectivity were enriched for a number of Gene Ontology ( GO ) categories associated with metabolism ( S3 and S4 Tables ) , consistent with the idea that molecular pathways associated with metabolism are involved in the expression of social behavior [56 , 57] . Previously , many of the proteins found to be widely present in social insect trophallactic fluid transferred from nurses to larvae were involved in sugar metabolism ( e . g . Glucose Dehydrogenase , several types of sugar processing proteins ) [15] . Along the same lines , many of the genes with with high social connectivity in our study are also annotated with terms associated with sugar metabolism ( S5 Table; e . g . Glycerol-3-phosphate dehydrogenase , Glucose dehydrogenase FAD quinone , Pyruvate dehydrogenase ) . Finally , we found that genes encoding for orthologs of cellularly-secreted proteins in Drosophila melanogaster ( possibly important for intercellular signaling ) tended to exhibit higher levels of social connectivity than their non-secreted counterparts ( Fig 3A ) . One gene that stands out in terms of being cellularly secreted and exhibiting a relatively high social connectivity is giant-lens , which inhibits EGFR signaling [53] . EGFR signaling affects eye and wing development [58] as well as body size in D . melanogaster [59] , caste development in the honey bee Apis mellifera [59 , 60] via the transfer of royalactin from nurses to larvae [59] , and worker body size variation in the ant Camponotus floridanus [61] . Further experimental work is necessary to ascertain whether giant-lens is actually orally secreted by nurses and transferred to larvae , but gene expression dynamics are consistent with the social transfer of giant-lens from nurses to larvae , followed by the inhibition of EGFR signaling at the end of larval development in worker-destined larvae ( Fig 3B ) . Importantly , this inhibition is not seen in reproductive-destined larvae ( S6 Fig ) . While caste in M . pharaonis is socially regulated in the first larval stage [49] , social inhibition of EGFR signaling could play a role in the regulation of worker body size [61] or secondary caste phenotypes such as wings [62 , 63] . In terms of broad evolutionary patterns , our study complements previous results suggesting genes with worker-biased expression tend to be rapidly evolving , evolutionarily young , and loosely connected in regulatory networks in comparison to genes with queen-biased expression [38 , 48 , 64–66] . Because pharaoh ant workers are obligately sterile , their traits are shaped indirectly by kin selection , based on how they affect the reproductive success of fertile relatives ( i . e . queens and males ) [23 , 67] . As a result , all-else-equal , genes associated with worker traits are expected to evolve under relaxed selection relative to genes associated with queen traits [68 , 69] . In general , the suite of genic characteristics commonly associated with worker-biased genes ( rapidly evolving , evolutionarily young , loosely connected ) are all consistent with relaxed selection acting on genes associated with workers [49] . Here , we show that within the worker caste , genes that appear to be functionally involved in the expression of social behavior ( i . e . nursing ) experience relaxed selective constraint relative to genes important for within-worker processes . Therefore , the combination of kin selection as well rapid evolution thought to be characteristic of social traits [25] likely act in concert to shape the labile evolutionary patterns commonly associated with worker-biased genes . Finally , it has also been suggested that plastic phenotypes such as caste recruit genes which were evolving under relaxed selection prior to the evolution of such plastic phenotypes [70–72] . Our results could also be consistent with this hypothesis , though the population genomic patterns we observe show that relaxed selective constraint is ongoing . In this study , we sought to reconstruct regulatory networks acting between nurses and larvae , beginning with the assumption that nurse gene expression changes as a function of the larval stage fed . This is more likely to be the case when nurses are specialized on feeding particular larval stages . According to a previous study , about 50% of feeding events are performed by specialists ( though note specialization is likely a continuous trait , and the 50% figure is the result of a binomial test ) [40] . Therefore , we expect our stage-specific nurse samples to comprise about 50% specialists . We also expect random nurse samples to contain 50% specialist nurses , but , crucially , the specialists should be relatively evenly divided among larval stages since random nurses were collected regardless of which larval stage they were observed feeding . Because our stage-specific nurse samples did not consist of 100% specialists , we expect that the signal of nurse-larva co-expression in our analysis is effectively diluted . In order to maximize the signal of nurse-larval co-expression dynamics , future studies would ideally focus entirely on specialists , as well as on tissues such as brains and the specific exocrine glands [73] known to be important for social behavior and communication . Despite these limitations , we were still able to observe transcriptomic signatures consistent with the social regulation of larval development . In this study , we uncovered putative transcriptomic signatures of social regulation and identified distinct evolutionary features of genes that underlie “social physiology” , the communication between individuals that regulates division of labor within social insect colonies [74 , 75] . Because we simultaneously collected nurses and larvae over a time series of interactions , we were able to elucidate the putative molecular underpinnings of nurse-larval social interactions . This is a promising approach that could be readily extended to study the molecular underpinnings of all forms of social regulation in social insect colonies , including regulation of foraging , regulation of reproduction , etc . . Furthermore , by adapting the methodology presented here ( i . e . simultaneous collection over the course of interactions followed by sequencing ) , the molecular mechanisms and evolutionary features of genes underlying a diverse array of social interactions , including courtship behavior , dominance hierarchy formation , and regulation of biofilm production could all be investigated . Overall , this study provides a foundation upon which future research can build to elucidate the genetic underpinnings and evolution of interacting phenotypes . To construct experimental colonies , we began by creating a homogenous mixture of approximately fifteen large source colonies of the ant Monomorium pharaonis . From this mixture , we created thirty total replicate experimental colonies of approximately equal sizes ( ~300–400 workers , ~300–400 larvae ) . We removed queens from ½ the study colonies to promote the production of reproductive-destined larvae . Reproductive caste is determined in M . pharaonis by the end of the first larval instar , likely in the egg stage [76] , and queen presence promotes culling of reproductive-destined L1 larvae . Removing queens halts this culling , but it is unknown which colony members actually perform such culling [76] . While we initially expected the presence of queens to impact the gene expression profiles of nurses , we detected 0 DEGs ( FDR < 0 . 1 ) between queen-present and queen-absent colonies for every sample type . This could indicate that nurses don’t perform culling and that worker developmental trajectories ( and nutritional needs ) are not appreciably different between queen-present and queen-absent colonies . Because queen presence did not substantially impact gene expression , in this study we pooled samples across queen-present and queen-absent colonies for all analyses . We pre-assigned colonies to one of five larval developmental stages ( labeled L1-L5 , where L1 and L2 refer to 1st-instar and 2nd-instar larvae and L3 , L4 , and L5 refer to small , medium , and large 3rd-instar larvae [77] ) . We identified larval stage through a combination of hair morphology and body size . L1 larvae are nearly hairless , L2 larvae have straight hairs and are twice the length of L1 larvae , and L3-L5 larvae have dense , branched hairs [78] . We separated 3rd-instar larvae into three separate stages based on body size [77] because the vast majority of larval growth occurs during these stages . We sampled individuals ( larvae as well as nurses ) across larval development time: beginning at the L1 stage , we sampled colonies assigned to each subsequent stage at intervals of 3–4 days , by the time the youngest larvae in colonies lacking queens were of the assigned developmental stage ( note that in colonies lacking queens , no new eggs are laid so the age class of the youngest individuals progressively ages ) . We sampled each colony once , according to the developmental stage we had previously assigned the colony ( e . g . for colonies that we labeled ‘L4’ , we waited until it was time to sample L4 larvae and nurses and sampled individuals from that colony at that time ) . From each colony , we sampled stage-specific nurses and worker-destined larvae , as well as random nurses from colonies with queens and reproductive-destined larvae from colonies without queens ( starting at the L2 stage , because at L1 caste cannot be distinguished [76 , 77] . Reproductive-destined larvae include both males and queens ( which cannot be readily distinguished ) , though samples are expected to be largely made up of queen-destined individuals given the typically skewed sex ratio of M . pharaonis [48] . See S1 Table for full sample list . For each time point in each assigned colony , we collected stage-specific nurses , nurses feeding larvae of the specified developmental stage ( L1 , L2 , etc ) . Concurrently , we collected random nurses , nurses we observed feeding a larva of any developmental stage . Rather than paint-marking nurses , we collected them with forceps as soon as we saw them feeding larvae . We collected random nurses as soon as we observed them feeding a larva of any developmental stage in the course of visually scanning the colony . We did not make an attempt to systematically collect nurses from different areas of the nest but did so haphazardly , such that the distribution of larval stages fed resembled overall colony demography . Nurses feed L1 and L2 larvae exclusively via trophallaxis ( i . e . liquid exchange of fluid ) , while nurses feed L3-L5 larvae both via trophallaxis and by placing solid food in larval mouthparts [79] . To get a representative sample of all types of nurses , we did not distinguish between nurses feeding liquid and solid food , though all L3-L5 samples contained a mixture of the two . After collecting nurses , we anaesthetized the colony using carbon dioxide and collected larvae of the specified developmental stage . All samples were flash-frozen in liquid nitrogen immediately upon sample collection . Note that workers in M . pharaonis are monomorphic [80] . We performed mRNA-sequencing on all samples concurrently using Illumina HiSeq 2000 at Okinawa Institute of Science and Technology Sequencing Center . Reads were mapped to the NCBI version 2 . 0 M . pharaonis assembly [38] , and we used RSEM [81] to estimate counts per locus and fragments per kilobase mapped ( FPKM ) for each locus . For further details on RNA extraction and library preparation , see [48] . We used an algorithm that categorizes genes based on their expression dynamics over time into a number of modules represented by pre-defined expression profiles [50]; see S2 Fig for workflow ) . To create modules , we started at 0 and either doubled , halved , or kept the expression level the same at each subsequent stage , resulting in 81 possible modules ( 333*3 = 81; four stages after L1 ) . To generate gene-specific expression profiles based on real results , we calculated the average log2 fold change in expression ( FPKM ) of the gene at each developmental stage compared to the initial expression level at stage L1 . We then assigned each gene to the closest module by Pearson correlation between gene expression profile and module expression profile [50] . To identify significantly-enriched modules , we generated null distributions of the number of genes present in each module ( based on permutation of expression over time ) , and retained modules with a significantly greater than expected number of genes based on these null distributions ( FDR < 0 . 05 after Bonferroni multiple correction [50] ) . We used the package EdgeR [82] to construct models including larval developmental stage and replicate and performed differential expression analysis for each sample type separately . We retained genes differentially expressed according to a nominal P-value of less than 0 . 05 ( i . e . no false discovery correction ) , as the purpose of this step was simply to identify genes that could be involved in interactions that shape larval development ( rather than spurious interactions arising from replicate-specific effects ) . See S1 Dataset for a list of all stage-specific nurse and larval differentially expressed genes . We normalized expression for each gene using the inverse hyperbolic sine transformation of FPKM . As input to the algorithm , we constructed “meta-samples” by combining expression data within the same replicate and time point from nurses and larvae and labeling genes according to the tissue they were expressed in , along the lines of host-symbiont studies [43 , 45] . We utilized the program GENIE3 [83 , 84] to construct two types of networks: those acting between larvae and nurse heads , and those acting between larvae and nurse abdomens . GENIE3 uses a random forest method to reconstruct regulatory connections between genes , in which a separate random forest model is constructed to predict the expression of each gene , with the expression of all other genes as predictor variables . The output of GENIE3 is a matrix of pairwise directional regulatory effects , where the regulatory effect of gene i on gene j is estimated as the feature importance of the expression of gene i for the random forest model predicting the expression of gene j ( i . e . regulatory effect is how important the expression of gene i is for determining the expression of gene j ) . These regulatory effects ( or strengths ) include both positive and negative as well as non-linear effects , though these different effect types are not distinguished . As a side note , a version of GENIE3 that was developed for time series data , dynGENIE3 [85] , does exist . However , we opted to utilize the original GENIE3 algorithm because we reasoned that the temporal spacing of developmental stages was likely too sparse for regulatory network reconstruction to incorporate time ( note also that the co-expression algorithm we used , STEM , was explicitly designed for short time series such as ours ) . While our method therefore does not explicitly incorporate temporal dynamics , we purposefully biased our results to emphasize larval development over differences between replicates by only utilizing genes differentially expressed across larval development ( or based on larval stage fed in the case of nurses ) . We repeated the entire regulatory reconstruction reconstruction process 1000 times and averaged pairwise connection strengths across runs , as the algorithm is non-deterministic . To capture the total effect of each gene on the transcriptome dynamics within tissues , we averaged the regulatory effects each gene had on all other 999 genes expressed in the same tissue ( “within-individual connectivity” ) . Similarly , to capture the effect each gene had on the transcriptome of social partners , we averaged regulatory effects each gene had on the 1000 genes expressed in social partners ( “social connectivity” ) . Previously , we performed whole-genome resequencing on 22 diploid M . pharaonis workers as well as one diploid M . chinense worker to serve as an outgroup [48] . We estimated selective constraint using MKtest2 . 0 [86] , assuming an equal value of alpha ( an estimate of the proportion of nonsynonymous substitutions fixed by positive selection ) across all genes . Selective constraint is the estimate of the proportion of nonsynonymous mutations that are strongly deleterious and thereby do not contribute to polymorphism or divergence [86] . Selective constraint is estimated using polymorphism data , so it represents the strength of purifying selection genes experience within the study population [54] . Phylostrata are hierarchical taxonomic categories , reflecting the most inclusive taxonomic grouping for which an ortholog of the given gene can be found [87–90] . We focused on distinguishing between genes that were evolutionarily “ancient” , present in non-insect animals , versus genes present in only insects , hymenopterans , or ants [49] . We constructed a database containing 48 hymenopteran available genomes , 10 insect non-hymenopteran genomes , and 10 non-insect animal genomes ( S2 Dataset ) . For outgroup genomes , we focused on well-annotated genomes which spanned as many insect orders and animal phyla as possible . Using this database , we estimated evolutionary age of genes based on the most evolutionarily distant identified BLASTp hit ( E-value 10−10 ) . We performed gene set enrichment analysis based on social connectivity for each gene in each tissue separately using the R package topGO [91] . We identified enriched gene ontology terms using Kolmogorov-Smirnov tests ( P < 0 . 05 ) . We performed all statistical analyses and generated all plots using R version in R version 3 . 4 . 0 [92] , aided by the packages “reshape2” [93] , “plyr” [94] , and “ggplot2” [95] .	Social interactions are fundamental to all forms of life , from single-celled bacteria to complex plants and animals . Despite their obvious importance , little is known about the molecular causes and consequences of social interactions . In this paper , we study the molecular basis of nurse-larva social interactions that regulate larval development in the pharaoh ant Monomorium pharaonis . We infer the effects of social interactions on gene expression from samples of nurses and larvae collected in the act of interaction across a developmental time series . Gene expression appears to be closely tied to these interactions , such that we can identify genes expressed in nurses with putative regulatory effects on larval gene expression . Genes which we infer to have strong social regulatory effects tend to have weak regulatory effects within individuals , and highly social genes tend to experience relatively weaker natural selection in comparison to fewer social genes . This study represents a novel approach and foundation upon which future studies at the intersection of genetics , behavior , and evolution can build .	[ "Abstract", "Introduction", "Results", "Discussion", "Conclusions", "Methods" ]	[ "medicine", "and", "health", "sciences", "abdomen", "medical", "personnel", "gene", "regulation", "health", "care", "developmental", "biology", "health", "care", "providers", "genome", "analysis", "genomics", "gene", "expression", "life", "cycles", "evolutionary", "genetics", "people", "and", "places", "professions", "anatomy", "gene", "regulatory", "networks", "nurses", "genetics", "transcriptome", "analysis", "population", "groupings", "biology", "and", "life", "sciences", "computational", "biology", "evolutionary", "biology", "larvae" ]	2019	Transcriptomic basis and evolution of the ant nurse-larval social interactome
Gene duplications are believed to facilitate evolutionary innovation . However , the mechanisms shaping the fate of duplicated genes remain heavily debated because the molecular processes and evolutionary forces involved are difficult to reconstruct . Here , we study a large family of fungal glucosidase genes that underwent several duplication events . We reconstruct all key ancestral enzymes and show that the very first preduplication enzyme was primarily active on maltose-like substrates , with trace activity for isomaltose-like sugars . Structural analysis and activity measurements on resurrected and present-day enzymes suggest that both activities cannot be fully optimized in a single enzyme . However , gene duplications repeatedly spawned daughter genes in which mutations optimized either isomaltase or maltase activity . Interestingly , similar shifts in enzyme activity were reached multiple times via different evolutionary routes . Together , our results provide a detailed picture of the molecular mechanisms that drove divergence of these duplicated enzymes and show that whereas the classic models of dosage , sub- , and neofunctionalization are helpful to conceptualize the implications of gene duplication , the three mechanisms co-occur and intertwine . In a seminal book , Susumu Ohno argued that gene duplication plays an important role in evolutionary innovation [1] . He outlined three distinct fates of retained duplicates that were later formalized by others ( for reviews , see [2] , [3] ) . First , after a duplication event , one paralog may retain the ancestral function , whereas the other allele may be relieved from purifying selection , allowing it to develop a novel function ( later called “neofunctionalization” ) . Second , different functions or regulatory patterns of an ancestral gene might be split over the different paralogs ( later called “subfunctionalization” [4] , [5] ) . Third , duplication may preserve the ancestral function in both duplicates , thereby introducing redundancy and/or increasing activity of the gene ( “gene dosage effect” [6] ) . Recent studies have shown that duplications occur frequently during evolution , and most experts agree that many evolutionary innovations are linked to duplication [7]–[10] . A well-known example are crystallins , structural proteins that make up 60% of the protein in the lenses of vertebrate eyes . Interestingly , paralogs of many crystallins function as molecular chaperones or glycolytic enzymes . Studies suggest that on multiple occasions , an ancestral gene encoding a ( structurally very stable ) chaperone or enzyme was duplicated , with one paralog retaining the ancestral function and one being tuned as a lens crystallin that played a crucial role in the optimization of eyesight [11] , [12] . The molecular mechanisms and evolutionary forces that lead to the retention of duplicates and the development of novel functions are still heavily debated , and many different models leading to Ohno's three basic outcomes have been proposed ( reviewed in [2] , [3] , [13] , [14] ) . Some more recent models blur the distinction between neo- and subfunctionalization [15] . Co-option models , for example , propose that a novel function does not develop entirely de novo but originates from a pre-existing minor function in the ancestor that is co-opted to a primary role in one of the postduplication paralogs [2] , [13] . Examples of such co-option models include the “gene sharing” or “Escape from Adaptive Conflict” ( EAC ) model [5] , [16]–[19] and the related “Innovation , Amplification and Divergence” ( IAD ) model [20]–[22] . The IAD model describes co-option as a neofunctionalization mechanism . A “novel” function arises in the preduplication gene , and increased requirement for this ( minor ) activity is first met by gene amplification ( e . g . , through formation of tandem arrays ) . After this , adaptive mutations lead to divergence and specialization of some of the duplicate copies . The EAC model , on the other hand , describes co-option rather as a subfunctionalization mechanism by which duplication allows a multifunctional gene to independently optimize conflicting subfunctions in different daughter genes . Another aspect in which various models differ is the role of positive selection . Some models emphasize the importance of neutral drift , while in other models adaptive mutations play an important role . For example , in the Duplication-Degeneration-Complementation ( DDC ) model of subfunctionalization [4] , degenerative mutations ( accumulated by neutral drift ) lead to complementary loss-of-function mutations in the duplicates , so that both copies become essential to perform all of the functions that were combined in the single preduplication gene . Whereas this type of subfunctionalization only involves genetic drift [4] , [8] , other subfunctionalization models , such as the EAC model , attribute an important role to positive selection for the further functional optimization of the postduplication paralogs [2] , [14] . There is a sharp contrast between the large number of detailed theoretical models of evolution after gene duplication , on the one hand , and the lack of clear experimental evidence for the various predictions made by these theories , on the other [2] . The key problem is the lack of knowledge about the functional properties of the ancestral , preduplication gene . Since these ancient genes and the proteins they encode no longer exist , many details in the chain of events that led from the ancestral gene to the present-day duplicates remain obscure . In most studies , the activities of the preduplication ancestor are inferred from unduplicated present-day outgroup genes that are assumed to have retained similar functional properties , but this is only an approximation . The central hurdle to surpass to obtain accurate experimental data on the evolution of gene duplicates involves rewinding the evolutionary record to obtain the sequence and activity of the ancestral proteins . Recent developments in sequencing and bio-informatics now enable us to reconstruct ancestral genes and proteins and characterize them in detail [23]–[31] . However , most ancestral reconstruction studies to date did not focus on the mechanisms that govern evolution after gene duplication . In this study , we used the yeast MALS gene family as a model system to gain insight in the molecular mechanisms and evolutionary forces shaping the fate of duplicated genes . The MALS genes encode α-glucosidases that allow yeast to metabolize complex carbohydrates like maltose , isomaltose , and other α-glucosides [32] , [33] . Several key features make this family ideal to study duplicate gene evolution . First , it is a large gene family encompassing multiple gene duplication events , some ancient and some more recent . Second , the present-day enzymes have diversified substrate specificities that can easily be measured [32] . Third , the availability of MALS gene sequences from many fungal genomes enabled us to make high-confidence predictions of ancestral gene sequences , resurrect key ancestral proteins , and study the selective forces acting throughout the evolution of the different gene duplicates . Fourth , the crystal structure of one of the present-day enzymes , Ima1 , has been determined [34] . Molecular modeling of the enzymes' binding pocket , combined with activity measurements on reconstructed and present-day enzymes , allowed us to investigate how mutations altered enzyme specificity and gave rise to the present-day alleles that allow growth on a broad variety of substrates . Combining these analyses , we were able to study the evolution and divergence of a multigene family to an unprecedented level of detail and show that the evolutionary history of the MALS family exhibits aspects of all three classical models of duplicate gene evolution proposed by Ohno ( gene dosage , neo- , and subfunctionalization ) . Some yeast species have evolved the capacity to metabolize a broad spectrum of natural disaccharides found in plants and fruits ( Figure 1 , tree adapted from [35] ) . The origin of this evolutionary innovation seems to lie in the duplication and functional diversification of genes encoding permeases and hydrolases [32] . The common Saccharomyces cerevisiae laboratory strain S288c , for example , contains seven different MALS genes ( MAL12 , MAL32 , and IMA1–5 ) , which originated from the same ancestral gene but allow growth on different substrates [32] , [33] . To understand how duplications led to functionally different MalS enzymes , we reconstructed , synthesized , and measured the activity of key ancestral MalS proteins . We used the amino acid ( AA ) sequences of 50 maltases from completely sequenced yeast species , ranging from Saccharomyces cerevisiae to Pichia and Candida species , for phylogenetic analysis and ancestral sequence reconstruction ( see Materials and Methods and Dataset S1 ) . A consensus amino-acid-based phylogenetic tree was constructed using MrBayes [36] under the LG+I+G model with four rate categories ( see Figure S1 , and see Materials and Methods for details ) . Trees constructed using MrBayes under other models of sequence evolution ( WAG , JTT ) generated largely identical results ( unpublished data ) . To further check the robustness of the AA tree inferred by MrBayes , we inferred a maximum likelihood ( ML ) tree under the LG+I+G model using PhyML ( Figure S2 ) [37] . With the exception of a few recent splits in the topology , the MrBayes and PhyML trees agree , increasing our confidence in the constructed tree . Codon-based tree reconstruction using MrBayes yielded similar results ( see further ) . Additional tests were performed to control for potential long branch attraction ( LBA ) artifacts , specifically to check the placement of the K . lactis branch as an outgroup to the Saccharomyces and Lachancea clades ( see Text S1 and Figures S3 , S4 , S5 , S6 ) . Next , we reconstructed the AA sequence of the ancestral maltases under several commonly used models of protein evolution ( LG , WAG , JTT; see Materials and Methods ) . All models support roughly the same ancestral protein sequences , increasing our confidence in the reconstructed ancestral sequences . In particular , all models identified the same residues for variable sites within 10 Å of the active center ( based on the crystal structure of the Ima1 protein ) , which are likely relevant sites with respect to enzymatic activity . The residues for a few other sites located further away from the active pocket vary between different models , but differences generally involve biochemically similar AAs ( see Table S1 ) . Synthesis of the ancestral enzymes was based on the reconstructed ancestral sequences obtained with the JTT model . For ambiguous residues ( i . e . , sites for which the probability of the second-most likely AA is >0 . 2 ) within 7 . 5 Å of the binding pocket , we constructed proteins containing each possible AA , while for ambiguous residues outside 7 . 5 Å we considered only the most likely AA . There is one ambiguous residue close to the active center in the ancestral proteins ancMalS and ancMal-Ima , namely residue 279 ( based on Saccharomyces cerevisiae S288c Ima1 numbering ) . We therefore synthesized two alternative versions of these proteins , one having G and one having A at position 279 . Whereas these alternative proteins show different activities for some substrates , the relative activities are similar and our conclusions are robust . Sequences for these reconstructed enzymes can be found in Dataset S2 . In the main figures , we show the variant with the highest confidence . Enzymatic data for all variants can be found in Table S2 . The activity of all resurrected ancestral enzymes was determined for different substrates ( see Materials and Methods , Text S1 , and Figure 2 ) . The results indicate that the very first ancestral enzyme , denoted as ancMalS , was functionally promiscuous , being primarily active on maltose-like substrates but also having trace activity on isomaltose-like sugars . The activity data presented in Figure 2 show how this promiscuous ancestral protein with relatively poor activity for several substrates evolved to the seven present-day enzymes that show high activity for a subset of substrates , and little or no activity for others . This confirms the existence of two functional classes of MalS enzymes that originated from ancient duplication events . First , Mal12 and Mal32 show activity against maltose-like disaccharides often encountered in plant exudates , fruits , and cereals , like maltose , maltotriose , maltulose , sucrose , and turanose ( a signaling molecule in plants ) . The five MalS enzymes of the second class ( Ima1–5 ) , which in fact result from two independent ancient duplication events giving rise to the Ima1–4 and Ima5 clades , show activity against isomaltose-like sugars including palatinose ( found in honey [38] ) and isomaltose . Differences in hydrolytic activity between members of the same ( sub ) class are more subtle or even absent , which is not surprising since some of these recent paralogs are nearly identical ( Mal12 and Mal32 , for example , are 99 . 7% identical on the AA level ) . The more recent ancestral enzymes also show a similar split in activity , with some enzymes ( ancMal ) showing activity towards maltose-like substrates , and others ( ancIma1–4 ) towards isomaltose-like substrates . Moreover , activity on isomaltose-like sugars ( isomaltose , palatinose , and methyl-α-glucoside ) changes in a coordinate fashion when comparing different enzymes , and the maltose-like sugars also group together . Careful statistical analysis reveals that the maltose-like group consists of two subgroups ( maltose , maltotriose , maltulose , and turanose , on one hand , and sucrose , on the other ) that behave slightly different , showing that the enzymes show quantitative differences in the variation of specificity towards these substrates ( two-way ANOVA analysis followed by Games-Howell test on log-transformed kcat/Km values; p values can be found in Table S3 ) . Interestingly , the most ancient ancestral enzymes do not show a clear split in activity towards either maltose-like or isomaltose-like sugars after duplication , and the transition of ancMalS to ancMal-Ima even shows an increase in activity for all substrates . This suggests that ( slight ) optimization for all substrate classes simultaneously was still possible starting from ancMalS . A clear divergence of both subfunctions occurred later , after duplication of ancMal-Ima , resulting in ancMal and ancIma1–4 . AncMal shows a significant increase in activity on maltose-like sugars accompanied by a significant drop in activity on isomaltose-like sugars compared to ancMal-Ima; and the reverse is true for ancIma1–4 ( see also Table S3 for exact p values for each enzyme–enzyme comparison on the different sugars tested ) . Together , this illustrates how , after duplication , the different copies diverged and specialized in one of the functions present in the preduplication enzyme . In two separate instances , a major shift in specificity is observed , from maltose-like sugars to isomaltose-like sugars ( transition from ancIMA5 to IMA5 , and from ancMAL-IMA to ancIMA1–4 ) . The shift in activity from ancMAL-IMA to ancIMA1–4 is particularly pronounced . The ancMAL-IMA enzyme hydrolyzes maltose , sucrose , turanose , maltotriose , and maltulose but has hardly any measurable activity for isomaltose and palatinose , whereas ancIMA1–4 can only hydrolyze isomaltose and palatinose ( and also sucrose ) . For the evolution of the maltase-like activity from the ancestral MalS enzyme to the present-day enzyme Mal12 , we see a 2-fold increase in kcat and a 3-fold decrease in Km for maltose , indicating an increase in both catalytic power and substrate affinity for this sugar . For the evolution of isomaltase-like activity in the route leading to Mal12 , kcat decreases more than 3-fold for methyl-α -glucoside . kcat for isomaltose and palatinose and the affinity for isomaltose and palatinose are so low that they could not be measured ( see Table S2 for the exact values of kcat and Km for each enzyme and each sugar; results of two-way ANOVA analysis followed by Games-Howell test comparing log-transformed kcat/Km values for different enzymes on each of the sugars can be found in Table S3 ) . To further explore the evolution of MALS genes and consolidate the measured activities of the ancestral enzymes , we expressed and purified additional present-day α-glucosidase alleles from other yeast species and measured their activities ( Figure 3 ) . We focused primarily on enzymes that are directly related to one of the ancestral proteins but did not undergo any further duplication events , and therefore have a higher probability of having retained a similar activity as their ( sub ) class ancestor . Indeed , the only present-day MalS enzyme of the yeast L . elongisporus has a broad but relatively weak activity comparable to the very first ancestral MalS enzyme , providing extra support for the accuracy of our ancestral reconstructions . Also in K . lactis , which contains two Mal alleles , one of the paralogs retains the broad specificity of ancMalS . The other paralog ( GI:5441460 ) has a deletion of five AAs close to the active pocket that likely explains the general lack of activity of this enzyme ( see Materials and Methods and Figure S7 ) . In contrast , yeasts that show multiple duplication events , like K . thermotolerans and S . cerevisiae , exhibit specialization , with some enzymes showing only activity for maltose-like substrates and others for isomaltose-like substrates . Moreover , the activities ( maltase- or isomaltase-like ) of homologs in S . cerevisiae and K . thermotolerans derived from the same intermediate ancestor are often similar , except in the IMA5 clade . Here , the K . thermotolerans and S . cerevisiae homologs have very different substrate specificities , indicating species-specific evolutionary trajectories and/or reciprocal paralog loss in the different species ( Figures 3 and 4 ) . Next , we investigated which mutations underlie the observed functional changes . We used the recently resolved crystal structure of Ima1 ( pdb entry 3A4A ) [34] as a template to study the molecular structure of the enzymes' substrate binding pocket ( see Materials and Methods ) . All enzymes share a highly conserved molecular fold , suggesting that changes in activity or substrate preference are likely caused by mutations in or around the substrate binding pocket . We identified nine variable AA residues within 10 Å of the center of the binding pocket in the various paralogs ( Figure 4 , right panel ) . Site-directed mutagenesis and crystallographic studies by Yamamoto et al . confirmed the importance of several of these residues for substrate specificity in the present-day Ima1 protein [39] , [40] . In particular , Yamamoto et al . [40] characterized the influence of residues 216-217-218 ( Ima1 numbering ) , which covary perfectly with each other and with the observed substrate specificity shifts across the phylogeny presented in Figure 4 . Sequence co-evolution analysis on 640 MAL12 homologs identified another cluster of three co-evolving residues among these nine residues ( positions 218 , 278 , and 279 in Ima1 ) , which we investigate here in detail . Together with residues 216 and 217 , residues 218 , 278 , and 279 seem to contribute to the activity shift observed in the evolution of Ima1–4 ( see Figures 4–6 , Figure S8 , and Supplementary Information for details ) . Molecular modeling of the mutations at 218-278-279 on the branch leading to ancIma1–4 ( see Figure 4 ) suggests that the change from alanine to glutamine at residue 279 shifts the binding preference of the pocket from maltose-like to isomaltose-like sugars ( Figure 5B–E ) . The two co-evolving residues at positions 218 and 278 are spatially close to AA 279 and cause subtle structural adaptations that help to better position the Q residue . To investigate if changes at all three positions are necessary for the observed shift in substrate specificity from ancMAL-IMA to ancIMA1–4 and to investigate the possible evolutionary paths leading to these three interdependent mutations , we synthesized all possible intermediate ancIMA1–4 enzyme variants with mutations at positions 218 , 278 , and 279 . We subsequently expressed , purified , and measured activity of these enzyme variants . Figure 5F depicts the results of these enzyme assays and shows that these residues indeed affect substrate specificity , with the largest shift depending on the A to Q change at position 279 , as expected from structural analysis . For one mutational path ( GVA to GVQ to SVQ to SMQ ) , we observe a gradual increase in activity towards isomaltose and palatinose , demonstrating that there is a mutational path that leads to a consistent increase in isomaltase activity without traversing fitness valleys . Moreover , in keep with the stabilizing role of the mutations at positions 218 and 278 , the A to Q change at position 279 along this path takes place before the two other mutations at positions 218 and 278 ( Figure 5F ) . Besides allowing the development of isomaltase activity in the Ima proteins , duplication also permitted further increase of the major ancestral function ( hydrolysis of maltose-like sugars ) in Mal12 and Mal32 . Structural analysis reveals that this increase in maltase activity , from ancMalS to Mal12/32 , is due to mutations D307E and E411D ( Figure 6G–J ) . These mutations increase the fit for maltose-like substrates but also completely block the binding of isomaltose-like substrates ( Figure 6 ) . Similar to what is seen for the evolution of AncMal-Ima to AncIma1–4 , changes that increase the binding stability of one type of substrate cause steric hindrance that prevents binding of the other class of substrates . These signs of incompatibilities between substrates indicate that it is difficult to fully optimize one enzyme for both maltose-like and isomaltose-like substrates , with the highly suboptimal ancMalS being a notable exception . After partial optimization of ancMalS , duplication of ancMAL-IMA likely enabled further optimization of the conflicting activities in separate copies . Interestingly , the transition from AncMalS to Ima5 shows a similar shift in substrate specificity as the transition of AncMal-Ima to AncIma1–4 . However , the residue at position 279 , a key factor in the evolution of AncMal-Ima to AncIma1–4 , remains unaltered in the evolution of AncMalS to Ima5 . Instead , L219 , a residue located proximal to position 279 , has changed into M219 in the Ima5 enzyme ( Figure 6C–F ) . How can such seemingly very different mutations yield a similar change in substrate specificity ? Structural analysis shows that the L-to-M mutation at position 219 in Ima5 causes a very similar structural change as the G279Q change in AncIma1–4 ( Figure 6 ) , indicating that different evolutionary routes may produce a similar shift in activity . In both cases , the evolution of isomaltase-like activity involved introducing a residue that can stabilize isomaltose-like substrates but causes steric hindrance for maltose-like sugars in the binding pocket . Based on the phylogeny of binding pocket configurations and on our enzyme activity tests , this functional shift in the IMA5 clade most likely occurred after a duplication in the common ancestor of S . kluyveri and S . cerevisiae ( Figures 3–4 ) . Next , we investigated the role of selective pressure during the different evolutionary transitions . We used MrBayes to construct a codon-based phylogeny under a GTR codon model of evolution , including 32 MALS genes that share the same nuclear genetic code . The resulting codon-based phylogeny was the same as the AA-based phylogeny generated using the LG+I+G protein model for all 50 sequences , apart from two exceptions in the ancIMA1–4 clade . First , S . mikitae IFO1815 c789 and S . paradoxus N45 branch off separately from S . kudriavzevii IFO1802 c1888 instead of together . Second , S . kudriavzevii IFO1802 c1565 now branches off separately instead of multifurcating with S . mikitae IFO1815 c633 and the branch leading to the S . cerevisiae IMA2–4 genes . Relative branch lengths between genes were similar to the branch lengths calculated under protein models of evolution . The topology of the codon-based tree is presented in Figure 4 . GA Branch analysis [41] identified a branch class with an elevated ω ( dN/dS ) rate ( ω = 0 . 66 ) but did not detect branch classes with ω>1 that would be considered strong proof for positive selection ( see Materials and Methods and Figure 4 ) . These results , combined with our activity test results and the observed sequence configurations around the active center , suggest , however , that positive selection might have been operating on specific sites in three specific postduplication branches associated with enzyme activity shifts , namely the ancIMA1–4 , ancIMA5b , and ancMAL branches , indicated with arrows on Figure 4 . We used the modified branch-site model A implemented in PAML to assess positive selection along these branches ( see Materials and Methods ) [42] . Results are presented in Table S4 . For both the ancIMA1–4 and ancIMA5b branches , p values and parameter estimates suggest that a proportion of sites has strongly elevated ω values , consistent with the GABranch results . On the branch from ancMAL-IMA to ancIMA1–4 , four sites show signs of positive selection , with a posterior Bayes Empirical Bayes ( BEB ) probability >0 . 95 , of which two , 216 and 279 , are within 10 Å of the active center and known to be important for substrate specificity . On the ancIMA5b branch , four sites show signs of positive selection ( BEB>0 . 95 ) , including again site 216 . For ancMAL , the null model ( no positive selection ) was not rejected at the 95% significance level . Both the corresponding parameter estimates and results of the GABranch analysis , however , suggest relaxation of purifying constraints on this branch . To get more support for the PAML branch-site test results , we performed an additional analysis using an alternative branch-site method that was recently implemented in the HyPhy package [43] . This method identified in total seven branches that possibly experienced positive selection: ancIMA1–4 ( p<0 . 0001 ) , ancIMA5b ( p = 0 . 0232 ) , ancMALS ( p = 0 . 0228 ) , S . kluyveri SAKL0A05698g ( p<0 . 0001 ) , K . thermotolerans GI: 255719187 ( p<0 . 0001 ) , the branch leading from ancIMA5 to the ancIMA5b branch ( p = 0 . 0168 ) , and finally the branch leading up to S . cerevisiae IMA2 , IMA3 , IMA4 , and YPS606 within the ancIMA1–4 clade ( p = 0 . 0353 ) . In other words , the ancMALS , ancIMA1–4 , and ancIMA5b branches are suggested to have evolved under positive selection , together with four other branches . The branch-site method implemented in HyPhy currently does not allow the identification of specific sites that may have evolved under positive selection on these branches . Together , our analyses indicate that some residues near the active pocket , in particular the key residues 216 and 279 that determine substrate specificity ( see above ) , may have experienced positive selection in the postduplication lineages leading to isomaltose-specific enzymes . It should be noted , however , that the specificity and sensitivity of the currently available methods for detecting positive selection , in particular branch-site methods , is heavily debated [42] , [44]–[47] . Possible pitfalls include fallacies in the assumption that synonymous substitutions are neutral , a reported increase in the number of false positives due to sampling errors when the number of ( non ) synonymous substitutions and sequences is low , and potential inadequacies in the null and alternative models that are being compared , leading to difficulties with completely ruling out other explanations for perceived positive selection . For these reasons , the positive selection test results reported here should be approached as indications rather than definitive proof . The previous results show how duplication of a promiscuous ancestral enzyme with limited activity towards two substrate categories allowed the evolution of separate enzyme clades that each show increased activities for a specific subset of substrates . The functional diversification of the different clades ensures their retention . However , why are recent , near-identical duplicates such as MAL12 and MAL32 conserved ? To investigate if selective pressure might protect the MAL12/MAL32 duplicates , we determined the fitness effect of inactivating each of them . The results in Figure S9 show that strains lacking just one of the MAL12 and MAL32 paralogs show a considerable fitness defect compared to a wild-type strain when grown on maltose . These results suggest that gene dosage may play a primary role in preserving these recent paralogs [6] . Dosage effects increasing maltase and/or isomaltase activity may also have played a role after the earliest MALS duplications , before the duplicates were optimized for different activities . One of the major issues in the field of molecular evolution is the plethora of theoretical models and variants of models concerning the evolution of gene duplicates , with few of the claims supported by solid experimental evidence . On many occasions , inherent properties of the evolutionary process make it extremely hard to find or generate experimental evidence for a given model . However , recent developments in genome sequencing , evolutionary genomics , and DNA synthesis open up exciting possibilities . Using these new opportunities , we were able to resurrect ancient MALS genes and the corresponding enzymes and provide a detailed picture of the evolutionary forces and molecular changes that underlie the evolution of this fungal gene family . The MALS gene family is an ideal model for the study of duplicate gene evolution , since it underwent several duplication events and encodes proteins for which we could accurately measure different activities . The availability of multiple fungal genome sequences provided sufficient data to robustly reconstruct ancestral alleles and study the selective forces that propelled divergent evolution of the paralogs . Additionally , the existence of a high-quality crystal structure of one of the present-day enzymes made it possible to predict the functional effects of mutations and to study the mechanistic basis of suspected adaptive conflicts between the maltase-like and isomaltase-like subfunctions . Our results paint a complex and dynamic picture of duplicate gene evolution that combines aspects of dosage selection and sub- and neofunctionalization ( see Figure 7 ) . The preduplication ancMalS enzyme was multifunctional and already contained the different activities found in the postduplication enzymes ( the basic idea of subfunctionalization ) , albeit at a lower level . However , the isomaltase-like activity was very weak in the preduplication ancestor and only fully developed through mutations after duplication ( increase of kcat/Km with one order of magnitude for isomaltase-like substrates from ancMalS to Ima1 ) , which resembles neofunctionalization . The ancestral maltase-like activity also improved substantially but to a lesser extent ( factor 6 . 9 on average from ancMalS to Mal12 ) , which therefore perhaps fits better with the subfunctionalization model . Moreover , our activity tests on Mal12/Mal32 mutants indicate that gene dosage may also have played a role in preserving MALS paralogs , especially right after duplication . This may not only have been the case for the recent MAL12–32 and IMA3–4 duplications but also for more ancient duplications involving multifunctional ancestors . In summary , whereas the classical models of dosage , sub- , and neofunctionalization are helpful to conceptualize the implications of gene duplication , our data indicate that the distinction between sub- and neofunctionalization is blurry at best and that aspects of all three mechanisms may intertwine in the evolution of a multigene family . Although it is difficult to classify our results decisively under one of the many models of evolution after gene duplication , most of our findings agree with the predictions of the “Escape from Adaptive Conflict” ( EAC ) model [5] , [16] , [17] , [19] , a co-option-type model in which duplication enables an organism to circumvent adaptive constraints on a multifunctional gene by optimizing the subfunctions separately in different paralogs . The EAC model makes three key predictions: ( i ) the ancestral protein was multifunctional , ( ii ) the different subfunctions could not be optimized simultaneously in the ancestral protein ( or at least not in an evolutionarily easily accessible way ) , and ( iii ) after duplication , adaptive changes led to optimization of the different subfunctions in separate paralogs [13] , [16] , [48] . In general , our findings fit with these predictions: ( i ) we find that several of the ancestral preduplication maltase enzymes ( ancMALS , ancMAL-IMA , and ancIMA5 ) were multifunctional; ( ii ) we provide evidence , through molecular modeling and activity tests of present-day enzymes , ancestors , and potential intermediates , that the maltase and isomaltase functions are difficult to optimize within one protein ( but see also below ) ; and ( iii ) we find that duplication resolved this adaptive conflict , and we find indications that positive selection might have driven key changes that optimized the minor isomaltase-like activity of the preduplication enzyme in one paralog , while the major maltase-like activity was further optimized in the other paralog . Figure 2 and the statistical analysis in Table S3 indicate that the activity of the different enzymes changes significantly at certain points along the evolutionary path . Interestingly , the overall image that emerges suggests that the enzymes developed activity towards either maltose-like or isomaltose-like sugars , but not both . This pattern is most clear in the evolution of ancMal-Ima to ancMal and ancIma1–4 . The postduplication improvement of the different activities present in the ancestral allele , with each of the new copies displaying increased activity for one type of substrate and concomitantly decreased activity towards the other substrate class , could be indicative of trade-offs in the evolution of the MALS gene family . However , the word “trade-off” implies that the two incompatible functions are both under selection , which is difficult to prove for the ancient enzymes . Moreover , our results indicate that for the ancient ancMalS enzyme , it is possible to simultaneously increase the activity towards both maltose-like and isomaltose-like substrates . Together , our analyses show that it is possible to optimize ( to a certain extent ) one function of a multifunctional enzyme without significantly reducing the other ( minor ) activity . However , analysis of the complete evolutionary path and molecular modeling of the active pockets of the enzymes shows that full optimization of both functions in a single enzyme is difficult to achieve , due to steric hindrance for one substrate class when fully optimizing the active pocket for binding of the other substrate type . This problem can be most easily overcome by duplication of the enzyme , allowing optimization of the different subfunctions in different paralog copies , as can be seen in the transition of ancMal-Ima to ancMal and ancIma1–4 . While most aspects of our data fit with the EAC model , some results are more difficult to reconcile with the EAC theory . Specifically , one of the pillars of the EAC model is that positive selection drives the specialization of both paralogs after duplication . While our data demonstrate that duplication of ancMAL-IMA has led to optimization of both subfunctions in different duplicate lineages ( maltase-like activity in ancMAL and isomaltase-like activity in ancIMA1–4 ) , our selection tests only reveal indications of positive selection in the ancIMA1–4 lineage but not in the ancMAL lineage . Moreover , as discussed above , positive selection is difficult to prove [44] , [49] , and we cannot exclude the possibility of both false positive and false negative artifacts . Recently , some other likely examples of the EAC mechanism have been described [16] , [17] , [50]–[52] . These studies also presented plausible arguments for ancestral multifunctionality , adaptive conflict , and/or adaptive optimization of subfunctions in different paralogs , but as in the present case , none could provide strong experimental evidence for all three predictions made by the EAC model [48] , [53] . Instead of classifying the evolutionary trajectory of particular gene duplicates into one of the many models for gene duplication , it may prove more useful to distill a more general picture of duplicate evolution across a gene family that includes aspects of dosage selection , and sub- and neofunctionalization , like the one depicted in Figure 7 . Our study is the first to investigate multiple duplication events in the same gene family in detail . Interestingly , we found that evolution has taken two different molecular routes to optimize isomaltase-like activity ( the evolution of ancMAL-IMA to ancIMA1–4 and ancIMA5 to IMA5 ) . In both cases , only a few key mutations in the active pocket are needed to cause shifts in substrate specificity . Some of these key mutations exhibit epistatic interactions . For example , the shift in substrate specificity occurring on the path from ancMAL-IMA to ancIMA1–4 depends in part on mutations at three co-evolving positions ( 218 , 278 , and 279 ) , but only one mutational path ( 279-218-278 ) shows a continuous increase in isomaltase-like activity . Interestingly , there is also a different path in the opposite direction ( 218-279-278 ) that shows a continuous increase in the ancestral maltase-like activity . This implies that the complex co-evolution at these three positions may be reversible . Interestingly , a recent study of the evolutionary history of plant secondary metabolism enzymes also identified AA changes that appear to be reversible [51] , in contrast to the situation for , for example , glucocorticoid receptor evolution , where evidence was found for an “epistatic ratchet” that prevents reversal to the ancestral function [54] . It is tempting to speculate that complex mechanisms like those driving the evolution of the MALS gene family may be a fairly common theme . Many proteins display some degree of multifunctionality or “promiscuity” [55]–[57] , just like the ancestral ancMal enzyme . Moreover , directed “in vitro” protein evolution experiments have shown that novel protein functions often develop from pre-existing minor functions [58] , [59] . Although the different functions within an enzyme often exhibit weak trade-offs , allowing optimization of the minor activity without affecting the original function of the enzyme [55] , [59] , [60] , this may not always be the case . If there are stronger trade-offs between different subfunctions , duplication may enable the optimization of the conflicting functions in different paralogs . While it is difficult to obtain accurate dating of the various duplication events , the duplication events studied here appear to postdate the divergence of Saccharomyces and Kluyveromyces clades , estimated to have occurred 150 mya [61] , but predate the divergence of Saccharomyces and Lachancea and the yeast whole genome duplication , about 100 mya . MALS diversification may thus have happened around the appearance and spread of angiosperms ( Early Cretaceous , between 140 and 100 mya [62] ) and fleshy fruits ( around 100 mya ) . Tentative dating results can be found in Table S6 , but these should be approached with caution ( see Text S1 ) . The major shift in the earth's vegetation caused by the rise of the angiosperms almost certainly opened up new niches , and it is tempting to speculate that duplication and diversification of the MALS genes may have allowed fungi to colonize new niches containing sugars hydrolyzed by the novel Mal ( Ima ) alleles . In other words , the availability of novel carbon sources in angiosperms and fleshy fruits could have provided a selective pressure that promoted the retention of MALS duplicates and the ensuing resolution of adaptive conflicts among paralogs . In total , the nucleotide and protein sequences of 169 extant maltases were collected for yeast species ranging from Saccharomyces cerevisiae to Pichia and Candida species . For Kluyveromyces thermotolerans , Saccharomyces kluyveri , and Kluyveromyces lactis , sequences were downloaded from Génolevures ( www . genolevures . org ) . Sequences for many of the Saccharomyces cerevisiae and Saccharomyces paradoxus genes were obtained from the sequence assemblies provided by the Wellcome Trust Sanger Institute ( http://www . sanger . ac . uk/research/projects/genomeinformatics/sgrp . html ) . All of the remaining extant maltase sequences were downloaded from NCBI ( www . ncbi . nlm . nih . gov/ ) . Sequences with greater than 92% pairwise protein sequence similarity to other sequences in the dataset were removed to reduce the phylogenetic complexity . All seven Saccharomyces cerevisiae S288c alleles were kept , however , yielding a final dataset of 50 sequences ( see also Dataset S1 ) . We used ProtTest 2 . 4 [63] to score different models of protein evolution for constructing an AA-based phylogenetic tree . All possible models with all improvements implemented in the program were taken into account . An initial tree was obtained by Neighbor-Joining ( BioNJ ) , and the branch lengths and topology were subsequently optimized for each evolutionary model independently . The LG+I+G model came out as best with a substantial lead over other protein models using −lnL , AIC , and AICc selection criteria ( AICc = 43 , 061 . 26 with AICw = 1 . 00 , while the second best model was WAG ( +I+G ) with AICc = 43 , 158 . 00 and AICw = 0 . 00 ) . Consequently , an AA-based phylogeny for the 50 sequences was determined using MrBayes 3 . 1 . 2 [36] with a LG invariant+gamma rates model ( four rate categories ) . Since the LG model is not implemented by default in MrBayes , we used a GTR model and fixed the substitution rate and state frequency parameters to those specified by the LG model . The BMCMC was run for 106 generations , sampling every 100 generations , with two parallel runs of four chains each . A burn-in of 2 , 500 samples was used , and the remaining 7 , 501 samples were used to construct a 50% majority-rule consensus phylogeny ( Figure S1 ) . The AWTY program [64] was used to check proper MCMC convergence under the given burn-in conditions . MrBayes AA tree constructions were also performed under other evolutionary models ( WAG , JTT ) . Additional tests were performed to exclude Long Branch Attraction ( LBA ) artifacts ( see Text S1 ) . We also inferred a maximum likelihood ( ML ) tree using PhyML under the LG+I+G model with four rate categories [37] . The initial tree was again obtained by BioNJ; tree topology , branch lengths , and rate parameters were optimized in a bootstrap analysis with 1 , 000 replicates . We also used MrBayes to construct a codon-based phylogeny , using a GTR codon model of evolution . The original dataset of 50 sequences contained 18 sequences for species that employ the alternative yeast nuclear genetic code ( all of them outgroup species ) . These sequences were removed from the dataset , resulting in a reduced dataset of 32 sequences . The codon alignment was obtained by translating the AA alignment obtained earlier . BMCMC analysis and consensus phylogeny construction were performed as described above for the AA trees . We contrasted models that did and did not allow for ω rate variation ( i . e . , the “Equal” versus “M3” codon model in MrBayes ) . AWTY analysis indicated that the latter was not able to converge properly , so we used the results of the Equal model . The PAML package [65] was used to infer the posterior AA probability per site in the ancestors of interest under several commonly used models of protein evolution ( LG , WAG , JTT ) , using the corresponding Bayesian consensus phylogenies . Both marginal and joint probability reconstructions were performed . The marginal reconstructions are presented in Table S1 . Protein sequences resulting from marginal reconstructions under the JTT model were used to synthetize ancestral enzymes . We performed tests for positive selection on the codon-based phylogeny obtained as described above . Various branch methods and branch-site methods included in the PAML [65] and HyPhy [66] packages were used . Co-evolving residues in the MALS gene family were detected using the framework described by [71] . The NCBI Blast server was used to collect Saccharomyces cerevisiae S288c MAL12 maltase homologs , with an E-value <10e-70 , resulting in a set of 1 , 211 sequences . Proteins were removed that were shorter than 400 AAs , longer than 800 AAs , and more than 95% similar to another protein in the dataset . This resulted in a dataset of 640 maltase homologs with sequence similarity >40% compared to Saccharomyces cerevisiae S288c MAL12 . These sequences were aligned with MAFFT and only the most reproducible residue–residue couplings ( present in at least 90% of the splits ) were retained . A two-way ANOVA using log-transformed kcat/Km ( to obtain values that are normally distributed ) as the variable and the different enzymes and sugars as factors was performed using the aovSufficient function from the HH package in R . This analysis was followed by pairwise comparisons using the Games-Howell post hoc test ( since samples had unequal variances , as demonstrated by Levene's test ) . Results can be found in Table S3 . Ancestral maltase genes were synthesized and cloned into vectors for overexpression in E . coli host cells by GENEART ( www . geneart . com ) . Sequences can be found in Table S1 and Dataset S2 . The inferred protein sequences were reverse translated in order to optimize their codon usage for E . coli . These gene sequences were synthesized including an N-terminal 6xHis tag ( ATGGGCAGCAGCCATCATCATCATCATCACAGCAGCGGCCTGGTGCCGCGCGGCAGCCAT ) and 5′UTR ( TCTAGAAATAATTTTGTTTAACTTTAAGAAGGAGA TATACC ) , cloned into in-house vectors at GENEART , and then sequenced . Subsequently , the inserts were subcloned into pET-28 ( a ) vectors ( Merck ) via XbaI/XhoI sites . All of the overexpression plasmids were transformed into E . coli strain BL21* . All E . coli strains were grown under selection in standard LB media+kanamycin ( Sigma Aldrich ) . Details on protein expression and purification can be found in Text S1 . The activities of the purified ancestral and present-day enzymes were determined by measuring glucose release from α-glucosides ( maltose , sucrose , turanose , maltotriose , maltulose , isomaltose , palatinose , and methyl-α-glucoside ) using a standard glucose oxidase/peroxidase coupled reaction . All sugars were purchased in their highest available purity . More information on the purchased sugars as well as a detailed protocol can be found in Text S1 . For each protein and substrate , the reaction velocity ( amount of glucose produced per time unit ) was determined . Subsequently , reaction velocities normalized by enzyme concentration as a function of substrate concentration were plotted and fitted using a nonlinear least squares fitting routine ( Levenberg-Marquardt algorithm ) both to Michaelis-Menten-style kinetics and Hill-style kinetics:The data fits were compared using an F statistic ( i . e . , Michaelis-Menten is a specific case of Hill kinetics with n = 1 ) , and the Michaelis-Menten model was rejected with α = 5% . From these fits , errors ( standard deviations ) were computed by jack-knifing over the individual substrate concentrations ( 12 data points in total ) . For numerical optimization , code was written in Python using NumPy . Model parameters of interest , along with their associated errors , were extracted ( i . e . , kcat and Km; see Table S2 ) . Processing ( http://processing . org ) was used to draw Figure 2 and Figure 5F by writing code . Enzyme efficiencies were plotted ( as vertical lines ) at different points on the tree , and values between were interpolated . Relative Malthusian fitness was determined by competing unlabelled WT ( KV1042 ) , mal12 ( KV1151 ) , and mal32 ( KV1153 ) strains against a reference strain ( KV3261 ) , expressing GFP from the TDH3p . Details can be found in the Supporting Information section . All molecular modeling was performed using the MOE 2010 . 10 package ( The Molecular Operating Environment , The Chemical Computing Group , Montréal , Canada ) . The recently released crystal structure of the Ima1 protein ( pdb entry: 3A4A ) , with glucose in the binding pocket , was used as a template to construct the different MALS homology models , with implementation of the Amber99 force field . Since the AAs contacting this glucose molecule are conserved within the different MALS subgroups , this glucose was used to model the different sugar substrates within the active sites , using the MOE 2010 . 10 ligX implementation . Full methods and any associated references can be found in the Supporting Information section .	Darwin's theory of evolution is one of gradual change , yet evolution sometimes takes remarkable leaps . Such evolutionary innovations are often linked to gene duplication through one of three basic scenarios: an extra copy can increase protein levels , different ancestral subfunctions can be split over the copies and evolve distinct regulation , or one of the duplicates can develop a novel function . Although there are numerous examples for all these trajectories , the underlying molecular mechanisms remain obscure , mostly because the preduplication genes and proteins no longer exist . Here , we study a family of fungal metabolic enzymes that hydrolyze disaccharides , and that all originated from the same ancestral gene through repeated duplications . By resurrecting the ancient genes and proteins using high-confidence predictions from many fungal genome sequences available , we show that the very first preduplication enzyme was promiscuous , preferring maltose-like substrates but also showing trace activity towards isomaltose-like sugars . After duplication , specific mutations near the active site of one copy optimized the minor activity at the expense of the major ancestral activity , while the other copy further specialized in maltose and lost the minor activity . Together , our results reveal how the three basic trajectories for gene duplicates cannot be separated easily , but instead intertwine into a complex evolutionary path that leads to innovation .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "biochemistry", "evolutionary", "biology", "enzymes", "biology", "evolutionary", "processes" ]	2012	Reconstruction of Ancestral Metabolic Enzymes Reveals Molecular Mechanisms Underlying Evolutionary Innovation through Gene Duplication
Point mutations in TBX1 can recapitulate many of the structural defects of 22q11 deletion syndromes ( 22q11DS ) , usually associated with a chromosomal deletion at 22q1 . 2 . 22q11DS often includes specific cardiac and pharyngeal organ anomalies , but the presence of characteristic craniofacial defects is highly variable . Even among family members with a single TBX1 point mutation but no cytological deletion , cleft palate and low-set ears may or may not be present . In theory , such differences could depend on an unidentified , second-site lesion that modifies the craniofacial consequences of TBX1 deficiency . We present evidence for such a locus in a mouse model . Null mutations of chordin have been reported to cause severe defects recapitulating 22q11DS , which we show are highly dependent on genetic background . In an inbred strain in which chordin−/− is fully penetrant , we found a closely linked , strong modifier—a mutation in a Tbx1 intron causing severe splicing defects . Without it , lack of chordin results in a low penetrance of mandibular hypoplasia but no cardiac or thoracic organ malformations . This hypomorphic Tbx1 allele per se results in defects resembling 22q11DS but with a low penetrance of hallmark craniofacial malformations , unless chordin is mutant . Thus , chordin is a modifier for the craniofacial anomalies of Tbx1 mutations , demonstrating the existence of a second-site modifier for a specific subset of the phenotypes associated with 22q11DS . In approximately 1 in 4000 human births , syndromic congenital malformations are associated with deletions in chromosomal region 22q11 . 2 . DiGeorge syndrome ( OMIM 188400 ) , velocardiofacial syndrome ( OMIM 192430 ) , and related syndromes are all associated with this deletion; these are collectively called the 22q11 deletion syndromes , 22q11DS [1] . At least 20 genes are contained within the region typically deleted , the DiGeorge Critical Region ( DCR ) . To understand the roles of particular DCR genes in the etiology of 22q11DS , the functions of many of these genes have been assessed in the mouse . Tbx1 null homozygotes show severe defects in all the structures variably affected in these syndromes [2] , [3] , [4] , while heterozygotes show aortic arch artery defects similar to some mildly affected patients [2] . Several different point mutations in TBX1 have been identified in patients with 22q11DS but without cytological deletions at 22q11 [5] , [6] . Thus , defective TBX1 function is a key factor in the pathogenesis of the 22q11DS malformations . Despite the identification of the DCR , and the key role of TBX1 in particular , the genetics of 22q11DS pathogenesis remains unclear . Significant numbers of 22q11DS patients don't possess deletions at 22q11 . 2 or knownTBX1 point mutations , while deletions in other regions of the genome have been observed [7] , [8] . These and related considerations suggest that other loci , as yet unknown , play an important role in the etiology of 22q11DS [9] . In mouse models , mutations in Crkl , which lies in 22q11 . 2 , and in Fgf8 , which is unlinked , have both been shown to modulate the developmental phenotype by enhancing the effects of a Tbx1 null mutation [10] , [11] . However , none of the genetic results to date suggest an explanation for the high variability of disease symptoms among 22q11DS patients , or why subsets of patients present with particular structural malformations but not others . Mice lacking chordin ( Chrd ) , a dedicated antagonist of Bone Morphogenetic Proteins ( BMPs ) , have been reported to show a phenotype recapitulating many structural features of 22q11DS , and very similar to that of Tbx1 null embryos [12] , [13] . Especially given that both genes are expressed in or around the pharyngeal endoderm during early organogenesis , this implied a mechanistic link between BMP antagonism and Tbx1 function . Consistent with this , reduced levels of Tbx1 message were observed in the pharyngeal region of Chrd−/− embryos; moreover , Chrd transcript injected into early Xenopus embryos could increase the endogenous Tbx1 transcriptional level [12] . Taken together , these results suggested that chordin acts in the pharyngeal region to protect Tbx1 expression from inhibition by local BMPs . However , as detailed below , our breeding of Chrd in different wildtype genetic backgrounds has indicated that in most cases , embryos can lack Chrd entirely , yet exhibit no resemblance to 22q11DS or Tbx1 mutants . Such results reveal that an additional , unknown genetic mechanism was at play in the generation of these phenotypes . Here we report our elucidation of the relationship between Chrd , Tbx1 , and the 22q11DS phenotypes . The previously reported Chrd−/− null phenotype is a completely penetrant constellation of defects – which we call hereafter the ‘full phenotype’ – that includes dysmorphic ear ( Figure 1A , B ) , absence of thymus ( athymia ) , persistent truncus arteriosus ( PTA ) , abnormal aortic arch artery structure ( Figure 1C–E ) , cleft palate ( Figure 1F–G ) . This spectrum of defects is virtually identical to those of Tbx1 null homozygotes ( see Figure S1 , available online ) . However , Chrd phenotypes are highly dependent on genetic background . We observed that Chrd−/− embryos in an inbred 129S6/SvEv ( 129S6 ) genetic background displayed 100% penetrance of this full phenotype . In contrast , mutant embryos in C57B6/J ( B6 ) and random outbred genetic backgrounds were viable and lacked all but a low penetrance of mild craniofacial phenotypes , primarily involving the mandible . This suggested the existence of a strain-specific modifier . To assess this , we interbred 129S6- and B6 . 129-Chrd+/− animals , generating F1 and F2 hybrid Chrd−/− embryos . We detected no phenotype in F1 Chrd−/− embryos , while F2 Chrd−/− embryos showed a 44 . 3% penetrance of phenotypes ( 43/97; Figure 1H ) , suggesting either a 129S6-derived recessive modifier or B6-derived dominant suppressor . Because random outbred Chrd homozygotes lacked DGS-like phenotypes , the former seemed more likely . The F2 hybrid Chrd−/− embryos showed a range of phenotypes , suggesting that the effect of a single modifier is incompletely penetrant , or that more than one modifier is present ( Figure 1I and Figure S2 ) . To begin locating the putative modifier ( s ) , we conducted a genome-wide scan for segregation of a particular 129S6-derived chromosomal region with the Chrd locus to result in the full phenotype . Although there was evidence of more than one potential modifier , the data suggested that such a region resides on the same chromosome as Chrd ( data not shown ) . In contrast to the situation in humans , Chrd is near the DCR in mice ( Figure 2A ) , reflecting a relative chromosomal translocation between Chrd and the DCR in these species . Since lesions within the DCR are implicated in similar phenotypes in mouse models [10] , [14] , we investigated whether a linked modifier might account for the strain-dependence of the Chrd phenotypes . Using several SSLP and SNP markers on our F2 recombinant genomic DNAs , we located a region of 129S6-derived chromosome segregating with the full Chrd phenotype; this region lay close to the marker rs4165069 , just proximal to the DCR ( Figure 2A ) . These data imply a physically-linked , recessive lesion specific to the 129S6-derived DCR area that is associated with the full Chrd−/− phenotype . Because the null phenotype of Chrd in the 129S6 background is essentially identical to that of Tbx1−/− , the simplest explanation is that this region contains a recessive , loss-of-function mutation affecting Tbx1 activity . If so , the Chrd mutant chromosome and linked modifier ( s ) should be unable to complement a Tbx1 null mutation . Accordingly , we crossed 129S6-Chrd+/− with B6-Tbx1tm1Bld/+ to generate double heterozygotes , with one Chrd allele and oneTbx1 allele being nulls created by gene targeting , in trans on the two cognate chromosomes . Among Chrd+/− , Tbx1tm1Bld/+ embryos , 5/12 showed the ‘full phenotype’ and 6/12 showed a partial DGS-like phenotype . This indicates the Tbx1 null allele interacts strongly with the Chrd mutant chromosome ( Figure 2B ) . In contrast , when B6 . 129-Chrd+/− was crossed with B6-Tbx1tm1Bld/+ mice , no significant phenotype was observed in double heterozygotes ( 0/15 , two cases of asymmetric thymus were detected ) . Thus the apparent interaction of theTbx1 null locus with Chrd depends on a linked sensitizer in the 129S6 background . The simplest explanation is a cryptic mutation in Tbx1 itself . Sequencing of Tbx1 revealed a point mutation ( G to T ) specific to the 129S6-Chrd+/− strain ( Figure 2C ) . We did not find any Tbx1 mutation in B6 . 129-Chrd+/− animals or in random outbred Chrd mice . Furthermore , we did not detect the mutation in 129S6 wild-type , other 129-substrains , or in the R1 ES cell line used for Chrd targeting ( Figure S3 , [12] ) . These data suggest that this novel mutation ( Tbx1G>T ) occurred spontaneously during initial Chrd+/− ES cell culture or in the establishment of the 129S6-Chrd colony . Utilizing the Tbx1G>T mutation as a SNP marker , we addressed how strongly it segregates with the full phenotype . The high LOD score suggests that the Tbx1 mutation is indeed a strong modifier ( Figure 2D ) . Since the mutation is in the boundary region of the second intron and is predicted to disrupt the splicing factor recognition sequence ( ESEfinder 2 . 0; Figure 2E ) [15] , [16] , we examined splicing ofTbx1 in 129S6-Chrd−/− embryos . We detected exon skipping and intron retention of Tbx1 mRNA from such embryos , leaving little of the correct form ( Figure 2F ) . This suggests that the Tbx1G>T mutation is at least partly responsible for the 22q11DS constellation of phenotypes observed in Chrd mutants in certain genetic backgrounds . Having determined that a Tbx1 lesion is a linked modifier of Chrd , we wondered how we observed 44% penetrance of the 22q11DS phenotypes in F2 Chrd−/− animals ( Figure 1H ) , instead of the expected value of 25% . Breeding records and retrospective Tbx1 genotyping indicated that this resulted from using a subset of F1 hybrid males as studs that by chance had most often inherited their mutant Chrd allele from the 129S6 background , and thus were also carriers of the tightly linked Tbx1G>T allele ( the expected phenotypic penetrance would be 50% if this were the case for all studs used ) . This caused a non-random bias in transmitting the Tbx1G>T allele to F2 hybrid animals . Our F1 and F2 genotyping also revealed that Chrd−/− , Tbx1+/G>T mice ( in which one Chrd null allele was inherited from the129S6 strain and one from B6 ) had very few 22q11DS phenotypes ( 1/53 for F2 animals; Figure 2D ) . To determine the phenotypes resulting from the Chrd null or the Tbx1G>T allele individually , we bred 129S6-Chrd+/− , Tbx1+/G>T with 129S6 wild-type mice to generate recombinant animals ( Figure 3A ) . We genotyped 540 offspring and identified one recombinant carrying only the Chrd null allele ( Chrd+/− , Tbx1+/+ ) , and three carrying only the Tbx1G>T allele ( Chrd+/+ , Tbx1+/G>T ) . We assessed the phenotypic consequences of the isolated mutations . Animals heterozygous for the Tbx1+/G>T mutation are healthy and fully viable , displaying no visible phenotype . We crossed Tbx1+/G>T animals to themselves and also to Tbx1+/null . Analysis of the three classes of Tbx1 mutant embryos for the DGS phenotype demonstrates genetic rescue: Tbx1G>T homozygotes show rescue of the major craniofacial defects of the Tbx1 null phenotype ( n = 18 ) , while the compound heterozygote ( Tbx1G>T/null ) is intermediate ( n = 13 ) , relative to the fully penentrant , strong DGS-like phenotype of the null homozygote . Thus Tbx1G>T is a hypomorphic allele ( Figure 3B ) . That craniofacial development is much less affected in the hypomorphic homozygote relative to the null demonstrates that craniofacial structures are more sensitive to Tbx1 dose than cardiovascular structures , consistent with a previous report [17] . We also prepared homozygotes for the Chrd allele alone , recombined away from Tbx1G>T . These Chrd null homozygotes did not exhibit phenotypes similar to 22q11DS , but showed a low penetrance of variable mandibular truncations , comparable in extent and frequency to what we have observed for B6 . 129-Chrd−/− embryos ( Figure 3C–H and Table S1 ) . A previous study revealed redundant but essential roles of chordin and noggin , another BMP antagonist , in mandibular outgrowth , during which these BMP antagonists promote cell survival in the developing 1st pharyngeal arch [18] . That study used the outbred Chrd strain , which does not carry the Tbx1 hypomorphic allele . Our study confirms a role for chordin in promoting mandibular development . Interestingly , we also observed a severe mandible truncation in one of the double mutant embryos ( Chrd−/− , Tbx1G>T/G>T , Table S1 ) . This result indicates that the craniofacial phenotype caused by loss of Chrd is not suppressed by the Tbx1 mutation , suggesting that Tbx1 is not in turn a modifier of Chrd . Our data indicating that Chrd deficiency is a modifier of Tbx1 action raises the issue of the molecular basis of this modifier effect . Previous work has suggested that Chrd activity promotes Tbx1 expression in the pharyngeal region of the mouse embryo . Tbx1 expression is reduced in this area of Chrd null embryos [12] . However , the pharyngeal tissues themselves are deficient from early stages in the Chrd mutant ( if the cryptic Tbx1 allele is also present ) . Nonetheless , ectopic Chrd activity in early Xenopus embryos can cause increased Tbx1 transcript levels [12] . In a separate study on the role of BMP signaling in heart development , we observe that BMP-soaked beads in the pharyngeal lumen reduce the expression of Tbx1 ( Choi et al . , manuscript in preparation ) . Such data support a model that Chrd expression in the dorsal pharyngeal endoderm promotes Tbx1 expression in the pharyngeal region , by antagonizing a repressive effect of BMP on Tbx1 transcription . To further test whether Chrd mutations alter Tbx1 expression in vivo , we measured the amount of Tbx1 transcripts by in situ and real-time quantitative PCR in the 129S6 strain ( in the presence of the Tbx1G>T mutation ) and in the B6 strain ( without the Tbx1 mutation ) . In both cases , there was reduced expression ( Figure 4 ) – about 30% less in B6 , despite pharyngeal tissues being fully formed and the Tbx1 locus presumably wildtype . In sum , our data indicate that Chrd has a modest but significant role in promoting Tbx1 expression . Such a decrease is insufficient to cause 22q11DS-like defects on its own , but may compound the consequences of a Tbx1 hypomorphic allele . Our search for the modifier of Chrd identified an unexpected , linked mutation in the Tbx1 locus in the 129S6-Chrd+/− strain , which causes mis-splicing . Comparison of phenotypes between the separated , isolated Tbx1G>T/G>T and Chrd−/− mutations and those associated with homozygosity for the original chromososome , i . e . Chrd−/− , Tbx1G>T/G>T , allow us to make important conclusions regarding the basis of the phenotypes we observed . First , we now know that the 22q11DS-like defects seen in the original 129S6-Chrd−/− mutant embryos were actually caused primarily by the mutation in Tbx1 . On its own , in both the 129S6 and B6 backgrounds , Chrd−/− causes a low penetrance of moderate to severe mandibular truncations . Secondly , embryos mutant for the hypomorphicTbx1G>T/G>T mis-splicing allele show several cardiac and pharyngeal defects nearly as severe as those of the Tbx1 null mutants , but without the major craniofacial phenotypes of the null ( reduced pinna and cleft palate ) . Finally , our results show that whereas homozygosity for the hypomorphic Tbx1G>T allele rarely results in major craniofacial defects alone , it does so when Chrd is absent . In contrast , the cardiovascular defects are unaffected . Thus , Chrd is a modifier specifically for the craniofacial phenotypes of Tbx1 lesions . The craniofacial defects associated with homozygosity for the cryptic double mutant reflect a synergistic relationship between the Chrd and Tbx1 lesions . Although there was a low penetrance of striking mandibular outgrowth defects in Chrd−/− mutants , most homozygotes ( 73% ) looked normal . The Tbx1G>T/G>T animals in turn showed a limited penetrance of craniofacial defects that are reminiscent of 22q11DS . Penetrance of low-set , reduced pinna ( outer ear ) was approximately 20% , while cleft palate was about 50% . However , when the two mutations are together , all mutant embryos develop fully penetrant , highly consistent craniofacial phenotypes; these defects appear to be identical to those of Tbx1−/− null animals , with cleft palate , reduced , low-set ears , athymia , persistant truncus arteriosis , etc ( Figure 5 ) . These phenotypes occur against a background of the low-penetrance of severe mandibular truncation defects caused by absence of Chrd per se , much more pronounced than the subtle mandibular hypoplasia reported for the Tbx1 null [3] . Therefore , the absence of Chrd leads to a synergistic ( as opposed to additive ) worsening of the defects caused by the hypomorphic Tbx1 allele , to generate the complete constellation of craniofacial defects seen in the Tbx1−/− null and reminiscent of 22q11DS . The Tbx1G>T mutation we identified behaves according to the genetic definition of a hypomorphic allele: Tbx1G>T/G>T homozygotes are less severely effected than homozygotes for a null allele , while the compound heterozygote is intermediate in severity . Molecular evidence to account for such reduced activity was apparent in the transcripts produced by the mutant allele . The mutation is predicted to disrupt normal splicing , and in fact we observed via PCR analysis both exon skipping and intron retention . In the latter case , reading frame-shifts result in nonsense codons . The transcript resulting from exon skipping would encode a truncated product lacking 25% of the T-box domain , essential for the proper function of the protein – if any protein is stably produced . In addition , these abnormal transcripts account for most of the spliceforms produced by the allele; we amplified little transcript corresponding to the correctly spliced wildtype version from Tbx1G>T/G>T homozygotes . The consequences of the cryptic Tbx1G>T/G>T allele linked to Chrd in the initial 129S6 background account for much of the phenotype of Chrd−/− mutants as previously reported [12] . Nevertheless , our genomic scan to assess the possibility of modifiers suggested more than one modifier was present . Thus there may be additional modifier ( s ) in the 129S6 background that influence the Chrd−/− , Tbx1G>T/G>T phenotype . Such lesions could be outside the Chrd , Tbx1 region , or even within it – the coding regions of these genes are approximately 2 million bases apart ( Mouse Genome Informatics ) . To prove that only the Chrd targeted mutation and the cryptic Tbx1 splice site mutation ( but no other 129S6 allele ) are sufficient for the full phenotype would require reconstituting only these mutations in a different inbred strain , a daunting task . We note that the behaviors of Chrd and Tbx1 alleles independently do not appear to be 129S6-specific . For Chrd−/− embryos , the penetrance and expressivity of the mandible phenotype in the 129S6 and B6 strains are quite comparable ( Table S1 ) . In the case of Tbx1 , alleles have been made and studied in multiple strains and produced very consistent phenotypes ( our observations , [2] , [3] ) . Altogether , it seems likely that if there are additional specific modifiers at play , they have a very minor role . Previous work and our unpublished observations suggest that Chrd functions to promote Tbx1 expression in the pharyngeal region . When we assayed Tbx1 expression in Chrd mutants with or without the cryptic Tbx1 hypomorphic mutation , consistent results were observed . In both 129S6- and B6-Chrd−/− embryos , there was mild but significant reduction of Tbx1 expression when compared to the wildtype embryos ( Figure 4 ) . Nevertheless , in all three backgrounds tested , Chrd null animals free of Tbx1G>T are viable and show no pharyngeal defects . Thus the decrease of Tbx1 expression in pure Chrd mutants is insufficient to cause a phenotype; however , loss of this activity could be a contributing factor in the functional synergy between the Chrd null and Tbx1G>T mutations in causing a more severe DGS-like phenotype than the hypomorphic Tbx1G>T mutation alone . We note that Chrd−/− , Tbx1+/G>T mice very rarely show Tbx1 mutant phenotypes ( 1/53 in the F2 hybrids ) . Thus it is possible that the reduction in functional Tbx1 protein caused by a single allele of Tbx1+/G>T , compounded by the moderately decreased Tbx1 expression caused by loss of Chrd , is in rare instances sufficient to generate phenotypes similar to 22q11DS . Tbx1 also shows genetic interactions with other mutations both within and outside the DCR [10] , [11] . However , no second-site mutation has been found previously that can account for why DCR or Tbx1 mutations are sometimes associated with a particular defect but sometimes not . The result reported here show that Chrd is a modifier for the craniofacial anomalies ofTbx1 mutations , demonstrating the existence of a second-site modifier for a specific subset of the phenotypes associated with 22q11DS . The Chrd null allele ( Chrdtm1Emdr ) was generated previously using R1 ES cells [12] . To generate outbred Chrd stock [18] , germline chimeras were mated to random outbred ICR females , with backcrossing of F1 founders to ICR . To generate 129S6 inbred stock , germline chimeras were mated to 129S6/SvEvTac ( Taconic ) wild type mice , and Chrd heterozygotes backcrossed for >10 generations . To generate B6 . 129S6-Chrd , a 129S6-Chrd+/− male was crossed to C57BL/6J wild-type ( Jackson Laboratory ) females . Resulting heterozygous males were subsequently backcrossed with C57BL/6J wild-type females for 10 generations . Chrd and Tbx1tm1Bld were genotyped via PCR as previously described [12] . To type the Tbx1G>T mutation , regions encompassing the mutation were PCR-amplified ( forward: 5′- AGCAGGGCAGGAACAGTCT-3′ , reverse: 5′- CTGCCTGGCCAGAGAAGTTA-3′ ) , cut with DpnII ( New England Biolab ) , and resolved by agarose gel electrophoresis . For partial genome scanning to find rough general locations of major Chrd modifiers , we searched for microsatellite marker differences between 129S6 and C57BL6 strains ( http://www . informatics . jax . org/searches/polymorphism_form . shtml; www . cidr . jhmi . edu/mouse/mmset . html ) . We used 154 markers , with an average interval of approximately 22 . 2Mb . We designed appropriate PCR primers and amplified corresponding regions from F2 hybrid animals . To assess Tbx1 splicing , total RNA was prepared from E9 . 5 embryos and subjected to RT-PCR . Pharyngeal tissues were homogenized and treated with TRIZOL Reagent ( Invitrogen ) . Remaining tissues were used for genotyping . After further purification , the RNA pellet was dried and resuspended in DEPC-water for subsequent reverse transcription ( RT ) with random hexamer and mouse mammary tumour virus ( MMTV ) RT ( Invitrogen ) . Resulting cDNA was used as a template for PCR . The following primers were used for testing Tbx1 splicing: TbxRT1F 5′-TTTGTGCCCGTAGATGACAA-3′ ( forward ) ; TbxRT2R 5′-TCATCCAGCAGGTTATTGGTC-3′ ( reverse ) ; TbxRT3R 5′-AATCGGGGCTGATATCTGTG-3′ ( reverse ) ; TbxRT4F 5′-TGTGGGACGAGTTCAATCAG-3′ ( forward ) . Skeletal tissues were prepared and visualized as previously described [18] .	A range of structural malformations is associated with 22q11 deletion syndrome ( 22q11DS ) , which is usually associated with microdeletions at chromosome 22q11 . 2 . Variable defects in cardiovascular , pharyngeal , and craniofacial structures occur , but the basis for such variability is unknown . Mutations in TBX1 , a gene within the region typically deleted , can recapitulate the structural anomalies of 22q11 . 2 deletions . However , even among family members with a single TBX1 point mutation , craniofacial defects are variable . In theory , such differences could depend on an unknown , second-site lesion that modifies the craniofacial consequences of TBX1 deficiency . We identify such a locus in mouse . In certain strains , lack of the chordin gene results in a phenotype resembling severe 22q11DS , also seen in mice lacking Tbx1 . We find that the chordin phenotype depends on a closely linked , strong modifier—a cryptic , partial-function mutation in Tbx1 . Without it , lack of chordin sometimes results in mandibular truncations but no cardiac or thoracic organ malformations . This novel Tbx1 allele per se results in defects resembling 22q11DS , but with a low frequency of hallmark craniofacial malformations , unless chordin is mutant . Thus chordin is a modifier for the craniofacial anomalies of Tbx1 mutations , demonstrating the existence of a second-site modifier for a specific subset of phenotypes associated with 22q11DS .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genetics", "and", "genomics/disease", "models", "developmental", "biology/organogenesis", "genetics", "and", "genomics/gene", "function", "developmental", "biology/developmental", "molecular", "mechanisms" ]	2009	Chordin Is a Modifier of Tbx1 for the Craniofacial Malformations of 22q11 Deletion Syndrome Phenotypes in Mouse
RNA-DEPENDENT RNA POLYMERASE 6 ( RDR6 ) is a key RNA silencing factor initially characterized in transgene silencing and virus resistance . This enzyme also contributes to the biosynthesis of endogenous short interfering RNAs ( siRNAs ) from non-coding RNAs , transposable elements and protein-coding transcripts . One class of protein-coding transcripts that have recently emerged as major sources of RDR6-dependent siRNAs are nucleotide-binding leucine-rich repeat ( NB-LRR ) proteins , a family of immune-receptors that perceive specific pathogen effector proteins and mount Effector-Triggered Immunity ( ETI ) . Nevertheless , the dynamic post-transcriptional control of NB-LRR transcripts during the plant immune response and the functional relevance of NB-LRRs in signaling events triggered by Pathogen-Associated Molecular Patterns ( PAMPs ) remain elusive . Here , we show that PTI is constitutive and sensitized in the Arabidopsis rdr6 loss-of-function mutant , implicating RDR6 as a novel negative regulator of PTI . Accordingly , rdr6 mutant exhibits enhanced basal resistance towards a virulent Pseudomonas syringae strain . We further provide evidence that dozens of CC-NB-LRRs ( CNLs ) , including the functionally characterized RPS5 gene , are post-transcriptionally controlled by RDR6 both constitutively and during PTI . These CNL transcripts are also regulated by the Arabidopsis microRNA miR472 and knock-down of this miRNA recapitulates the PTI and basal resistance phenotypes observed in the rdr6 mutant background . Furthermore , both miR472 and rdr6 mutants were more resistant to Pto DC3000 expressing AvrPphB , a bacterial effector recognized by the disease resistance protein RPS5 , whereas transgenic plants overexpressing miR472 were more susceptible to this bacterial strain . Finally , we show that the enhanced basal and RPS5-mediated resistance phenotypes observed in the rdr6 mutant are dependent on the proper chaperoning of NB-LRR proteins , and might therefore be due to the enhanced accumulation of CNL proteins whose cognate mRNAs are no longer controlled by RDR6-dependent siRNAs . Altogether , this study supports a model whereby the miR472- and RDR6-mediated silencing pathway represents a key regulatory checkpoint modulating both PTI and ETI responses through the post-transcriptional control of disease resistance genes . To defend themselves against pathogens , plants have evolved potent inducible immune responses . The first line of active defense relies on the recognition of common features of microbial pathogens , such as flagellin ( the major protein of bacterial flagellum ) , lipopolysaccharides , glycoproteins and chitin [1] . These microbial determinants are referred to as Pathogen- or Microbe- Associated Molecular Patterns ( PAMPs/MAMPs ) and are sensed by host-encoded Pattern-Recognition Receptors ( PRRs ) or surface receptors , which encode transmembrane receptor-like kinases . Upon PAMP detection , PRRs trigger a series of immune responses including , for instance , MAPK ( mitogen-activated protein kinase ) activation , reactive oxygen species ( ROS ) production , differential expression of genes , callose ( β-1->3 glucose polymer ) deposition and stomatal closure , which ultimately leads to basal immunity or PAMP-Triggered Immunity ( PTI ) [2]–[5] . To enable disease , pathogens produce a large array of divergent virulent determinants known as pathogen effectors that suppress different steps of PTI , resulting in disease susceptibility [6] , [7] . As a counter-counter defense strategy , plants have evolved a repertoire of immune receptors , called disease resistance ( R ) proteins that can sense effector proteins and establish effector-triggered-immunity ( ETI ) [1] . The largest class of R proteins is composed of intracellular receptors that share structural homologies with mammalian innate immune receptors , such as NUCLEOTIDE-BINDING OLIGOMERIZATION DOMAIN-CONTAINING PROTEIN 1 ( NOD1 ) and NOD2 , which perceive bacterial PAMPs [8] . Plant NOD-like receptors ( NLRs ) are composed of nucleotide-binding ( NB ) and leucine-rich repeat ( LRR ) domains . They additionally contain an N-terminal domain that is composed of either a Toll/interleukin1 receptor ( TIR ) or a coiled-coil ( CC ) module , and are thus referred to as TNLs or CNLs , respectively [9] . These R proteins can directly sense pathogen effectors [1] , however , in most cases they recognize indirectly these virulent determinants by detecting their effects on plant target proteins called ‘guardees’ [10] . Upon pathogen effector recognition , R proteins trigger a series of immune responses that significantly overlap with PTI responses , albeit with a stronger amplitude , and often result in a form of programmed cell death known as the hypersensitive response ( HR ) [1] . Importantly , constitutive expression or activation of R proteins often leads to constitutive cell death as well as severe developmental defects in the absence of pathogen [11]–[16] , indicating that R genes and their products must be under tight negative control in unchallenged conditions . Consistent with this idea , transcriptional regulation , RNA processing , protein modifications , protein stability , and nucleocytoplasmic trafficking were shown to play a critical role in controlling R-mediated autoimmune responses [17] . More recently , RNA silencing has also emerged as a key regulatory mechanism that negatively regulates R gene expression [18]–[24] . RNA silencing is an ancestral gene regulatory mechanism that controls gene expression at the transcriptional ( TGS , Transcriptional Gene Silencing ) and post-transcriptional ( PTGS , Post-transcriptional Gene Silencing ) levels . The core mechanism of RNA silencing starts with the production of double stranded RNAs ( dsRNAs ) that are processed by RNase-III enzymes DICERs into 20–24 nt small RNA duplexes . One selected strand is subsequently incorporated into an RNA-induced silencing complex ( RISC ) containing an argonaute ( AGO ) protein , and guides these complexes onto sequence complementary RNA/DNA targets . The plant model Arabidopsis thaliana encodes 4 DICER-like proteins and 10 AGOs . DCL1 processes miRNA precursors into mature microRNAs that are mostly incorporated into the AGO1-RISC that guides mRNA degradation and/or translation inhibition of sequence complementary mRNA targets . DCL2 , DCL3 and DCL4 are involved in the biogenesis of short interfering RNAs ( siRNAs ) from extensive dsRNAs produced from read through , convergent or overlapping transcription , endogenous hairpins as well as some miRNA precursors [25] , [26] . As an example , overlapping sense and antisense transcripts that are produced at a functionally relevant disease resistance gene cluster , were found to be processed into siRNAs , leading to the down-regulation of several disease resistance gene transcripts within this cluster [18] . In addition , a large proportion of dsRNAs are produced by RNA-dependent RNA polymerases ( RDRs ) that convert single stranded RNAs into dsRNAs . RDR6 , which is one out of six Arabidopsis RDRs , produces dsRNAs from viral and transgene transcripts as well as some endogenous transcripts [27] . These dsRNAs are processed in part by DCL4 into 21 nt siRNAs that direct PTGS of endogenous mRNA targets or exogenous RNAs derived from sense-transgenes or viral RNAs [28]–[31] . In addition to the biogenesis of primary siRNAs , plants have evolved the production of secondary siRNAs as a feed-forward amplification of silencing signals . These siRNAs are produced by the combined action of primary siRNA/miRNA-directed transcript cleavage and the activity of RDRs that use the target transcripts as template to generate dsRNAs [32] . In plants , the best-characterized endogenous secondary siRNAs are termed trans-acting siRNAs ( tasiRNAs ) [33] , [34] . The biogenesis of these small RNA molecules is initiated by 22 nt long miRNAs that direct AGO1-mediated cleavage of a non-coding TAS primary transcript [35] , [36] . One of the cleavage products is then converted by RDR6 into dsRNAs , which are processed by DCL4 into 21-nt phased siRNA duplexes . These secondary siRNAs guide an AGO protein to silence sequence complementary mRNA targets in trans . Importantly , this phenomenon is not restricted to non-coding transcripts but also targets protein-coding transcripts and both TNLs and CNLs have emerged as major targets of this silencing pathway [20] , [21] , [22] . For example , two 22 nt long miRNAs that initiate the production of RDR6-dependent secondary siRNAs , were found to directly control the tobacco disease resistance gene N , which recognizes the C-terminal helicase domain of the Tobacco Mosaic Virus ( TMV ) replicase protein [22] . These miRNAs play a functional role in N-regulation because their overexpression was shown to compromise N-mediated resistance to TMV [22] . Another recent study conducted in Solanum lycopersicum showed that miR482 , a 22 nt long conserved miRNA that targets dozen of CNLs , was down-regulated in response to unrelated viruses as well as to a bacterium that encode RNA silencing suppressors [21] . Interestingly , this phenomenon was associated with the derepression of some CNLs that are targeted by miR482 , suggesting that pathogen-triggered suppression of RNA silencing likely derepresses a whole repertoire of immune receptors during infection that might contribute to plant immunity [21] . Recent findings have thus revealed a critical role of miRNA-directed phased siRNA production in controlling the expression of R gene transcripts in the context of pathogen infection . Nevertheless , the interplay between the dynamic regulation of the RNA silencing machinery involved in miRNA-directed secondary siRNA production and the post-transcriptional regulation of R gene transcripts that are targeted by these small RNA species remains unknown . In addition , whereas some intracellular immune-receptors have recently been characterized in basal defense as well as plant defense against a disarmed bacterium very little is known on the functional relevance of plant NLRs in PTI [21] . The present study addresses some of these important issues by studying the regulation of RDR6 during antibacterial defense and the role of this silencing factor in the control of CNLs that are targeted by the Arabidopsis miR472 , a miRNA related to miR482 . Although ARGONAUTE 1 ( AGO1 ) and DICER-LIKE 1 ( DCL1 ) were previously shown to contribute to PTI [37] , [38] , their regulation during the plant innate immune response has not been determined . To get a first insight into the regulation of components of PTGS during plant defense , we examined the expression levels of well-characterized PTGS factors in multiple conditions known to trigger PTI responses ( Genevestigator database: https://www . genevestigator . com ) . Results from this analysis revealed that RDR6 , AGO1 and SUPPRESSOR OF GENE SILENCING 3 ( SGS3 ) mRNAs [39] were all down-regulated , with RDR6 showing the highest difference ( consistently more than 2-fold in the various conditions analyzed ) ( Figures S1A , S1B ) . Accordingly , Reverse-Transcriptase Quantitative Polymerase chain reaction ( RT-qPCR ) analyses revealed a significant decrease in RDR6 mRNA levels in Arabidopsis leaves and seedlings treated with the flagellin-derived peptide flg22 ( Figure 1 ) , with a decrease in RDR6 transcripts starting at 10 min in Arabidopsis elicited seedlings ( Figure S1C ) . A similar effect was observed with the type-three secretion ( TTS ) defective mutant Pto DC3000 hrcC− , which can elicit , but not suppress , PTI responses due to its inability to inject effector proteins within host cells ( figure S1C ) . The PAMP-triggered dynamic regulation of RDR6 transcripts therefore suggested a potential role for RDR6 in orchestrating PTI responses . To test this idea , we first monitored the effect of the rdr6-15 loss-of-function mutation on the production of reactive oxygen species ( ROS ) , one of the earliest cellular responses following PAMP perception , which is known to orchestrate the establishment of different defensive barriers against biotrophic pathogens [40] . We observed a more pronounced flg22-triggered oxidative burst in the rdr6 mutant as compared to WT-elicited plants ( Figure 2A ) . However , given that the kinetics of flg22-triggered ROS production precedes the down-regulation of RDR6 transcripts in wild-type treated plants ( Figure 1 ) , these results suggest that the repression of RDR6 mRNAs is unlikely causative for this early PTI response . We also monitored the expression of PTI marker genes and found a primed induction of Flg22 RECEPTOR KINASE 1 ( FRK1 ) in the rdr6-elicited mutant ( Figure 2B , [41] ) . Of note , induction of FRK1 as well as the two other early PTI marker genes WRKY22 and WRKY29 was also moderately sensitized upon syringe infiltration of water in rdr6- versus WT-leaves ( Figure 2B ) , suggesting that RDR6 may additionally repress a wounding response caused by mechanical stress . We further monitored the flg22-triggered formation of cell wall depositions of callose , a late PTI response that plays a critical role in the establishment of basal immunity [42] , [43] . An increase in flg22-induced callose depositions was observed in the rdr6-15 mutant as compared to WT plants , reinforcing a role for RDR6 in repressing this late PTI response ( Figure 2C ) . It is noteworthy that a higher number of callose deposits were also observed in mock-treated rdr6-15 mutant versus WT plants , but not in untreated rdr6-15 mutant leaves ( data not shown ) , suggesting that RDR6 may additionally prevent callose deposition upon wounding caused by syringe infiltration . Natural surface openings , such as stomata , are important entry sites for bacterial plant pathogens such as Pto DC3000 and previous studies have shown that stomata closure plays an active role in limiting bacterial invasion as part of PTI responses [4] . Furthermore , fls2 mutants were found to be more susceptible to Pto DC3000 upon spray inoculation , although no discernible phenotype was observed using classical syringe infiltration assay , which bypassed basal immunity present at the leaf surface [44] . Given that the rdr6-15 mutant was sensitized for multiple flg22-triggered PTI responses , we reasoned that such silencing-deficient mutant might display enhanced resistance to Pto DC3000 upon spray inoculation . Consistent with this hypothesis , we found ∼10 times lower bacterial titer on rdr6-15 mutant as compared to WT plants spray inoculated with Pto DC3000 ( Figure 2D ) . Collectively , these data provide evidence that the RNA silencing factor RDR6 acts as a negative regulator of basal immunity . These results also suggest that some positive regulators of plant defense are likely to be directly controlled by RDR6-dependent siRNAs . Besides generating siRNAs directed against viral- , transgene- and transposon-derived RNAs , RDR6 is known to produce secondary 21 nt siRNAs from several endogenous loci including TAS genes [25] , [26] . We thus searched for candidate defense gene transcripts that would be directly controlled by RDR6-dependent siRNAs . We used publicly available small RNA libraries derived from WT and rdr6 mutant leaves and selected candidate genes with a significant reduced amount of 21 nt siRNAs in the rdr6 as compared to the WT background . Using such criterion , we identified 75 loci that were likely targeted by RDR6-dependent siRNAs . Among those , 27 were previously annotated as TAS genes or tasiRNA targets . The remaining 48 protein-coding genes were enriched in GO categories ‘response to stress’ ( http://bar . utoronto . ca/welcome . htm ) , ( Figure S2 ) , and include well-characterized RDR6-dependent targets such as AGO1 , which is targeted by miR168-directed secondary siRNAs [45] . In addition , thirteen other candidate genes were annotated as miRNA targets and include multiple disease resistance gene transcripts that were previously identified as targets of miR472 ( Figures S3 , S4 ) , a 22 nt long miRNA that is at least in part loaded into AGO1-RISC [35] , [46] . These R genes are phylogenetically related to the functionally relevant disease resistance gene RPS5 , which was previously characterized in ETI [47] . Given that At1g51480 and At5g43730 were among the CNL transcripts with the most matching secondary siRNAs ( Figure S4 ) , we decided to further characterize their regulation by RDR6 in both naïve and flg22-challenged conditions . These candidate genes are referred to here as Resistance Silenced Gene 1 ( RSG1 , At1g51480 ) and Resistance Silenced Gene 2 ( RSG2 , At5g43730 ) . We also included RPS5 in this analysis , which was previously validated as miR472 target in Parallel Analysis of RNA Ends ( PARE ) datasets . We found a mild enhanced accumulation of these three candidate transcripts in unchallenged rdr6 mutant as compared to non-treated WT seedlings ( Figure 3A ) , suggesting that these mRNAs are weakly controlled by RDR6-dependent siRNAs in naïve conditions , presumably due to their low basal transcriptional level in unchallenged conditions as previously observed for several disease resistance genes [17] , [48] . We next monitored the levels of these mRNAs upon flg22 treatment in both WT and rdr6 mutant backgrounds . Whereas a mild increased induction of these transcripts was found in WT-elicited background , as observed in publicly available datasets ( Figure S5 ) , a 10- to 20-fold enhanced accumulation of these transcripts was obtained in the rdr6-elicited mutant seedlings , indicating cell priming in the absence of RDR6-dependent siRNAs ( Figure 3B ) . These results therefore indicate that RDR6-dependent secondary siRNAs negatively regulate these CNL transcripts and that this post-transcriptional regulatory control is particularly relevant during PTI , when these disease resistance genes are presumably transcriptionally activated . Given that miR472 was shown to target the above CNL mRNAs and to initiate the production of RDR6-dependent secondary siRNAs at these loci [20] , [21] , [22] , we next characterized the role of this particular miRNA in the regulation of these candidate CNL transcripts as well as other orphan targets . For this purpose , we first transformed Arabidopsis with a construct containing AtmiR472 driven by the strong Cauliflower Mosaic Virus ( CaMV ) 35S promoter and selected a reference line ( referred to as miR472OE line ) exhibiting high miR472 accumulation compared to WT ( Figure 4A ) . This line displayed a 25% and 30% reduction in the accumulation of RPS5 and RSG1 transcripts , respectively ( Figure S6 ) , providing further evidence that miR472 targets these CNL mRNAs in unchallenged conditions . Furthermore , genome-wide small RNA deep sequencing analyses revealed a drastic enhanced accumulation of secondary siRNAs at the 3′ ends of miR472 target sites for RPS5 , RSG1 and RSG2 mRNAs as well as for 16 other CNL transcripts in the miR472OE line as compared to wild type seedlings ( Figures 4B , 4C , S7 ) . It is noteworthy that no siRNA were identified upstream the miR472 target site , which is in agreement with the rapid degradation of this region after miRNA-guided cleavage [49] . Furthermore , normal levels of tasiRNAs were identified in miR472OE as compared to WT seedlings ( Figure S8 ) , proving evidence that the enhanced accumulation of CNL-derived secondary siRNAs are not due to a general activation of the RDR6-dependent pathway in this transgenic line . Collectively , these results strongly reinforce a role for miR472 in initiating the biosynthesis of RDR6-dependent secondary siRNAs at our candidate CNL transcripts and revealed additional CNLs that are directly targeted by this regulatory process including the other functionally relevant disease resistance gene SUMM2 ( Figure S9 ) [50] . We next analyzed the mRNA accumulation of two candidate CNLs in the miR472OE line challenged with flg22 . Flg22-triggered induction of RPS5 and RSG1 mRNAs was significantly impaired in miR472OE line as compared to WT-elicited control ( Figure 5A ) , supporting a role for miR472 in regulating the accumulation of these targets during flg22 elicitation . We also examined different PTI features in the miR472OE reference line by monitoring ROS production and callose deposition upon flg22 treatment . While this transgenic line displayed a normal flg22-triggered ROS production as compared to WT-elicited control ( Figure 5B ) , we found a reduced number of flg22-induced callose deposits relative to WT-treated plants ( Figure 5C ) , indicating that the miR472OE reference line is altered in the latter PTI response . It is noteworthy that similar PTI phenotypes were observed in another independent transgenic line overexpressing miR472 ( Figure S10 ) . To get further insights into the role of miR472 in the regulation of CNL transcripts and PTI responses , we further characterized a transgenic line carrying a T-DNA insertion within the promoter of the AtmiR472 locus ( Salk_087945 , referred to as miR472m ) . This line displayed a drastic decrease in the accumulation of the mature form of miR472 relative to the levels of this miRNA in WT background ( Figure S11 ) . Furthermore , a primed induction of RPS5 and RSG1 transcripts was found in the miR472m line relative to WT background treated with flg22 ( Figure 5A ) , supporting a role for miR472 in repressing mRNA accumulation of these CNL mRNAs during flg22 elicitation . Further phenotypic analyses in this line revealed a more pronounced flg22-induced ROS production and callose deposition , thereby mimicking the primed PTI responses observed in the rdr6-elicited mutant ( Figures 5B , 5C ) . We thus conclude that miR472 and RDR6-dependent secondary siRNAs regulate PTI responses likely by targeting a whole repertoire of CNL transcripts . Finally to determine the role of miR472 in basal resistance , we inoculated the virulent Pto DC3000 strain on miR472OE and miR472m lines and monitored bacterial titers in these genetic backgrounds as compared to WT-infected control . We found an increased Pto DC3000 titer in the miR472OE line , and , conversely , a reduced growth of this bacterium in the miR472m line as compared to WT-infected control ( Figures 5D , S10 ) . These results indicate that miR472 not only represses PTI responses but also negatively regulates basal resistance against Pto DC3000 . These results also suggest that a subset of CNLs , which are targeted by miR472 and RDR6-dependent secondary siRNAs , may control basal resistance against Pto DC3000 . The effective targeting of RPS5 mRNAs by miR472 and RDR6-dependent secondary siRNAs ( Figure 4B , 4C ) , together with the well-characterized role of RPS5 in recognizing the bacterial effector AvrPphB and mounting ETI [47] , prompted us to investigate the role of miR472 and RDR6 in RPS5-mediated resistance . For this purpose , the rdr6-15 and miR472m lines were first inoculated with a Pto DC3000 strain carrying AvrPphB and bacterial titers were monitored at 4 days post-inoculation . Results from these analyses indicated a significant enhanced RPS5-mediated resistance in both rdr6 and miR472m as revealed by lower bacterial titers in these mutants as compared to WT-infected plants ( Figure 6A ) , which is consistent with the enhanced accumulation of RPS5 transcripts in these mutant backgrounds ( Figures 3A , 3B , 5A ) . Of note , this phenomenon was specific to RPS5-mediated resistance , because no phenotype was observed upon inoculation of rdr6 and miR472m lines with Pto DC3000 expressing AvrRpt2 , a bacterial effector that is recognized by another CNL that is not targeted by miR472 ( Figure S12 , [13] ) . We next inoculated the Pto DC3000 ( AvrPphB ) strain on the miR472OE reference line and monitored bacterial titers as well as disease symptoms at 4 days post-inoculation . Interestingly , we found a significant enhanced Pto DC3000 ( AvrPphB ) titer in the miR472OE line as compared to WT-infected plants ( Figure 6B ) , which was associated with a rescue of both chlorotic and necrotic disease symptoms in this transgenic plants ( data not shown ) , thereby mimicking the phenotypes observed in rps5 loss-of-function mutants ( Figure 6A , [47] ) . We conclude that overexpression of miR472 is sufficient to compromise RPS5-mediated resistance , which is consistent with the reduced levels of RPS5 mRNAs in this transgenic line ( Figures 5A , S6 ) . Collectively , these results indicate that miR472 and RDR6 negatively regulate not only PTI but also RPS5-mediated resistance , suggesting a critical role for RPS5 and other CNLs in basal and race-specific immunity . The predicted target site of miR472 is embedded within a region encoding the P-loop domain , which is highly conserved in a large repertoire of CNL disease resistance proteins [9] . It is therefore likely that multiple CNLs are controlled by this particular miRNA and , in agreement , 19 CNL transcripts were experimentally validated as miR472 targets in Arabidopsis seedlings overexpressing miR472 ( Figures 4 , S7 ) . This suggests that the enhanced basal resistance phenotype observed in the rdr6 and miR472m mutants might not only be due to the constitutive expression and/or primed induction of the few CNLs that have been characterized in these mutant backgrounds ( e . g . RPS5 ) , but also likely to multiple other relatives that are targeted by these small RNAs , rendering the functional characterization of these CNLs challenging . To circumvent this issue , we first introduced , in the rdr6 mutant background , mutations that abolish CNL-mediated signaling , and subsequently monitored Pto DC3000 titer in these double mutant backgrounds . Since several CNLs are known to trigger SA-signaling/biosynthesis [51] , including RPS5 [52] , we hypothesized that the SA-dependent defense response might be constitutive in the rdr6 mutant background . Consistent with this idea , we found a constitutive expression of the SA-dependent marker gene PATHOGENESIS-RELATED 1 ( PR1 ) and the ISOCHORISMATE SYNTHASE1 ( ICS1 ) ( Figures 7A , 7B ) [51] , [53] , as well as an enhanced resistance to Pto DC3000 , in the rdr6 mutant as compared to WT control ( Figures 7C , 7D , 7E , 7F ) . Importantly , this increased resistance to Pto DC3000 was abolished by introducing mutations that compromise SA-biosynthesis ( the sid2-2 mutation , [53] ) or SA-signaling ( the npr1-1 mutation , [54] ) in the rdr6-15 mutant background ( Figures 7C , 7D ) . These results therefore indicate that the enhanced basal resistance achieved in the rdr6 mutant relies on the constitutive activation of the SA-dependent defense response , which might be initially triggered by the enhanced accumulation of CNLs that are no longer controlled by RDR6-dependent secondary siRNAs in this mutant background . To get further insights into the role of these CNLs in the enhanced basal resistance phenotype observed in the rdr6 mutant , we took advantage of the property of the REQUIRED FOR MLA12 RESISTANCE ( RAR1 ) protein . RAR1 is part of a molecular chaperone complex , containing HEAT SHOCK PROTEIN 90 ( HSP90 ) and SUPPRESSOR OF G-TWO ALLELE OF SKP1 ( SGT1 ) , and plays a major role in NLR protein stability and activity [55]–[65] . Importantly , the steady-state accumulation of several CNL proteins , including RPS5 , was shown to be dramatically impaired in rar1 loss-of-function mutants [58] , [64]–[68] . We thus reasoned that by introducing a rar1 loss-of-function mutation in the rdr6 mutant background , we would destabilize CNL proteins whose cognate mRNAs are targeted by RDR6-dependent siRNAs , and therefore potentially restore disease susceptibility . Consistent with this hypothesis , we found that the increased resistance achieved in the rdr6 mutant was abolished in the rdr6-rar1 double mutant ( Figure 7E ) . It is noteworthy that an enhanced Pto DC3000 titer was also found in the single rar1 and double rdr6-rar1 mutants as compared to WT control , indicating that RAR1 contributes to basal resistance as previously reported [64] . Given that RDR6 was found to negatively regulate RPS5-mediated resistance ( Figure 6 ) , we also monitored Pto DC3000 ( AvrPphB ) titer in the single rdr6 mutant as compared to the rdr6-rar1 double mutant . Results from these analyses indicated that the enhanced RPS5-mediated resistance observed in rdr6 mutants was partially compromised in the rdr6-rar1 mutant ( Figure 7F ) . Collectively , these results indicate that the increased basal and specific resistance observed in the rdr6 mutant is dependent on the proper chaperoning of CNL proteins ( e . g . RPS5 ) , and might therefore be due to the enhanced accumulation of CNL proteins whose cognate mRNAs are no longer controlled by endogenous secondary siRNAs in this silencing-defective mutant . RDR6 has been clearly implicated as a positive regulator of virus and viroid resistance . Indeed silencing of RDR6 in Nicotiana benthamiana results in hyper-susceptibility to some viruses and viroids [29] , [30] . Moreover , in situ hybridization shows that viruses and viroids invade floral and vegetative meristems of N . benthamiana rdr6 RNAi plants [69] , [70] . Here , by combining microbiological , genetic , genomic and molecular techniques , we demonstrate that RDR6 also acts as a negative regulator of PTI , basal defense as well as RPS5-mediated resistance . Indeed , we first showed that knock-out of RDR6 renders the plants more resistant to the hemibiotrophic pathogen Pto DC3000 and to the avirulent Pto DC3000 ( AvrPphB ) strain ( Figures 2D , 6A , 7C , 7D , 7E ) . Furthermore , classical PTI responses such as ROS production , mRNA accumulation of PAMP-response genes as well as callose deposition were increased in rdr6 plants as compared to WT plants upon flg22 treatment ( Figures 2A , 2B , 2C ) [71] . Our results are thus in sharp contrast with the previously reported PTI phenotypes observed in ago1 loss-of-function mutants [38] . Why is there such a discrepancy between these PTGS-defective mutant phenotypes during PTI ? One would argue that AGO1 is not only involved in the siRNA pathway but also in the canonical miRNA pathway . AGO1 impairment has thus additional consequences on the action of several miRNAs necessary for PTI [38] , [72] , thereby leading to the previously reported compromised PTI responses in ago1 loss-of-function mutants such as in other miRNA-defective mutants [37] , [38] . It is also possible that RDR6-derived siRNAs that target disease resistance genes may not only be loaded into AGO1-RISC but also into other as-yet unknown AGO-RISCs , thereby contributing in part to the post-transcriptional regulation of CNLs in an AGO1-independent manner . We also observed a constitutive activation of the SA defense marker gene PR1 and an enhanced expression of ICS1 in the rdr6 loss-of function mutant ( Figures 7A , 7B ) . To examine the involvement of the SA-dependent defense in the enhanced disease resistance phenotype observed in rdr6 mutant , the rdr6-15 mutation was combined with the sid2-2 , a loss-of-function mutation in ICS1 also referred to as SID2 [53] . Inactivation of ISC1/SID2 abolishes rdr6 resistance to Pto DC3000 and similar results were obtained in the npr1 mutant , which is impaired in SA signaling [54] ( Figures 7C , 7D ) . Therefore , the SA-dependent defense pathway plays a critical role in the enhanced basal resistance phenotype observed in the rdr6 mutant . Such constitutive SA-dependent defense response might result from a derepression of a subset of CNL transcripts ( e . g . RPS5 mRNAs ) that are no longer regulated by secondary siRNAs in this silencing-defective mutant . Additionally , it may result from the post-translational activation of R proteins that would be constitutively present in a protein complex with RDR6 and active in the absence of this silencing factor , as observed in classical ‘guardee’ mutants [10] . Further investigations will be necessary to address these possibilities . Moreover , additional experiments will be required to determine whether the constitutive SA-dependent defense response observed in the rdr6 mutant is linked with the mild constitutive PTI responses in this silencing-defective mutant or whether both processes remain independent . We observed a higher expression of PAMP-response marker genes in unchallenged rdr6 mutant as compared to WT seedlings and a significant hyper-induction of FRK1 in the rdr6-elicited mutant ( Figure 2B ) . Furthermore , a more pronounced callose deposition as well as ROS production were observed in the rdr6 mutant challenged with flg22 as compared to WT-elicited seedlings ( Figures 2A , 2C ) , indicating that this silencing-defective mutant is in a physiological situation known as “primed” state [73] . Those results also indicate that RDR6 encodes a novel negative regulator of PTI and further reinforce the idea that PTI is under a tight negative regulatory control as previously reported [2] , [74] , [75] , [76] . Interestingly , an analogous RNA silencing-dependent regulatory phenomenon has been recently described in the transcriptional control of a disease resistance gene during PTI [24] . In this case , flg22 was shown to trigger the repression of a subset of RNA-directed DNA methylation factors and this process was associated with TGS release and with the transcriptional activation of this immune receptor , which is targeted by siRNA-directed DNA methylation in its promoter region [24] . Although RDR6 mRNAs were down-regulated in response to flg22 ( Figure 1 ) , it remains to be tested whether this molecular effect could be accompanied with a decrease in RDR6 protein levels as well as an eventual global release of RDR6-silencing as part of PTI responses . How does RDR6 repress PTI , basal resistance and RPS5-mediated resistance ? A first in silico analysis of small RNA populations derived from rdr6 mutant as compared to wild-type leaf samples allowed us to identify R gene mRNA candidates that are targeted by RDR6-dependent secondary siRNAs ( Figure S4 ) . However , the low abundance of secondary siRNAs in the majority of cases limited the identification of such miR472/RDR6 targets . By contrast , the use of our transgenic line overexpressing miR472 was instrumental in identifying with confidence 19 bona fide CNL target transcripts that contain the miR472 recognition sites as well as a large number of secondary siRNAs located downstream of their miR472-guided cleavage site ( Figures 4 , S9 ) . Among these candidates , we have identified RPS5 and SUMM2 transcripts , which encode functionally relevant disease resistance proteins with well-characterized role in ETI [47] , [50] , [56] . These results therefore suggested that the miR472/RDR6-silencing pathway inhibits the accumulation not only of disease resistance gene transcripts encoding R proteins required for PTI and basal resistance but also of transcripts encoding immune receptors required for ETI . This implicates miR472 and RDR6 in a central regulatory pathway that modulates both ETI and PTI responses . Consistent with this , we found that RDR6 and miR472 act not only as negative regulators of PTI and basal immunity but also as repressors of RPS5-mediated resistance ( Figure 6 ) . In addition , the use of the rar1 mutant , which destabilizes disease resistance proteins including RPS5 [58] , [59] was useful to provide genetic evidence that the enhanced disease resistance phenotypes observed in the rdr6 mutant is likely the result of a higher accumulation of NB-LRR proteins in this silencing-defective mutant ( Figure 7 ) . The present work also provides genetic evidence that miR472- and RDR6-dependent secondary siRNAs efficiently control the steady state levels of three CNL transcripts . Indeed , we first showed that RPS5 , RSG1 and RSG2 mRNAs were moderately up-regulated in untreated rdr6 mutant and significantly hyper-induced in this silencing-defective mutant challenged with flg22 ( Figure 3B ) . Accordingly , a lower level of RPS5 and RSG1 mRNAs was detected in the miR472OE line ( Figure S6 ) and a compromised induction of these CNL transcripts was also observed in this transgenic line challenged with flg22 ( Figures 5A , 5B , 5C ) . Conversely , the miR472 knock-down line displayed higher accumulation of CNL mRNAs , which was associated with increased PTI responses ( Figure 5 ) , therefore mimicking the phenotypes observed in the rdr6 mutant ( Figure 2 ) . Collectively , these results indicate that both Arabidopsis RDR6 and miR472 negatively regulate the steady state levels of these candidate CNL transcripts in normal growth conditions and during PTI , although these effects appear more pronounced during the elicitation possibly due to the concomitant transcriptional activation of these R genes as previously demonstrated for other biotic stress responsive disease resistance genes [17] , [18] , [48] . Based on these results , we propose a model , which integrates the contribution of the miR472/RDR6-dependent PTGS pathway in plant immunity ( Figure 8 ) . In unchallenged conditions , both miR472 and RDR6 are constitutively expressed and negatively regulate a subset of CNL mRNAs at the post-transcriptional level ( Figure 8 ) . MiR472 guides cleavage of RPS5 , RSG1 , RSG2 and at least 16 other CNL transcripts that carry miR472 recognition sites and RDR6 uses 3′ cleavage products as substrates to generate dsRNAs that are presumably processed by DCL4 into 21 nt siRNAs ( Figure 8 ) . These secondary siRNAs can act in cis by guiding mRNA degradation of the CNL transcripts from which they are produced , but also likely in trans presumably by targeting CNLs as well as unrelated mRNAs that display sequence complementary to these small RNA species as was recently suggested in tomato ( [21] , Figure S7 ) . It is also likely that the genes encoding the above immune receptors remain at a transcriptionally inactive state in unchallenged conditions as demonstrated for several other disease resistance genes [17] , [48] . In this case , the concomitant low basal transcriptional expression of CNLs and the miR472/RDR6-dependent post-transcriptional regulatory process would effectively deplete immune receptor mRNAs in the absence of pathogens , thus preventing an autoimmune response that would have detrimental consequences on plant fitness [1] , [77] . This is reminiscent of recent findings on other 22 nt miRNAs/secondary siRNAs that target NLR transcripts in different plant species [20] , [21] , [22] , as well as with the observation that the production of siRNAs at the disease resistance RPP4 cluster repress basal expression of several R gene transcripts within this cluster and likewise prevent constitutive activation of the SA-dependent defense pathway [18] . Our model also suggests that the mature form of miR472 is down-regulated during PTI , as a 4-fold decrease in the accumulation of this microRNA was observed in small RNA libraries generated by Li et al [38] upon flg22 treatment , which was confirmed in Arabidopsis leaves and seedlings treated with flg22 ( Figure S13 ) . We thus propose that upon pathogen detection , and perhaps also perception of non-adapted microbes , microbe-associated molecular patterns trigger the down-regulation of miR472 , which in concert with the eventual transcriptional activation of CNLs , may contribute to the transient enhanced accumulation of CNL mRNAs/proteins at an early phase of the elicitation ( Figure 8 ) . This gene regulatory mechanism may also be reinforced by the down-regulation of RDR6-dependent silencing pathway as suggested by the rapid repression of RDR6 mRNAs during PTI ( Figure 1 ) . At a later phase of the elicitation , we propose that this double post-transcriptional layer of regulation mediated by miR472 and RDR6 likely trigger a robust resilencing of these CNL transcripts to prevent a sustained activation of the plant immune response . Although R proteins have been extensively characterized in ETI [10] , there is increasing evidence that these immune receptors can also contribute to basal defense as well as PTI responses in plants [78] . For example , a compromised basal resistance to virulent Pto DC3000 was previously reported in a rar1 loss-of-function mutant [64] , and confirmed in the present study ( Figure 7E ) , suggesting that plant NLRs contribute to basal immunity . More recently , a subclade of CNL proteins , characterized as ‘helper NB-LRR’ , where not only required for ETI but also for basal resistance and this process was independent of their P-loop motifs [79] . Importantly , these CNLs additionally regulate PAMP-triggered SA-accumulation in response to a disarmed P . syringae strain , which provides evidence that plant NLRs contribute to PTI [79] . Nevertheless , these CNLs do not control early events of PTI responses triggered by flg22 or the elongation factor-derived peptide elf18 , indicating that these immune receptors likely act downstream or independently of these early PTI signaling events [79] . In the present work , we showed that another subclade of CNLs , which are targeted by miR472 and RDR6-dependent siRNAs , possibly contribute to multiple PTI signaling events , including potentially flg22-triggered callose deposition and ROS production ( Figure 3B ) . Interestingly , the product of one if this mRNA target , the RPS5 protein , was previously shown to reside in the same protein complex as the PTI receptor FLS2 [80] , further supporting a molecular link between ETI and PTI components . In conclusion we have established a direct link between miR472/RDR6-dependent PTGS and plant immunity . We showed that both miR472 and RDR6 act as negative regulators of PTI and ETI , presumably by repressing a subset of CNLs at the post-transcriptional level . Our data therefore sustain previous anticipations suggesting that in addition to their role in specific resistance , R proteins contribute to PTI [10] , [64] , [78] , [81] , [82] . Furthermore , given that flg22 as well as disarmed bacteria were shown to trigger Systemic Acquired Resistance ( SAR ) , such as in response to pathogens expressing Avr products [83]–[86] , we speculate that a potential release of miR472- and eventually of RDR6-dependent PTGS may also occur in distal tissues , and thereby might contribute to the transient derepression of a whole repertoire of disease resistance genes as part of the SAR response . Arabidopsis thaliana seeds from the Col-0 accession were used as wild-type , the rdr6-15 T-DNA insertion line has been previously described in Xi et al [87] . We also used sid2-2 , npr1-1 and rar1-21 mutant alleles . Plants were genotyped with the following primers and conditions: rdr6 ( RDR6_LP:TGAATCCATTCCTGAACAAGC; RDR6_RP: CAATGCAACCTCATCTTGGATG; LB3: TAGCATCTGAATTTCATAACCAATCTCGATACAC ) , npr1 ( 1g64280_F: AGGGGATATACGGTGCTTCAT; 1g64280_R: GAGCAGCGTCATCTTCAATTC ) ; sid2 ( sid2_F:CAGTCCGAAAGACGACCTCGAGTT;sid2_R:CTCATCATCTTCCTTCGTAAGTCTCC ) ; rar1 ( 5g51700_F: AAGCAGGGAGTAAGTCAAATTTAC; 5g51700_R CAAACTGAAATCATGACTTCTTTG ) . All plants were grown in short days conditions subjected to a cycle of 8 h and 16 h of light and darkness , respectively , at a day/night temperature of 22 . 5/18 . 5° with 50–60% humidity for about 5–6 weeks . The plants were watered 16 h before inoculation to promote stomatal opening , thereby facilitating inoculation . Pseudomonas syringae pv . tomato DC3000 ( Pto DC3000 ) was grown at 28°C on NYGB medium ( 5 g L−1 bactopeptone , 3 g L−1 yeast extract , 20 ml L−1 glycerol ) containing kanamycin ( 50 mg mL−1 ) and rifampicin ( 25 mg mL−1 ) for selection . Pto DC3000 , Pto DC3000 AvrPphB and Pto DC3000 AvrRpt2 from overnight culture were collected , washed once and resuspended in 10 mM MgCl2 at a concentration of 5×105 colony-forming units ( CFU ) mL−1 . A . thaliana leaves were infiltrated with bacterial suspensions using a needleless syringe . Leaves were harvested immediately ( 0 dpi ) or after 4 days . Two leaf discs ( d = 0 . 4 mm ) from two different leaves were washed in 10 mM MgCl2 and then ground with a Microfuge pestle . After grinding of the tissue , the samples were diluted 1∶10 serially . Samples were plated on NYGA solid medium ( NYGB with 10 g L−1 agar ) supplemented with antibiotics . Plates were placed at 28°C for 4 days and the CFU were counted . For spray inoculation bacteria were resuspended in 10 mM MgCl2 at OD600 of 0 . 2 ( 108 CFU/mL ) and Silwet was added to a final concentration of 0 . 04% . All experiments presented were repeated three times and statistical differences were detected with a Wilcoxon test ( * , P<0 . 05; ** , P<0 . 01 ) . Reactive oxygen species released by leaf discs were assayed by H2O2-dependent luminescence of luminal [88] . Leaf discs were deposed into 96-well plate and incubated overnight in 200 µL H2O in a growth chamber . The next morning , 100 µL H2O containing 20 µM luminol and 1 µg horseradish peroxidase ( Sigma ) with or without 100 nM flg22 were added . Luminescence was immediately measured for 45 min using a Tristar LB 941 plate reader ( Berthold technologies , Thoiry ) . At least 25 to 30 discs were tested by conditions . For callose detection , leaves were infiltrated with 100 nM flg22 or water using a needleless syringe . After 15 h , about ten leaves from at least four independent plants were cleared by immersion in an alcoholic lactophenol solution by the method of Shipton and Brown [89] modified by Adam and Sommerville [90] . They were rinsed in 50% ethanol , then in water . Callose was detected by staining for 30 min in 150 mM K2HPO4 ( pH 9 . 5 ) buffer containing 0 . 01% aniline blue ( Sigma-Aldrich ) . After staining each leaf was mounted in 50% glycerol and examined with an Olympus Macro Zoom System Microscope MVX10 fluorescent microscope ( excitation filter 365 nm and barrier filter 420 nm ) . Representative pictures are shown . The number of callose deposits per picture was determined using ImageJ ( National Institutes of Health , Bethesda , MD , U . S . A . ) and compared using a Wilcoxon test ( P<0 . 05 ) . We analyzed 25 to 30 pictures corresponding to more than five independent leaves for each treatment . For RNA extraction , leaves or seedlings were collected , immediately frozen in liquid nitrogen , and then stored at −80°C . Total RNA was prepared by TRIzol ( Invitrogen ) extraction as recommended by the supplier ( Invitrogen ) . For RT-PCR analysis , first-strand cDNA was synthesized using Superscript reverse transcriptase ( Invitrogen , ) from 1 µg of RNase-free DNaseI-treated ( Promega ) total RNA in a 20 µl reaction volume . Quantitative PCR reactions were performed on 1/40 of cDNA , 300 nM final concentration of each primer pair and LightCycler 480 SYBR Green I Master 2× conc . ( Roche ) . PCR was performed in 384-well optical reaction plates heated at 95°C for 10 min , followed by 45 cycles of denaturation at 95°C for 15 s and annealing and elongation at 60°C for 30 s . A melting curve was performed at the end of the amplification by steps of 1°C ( from 95°C to 50°C ) . Each experiment was repeated two to three times . Transcript levels were normalized to that of At2G36060 , At4G29130 and At5G13440 genes . These reference genes display invariant expression over hundreds of publicly available microarray experiments . The gene-specific primers used in this analysis were listed in Figure S14 . For miR472 quantification , total RNA was isolated from plants using TRIzol reagent ( Invitrogen ) and treated with RNase-free DNaseI ( Promega ) . Small RNAs were polyadenylated with ATP by poly ( A ) polymerase following the manufacturer's directions for the Poly ( A ) Tailing Kit ( Ambion ) . After phenol-chloroform extraction and ethanol precipitation , the RNAs were reverse-transcribed with 200 U SuperScript III Reverse Transcriptase ( Invitrogen ) and 0 . 5 µg poly ( T ) adapter ( Figure S14 ) according to the manufacturer's protocols ( Invitrogen ) . The cDNAs were used for qPCR with miR472 as one primer and the reverse primer as described by Shi and Chiang [91] . 5 , 8S ribosomal RNA gene was used as internal control as previously described [91] . Sequences of miR472 , reverse primer , poly ( T ) adapter and 5S primers are listed in Figure S14 . Total cellular RNA ( 5 µg ) , extracted using TRIzol reagent ( Invitrogen ) was processed into sequencing libraries using adapted Illumina protocols and sequenced at Fasteris ( http://www . fasteris . com , Switzerland ) using the Hi-seq 2000 sequencer . All next-generation sequencing data have been deposited to the NCBI Gene Expression Omnibus ( GEO ) . We took advantage of publicly available sRNA libraries from leaf tissue [92] . These data correspond to 2 replicates of WT and rdr6 sRNA sequenced using Illumina Genome Analyser technology . Replicates were pooled and sequence reads were matched against the Arabidopsis thaliana genome ( TAIR10 ) using MUMmer v3 . 0 [93] . Only 15 to 30-nt long sRNAs reads with perfect match over their entire length were analysed further ( 2 434 780 and 1 753 064 for WT and rdr6 respectively ) . The number of 20–22 nt reads matching TAIR10 annotated protein coding genes locus or tasiRNA were then compared between WT and rdr6 libraries by differential analysis with NOISeq [94] using the parameters indicated below: k = NULL , norm = “rpkm” , long = 1000 , q = 0 . 90 , pnr = 0 . 5 , nss = 1000 , v = 0 . 02 , lc = 1 . The WT and miR472OE sRNA libraries , containing 17 828 872 and 30 869 878 sRNA reads respectively , were processed using the same methods . Over those reads , 88 . 9% are 15 to 30-nt long and can be perfectly aligned to Arabidopsis genome . The miR472OE line was validated by comparing the number of reads mapping to all miRNA stem-loop loci ( miRBase release 19; [95]–[98] ) between WT and mutant sRNA libraries . Differential analysis of 20–22 nt reads in genes has then been done as described in the previous paragraph .	Virus resistance relies in some plant-viral interactions on the RNA-DEPENDANT RNA POLYMERASE 6 ( RDR6 ) , a major actor of RNA silencing that acts at the post-transcriptional level . Here , we demonstrate that RDR6 also plays a role in basal defense and race-specific resistance . RDR6 and the microRNA miR472 , which targets the mRNAs of disease resistance genes of coiled-coil nucleotide-binding leucine-rich-repeats family ( e . g . RPS5 ) , act in cooperation to control post-transcriptionally these immune receptors . Induction of these resistance genes is primed in rdr6- and miR472-elicited mutants and this effect is associated with an enhanced basal and race-specific immunity in these backgrounds .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "plant", "science", "genomics", "immunology", "microbiology", "biology" ]	2014	The Arabidopsis miR472-RDR6 Silencing Pathway Modulates PAMP- and Effector-Triggered Immunity through the Post-transcriptional Control of Disease Resistance Genes
Protozoan parasites infect and kill millions of people worldwide every year , particularly in developing countries where access to clean fresh water is limited . Among the most common are intestinal parasites , including Giardia lamblia and Entamoeba histolytica . These parasites wreak havoc on the epithelium lining the small intestines ( G . lamblia ) and colon ( E . histolytica ) causing giardiasis and amebiasis , respectively . In addition , there are less common but far more deadly pathogens such as Naegleria fowleri that thrive in warm waters and infect the central nervous systems of their victims via the nasal passages . Despite their prevalence and associated high mortality rates , there remains an unmet need to identify more effective therapeutics for people infected with these opportunistic parasites . To address this unmet need , we have surveyed plants and traditional herbal medicines known throughout the world to identify novel antiparasitic agents with activity against G . lamblia , E . histolytica , and N . fowleri . Herein , we report Larrea tridentata , known as creosote bush , as a novel source for secondary metabolites that display antiparasitic activity against all three pathogens . This report also characterizes the lignan compound classes , nordihydroguairetic acid and demethoxyisoguaiacin , as novel antiparasitic lead agents to further develop more effective drug therapy options for millions of people worldwide . Intestinal protozoan parasite infections , through contaminated water and food supplies , are global health problems affecting hundreds of millions of people annually . The two most common intestinal parasites are Giardia lamblia and Entamoeba histolytica , which can lead to giardiasis or invasive amebiasis , respectively . G . lamblia and E . histolytica have simple infection life cycles that begin with ingesting viable cysts , excystation , followed by trophozoite multiplication in the small intestine or trophozoite migration and invasion in the colon ( Fig 1A ) [1–3] . Annually , giardiasis , has an estimated worldwide prevalence of 200 million cases [4] , and according to the World Health Organization ( WHO ) giardia infections contribute substantially to the 846 , 000 deaths annually from diarrheal disease [5 , 6] . Once G . lamblia has excysted in the small intestines , trophozoites attach to epithelial cells and elicit aberrant signaling events that disrupt organ function including the induction of programed cell death or apoptosis [3] . Although less prevalent than G . lamblia , E . histolytica infections lead to 50 million cases of invasive disease and up to 100 , 000 deaths , annually [7] . Invasive amebiasis is characterized by profound intestinal tissue damage and ulceration [8] . Recently , Ralston and colleagues determined trogocytosis as the mechanism by which E histolytica feeds on its host . The term ‘trogocytosis’ was taken from the Greek word trogo which means to nibble [8 , 9] . The amebae damage and consume the intestinal mucosa epithelium by nibbling away at epithelial cell membranes , triggering cell death . Interestingly , Ralston et al . concluded that amebae feed on bacteria in the gut for nutrition but that host cell ingestion is done by the amebae to create a more spacious environment [8] . Free-living ameba Naegleria fowleri has been described as the cause of primary amebic meningoencephalitis ( PAM ) in more than 16 countries [10] . Until 2012 , about 310 cases have been reported globally with a fatality rate of more than 95% [11] . According to the Centers for Disease Control and Prevention ( CDC ) , 138 cases of PAM have been reported in the U . S . A . from 1962–2015 with a 98% mortality rate . PAM results from water containing N . fowleri entering the nasal cavity followed by migration of the amebae to the brain ( Fig 1B ) [12–17] . Within the brain , N . fowleri causes extensive inflammation , hemorrhage , and necrosis . The time from initial exposure to onset of illness is usually 5–7 days but may be as early as 24 h , leading to death in 3 to 7 days [18] . Treatment for giardiasis and invasive amebiasis is largely limited to the nitroimidazole drug class ( e . g . metronidazole ) [19] . Metronidazole , is the primary drug of choice , which requires a relatively long treatment time and high dosage to eradicate intestinal parasite infections [20] . Moreover , metronidazole is both mutagenic and carcinogenic and its use presents other significant adverse effects [21 , 22] . In addition , G . lamblia and E . histolytica drug resistance and treatment failures remain an increasing problem [23–26] . Amphotericin B remains a cornerstone of therapy for PAM but is not FDA-approved for this indication and has had limited success despite its worldwide use [27] . Treatment with amphotericin B requires a high dosage and its use is frequently associated with renal toxicity and anemia , among other adverse effects [27] . Recently , an investigational drug , miltefosine , clinically used to treat leishmaniasis , has shown some promise in combination with other drugs as a treatment for PAM [28] . The CDC , through an established protocol with the FDA , is now directly providing miltefosine to the clinicians as a treatment option for PAM . However , it is still not FDA approved and has limited availability in the U . S . A . Furthermore , G . lamblia , E . histolytica and N . fowleri are listed by the United States National Institutes of Health and the Centers for Disease Control as a category B biodefense/bioterrorism pathogens due to their low infectious dose and potential for dissemination through compromised food and water supplies . Given the prevalence and mortality caused by these protozoan pathogens , compounded by their potential bioterrorism threat , more effective antiparasitic agents is a critical unmet need to treat the current pandemic and avert future outbreaks and deaths . Natural products have played an important role throughout history in the treatment of human disease through traditional medicines and as a source for effective pharmaceutical development [29 , 30] . In particular , plants have been a vast source of secondary metabolites that display potent antiparasitic activity , including protozoan parasites [30–33] . For example , G . lamblia and E . histolytica are endemic to Mexico and infections are prevalent [34 , 35] . Moreover , nitroimidazole drugs display limited efficacy in the Mexican population [36] . Therefore , scientists have turned to native plants used as Mexican traditional medicines for intestinal diseases in the search for novel more effective antiparasitic agents [37 , 38] . Similarly , using our established assays [39 , 40] , we have surveyed plants used as traditional medicines from around the world and that are common to the southwestern United States and throughout Mexico . Herein , we report the discovery of Larrea tridentata , commonly known as creosote bush or chaparral , as a novel source for antiparasitic secondary metabolites [41] . Though the extract of L . tridentata earlier showed antiparasitic activity against Trypanosoma brucei rhodesiense , T . cruzi , Leishmania donovani and Plasmodium falciparum [42] , this is the first report to show their activity against a free-living amoeba N . fowleri and against diarrhea causing parasites E . histolytica and G . lamblia . We have identified seven known compounds 1–7 ( Fig 2 ) with 1–6 displaying antiparasitic activity against E . histolytica , G . lamblia , and N . fowleri . Compounds 1 and 2 showed better activity against N . fowleri than the current drug miltefosine . In addition , we have identified two secondary metabolites , compounds 8 and 9 ( Fig 2 ) , that we isolated from the same active fractions as 1–7 that appeared to have novel structures . Compound 9 displayed modest antiparasitic activity against G . lamblia and N . fowleri . An examination of the literature indicated that 8 and 9 structures have been reported [43 , 44] . Interestingly , compound 8 has not previously been isolated or structurally characterized from the creosote plant , rather , Cho and colleagues used Larreatricin 3’-hydroxylase enzyme purified from creosote and the known secondary metabolite from creosote , larreatricin , to enzymatically prepare 8 , albeit in very low yield [44] . However , the structure of 9 was dubiously deduced from Graziela mollisima as an impure mixture with insufficient analytical data to accurately characterize the structure [43] . Therefore , this is the first report to unambiguously characterize the novel secondary metabolites 8 and 9 from L . tridentata . Since compounds 1 and 2 were found more active against N . fowleri than miltefosine , we selected these two compounds to investigate their ability to inhibit N . fowleri cysteine protease , an enzyme shown to play an important role in host tissue invasion by N . fowleri [45] . 1H , 13C and 2D NMR spectra were recorded on a Bruker Avance III spectrometer ( 400 MHz for 1H NMR and 100 MHz 13C NMR ) . Chemical shifts are recorded in ppm ( δ ) using residual solvent signal as internal reference , and coupling constants ( J ) are reported in Hz . The following splitting abbreviations were used for NMR signals: s = singlet , d = doublet , t = triplet , q = quartet , m = multiplet , br = broad . High-resolution mass spectra ( HRMS ) were recorded on a Bruker Q-TOF-2 Micromass spectrometer equipped with lock spray , using ESI with methanol as the carrier solvent . Accurate mass measurements were performed using leucine enkephalin as a lock mass and the data were processed using MassLynx 4 . 1 . Exact m/z values are reported in Daltons . Optical rotations were measured in CH3OH on a JASCO P1010 polarimeter at 589 nm ( Na D-line ) with a path length of 1 dm and are reported with implied units of 10−1 deg cm2 g-1 . Concentrations ( c ) are given in g/100 mL . UV was measured in CH3OH on an Agilent 8453 UV-Visible Spectrophotometer . Analytical and preparative HPLC were performed on a Shimadzu Prominence HPLC system equipped with LC-6AD pumps , an autosampler ( SIL-20AC ) and manual injection port ( Rheodyne , 3725i ) , a column oven ( CTO-20A , temperature set at 27°C ) , a photo diode array detector ( SPD-M20A , using a Deuterium lamp and a tungsten lamp as light sources ) and a system controller ( CBM-20A ) . A Phenomenex Kinetex C18 reversed phase column ( 5 μm , 100 Å , 250 ✕ 4 . 6 mm ) fitted with a guard cartridge , with a flow rate of 0 . 7 mL/min was used for analytical chromatography , and a Phenomenex Kinetex C18 reversed phase column ( 5 μm , 100 Å , 250 ✕ 21 . 1 mm ) fitted with a guard cartridge with a flow rate of 5 . 0 mL/min was used for preparative chromatography . The HPLC data were processed using LabSolutions Lite software ( version 5 . 22 ) . The dried powdered material ( 11 . 0 g ) of L . tridentata ( Mountain Rose Herbs ) was extracted with methanol at room temperature for 72 h . After filtration through Celite , the methanol extract was concentrated under reduced pressure to give a crude residue ( 2 . 55 g ) . The extract residue ( 2 . 53 g ) was treated with water ( 150 mL ) and partitioned against hexane ( 150 mL × 3 ) , ethyl acetate ( 150 mL × 3 ) and n-butanol ( 150 mL × 2 ) successively to yield a hexane fraction ( 128 . 6 mg ) , an ethyl acetate fraction ( 1 . 5 g ) , a n-butanol fraction ( 411 . 7 mg ) , and a water fraction ( 504 . 2 mg ) , respectively . The parasite active ethyl acetate fraction ( 808 . 5 mg ) was then chromatographed on a Sephadex LH-20 column eluted with 20% hexane in CH2Cl2 ( 200 mL ) , 60% CH2Cl2 in acetone ( 400 mL ) , 20% CH2Cl2 in acetone ( 200 mL ) , 20% CH2Cl2 in methanol ( 200 mL ) , and 100% methanol ( 200 mL ) . Ten fractions were collected: fractions A ( 12 . 9 mg ) and B ( 13 . 4 mg ) from 20% hexane in CH2Cl2; fractions C ( 28 . 6 mg ) , D ( 386 . 1 mg ) , E ( 181 . 9 mg ) , and F ( 81 . 6 mg ) from 60% CH2Cl2 in acetone; fractions G ( 48 . 7 mg ) and H ( 35 . 9 mg ) from 20% CH2Cl2 in acetone; fraction I ( 52 . 6 mg ) from 20% CH2Cl2 in methanol and fraction J ( 2 . 1 mg ) from 100% methanol . Fraction E ( 138 . 9 mg ) was chromatographed on reverse phase preparative HPLC and eluted with gradient 20–100% acetonitrile in water for 40 min to yield 1 ( 11 . 7 mg ) and 3 ( 43 . 3 mg ) as yellowish resinous solid along with sub-fraction E1 ( 11 . 0 mg ) . Sub-fraction E1 was re-chromatographed under similar HPLC conditions to afford 2 ( 4 . 0 mg ) , 7 ( 3 . 3 mg ) , and 8 ( 1 . 8 mg ) . Fraction D ( 386 . 1 mg ) was chromatographed on silica gel column ( 13 . 0 g ) eluted with increasing amounts of methanol in CH2Cl2 to afford seven fractions , D1 ( 0 . 6 mg ) , D2 ( 181 . 1 mg ) , D3 ( 73 . 5 mg ) , D4 ( 65 . 6 mg ) , D5 ( 5 . 6 mg ) , D6 ( 5 . 5 mg ) , D7 ( 7 . 5 mg ) . Fraction D2 ( 133 . 0 mg ) was chromatographed on preparative HPLC and eluted with isocratic 50% acetonitrile in water to yield 2 ( 29 . 7 mg ) , 4 ( 14 . 6 mg ) , 5 ( 2 . 0 mg ) and 6 ( 4 . 5 mg ) as yellow resinous solids . ( 7R , 7’R ) -7 , 7’-bis ( 4’ , 3 , 4-trihydroxyphenyl ) - ( 8R , 8’S ) -8 , 8’-dimethyltetrahydrofuran ( 8 ) : colorless oil; [α]D25–88 . 1 ( c 0 . 16 , CH3OH ) ; UV ( MeOH ) λmax ( log ε ) 211 ( 3 . 44 ) ; 236 ( 2 . 54 ) , 282 ( 1 . 64 ) ; 1H and 13C NMR data , see Table 1; HRESIMS m/z 301 . 1506 [M + H]+ ( calcd for C18H21O4 , 301 . 1439 ) 3-Methoxy-6 , 7 , 4’-trihydroxyflavonol ( 9 ) : Yellow solid; UV ( MeOH ) λmax ( log ε ) 211 ( 5 . 06 ) , 266 ( 4 . 90 ) , 348 ( 4 . 86 ) ; 1H and 13C NMR data , see Table 2; HRESIMS m/z 301 . 0690 [M + H]+ ( calcd for C16H13O6 , 301 . 0712 ) . Trophozoites of E . histolytica HM1: IMSS and G . lamblia WB strains were axenically maintained in TYI-S-33 medium supplemented with penicillin ( 100 U/ml ) , streptomycin ( 100 μg/ml ) [46 , 47] . Trophozoites of N . fowleri strain KUL were axenically cultured in Nelson’s medium supplemented with 10% FBS at 37°C [45] . All experiments were performed using trophozoites harvested during the logarithmic phase of growth . Four solvent partitioned fractions of an aqueous methanolic extract of L . tridentata and compounds 1–9 were screened for activity against E . histolytica , G . lamblia , and N . fowleri . For primary screening , the positive control for E . histolytica and G . lamblia was 5 μg/mL of metronidazole ( Sigma-Aldrich ) and 46 μg/mL of amphotericin B for N . fowleri ( Sigma-Aldrich ) . Test samples were diluted to 10 mg/mL of extracts , HPLC fractions , and pure compounds in DMSO . Finally , 0 . 5 μL of diluted sample was transferred to white , solid bottom tissue culture 96-well plates ( E&K Scientific ) followed by addition of 99 . 5 μL trophozoites ( 5 , 000 E . histolytica and G . lamblia , and 10 , 000 N . fowleri ) in TYI-S-33 medium or Nelson’s medium . The final concentration for test compounds was 50 μg/mL and 0 . 5% DMSO , which was the negative control and compound vehicle that we have shown has no effect the growth rate of trophozoites [39 , 40 , 48] . Assay plates were incubated for 48 h at 37°C . E . histolytica and G . lamblia plates were kept in the GasPak EZ Anaerobe Gas Generating Pouch System ( VWR ) to maintain anaerobic condition throughout the incubation period . Screening was performed in duplicate using the CellTiter-Glo assay ( Promega ) and luminescence was measured using an EnVision plate reader ( PerkinElmer ) [40 , 48] . The antiparasitic activity of 1–6 and 9 were confirmed by EC50 dose response experiments , using the CellTiter-Glo assay , conducted in triplicate over a concentration range from 5-to-700 μM against trophozoites ( Table 3 ) . Miltefosine and metronidazole , current drugs for the treatment of PAM and amebiasis and giardiasis were also tested in triplicate as positive controls for EC50 determination ( Table 3 ) . Dose response curves including standard deviation ( SD ) calculation were processed using GraphPad Prism software 5 . 0 . Percent inhibition relative to maximum and minimum reference signal controls was calculated using the formula: % Inhibition = [ ( mean of Maximum Signal Reference Control—Experimental Value ) / ( mean of Maximum Signal Reference Control—mean of Minimum Signal Reference Control ) ] × 100 . The HUVEC-TERT2 cell line was purchased from Evercyte GmbH ( Vienna , Austria ) and cultured and maintained in endothelial cell basal medium ( Lonza ) as described previously [49 , 50] . Briefly , cells were seeded into a white 384-well solid bottom plate ( Nunc , ThermoFisher ) at a density of 1000 cells/well in 39 μL of media using a Janus liquid handler ( PerkinElmer ) . Serial dilutions using 1 μL of compound 1 and 2 at varying concentrations were dispensed into each well in triplicate . After 48 h incubation , 40 μL of CellTiter-Glo reagent ( Promega ) was added into each well . The contents were mixed for 2 min on a microplate shaker to induce cell lysis and further incubated at room temperature for 10 min to stabilize the luminescent signal . Luminescence was measured with an EnVision plate reader ( PerkinElmer ) and %inhibition calculations were performed using the following formula for single-point normalization: %Inhibition = ( 1-Raw Sample Value/Mean of DMSO Signal Reference Value ) × 100 . Dose response curves including EC50 calculations were processed using GraphPad Prism software . To prepare the cell lysate , N . fowleri trophozoites were removed from the culture flask surface by incubating in an ice bath for 10 min , centrifuged at 300 g for 10 min , and washed twice with PBS ( pH 7 . 2 ) . The cells were disrupted by four cycles of freeze thawing in PBS [51] . Protein concentration was quantified by the method of Bradford ( Bio-Rad ) . The activity of the cysteine protease present in the crude extract after incubating in presence and absence of different concentrations of compounds 1 and 2 was assayed by the liberation of the fluorescent leaving group , 7-amino-4-methyl coumarin ( AMC ) , from the peptide substrate Z–Phe–Arg–AMC ( 40 μM ) ( where Z is benzyloxycarbonyl , R&D Systems ) [45] . The assay was performed at 25°C in an automated microtiter plate spectrofluorometer ( EnVision , PerkinElmer ) with excitation wavelength at 355 nm and emission wavelength at 460 nm [52] . Enzyme samples were added to the reactivation buffer ( 10 mM Tris , 5 mM EDTA , 50 mM NaCl , pH 7 . 4 , 10 mM DTT ) , and preincubated for 20 min at 37°C prior to the hydrolysis of substrate . The rate of substrate hydrolysis at ambient temperature was determined from the rate of increase of fluorescence , monitored on a continuously recording spectrofluorometer and measured as RFU/min/μg protein . An aqueous methanolic extract of the creosote plant was partitioned against hexane , ethyl acetate and n-butanol successively to obtain four solvent partitioned fractions . These fractions were tested for antiparasitic activity , the ethyl acetate fraction showed activity at 50 μg/mL and was selected for further study . It was fractionated on Sephadex LH-20 and the fractions were subjected to chromatographic separation by HPLC to yield 1–9 as pure compounds . Compound 1 was obtained as a yellow resinous mass . The 1H , 13C , and HMQC NMR ( acetone-d6 ) indicated 9 carbon resonances and corresponding proton signals , consisting of one methyl [δH 0 . 83 d ( 6 . 6 ) ] , four methines [δH 1 . 74 m] , three aromatic signals displaying an ABC splitting pattern [δH 6 . 52 dd ( 7 . 9 , 1 . 8 ) ; δH 6 . 69 d ( 1 . 8 ) ; and δH 6 . 73 d ( 7 . 9 ) ] , and one methylene [δH 2 . 21 dd ( 13 . 3 , 9 . 2 ) ; δH 2 . 70 dd ( 13 . 3 , 5 . 0 ) ] . These data were identical with the known creosote secondary metabolite , nordihydroguairetic acid ( NDGA ) ( Table S1 and Fig . S1-S3 in S1 Appendix ) [53] . Next , we identified known compound 2 as 3’-O-methylnordihydroguairetic acid ( 3’-O-methyl-NDGA ) [54] . Although similar in structure to 1 , compound 2 is non-symmetrical , which revealed the full 19 carbon resonances and corresponding proton signals as follows: two methyls [δH 0 . 82 d ( 6 . 6 ) , 0 . 83 d ( 6 . 6 ) ] , eight methines ( δH 1 . 74 m , 2H ) , six aromatics [δH 6 . 58 dd ( 8 . 0 , 2 . 0 ) , δH 6 . 61 d ( 1 . 9 ) , δH 6 . 64 dd ( 8 . 0 , 1 . 9 ) , δH 6 . 67 d ( 2 . 0 ) , δH 6 . 77 d ( 8 . 0 ) , and δH 6 . 82 d ( 8 . 0 ) ] , and two methylenes [δH 2 . 25 dd ( 13 . 1 , 9 . 3 ) , δH 2 . 71 dd ( 13 . 3 , 4 . 8 ) , δH 2 . 25 dd ( 13 . 1 , 9 . 4 ) , δH 2 . 68 dd ( 13 . 3 , 5 . 0 ) ] . In addition , DEPT-135 and HMQC supported the presence of two methyls ( δc 16 . 6 , 16 . 4 ) , eight methines of which two aliphatic ( δc 39 . 3 , 39 . 1 ) and six aromatic ( δc 113 . 2 , 115 . 4 , 115 . 8 , 116 . 9 , 121 . 2 , 122 . 3 ) , two methylenes ( δc 40 . 0 , 39 . 2 ) and six quaternary aromatic ( δc 134 . 1 , 134 . 3 , 143 . 8 , 145 . 4 , 145 . 7 , 48 . 1 ) ( Table S2 and Fig . S4-S8 in S1 Appendix ) . We identified compound 3 as Nor-3’-demethoxyisoguaiacin and 4–6 as analogs of 3 that have a tetrahydronaphthalene ring system [54 , 55] . The 1H NMR ( CDCl3 ) displayed the following signals: two methyls [δH 0 . 88 d ( 6 . 9 ) , 0 . 89 d ( 6 . 9 ) ] , nine methines including three aliphatic [δH 3 . 57 d ( 6 . 2 ) , 1 . 89 m , 1 . 99 m] , two aromatic singlets ( δH 6 . 60 s , δH 6 . 29 s ) resulting from an A2B2 tetra-substituted phenyl ring , four signals giving an A2B2 splitting pattern [δH 6 . 86 ( 2H , d 8 . 5 ) , δH 6 . 69 ( 2H , d 8 . 5 ) ] due to a 1 , 4-disubstituted phenyl , and one methylene [δH 2 . 83 dd ( 16 . 4 , 5 . 5 ) , δH 2 . 41 dd ( 16 . 4 , 7 . 2 ) ] . The 13C NMR ( acetone-d6 ) displayed eighteen signals and HMQC supported the presence of two methyls ( δc 16 . 1 , 16 . 3 ) , three methines ( δc 50 . 8 , 41 . 8 , 30 . 1 ) , one methylene ( δc 35 . 7 ) , one A2B2 substituted phenyl ( δc 115 . 9 d , 117 . 7 d , 128 . 1 s , 130 . 7 s , 140 . 0 s , 144 . 4 s ) , and one 1 , 4-disubstituted phenyl [δc 115 . 7 ( 2C , d ) , 130 . 8 ( 2C , d ) , 139 . 3 s , 156 . 3 s] ( Table S3 , Fig . S9-S13 in S1 Appendix ) . 4–6 were easily dereplicated due to varying methoxy and phenol substituents . Specifically , compound 4 ( Nor-isoguaicin ) has a methoxy in the 3’-position , which was determined by the ABC proton splitting pattern [δH 6 . 79 d ( 8 . 0 ) , δH 6 . 52 d ( 1 . 8 ) , δH 6 . 50 dd ( 8 . 0 , 1 . 8 ) ] from the tri-substituted phenyl ring ( Table S4 and Fig . S14-S18 in Appendix ) . Conversely , compounds 5 ( 3’-Demethoxyisoguaiacin ) has a methoxy group in the 7 position of the tetra-substituted ring ( Table S5 and Fig . S19-S22 in S1 Appendix ) and 6 ( 6 , 3'-Di-O-demethylisoguaiacin ) which contains a 3’ , 4’-dihydroxy phenyl moiety were determined by comparison with the reported chemical shifts ( Table S6 and Fig . S23-S25 in S1 Appendix ) [54 , 56] . Finally , 7 was purified as a colorless oil and identified as 3-hydroxy-4-epi-larreatricin with 1H and 13C NMR matching the known literature structure ( Table S7 and Fig . S26-S30 in S1 Appendix ) [57] . During the purification of 1–7 we identified lignan 8 and flavanol 9 , however , these secondary metabolites have never been isolated from the creosote plant ( 8 ) or were not structurally well characterized ( 9 ) . Therefore , we report herein the isolation and structure elucidation from the creosote plant . Compound 8 , was purified as a colorless oil and the molecular formula was deduced from the HRMS and 13C NMR as C18H20O4 . The 1H NMR ( Table 1 ) displayed signals attributable to two methyl groups [δH 0 . 97 d ( 6 . 6 ) , δH 0 . 57 d ( 7 . 1 ) ] , and eleven methines , including: two oxygenated aliphatic protons [δH 5 . 38 d ( 4 . 2 ) , δH 4 . 54 d ( 9 . 4 ) ] , two aliphatic protons [δH 2 . 38–2 . 44 , m , 2H] , four aromatic protons giving an A2B2 splitting pattern [δH 7 . 17 d ( 8 . 1 ) , δH 6 . 81 d ( 7 . 8 ) ] , and three aromatic protons giving an ABC splitting pattern [δH 6 . 91 br s , δH 6 . 81 br dd ( 7 . 2 ) , δH 6 . 72 d ( 7 . 2 ) ] . The 13C NMR revealed the occurrence of eighteen carbons resonances , DEPT-90 in conjunction with HMQC supported the presence of seven aromatic methines , including A2B2 [δc 128 . 0 x 2 and δc 115 . 5 x 2] and ABC splitting patterns ( δc 118 . 5 , δc 115 . 7 and δc 114 . 0 ) . Further , we observed two oxygenated [δc 86 . 2 and δc 85 . 2] and two non-oxygenated ( δc 48 . 4 and δc 44 . 0 ) methines as well as two methyl functional groups ( δc 12 . 2 and δc 9 . 7 ) . The remaining five quaternary 13C NMR signals were indicative of aromatic chemical shifts ( δc 157 . 0 , 146 . 0 , 145 . 1 , 136 . 5 and 132 . 8 ) . These NMR data were identical with the previously reported enzymatically synthesized ( ± ) 3-hydroxy-larreatricin [44] . We observed HMBC correlations from aromatic H-2 ( δH 6 . 91 ) of the tri-substituted phenyl ring to C-7 ( δc 86 . 2 ) of the furan ring . In addition , HMBC correlations from H-2’/H-6’ ( δH 7 . 17 ) of the 1 , 4-di-substituted phenyl ring to C-7’ ( δc 85 . 2 ) of furan ring proved the attachment of two phenyl rings at C-7 and C-7’ of furan ring , respectively ( Fig 3A ) . These assignments were further confirmed by the HMBC correlations of H-7/C-2 and H-7’/ C-2’ , C-6’ . The position of two methyls of furan ring was elucidated using HMBC cross peaks between methine H-7’ ( δH 5 . 38 ) and methyl C-9’ ( δc 9 . 7 ) and between methine H-7 ( δH 4 . 54 ) and methyl C-9 ( δc 12 . 2 ) . Finally , the relative stereochemistry of four stereogenic centers in furan ring was assigned by the 1D nuclear Overhauser effect ( NOE ) experiment ( Fig 3B ) . Irradiation at δH 4 . 54 ( H-7 ) gave enhanced signals at δH 6 . 92 ( H-2 ) , δH 0 . 97 ( H-9 ) and δH 0 . 57 ( H-9’ ) , indicating the spatial proximity of H-2 , H-9 and H-9’ . In addition , irradiation at δH 5 . 38 ( H-7’ ) gave enhanced signal exclusively at δH 7 . 17 ( H-2’/H-6’ ) , the absence of correlations between H-7’ and H-7 clearly indicated the trans configuration of the 2-substituted phenyl ring . Accordingly , the structure of 8 was established as ( 7R , 7’R ) -7 , 7’-bis ( 4’ , 3 , 4-trihydroxyphenyl ) - ( 8R , 8’S ) -8 , 8’-dimethyltetrahydrofuran ( Fig . S31-S38 in S1 Appendix ) , which is a stereoisomer of 7 . Compound 9 was obtained as yellow solid and its molecular formula , C16H12O6 , was deduced by HRMS as well as 1H and 13C NMR analysis . In the 1H NMR ( CDCl3 + CD3OD ) spectrum , a methoxy functionality [δH 3 . 74 s] was observed as well as six aromatic methines including two singlets [δH 6 . 35 s , 6 . 20 , s] and an A2B2 splitting pattern [δH 7 . 93 d ( 8 . 6 ) , 2H; 6 . 88 d ( 8 . 6 ) , 2H] resulting from a 1 , 4-disubstituted phenyl ring . The 13C NMR ( Table 2 ) showed sixteen carbon signals and DEPT-90 in conjunction with HMQC supported the presence of one methoxy ( δc 60 . 1 ) and six aromatic methines of which four [δc 130 . 3 x 2 and 115 . 6 x 2] correlated to two doublet signals giving an A2B2 pattern . In addition , we observed two signals that correlated with two aromatic proton singlets of the tetra-substituted phenyl ring ( δc 98 . 9 and 94 . 1 ) . The remaining nine quaternary 13C NMR signals include a carbonyl ( δc 178 . 8 ) , six aromatic and two olefinic carbons ( δc 163 . 9 , 161 . 5 , 159 . 7 , 157 . 0 , 156 . 5 , 138 . 4 , 121 . 7 and 105 . 2 ) . These NMR data were consistent with a flavonol ring system containing three hydroxyls and one methoxy group . The HMBC cross peaks observed between the aromatic protons in the A-ring with H-8 ( δH 6 . 35 ) , C-7 ( δc 163 . 9 ) , C-8a ( δc 157 . 0 ) , and C-5a ( δc 105 . 2 ) ( Fig 4 ) . Cross peaks were also observed between proton H-5 ( δH 6 . 20 ) , C-6 ( δc 161 . 5 ) , and C-5a ( δc 105 . 2 ) suggesting the attachment of two hydroxyl groups at C-7 ( δc 163 . 9 ) and C-6 ( δc 161 . 5 ) . In addition , these cross peaks indicated an oxygen attachment to C-8a ( δc 157 . 0 ) , signifying the O-1 position of the flavonol C-ring . The flavonol B and C ring connectivity were elucidated using HMBC correlations between protons H-2’/H6’ ( δH 7 . 93 ) and carbons C-2 ( δc 156 . 5 ) , and C-4’ ( δc 159 . 7 ) . The phenolic substitution on ring B was indicated through H-3’/H-5’ ( δH 6 . 88 ) and carbon C-1’ ( δc 121 . 7 ) correlations . Finally , the HMBC cross peak between methoxy protons ( δH 3 . 74 ) and C-3 ( δc 138 . 4 ) indicated that attachment at the C-3 position of the flavonol C-ring ( Fig . S23-S33 in S1 Appendix ) [58] . Therefore , we have precisely determined compound 9 to be 3-methoxy-6 , 7 , 4’-trihydroxyflavonol . We previously developed a high-throughput screening CellTiter-Glo ATP bioluminescence-based assay to assess antiparasitic activity [48] , and used this assay to test compounds 1–9 against the trophozoite stage of E . histolytica , G . lamblia , and N . fowleri . Compounds 1–6 displayed dose response antiparasitic activity against all three pathogens by reducing the culture density by 50% ( EC50 ) compared to untreated trophozoite cultures ( Table 3 ) . Compound 1 proved to be the most potent against both G . lamblia and N . fowleri ( EC50 = 36 μM ) ( Fig 5 ) . However , 1 and 2 display similar EC50 values , and both exhibited only moderate activity against E . histolytica with EC50 values of 103 μM and 171 μM , respectively . Both compounds 1 and 2 were found to be about 1 . 5-fold more active relative to the current standard drug miltefosine ( EC50 = 54 . 5 μM ) against N . fowleri . Compound 3 was more active against G . lamblia ( EC50 = 49 μM ) than E . histolytica ( EC50 = 94 μM ) or N . fowleri ( EC50 = 73 μM ) , whereas compound 4 had similar activity against all three pathogens with EC50 values from 74 μM to 83 μM . Compounds 5 and 6 had comparatively weak activity against the three pathogens . Similarly , 9 displayed modest antiparasitic activity against G . lamblia ( EC50 = 153 μM ) and N . fowleri ( EC50 = 235 μM ) ( Table 3 ) . Larreatricin derivatives and stereoisomers 7 and 8 displayed no antiparasitic activity . To further assess the therapeutic potential of 1 and 2 , which displayed the most potent antiparasitic activity agains N . fowleri , we conducted a cytotoxicity study with human umbilical vein endothelial cells ( HUVEC ) , using the same CellTiter-Glo assay and time course that we used for assessing trophozoite toxicity ( Fig 5B ) . Compounds 1 and 2 inhibit HUVEC cell viability with EC50 values of 86 μM and 59 μM , respectively . Thus , 1 and 2 are correspondingly 2 . 4 fold and 1 . 6 fold less toxic to human cells compared to N . fowleri , which is statistically significant ( P<0 . 0001 ) ( Fig 5C ) . NDGA was previously shown to inhibit cysteine protease in cancer [59] , and recent studies linked the involvement of cysteine protease in the pathogenesis of N . fowleri [45] . Thus , we investigated the effects of compounds 1 and 2 on cysteine protease activity present in total crude lysate of N . fowleri over a concentration range from 1 . 875-to-30 μM . The dose dependent effect varied between 1 and 2 , however , both inhibited the cysteine protease activity by almost 50% at 1 . 875 μM ( Fig 6 ) . This data indicates that the activity of compounds 1 and 2 against whole cell N . fowleri may be due to the modulation of cysteine protease activity present in the trophozoites . Because lignans 1–6 are from the same structural class of compounds we could assess notable structure activity relationships ( SAR ) . For example , 1 and 2 displayed overall more potent activity compared to 3–6 , which may be a result of the more flexible straight chain structure that offers more conformational flexibility compared to 3–6 . In addition , introducing a methoxy group in the 3’-position of 2 appears to be negligible with regard to SAR . Conversely , 3 and 4 only differ by one methoxy group in the 3’ position ( i . e . compound 4 ) , which reduced the antiparasitic activity against G . lamblia by ~2 fold . However , this functional group was dispensable when comparing the activity between E . histolytica and N . fowleri . Similarly , introducing a phenol in the 3’ position as in 6 also results in reduced activity compared to 3 . The most striking SAR is observed by introducing a methoxy group in the seven position such as in 5 , which results in a substantial loss of activity compared to 3: ~3 fold ( E . histolytica ) , 4-fold ( G . lamblia ) , and ~ 2 fold ( N . fowleri ) . Although 1–6 are proposed to be biosynthesized from 7 and 8 [44] and share many of the same structural features , these compounds displayed no antiparasitic activity . To better understand this SAR we compared the calculated LogP values for 1–9 . Compounds 7 and 8 are 10 fold more hydrophilic ( CLogP = 3 . 5 ) compared to 1–6 ( CLogP = 4 . 5 ) . However , the flavonol 9 ( CLogP = 1 . 1 ) is 1 , 000 fold more hydrophilic compared to 7 and 8 . Interestingly , flavonoids are known to actively diffuse through organism membranes via membrane transporters such as the ATP-binding cassette ( ABC ) transporters [60] . Moreover , parasitic protozoa are known to express these ABC transporters and other relevant transporters utilized by flavonoids [61] , which may explain the activity of 9 compared to 7 and 8 . Thus , it is plausible that the difference in hydrophilicity may be a physical property of 7 and 8 preventing diffusion into the parasite trophozoites , explaining their inactivity compared to 1–6 and 9 . Compounds 1 and 2 did not display more potent activity against E . histolytica and G . lamblia compared to metronidazole , but both compounds where 1 . 5 fold more potent against N . fowleri compared to miltefosine , which is used for the treatment of PAM . Therefore , we selected N . fowleri for follow-up studies with compounds 1 and 2 . Interestingly , although NDGA has been shown to be cytotoxic to tumor cells by inducing apoptosis and possess antiviral activity [62 , 63] , it has also been shown to be a neuroprotective agent and protective of human monocytes and other human cells and tissues through its powerful antioxidant activity [62–65] . However , at high doses , NDGA has been shown to display nephrotoxicity and hepatotoxicity [62] . Importantly , our data and the collective literature reports described herein indicate that NDGA and derivatives have some therapeutic potential against N . fowleri . Next , we investigated a potential molecular target of NDGA by review of the literature . A report by Huang et al . showed that the NDGA derivatives significantly inhibited cysteine protease activity [59] . Recent studies have also reported that N . fowleri lysate contains cysteine proteases such as cathepsin B-like protease that are important virulence factors of N . fowleri . Cysteine cathepsins are also critical to invasion , evasion , immunomodulation and are implicated in the attachment mechanism to the host tissue [45 , 51 , 66] . Moreover , cathepsins also potentiate N . fowleri growth [66] . Based on these studies we hypothesized that 1 and 2 may be inhibitors of cysteine protease activity present in N . fowleri . Indeed , our results validated this hypothesis and show that NDGA/derivatives inhibit 50% to 80% of N . fowleri cysteine protease activity between 1 . 875–7 . 5 μM ( Fig 6 ) , which are potencies that are consistent with our antiproliferative and antiparasitic data against N . fowleri ( Table 3 and Fig 5 ) . In conclusion , lignans 1–8 and flavonol 9 represent two well-known classes of plant secondary metabolites [67 , 68] . The well-studied flavonoid class of natural products such as 9 display a broad range of biological activity including antiparasitic activity [68 , 69] . Likewise , lignan natural products have received strong interest and have been intensely studied due to their broad clinically relevant biological activity , including: antioxidant , antiviral , antibacterial , immunosuppressive , anti-inflammatory , and anticancer properties [67 , 70] . Only one previous study reported in 1978 demonstrated that NDGA isolated from L . tridentata had inhibitory effect on the growth of non-pathogenic Entamoeba invadens [71] . Our report , for the first time , demonstrates that lignans isolated from L . tridentata are active against pathogenic E . histolytica and G . lamblia , which directly cause human amebiasis and giardiasis , respectively . Moreover , literature reports of natural products effective against N . fowleri growth have been limited [28 , 72] and our study has identified relatively potent compounds from L . tridentata that have amebicidal activity against N . fowleri , which we show may be due to inhibiting cysteine protease activity present in the lysate of N . fowleri . Therefore , lignan secondary metabolites from the creosote bush represent a class of natural products pharmacophore that can be optimized through medicinal chemistry to translate more effective therapeutic options for amebiasis , giardiasis , and PAM .	Entamoeba histolytica , Giardia lamblia , and Naegleria fowleri pathogens are widespread throughout the world infecting and killing hundreds of thousands of people every year . They are also listed as category B bioterrorism agents by the NIH and the CDC . However , there is a serious unmet need to develop more effective therapies to treat these deadly pathogens . Herein we describe that lignans isolated from the creosote bush , common to the southwestern U . S . A . and throughout Mexico , display relatively potent antiparasitic activity against E . histolytica , G . lamblia , and N . fowleri .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "trophozoites", "parasite", "groups", "medicine", "and", "health", "sciences", "giardia", "enzymes", "enzymology", "parasitic", "diseases", "parasitic", "protozoans", "parasitology", "apicomplexa", "protozoans", "cysteine", "proteases", "naegleria", "fowleri", "giardia", "lamblia", "parasitic", "intestinal", "diseases", "entamoeba", "histolytica", "proteins", "protozoan", "infections", "primary", "amoebic", "meningoencephalitis", "biochemistry", "biology", "and", "life", "sciences", "proteases", "organisms" ]	2017	Larrea tridentata: A novel source for anti-parasitic agents active against Entamoeba histolytica, Giardia lamblia and Naegleria fowleri
Clonorchiasis , one of the most important food-borne trematodiases , affects more than 12 million people in the People’s Republic of China ( P . R . China ) . Spatially explicit risk estimates of Clonorchis sinensis infection are needed in order to target control interventions . Georeferenced survey data pertaining to infection prevalence of C . sinensis in P . R . China from 2000 onwards were obtained via a systematic review in PubMed , ISI Web of Science , Chinese National Knowledge Internet , and Wanfang Data from January 1 , 2000 until January 10 , 2016 , with no restriction of language or study design . Additional disease data were provided by the National Institute of Parasitic Diseases , Chinese Center for Diseases Control and Prevention in Shanghai . Environmental and socioeconomic proxies were extracted from remote-sensing and other data sources . Bayesian variable selection was carried out to identify the most important predictors of C . sinensis risk . Geostatistical models were applied to quantify the association between infection risk and the predictors of the disease , and to predict the risk of infection across P . R . China at high spatial resolution ( over a grid with grid cell size of 5×5 km ) . We obtained clonorchiasis survey data at 633 unique locations in P . R . China . We observed that the risk of C . sinensis infection increased over time , particularly from 2005 onwards . We estimate that around 14 . 8 million ( 95% Bayesian credible interval 13 . 8–15 . 8 million ) people in P . R . China were infected with C . sinensis in 2010 . Highly endemic areas ( ≥ 20% ) were concentrated in southern and northeastern parts of the country . The provinces with the highest risk of infection and the largest number of infected people were Guangdong , Guangxi , and Heilongjiang . Our results provide spatially relevant information for guiding clonorchiasis control interventions in P . R . China . The trend toward higher risk of C . sinensis infection in the recent past urges the Chinese government to pay more attention to the public health importance of clonorchiasis and to target interventions to high-risk areas . Clonorchiasis is an important food-borne trematodiases in Asia , caused by chronic infection with Clonorchis sinensis [1 , 2] . Symptoms of clonorchiasis are related to worm burden; ranging from no or mild non-specific symptoms to liver and biliary disorders [3 , 4] . C . sinensis is classified as a carcinogen [5] , as infection increases the risk of cholangiocarcinoma [6] . Conservative estimates suggest that around 15 million people were infected with C . sinensis in 2004 , over 85% of whom were concentrated in the People’s Republic of China ( P . R . China ) [6–8] . It has also been estimated that , in 2005 , clonorchiasis caused a disease burden of 275 , 000 disability-adjusted life years ( DALYs ) , though light and moderate infections were excluded from the calculation [9] . Therefore , two national surveys have been conducted for clonorchiasis in P . R . China; the first national survey was done in 1988–1992 and the second national survey in 2001–2004 . Of note , the two surveys used an insensitive diagnostic approach with only one stool sample subjected to a single Kato-Katz thick smear . The first survey covered 30 provinces/autonomous regions/municipalities ( P/A/M ) with around 1 . 5 million people screened , and found an overall prevalence of 0 . 37% [10] . Data from the second survey , which took place in 31 P/A/M and screened around 350 , 000 people , showed an overall prevalence of 0 . 58% [7] . Another dataset in the second national survey is a survey pertaining to clonorchiasis conducted in 27 endemic P/A/M using triplicate Kato-Katz thick smears from single stool samples . The overall prevalence was 2 . 4% , corresponding to 12 . 5 million infected people [8] . Two main endemic settings were identified; the provinces of Guangdong and Guangxi in the south and the provinces of Heilongjiang and Jilin in the north-east [1 , 2 , 6] . In the latter setting , the prevalence was especially high in Korean ( minority ) communities . In general , males showed higher infection prevalence than females and the prevalence increased with age [6 , 8] . The life cycle of C . sinensis involves specific snails as first intermediate hosts , freshwater fish or shrimp as the second intermediate host , and humans or other piscivorous mammals as definitive hosts , who become infected through consumption of raw or insufficiently cooked infected fish [1 , 2 , 11 , 12] . Behavioral , environmental , and socioeconomic factors that influence the transmission of C . sinensis or the distribution of the intermediate hosts affect the endemicity of clonorchiasis . For example , temperature , rainfall , land cover/usage , and climate change that affect the activities and survival of intermediate hosts , are considered as potential risk factors [13 , 14] . Socioeconomic factors and consumption of raw freshwater fish are particularly important in understanding the epidemiology of clonorchiasis [15] . Consumption of raw fish dishes is a deeply rooted cultural practice in some areas of P . R . China , while in other areas it has become popular in recent years , partially explained by perceptions that these dishes are delicious or highly nutritious [1 , 2 , 16 , 17] . Treatment with praziquantel is one of the most important measures for the management of clonorchiasis , provided to infected individuals or entire at-risk groups through preventive chemotherapy [18 , 19] . Furthermore , information , education , and communication ( IEC ) , combined with preventive chemotherapy , is suggested for maintaining control sustainability [20] . Elimination of raw or insufficiently cooked fish or shrimp is an effective way for prevention of infection , but this strategy is difficult to implement due to deeply rooted traditions and perceptions [1] . Environmental modification is an additional way of controlling clonorchiasis , such as by removing unimproved lavatories built adjacent to fish ponds in endemic areas , thus preventing water contamination by feces [1 , 21] . Maps displaying where a specific disease occurs are useful to guide prevention and control interventions . To our knowledge , only a province-level prevalence map of C . sinensis infection is available for P . R . China , while high-resolution , model-based risk estimates based on up-to-date survey data are currently lacking [1] . Bayesian geostatistical modeling is a rigorous inferential approach to produce risk maps . The utility of this method has been demonstrated for a host of neglected tropical diseases , such as leishmaniasis , lymphatic filariasis , schistosomiasis , soil-transmitted helminthiasis , and trachoma [22–28] . The approach relies on the qualification of the association between disease risk at observed locations and potential risk factors ( e . g . , environmental and socioeconomic factors ) , thus predicting infection risk in areas without observed data [28] . Random effects are usually introduced to the regression equation to capture the spatial correlation between locations via a spatially structured Gaussian process [26] . Here , we compiled available survey data on clonorchiasis in P . R . China , identified important climatic , environmental , and socioeconomic determinants , and developed Bayesian geostatistical models to estimate the risk of C . sinensis infection at high spatial resolution throughout the country . This work is based on clonorchiasis survey data extracted from the peer-reviewed literature and national surveys in P . R . China . All data were aggregated and do not contain any information at individual or household levels . Hence , there are no specific ethical issues that warranted attention . A systematic review was undertaken in PubMed , ISI Web of Science , China National Knowledge Internet ( CNKI ) , and Wanfang Data from January 1 , 2000 until January 10 , 2016 to identify studies reporting community , village , town , and county-level prevalence data of clonorchiasis in P . R . China . The search terms were “clonorchi” ( OR “liver fluke” ) AND “China” for Pubmed and ISI Web of Science , and “huazhigaoxichong” ( OR “ganxichong” ) for CNKI and Wanfang . Government reports and other grey literature ( e . g . , MSc and PhD theses , working reports from research groups ) were also considered . There were no restrictions on language or study design . County-level data on clonorchiasis collected in 27 endemic P/A/M in the second national survey were provided by the National Institute of Parasitic Diseases , Chinese Center for Disease Control and Prevention ( NIPD , China CDC; Shanghai , P . R . China ) . Titles and abstracts of articles were screened to identify potentially relevant publications . Full text articles were obtained from seemingly relevant pieces that were screened for C . sinensis infection prevalence data . Data were excluded if they stemmed from school-based surveys , hospital-based surveys , case-control studies , clinical trials , drug efficacy studies , or intervention studies ( except for baseline or control group data ) . Studies on clearly defined populations ( e . g . , travellers , military personnel , expatriates , nomads , or displaced or migrating populations ) that were not representative of the general population were also excluded . We further excluded data based on direct smear or serum diagnostics due to the known low sensitivity or the inability to differentiate between past and active infection , respectively . All included data were georeferenced and entered into the open-access Global Neglected Tropical Diseases ( GNTDs ) database [29] . Environmental , socioeconomic , and demographic data were obtained from different accessible data sources ( Table 1 ) . The data were extracted at the survey locations and at the centroids of a prediction grid with grid cells of 5×5 km spatial resolution . Land cover data were re-grouped to the following five categories: ( i ) forests , ( ii ) scrublands and grass , ( iii ) croplands , ( iv ) urban , and ( v ) wet areas . They were summarized at each location ( of the survey or grid cell ) by the most frequent category over the period 2001–2004 for each pixel of the prediction grid . Land surface temperature ( LST ) and normalized difference vegetation index ( NDVI ) were averaged annually . We used human influence index ( HII ) , urban extents , and gross domestic product ( GDP ) per capita as socioeconomic proxies . The latter was obtained from the P . R . China yearbook full-text database at county-level for the year 2008 and georeferenced for the purpose of our study . Details about data processing are provided in Lai et al . [26] . We georeferenced surveys reporting aggregated data at county level by the county centroid and linked them to the average values of our covariates within the specific county . The mean size of the corresponding counties was around 2 , 000 km2 . We grouped survey years into two categories ( before 2005 and from 2005 onwards ) . We selected 2005 as the cutoff year because after the second national survey on important parasitic diseases in 2001–2004 , the Chinese government set specific disease control targets and launched a series of control strategies [7 , 30] . We standardized continuous variables to mean zero and standard deviation one ( SD = 1 ) . We calculated Pearson’s correlation between continuous variables and dropped one variable among pairs with correlation coefficient greater than 0 . 8 to avoid collinearity , which can lead to wrong parameter estimation [31] . Researchers have suggested different correlation thresholds of collinearity ranging from 0 . 4 to 0 . 85 [31] . To test the sensitivity of our threshold , we also considered two other thresholds , i . e . , 0 . 5 and 0 . 7 . Three sets of variables were obtained corresponding to the three thresholds and were used separately in the variable selection procedure . Furthermore , continuous variables were converted to two- or three-level categorical ones according to preliminary , exploratory , graphical analysis . We carried out Bayesian variable selection to identify the most important predictors of the disease risk . In particular , we assumed that the number of positive individuals Yi arises from a binominal distribution Yi∼Bn ( pi , ni ) , where ni and pi are the number of individuals examined and the probability of infection at location i ( i = 1 , 2 , … , L ) , respectively . We modeled the covariates on the logit scale , that is logit ( pi ) =β0+∑k=1βk×Xi ( k ) , where βk is the regression coefficient of the kth covariate X ( k ) . For a covariate in categorical form , βk is a vector of coefficients {βkl} , l = 1 , … , Mk , where Mk is the number of categories , otherwise it has a single element βk0 . We followed a stochastic search variable selection approach [32] , and for each predictor X ( k ) we introduced a categorical indicator parameter Ik which takes values j , j = 0 , 1 , 2 with probabilities πj such that π0 + π1 + π2 = 1 . Ik = 0 indicates exclusion of the predictor from the model , Ik = 1 indicates inclusion of X ( k ) in linear form and Ik = 2 suggests inclusion in categorical form . We adopted a mixture of Normal prior distribution for the parameters βk0 , known as spike and slab prior , proposing a non-informative prior βk0∼N ( 0 , σB2 ) with probability π1 in case X ( k ) is included in the model ( i . e . , Ik = 1 ) in linear form ( slab ) and an informative prior βk0∼N ( 0 , ϑ0σB2 ) with probability ( 1 − π1 ) , shrinking βk0 to zero ( spike ) if the linear form is excluded from the model . ϑ0 is a constant , fixed to a small value i . e . , ϑ0 = 0 . 00025 forcing the variance to be close to zero . In a formal way the above prior is written βk0∼δ1 ( Ik ) N ( 0 , σB2 ) + ( 1−δ1 ( Ik ) ) N ( 0 , ϑ0σB2 ) where δj ( Ik ) is the Dirac function taking the value 1 if Ik = j and zero otherwise . Similarly , for the coefficients {βkl} , l = 1 , … , Mk corresponding to the categorical form of X ( k ) with Mk categories , we assume that βkl∼δ2 ( Ik ) N ( 0 , σBl2 ) + ( 1−δ2 ( Ik ) ) N ( 0 , ϑ0σBl2 ) . For the inclusion/exclusion probabilities πj , we adopt a non-informative Dirichlet prior distribution , i . e . ( π0 , π1 , π2 ) T∼Dirichlet ( 3 , a ) , a = ( 1 , 1 , 1 ) T . We also used non-informative inverse gamma prior distributions , IG ( 2 . 01 , 1 . 01 ) for the variance hyperparameters σB2 and σBl2 , l=1 , … , Mk . We considered as important , those predictors with posterior inclusion probabilities of πj greater than 50% . The above procedure fits all models generated by all combinations of our potential predictors and selects as important those predictors which are included in more than 50% of the models . Bayesian geostatistical logistic regression models were fitted on C . sinensis survey data to obtain spatially explicit estimates of the infection risk . The predictors selected from the variable selection procedure were included in the model . The model extended the previous formulation by including location random effects on the logit scale , that is logit ( pi ) =β0+∑k=1βk×Xi ( k ) +εi , where covariate X ( k ) are the predictors ( with functional forms ) that have been identified as important in the variable selection procedure . We assumed that location-specific random effects ε = ( ε1 , … , εL ) T followed a multivariate normal prior distribution ε∼MVN ( 0 , Σ ) , with exponential correlation function Σij=σsp2exp⁡ ( −ρdij ) , where dij is the Euclidean distance between locations , and ρ is the parameter corresponding to the correlation decay . We also considered non-informative normal prior distributions for the regression coefficient βkl , l=0 , 1 , … , Mk , that is βkl∼N ( 0 , 102 ) , an inverse gamma prior distribution for the spatial variance σsp2∼IG ( 2 . 01 , 1 . 01 ) , and a gamma prior for the correlation decay ρ∼G ( 0 . 01 , 0 . 01 ) . We estimated the spatial range as the minimum distance with spatial correlation less than 0 . 1 equal to −log ( 0 . 1 ) /ρ . We formulated the model in a Bayesian framework and applied Markov chain Monte Carlo ( MCMC ) simulation to estimate the model parameters in Winbugs version 1 . 4 ( Imperial College London and Medical Research Council; London , United Kingdom ) [33] . We assessed convergence of sampling chains using the Brooks-Gelman-Rubin diagnostic [34] . We fitted the model on a random subset of 80% of survey locations and used the remaining 20% for model validation . Mean error and the percentage of observations covered by 95% Bayesian credible intervals ( BCIs ) of posterior predicted prevalence were calculated to access the model performance . Bayesian kriging was employed to predict the C . sinensis infection risk at the centroids of a prediction grid over P . R . China with grid cells of 5 × 5 km spatial resolution [35] . This spatial resolution is often used for estimation of disease risk across large regions as it is a good trade-off between disease control needs and computational burden . Furthermore , predictions become unreliable when the grid cells have higher resolution than that of the predictors used in the model . Population-adjusted prevalence ( median and 95% BCI ) for each province was calculated using samples of size 500 from the predictive posterior distribution estimated over the gridded surface . These samples available for each grid cell were converted to samples from the predictive distribution of the population-adjusted prevalence for each province by multiplying them with the gridded population data , summing them over the grid cells within each province and divided them by the province population . The samples from the population-adjusted prevalence for each province were summarized by their median and 95% BCI . Our disease data consist of point-referenced ( village- or town-level ) and areal ( county-level ) data . Analyses ignoring the areal data may loss valuable information , especially in regions where point-referenced data is sparse . Here , we assumed a uniform distribution of infection risk within each survey county and treated the areal data as point-referenced data by setting the survey locations as the centroids of the corresponding counties . To assess the effect of this assumption on our estimates , we simulated data over a number of hypothetical survey locations within the counties and compared predictions based on approaches using the county aggregated data together with the data at individual georeferenced survey locations and using the data at individual georeferenced survey locations only ( excluded the county aggregated data ) . The former approach gave substantially better disease risk prediction compared to the later one . The methodology for the simulation study and its results are presented in Supplementary Information S1 Text and S1 Fig , respectively . A data selection flow chart for the systematic review is presented in Fig 1 . We identified 7 , 575 records through the literature search and obtained one additional report provided by NIPD , China CDC ( Shanghai , P . R . China ) . According to our inclusion and exclusion criteria , we obtained 143 records for the final analysis , resulting in 691 surveys for C . sinensis at 633 unique locations published from 2000 onwards . A summary of our survey data , stratified by province , is provided in Table 2 . The geographic distribution of locations and observed C . sinensis prevalence are shown in Fig 2B . We obtained data from all provinces except Inner Mongolia , Ningxia , Qinghai , and Tibet . We collected more than 50 surveys in Guangdong , Guangxi , Hunan , and Jiangsu provinces . Over 45% of surveys were conducted from 2005 onwards . Around 90% of surveys used the Kato-Katz technique for diagnosis , while 0 . 14% of surveys had no information on the diagnostic technique employed . The overall raw prevalence , calculated as the total number of people infected divided by the total number of people examined from all observed surveys , was 9 . 7% . We considered a total of 12 variables ( i . e . , land cover , urban extents , precipitation , GDP per capita , HII , soil moisture , elevation , LST in the daytime , LST at night , NDVI , distance to the nearest open water bodies , and pH in water ) for Bayesian variable selection . Elevation , NDVI , distance to the nearest open water bodies , and land cover were selected for the final geostatistical logistic regression model . The variables that were selected via the Bayesian variable selection method are listed in Supporting Information S1 Table . The list was not affected by the collinearity threshold ( i . e . , 0 . 5 , 0 . 7 , and 0 . 8 ) we have considered . The parameter estimates arising from the geostatistical model fit are shown in Table 3 . The infection risk of C . sinensis was higher from 2005 onwards than that before 2005 . Elevation had a negative effect on infection risk . People living at distance between 2 . 5 and 7 . 0 km from the nearest open water bodies had a lower risk compared to those living in close proximity ( <2 . 5 km ) . The risk of C . sinensis infection was lower in areas covered by forest , shrub , and grass compared to crop . Furthermore , NDVI was positively correlated with the risk of C . sinensis infection . Model validation indicated that the Bayesian geostatistical logistic regression models were able to correctly estimate ( within a 95% BCI ) 71 . 7% of locations for C . sinensis . The mean error was -0 . 07% , suggesting that our model may slightly over-estimate the infection risk of C . sinensis . Fig 2A shows the model-based predicted risk map of C . sinensis for P . R . China . High prevalence ( ≥20% ) was estimated in some areas of southern and northeastern parts of Guangdong province , southwestern and northern parts of Guangxi province , southwestern part of Hunan province , the western part of bordering region of Heilongjiang and Jilin provinces , and the eastern part of Heilongjiang province . Most regions of northwestern P . R . China and eastern coastal areas had zero to very low prevalence ( <0 . 01% ) . The prediction uncertainty is shown in Fig 2C . Table 4 reports the population-adjusted predicted prevalence and the number of individuals infected with C . sinensis in P . R . China , stratified by province , based on gridded population of 2010 . The overall population-adjusted predicted prevalence of clonorchiasis was 1 . 18% ( 95% BCI: 1 . 10–1 . 25% ) in 2010 , corresponding to 14 . 8 million ( 95% BCI: 13 . 8–15 . 8 million ) infected individuals . The three provinces with the highest infection risk were Heilongjiang ( 7 . 21% , 95% BCI: 5 . 95–8 . 84% ) , Guangdong ( 6 . 96% , 95% BCI: 6 . 62–7 . 27% ) , and Guangxi ( 5 . 52% , 95% BCI: 4 . 97–6 . 06% ) . Provinces with very low risk estimates ( median predicted prevalence < 0 . 01% ) were Gansu , Ningxia , Qinghai , Shanghai , Shanxi , Tibet , and Yunnan . Guangdong , Heilongjiang , and Guangxi were the top three provinces with the highest number of people infected: 6 . 34 million ( 95% BCI: 6 . 03–6 . 62 million ) , 3 . 05 million ( 2 . 52–3 . 74 million ) , and 2 . 08 million ( 1 . 87–2 . 28 million ) , respectively . To our knowledge , we present the first model-based , high-resolution estimates of C . sinensis infection risk in P . R . China . Risk maps were produced through Bayesian geostatistical modeling of clonorchiasis survey data from 2000 onwards , readily adjusting for environmental/climatic predictors . Our methodology is based on a rigorous approach for spatially explicit estimation of neglected tropical disease risk [27] . Surveys pertaining to prevalence of C . sinensis in P . R . China were obtained through a systematic review in both Chinese and worldwide scientific databases to obtain published work from 2000 onwards . Additional data were provided by the NIPD , China CDC . We estimated that 14 . 8 million ( 95% BCI: 13 . 8–15 . 8 million; 1 . 18% ) people in P . R . China were infected with C . sinensis in 2010 , which is almost 20% higher than the previous estimates of 12 . 5 million people for the year 2004 , based on empirical analysis of data from a large survey of clonorchiasis conducted from 2002–2004 in 27 endemic P/A/M . The mean error for the model validation was slightly smaller than zero , suggesting that our model might somewhat over-estimate the true prevalence of clonorchiasis . The overall raw prevalence of the observed data was 9 . 7% . This can be an over-estimation of the overall prevalence as many surveys were likely to have been conducted in places with relatively high infection risk ( preferential sampling ) . Our population-adjusted , model-based estimates was much lower ( 1 . 18% , 95% BCI: 1 . 10–1 . 25% ) and it should reflect the actual situation because it takes into account the distribution of the population and of the disease risk across the country . Indeed , geostatistical models get their predictive strength from regions with large amount of data that allow more accurate estimation of the relation between the disease risk and its predictors , therefore they are the most powerful statistical tools for predicting the disease risk in areas with sparse data . Still , the estimates in regions with scarce data should be interpreted with caution . However , even though our data did not include surveys from four provinces ( Inner Mongolia , Ningxia , Qinghai , and Tibet ) , our model obtained low or zero prevalence estimates which are consistent with data summaries of the second national survey aggregated at provincial level for these four provinces [7] . On the other hand , our model may overestimate the overall infection risk for Heilongjiang province , as the high risk areas in the southeastern and southwestern parts of the province may influence the prediction in the northern part , where no observed data were available . We found an increase of infection risk of C . sinensis for the period from 2005 onwards , which may be due to several reasons , including higher consumption of raw fish , lack of self-protection awareness of food hygiene , low health education , and rapid growth of aquaculture [13 , 36] . Consumption of raw freshwater fish is related to C . sinensis infection risk [15 , 37] , however , such information is unavailable for P . R . China . Elevation was one of the most important predictors in our model . Different elevation levels correspond to different environmental/climatic conditions that can influence the distribution of intermediate host snails . Our results show a positive association of NDVI and the prevalence of C . sinensis . We found that distance to the nearest water bodies was significantly related to infection risk . Traditionally , areas adjacent to water bodies were reported to have a higher prevalence of C . sinensis , however , due to improvement of trade and transportation channels , this situation may be changing , which may explain our result showing a non-linear relationship between distance to nearest water bodies and infection risk [2 , 13] . Furthermore , our analysis supports earlier observations , suggesting an association between land cover type and infection risk [13 , 14] . Interestingly , the risk of infection with other neglected tropical diseases , such as soil-transmitted helminthiasis and schistosomiasis , has declined in P . R . China over the past 10–15 years due to socioeconomic development and large-scale interventions [38] . However , clonorchiasis , the major food-borne trematodiases in P . R . China , shows an increased risk in recent years , which indicates the Chinese government needs to pay more attention to this disease . Several areas with high infection risk in P . R . China are indicated ( Supporting Information S2 Fig ) , where control strategies should be focused . The recommended treatment guidelines for clonorchiasis of the WHO advocate praziquantel administration for all residents every year in high endemic areas ( prevalence ≥20% ) and for all residents every two years or individuals regularly eating raw fish every year in moderate endemic areas ( prevalence <20% ) [19] . As re-infection or super-infection is common in heavy endemic areas , repeated preventive chemotherapy is necessary to interrupt transmission [18] . On the other hand , to maintain control sustainability , a comprehensive control strategy must be implemented , including IEC , preventive chemotherapy , and improvement of sanitation [20 , 21] . Through IEC , residents may conscientiously reduce or stop consumption of raw fish . Furthermore , by removing unimproved latrines around fish ponds , the likelihood of fish becoming infected with cercariae declines [39] . A successful example of comprehensive control strategies is Shangdong province , where clonorchiasis was endemic , but after rigorous implementation of comprehensive control programs for more than 10 years , the disease has been well controlled [40] . The Chinese Ministry of Health set a goal to halve the prevalence of clonorchiasis ( compared to that observed in the second national survey in 2001–2004 ) in highly endemic areas by 2015 using integrated control measures [30] . In practice , control measures are carried out in endemic villages or counties with available survey data . However , large-scale control activities are lacking in most endemic provinces , as control plans are difficult to make when the epidemiology is only known at provincial level [41] . Our high-resolution infection risk estimates provide important information for targeted control . Our analysis is based on historical survey data compiled from studies that may differ in study design , diagnostic methods and distribution of age groups . As more than 90% of surveys applied Kato-Katz as diagnostic method , we assumed similar diagnostic sensitivity across all surveys . However , the sensitivity may vary in space as a function of infection intensity . Most of the survey data are aggregated over age groups , thus we could not obtain age-specific risk estimates . Moreover , bias might occur when age distribution in survey population differ across locations as different age group may have different infection risk . In conclusion , we present the first model-based , high-resolution risk estimates of C . sinensis infection in P . R . China , and identified areas of high priority for control . Our findings show an increased risk from 2005 onwards , suggesting that the government should put more efforts on control activities of clonorchiasis in P . R . China .	Clonorchiasis is an important food-borne trematodiases and it has been estimated that more than 12 million people in China are affected . Precise information on where the disease occurs can help to identify priority areas for where control interventions should be implemented . We collected data from recent surveys on clonorchiasis and applied Bayesian geostatistical models to produce model-based , high-resolution risk maps for clonorchiasis in China . We found an increasing trend of infection risk from 2005 onwards . We estimated that approximately 14 . 8 million people in China were infected with Clonorchis sinensis in 2010 . Areas where the high prevalence of C . sinensis was predicted were concentrated in the provinces of Guangdong , Guangxi , and Heilongjiang . Our results suggest that the Chinese government should pay more attention on the public health importance of clonorchiasis and that specific control efforts should be implemented in high-risk areas .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "invertebrates", "medicine", "and", "health", "sciences", "helminths", "china", "tropical", "diseases", "geographical", "locations", "vertebrates", "parasitic", "diseases", "animals", "simulation", "and", "modeling", "trematodes", "freshwater", "fish", "clonorchis", "sinensis", "foodborne", "trematodiases", "probability", "distribution", "mathematics", "neglected", "tropical", "diseases", "infectious", "disease", "control", "research", "and", "analysis", "methods", "infectious", "diseases", "fishes", "flatworms", "clonorchiasis", "research", "assessment", "probability", "theory", "people", "and", "places", "helminth", "infections", "asia", "clonorchis", "systematic", "reviews", "biology", "and", "life", "sciences", "physical", "sciences", "organisms" ]	2017	Risk mapping of clonorchiasis in the People’s Republic of China: A systematic review and Bayesian geostatistical analysis
The adhesion of Plasmodium falciparum-infected erythrocytes ( IRBC ) to receptors on different host cells plays a divergent yet critical role in determining the progression and outcome of the infection . Based on our ex vivo studies with clinical parasite isolates from adult Thai patients , we have previously proposed a paradigm for IRBC cytoadherence under physiological shear stress that consists of a recruitment cascade mediated largely by P-selectin , ICAM-1 and CD36 on primary human dermal microvascular endothelium ( HDMEC ) . In addition , we detected post-adhesion signaling events involving Src family kinases and the adaptor protein p130CAS in endothelial cells that lead to CD36 clustering and cytoskeletal rearrangement which enhance the magnitude of the adhesive strength , allowing adherent IRBC to withstand shear stress of up to 20 dynes/cm2 . In this study , we addressed whether CD36 supports IRBC adhesion as part of an assembly of membrane receptors . Using a combination of flow chamber assay , atomic force and confocal microscopy , we showed for the first time by loss- and gain-of function assays that in the resting state , the integrin α5β1 does not support adhesive interactions between IRBC and HDMEC . Upon IRBC adhesion to CD36 , the integrin is recruited either passively as part of a molecular complex with CD36 , or actively to the site of IRBC attachment through phosphorylation of Src family kinases , a process that is Ca2+-dependent . Clustering of β1 integrin is associated with an increase in IRBC recruitment as well as in adhesive strength after attachment ( ∼40% in both cases ) . The adhesion of IRBC to a multimolecular complex on the surface of endothelial cells could be of critical importance in enabling adherent IRBC to withstand the high shear stress in the microcirculations . Targeting integrins may provide a novel approach to decrease IRBC cytoadherence to microvascular endothelium . Cell-cell interaction in the microvasculature is a complex process that involves multiple ligands and receptors that mediate different types of adhesive behavior in a sequential manner . The adhesive cascade is best studied in leukocyte-endothelial cell interactions that includes leukocyte tethering , crawling , rolling and adhesion on endothelium , followed by transmigration of leukocytes into extravascular tissues [1] . The strength of the interaction between ligands and receptors at each stage of the cascade can be qualitatively or quantitatively regulated by molecular events such as conformational change of the adhesion molecules , and/or intracellular signaling in both leukocytes and endothelial cells leading to modification of biological processes such as calcium flux , protein phosphorylation , cytoskeletal rearrangement and cell migration [2] . The adhesive interaction between Plasmodium falciparum-infected erythrocytes ( IRBC ) with vascular endothelium , the most consistent pathological finding in the human infection , is also governed by similar molecular mechanisms . Based on our findings with clinical parasite isolates obtained directly from infected patients , we have previously proposed a paradigm for IRBC cytoadherence under flow conditions that consists of a recruitment component that involves tethering , rolling and adhesion of IRBC that is mediated largely by P-selectin , ICAM-1 and CD36 respectively on primary human dermal microvascular endothelium ( HDMEC ) [3] , [4] . In addition , we detected post-adhesion signaling events involving Src family kinases and the adaptor protein p130CAS in endothelial cells that lead to receptor clustering and cytoskeletal rearrangement which in turn enhance the magnitude of the adhesive strength , allowing adherent IRBC to withstand higher shear stress [5] , [6] . Intracellular signaling in endothelial cell lines has also been shown for parasite lines and clones selected to adhere to ICAM-1 [7] . Together , these findings underscore the complexity of the cytoadherence process in the vasculature that might not be appreciated when studied as isolated ligand-receptor interactions on recombinant proteins or transfectants [8] , [9] , [10] . The class B scavenger molecule CD36 has a unique relationship with IRBC . CD36 is highly expressed in vital organs such as the lung , liver and kidney [11] , and scantily in the brain [12] . CD36-mediated cytoadherence may contribute to dysfunction in these organs by impairing microcirculatory blood flow , although a direct link between IRBC adhesion to CD36 or any other receptor on primary endothelium under flow conditions has not been established . A recent report on P . berghei infection in mice suggests that CD36-dependent tissue sequestration may also promote parasite growth and other parasite survival benefits [13] . This long suspected association makes teleological sense as cytoadherence has likely evolved as a mechanism for host evasion . On the other hand , platelets have been shown to have a direct cytotoxic effect on IRBC adherent on CD36 through the release of platelet factor 4 ( PF4 ) that binds to the Duffy blood group antigen on erythrocytes[14] . PF4 acts by its lytic activity on the food vacuole of the intraerythrocytic parasite while sparing the red cell membrane [15] . Collectively , these findings indicate that IRBC can interact with CD36 on different host cells with diverse biological effects . An important question regarding IRBC–host cell interaction that has not been addressed is whether CD36 supports IRBC adhesion alone , or as part of an assembly of membrane receptors as it does in response to fibrillar α-amyloid [16] , [17] , [18] . The engagement and focal aggregation of the receptors following initial IRBC adhesion may lead to the formation of a functional complex which increases the strength of the adhesive interactions critical for determining adhesion in the microvasculature in vivo . IRBC could bind directly to multiple host surface molecules through different domains on the cytoadherent parasite ligand Plasmodium falciparum erythrocyte membrane protein 1 ( PfEMP1 ) [19] . Alternatively , the involvement of other membrane receptors may occur downstream of CD36 ligation by being recruited to the site of adhesion where cross-talk between signaling molecules is facilitated [20] . In either scenario , integrins , a family of heterodimeric , non-covalently bound cell surface receptors , are likely candidate molecules to be involved , as they promote adhesion to other cells and matrix proteins , and are often associated physically and functionally with CD36 [21] . Indeed , CD36 is known to guide integrins into signaling rafts , and in so doing , may regulate integrin function . IRBC may bind to integrins directly through the tri-amino acid motif arginine-glycine-aspartic acid ( RGD ) present on PfEMP1 [22] , [23] , [24] , or interact with integrins through binding to thrombospondin-1 ( TSP-1 ) [25] . In support of a role for integrins in cytoadherence , an anti-αv antibody has been reported to partially inhibit the adhesion of a laboratory-adapted parasite line to HDMEC under flow conditions in vitro [26] . There is also evidence that IRBC and apoptotic leukocytes could downregulate dendritic cell function through CD36 or αvβ3 [27] . As over 250 anti-integrin drugs have now entered clinical trials [28] , an understanding of the role of integrins in cytoadherence may lead to the novel application of some of the available agents as adjunctive therapy in the treatment of severe falciparum malaria . In this study , we used a combination of flow chamber assay , atomic force microscopy ( AFM ) and confocal microscopy to investigate the role of integrins in cytoadherence . Our results reveal a novel and robust co-operative association of CD36 with the integrin α5β1 in microvascular endothelium as a result of Src family kinase signaling . The formation of an IRBC-endothelial cell synapse consisting of multiple adhesion molecules contributes significantly to overall IRBC recruitment and adhesive strength . Identification of the dynamic interactions among all binding partners for IRBC on primary endothelial cells will provide a theoretical basis for the rational design of anti-adhesive therapy . A number of integrins are expressed on microvascular endothelial cells [29] , including αvβ3 , αvβ1 , α2β1 , α3β1 and α5β1 that recognize RGD motifs on their ligands [30] . In this study , we focused on αv , α5 and β1 that have been shown to be associated with CD36 on HDMEC by both an antibody array and by co-immunoprecipitation [31] . All 4 proteins were shown to be surface expressed by flow cytometry ( Figure S1A ) and by confocal immunofluorescence microscopy ( Figure S1B ) . The latter findings indicate that the integrin molecules are expressed on the luminal surface of endothelial monolayers and are therefore available for interaction with IRBC . To begin to assess a role for integrins in IRBC adhesion , HDMEC monolayers were pretreated with 5 to 50 µM of the integrin antagonist RGD ( H-Gly-Arg-Gly-Asp-Ser-Pro-OH ) for 30 min at 37°C before the infusion of IRBC in a flow chamber assay . The IRBC used was a lab adapted clone 7G8 that binds to both CD36 and ICAM-1 . The results indicate that RGD inhibited IRBC adhesion in a dose-dependent manner ( Figure 1A ) , while the control peptide RAD ( H-Gly-Arg-Ala-Asp-Ser-Pro-OH ) had no effect . The decrease in the total number of adherent IRBC on RGD treated monolayers was associated with an increase in the number of rolling cells that did not adhere ( control 52±8 vs RGD 95±12 cells/7 min , n = 6 , p = 0 . 0034 ) , suggesting that integrins are critical for firm adhesion to occur . It has been previously reported that an anti-αv antibody inhibited IRBC adhesion to HDMEC [26] . The integrin was assumed to be αvβ3 by the authors and in subsequent publications [24] , [32] , [33] , but the finding has never been confirmed . To determine if αvβ3 could mediate IRBC adhesion , we pre-incubated HDMEC with the cylic peptide cRGDfV ( cyclo ( -Arg-Gly-Asp-D-Phe-Val ) ) that is a specific inhibitor for the αvβ3 and αvβ5 integrins [34] . In contrast to RGD , the cyclic peptide at 20 µM had no effect on cytoadherence of 7G8 parasites ( Figure 1B ) . To confirm that the ability of IRBC to interact with integrins is not acquired as a result of prolonged passage in the laboratory , flow chamber experiments were performed with clinical parasite isolates . As in the case with 7G8 , adhesion of cryopreserved IRBC from acutely infected patients , cultured for 24 to 36 h , was inhibited by RGD at 20 µM ( Figure 1C ) . Moreover , cytoadherence of the clinical isolates was reduced by the inhibitory anti-β1 mAb TDM29 ( 10 µg/ml ) but not the activating anti-β1 mAb TS2/16 ( 10 µg/ml ) ( Figure 1D ) . Inhibition of adhesion was also seen with the anti-α5 mAb JBS5 ( 10 µg/ml ) ( Figure 1E ) , suggesting that binding of IRBC is to an epitope involving both subunits of the heterodimer . Consistent with the results obtained with cRGDfV , an inhibitory anti-αvβ3 mAb 23C6 ( 10 µg/ml ) did not affect cytoadherence ( Figure 1F ) . We next used atomic force microscopy ( AFM ) to directly measure the adhesive force between IRBC and HDMEC at the single cell level as previously described [6] . An IRBC attached to the cantilever by dopamine hydrochloride was brought into contact with an endothelial cell using a contact force of 150 pN at a steady velocity of 1 . 5 µm sec−1 . These parameters were chosen because they approximate the marginating force experienced by a more rigid IRBC as it is pushed from centreline blood flow towards the vessel wall by NRBC . Upon contact , the IRBC and HDMEC were allowed to remain in contact for 5 min before it was withdrawn at the same velocity . Adhesive strength was measured as the force required for detachment of the IRBC from an HDMEC ( Figure 2A ) . In a previous study , we found that the detachment force between IRBC and HDMEC increased rapidly with the duration of contact , so that the magnitude of the force could be 5 to 6 fold higher by the end of 5 min [6] . The resulting increase in adhesion strength enabled adherent IRBC to withstand shear stress of up to 20 dynes/cm2 . To determine if integrins contribute to this enhancement of adhesive force following contact , force measurements were performed on HDMEC monolayers pre-treated with RAD or RGD using 7G8 parasites . The results indicate that RGD ( Figure 2B ) and the inhibitory anti-β1 mAb TDM29 ( Figure 2C ) both inhibited adhesive force of IRBC by approximately 40% while RAD or the activating anti-β1 mAb TS2/16 had no effect . As in the case of adhesion in the flow chamber assay , cRGDfV had no effect on detachment force ( Figure 2D ) . Together , our results indicate a significant role for integrins both in the firm adhesion of IRBC under flow conditions and in the post-contact increase in adhesive strength between IRBC and endothelial cells . As cytoadherence occurs in a proinflammatory environment during acute P . falciparum infection [35] , [36] , the inhibitory effect of RGD on IRBC adhesion in the flow chamber assay was studied on HDMEC monolayers that had been pre-stimulated with 1 ng/ml of TNF-α for 20 to 24 h . The results are shown in Figure 3 . In accordance with what we reported previously [3] , TNF-α stimulation of HDMEC upregulated ICAM-1 expression , and induced VCAM-1 in a small percentage of cells ( Figure 3A ) . The expression of CD36 , α5 and β1 remained essentially unchanged . Stimulation of HDMEC with TNF-α at 1 ng/ml did not lead to an increase in the number of adherent IRBC of the 7G8 parasite line ( Figure 3B ) . IRBC adhesion was reduced by RGD as on unstimulated endothelium , and was inhibited by >90% by the anti-CD36 mAb FA6 . We next tested two of the three clinical isolates that were studied in Figure 1 . IRBC adhesion increased by almost two fold on cytokine-stimulated HDMEC ( Figure 3C ) . Adhesion was inhibited to a similar extent by both RGD and by an anti-ICAM-1 mAb . However , there was no additive effect when monolayers were pre-treated with a combination of RGD and anti-ICAM-1 . The effect of TNF-α stimulation on the strength of IRBC adhesion was investigated by AFM . For the 7G8 parasite line as well as the 2 clinical isolates , there was no difference in the magnitude of the adhesive force generated over 5 minutes on TNF-α-stimulated endothelium compared to unstimulated controls ( Figure 3D ) . As well , pre-incubation with RGD inhibited 5adhesive strength as on undstimulated endothelium . Collectively , the results suggest that integrins contribute to both an increase in IRBC adhesion and adhesive strength on cytokine-stimulated HDMEC . To more specifically assess the role of β1 integrin in the increase in IRBC recruitment and the adhesive strength of the IRBC-integrin interaction , we performed gene knock down of β1 integrin in HDMEC by siRNA . Loss of β1 integrin protein production was confirmed by Western blot ( Figure 4A and B ) . The targeted deletion of β1 integrin did not alter endothelial CD36 or ICAM-1 expression ( data not shown ) . The loss of β1 integrin led to a reduction in IRBC adhesion in the flow chamber assay ( Figure 4C ) as well as the adhesive strength as measured by AFM ( Figure 4D ) . These results confirmed a functional role for β1 integrin in mediating IRBC cytoadherence . Interestingly , when the α5 integrin was similarly knocked down ( Figure 5A and B ) , a reduction in IRBC adhesion in the flow chamber assay was observed ( Figure 5C ) while adhesive strength by AFM was unaffected ( Figure 5D ) . The lack of effect of α5 knock down on adhesive force was consistent with the inability of the mAb JBS5 to inhibit adhesive force ( Figure 5E ) . The inhibitory effect of an antagonist peptide and mAb to the integrin α5β1 suggest that IRBC may be binding directly to the integrin . Indeed , RGD motifs have been demonstrated in several DBL domains of the cytoadherent ligand PfEMP1 in a number of parasite lines [22] , [23] , [24] . These parasite RGD motifs could potentially interact with endothelial integrins and contribute to the overall adhesive force between IRBC and endothelium . To determine if IRBC can bind to α5β1 under flow conditions , IRBC were infused over HMEC-1 monolayers . This endothelial cell line , derived from HDMEC , does not express CD36 [29] , or only at a very low level ( Figure 6A ) . In contrast , the α5 and β1 integrins are highly expressed ( Figure 6A ) . We found that <5 IRBC/mm2 rolled and/or adhered on HMEC-1 for the total duration of a 7-min infusion in the flow chamber assay ( Figure 6B ) . However , when HMEC-1 was tranduced with CD36-GFP , but not GFP alone , there was a dramatic increase in IRBC adhesion that was reduced by pre-treatment of the monolayer with RGD . A similar effect of CD36 on β1 integrin function was detected by AFM , as indicated by an increase in adhesive force on CD36 transduced cells that was partially inhibited by pre-treatment of the monolayer with RGD ( Figure 6C ) . These results suggest that in its resting state , β1 integrin by itself is unable to support IRBC adhesion under flow conditions . The presence of CD36 may lead to proadhesive changes of β1 integrin through changes in i ) integrin surface expression , ii ) conformation or iii ) subcellular localization . To determine the mechanism of CD36-β1 integrin interaction with respect to cytoadherence , we first determined if the presence of CD36 modulated surface expression of β1 integrin using HMEC-1 cells . No increase in β1 integrin expression on HMEC-1 was seen by flow cytometry after the transduction of CD36 ( data not shown ) . The contribution of conformational changes was also unlikely as neither the activating β1 antibody TS2/16 ( Figure 1D ) nor the addition of 1 mM Mn2+ ( data not shown ) , both of which activate β1 integrin [37] , had any effect on IRBC adhesion . The most likely mechanism is integrin clustering . Indeed , clustering of α5β1 in mouse embryonic fibroblasts as a result of lateral diffusion has been demonstrated to increase adhesive strength between the integrin and its natural ligand fibronectin , while clustering of αvβ3 contributes to mechanotransduction [38] . To determine if β1 integrin is recruited upon IRBC adhesion , IRBC purified on a magnetic bead separation column ( Miltenyi Biotec , Auburn , CA ) were allowed to adhere to HDMEC . A labeled anti-β1 antibody was added and endothelial cells were observed by live cell imaging . Figure 7A shows that β1 integrin was recruited to the site of IRBC adhesion within minutes of the adhesion process . Moreover , the recruited β1 integrin forms a cup-shaped protrusion around the IRBC on the cell surface ( Figure 7B ) . Interestingly , β1 integrin clustering was also seen with polystyrene beads coated with an anti-CD36 mAb FA6 ( Figure 7C ) , or the recombinant PpMC-179 peptide representing the minimal binding domain of PfEMP1 ( residues 88–267 of the CIDR domain of the parasite strain Malayan Camp ) for CD36 [39] ( Figure 7D ) . The results suggest that β1 integrin recruitment likely occurred as a result of CD36 ligation , either passively as a member of a molecular complex with CD36 , or as a result of recruitment downstream of CD36 ligation . The recruitment of β1 integrin by CD36 was specific , as anti-CD36 coated beads did not recruit ICAM-1 on HDMEC transduced with GFP-ICAM-1 ( Figure 7E ) . Compared to anti-CD36 coated beads , anti-ICAM-1 coated beads recruited much less β1 on TNF-α-stimulated endothelium ( Figure 7F and 7G ) . The dependence of integrin clustering on CD36 was further demonstrated with HMEC-1 cells . When FA6-coated beads were added to HMEC-1 , no β1 integrin clustering was seen on confocal microscopy ( Figure 8 ) . In contrast , CD36 and integrin clustering at the site of adhesion was clearly observed after HMEC-1 were transduced with CD36-GFP . If α5β1 clustering occurs downstream of CD36 ligation by IRBC , one would expect that the process would be mediated by Src family kinases that are activated by the binding of IRBC to CD36 [5] , [6] . The possibility was investigated with anti-CD36 coated beads using PP1 , a specific Src family kinase inhibitor and its inactive analog PP3 . The requirement for Ca2+ was also investigated by preincubating HDMEC with the intracellular calcium chelator BAPTA-AM . Similar to the effects of the inhibitors on CD36 and actin recruitment [6] , β1 integrin recruitment in response to anti-CD36 coated beads was significantly reduced by these inhibitors ( Figure 9A and B ) . In contrast , neither RGD nor anti-β1 antibody had any effect on the recruitment of β1 integrin or CD36 ( Figure 9C ) , or actin and phosphorylated Src ( Figure 9D ) . Consistent with these observations , neither RGD nor anti-β1 mAb had any effect on the force of detachment of anti-CD36 coated beads as measured by AFM ( data not shown ) . Together , these results indicate that β1 integrin clustering occurs as a downstream event of CD36 ligation and subsequent signaling events , and not through extracellular integrin activation . Lateral diffusion of the β2 integrin LFA-1 , as detected by single molecule tracking , can be induced by phorbol-12-myristate-13-acetate ( PMA ) and low dose cytochalasin D in lymphocytes [40] and monocytes [41] . The resulting integrin clustering leads to an increase in the number of adherent leukocytes on ICAM-1 . To determine if PMA could also affect integrin interaction with IRBC , we performed the flow chamber assay with HMEC-1 monolayers with or without pre-treatment with PMA ( 50 ng/ml ) for 30 min at 37°C . HMEC-1 cells were utilized in these experiments to allow for an assessment of IRBC adhesion directly to integrins in response to PMA in the absence of CD36 . Fewer than 5 IRBC adhered on either untreated or treated monolayers under flow conditions ( Figure 10A ) . However , PMA treatment significantly increased the number of IRBC that adhered in a static binding assay . In these experiments , monolayers were pre-treated with either 50 ng/ml of PMA alone for 30 min at 37°C , or 100 µM RGD or RAD peptide added after 10 min of incubation . At the end of 30 min , 1 ml of a 1% hematocrit suspension of IRBC in RPMI at 5–9% parasitemia was added and allowed to adhere for 20 min . The 35 mm dish with the HMEC-1 monolayer and IRBC was then mounted into the flow chamber system . HBSS was infused at 1 dyne/cm2 , and the number of adherent cells was counted at 30 sec intervals for a total of 4 min and the mean taken . The results suggest that while PMA stimulation did not increase IRBC adhesion under flow conditions , the adhesion of IRBC to PMA-stimulated HMEC-1 was significantly increased when the IRBC were allowed to bind under static conditions . Moreover , the enhancing effect of PMA was abrogated by pre-treatment of the HMEC-1 monolayers with RGD but not RAD . The increase in the number of adherent IRBC on PMA-stimulated monolayers was not associated with an increase in CD36 , β1 integrin or ICAM-1 expression ( Figure 10B ) . The adhesion of P . falciparum-infected red cells to host endothelial cells in vital organs such as the brain and lung plays a fundamental role in the progression and outcome of the infection [42] . In this report , we showed for the first time by loss- and gain-of function assays that the integrin α5β1 may have a significant role in this pathological process on human microvascular endothelium . Our data suggests that in the resting state , α5β1 does not support adhesive interactions between IRBC and HDMEC . Upon IRBC adhesion to CD36 , the integrin is either recruited passively as part of a molecular complex with CD36 , or recruited actively to the site of IRBC attachment on CD36 ligation through phosphorylation of Src family kinases , a process that is Ca2+-dependent . Clustering of b1 integrin is associated with an increase in IRBC recruitment under flow conditions as well as an increase in adhesive strength after attachment on both unstimulated and TNF-α-stimulated endothelium . Conformational change of the integrin does not appear to play a role , as neither an activating antibody nor Mn2+ had any effect on IRBC adhesion . The binding of IRBC to clustered β1 could be inhibited by RGD and inhibitory antibodies to β1 and α5 integrins . As adhesion molecule expression can be affected by cytokines in the microenvironment , we also tested if integrins would have a role in IRBC adhesion to TNF-α-stimulated HDMEC . We found that TNF-α did not affect the surface expression of α5 or β1 , and the RGD peptide inhibited adhesion as on unstimulated endothelium . Moreover , anti-CD36 beads recruited β1 integrin on stimulated endothelium . Anti-ICAM-1 and RGD inhibited IRBC adhesion to a similar degree , but their effects were not additive , suggesting that α5β1 and ICAM-1 have a similar functional role in IRBC adhesion under flow conditions . This finding is consistent with our previous in vitro [3] and in vivo [4] observations that for IRBC that adhere to CD36 , such as most clinical isolates from Thailand , ICAM-1 plays an important accessory role in increasing the number of adherent cells by reducing their rolling velocity . The relative contribution of integrins to IRBC cytoadherence compared to adhesion molecules that are upregulated or induced on stimulated endothelium suggested by our current findings will need to be confirmed with a large number of clinical parasite isolates with different binding phenotypes . An association between integrins and CD36 has been described in several human cell types , including melanoma cells [21] , microglial cells [16] , platelets [43] , and HDMEC [31] . In melanoma cells , the association between CD36 and b1 integrins requires the extracellular domain of the CD36 molecule . The association may occur within raft domains , since ectopic expression of CD36 increases the proportion of β1 integrins found within this fraction . In microglial cells , CD36 , α6β1 and CD47 form a receptor complex for fibrillar β-amyloid , and antagonists specific for each component inhibits phagocytosis of β-amyloid to the same extent , suggesting that each component of the receptor complex is required but not sufficient for uptake of β-amyloid . In both platelets [43] and HDMEC [31] , two distinct pools of CD36 in the cell membrane have been identified . One pool of CD36 is distributed in low-density lipid rafts and co-localizes with the Src family kinases . The other pool is in the high density soluble fraction that also contain β1 integrin , VEGFR-2 , Syk and tetraspanins . Thrombospondin 1 ( TSP-1 ) is also strongly detected in the high-density fraction , and participates in the formation of the VEGFR-2-Syk-CD36 complex that regulates angiogenesis . The 40% reduction of IRBC adhesion and adhesive strength by β1 integrin knock down may be mediated by this high density pool of CD36 in the cell membrane . Collectively , the evidence points to a close functional and subcellular association of CD36 with β1 integrin in multiple cell types . The integrin α5β1 is one of 24 known members of the integrin family of adhesion molecules that are formed by noncovalent linkage of an α and β subunit with the ligand-binding ‘head’ region of the integrin being formed by both subunits [44] . Integrin α5β1 can exist in multiple conformational states , i . e . inactivated , intermediate activated , and fully activated . It can be activated by inside-out signaling or by non-physiological stimuli such as an activating antibody or extracellular Mn2+/Mg2+ . The natural ligand of α5β1 is the extracellular matrix protein fibronectin ( FN ) , a dimeric glycoprotein composed of two subunits each with multiple homologous domains named FNI , FNII , and FNIII . Optimal binding of FN to the integrin requires both the RGD motif present in FNIII domain 10 ( FN10 ) and a synergy site located in FN9 [45] . The RGD motif binds to the β1 subunit while the synergy site interacts with α5 . AFM studies have shown that FN with RGD deletion binds weakly to α5β1 , while the force of detachment is only slightly less than wild type when the synergy site is mutated , and there is no enhancement of binding upon integrin activation [46] . These findings would support our results on the differential effect of β1 and α5 knock down on IRBC adhesion and adhesive strength . Interaction with both subunits appeared to be essential for IRBC adhesion under shear stress in the flow chamber assay . In the AFM experiments , IRBC and HDMEC were brought together mechanically at a constant rate and kept in contact by a constant force . In this situation , the interaction with α5 appeared dispensable , as evidenced by the normal force of detachment in α5 knockdown cells , and in cells pre-treated with the anti-α5 mAb JBS5 . As endothelial cells are surrounded by an extracellular glycocalyx of approximately 30 to 50 nm in thickness that cannot be breached by either the bent ( 10 nm ) or extended ( 20 nm ) integrin forms [47] , the initial attachment to CD36 may in addition to clustering lead to compression of the glycocalyx , bringing α5β1 integrin to close proximity of ligands on IRBC . Several mechanisms that underlie the change in affinity or avidity of integrins for their ligands have been proposed [48] . The importance of clustering in promoting the avidity of integrin-ligand interactions , i . e . an increase in adhesiveness independently of integrin conformational changes , has been demonstrated by single molecule tracking . Using this technique , PMA was seen to induce a 10-fold increase in the lateral diffusion rate of LFA-1 in EBV transformed B lymphocytes [40] . The movement induced was random instead of directed , indicating that it was due to a release of the integrin from cytoskeletal attachment and subsequent free diffusion rather than a directional movement due to the application of forces . As a corollary , it would appear that the nonadhesive state of integrins is actively maintained by the cytoskeleton . The movement of LFA-1 was considered an important early step in the adhesion of monocytes to immobilized ICAM-1/E-selectin under flow conditions [41] . However , the bond formed by IRBC with clustered integrins on HMEC-1 induced by PMA appeared to be unable to withstand shear flow . Whether ligation by IRBC results in integrin activation and/or outside-in signaling that further modifies endothelial cell functions such as barrier function or proinflammatory mediator production remains to be determined . Using an αvβ3/αvβ5-specific antagonist ( cRGDfV ) and an inhibitory mAb to αvβ3 , we were unable to confirm a role for αvβ3 integrin in mediating cytoadherence on HDMEC . Nevertheless , αvβ3 or other integrins may play an important role in IRBC adhesion to endothelial cells in other anatomical locations or other cell types , as already demonstrated for dendritic cells [27] . Integrins on the surface of platelets also play an active role in many physiological process , e . g . αIIbβ3 in thrombus formation [49] , and may quite likely participate in the interaction of IRBC and platelets . The ligand ( s ) on IRBC that interacts with α5β1 integrin remains to be determined . The cytoadherent ligand PfEMP1 appears to be the most likely candidate , but ligands of host cell origin such as phosphatidylserine that is exposed by changes in membrane topography induced by an intracellular parasite may also play a part [50] . The presence of RGD motifs on PfEMP1 was noted in the initial reports on the cloning of the gene in the lab-adapted parasite clones MC and FCR3 [22] , [23] . While RGD motifs did not occur in each protein sequence examined , their positions varied ( DBL1-4 ) when they did occur . More recently , an analysis of seven P . falciparum genomes revealed that RGD motifs are overrepresented in highly conserved positions in the DBLα0 domain in close proximity to several cysteine residues [24] . This observation raises the interesting question of why the localization of RGD motifs on PfEMP1 is conserved in 1a parasite protein that is otherwise highly structurally variant . Further studies into the process of ligand recognition between IRBC and integrins may shed light not only on IRBC adhesion to endothelial cells , but to other host cells such as monocyte/macrophages and platelets . In summary , we have demonstrated that the integrin α5β1 acts synergistically with CD36 in mediating cytoadherence of IRBC to primary human microvascular endothelium . Together with our previous finding of IRBC-induced CD36 clustering and actin cytoskeletal rearrangement [6] , a picture is emerging of the formation of a cytadherence synapse involving multiple adhesion molecules and ligand ( s ) on IRBC . This type of cooperative adhesive interactions may be of critical importance in enabling IRBC adhesion in the microcirculations . The ultimate goal for elucidating the molecules and processes involved in cytoadherence to primary endothelial cells is to develop rational treatments that could target key receptor molecules that come into play under physiological shear stress . Integrins are known to mediate immune cell recruitment and tumor cell migration that are associated with autoimmune diseases such as multiple sclerosis and different types of cancer respectively , and have become common therapeutic targets for these diseases [26] . As CD36 on different host cells may have both a protective and a pathological role against P . falciparum in the human host , anti-adhesive therapy targeted against its functional partners such as integrins with antibodies or small peptides rather than CD36 itself may provide a mechanism to decrease IRBC cytoadherence to microvascular endothelium while preserving the beneficial effects of this molecule against P . falciparum . The collection of P . falciparum-infected blood specimens was approved by the Ethics Committee of the Faculty of Tropical Medicine , Mahidol University , Bangkok , Thailand . Written informed consent was obtained from all patients and/or their legal guardians according to the Declaration of Helsinki . The collection of discarded foreskins for the isolation of endothelial cells and red blood cells from normal donors was approved by the Conjoint Ethics Board of Alberta Health Services and The University of Calgary , Alberta , Canada . Unless otherwise specified , all tissue culture and PCR reagents were obtained from Invitrogen Life Technologies Canada Inc . ( Burlington , ON ) and chemical reagents were purchased from Sigma-Aldrich Co . ( St . Louis , MO ) . The Src-family kinase inhibitor PP1 and the inactive analog PP3 were purchased from Enzo Life Sciences International , Inc . ( Plymouth Meeting , PA ) . Chemiluminescence HRP substrate was purchased from Millipore Corp . ( Billerica , MA ) . Endothelial basal medium ( EBM ) was purchased from Lonza Walkersville , Inc . ( Walkersville , MD ) . The following mAb were used: anti-human CD36 clone FA6-152 ( Beckman Coulter Canada , Inc . , Mississauga , ON ) ; anti-human integrin β1 clones TDM 29 and TS2/16 ( Millipore ) ; FITC- and PE-labelled anti-human integrin β1 clone MEM-101A ( Abcam , Cambridge , MA ) ; anti-human α5 clones JBS5 ( Millipore ) ; anti-human ICAM-1 clone 84H10 ( R&D Systems , Inc Minneapolis , MN ) ; anti-human αvβ3 clone 23C6 ( Chemicon International ) ; mouse IgG1 clone 11711 ( R&D Systems ) ; anti-phospho-Tyr418Src ( BioSource; Invitrogen ) , anti-His-tag ( His-probe ( H-15 ) ) ( Santa Cruz Biotechnology Inc . , Santa Cruz , CA ) ; FITC goat anti-mouse IgG1 ( Becton Dickinson , San Diego , CA ) ; and Alexa Fluor 488 or 568 goat anti-mouse IgG1 antibodies and rhodamine-phalloidin ( Molecular Probes , Invitrogen ) . Horseradish peroxidase ( HRP ) -conjugated secondary antibodies were purchased from Jackson ImmunoResearch Laboratories ( West Grove , PA ) . The RGD ( H-Gly-Arg-Gly-Asp-Ser-Pro-OH ) and RAD ( H-Gly-Arg-Ala-Asp-Ser-Pro-OH ) peptides were purchased from Calbiochem , EMD Bioscience Inc , La Jolla , CA . cRGDfV ( cyclo Arg-Gly-Asp-D-Phe-Val and cRADfV ( cyclo Arg-Ala-Asp-D-Phe-Val ) were from Bachem Inc . , Torrance , CA . The majority of the experiments was performed with the parasite line 7G8 that binds to both CD36 and ICAM-1 . The stock culture was shown to be free of mycoplasma contamination by RT-PCR ( MycoAlert , Lonza Walkersville , Inc . , Walkersville , MD ) . Frozen aliquots were thawed and cultured for 24 to 30 h at 37°C and 5% CO2 until the late trophozoite/early schizont stage as determined by light microscopy . IRBC cultures were used in single experiments and then discarded . Experiments were also performed with cryopreserved clinical parasite isolates obtained from adult Thai patients with acute falciparum malaria at the Hospital for Tropical Diseases , Bangkok , Thailand [6] . Primary human dermal microvascular endothelial cells were harvested from discarded neonatal human foreskins using 0 . 5 mg/ml Type IA collagenase ( Roche Diagnostics , Indianapolis , IN ) as described [3] . Harvested cells were seeded in 60 mm tissue culture dishes in endothelial basal medium ( EBM ) with supplements provided by the manufacturer . When cells were confluent , they were further purified on a magnetic bead separation column using CD31-coated beads ( Miltenyi Biotec , Auburn , CA ) . Only cell preparations which were >95% positive for CD36 expression by flow cytometry were maintained for experiments . Experiments were performed with cells from passages two to five that were demonstrated to consistently support IRBC adhesion . HMEC-1 , an immortalized cell line derived from HDMEC [29] , was a kind gift of F . J . Candal at Emory University , Atlanta , Georgia . The cell line was maintained in EBM as for primary cells . GFP-labeled CD36 was produced using the AdEasy adenoviral system ( Stratagene , La Jolla , CA ) . GFP-labeled ICAM-1 was produced using the Virapower adenoviral expression system ( Invitrogen ) . The selected recombinant was used to transfect HEK293 cells where deleted viral assembly genes were complemented in vivo . Harvested virus titers were adjusted to 1 . 0×1010 plaque-forming units ( pfu ) per ml . Viruses with the adeno-GFP construct were used as the control . Transduction was carried out using 1 . 0–2 . 0×107 pfu/ml . The recombinant adenoviruses were routinely tested for the presence of endotoxin using the Kinetic QCL Limulus Amebocyte Lysate assay ( Lonza , Walkersville , MD ) and contained <0 . 3 endotoxin units/ml . Expression of CD36 and ICAM-1 on transduced cells was confirmed by both immunofluorescence microscopy and flow cytometry . HDMEC were seeded in 35 mm tissue culture dishes at 2×105 cells/dish and transfected 24 h later when the cells were 50 to 60% confluent . At the time of transfection the medium was aspirated and replaced with 1 . 0 ml of Optimem . The transfection mixture of 10 µl HiPerfect ( Qiagen . GmBH , Hilden , Germany ) and 20 nM siRNA for β1 integrin ( Qiagen ) or α5 integrin or 20 nM scrambled siRNA ( All Stars Negative Control , Qiagen ) was added in 100 µl of Optimem . Four hours after transfection , 1 . 0 ml of EBM was added to each dish . Monolayers were used for flow chamber studies 72 h after transfection . Cell lysates collected at the same time were used to confirm gene knockdown by Western blot analysis . For AFM experiments , transfected cells were trypsinized after 24 h and seeded on 25 mm glass coverslips ( see below ) . The recombinant PpMC-179 protein representing the minimal binding domain of PfEMP1 ( residues 88–267 of the CIDR domain of the parasite strain Malayan Camp ) for CD36 [39] with a His6-tag on the C terminus was used to coat carboxyl functionalized polystyrene beads with a diameter approximating that of red blood cells ( 6 . 37±0 . 37 mm , Bangs Laboratories Inc . , Fishers , IN ) [6] . Beads were first covalently coated with an anti-His-tag antibody according to the instructions of the manufacturer . Prior to being used , the beads were washed in PBS and resuspended at a ratio of 1 µl of beads to 1 µg of PpMC-179 peptide in 20 µl of PBS for 1 h at room temperature . The presence of PpMC-179 on extensively washed beads was confirmed by Western blot . Polystyrene beads were also coated with FA6-152 , a mAb known to inhibit IRBC binding to CD36 , or 84H10 , a mAb known to inhibit IRBC binding to ICAM-1 . One microgram of antibody in 10 µl PBS was added to 1 µl of beads for 2 h on a rocker and then let stand for 1 h . After washing in PBS , the beads were blocked with 100 µl of 0 . 1% BSA/PBS for 30 min . Beads were used within 48 h of preparation . The presence of antibody on the beads was confirmed by flow cytometry using goat-anti-mouse IgG-Alexa 488 . Mouse IgG1-coated beads were used as controls . HDMEC were seeded into μ-slideVI0 . 4 ( Ibidi GmbH , Munich , Germany ) at 2×104 cells/chamber . When cells were 95% confluent ( 48 h ) , they were transduced with adenoviral GFP-CD36 . For live cell imaging , MACS purified IRBC at 0 . 1% hematocrit were added to the chamber 24 hours after transduction , followed by the addition of FITC-labelled anti-β1 integrin . IRBC and HDMEC interactions were imaged in an enclosed , humidified chamber maintained at 37°C and 5% CO2 . For experiments with antibody or peptide-coated beads , 0 . 5 µl of antibody- or peptide-coated beads resuspended in 60 µl HBSS were incubated with monolayers at 37°C and 5% CO2 for 30 min . After unbound beads were removed with HBSS , the monolayers were fixed with 1%PFA for 30 min at room temperature . They were blocked with 1%BSA and stained , or permeabilized with 0 . 2% Triton X-100 for 5 min prior to labeling with antibodies . Quantification of β1 integrin clustering was performed by randomly selecting 3 microscopic fields at 60× magnification per condition per experiment as described [6] . Except for IgG1 or anti-His-tag coated beads , each field contained an average of 15 to 20 beads . Adherent beads associated with localized patches of fluorescence or discrete fluorescent rings were scored as positive . Beads that were partially visible in the field of view or rings that were present without beads were excluded . Each field was scored independently by 2 blinded observers . Results are expressed as per cent beads positive for protein recruitment . IRBC-endothelial cell interactions at fluid shear stresses approximating those in the microvasculature were studied using a parallel plate flow chamber as described [3] . A 1% IRBC suspension was infused over confluent HDMEC monolayers at 1 dyne/cm2 that allowed us to optimally visualize the adhesive interactions in real time . Experiments were recorded and analyzed off-line . A rolling IRBC was defined as one which displayed a typical end-on-end rolling motion at a velocity of <150 mm/sec , compared to a centerline red blood cell flow rate of >1000 mm/sec , and a velocity of >150 mm/sec for non-interacting cells in close proximity to the endothelial monolayer . An adherent IRBC was defined as one which remained attached for >10 sec . Results were expressed as the number of adherent IRBC/mm2 . Force spectroscopy was performed using a Nanowizard II atomic force microscope ( AFM ) equipped with a CellHesion Module ( JPK Instruments , Berlin , Germany ) as described [6] . The AFM was mounted on the stage of a Zeiss Axiovert 200 inverted light microscope ( Carl Zeiss , Thornwood , NY ) both of which were enclosed in a humidified chamber maintained at 37°C and 5% CO2 . Force measurements were obtained using relative contact mode with a relative set point of 150 pN , extend/retract rates of 1 . 5 µm/s , Z-length of ∼50 µm , and a sample rate of 256 Hz . The baseline vertical offset was adjusted prior to every AFM reading and the software was set to correct for nonlinearity and hysteresis of the piezo . Analysis of data to determine the maximum detachment force was by software provided by JPK Instruments . Force measurement experiments were performed with 2×104 HDMEC seeded on one half of a 25 mm round glass coverslips pre-coated with 0 . 2% gelatin . At the time of confluence ( 48 h ) , the coverslips were loaded into a custom-built liquid cell that allowed the force measurements to be performed in fluid phase . The monolayers were washed twice with HBSS . A volume of 1 µl of IRBC at 0 . 5% hematocrit and 5 to 10% parasitemia was added to the half of the coverslip not seeded with HDMEC . The liquid cell was placed on the stage of the inverted microscope . At the start of the experiment , the cantilever that had been functionalized with dopamine hydrochloride was lowered onto an IRBC that was allowed to adhere for 60 sec before the cantilever with the adherent IRBC was moved to the monolayer and brought into contact with an endothelial cell . In experiments in which inhibitors and antibodies were used , PP1 ( 10 µM ) and PP3 ( 10 µM ) were diluted from 1000× stock solutions in DMSO to the working concentration in culture medium and incubated with HDMEC monolayers for 30 min at 37°C . After removal of inhibitors , monolayers were washed 2× with HBSS before force measurements were carried out . For antibodies , the HDMEC monolayers were loaded into the AFM chamber , washed twice with HBSS , and then incubated with 200 µl of antibody at 5 to 10 µg/ml in HBSS for 30 min at 37°C . Prior to force measurements , the monolayers were washed 2× with HBSS . Statistical analysis was performed using GraphPad Prism ( version 4 , GraphPad Software Inc . ) . Data are expressed as mean ± SEM unless otherwise stated . Data from two groups were compared using Student's 2-tailed t-test for paired samples unless otherwise specified . p values of 0 . 05 or less were considered statistically significant .	Of the several species of malaria parasites that infect humans , disease caused by Plasmodium falciparum is responsible for most of the deaths . The unique pathological finding of this infection is the intense adhesion of infected red blood cells ( IRBC ) in the microcirculation , resulting in obstruction to blood flow and organ dysfunction . The focus of our research is to identify the molecules on host endothelial cells that support the adhesion as potential therapeutic targets . In this report , we showed for the first time a functional association between CD36 , a well studied adhesion molecule for parasite adhesion , and α5β1 , a member of the integrin family of adhesion molecules that are important for adhesion of leukocytes to blood vessels and cell adhesion to the extracellular matrix . We found that by itself , α5β1 integrin does not support IRBC adhesion . When IRBC adhere to CD36 , the integrin is recruited to the site of adhesion through activation of the Src family kinase signaling pathway . As a result , both the number of adherent IRBC and the strength with which they adhere is greatly increased . These results demonstrate that IRBC adhesion is a dynamic and complex process . We need to identify each of the functional components to design anti-adhesive therapy .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "medicine", "biology" ]	2013	CD36 Recruits α5β1 Integrin to Promote Cytoadherence of P. falciparum-Infected Erythrocytes
Genome-wide association studies ( GWAS ) have transformed our understanding of the genetics of complex traits such as autoimmune diseases , but how risk variants contribute to pathogenesis remains largely unknown . Identifying genetic variants that affect gene expression ( expression quantitative trait loci , or eQTLs ) is crucial to addressing this . eQTLs vary between tissues and following in vitro cellular activation , but have not been examined in the context of human inflammatory diseases . We performed eQTL mapping in five primary immune cell types from patients with active inflammatory bowel disease ( n = 91 ) , anti-neutrophil cytoplasmic antibody-associated vasculitis ( n = 46 ) and healthy controls ( n = 43 ) , revealing eQTLs present only in the context of active inflammatory disease . Moreover , we show that following treatment a proportion of these eQTLs disappear . Through joint analysis of expression data from multiple cell types , we reveal that previous estimates of eQTL immune cell-type specificity are likely to have been exaggerated . Finally , by analysing gene expression data from multiple cell types , we find eQTLs not previously identified by database mining at 34 inflammatory bowel disease-associated loci . In summary , this parallel eQTL analysis in multiple leucocyte subsets from patients with active disease provides new insights into the genetic basis of immune-mediated diseases . Most of the hundreds of disease-associated single nucleotide polymorphisms ( SNPs ) identified by genome-wide association studies ( GWAS ) lie outside protein-coding regions , and are presumed to act by regulating gene expression [1 , 2] . Investigating the effects of allelic variation on transcription by expression quantitative trait locus ( eQTL ) mapping provides insights into how risk loci influence disease susceptibility , and may identify pathways amenable to pharmacological intervention . eQTLs vary considerably between tissues and cell types [3–5] , and so when attempting to interpret GWAS signals through eQTL data , the context in which eQTLs are present is critically important . eQTLs have been previously examined in cell lines [6 , 7] or in one or two primary immune cell types [5 , 8 , 9] , but a comparative analysis of eQTLs across a broad range of the major leucocyte subsets implicated in immune-mediated disease pathogenesis has not yet been carried out . Most autoimmune diseases exhibit less than 50% concordance in monozygotic twins , highlighting the importance of environmental factors in their pathogenesis [10] . Studies in model organisms show that eQTLs vary in different environmental conditions [11 , 12] , and in vitro stimulation of primary human immune cells can both abrogate and induce eQTLs [13–15] . These experiments cannot , however , reproduce the in vivo inflammation that characterizes human autoimmune and inflammatory disease . Moreover , these studies have typically been limited to a few hundred genes ( e . g . ref . s [14 , 15] ) . We hypothesised that an analysis of gene expression across different immune cell types in both health and active inflammatory disease could provide additional insight into associations between genotype and phenotype . Our study across five immune cell types provides the most comprehensive comparison to date of the cells known to play roles in immune-mediated disease , and includes neutrophils , a key immune cell type for which a systematic eQTL analysis has not been reported . We examined eQTLs across approximately 8 , 000 genes selected in an unbiased manner , and by including both patients with active inflammatory disease and healthy controls , we reveal eQTLs present only in the context of human in vivo inflammation . We anticipate that such eQTLs may be important in understanding the heterogeneity in immune responses between individuals , and may have implications for understanding the inter-individual variation in clinical course seen in autoimmune and infectious diseases . Limiting the initial analysis to 91 IBD patients and 43 HVs genotyped on the Illumina Human OmniExpress12v1 . 0 BeadChip ( see Table 1 for sample sizes by cell type ) , we mapped local-acting ( cis ) eQTLs using an additive genetic model for 9 , 041 probesets corresponding to 7 , 552 unique Entrez Gene IDs selected by highest variance ( see Methods ) . Using separate analysis of each cell type , we found a cis eQTL in at least one cell type for 4 , 524 ( 59 . 9% ) of the 7 , 552 genes analysed ( 5% FDR , see S1 Data for summary statistics ) . We next investigated sharing of eQTLs across cell types . Previous reports have claimed that eQTLs are highly specific to different leucocyte subsets [5] . These conclusions were based on separate eQTL analysis in each cell type , followed by comparison of the resulting lists of significant hits in each . As Flutre et al [16] highlighted , this approach fails to account for incomplete power and leads to exaggerated claims of cell-type specificity . An eQTL truly present in two cell types may pass the significance threshold in one but may fail to meet significance in another , and thus falsely be called a cell-type specific eQTL . This may be due to random noise , or because there is insufficient power to detect an eQTL which is weaker but nonetheless present in the second cell type . In order to more accurately estimate the degree of sharing of eQTLs across immune cell subsets , we performed joint analysis across the 5 cell types using a Bayesian model ( eQTL Bayesian Model Averaging , or ‘eQTLBMA’ [16] ) that , for each SNP-gene pair , assesses the evidence for each possible configuration of eQTL absence/presence across the 5 cell types . There are thus 25 − 1 = 31 possible configurations for an eQTL to be present in at least one of 5 tissues . Through joint analysis of all 5 cell types , we found that 3 , 440 probesets ( corresponding to 3 , 185 unique Entrez Gene IDs ) had a significant cis eQTL ( 5% Bayes FDR ) . For each of these probesets , we took the SNP with the highest posterior probability of being the eQTL . Then for the resulting list of SNP-probeset pairs , we compared the posterior probabilities for all possible configurations of eQTL presence/absence across cell types . This approach revealed that 45 . 1% of eQTLs have the highest posterior probability for presence across all 5 cell types , whereas only 9 . 9% have the highest posterior probability for presence of the eQTL in just one cell type ( Fig 1A ) . In stark contrast , using the naïve approach of separate analysis of each cell type , followed by comparison of the lists of significant eQTLs found in each , over half of eQTLs identified ( 5% FDR ) were detected in only one cell type ( S1 and S2 Figs and see S1 Text ) . This finding suggests that previous reports of marked immune cell-type eQTL specificity from separate tissue analysis [5] were overstated . For the AAV patients ( n = 46 ) , expression data were not available for B cells , so we used eQTLBMA to perform eQTL modelling jointly across 4 cell types . This analysis detected eQTLs for 899 probesets ( corresponding to 845 genes ) , using a 5% Bayes FDR . Consistent with the analysis of the IBD-HV data , 44 . 4% of eQTLs had the best posterior probability for presence across all 4 cell types , and only 12 . 1% were specific to a single cell type ( Fig 1B ) . However , there was a greater proportion of probesets with neutrophil-specific eQTLs in the AAV analysis ( 10 . 8% , versus 4 . 3% in the IBD-HV data ) , perhaps reflecting the increased neutrophil turnover or activation that is prominent in AAV . This observation did not simply reflect that fewer cell types were examined in the AAV analysis; re-analysis of the IBD-HV data excluding B cells found that 4 . 8% were neutrophil-specific ( S1 Text , S3 Fig ) . The finding of a higher proportion of neutrophil-specific eQTLs was also apparent from the one-at-a-time cell-type analysis; in AAV 27 . 2% of eQTLs were detected in neutrophils only versus 9 . 7% in the IBD-HV data ( S1 Fig ) . Hierarchical clustering of cell types by their eQTL test statistics from separate cell-type analyses recapitulated their haematopoietic differentiation; the myeloid and lymphoid cells cluster together , with CD4 and CD8 T lymphocytes most closely related ( Fig 2A ) . Examples of this sharing of eQTLs between related cell types include a myeloid-specific eQTL for KSR1 ( at a GWAS locus for Crohn’s disease ) and T cell-specific eQTLs for DEPTOR and HTR6 ( Fig 2B , S4 Fig ) . We quantified the similarity in eQTL profiles between cell types using the Jaccard coefficient . The Jaccard coefficient is defined as the intersect divided by the union of two sets ( here , the number of eQTLs common to a pair of cell types , as a proportion of the total number of eQTLs identified in either or both cell types ) . Using the Bayesian method of joint analysis across cell types , the Jaccard coefficient for CD4 and CD8 T cells was 100% for both the IBD-HV dataset and the AAV dataset ( S5 Fig ) . Sharing of eQTLs between lymphoid and myeloid cells was less common , with , for example , Jaccard coefficients of 48% and 45% between CD4 T cells and neutrophils in the IBD-HV and AAV data respectively . For comparison , we also calculated the Jaccard coefficients based on eQTLs detected through separate analysis of each cell type . To do this , we restricted the analysis to individuals for whom expression data were available for all cell types being compared ( n = 65 for the 5 celltypes in the IBD-HV analysis , and n = 33 for the 4 cell types in the AAV analysis ) . By using the same set of individuals , the sample size and genotype matrix ( predictor variables ) were identical for each cell type . This both ensures equal statistical power for each cell type and controls for inter-individual variability . Nevertheless , as highlighted by Flutre et al [16] , separate tissue analysis results in under-estimation of eQTL sharing as a consequence of incomplete power ( as occurs with any finite sample size ) , even if power is equal for each tissue . Using separate cell-type analysis , CD4 and CD8 T cells again had the most similar eQTL profiles , but the Jaccard coefficient was 40% in the IBD-HV analysis , and only 31% in AAV ( S1 Fig ) . The lower proportion in the AAV dataset highlights how , as sample size and power decrease , the under-estimation of eQTL sharing that results from separate cell-type eQTL analysis worsens . This confirms that separate cell-type analysis identifies fewer shared eQTLs than joint analysis across cell types with the Bayesian method . To investigate whether the Bayesian model unduly favoured declaring eQTLs as shared across all cell types , we randomly permuted the sample labels for the CD4 T cell expression data , leaving the genotype data and the expression data for the other cell types unchanged , and then re-ran the analysis with eQTLBMA . After permutation of the CD4 T cell expression data , very few eQTLs were declared in CD4 T cells , and the Jaccard coefficient between CD4 and CD8 T cells was 0 . 1% ( S1 Text , S6 Fig ) , indicating that eQTLBMA is behaving appropriately . To explore the possibility that eQTL cell-type specificity reflected lack of gene expression in some cell types , we re-ran the joint eQTL mapping across cell types in the IBD-HV data , limiting analysis to 5 , 186 probesets with evidence of robust expression in all 5 cell types ( Methods ) . This showed that eQTL sharing was greater amongst these genes , with 64% of eQTLs found in all 5 cell types ( S7 Fig ) . Therefore , lack of expression accounts for some , but not all , of the eQTL cell-type specificity that we observed . Finally , whilst most eQTLs shared across cell types had the same direction of effect , a small proportion of these “shared” eQTLs acted in opposing directions ( Fig 3A , S8–S11 Figs ) . Examples include eQTLs for CD52 and CD101 ( Fig 3B ) . Independent analysis of the AAV cohort confirmed this observation , indicating that these eQTLs with discordant effects between cell types are unlikely to be false positives ( S1 Text ) . To identify eQTLs that are specific for active inflammatory disease in a statistically robust manner , we analysed the IBD-HV data using a linear model with a genotype × disease interaction ( G×D ) term . A significant G×D interaction term for a SNP-gene pair indicates that the effect of genotype on expression is significantly different in IBD versus health . In biological terms , this includes ( i ) eQTLs present in health but abrogated in IBD , ( ii ) eQTLs present in IBD but not in health , ( iii ) eQTLs with opposing directions of effect in health compared to IBD , and ( iv ) eQTLs whose effects in health and IBD are in the same direction , but of significantly different magnitudes . More formally , the interaction term assesses whether there is a significant difference in the slope of the genotype-expression regression line between healthy individuals and IBD patients i . e . whether the effect size of a unit change in allele dose on expression is significantly different between health and disease ( S12 Fig ) . Fitting a model with an interaction term is statistically more robust than the naïve approach of separate analysis of IBD and HV cohorts , followed by comparison of the resulting lists to find eQTLs specific to one group or the other ( see Methods and S1 Text ) . We used a ‘two-step’ procedure to increase power ( Methods ) and a 15% FDR as the significance threshold . Given the relative lack of statistical power to detect interaction effects compared to main effects , and our relatively modest sample sizes , we felt a 15% FDR provided a reasonable balance between false positives and false negatives . Using this threshold , we identified 13 genes with an eQTL exhibiting a G×D interaction effect across the cell types examined ( Fig 4 , S1 Table ) . In CD4 T cells , for example , rs11230584 , a SNP located between CD5 and CD6 was associated with expression of both genes in IBD patients but not in healthy individuals ( Fig 4A ) . For CD6 , uncorrected p-values , Benjamini-Hochberg adjusted p-values and q-values for the G×D interaction were 7 . 2×10−7 , 0 . 00048 , and 0 . 00048 respectively , and for CD5 5 . 6×10−5 , 0 . 0038 , and 0 . 0019 . CD5 and CD6 play important roles in lymphocyte signalling ( see Discussion ) , and therefore this inflammatory disease-specific eQTL is likely to lead to differences in immune responses according to genotype . Interestingly , not all the genes with disease-specific eQTLs have strictly immunological functions . For example , in neutrophils we identified an eQTL for CTDP1 in both the IBD and the AAV cohorts that was absent in health ( Fig 4E ) . CTDP1 encodes a protein phosphatase , FCP1 , which regulates gene expression by dephosphorylating the C-terminus of the largest subunit of RNA polymerase II . Thus this analysis reveals the novel and potentially clinically relevant observation of eQTLs present only in the context of inflammatory disease . For patients with AAV , in addition to the baseline samples , we also measured gene expression in peripheral blood monocytes and neutrophils taken 3 months and 12 months into treatment ( Table 2 ) . The treatment of active AAV begins with induction therapy with intensive immunosuppression . This consists of high-dose corticosteroids , and cyclophosphamide ( a cytotoxic agent ) or rituximab ( a monoclonal antibody against CD20 which causes B cell depletion ) . The corticosteroid dose is slowly weaned , and once a period of stable remission has been achieved , patients are switched to maintenance immunosuppression ( typically low-dose corticosteroids plus azathioprine ) . In the example of CTDP1 , we found that the eQTL present in active IBD and AAV is absent in AAV patients at 3 or 12 months , providing convincing evidence that the eQTL is inflammation- rather than disease-specific . To compare eQTL profiles in AAV pre- and post-treatment more generally , we used the same Bayesian model averaging method ( eQTLBMA ) that we used for the analysis of cell-type specificity . For this analysis , the subgroups were not cell types , but instead were time points . These were a ) time 0 ( pre-treatment , when vasculitis is newly diagnosed or flaring ) , b ) time 3 months ( 3 months into induction therapy ) , and c ) time 12 months ( patients on maintenance immunosuppressive therapy ) . In mononcytes , no time zero-specific eQTLs were identified . In neutrophils , joint analysis of the three timepoints identified eQTLs for 288 probesets corresponding to 262 unique genes ( 5% Bayes FDR ) . For 14% of these genes , the ‘best’ model ( i . e . the configuration with the highest posterior probability ) was that eQTL is only present at time zero ( Fig 5A ) . Examples include MTOR , NPHP3 , and SREBF1 ( Fig 5 ) . Thus we show that eQTLs present in active untreated AAV can disappear after treatment . These eQTLs are probably inflammation-dependent , although their absence post-treatment could reflect drug effects independent of the resolution of inflammation . To identify the effects of disease-associated SNPs on gene expression , we took SNPs associated with complex traits listed in the National Human Genome Research Institute GWAS catalogue , and proxy SNPs in high linkage disequilibrium ( LD ) ( r2 >0 . 8 ) , and examined their overlap with eQTL SNPs ( eSNPs ) from our analyses ( IBD-HV S2 Table , AAV S3 Table ) . We initially focused on IBD as its genetic basis has been extensively studied; a recent GWAS meta-analysis identified 163 independent risk loci [17] . At 34 IBD-associated loci ( CD , UC , or both ) we identified eQTLs from the analysis of the IBD-HV data that were not previously apparent from eQTL database mining [17] ( 18 where no eQTL was identified previously , and 16 where we found eQTLs for additional genes; S13 and S14 Figs ) . When combined with those described in ref . [17] , our findings increase the number of IBD-risk loci with eQTLs from 64 to 82 . Disease-associated SNPs are often assumed to exert their effect through the nearest gene but we found that intronic disease-associated SNPs may instead affect the expression of other genes in the region . For example , rs8049439 , an intronic SNP in ATXN2L associated with early-onset IBD [18] , is an eQTL for the neighbouring TUFM in all five cell types examined ( Fig 6A ) , but not for ATXN2L itself . These data allow the creation of a revised list of ‘IBD-associated genes’ based not on proximity to disease-associated SNPs , but instead upon the biological effects that these SNPs have on gene expression ( S15 and S16 Figs , S4 Table ) . Analysis of the direction of effect and cell-type specificity of eQTLs can provide important insights into the role of genetic variation on disease pathogenesis . To illustrate this , three examples will be described . The first demonstrates how detailed eQTL analysis can revise assumptions about genetic contributions to pathogenesis . rs102275 , an intronic variant in TMEM258 associated with Crohn’s disease [19] , is an eQTL for FADS2 in all cell types examined ( Fig 6B ) . FADS2 encodes fatty acid desaturase 2 , a rate-limiting enzyme in the conversion of linoleic acid to pro-inflammatory arachidonic acid . FADS2 knockout mice develop duodenal and ileocecal ulceration [20] , leading to speculation that FADS2 expression is protective against IBD [19] . However , the risk allele ( G ) for Crohn’s disease is associated with increased , not decreased , FADS2 expression ( Fig 6B ) , suggesting that higher FADS2 expression may in fact increase IBD susceptibility . Simply demonstrating that a SNP is associated with both disease and gene expression is not sufficient to infer causality as there may be two distinct causal variants in LD . We therefore performed colocalisation testing [21] of the eQTL for FADS2 and for IBD . In all five cell types , this provided strong evidence that the causal variant for the eQTL and Crohn’s disease susceptibility was the same ( posterior probability >98% ) . Detailed eQTL mapping may also help differentiate between multiple biologically plausible candidate genes . For example , GWAS have identified an association between rheumatoid arthritis and rs3761847 , a SNP in an LD block that encompasses two genes that have been implicated in chronic inflammation: TRAF1 and C5 [22 , 23] ( Fig 7A , S17 Fig ) . Our data reveal that this SNP is an eQTL for TRAF1 expression in B cells ( Fig 7B ) and C5 expression in monocytes . In B cells rs3761847 is the most significant eSNP for TRAF1 expression ( p 4×10-5 ) . However , rs3761847 is in weak LD ( r2 0 . 35 ) with the most significant eSNP for monocyte C5 expression ( rs10818504 , p 3×10-9 ) . After conditioning on rs10818504 , there is no residual association of the rs3761847 with monocyte C5 expression ( p 0 . 98 ) , leading to the conclusion that rs3761847 is likely to drive disease susceptibility through TRAF1 in B cells , and not through modulating C5 . Finally , comprehensive eQTL mapping may implicate new candidate genes . NOD2 is central to responses to intracellular pathogens , and SNPs in the NOD2 region have been linked to leprosy and Crohn’s disease [17 , 24] . rs9302752 lies between NOD2 and SNX20 in a susceptibility locus for leprosy but not Crohn’s disease , and has been presumed to predispose to leprosy via altering NOD2 expression [24 , 25] . We find an eQTL at this locus not only for NOD2 but also for SNX20 in neutrophils , monocytes and CD4 T cells , with the direction of effect on expression of both genes in neutrophils opposite to that in monocytes and CD4 T cells ( Fig 8 ) . Independent analysis of the AAV data confirmed the discordant effect of the eQTL in neutrophils . NOD2 and SNX20 are transcribed in opposite directions , and so may share a common regulatory element . Consistent with this , the Pearson correlation coefficients between NOD2 and SNX20 mRNA levels are 0 . 69 , 0 . 54 , 0 . 54 in neutrophils , monocytes and CD4 T cells respectively ( PEER-adjusted IBD-HV expression data ) , and ENCODE data suggests active regulatory elements in this region ( Fig 8 ) . SNX20 cycles P-selectin glycoprotein ligand-1 ( PSGL1 ) into endosomes , controlling its interaction with the cell adhesion molecules P- , E- and L- selectin on myeloid cells and activated T cells . Whilst NOD2 remains the leading candidate gene at this locus , the eQTL data indicates consideration should also be given to the role of SNX20 . We performed eQTL mapping in the context of either active inflammatory disease or health in five leucocyte subsets that play key roles in the pathogenesis of immune-mediated diseases . Previous reports have suggested that eQTLs are highly specific to immune cell subsets [5] . A naïve interpretation of our data would lead to the same conclusion . Performing eQTL analysis separately in each cell type , and comparing the results obtained from each , suggested that just 5% of genes share eQTLs across all the cell types examined . While closely related cell types appeared to share more eQTLs , even among CD4 and CD8 T cells less than half of eQTLs were detected in both . However , this approach fails to accounts for incomplete power and leads to substantial underestimates of eQTL sharing . When our data are analysed with a more sophisticated joint modelling approach , estimates of the proportion of eQTLs present in all five cell types examined rises to 45% . eQTL analysis of multiple cell types , whether performed with joint modelling across cell types or through one-at-a-time analysis of each cell type , identified many more eQTLs than analysis of any given single cell type . This highlights the importance of studying eQTLs in relevant cell types , and emphasises that studies using a single tissue or mixed cell populations such as whole blood [26] will fail to detect the effects of many GWAS loci on transcription . We identified eQTLs with opposing directions of effect between cell types , including eQTLs for CD52 and CD101 . Soluble CD52 inhibits activated T cells through SIGLEC-10 [27] , while CD101 is an inhibitor of T cell proliferation . This implies that a single genetic variant with contrasting effects on gene expression in different cell types could result in complex downstream impacts on immunity . eQTLs with discordant effects between cell types tended to have smaller effect sizes than eQTLs where the direction of effect was consistent ( S8–S11 Figs ) . We suspect this general pattern is biologically meaningful . eQTLs with large effect sizes that are concordant between cell types may reflect a different underlying regulatory mechanism to the more subtle regulatory variants that can have different directions of effect in different contexts . Our eQTL analysis provides new insights into disease-associated genetic variants . At 34 IBD-associated loci we identified eQTLs that were not previously apparent from eQTL database mining [17] , highlighting the power of studying multiple leucocyte subsets from patients with active inflammatory disease . This eQTL study not only helps prioritise candidate genes , but also provides information on whether disease-associated SNPs up- or down-regulate these genes . This is important because it may challenge current assumptions about the roles such candidate genes play in pathogenesis . For example , our data indicate that higher FADS2 expression may increase IBD susceptibility , which was not apparent from the phenotype of FADS2 knock out mice . This study reveals eQTLs present only in the context of inflammatory disease . For example , we identified a SNP , rs11230584 , that is associated with both CD5 and CD6 expression in patients with active IBD , but not in healthy individuals . CD5 and CD6 are both transmembrane glycoproteins expressed on T cells and on the B1 subset of B cells . CD5 acts a negative regulator of T cell receptor signalling [28] . In addition , CD5 expression activates multiple intracellular pathways , leading to the production of the type II cytokines IL-5 , IL-10 and IL-13 . CD6 is involved in the immunological synapse acting as a T cell co-stimulatory molecule , and anti-CD6 monoclonal antibodies have been used in clinical trials for the treatment of psoriasis [29] . Accordingly , this eQTL may result in genotype-dependent differences in immune responses , and perhaps even in responses to anti-CD6 therapy . In addition to identifying eQTLs dependent on disease state , we also show that eQTLs present in active inflammatory disease can disappear following treatment . In order to identify inflammatory disease-dependent eQTLs in a statistically rigorous manner , we used a linear model with a genotype × disease ( G×D ) interaction term . The statistical power to detect interaction effects is lower than for so-called ‘main’ effects ( in this study , the marginal effect of genotype on expression ) . A limitation of this study was the relatively modest sample sizes , and we highlight that we used a 15% FDR for the G×D interaction analysis compared to a 5% FDR throughout the rest of the analyses . Because our power to detect G×D effects was limited , it is likely that a higher proportion of eQTLs identified in our analyses of the IBD and AAV cohorts are disease-specific than is apparent from the results of the interaction analysis . Of note , the proportion of neutrophil-specific eQTLs in the AAV analysis was over twice that in the joint IBD-HV analysis . Given the prominent neutrophil turnover and activation seen in AAV , we hypothesise that this observation may be driven by AAV-specific eQTLs in neutrophils . Whilst eQTLs present only after in vitro activation of immune cells have been described [13–15] , such experiments are unphysiological and cannot fully recapitulate the complexity of cellular activation in vivo . The eQTLs we identified , using relevant tissues in the context of in vivo inflammation , are important to the appropriate interpretation of the functional effects of SNPs identified through GWAS . Clinical outcomes vary considerably between patients with the same autoimmune disease , with some individuals experiencing a relatively indolent course , while others suffer from more aggressive disease . The majority of genetic studies to date have focussed on identifying genetic variants that predispose to disease , but candidate SNP studies have revealed differential outcomes according to genotype in Crohn’s disease , rheumatoid arthritis , malaria and tuberculous meningitis [30 , 31] . Genetic variants that drive outcome once disease is established may only act as eQTLs in patients with active inflammatory disease , and not in healthy individuals . Therefore inflammatory-disease specific eQTLs , such as the eQTL for CD5 and CD6 , may be important for understanding the heterogeneity of clinical course in autoimmune and infectious diseases . In summary , we reveal that the consequences of genetic variation on gene expression are complex , and may vary between cell types , between health and inflammatory disease , and before and after treatment . This study , using parallel eQTL mapping in a broad panel of relevant cell types from both healthy individuals and patients with active inflammatory disease , provides increased understanding of the biology underlying the genetic basis of immune-mediated diseases . Ethical approval for this work was obtained from the Cambridgeshire Regional Ethics Committee ( REC08/H0306/21 ) . All participants provided written informed consent . We recruited a cohort of adult healthy volunteers , free from autoimmune disease or any other chronic condition and not taking any regular medication at the time of enrolment . Adult patients with active inflammatory bowel disease ( Crohn’s disease or ulcerative colitis ) were recruited from a specialist inflammatory bowel disease clinic at Addenbrooke’s hospital in Cambridge , UK , prior to commencing treatment , as described previously [32] . Diagnosis was made using standard endoscopic , histopathological , and radiological criteria . Patients receiving immunomodulators or corticosteroids were excluded to avoid potential confounding effects on gene expression . Patients with a diagnosis of ANCA-associated vasculitis ( either granulomatosis with polyangiitis , formerly Wegener’s granulomatosis , or microscopic polyangiitis , but not eosinophilic granulomatosis with polyangiitis , formerly Churg-Strauss syndrome ) and evidence of active disease were recruited from a specialist clinic at Addenbrooke’s Hospital . Active disease was defined by Birmingham Vasculitis Activity Score ( BVAS ) , elevated acute phase markers , and the clinical assessment that induction immunosuppression would be required to manage disease , as described previously [33] . 27 of 46 AAV patients were venesected at the time of first clinical presentation prior to treatment . The remainder had clearly defined flares of previously quiescent disease on stable maintenance therapy , and required treatment with induction immunosuppression . See S5 Table for more detailed demographics . Cell separations were performed using positive selection with magnetic-activated cell sorting ( MACS ) as previously described [34] . Blood samples ( 100 ml ) were collected into 4% sodium citrate and PBMC isolated by centrifugation over Histopaque 1077 ( Sigma ) . The PBMC sample was split into two aliquots and CD14 monocytes were isolated from one aliquot and CD19 B cells from the other by magnetic cell sorting using CD14 and CD19 microbeads ( Miltenyi Biotec ) according to the manufacturer’s instructions . CD4 and CD8 T cells were then isolated from the CD14 and CD19 negative fractions , respectively , by magnetic cell sorting using CD4 and CD8 microbeads as described by the manufacturer . CD16 neutrophils were obtained from the red cell/granulocyte pellet by red cell lysis followed by positive selection using CD16 microbeads as described by the manufacturer . RNA was extracted from cell lysates using either Qiagen RNEasy Mini Kits or Qiagen Allprep kits according to the manufacturer’s instructions . RNA quality was assessed using an Agilent Bioanalyser 2100 ( Agilent Technologies ) and quantified by spectrophotometry using a NanoDrop ND-1000 spectrophotometer ( Thermo Scientific ) . Genomic DNA was extracted using Qiagen Allprep kits according to the manufacturer’s instructions and quantified as above . Patients with IBD and healthy controls were genotyped using the Illumina Human OmniExpress12v1 . 0 BeadChip at the Wellcome Trust Sanger Institute according to the manufacturer’s protocol . After genotype calling with Illumina GenCall software , SNP data was processed in the following sequence using PLINK [35] . Samples with a sex mismatch , abnormal heterozygosity , proportion of missing SNPs >5% , and duplicated or related samples ( identified using Identity by State ) were removed . SNPs with missing calls >5% , extreme deviation from Hardy-Weinberg Equilibrium ( p-value <1×10-8 ) , or which were monomorphic were removed . Finally , principal components analysis of the post-QC genotype calls combined with calls from HapMap3 founder individuals ( downloaded from http://hapmap . ncbi . nlm . nih . gov/ ) , using the R/Bioconductor snpStats package , confirmed all samples were of European ancestry ( S18 Fig ) . Genotype data was read into R version 3 . 02 ( http://www . R-project . org/ ) using the snpStats package , and stored as a SnpMatrix object . SNPs with minor allele frequency ( MAF ) <5% were excluded from eQTL analysis , leaving 626 , 858 autosomal SNPs . These data have been deposited in the European Genome-phenome Archive ( EGA ) and are available on request ( EGAS00001001251 ) . Genotype data for AAV patients were available from a previous GWAS [36] . Briefly , genotyping was performed by AROS Applied Biotechnology ( Aarhus , Denmark ) using the Affymetrix SNP 6 . 0 Platform . SNP genotypes were called using the CRLMM algorithm [37] , and standard QC performed as described previously [36] . 200 ng RNA was processed for hybridization onto Affymetrix Human Gene ST 1 . 1 microarrays , according to the manufacturer’s instructions , prior to scanning . Microarrays were processed with the automated GeneTitan instrument which uses a 96-well plate . Chip probe intensity files ( . CEL ) were read into R version 3 . 02 using the Bioconductor oligo package [38] . Raw intensity values were pre-processed with the Robust Multi-array Average ( RMA ) algorithm which performs background correction , quantile normalization , log2 transformation and summarization to gene level via a median polish . Probe annotation was obtained from the pd . hugene . 1 . 1 . st . v1 Bioconductor package . The resulting expression matrix and phenotype data was stored as an ExpressionSet object . The raw intensities for the expression data for each cell type were read in and normalised separately . The arrayQualityMetrics package was run on the normalised expression matrices before and after batch correction with ComBat ( see below ) and poor quality samples removed . Microarray data are available in the ArrayExpress database ( www . ebi . ac . uk/arrayexpress ) under accession number E-MTAB-3554 . Principal components analysis of the expression matrices indicated batch effects . To address this , we used two alternative approaches . The first approach was removal of unwanted expression heterogeneity by adjustment for latent factors using Probabilistic Estimation of Expression Residuals ( PEER ) [39] . The use of PEER has increased eQTL discovery compared to standard methods in a number of studies [4 , 40] . PEER can model both known confounders and hidden latent factors in the expression data . We ran PEER specifying batch and sex as known confounders , and used the expression residuals as the dependent variable in the subsequent eQTL scan . For the analysis of cell-type specificity , comparison of direction of effect between cell types , and for the intersection of eQTLs with GWAS SNPs , we used eQTLs detected from association scans using PEER residuals . Alternatively , we performed batch correction using the ComBat function from the sva ( Surrogate Variable Analysis ) Bioconductor package [41] . ComBat was implemented specifying diagnosis as a covariate of biological interest whose effect should be preserved . This method was used for the genotype × disease interaction analysis and the pre- and post-treatment analysis . The GGtools Bioconductor package [42] was used to combine SnpMatrix and ExpressionSet objects for patients for whom we had both good quality genotype and expression data to create an smlSet object . Table 1 shows the numbers of samples available for analysis by disease and cell type . For the Bayesian joint analysis across cell types presented in Fig 1 , we used all available samples , even if this meant differing sample sizes for each cell type . For the estimates of eQTL cell-type specificity from one-at-a-time cell-type analysis , we confined eQTL mapping to individuals with expression data available for all cell types to ensure equal power for each cell type . In the latter scenario , subsetting was performed after pre-processing and adjustment for latent factors or batch correction . In order to make direct comparisons with results from separate cell-type analysis , we also used the Bayesian method on the subset of individuals with expression data available for all cell types ( S19 Fig ) . For the detection of eQTLs to intersect with GWAS SNPs , we used all available samples to maximise power , even if this meant differing sample sizes for each cell type . The expression versus genotype boxplots in Figs 2 , 3 , 6 , 7 and 8 show all individuals for whom we have good quality expression and genotype data rather than restricting to the subset of individuals with expression data available for all cell types . Filtering of probesets in the expression data was performed prior to the eQTL scan . Filtering of expression data has been shown to increase power to detect differentially expressed genes , and should not bias FDR estimates as long as the variable of interest is not used in the filtering step [43] . Probesets on sex chromosomes and those which did not map to an Entrez gene ID were removed . The remaining probesets were then filtered by variance ( Bioconductor genefilter package ) . As well as increasing power , filtering by variance has two additional benefits . First , genes which are expressed homogeneously are not informative in an eQTL analysis . Second , unexpressed genes will show little variability between samples and thus filtering on the basis of lack of expression variability is a reliable method for detecting and excluding unexpressed genes from further analysis [44] . In the downstream analysis we compared eQTLs detected in each cell type . In order make these comparisons valid , it was necessary to perform eQTL mapping on the same filtered list of probesets in each cell type . To achieve this , we retained any gene whose variance across samples was in the upper quartile in any one of the cell types . For the joint IBD-HV analysis , this resulted in 9 , 041 probesets when using all available samples , and 9 , 242 when restricting to the 65 individuals with expression data available for all cell types . For the AAV analysis , the same filtering procedure resulted in 8 , 916 probesets when using all available samples , and 8 , 979 when restricting analysis to the 33 patients with complete expression data available for CD4 T cells , CD8 T cells , monocytes and neutrophils . This analysis was performed using the IBD-HV dataset , restricted to the individuals with expression data available for all 5 cell types ( n = 65 ) . To identify probesets targeting robustly expressed genes in each cell type , the following procedure was performed . First we calculated the mean expression of each probeset across all samples from a given cell type ( using expression data that had been RMA-normalised and then batch-corrected with ComBat ) . To determine thresholds for declaring probesets as expressed or non-expressed in that cell type , we used the normalmixEM function from R package mixtools to fit a mixture of two Gaussian distributions to the vector of mean probeset expression values . Starting values for the expectation–maximization ( EM ) algorithm were chosen assuming 50% of expressed probesets , and the initial mean values of the underlying Gaussian distributions were set equal to the first and third quartile of the probeset expression distribution . The distribution with the lowest mean expression was considered as that corresponding to non-expressed genes . A detection threshold was set by taking the 95th percentile of this distribution . Probesets whose mean expression was above the detection threshold were considered to be expressed in that cell type . Subsequently , probesets that did not map to an Entrez gene ID were removed . The lists of probesets expressed in each cell type were then intersected to find those expressed in all cell types . Probesets on the sex chromosomes , and those which did not map uniquely to an Entrez ID were removed . This resulted in 5 , 186 probesets . The matrix of PEER-adjusted expression data was then subsetted to retain only these probesets , and the eQTL scan was performed using the residuals from PEER as the dependent variable . We performed cis eQTL mapping with the All . cis function in the GGtools Bioconductor package . Previous work has suggested that the most cis eQTLs are approximately centred around the transcribed region , with few eQTLs lying more than 50kB from the gene [6 , 7 , 45] . For each gene , we defined cis SNPs as those lying in the region spanning 100kB up and downstream from the gene start and end positions respectively . SNP location information was obtained using the Bioconductor package SNPlocs . Hsapiens . dbSNP . 20120608 ( dbSNP Build 137 ) . Sensitivity analysis using varying distances to define cis found no increase in eQTL discovery for cis distances greater than 100 kB using a 5% FDR significance threshold . The All . cis function calculates association statistics ( score tests ) for each cis SNP-transcript pair . Specifically , the function fits a generalized linear model with transcript abundance as the continuous dependent variable and genotype as the predictor variable . Optionally , potential confounders of an expression-genotype association can be included as covariates . Genotype is a discrete variable coded 0 , 1 , or 2 according to allelic dose ( i . e . homozygous SNP genotypes are coded 0 or 2 and heterozygous genotypes are coded 1 ) . For each gene , each SNP in the cis region was tested in turn for association with mRNA abundance ( after adjustment with PEER ) using an additive genetic model; after the base model was fitted ( i . e . a model with an intercept and any specified covariates ) , a score test was then performed for addition of genotype to the model . The tests used are asymptotic chi-squared tests based on the vector of first and second derivatives of the log-likelihood with respect to the parameters of the additive model . Where the dataset comprised more than one disease grouping ( i . e . IBD and healthy individuals ) , the diagnostic grouping was included as an additional covariate . We did not include sex as a covariate , as sex had already been specified as a known confounder when running PEER . GGtools provides estimates of the FDR through estimation of the distribution of statistics under the null by breaking expression-genotype relationships through permutation of sample labels ( the ‘plug-in FDR’ method [46]- see S1 Text ) . A 5% FDR was used to declare statistical significance . To evaluate eQTL cell-type specificity , we used the eQTLBMA package which implements the method described by Flutre et al [16] . We ran this according to the instructions in the user manual . Briefly , we first transformed the expression level of each gene into the quantiles of a standard Normal distribution . Ties were broken randomly . We then ran the eqtlbma_bf programme , to compute the Bayes factors assessing the support in the data for each probeset-SNP pair being an eQTL . We analysed the IBD patient and HV data together , specifying disease status as a covariate . To maximise power , we used all available samples ( from 91 IBD patients and 43 HVs ) , even if this meant that for some individuals we did not have full expression data for all cell types . In this scenario we used the option --error hybrid . Samples from AAV patients were analysed separately to avoid difficulties that might arise from combining genotype data from a separate platform . Again , we used all available samples . The output file containing the Bayes factors was then fed into the eqtl_hm programme , to fit the hierarchical model with an EM algorithm , and get to maximum-likelihood estimates of hyper-parameters and the configuration probabilities . To obtain the posterior probabilities , we need an estimate of the probability for a gene to have no eQTL in any tissue , π0 . To do this , we implemented the EBF procedure , using the gene-level Bayes factors averaged over the grid and configuration weights ( estimated via the EM algorithm ) . Finally we ran eqtlbma_avg_bfs to obtain ( i ) the posterior probability ( PP ) for the gene to have an eQTL in at least one cell type , ( ii ) the PP for a SNP to be “the” eQTL , assuming one eQTL per gene , ( iii ) the PP for the SNP to be an eQTL , ( iv ) the PP for the eQTL to be active in a given tissue , and ( v ) the PP for the eQTL to be active in a given configuration . We used a PP corresponding to a 5% Bayes FDR as the significance threshold . To evaluate estimates of eQTL cell-type specificity from one-at-a-time cell-type analysis , we limited the IBD-HV analysis to 65 individuals ( 47 IBD patients , 18 HVs ) in whom samples passing expression quality control were available for all five cell types . By using the same set of individuals , the sample size and genotype matrix ( predictor variables ) were identical for each cell type . In order to make a fair comparison between the results from the one-at-a-time cell-type analysis and those from eQTLBMA , we re-ran eQTLBMA on these same 65 individuals with full expression data for all 5 cell types . In this latter scenario we used the option --error mvlr ( S19 Fig and S1 Text ) . For all possible pairings of the cell types , we took those eSNP-gene associations that were statistically significant in both members of the pair in the one-at-a-time cell-type analyses ( FDR <0 . 05 ) , and compared the direction of effect on expression in each . We did this by plotting the estimated coefficient ( beta ) for the genotype term in the first cell type against that in the second . The betas were estimated from regressing PEER-adjusted expression on genotype using the lm function in R . For the IBD-HV analysis , disease status was included as a covariate . eQTLs with opposing directions of effect between cell types have a positive beta in one cell type but a negative beta in the other ( see S1 Text for more detailed discussion ) . To identify eQTLs influenced by the presence or absence of inflammatory disease we analysed the joint IBD-HV dataset using a linear model with a genotype × disease interaction term . Expression data for each cell type was RMA normalised separately , but within each cell type expression data from healthy individuals and IBD patients were normalised together . We used expression data that had been batch-corrected with ComBat , as adjustment for latent factors with PEER might remove genuine disease effects . We again restricted testing for each gene to cis SNPs . For each gene-SNP pair , a linear model was fitted using expression as the dependent variable , and genotype ( coded 0 , 1 , or 2 ) , disease status ( coded 0 and 1 for healthy and IBD respectively ) , and a genotype × disease interaction term as the predictor variables . This was performed in R using the lm function . Fitting a model with an interaction term is statistically more robust than the naïve approach of separate analysis of IBD and HV cohorts , followed by comparison of the resulting lists to find eQTLs common to both groups , and those identified only in IBD or only in HVs . The latter approach would result in an excess of false declarations of group-specific eQTLs . In particular , the comparison of lists approach is unduly influenced by the difference in statistical power to detect eQTLs in each group when the group sample sizes are different . The IBD group was larger than the HV group , and so there was greater power to detect eQTLs in the former . This would have led to many false declarations of IBD-specific eQTLs had we used the comparison of lists approach ( see S1 Text for more detailed discussion ) . The analysis was performed for CD4 T cells , CD8 T cells , monocytes and neutrophils . B cells were not analysed due to the low number of samples available from healthy individuals . For each cell type , we used all available IBD and HV samples ( see Table 1 for sample sizes ) . We limited testing of SNPs to those with MAF of 10% or higher , as power decreases with decreasing MAF . Probesets were filtered in the following order: first , removal of probesets that were non/lowly expressed in that cell type , and those which did not map to an Entrez gene ID; second , filtering of the remaining probesets by variance to retain only those in the upper 40% . Interaction effects are harder to detect than main effects , so to improve our power we employed a ‘two-step’ procedure adapted from methods used to detect genotype-environment interactions in GWAS [47] . ‘Two-step’ procedures aim to reduce the multiplicity burden associated with having a large number of variables to explore by applying some filter ( step 1 ) on the number of variables actually tested ( step 2 ) . In order to prevent biasing of FWER/FDR estimates , the filtering procedure used in step 1 should be independent of the final test statistics in step 2 . In step 1 we performed linear regression of expression on genotype ( main effect of genotype ) , with no covariates . Importantly , by not including disease status as a covariate , we ensured independence of the step 2 statistics [47] . SNP-gene pairs passing the step 1 threshold α1 were eligible for step 2 testing . We used a p-value of 5×10-5 for α1 . A more liberal α1 allows more SNP-gene pairs through to step 2 , but at the price of more multiple testing and a more stringent threshold in step 2 . In addition , because there were multiple correlated SNPs due to LD , where multiple SNPs were significantly associated with the same gene at the threshold α1 , only the most significant SNP per gene was taken forward to step 2 . In step 2 , we fitted the full model with regression of expression on genotype , disease and a genotype × disease interaction term . P-values for the interaction terms were then adjusted for the number of tests performed in step 2 with the p . adjust function in R using the Benjamini-Hochberg procedure , and a 0 . 15 FDR threshold used to declare significance . Q-values were also calculated . Each cell type was analysed separately . We used eQTLBMA to jointly analyse expression data from the 3 timepoints: baseline , 3 months , and 12 months . We used we used the option --error hybrid as the individuals in each subgroup were overlapping but not identical . For genes with a significant eQTL ( 5% Bayes FDR ) , we took the SNP with the highest posterior probability for being the eQTL . For this list of SNP-gene associations , we then assessed the posterior probabilities for each of the 23 − 1 = 7 possible configurations . To identify the effects of disease-associated SNPs on gene expression , and how this varies between cell types , we examined the overlap of SNPs associated with complex traits from the NIH National Human Genome Research Institute GWAS catalogue ( http://www . genome . gov/gwastudies/ ) ( download date 20 May 2015 ) with eQTL SNPs ( eSNPs ) discovered in each cell type in our analysis . The SNPs identified in eQTL studies or GWAS are not necessarily themselves causal variants , but instead may be in LD with them . Therefore GWAS tag SNPs , and proxy SNPs in high LD with them ( r2 >0 . 8 ) , were cross-referenced with cis eQTLs found in each cell type . SNAP was used to identify proxy SNPs ( http://www . broadinstitute . org/mpg/snap/ ) , using the 1000 Genomes ( http://www . 1000genomes . org/about ) pilot 1 data ( CEU individuals ) as the reference population . To define the eSNPs for cross-referencing , we repeated the analysis using two levels of stringency for declaring colocalisation of eQTL and GWAS signals . For the basic criteria , a GWAS signal was declared to colocalise with the eQTL if the GWAS SNP or one of its proxies was a significant eQTL ( FDR <0 . 05 ) . For the more stringent criteria , in addition to the basic criteria , the GWAS SNP or one of its proxies had to be the most significant cis eSNP for that gene ( S1 Text , S16 Fig ) . For the detailed examination of the effects of IBD-associated SNPs on gene expression , we took SNPs identified in the IBD GWAS meta-analysis [16] as hits in either Crohn’s disease ( CD ) , ulcerative colitis ( UC ) or both ( IBD ) ( Supplementary Table 2 of that paper ) , and cross-referenced these SNPs and proxies ( r2 >0 . 8 ) with the eQTL SNPs from our analysis as above ( see S1 Text ) . Colocalisation testing was performed using the coloc package [21] . We used the publically available summary statistics for IBD , CD and UC GWAS meta-analysis by Jostins et al [17] along with our eQTL data . 1 ) Expression data has been deposited into ArrayExpress ( accession number E-MTAB-3554 ) . 2 ) Genotype data has been deposited into the EGA ( study accession number EGAS00001001251 , url https://www . ebi . ac . uk/ega/studies/EGAS00001001251 ) . Controlled access for scientific researchers is available via requests to the data access committee ( EGAC00001000338 ) . 3 ) Summary statistics for all significant eQTLs for each cell type and dataset are included as a compressed tarball ( S1 Data ) .	The human immune system has evolved to protect us from infection and cancer , whilst avoiding damage to healthy tissue . If this complex system goes wrong , immune cells may cause inappropriate inflammation and damage , resulting in clinical disease . Examples include inflammatory bowel disease and autoimmune vasculitis , characterised by inflammation in the gut and blood vessels respectively . Genetic studies have identified many variants in our DNA code that predispose to such immune-mediated diseases . The majority of these variants lie outside protein-coding regions , and so how they influence disease risk remains largely unclear . Examining how genetic variants affect gene expression can help bridge this gap in our knowledge , but these effects are highly dependent on the cellular or environmental context such as tissue type or cellular activation status . We investigated the genetic control of gene expression in five white blood cell subtypes taken from patients with active inflammatory bowel disease and autoimmune vasculitis , and from healthy controls . We report the novel observation of distinct variants that only affect gene expression in patients with active inflammatory disease , and show that these effects can disappear following treatment . These findings provide new insights into the genetic basis of important immune-mediated diseases .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "blood", "cells", "t", "helper", "cells", "genome-wide", "association", "studies", "medicine", "and", "health", "sciences", "immune", "cells", "immunology", "gastroenterology", "and", "hepatology", "genome", "analysis", "inflammatory", "bowel", "disease", "cytotoxic", "t", "cells", "neutrophils", "white", "blood", "cells", "animal", "cells", "gene", "expression", "t", "cells", "cell", "biology", "monocytes", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "genomics", "computational", "biology", "human", "genetics" ]	2016	Insight into Genotype-Phenotype Associations through eQTL Mapping in Multiple Cell Types in Health and Immune-Mediated Disease
The recent development of whole genome association studies has lead to the robust identification of several loci involved in different common human diseases . Interestingly , some of the strongest signals of association observed in these studies arise from non-coding regions located in very large introns or far away from any annotated genes , raising the possibility that these regions are involved in the etiology of the disease through some unidentified regulatory mechanisms . These findings highlight the importance of better understanding the mechanisms leading to inter-individual differences in gene expression in humans . Most of the existing approaches developed to identify common regulatory polymorphisms are based on linkage/association mapping of gene expression to genotypes . However , these methods have some limitations , notably their cost and the requirement of extensive genotyping information from all the individuals studied which limits their applications to a specific cohort or tissue . Here we describe a robust and high-throughput method to directly measure differences in allelic expression for a large number of genes using the Illumina Allele-Specific Expression BeadArray platform and quantitative sequencing of RT-PCR products . We show that this approach allows reliable identification of differences in the relative expression of the two alleles larger than 1 . 5-fold ( i . e . , deviations of the allelic ratio larger than 60∶40 ) and offers several advantages over the mapping of total gene expression , particularly for studying humans or outbred populations . Our analysis of more than 80 individuals for 2 , 968 SNPs located in 1 , 380 genes confirms that differential allelic expression is a widespread phenomenon affecting the expression of 20% of human genes and shows that our method successfully captures expression differences resulting from both genetic and epigenetic cis-acting mechanisms . Understanding the genetic causes of phenotypic variation in humans still remains a major challenge for human genetics . In hundreds of cases , a single DNA sequence polymorphism affecting a protein coding sequence has been linked to a clear simple Mendelian phenotype ( see e . g . [1] ) and , for a much smaller but increasing number of cases , to more complex phenotypes [2]–[4] . Recent developments in high-density genotyping technologies have led to the completion of several whole genome association studies that test hundreds of thousands of markers for a specific disease . While earlier studies essentially focused on variants in coding sequences and regions immediately surrounding candidate genes , whole genome scans interrogate , in an unbiased way , most of the human genome including large regions of non-coding DNA that had not been studied previously . Interestingly , some of the strongest signals observed in these association studies are located in non-coding regions , either in large introns ( e . g . [5]–[7] ) or far away from any annotated loci ( e . g . [8] and references therein ) . The mechanisms connecting these polymorphisms to the etiology of the diseases are still unclear but regulation of gene expression remains an obvious candidate . It is thus becoming particularly important to have a powerful and reliable method to easily test the influence of DNA polymorphisms on gene expression . One of the approaches commonly used to identify regulatory polymorphisms is to look for statistical associations between variation in gene expression and individual genotypes [9] , [10] . This method offers the advantage of simultaneously analyzing thousands of genes using gene expression arrays and has yielded fascinating results in yeast [11] , [12] and mouse [13]–[16] . Its application in humans [17]–[24] suffers from relatively low statistical power due to potential inter-individual differences in a large number of causal variants involved in the regulation of a specific gene [25] , their modest effects and the burden of the multiple testing correction necessary to take into account the large number of independent tests performed . In addition , since this approach requires extensive genotype information for all individuals , it is costly to apply to new samples . An alternative approach is to compare the relative expression of the two alleles in one individual: the effect of a polymorphism affecting in cis the regulation of a particular transcript can be detected by measuring the relative expression of the two alleles in heterozygous individuals using a transcribed SNP as a marker [26]–[32] . Several studies have used this approach in humans but have been criticized for their low throughput or the apparent high variability . Here we describe a novel array-based method that allows high-throughput assessment of differential allelic expression . We used a modified version of the Illumina GoldenGate genotyping platform , the Allele-Specific Expression ( ASE ) assay , to assess the extent of differential allelic expression for over 1300 genes in more than 80 human lymphoblastoid cell lines ( LCLs ) . Our analyses include 352 genes located in ENCODE regions and chromosome 21 that have been previously screened for cis-regulatory polymorphisms using total gene expression [19] . This allows us to directly compare the advantages and drawbacks of the two approaches in terms of range , sensitivity and robustness . We specifically address the issue of experimental noise and reproducibility of the findings and show that biology , not experimental variability , is responsible for the patterns observed . We discuss the relevance of our results for the identification of the molecular mechanisms regulating gene expression , as well as their implications for future genetic studies . We first assessed the extent of differential allelic expression at 1 , 432 exonic SNPs using 81 individual LCLs with the Illumina ASE technology ( Figure 1 and Table S1 for the composition of the Illumina ASE Cancer Panel ) . This technology uses primer extension assays with fluorescence-labeled allele-specific primers to measure the proportion of each allele separately at the genomic and transcriptomic levels ( Figure 2 ) . Five hundred and twelve SNPs ( in 345 genes ) displayed an expression level significantly higher than background in at least three heterozygous individuals and were further investigated ( see Materials and Methods for details ) . The extent of differential allelic expression at each SNP was obtained by comparing the relative amount of each allele in RNA to the ratio observed in DNA . As a first effort to determine if the assay could reliably be used to assess differential expression we generated spike mixes using varying proportions of total RNA extracts from two individuals . For 20 exonic SNPs located in expressed transcript , the two individuals are homozygous for the different alleles ( i . e . respectively AA and BB ) , while for 192 SNPs one individual is heterozygous and the other homozygous ( i . e . either AB and AA , or AB and BB ) . Since the expression of each gene may differ between the two individuals , one does not expect to observe an exact translation of the proportions of total RNA mixed to the “allelic” expression level . However , the allelic expression differences estimated for the different spike mixes should be the proportional to each other . For all homozygous/homozygous mixes ( 20 out of 20 SNPs ) and 83% of the heterozygous/homozygous mixes ( 159 out of 192 ) , we observed a significant linear correlation ( p<0 . 05 ) between the proportion of mixed total RNAs and the “allele-specific expression” estimated by the assay ( Figure 3 and Figure S1 ) . We then tried to assess the threshold above which differential allelic expression would be genuine: even if the two alleles are equally expressed in one individual , we expect the ratio of allelic expression measured at a given marker to deviate stochastically from 50∶50 due to experimental variability . In order to differentiate technical noise from biological signal ( i . e . the differences in allelic expression due to differential cis-acting regulation ) , we evaluated the extent of experimental variability in the assay by comparing independent estimates of allelic imbalance for duplicates of individual RNA . We used duplicated measurements from 81 individuals at all SNPs expressed to determine a robust estimation of the experimental variability ( N = 31 , 503 duplicates ) . After averaging duplicate differences for each SNP over all individuals , we observed than less than 3% of the SNPs show a population average variability greater than 10% ( see Materials and Methods for more details and Figure S2 ) . This level of experimental variability corresponds to a ratio of allelic expression of 60∶40 ( i . e . 1 . 5-fold difference ) . Thus , population-average allelic expression ratio at any SNP lower than 60∶40 can be explained by experimental noise , while a SNP displaying a population-average differential allelic expression greater than this threshold most likely reflects a biological process affecting cis-regulation . We are interested in the present study in identifying loci with common allelic expression differences and we thus focused on population-average differential allelic expression: the average over all heterozygous individuals of the extent of allelic expression differences , regardless of which allele is over expressed ( this is addressed later ) . The identification of a single individual with dramatic allelic expression difference is also possible using the same approach ( but a different detection cut-off ) but is beyond the scope of this paper . Among the 345 genes expressed in this first panel ( 512 SNPs ) , 72 ( 87 SNPs ) displayed an average level of allelic imbalance larger than this 40∶60 cut-off and were thus considered to display significant differences in allelic expression ( Figure 4 ) . These analyses rely on the observation of the three genotypes in the population ( i . e . AA , AB and BB ) . To also include SNPs with a lower minor allele frequency for each it was not possible to observe homozygotes for the minor allele in our small sample , we designed a second analysis method using solely the heterozygous individuals . If the alleles are differentially regulated we expect to observe in some cases a very large variance in the ratio of the two alleles in the population . We used this approach to determine SNPs for which the heterozygous individuals harbor a variance of the allelic ration higher than expected using a Maximum expectation algorithm ( see Materials and Methods for details ) . This approach does not allow us to quantity the overall extent of differential allelic expression but identifies 8 genes with differential allelic expression that were not identified by the previous method . When one considers the estimates of allelic expression obtained using different SNPs in a same transcript showing significant differences in allele expression ( i . e . with a population-average ratio greater than 60∶40 ) , we note that 36 out of 44 correlations between individuals estimates are significant ( for an average r2 of 0 . 83 ) . Individuals showing a large allelic expression difference at one SNP display similar patterns at all heterozygous positions of the transcript ( an example is shown on Figure 5 ) . This observation supports our findings that the experimental variability is low in the Illumina ASE assay and that this assay allows quantitative assessment of differential allelic expression . Consequently , the population-average estimates of allelic imbalance obtained with different markers in the same transcript tend to be similar ( Table S2 ) but can vary since different individuals will be included in the average ( depending on whether they are heterozygotes at this marker ) . To further assess the validity of our results , we randomly selected 25 genes tested on the Illumina ASE platform and used quantitative sequencing of RT-PCR products [33] to measure allelic imbalance for the same SNP in the same individuals ( Figure S3 ) . The selected genes consisted of eight autosomal genes with significant allelic imbalance and 17 genes for which the level of differential allelic expression did not reach our significance threshold . We analyzed the same 81 individual LCLs using RNA from the same extract as for the Illumina assay . Overall , we observed a strong correlation between the estimates of allelic imbalance obtained for each individual using the two methods for the genes with a ratio of allelic expression larger than our 40∶60 cut-off ( r2>0 . 8 for 6 out of 8 genes , see Figure 6A as a example ) . The correlations were not statistically significant for the genes for which the average difference in allelic imbalance did not reach our significance threshold ( 16 out of 17 genes , Figure 6B ) : minor deviations observed in the allelic ratio for these genes likely correspond to random variations and are therefore not expected to be reproducible . The strength of the correlation ( measured by Pearson's r2 ) for all 25 genes is shown on Figure 4 . One SNP in CD44 ( rs8193 ) displayed a low but significant correlation between our estimates of allelic imbalance obtained from the Illumina assay and those using quantitative sequencing ( p<10−4 , r2 = 0 . 4075 ) , even though the average level of allelic imbalance was below our significance cut-off on the Illumina platform . Allelic imbalance at CD44 has been previously reported [30] and it is likely that the signal observed at that gene is real but corresponds to a low level of differential expression . Two genes ( ABL2 , XRCC1 ) showed significant allelic imbalance in the Illumina ASE assay ( with a mean allelic ratio of , respectively , 70∶30 and 65∶35 ) but were not validated by quantitative sequencing . Manual inspection of the Illumina results for these genes revealed that the allelic expression ratios were estimated using a small number of homozygotes for the minor allele ( respectively , 1 and 2 individuals ) which led to an incorrect estimation of the expected dye ratio for heterozygotes and to a general over-estimation of allelic imbalance . For further analyses , we manually curated the list of all genes with significant differential allelic expression to remove potential false positives due to low number of homozygous individuals . Our study uses lymphoblastoid cell lines ( LCLs ) and it remains controversial whether culture conditions could artefactually generate differential allelic expression . We therefore tested whether allelic imbalance is influenced by harvesting the cells after different numbers of passages . This allowed us to control the effect of changes in the culture environment including pH , nutrient concentration and cell density at the time of harvest . We compared our estimation of allelic imbalance for three genes with significant population-average allelic imbalance in 47 individuals using recently thawed LCLs harvested after the 2nd , 4th and 6th successive passages ( respectively , “growths” 1 , 2 and 3 ) . The correlations between the allelic imbalance estimations are displayed in Figure S4 for the comparison of growths 2 and 3 . The estimations of allelic imbalance after different passages were very similar to each other ( r2>0 . 9 ) , supporting the idea that differential allelic expression is little influenced by variations in culture environment . An alternative approach to identifying cis-regulatory polymorphisms is to test for statistical association in a population between total gene expression measurements and the genotypes at markers in or surrounding the transcript . Interestingly , one of the genes showing the most marked difference in allele expression in our analysis is one of the 14 genes identified by Cheung and colleagues in a previous genome-wide study [17] . One comprehensive analysis was recently conducted for 512 RefSeq genes in ENCODE regions and chromosome 21 using LCLs from 60 unrelated individuals genotyped by the HapMap project [19] . In order to compare the respective strengths and weaknesses of total gene expression mapping and differential allelic expression , we designed a second panel that includes SNPs in the same genomic regions to analyze the same individual LCLs ( Figure 1 and Table S1 for details ) . Using the information from the HapMap phase I ( release 16 ) to select common exonic SNPs , we were able to include 228 and 124 genes from , respectively , ENCODE regions and chromosome 21 , while Stranger and colleagues selected 321 and 191 genes ( after screening for genes with high variance in their expression among individuals , see Materials and Methods for details ) . From the regions analyzed by Stranger and al . , two-hundreds and ninety SNPs ( in 170 genes ) showed an expression level significantly higher than the background in three or more heterozygous individuals and were further investigated . Forty-nine out of 170 genes show significant level of differential allelic expression including 6 out of the 21 genes identified by Stranger and colleagues and present in our panel . Additionally , TTC3 which shows significant association between total gene expression and genotypes in the study by Stranger et al . shows patterns of allelic expression consistent with differential allelic expression ( including a very high correlation between the extent of differential allelic expression estimated using different SNPs ) on the Illumina ASE assay , even though it did not pass the significance cut-off . Overall in this second panel , 497 SNPs in 317 genes were expressed in three or more heterozygous individuals ( out of 1536 SNPs in 674 genes ) and 78 SNPs in 65 genes showed a significant level of differential allelic expression ( Figure 1 ) . To test whether intronic SNPs could be used instead of exonic SNP , we included for each gene on the second panel one intronic SNP . In general , intronic SNPs were less successfully analyzed and passed our expression threshold only for genes highly expressed in LCLs ( Figure S5 ) . This finding is consistent with previous observations [30] and the low proportion of unspliced mRNA ( heteronuclear RNA ) in cells relative to spliced transcripts . If the intronic SNP of a gene was detected in the RNA extract , it typically yields estimates of differential allelic expression very similar with those obtained using exonic SNPs . Overall , 177 out of 1 , 009 expressed SNPs ( in 140 out of 643 genes expressed , 22% ) display population-average ratios of allelic expression larger than 40∶60 or an higher than expected variance in allelic expression among heterozygous individuals and are thus unlikely to result solely from stochastic variation in the experiment ( Figure S6 ) . Table 1 shows the 133 SNPs ( 100 genes ) with significant allelic imbalance after manual curation to remove possible false positives due to a low number of individual homozygous for the minor allele ( this list is likely over-conservative and the complete data is presented in Table S2 ) . Many of the genes with the highest extent of allelic imbalance in LCLs are located on the X-chromosome . While it is known that one allele at most X-linked genes is silenced in females by inactivation of one entire chromosome [34] , [35] , we would expect that a polyclonal cell population ( in which half of the cells inactivate one X chromosome and the other 50% inactivate the alternate X chromosome ) would give a similar level of expression for both alleles . However , all X-linked genes on our two SNP panels ( 22 SNPs in 12 genes ) were among the top 5% of genes with most dramatic allelic imbalance patterns . The extent of allelic imbalance at a given gene varies among individual LCLs but interestingly , the patterns of allelic imbalance are very consistent across genes for a given individual ( Figure S7 ) . Additionally , the inheritance of the expressed allele ( determined , when possible , using the pedigree information for the two families included in this study ) appeared random . It has been previously proposed that the extent of clonality of a cell line could explain the patterns of allelic imbalance at genes with random mono-allelic expression [30]: clonal cells will all have the same X chromosome inactivated and thus display very high ratios of allelic imbalance . In contrast , cell-lines composed of a polyclonal population of lymphoblasts will have one or the other of their X chromosomes inactivated in different cells and thus an apparent expression of both alleles ( i . e . , a low extent of differential allelic expression ) . Our observations at X-linked genes are consistent with this hypothesis and the biased clonality of these LCLs , which were created over 20 years ago and passaged numerous times ( see also [30] ) . The two autosomal genes displaying the most dramatic allelic imbalance patterns have previously been shown to be imprinted in humans: PEG10 [36] and SNRPN [37] . In addition , KCNQ1 , MEST and ZNF215 which are imprinted in humans [38]–[40] also show significant differences in allelic expression ( Table 1 ) . The mode of inheritance of the expressed allele also corresponds , in each case , to what has been described for the expression of these genes: for PEG10 and SNRPN , heterozygous individuals express the paternally-inherited allele ( i . e . maternally imprinted ) while for KCNQ1 the maternally-inherited allele is expressed . Our limited pedigree information is not conclusive for MEST and ZNF215 . The only other known imprinted gene analyzable in our panel , PLAGL1 [41] , [42] did not pass the significance threshold ( i . e . an allelic ratio greater than 60∶40 ) but shows a population average allelic imbalance larger than 55∶45 and a high correlation between the two SNPs analyzable in the panel ( rs2076684 and rs9373409 ) ; therefore it likely represents a significant difference in allelic expression . The 83 remaining genes ( 103 SNPs ) with significant population-average allelic imbalance included several genes for which allelic imbalance had been shown in previous studies ( e . g . IL1A or IGF1 described in [30] ) . For some genes ( e . g . , CHI3L2 ) , one allele/haplotype is clearly expressed more than the other in heterozygotes and the inheritance pattern in families supports a genetic cause for allelic imbalance . For other genes , neither the direction of allelic imbalance nor the pedigree analysis allowed us to easily differentiate the genetic/epigenetic cause of the differential allelic expression ( Table 1 ) . For 56 genes with significant differences in allelic expression we tested whether differential allelic expression could be statistically associated to one of the SNP in the vicinity of the gene genotyped by the HapMap project ( see Materials & Methods for details ) . The results of these tests for SERPINB10 and ABCG1 are shown on Figure 7 and the strongest nominal association for each gene is displayed on Table 2 . Twenty-three genes still display statistical significant associations after Bonferroni correction for multiple testing ( highlighted in green on Table 2 ) showing a clear enrichment relative to the 2–3 associations expected by chance . Our power to detect a significant association between a HapMap SNP and the under-/over-expressing chromosome in this setting is low due to our reduced sample size ( only the heterozygous individuals are taken into account in this analysis ) and the number of regulatory haplotypes identified is thus likely underestimated . Additionally , many SNPs are tested for each gene and it is thus possible that some of the regulatory haplotypes result from spurious associations ( i . e . they are false positives ) . One argument against a very high rate of false positive in our analysis is that imprinted genes such as MEST or PEG10 do not show any signal of association ( Figure S8 ) consistent with the fact that the cis-regulatory mechanism at these genes is not encoded in the DNA sequence . To further investigate the validity of our association , we attempted to independently confirm these regulatory haplotypes by testing for the statistical association between one SNP in the regulatory haplotype and gene expression level . We used gene expression measurements performed at the Wellcome Trust Sanger Institute ( kindly provided by M . Dermitzakis ) on the same individual cell lines assayed by Illumina gene expression arrays . For each gene , we tested whether the homozygotes for the regulatory haplotype associated with low allelic expression in heterozygotes show a significantly lower gene expression level than the homozygous individuals for the regulatory haplotype associated with high allelic expression . We also performed locus-specific RT-PCR and quantified the level of gene expression using SYBR-Green for eleven genes for which differential allelic expression was significantly associated with allelic expression but for which expression data were not available ( 2 genes ) or genes with strong association with a regulatory haplotype but that were not validated using the Sanger dataset ( 9 genes ) . Overall , out of the 47 genes with a significant association between a SNP ( or several , defining the regulatory haplotype ) and differential allelic expression at the nominal cut-off , 10 were confirmed using gene expression measurements while 5 other genes showed a trend but did not reach statistical significance ( Table 2 ) . We analyzed differences in relative allelic expression ( or allelic imbalance ) at 1 , 380 human genes using 2 , 968 SNPs and more than 80 lymphoblastoid cell lines from individuals with European ancestry . Using quantitative sequencing we validated our results for a subset of genes and showed that the experimental variability in both settings is low and that the Illumina ASE assay and quantitative sequencing of RT-PCR products yield reproducible estimates of allelic imbalance consistent with each other . Overall , the experimental noise is much lower than the difference in relative allelic expression observed at many loci and therefore cannot be responsible for it . Additionally , the high concordance of the results obtained using different SNPs in the same transcript supports our findings that allelic imbalance , as we estimated it , is not an experimental artefact but reflects inherent biological differences in the relative expression of both alleles in heterozygous individuals . We also showed that lymphoblastoid cell lines , despite being simplified biological materials , are suitable resources to investigate mechanisms of gene regulation . Here , we demonstrated that our estimation of allelic imbalance is little affected by growth conditions and that LCLs harvested from different passages yield very similar results . Finally , the results efficiently recapitulate the consequences of the epigenetic mechanisms established in the individuals from which the cells have been derived ( see also [43] ) . We are therefore confident that , overall , the patterns of allelic imbalance we observed are neither experimental artifacts , nor specific to the material studied , but represent a common biological phenomenon affecting human gene expression . We showed that LCLs derived from female individuals still harbor the consequences of X-inactivation at all X-linked genes investigated , with one allele being transcriptionally silenced [34] . The extent of allelic imbalance detected at X-linked genes can vary among LCLs due to the various degrees of clonality of these cells but clonal LCLs consistently show complete silencing of one allele at all X-linked genes investigated ( Figure S7 ) . In addition , imprinting , established in the germ lines of the parents of the individuals from which the cells are derived [44] , is also maintained in LCLs . In our experiments , PEG10 , SNPRN , MEST and KCNQ1 show reduced or absent expression of one allele and , when the mode of inheritance can be determined , it corresponds to the imprinting mechanism described in the literature ( i . e . PEG10 and SNPRN are maternally imprinted , KCNQ1 is paternally imprinted ) . We thus observe extensive differential allelic expression ( i . e . allelic ratio larger than 70∶30 ) for all genes whose expression is known to be epigenetically regulated . This clearly shows that analysis of differential allelic expression is a suitable method for identifying the consequences of epigenetic mechanisms of gene regulation . The Illumina ASE assay would thus provide an efficient method to screen tumor tissues and identify patterns of differential allelic expression resulting from aberrant methylation or loss of imprinting that are known to be involved in the etiology of cancers [45]–[47] . Interestingly , IMPACT which shows significant extent of allelic imbalance at two SNPs ( rs677688 and rs1053474 ) in our study , is known to be imprinted in mice [48] but not in humans [49] . The mode of inheritance of the over-expressed alleles could not be determined using the two families available in our study ( i . e . the parents were always homozygous for the same allele ) . The attempt to map differential expression to a regulatory haplotype was not successful and is consistent with an epigenetic mechanism of gene regulation . More investigations are required to determine whether the pattern of allelic imbalance observed for IMPACT results from incomplete silencing of one allele following imprinting in the parental germ-lines or whether it results from random mono-allelic expression or another mechanism of gene expression regulation . Our analysis of 643 genes expressed in LCLs shows that , for a large proportion of them ( ∼20% ) , the two alleles are differentially expressed in most heterozygous individuals . For 18 genes , differential expression resulted from a known epigenetic silencing of one of the two alleles , either through X-inactivation in females or imprinting . The mechanisms leading to allele-specific expression at all other genes could be driven by a polymorphism affecting the cis-acting regulation ( e . g . a SNP in a transcription factor or a miRNA biding site ) or simply result from random silencing of one of the two alleles . We tested 56 genes for association of differential allelic expression patterns observed with a cis-acting regulatory polymorphism using genotypes generated by the HapMap project ( see Materials and Methods for details ) . For 23 of these genes we identified a region statistically associated with differences in allele expression that could indicate the existence of a regulatory haplotype ( i . e . , a region of one chromosome likely containing the polymorphism ( s ) causing the differential cis-regulation ) . These regions are often tens of kb long , consistent with previous descriptions of the linkage disequilibrium patterns in humans [50] . Although this approach does not identify the actual polymorphism ( s ) responsible for the differential cis-regulation , examination of these regulatory haplotypes provides some valuable insights on the mechanisms leading to differential expression and can guide future investigations . For example , the regulatory haplotype for GAS7 is almost exclusively restricted to the 3′UTR of the gene and may indicate that the patterns of allelic imbalance observed are due to differential mRNA processing , stability or the presence of a 3′ enhancer . In contrast , the regulatory haplotype identified for MGC33648 is located in the 5′ region and does not seem to overlap with the gene itself . This might be indicative of alternative promoter usage or differential transcription efficiency ( e . g . due to differential transcription factor binding site affinity ) . Several recent studies have used large-scale associations between gene expression and extensive genotype information to investigate gene regulation in humans , some of them using cell lines included in our study . In particular , Stranger and colleagues analyzed 630 genes located in ENCODE regions , on chromosome 21 and in one portion of chromosome 20 . They found evidence of cis-acting regulation for 63 genes [19] . 2005 ) . We were able to analyze 21 of these genes in our experiment . Six of them also showed evidence of cis-acting regulation ( e . g . SERPINB10 or TSGA2 ) in our study while a seventh gene ( TTC3 ) showed patterns consistent with differential allelic expression but did not reach our significance threshold . The remaining 14 genes did not show evidence of differential allelic expression in our analysis . Alternatively , we identified 10 new genes located in ENCODE region or chromosome 21 that showed significant level of differential allelic expression but were not detected in the Stranger study . Several non-exclusive reasons could explain the discrepancies between the results of the two approaches . First , it is worth noting that , even if the same individuals are analyzed by allelic-specific expression and gene expression association , the power to detect cis-acting effect differs depending on the allele frequency of the marker used: in gene expression association analysis all individuals are analyzed but the power in the regression analysis depends on their genotypes ( e . g . the genotypes AA , AB and BB are encoded in the linear regression as 0 , 1 and 2 ) while in allelic expression analysis only the individuals heterozygotes at the marker considered are analyzed . This can become particularly problematic to study differential allelic expression at some genes since it requires a relatively common exonic SNP to detect allelic imbalance . In this context , it is worth noting that intronic SNPs can successfully be used for genes that are highly expressed ( see also [30] ) . Second , associations of gene expression to genotypes depends greatly on the linkage disequilibrium ( LD ) patterns and requires extensive genotype information from all the individuals in order to include one marker in LD with the regulatory polymorphism . Allelic expression , on the other hand , directly investigate cis effect directly at the gene level and thus only requires physical link between the gene and the regulatory polymorphism affecting it ( i . e . they need to be on the same chromosome ) . Finally , the differences between allelic expression and gene expression mapping might indicate that some genes are also regulated by trans-acting mechanisms that differ among individuals: differential allelic expression is influenced only by cis-acting mechanisms of gene regulation while gene expression is influenced by cis- and trans-acting gene regulation . It is thus not unlikely that individual differences in trans-acting regulation swamp the signal from cis-acting polymorphisms . In this context , it is noteworthy that total gene expression mapping has been much more successful in mice and yeast for which the genetic heterogeneity is much lower and can be controlled ( reviewed in [9] , [10] , [51] ) . In humans , or in any other outbred population , genetic heterogeneity greatly limits the identification of cis-acting mechanisms using gene expression data while measurements of differential allelic expression are unaffected . We showed here that allelic expression assays are complementary from gene expression mapping and that the Illumina ASE assay overcomes two of the major limitations and criticisms of the former methodologies used to assess differential allelic expression: it allows a robust and high-throughput estimation of allelic imbalance: it is now possible to reliably screen hundreds of RNAs for several hundreds of genes in a couple of days . Additionally , when several SNPs can be used to assess differential allelic expression , the assay becomes very robust since each marker provides an independent estimation and one can test the correlation among estimates obtained at different positions . It is worth noting here that since this assay relies on the comparison of allelic ratio in DNA and RNA of each individual , it internally controls for the existence of polymorphisms in the primer sites or copy number variation encompassing the gene studied ( that will affect equally DNA and RNA ) . Likely , the greatest advantage of the analysis of differential allelic expression over total gene expression is its flexibility . To identify differential regulation of gene expression using total gene expression , one needs extensive genotype information to test whether , at any polymorphic position , the gene expression differences among individuals segregate according to their genotype . This precludes a quick assessment of the expression of one locus in one cohort of particular interest or using a specific tissue . In contrary , differential allelic expression offers the advantage that any one gene can be quickly assessed in any cohort or tissue by simply comparing the expression of the two alleles in each individual ( the amount of genetic information recently made available by the HapMap project allows a quick and easy selection of markers likely to be polymorphic for a given gene ) . The determination of regulatory haplotypes would still require extensive information concerning surrounding polymorphisms but the initial screening to determine whether one transcript is differentially cis-regulated can be done very efficiently with a handful of markers . We showed that differential allelic expression is a robust approach to identify cis-acting mechanism of gene regulation . It complements gene expression association studies and offers additional perspectives , notably on epigenetic mechanisms of gene regulation . It could thus be particularly interesting to apply this assay to tumors to detect mis-regulated genes due to aberrant methylation patterns or loss of imprinting . In addition , our approach is applicable to any new cohort or tissue since it is self-sufficient to identify differential cis-regulation and does not require additional genotyping . It can be easily used to follow-up interesting non-coding regions associated to a particular disease and test if they are involved in the etiology of the disease through some regulatory effects on neighboring genes . 83 lymphoblastoid cell lines ( LCL ) derived from blood samples from the CEPH collection were selected for this project . They included 60 unrelated individuals obtained from Utah residents with ancestry from western and northern Europe for which DNA was genotyped for millions of SNPs covering the entire genome by the International HapMap Project . Additionally , 21 LCLs from CEPH pedigrees 1420 and 1444 were included to provide complete information on two three-generation CEPH families . Cells were grown at 37°C and 5% CO2 in RPMI 1640 medium ( Invitrogen , Burlington , Canada ) supplemented with 15% heat-inactivated fetal bovine serum ( Sigma-Aldricht , Oakville , Canda ) , 2 mM L-glutamine ( Invitrogen , Burlington , Canada ) and penicillin/streptomycin ( Invitrogen , Burlington , Canada ) . The cell growth was monitored with a hemocytometer and the cells were harvested when the density reached 0 . 8–1 . 1 × 106 cells/mL . Cells were then resuspended and lysed in TRIzol reagent ( Invitrogen , Burlington , Canada ) . For all LCLs , three successive growths were performed ( corresponding to the 2nd , 4th and 6th passages ) after thawing frozen cell aliquots . We estimated allelic imbalance at 1 , 380 genes ( two panels of ∼1 , 500 SNPs , Figure 1 ) using the Illumina ASE assay ( Figure 2 ) . The experiment is similar to the one used for large-scale SNP genotyping [52] and gene expression profiling [53] except that DNA and RNA are independently assessed and compared to each other . RNA was first converted into biotinylated cDNA [53] while DNA was treated according to the usual GoldenGate assay protocol [52] . Biotinylated DNA ( derived from genomic DNA or mRNA ) was immobilized on paramagnetic beads and pooled SNP-specific oligonucleotides were annealed on the DNA . Hybridized oligonucleotides were then extended and ligated to generate DNA templates , which were amplified using universal fluorescently-labeled primers . Finally , single-stranded PCR products were hybridized to a Sentrix Array Matrix [52] , and the arrays were imaged using the BeadArray Reader Scanner [54] . 96 samples ( DNA or RNA ) were analyzed per Sentrix Array for ∼1 , 500 SNPs . All RNA measurements were performed in duplicates . To estimate the extent of allelic imbalance in heterozygote individuals at each SNP of the Illumina ASE panel , we developed algorithms using two different approaches: i ) we used information from individuals of all three genotypes ( AA , AB and BB ) , and/or ii ) we used only the heterozygote individuals . We first determined whether a given gene was expressed above a determined background in a given individual . To do so , we made use of the fact that the genotypes were known ( from the DNA analysis ) and developed a locus-specific expression background cut-off: homozygote individuals ( i . e . AA or BB ) can only express the corresponding allele , respectively A or B , at the RNA level ( if at all ) . We thus determined a background fluorescence level ( i . e . corresponding to random noise ) for each allele ( i . e . A and B ) by measuring the emission in the corresponding dye ( respectively , Cy3 and Cy5 ) in individuals homozygous for the other allele ( respectively BB and AA ) . This is represented schematically on Figure S9 . To avoid false positive results due to the inclusion of transcripts not expressed in the cell lines considered , we used a conservative approach and arbitrarily fixed the background emission cutoff to the maximum emission of the absent allele of all homozygotes , plus the mean emission of the absent allele divided by the number of homozygotes ( to weight the uncertainty in the determination of the “maximum noise” by the numbers of individuals used to determine it ) . This procedure allowed us to independently estimate the background emission of each allele/dye specifically for each SNP , which is particularly important because the fluorescence emission can differ drastically between the dyes and among loci ( data not shown ) . We then proceeded to the detection call using the background cut-offs: individuals with genotypes AA were considered to express a given transcript if the emission was larger than twice the cutoff background emission of A , individuals with genotypes AB if the fluorescence was larger than the sum of the background emission of A and the background emission of B , and individuals with genotypes BB if the emission is larger than twice the background emission of B . Since the inclusion in the analyses of transcripts expressed at low level ( or not expressed at all ) is very problematic , we excluded from our analyses all loci for which less than 75% of the individuals had discordant replicate expression ( i . e . , one replicate above expression background , the other under the cut-off value ) . The first method used to determine whether some heterozygote individuals expressed significantly differently the two alleles is locus-specific but requires having at least one individual expressed from each homozygote genotype ( AA and BB ) . In this case , we determined the median log ratio of the two dyes for each homozygote clusters at the DNA and RNA level ( ) as well as the median absolute deviations ( MAD ) . We used medians and MADs , instead of means and standard deviations , to down weight the influence of possible outliers . We then determined a range of “expected” ( i . e . non significant ) variation of allelic expression for the heterozygote individuals . We calculated the equation of the lines joining the median values plus/minus two MAD of AA and BB and estimated the range , for the log ratio of the dyes at the RNA level , between the lines at the value corresponding to the median of DNA in heterozygote individuals ( Figure S10 ) . If the observed log ratio of dyes for a given heterozygote individual fell outside the expected range of variation in absence of AI ( Figure S8 ) , we scored each heterozygote individual separately to obtain a quantitative estimation of allelic imbalance using the ratio:This simple estimate indicates both the magnitude of the allelic imbalance ( i . e . the fold difference ) and its direction ( i . e . which allele is more expressed than its counterpart ) . In order to assess allelic imbalance for SNPs with low minor allele frequencies ( for which homozygote individuals with the minor allele may not be present in a small sample size panel ) , we developed a second method based solely on the heterozygote individuals . If a given transcript is affected by allelic imbalance we expect that either the variance of the log ratio of dyes for heterozygote RNAs to be greatly increased relative to the variance of homozygote RNAs , or , if one allele is systematically more expressed than the other , the mean value of these log ratios to be drastically shifted from its expected intermediate position ( between the mean for AA and the mean for BB homozygote RNAs ) . For all SNPs with at least five individuals with the same genotype expressed , we estimated the standard deviation of the log ratio of dyes for DNA and RNA . The distribution of the log ratio of the standard deviations ( i . e . log σDNA/σRNA ) over all loci for heterozygous individuals differed from those observed using homozygous individuals and did not seem to fit a normal distribution ( Figure S11 ) . Based on the assumption that this distribution may include some loci in allelic imbalance ( and thus with a higher than expected RNA variance ) , we fitted a mixture of two Gaussians on our dataset ( i . e . , one corresponding to the loci with allelic imbalance , the second including all other loci ) using a Maximum Expectation algorithm implemented in R ( mixdist package ) . For our data , the best fit was obtained with a minor distribution ( including ∼3% of the loci ) corresponding to the most extremely negative log ratios of variances ( i . e . , that the RNA standard deviation was larger than expected ) . For each locus , we then used the probability of belonging to the “higher-than-expected RNA variance” distribution as an indication of allelic imbalance . We assessed the extent of allelic imbalance by quantitative sequencing following the method described in Ge et al . [33] . Briefly , we isolated RNA using TRIzol reagent following the manufacturer's instructions . We assessed RNA quality with an Agilent 2100 Bioanalyzer ( Agilent , Palo Alto , USA ) before synthesizing first strand cDNA using random hexamers ( Invitrogen , Burlington , Canada ) and Superscript II reverse transcriptase ( Invitrogen , Burlington , Canada ) . For each locus , we designed locus-specific primers , in the exon/UTR containing the SNP analyzed , at least 50 bp away from the SNP studied . 5 ng of genomic DNA and 10 ng of total cDNA were then amplified by PCR using Hot Start Taq Polymerase ( Qiagen , Mississauga , Canada ) with an activation step ( 95°C for 15 minutes ) followed by 40 cycles ( 95°C for 30 s , 55°C for 30 s and 72°C for 45 s ) and a final extension step ( 72°C for 6 minutes ) . PCR products were purified using Exonuclease I and Shrimp Alkaline Phosphatase ( USB , Cleveland , USA ) and sequenced using either one of the former primers or a nested primer , on an Applied Biosystems 3730xl DNA analyzer . We used PeakPeaker v . 2 . 0 [33] with the default settings to quantify the relative amount of the two alleles measured from the chromatogram after peak intensity normalization . To estimate the experimental variability of the entire experimental setup we used a hierarchical strategy for two genes ( cf . Figure S12 ) : for two/three individual cell lines , we extracted independently RNAs three times and performed , on each extract , three independent RT-PCRs . All cDNA obtained were then split into three aliquots , each amplified independently by locus-specific PCR . These PCR products were finally sequenced each three times ( i . e . three independent sequencing reactions ) . To estimate the variability at each experimental stage we calculated the mean standard variation normalized to the mean using the independent triplicates . To calculate the variance in the higher hierarchical levels ( PCR , RT-PCR ) , we averaged the values from the lower level ( e . g . , to estimate the variability at the PCR level , we compared the means of the three sequencing values performed on each of the three PCRs: [s1 , s2 , s3] vs [s4 , s5 , s6] vs [s7 , s8 , s9] ) . The results are presented in Text S1 . We attempted to map allelic imbalance to regulatory haplotypes for all genes with significant differences in allelic expression that fulfilled these criteria: i ) they are mapped on the build 34 of the human genome , ii ) the SNP used in the Illumina ASE assay has also been genotyped by the HapMap [55] and iii ) there are more than four HapMap individuals heterozygous at the marker SNP . For each gene , we retrieved the haplotype information from the phased chromosomes of each of the 57 HapMap CEPH individuals for 100 , 000 bp upstream and downstream of the SNP used to assess allelic imbalance . When a transcript contains more than one SNP or if two SNPs used to assess allelic imbalance at two transcripts are separated by less than 200 , 000 bp , the region retrieved spans from the most upstream marker plus 100 , 000 bp to the most downstream marker minus 100 , 000 bp . For each individual LCL , the over expressed and under expressed haplotype/chromosome were identified and each SNP was tested for segregation of the alleles in under- and over-expressed chromosomes using a Fischer's exact test . Between 47 and 592 SNPs were tested for each gene ( mean = 229 ) and the associations remaining significant after Bonferroni correction for multiple testing are shown in green in Table 2 . Illumina total gene expression data were obtained from the Wellcome Trust Sanger Institute for the 60 unrelated CEPH individuals genotyped by the HapMap project and included in our assay . We also determined the total expression for 10 genes using Real-Time PCR and SYBR Green labeling on an ABI 7900HT ( Applied Biosystems , Foster City , CA ) instrument . 8–10 ng of first strand cDNA were amplified using 0 . 32 µM of gene specific primers and Power SYBR Green PCR master mix ( Applied Biosystems ) according to the manufacturer's instructions . The amplifications started by 95°C for 10 min followed by 40 cycles at 95°C for 20 s , 58°C for 30 s and 72°C for 45 s . We performed the Real-Time PCR assays for the 60 individuals LCLs genotyped by the HapMap projects and analyzed 6 replicates per each sample . A standard curve was established using a dilution series of total cDNA of known concentration . The Ct for each replicate was transformed to a relative concentration using the estimated standard curve function ( SDS 2 . 1 , Applied Biosystems ) and normalized based on 18S rRNA Taqman ( Applied Biosystems ) expression data obtained for each sample to account for well to well variability . All analysis scripts are available upon request . PeakPicker v . 2 . 0 is available at http://www . genomequebec . mcgill . ca/EST-HapMap/ .	We describe a new methodology to identify individual differences in the expression of the two copies of one gene . This is achieved by comparing the mRNA level of the two alleles using a heterozygous polymorphism in the transcript as marker . We show that this approach allows an exhaustive survey of cis-acting regulation in the genome; we can identify allelic expression differences due to epigenetic mechanisms of gene regulation ( e . g . imprinting or X-inactivation ) as well as differences due to the presence of polymorphisms in regulatory elements . The direct comparison of the expression of both alleles nullifies possible trans-acting regulatory effects ( that influence equally both alleles ) and thus complements the findings from gene expression association studies . Our approach can be easily applied to any cohort of interest for a wide range of studies . It notably allows following up association signals and testing whether a gene sitting on a particular haplotype is over- or under-expressed , or can be used for screening cancer tissues for aberrant gene expression due to newly arisen mutations or alteration of the methylation patterns .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genetics", "and", "genomics/functional", "genomics", "genetics", "and", "genomics/gene", "expression", "genetics", "and", "genomics/complex", "traits", "genetics", "and", "genomics/genetics", "of", "disease", "genetics", "and", "genomics/epigenetics", "genetics", "and", "genomics", "genetics", "and", "genomics/cancer", "genetics" ]	2008	Differential Allelic Expression in the Human Genome: A Robust Approach To Identify Genetic and Epigenetic Cis-Acting Mechanisms Regulating Gene Expression
Ambiguity in genetic codes exists in cases where certain stop codons are alternatively used to encode non-canonical amino acids . In selenoprotein transcripts , the UGA codon may either represent a translation termination signal or a selenocysteine ( Sec ) codon . Translating UGA to Sec requires selenium and specialized Sec incorporation machinery such as the interaction between the SECIS element and SBP2 protein , but how these factors quantitatively affect alternative assignments of UGA has not been fully investigated . We developed a model simulating the UGA decoding process . Our model is based on the following assumptions: ( 1 ) charged Sec-specific tRNAs ( Sec-tRNASec ) and release factors compete for a UGA site , ( 2 ) Sec-tRNASec abundance is limited by the concentrations of selenium and Sec-specific tRNA ( tRNASec ) precursors , and ( 3 ) all synthesis reactions follow first-order kinetics . We demonstrated that this model captured two prominent characteristics observed from experimental data . First , UGA to Sec decoding increases with elevated selenium availability , but saturates under high selenium supply . Second , the efficiency of Sec incorporation is reduced with increasing selenoprotein synthesis . We measured the expressions of four selenoprotein constructs and estimated their model parameters . Their inferred Sec incorporation efficiencies did not correlate well with their SECIS-SBP2 binding affinities , suggesting the existence of additional factors determining the hierarchy of selenoprotein synthesis under selenium deficiency . This model provides a framework to systematically study the interplay of factors affecting the dual definitions of a genetic codon . Stop codons can be reassigned to encode amino acids [1 , 2] . Failures in stop codon reassignment leads to the production of prematurely terminated proteins [3 , 4] , but how cellular factors influence alternative definitions of stop codons is not fully understood . While some stop codon reassignments are confined to certain species or organelles , redefinition of UGA to selenocysteine ( Sec ) in selenoprotein synthesis occurs in all three domains of life [5] . Selenoproteins are proteins that contain the Sec amino acid residue . Translating UGA to Sec requires Sec-tRNASec ( Sec-specific tRNA charged with Sec ) , the Sec insertion sequence ( SECIS ) element at the 3’ untranslated region ( 3’UTR ) of selenoprotein mRNAs [4 , 6 , 7] , and other regulatory factors such as SBP2 [8–10] and EFSec [11 , 12] . Failed UGA to Sec decoding results in translation termination , with UGA being recognized by a release factor ( RF ) instead . RFs trigger the hydrolysis of ester bonds in peptidyl-tRNA and corresponding release of translated proteins from the ribosome [13 , 14] . Translating UGA to Sec is inefficient [15–17] and influenced by the abundance of selenoprotein mRNA , Sec-tRNASec , selenium , SBP2 and the intrinsic properties of SECIS elements [8 , 17–21] . Overexpression of selenoprotein mRNA reduces UGA-to-Sec decoding [18 , 22] , but this effect could be rescued by co-expression of uncharged Sec-specific tRNA ( tRNASec ) [18 , 22] or SBP2 [8] . The efficiency of Sec incorporation has been shown to be positively correlated with tRNASec or selenium supply in cells [20] yet differs among seleoproteins [23] . There are at least 25 selenoproteins in the human proteome [24] and their difference in Sec incorporation efficiency leads to a “selenoprotein hierarchy” under selenium deficiency [23]: proteins with higher Sec incorporation efficiency exploit more Sec-tRNASec and are more rapidly synthesized . It is well known that hierarchical selenoprotein expression depends on the SECIS-SBP2 interaction [8] , but whether this interaction is the sole determinant for selenoprotein hierarchy remains unclear . Despite the aforementioned rich studies in selenoprotein translation , a systematic and quantitative characterization of the joint effects of various regulatory factors has not yet been reported . To fill this gap , we developed a simple mechanistic model that captures the quantitative characteristics of the UGA translation process and applied this model to experimental data to investigate how various regulatory factors influence the definition of UGA . We utilized differential protein half-lives from full-length and truncated selenoproteins , retrieved from a single-cell-based global protein stability ( GPS ) assay [25] , to infer UGA definitions under cell culture conditions with variations in selenium supply and selenoprotein expression levels , and used those inferred quantities to estimate the model parameters . We found that the qualitative behavior of selenoprotein translation derived from our model closely resembles that from experimental data . Moreover , we re-capitulated the selenoprotein hierarchy by measuring and comparing the stability of proteins expressed from constructs with SECIS elements of four distinct selenoproteins . The estimated Sec incorporation rates are incongruent with the reported SECIS-SBP2 binding equilibrium constants , suggesting the existence of additional factors to explain selenoprotein hierarchy . Our model provides a framework to quantitatively study the regulation of UGA codon redefinition and selenoprotein synthesis . Selenoprotein synthesis serves as a remarkable model to study how cellular and environmental factors influence the definition of a dual-use codon . We have proposed a concise mathematical model of selenoprotein synthesis that matches well with both qualitative and quantitative characteristics of experimental results . By combining the power of biological experiments and computational modeling , we have revealed how multiple cis and trans regulatory factors collectively influence the definition of UGA . The characteristics of experimental data can be explained by the competition between RF and Sec-tRNASec for UGA codons of limited selenoprotein mRNAs , as well as the limited abundance of tRNASec . We formulated these two types of resource limitation as a quantitative , mechanistic model . Simulations according to this model successfully reproduced qualitative characteristics of the experimental data ( Fig 5 ) . Beyond qualitative matching , we also proposed an algorithm to estimate model parameters from experimental data . The model derived from the estimated parameters fit well with the experimental data ( Fig 6A , S1 Table and S2 Table ) . Previous work on the importance of SECIS-SBP2 interactions for the selenoprotein expression hierarchy remains inconclusive . Some studies have indicated that SECIS-SBP2 interactions dictate the selenoprotein hierarchy [8] , whereas others have suggested that those interactions alone are insufficient to determine Sec incorporation efficiency [21 , 32] . Our deduced Sec incorporation rates attributed to distinct SECIS elements did not correlate well with reported SECIS-SBP2 binding affinities ( Table 4 ) . SEPHS2 and GPX1 had substantially higher Sec incorporation rates than SEPX1 and SELK , yet the SECIS-SBP2 binding of SELK was the strongest among the four SECIS elements . Thus , we provide evidence to support the presence of other determining factors for selenoprotein hierarchy . The order of predicted GFP-RFP curves among the four SECIS elements is consistent with the order of the corresponding experimental curves except for zero selenium concentration ( S6 Fig ) . At zero selenium concentration , the predicted curves of all SECIS elements coincide and are considerably lower than all the experimental curves . This is likely due to the existence of residual selenium in cells even at zero external selenium supply . The parameters in our model conform to some of the fundamental quantitative features of cell biology , such as the translation and degradation rates of proteins , incorporation rates of Sec-tRNASec and RFs , and the quantities of tRNASec and RFs in cells . Few of these quantities have been reported for mammalian cells , so it is not possible to verify the accuracy of the estimated parameters from existing information . Thus , a thorough verification of the estimated parameter values remains to be conducted . The concise selenoprotein synthesis model we propose circumvents detailed mechanistic description . It is now possible to build a more detailed , mechanistic model by including all the intermediate steps in the pathway . However , introducing additional free parameters without concomitant measurements merely complicates the model with little improvement in accuracy . Importantly , in our simplified equations , we reveal the existence of a limiting factor beyond selenium concentration in Sec-tRNASec synthesis . Which enzymes or substrates constitute the true limiting factor warrants further investigation . Likewise , incorporation of tRNASec or RFs at a UGA site involves binding of multiple molecules [8–12 , 33 , 34] . Some of them could possibly be limiting factors additional to excess mRNA and tRNASec supplies . Despite Sec incorporation being a very specialized process , the process of synthesizing and degrading multiple products with shared and limited resources is ubiquitous in biochemical systems . Some instances include dichotomy between growth and production of organisms , competitive binding of transcription factors and their repressors on promoters , and biosynthesis of metabolites from multiple pathways with shared substrates . Although the models capturing those phenomena may have very different formulations than the models described in this study , the methodology we introduced may be extended to other systems with similar characteristics . Furthermore , presence of multiple exogenous and endogenous limiting factors , such as selenium , selenoprotein transcripts and tRNASec in our study , may yield a more complicated system behavior than the cases with single or no limiting factors . To generate the SEPHS2 and SEPW1 GPS reporter construct , SEPHS2 and SEPW1 cDNA from the Mammalian Gene Collection ( GE Healthcare Dharmacon Inc . , Lafayette , CO , USA ) was cloned into a lentiviral vector carrying the RFP-IRES-GFP GPS cassette using Gateway technology ( Life Technologies , Carlsbad , CA , USA ) . To generate SEPHS2 and SEPW1 mutants that exclusively express PL or PS , the TGA/Sec codon on SEPHS2 and SEPW1 cDNA was mutated into TGT/Cys or TAA/stop by site-directed mutagenesis ( Stratagene , Santa Clara , CA , USA ) , respectively . To replace the SECIS element of SEPHS2 with that of other selenoproteins , SECIS elements of GPX1 , SELK and SEPX1 were amplified from corresponding selenoprotein cDNAs and cloned into the SEPHS2 reporter using Gibson Assembly ( New England Biolabs Inc . , Ipswich , MA , USA ) . HEK293T cells were maintained in DMEM with 10% fetal bovine serum ( FBS , purchased from Hyclone Laboratories , Logan , UT , USA ) and antibiotics in a 6% CO2 atmosphere at 37°C . FBS is the main source of selenium in cell culture . To control selenium supply , cells were first depleted of selenium in FBS-free DMEM supplemented with 10 μg/mL insulin and 5 μg/mL transferrin for 24 hrs . Cells were then balanced with indicated concentrations of sodium selenite ( Na2SeO3 , Sigma-Aldrich , St . Louis , MO , USA ) for another 24 hrs . All tissue culture media and supplements were purchased from Gibco Life Technologies , unless otherwise indicated . To produce lentiviruses , HEK293T cells were transfected with pHAGE , pHIV gag/pol , pVsvg , pRev and pTat using TransIT-293 reagent ( Mirus Bio LLC , Madison , WI , USA ) . Viruses were harvested 48 hrs after transfection . To generate GPS reporter cell lines , cells were infected with lentiviruses carrying GPS reporter constructs . Infection was carried out in media with 8 μg/mL polybrene ( Sigma-Aldrich ) . To collect reporter cell lines with a series of SEPHS2 synthesis levels , cells were infected stepwise with lentiviruses carrying GPS reporter constructs . To prepare samples for FACS analysis , cells were washed with PBS , trypsinized and resuspended in medium containing 2% FBS and analyzed using a BD LSR Fortessa system ( BD Biosciences , San Jose , CA , USA ) . 106 cells were recorded for each sample . FlowJo ( Ashland , OR , USA ) was used for primary FACS data analysis . Cells were harvested in cold PBS and lysed in RIPA buffer ( 150 mM NaCl , 1 . 0% IGEPAL®CA-630 , 0 . 5% sodium deoxycholate , 0 . 1% SDS , and 50 mM Tris , pH 8 . 0 ) . Standard procedures were used for Western blotting . Antibody against GFP ( JL-8 ) was purchased from Clontech Laboratories ( Mountain View , CA , USA ) . The single-cell-based GPS data consists of 106 pairs of RFP-GFP intensities for individual cells . The RFP-GFP relationship in each cell manifests a high level of variation . However , for each small range of RFP values , the corresponding GFP values typically have a Gaussian distribution with a variance proportional to the RFP value . Therefore , we treated the GPS data as instantiations of the following random variables: y = f ( x ) + ϵ , where x denotes a random variable of RFP intensities with an unspecified distribution and y denotes a random variable of GFP intensities and is a function of x with an additive noise ϵ . ϵ∼N ( 0 , xσ2 ) follows a Gaussian distribution with zero mean and xσ2 variance . To reduce data noise and size , we applied two filtering procedures to the GPS data . First , we divided the range of RFP and GFP values into 2000 grids and discarded the data points in grids comprising fewer than 30 data points . Second , we sorted the RFP values and selected 0 . 4% data points . The processed data thereby consisted of about 3000 pairs of RFP and GFP values for each selenium concentration . The basic assumptions and reactions of the model are described in the Results and illustrated in Fig 1 . Here , we demonstrate the mathematical formulation of the model . We first introduce the following notations: We developed a grid-search algorithm to find the parameter values that best fit the experimental data . Among the six undetermined parameters , ρp is an arbitrary parameter that only affects the scale of selenoprotein expression but not the behavior of the translation process in simulation . Thus , we first excluded ρp in the fitting algorithm and manually adjusted ρp after fitting . We generated grids with different combinations of parameters and calculated the fitness of the predicted ( RFP , GFP ) intensities generated by these parameters with the experimental results . The grids were first generated by logarithmically dividing each parameter into 12 intervals within their boundaries ( the range of each parameter value is shown in Table 4 ) . These parameter sets were applied to the mathematical model to convert RFP values into PL and PS in the loss function Q2: TQ2=∑i=1nQi2 ( k1 , k3 , kf , Ttotal , ρ , ρp ) =∑i=1n ( Ptotali−PLi−PSi ) 2 Q2={Ptotal–[ρ⋅SPL ( 1+kF ) kF]k1⋅mf⋅k3⋅Se⋅Ttotal[ ( 1+k3⋅Se ) ( 1+k1⋅mf ) ]–[ρ⋅SPS ( 1+kF ) kF]⋅kF⋅mf}2 ( 17 ) The total loss function TQ2 is summed over all data points indexed by i . Ptotal is calculated by transforming GFP intensities using ρp . The loss function has a complicated nonlinear form and thus contains many local optima . Analytic algorithms such as gradient descent will likely find suboptimal solutions whose loss is far from the global minimum . We devised a variation of the divide-and-conquer heuristic approach to alleviate this problem . We started by partitioning the log-scale range of each parameter value by coarse-grained intervals . A small number of multi-dimensional grids were generated from the partitioned parameter space . We then recursively performed the following computations: ( 1 ) evaluation of loss function values of parameter configurations on the grids , ( 2 ) selection of the top 30 parameter configurations , and ( 3 ) subdivision of the selected grids into smaller intervals . Recursion stopped when the grid sizes reached the required resolution of parameter values . The criteria for selecting the parameter configuration from the top-ranking solutions are reported in S1 File . The Matlab codes of the parameter estimation algorithm are reported in S2 File . The GPS data of SEPHS2 , GPX1 , SEPX1 , SELK and SEPW1 are reported in S3–S7 Files respectively . The top ranking solutions of the four SECIS element constructs and SEPW1 are reported in S8 File . We randomly generated 100 parameter sets within each parameter boundary by the following function: 10log10UBi+ ( log10UBi−log10LBi ) X ( 18 ) Where UB and LB are the upper and lower bounds , respectively , of each parameter and X is a random number uniformly distributed on the open interval ( 0 , 1 ) . For each parameter set , about 1000 corresponding RFP and GFP values were generated by the mathematical model . The parameter estimation algorithm was applied to the simulated data , and the estimated parameter values were compared with the parameter values from which the simulated data were generated . We also introduced additive noise to the simulated data with the following formula: GFPnew=GFPoriginal+GFPoriginal⋅NorR ( 19 ) Where GFPoriginal denotes the GFP values calculated from the model . NorR is randomly drawn from a normal distribution with a mean equal to 0 . The standard deviation of NorR varied from 0 . 3 to 5 . 0 ( Table 2 ) . The performance of our algorithm was evaluated by the log10 ratios between predicted and underlying parameter values: Error=\|log10PpredictPanswer\| ( 20 ) Where Ppredict denotes the parameters predicted by the algorithm and Panswer are the true parameters . A parameter value prediction was labeled successful if the error of at least one of the predicted parameter set was smaller than 1 among the top 15 answers reported by the algorithm . The recovering rate indicates the ratio of successful predictions among 100 test sets .	The “code book” of protein translation maps 43 = 64 triplets of RNA sequences ( codons ) into 20 canonical amino acids and the stop signal . This code book is universal in almost all organisms on earth . Selenoproteins consist of selenium-containing amino acids–selenocysteines ( Sec ) –that are not among the 20 canonical amino acids . The cells “borrow” a stop codon UGA to translate selenocysteines . Since UGA maps to two possible outcomes , the translation machinery can synthesize both full-length selenoproteins ( when UGA encodes selenocysteine ) and truncated peptide chains ( when UGA encodes translational termination ) . Despite extensive study about selenoprotein synthesis mechanisms , a quantitative model for how cells allocate resources to synthesize each species is yet to appear . We propose a quantitative model that can explain the dependency of experimental observables such as protein stability and Sec incorporation efficiency by various factors such as selenium concentration and mRNA levels . Saturation of those quantities implies the existence of limiting factors such as mRNA transcripts and Sec-specific tRNAs . The match between model simulations and experimental data suggests that the cellular decision making of synthesizing the two species of proteins may follow simple first-order kinetics .	[ "Abstract", "Introduction", "Discussion", "Materials", "and", "methods" ]	[ "transfer", "rna", "applied", "mathematics", "messenger", "rna", "protein", "abundance", "simulation", "and", "modeling", "algorithms", "protein", "synthesis", "mathematics", "chemical", "synthesis", "research", "and", "analysis", "methods", "selenium", "proteins", "gene", "expression", "chemistry", "proteomics", "biosynthetic", "techniques", "biochemistry", "rna", "nucleic", "acids", "protein", "translation", "genetics", "biology", "and", "life", "sciences", "physical", "sciences", "non-coding", "rna", "chemical", "elements" ]	2017	A quantitative model for the rate-limiting process of UGA alternative assignments to stop and selenocysteine codons
Antibody effector functions , such as antibody-dependent cellular cytotoxicity , complement deposition , and antibody-dependent phagocytosis , play a critical role in immunity against multiple pathogens , particularly in the absence of neutralizing activity . Two modifications to the IgG constant domain ( Fc domain ) regulate antibody functionality: changes in antibody subclass and changes in a single N-linked glycan located in the CH2 domain of the IgG Fc . Together , these modifications provide a specific set of instructions to the innate immune system to direct the elimination of antibody-bound antigens . While it is clear that subclass selection is actively regulated during the course of natural infection , it is unclear whether antibody glycosylation can be tuned , in a signal-specific or pathogen-specific manner . Here , we show that antibody glycosylation is determined in an antigen- and pathogen-specific manner during HIV infection . Moreover , while dramatic differences exist in bulk IgG glycosylation among individuals in distinct geographical locations , immunization is able to overcome these differences and elicit antigen-specific antibodies with similar antibody glycosylation patterns . Additionally , distinct vaccine regimens induced different antigen-specific IgG glycosylation profiles , suggesting that antibody glycosylation is not only programmable but can be manipulated via the delivery of distinct inflammatory signals during B cell priming . These data strongly suggest that the immune system naturally drives antibody glycosylation in an antigen-specific manner and highlights a promising means by which next-generation therapeutics and vaccines can harness the antiviral activity of the innate immune system via directed alterations in antibody glycosylation in vivo . Mounting evidence points to a critical role for non-neutralizing antibody effector function , such as antibody-dependent cellular cytotoxicity ( ADCC ) , antibody-dependent cellular phagocytosis ( ADCP ) and complement-dependent cytotoxicity ( CDC ) , in protection against [1] , and control of HIV [2] , influenza [3] , Ebola virus [4] , and bacterial infections [5] . Earlier work suggests that potent , long-lived antibody effector activity is driven by IgG1 antibodies [6] , the dominant subclass in the blood [7] . However , as all vaccinated and infected individuals ultimately produce IgG1 antibodies , it is unclear why some IgG1 responses provide protective immunity while others provide limited immunity at the same titers . While emerging data suggest that the co-selection of additional antibody subclasses , such as the most functional subclass , IgG3 , may collaborate to direct more effective immune complex–based activity [8] , IgG3 is cleared rapidly from the systemic circulation [9] , arguing that sustained levels of some , but not other IgG1 antibodies may represent the critical determinant of protective immunity against HIV . Thus , defining how the immune system naturally tunes IgG1 represents a critical step for the development of more effective strategies to harness the immune system to prevent or control HIV infection . Every IgG antibody is glycosylated at a single asparagine residue within the CH2 domain of the constant region ( in the crystallizable fragment , Fc ) , and data from the monoclonal therapeutic community suggest that these changes potently alter the inflammatory profile and effector functions of the antibody [10] . The antibody glycan consists of variable levels of four sugar subunits ( galactose , sialic acid , fucose and an N-acetylglucosamine that bisects the arms of the structure ( b-GlcNAc ) ) , each of which alters the affinity of the antibody for innate immune receptors , including Fc receptors found on all innate immune cells [11] . For example , changes in fucose and the b-GlcNAc play a critical role in modulating monoclonal therapeutic antibody effector function , where a lack of fucose [12] , the addition of the b-GlcNAc [13] , and elevated sialic acid [14] increases ADCC activity . In contrast , agalactosylated polyclonal antibodies are associated with increased inflammation in HIV [15 , 16] and chronic autoimmune conditions [17] and agalactosylated monoclonal therapeutics are known to drive enhanced complement binding and activation [18] . Conversely , the presence of higher levels of galactose provides the scaffold for the addition of terminal sialic acid groups , that are thought to drive anti-inflammatory activity through binding to lectin-like receptors [19] , though there is some controversy in the field as to whether IVIG’s anti-inflammatory effect is due to sialylation alone [20–22] . Thus , while the antibody therapeutics field has clearly demonstrated that alterations in antibody glycosylation is a critical mechanism for improving therapeutic efficacy via the augmentation of effector function [6 , 13 , 23] or through the alteration of inflammation in rheumatoid arthritis treatment [19] , it is still unclear whether antibody glycosylation is actively regulated in vivo . While , recent studies on antigen specific antibodies have shown that antigen-specific antibodies are induced with distinct antibody glycan profiles , it is still unclear whether distinct antibodies within the same individual are programmed with unique glycosylation profiles aimed at enhancing particular effector functions . However , given the emerging data pointing to distinct antibody glycan profiles on antigen-specific antibodies compared to bulk circulating antibodies [15 , 24] , it is possible that antibody glycosylation may be actively controlled by the immune system . Moreover , over 30 different glycan structures have been identified in naturally produced antibodies , each with the theoretical capacity to drive distinct effector functional profiles [25 , 26] , that may be selected immunologically in disparate manners to drive unique effector functions . Thus this study aimed to determine whether antibody glycosylation is differentially tuned against specific pathogens and/or antigenic targets and whether antibody glycosylation could be actively directed through immunological priming . We demonstrate different glycoprofiles on particular antigen- and pathogen-specific antibodies , clearly illustrating unique antibody glycan profiles against individual antigens , each of which was distinct from that present in bulk circulating antibodies , linked to distinct antibody effector functions , pointing to antigen-specific regulation of antibody glycosylation . Furthermore , while bulk circulating antibody Fc glycosylation was dramatically different in geographically distinct populations , immunization with a viral vector-based vaccine induced remarkably similar antigen-specific antibody glycan profiles on vaccine-specific antibodies induced at all three geographic sites . Conversely , distinct antigen-specific IgG glycosylation profiles were induced by a protein–based HIV vaccine , suggesting that antibody glycosylation is selectively tuned at the time of vaccination by distinct inflammatory signals co-delivered at the time of B cell priming . Collectively , these data argue that IgG glycosylation is elicited differently by specific pathogens , antigens , and immune signals to selectively and specifically induce the targeted antibody functional profiles . Thus the regulation of antibody glycosylation represents a potentially novel means by which next-generation therapeutic or vaccine strategies may selectively direct the immune regulatory and killing activity of antibodies . Inflammatory diseases [27] and viral infections [10 , 11] drive an overall shift in bulk circulating antibody glycosylation . Likewise , recent studies have highlighted that unique glycan profiles emerge on antigen-specific antibodies [15 , 24] . However , whether all antigen-specific antibodies exhibit the same glycan profiles , tuned exclusively by inflammation , or whether antigen-specific antibody populations are tuned in an antigen- and/or pathogen- specific manner is unclear . Thus , HIV envelope ( gp120 ) - , HIV capsid ( p24 ) - , and influenza envelope ( HA ) -specific antibodies were selectively enriched from a population of 193 HIV-infected patients , and glycan profiling was performed on enzymatically removed glycans by capillary electrophoresis ( S1 Fig , S1 Table ) . Remarkably , antibody glycosylation differed not only between antigen-specific antibodies and bulk circulating antibodies , but also among the three different antigen-specific antibody populations ( Fig 1A ) . Specifically , as previously reported , gp120-specific antibodies possessed elevated levels of agalactosylated glycans and slightly increased fucosylated and bisected glycans with a decrease in sialylated structures compared to bulk antibodies ( Fig 1A ) . Collectively , this combination of sugars points to the induction of a more functional ( elevated b-GlcNAc ) and more inflammatory ( low galactose and sialic acid ) glycan on gp120-specific antibodies [15 , 16] . Thus , based on known glycan structure:function relationships [13 , 28] , gp120-specific antibodies exhibit a slightly inflammatory asialylated glycan poised to direct ADCC and complement-mediated killing via elevated b-GlcNAc levels . However , p24-specific antibodies exhibited an even more exaggerated inflammatory profile than gp120-specific antibodies ( Fig 1A ) . p24-specific antibody glycans included significantly higher levels of agalactosylated glycans and slightly more fucosylation , as compared to bulk antibody glycans ( Fig 1A ) . These antibodies exhibited low levels of sialic acid and comparable levels of b-GlcNAc to the bulk circulating antibodies . Thus , p24-specific antibodies are selectively tuned to express a highly inflammatory agalactosylated glycan . In contrast to gp120- and p24-specific antibodies , influenza-specific antibodies exhibited a significantly different glycan profile , marked by significantly increased galactosylation ( Fig 1A ) and sialylation and reduced b-GlcNAc . Thus , influenza-specific antibodies exhibited a third glycan profile , that like IVIG , may be tipped towards an anti-inflammatory glycan [10 , 19] that may be deliberately tuned to drive enhanced ADCC via reduced fucose [28] . Interestingly , no correlation was observed among glycan patterns selected on gp120- or HA-specific antibodies ( S2 Fig ) arguing that antigen-specific antibody glycosylation is selected independently in each individual and is not influenced by the host’s genetic or pre-infection background . Thus , based on univariate analyses , comparing the incorporation of individual sugars into the antibody glycan , antibody glycosylation varies significantly among antigen-specificities in HIV infected persons . Beyond differences at the antigen-specific level , differences were previously observed in gp120-specific antibody glycosylation profiles among a small group of HIV infected patients with differential clinical progression profiles [15] . Similar differences in antibody glycan-profiles were observed within this larger patient population ( Fig 1B ) , with elevated agalactosylation among all spontaneous controllers and elevated bisection among the chronic treated patients . However , interestingly , no between HIV group differences were observed among the influenza-specific antibody glycan profiles ( S3 Fig ) , demonstrating disease specific nature of antibody glycan tuning . While changes in antibody glycosylation and their effect on antibody effector function has been clearly illustrated in the context of monoclonal therapeutics [6] , less is known about the impact of glycan-structure changes , which are often small , in polyclonal antibody populations . Thus we next sought to define whether the observed glycan-profile differences impacted antibody effector function . Seven gp120-specific antibody effector functions , including: antibody dependent complement deposition ( ADCD ) , ADCC , NK degranulation associated CD107a surface expression , interferon-γ ( IFN-γ ) and macrophage inflammatory protein 1β ( MIP-1β ) release , ADCP and antibody dependent cellular viral inhibition ( ADCVI ) were assessed in a group of chronically untreated HIV infected patients for whom sufficient amounts of plasma were available for functional profiling . Thus gp120-specific antibody glyco-profiles and gp120-specific antibody functionality were assessed in parallel to determine the relationship between glycosylation and antibody functionality against the same antigenic target in polyclonal pools of antibodies . The glycan profile:function correlational analyses showed a number of relationships ( Fig 1C ) . Interestingly , known relationships previously demonstrated for monoclonal therapeutics [12 , 13 , 29] , such as the association between: 1 ) low fucose and high ADCC , high bisecting GlcNAc and high ADCC , and 3 ) low galactose and complement activation , were observed in polyclonal gp120-specific antibodies ( Fig 1C ) . Additionally , novel correlations were also observed including a significant positive association between di-sialylation and ADCC , agalactosylation and bisection were positively correlated with phagocytic activity , and mono-sialylation and di-galactosylation were associated with antibody mediated viral inhibition ( Fig 1C ) . These data suggest that polyclonal antibody glycan structure shifts clearly result in alterations in antibody effector function that may be actively regulated during an immune response to direct enhanced clearance and control in an antigen-specific manner . To gain a more complete understanding of the differences in glycosylation between different antigen-specific antibody populations , we used principle component analysis ( PCA ) to generate integrated multivariate glycan profiles for each antibody population . PCA linearly transforms multi-dimensional measurements into linear coordinates , called principle components . We can plot the first two principle components , which account for the greatest variation , onto a two-dimensional plot to define multivariate differences among groups ( overlapping dots reflect similar profiles whereas non-overlapping dots represent different overall glycan profiles ) . Specifically , PCA demonstrated that the three antigen specificities separated as distinct antigen-specific antibody glycan profiles ( Fig 1D ) , with limited overlap , suggesting that each antigen-specific antibody population is induced with a unique glycan profile . Moreover , the vectors on the loadings plot ( Fig 1E ) , illustrate the strength of the contribution of each glycan structure in driving the separation in the overall glycan profiles , highlighting the unique nature of selective enrichment of di-sialylated glycans among HA-specific antibodies , elevated G0F/G1B glycans among the p24-specific antibodies , and G1/G0FB glycans in the gp120-specific antibody population . Overall , these data strongly argue that antibody glycosylation is calibrated at an antigen-specific level within the same individual and that that each profile is distinct from one another and from bulk circulating antibodies ( Fig 1 ) . These data therefore suggest that antibody glycosylation may be tuned actively during the induction of an immune response , to generate an antigen/pathogen appropriate effector response . However whether antibody glycosylation can be actively manipulated is still unclear . To determine whether antibody glycosylation is programmable in a reproducible manner we turned to a vaccine trial using the experimental Ad26/Ad35 expressing an HIV Envelop protein A ( the B003/IPCAVD-004/HVTN091 trial ) that was conducted at sites in the United States , Kenya/Rwanda ( East Africa ) , and South Africa . As IgG bulk and antigen-specific glycosylation differ dramatically in inflammatory diseases [27] and infectious diseases [15] , bulk circulating antibody Fc glycosylation profiles were initially compared across the sites to ascertain baseline differences between the vaccine populations . Significant differences in bulk IgG Fc galactosylation and sialylation were observed among individuals in the three regions ( Fig 2A ) . In particular , individuals from both African regions exhibited significantly higher proportions of agalactosylated ( G0 ) Fcs . While both African groups displayed lower sialylation than US vaccinees , East Africans had the lowest bulk antibody sialylation . Given the role of low galactose and sialylation in determining the inflammatory activity of antibodies [10] , these data suggest that bulk antibody glycosylation in Africans is associated with inflammatory glycosylation , with East Africans having the most inflammatory profile . Since the bulk IgG population is made up of a large array of antigen-specific antibodies corresponding to the pathogens encountered by an individual , these differences may correspond to differences in genetic background , diet , and/or exposure to pathogens at each geographical location . As mentioned above , changes in fucose and b-GlcNAc alter antibody function [12 , 13] . Interestingly , in the South African cohort , we observed higher fucosylation of bulk Fc compared to either of the two other groups ( Fig 2A ) . Additionally , South Africans had lower b-GlcNAc compared to both groups . Given the low functional activity of antibodies with high fucose and low b-GlcNAc containing glycans , these results highlight that even within a single continent , significant differences may arise in antibody glycosylation of bulk antibodies , not only in sugars that modulate inflammation , but also among sugars that are critical for driving antibody functionality . Principle component analysis demonstrated largely non-overlapping antibody glycosylation profiles among vaccinees at each of the three sites in their bulk circulating antibody glycosylation profiles ( Fig 2B ) , demonstrating that fundamentally different glycosylation profiles exist within each geographic region . The separation was largely driven by inflammatory glycan structures as shown by the length of the vectors on the loadings plot . For example , G0 structures , associated with inflammation [17] , were largely associated with Kenya/Rwanda , whereas G2 structures , that are thought to be less inflammatory , were enriched among vaccinees from the US . Interestingly , the East African vaccinees separated completely from the other groups , while the South African and American recipients overlapped slightly in their bulk Fc-glycan profiles , suggesting that while variation in bulk Fc glycosylation exists among all groups , there are greater differences between East Africans and the other two populations , which are potentially related to a multitude of variables including distinct genetics [25] , endemic infections [30] , differences in diet [31] , and other environmental factors . Overall , these data strongly suggest that baseline inflammation may alter the bulk antibody glycan profiles , potentially pre-determining the glycan profile of newly elicited antibodies . We next aimed to determine whether antigen-specific antibodies induced via vaccination exhibited distinct glycosylation profiles to bulk circulating antibodies , in a manner analogous to the distinct antigen-specific antibodies seen in HIV-infected subjects ( Fig 1 ) Thus , the viral vector–induced antigen-specific antibodies were enriched from bulk IgG using vaccine-matched gp120 antigens in a subset of vaccinees from each region . Strikingly , the viral vector–elicited antibodies were significantly different from the bulk antibody glycans by PCA analysis ( Fig 3A ) . Specifically , no overlap was observed in the overall glycan profiles of antigen-specific ( maroon ) and bulk antibodies ( blue ) ( Fig 3A , left ) . Furthermore , all glycan types were found to be significantly different in vaccine-specific antibodies compared to bulk antibody glycosylation , with vaccine-specific glycans being significantly more agalactosylated , mono-galactosylated , sialylated and bisected but less fucosylated and di-galactosylated ( Fig 3B ) . As illustrated in the loadings plot ( Fig 3A , right ) , this separation was driven primarily by differences in fucosylation and sialylation . This analysis shows that antigen-specific antibody glycosylation is specific and driven by the immune signals delivered at the time of immunization . To ultimately determine whether antibody glycosylation is programmed at an antigen-specific level in a reproducible manner independent of regional differences in bulk circulating Fc glycosylation , we next aimed to determine whether the regional differences observed in bulk Fc glycosylation were also present in antigen-specific antibody glycoprofiles . Remarkably , no differences among all three geographical locations were observed for any sugar in the antigen-specific antibody glycans ( Fig 3C ) , strongly arguing that the viral vector-based vaccine induced a single glycan profile on vaccine-induced antibodies that is independent of pre-existing antibody glycan profiles . While a trend toward reduced addition of the b-GlcNAc was observed in the antigen-specific antibodies in the Kenyan/Rwandan vaccinees , these levels were opposite to those observed in the bulk antibody profiles ( Fig 2A ) , arguing for a directed change away from the bulk antibody glycan profiles among the vaccine-induced antibodies . In addition , as expected , HA-specific antibody glycans were different across regions at each timepoint tested ( cross-sectionally ) , but did not change due to vaccination ( Fig 3D ) , indicating that vaccine-induced glycan changes are restricted to the vaccine antigen-specific antibody population , and does not globally affect the humoral immune response . Thus collectively , these data strongly argue that vaccine-induced glycosylation changes are highly restricted to the de novo-induced antigen-specific antibody response independent of baseline geographical regions in antibody glycosylation , suggesting that antigen-specific antibody glycosylation is induced at the time of B cell priming in a uniform and directed manner . Given that the adenovirus-vectored vaccines elicited similar glycosylation patterns on vaccine-specific antibodies in all tested vaccinees , we aimed to determine whether the observed antigen-specific glycan profiles are conserved across all de novo-induced antibodies or if they are tuned differentially by distinct immunogens at the time of vaccination . Antigen-specific glycan profiles of antibodies isolated from recipients of the VAX003 vaccine trial ( an alum-adjuvanted recombinant gp120 vaccine that induced high titer vaccine-specific antibody responses ) were compared with glycan profiles induced in the B003/IPCAVD-004/HVTN091 experimental vaccine trial . Interestingly , using PCA , we observed significant separation of the gp120-specific antibody glycosylation profiles across the two vaccines for galactose , sialic acid , and b-GlcNAc ( Fig 4A ) . Specifically , the B003/IPCAVD-004/HVTN 091 vaccine trial induced more anti-inflammatory glycan structures , with higher proportions of di-galactosylated , and sialylated glycans compared to antibodies induced in the VAX003 trial ( Fig 4B ) . The viral vector used in B003/IPCAVD-004/HVTN 091 also induced an increased proportion of bisected glycan structures , which are known to elicit greater ADCC activity and therefore increased functionality [13] . However , no differences were observed in fucose content across the two vaccine trials . Thus , the viral vector used in B003/IPCAVD-004/HVTN091 induced a less inflammatory but highly functional antibody glycan profile compared to the poorly functional but highly inflammatory antibody glycan profiles induced in the alum-adjuvanted VAX003 study . All together , these data demonstrate that different immune signals delivered at the time of antigen exposure can specifically tune the glycosylation of antigen-specific antibodies in a signal-specific manner . Furthermore , the same antigen , HIV gp120 , can induce differentially glycosylated antibodies in the presence of distinct immune signals . Given the importance of glycosylation in modulating antibody effector functions , these data suggest that inflammatory signals at the time of B cell priming can direct antibody glycosylation , aimed at tuning antibody effector function to target antigens in a pathogen-specific manner . Importantly , these data highlight that next-generation vaccine design strategies may selectively tune antibody effector function via modulation of antibody glycosylation . Unlike subclass selection , which irreversibly changes the constant domain , antibody glycosylation represents a flexible and powerful mechanism by which the immune system naturally finely tunes antibody effector function . However , while significant changes in antibody glycosylation have been reported on bulk circulating antibodies in the setting of chronic inflammatory diseases [27] and on antigen-specific antibodies [15 , 32] , it is still unclear whether the immune system naturally and selectively tunes antibody glycosylation in an antigen-specific manner . Here , we show differential glycosylation on distinct antigen- and pathogen-specific antibodies isolated from the same individuals ( Fig 1 ) , suggesting that the selection of antibody glycan profiles may be determined at the time of B cell priming as a means to specifically tune antibody effector activity to eliminate individual targets in an antigen/pathogen-appropriate manner . Moreover , we show that antibody glycosylation can be actively influenced via vaccination , overcoming different baseline circulating antibody glycome differences among vaccinees ( Fig 2 ) , to generate a specific antibody glycan profile within the vaccine-specific antibody subpopulation ( Fig 3 ) . Finally , distinct differences were observed in antibody glycan profiles among antibodies induced by different vaccines ( vectored versus protein-only ) , highlighting the critical nature of distinct priming signals in directing the glycan profiles of antigen-specific antibodies ( Fig 4 ) . Given that antigen-specific antibodies represent only a small percent of the total circulating antibodies , which are composed of swarms of distinct epitope-specific antibodies , it is unlikely that antigen-specific antibody glycan shifts would influence the overall circulating glycome . However , differential antigen-specific antibody glycosylation clearly reflects differences in selective immune programming directed against distinct pathogens/antigens aimed at harnessing the broad Fc effector functional potential of the humoral immune response . The differential selection of antigen-specific antibody glycosylation profiles may be a means by which the humoral immune system customizes and selectively arms antibodies with extra-neutralizing effector functions that are more effective in controlling and clearing particular pathogens . While the magnitude of the observed glycan changes appear small , they are highly significant , and even small shifts were clearly associated with robust changes in antibody effector functions ( Fig 1C ) . Given that antibodies function as polyclonal swarms in immune-complexes , small glycan changes may have profound effects on innate immune Fc-effector cell functionality by simply skewing Fc-receptor/complement activation towards more desirable functions . Interestingly , some of the associations reflect previously identified relationships described for monoclonal therapeutic functional enhancement ( low fucose = ADCC ) . However , additionally , novel glycan profile shifts that tracked with enhanced viral inhibition ( G1 ) , phagocytosis ( G0 ) , ADCC ( SA ) , and complement activation ( F and SA ) . Additionally , HA-specific antibodies exhibited more galactosylated , sialylated ( less inflammatory ) , and afucosylated ( more ADCC ) glycan profiles ( Fig 1A ) , suggestive of an antibody glycan profile tuned to promote rapid NK cell cytotoxicity via FCGR3A [12] in the absence of high levels of inflammation [33] . By contrast , gp120-specific antibodies exhibited a more inflammatory profile marked by low galactosylation and sialylation with higher levels of fucosylation and b-GlcNAc incorporation , the latter of which has been implicated in driving ADCC and complement activation [12 , 13] . Because the containment of influenza virus occurs predominantly within the lung [34] , where rapid killing in the absence of excessive inflammation may be required to avoid immunopathology , an anti-inflammatory but ADCC-inducing antibody glycan would be highly protective . Conversely , non-neutralizing control of HIV may occur both in the blood as well as in lymphoid and gut tissues [35] , where excessive pathology has been documented [36 , 37] . However , whether less inflammatory and functional antibodies targeting HIV could provide enhanced control of HIV is unclear . Defining protective glycan structures against other pathogens that infect via mucosal tissues may point to potential antibody effector profiles that could provide enhanced early containment of HIV . Yet , differences were observed in antigen-specific antibody glycosylation , strongly suggesting that the immune response evolves a highly specific antibody glycoprofile to target each pathogen , engineering the correct glycans to generate antibodies that are the most functional for particular pathogen . Thus , since the choice of N-linked glycan on the antibody is both directable and crucial to determining the functionality of antibodies , a deeper understanding of how and why particular antibody glycans are built up is crucial for understanding mechanisms to actively manipulate and control the bioactivity of antibodies via vaccination . Moreover , as vaccine-specific antibody glycosylation profiles were similar within vaccinees , irrespective of their baseline bulk antibody glycosylation profile , suggests that B cell programming of antibody glycosylation must occur in an environment that is unaffected by baseline inflammatory differences that may be driven by different diets , microbiomes , or co-infections , [25 , 30 , 31] . Furthermore , alterations in antibody glycosylation during vaccination clearly occurs in an antigen-specific manner , as HA-specific antibody glycosylation profiles were unaltered , suggesting that antibody glycosylation must require some level of B cell receptor triggering . Yet , because distinct vaccinations elicited discrete antibody glycan profiles , it is likely that unique inflammatory signals , delivered via TLRs or cytokines , at the time of BCR engagement , must play a critical role in tuning the antibody glycoprofile towards particular effector function , potentially allowing B cells to integrate information and program function according to the microbial origin of the antigen to which they are selected . Previous genome wide data have identified a small number of single nucleotide polymorphisms that track with altered systemic antibody glycosylation [38] . However , here we observed highly significant regional differences in the circulating antibody glycome , that extend far beyond the frequencies at which these SNPs occur within populations , suggesting strongly the critical importance of environmental influences on modulating and shaping circulating humoral immune profiles . Whether these antibody-glycome profiles are imprinted gestationally , related to nutritional differences or linked to differences in co-infections remains to be determined . Furthermore , whether this profile is reversible ( upon relocation ) , and if it fluctuates over time also remains unclear . Given the critical nature of non-specific antibody glycosylation in therapeutically reducing inflammation , as is observed with IVIG [19 , 20] , understanding the key modulators of bulk glycosylation may not only help tune immunologic inflammation but also provide insights into populations at risk for particular infection , malignancies , or autoimmune disease . Irrespective of the baseline circulating antibody glycome differences , vaccination resulted in the rapid selection of a single vaccine-specific glycan profile , suggesting that vaccination can select for uniform protective humoral immune profiles globally . However , whether these antibody glycan profiles change over time and whether they are preserved and can be recalled is unclear . Because humoral immune responses are composed of different waves of plasmablasts , it is plausible that the peak vaccine-specific antibody glycan profile may represent a particular wave of glycan profiles that later matures over time , concurrent with the loss of specific plasmablast populations . Thus future longitudinal vaccine studies may provide enhanced insights into the specific antibody-glycoprofiles that may be seeded into the long-lived plasma cell pool , which are aimed at conferring immunologic memory until pathogen re-exposure . Moreover , understanding how antibody glyco-profiles may be tuned in memory-B cells that give rise to new waves of plasmablasts , through prime-boosting or via reaction to distinct adjuvants , offers a unique opportunity to fine tune antibody effector activity . Thus these data highlight the critical need to more fully understand the signals and mechanisms that program glycan profiles naturally and via vaccination to define the specific vaccines , adjuvants , and/or vectors that can selectively induce glycan perturbations linked to desired antibody effector functions . This study is the first to clearly demonstrate that the immune system actively tunes antibody glycosylation in an antigen-specific manner via the delivery of specific signals at the time of antigen exposure , in a BCR-dependent manner , independent of baseline inflammatory differences . Whether host-genetics or baseline risk factors may impact B cell programming and thereby the ability to generate specific antibody glycan profiles is unknown . However given the emerging interest in modulating antibody effector activity against HIV , and other pathogens , novel strategies able to actively modulate antibody effector function in vivo are highly desirable . Thus the data presented here demonstrate that vaccination can harness antibody effector function via the regulation of antibody glycosylation offering a novel route to improve future vaccines that may provide protection from HIV and beyond . HIV-positive patients were recruited through the Ragon Institute at Massachusetts General Hospital . A total of 197 HIV-positive subjects , balanced for sex and age , were included in this study , including 48 elite controllers ( <75 copies RNA/ml ) [39] , 64 viremic controllers ( 75–2000 copies of RNA/ml ) , 44 HIV-positive patients on antiretroviral therapy ( <70 copies RNA/ml ) , and 41 untreated HIV-positive patients ( >70 copies RNA/ml ) . The B003/IPCAVD-004/HVTN 091 vaccine trial ( clinicaltrials . gov ID: NCT01215149 ) was a safety and immunogenicity trial of a combination of adenovirus vectors ( Ad26 and Ad35 ) expressing gp120 ENVA . Adenoviral vectors were administered at seven sites in three regions: the United States , East Africa , and South Africa . Low-risk , HIV-negative adults received two intramuscular doses of one or the other vector , and all samples used in this study were collected at peak immunogenicity , two weeks after final immunization ( manuscript in preparation ) . The VAX003 trial was a phase III efficacy trial administered in Thailand in a high-risk population of intravenous drug users ( clinicaltrials . gov ID NCT00006327 ) . This trial used seven doses of AIDSVAX B/E , a recombinant gp120 clade B/E , with alum as the adjuvant . All samples used in this study were collected at peak immunogenicity , two weeks after the final vaccination . The study was reviewed and approved by the Massachusetts General Hospital Institutional Review Board , and each subject gave written informed consent . All samples were collected and used with approval from local the Institutional Review Boards and appropriate national regulatory authorities [18] . The HIV positive patient cohorts were approved by the Massachusetts General Hospital institutional review board . The ICPAVD 004 trial was approved by the Harvard Medical School institutional review board and each subject gave written informed consent . The VAX003 study was approved by the Bangkok Metropolitan Administration institutional review board and each subject gave written informed consent . Plasma was collected by the vaccine trial staff and stored at -80°C until use . IgG was isolated using Melon Gel IgG purification resin ( Thermo Fisher ) according to the manufacturer's instructions . Fc glycans were released and analyzed as described [40] . Briefly , whole antibodies were treated with IdeS protease ( Genovis ) to separate the Fab from the Fc and Fc glycans were released by PNGaseF and then dried down prior to fluorescent labeling with 8-aminopyrene-1 , 3 , 6-trisulfonate . Labeled glycans were run on a capillary electrophoresis machine and glycan peaks were assigned using labeled standards . A representative glycan spectra is depicted in S1 Fig and a table of peak assignments and names are listed in S1 Table . Isolated IgG was passed over gp120 embedded columns ( YU2 for HIV-positive subjects , EnvA for IPCAVD vaccinees , and a 1:1 mix of A244 and MN for VAX003 vaccinees; Immune Technology Corp . ) , p24 ( HXBc2; Immune Technology Corp . ) or HA ( mix of HAΔTM H1N1 A/Solomon Islands/3/2006 , HAΔTM H3N2 A/Wisconsin/67/x161/2005 , HAΔTM H1N1 A/Brisbane/59/2007 , HAΔTM H3N2 A/Brisbane/10/2007 , HA1 H1N1 A/New Caledonia/20/99 , HA B/Malaysia/2506/2004/0054P and HAΔTM B/Florida/2006; Immune Technology Corp . ) . The bound antibody was eluted from the column in 0 . 1 M citric acid , pH 2 . 9 . The purified IgGs were treated with PNGaseF to release the attached N-linked glycans for analysis . To compare these antibodies to the bulk fraction , a subset of bulk IgG was processed without Fc separation as described above . Functional profiling was performed as described previous [41] Briefly , bulk IgG from chronic untreated HIV patients was purified using a Melon Gel IgG purification Kit ( Thermo Scientific ) . Complement activation was measured via the recruitment of C3d to CEM-NKR target cells by flow cytometry following antibody labeling of gp120-adsorbed target cells . ADCC was quantified following the elimination of fluorescently labeled antibody-coated-gp120-adsorbed CEM-NKR cells by purified primary NK cells [8] . NK cell activation was assessed following the addition of purified NK cells to antibody adsorbed gp120-coated 96-well plates by flow-cytometry as the frequency of NK cells degranulating ( CD107a upregulation ) , chemokine ( MIP-1β ) or cytokine ( IFN-γ ) secreting cells [8] . For ADCP was measured as the level of antibody-induced-gp120-coated fluorescent bead uptake by flow cytometry by THP-1 cells [42] . Finally , ADCVI was quantified as the difference in HIV-JRCSF replication in the presence or absence of antibodies in activated CD4 T cells in the presence of autologous primary NK cells and antibodies over 7 days [43] . Univariate data were analyzed using GraphPad Prism Version 6 . 0e for Mac ( GraphPad Software , San Diego , California ) for statistical significance and graphical representation . The statistical tests used are indicated for each figure . Heat maps were constructed using GENE-E ( Brode Institute , Cambridge , MA ) . Multivariate analyses were performed using MATLAB and Statistics Toolbox Release 2013b ( the MathWorks Inc . , Natick , Massachusetts ) and JMP Pro 11 . 00 ( SAS , Cary , North Carolina ) .	Accumulating evidence points to a critical role for non-neutralizing antibody functions in protective immunity against a variety of pathogens , including HIV . Non-neutralizing antibody function is controlled by antibody constant domain interactions with Fc receptors , which itself is regulated via changes in antibody subclass/isotype selection or antibody glycosylation . This study specifically aimed to determine whether glycosylation of IgG is naturally tuned to target distinct pathogens or antigens and whether this activity can be actively modulated to direct antibody effector function . The study clearly demonstrates that the immune system naturally exploits unique IgG glycosylation profiles to target distinct pathogens and antigens and that this activity can be actively manipulated via vaccination . Moreover , because different vaccines drive unique glycosylation profiles , future studies that define the specific signals that control antibody glycosylation may lead to the generation of next-generation therapeutic interventions that can leverage and specifically direct the killing activity of the innate immune system , targeting HIV and beyond .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "hiv", "infections", "medicine", "and", "health", "sciences", "immune", "physiology", "pathology", "and", "laboratory", "medicine", "pathogens", "immunology", "microbiology", "retroviruses", "viruses", "immunodeficiency", "viruses", "vaccines", "preventive", "medicine", "signs", "and", "symptoms", "rna", "viruses", "glycosylation", "antibodies", "vaccination", "and", "immunization", "public", "and", "occupational", "health", "immune", "system", "proteins", "infectious", "diseases", "inflammation", "proteins", "medical", "microbiology", "hiv", "antigens", "microbial", "pathogens", "immune", "response", "biochemistry", "post-translational", "modification", "physiology", "viral", "pathogens", "biology", "and", "life", "sciences", "viral", "diseases", "lentivirus", "glycobiology", "organisms" ]	2016	Antigen-Specific Antibody Glycosylation Is Regulated via Vaccination
In the northwest of Ethiopia , at the South Gondar region , there was a visceral leishmaniasis ( VL ) outbreak in 2005 , making the disease a public health concern for the regional health authorities ever since . The knowledge on how the population perceives the disease is essential in order to propose successful control strategies . Two surveys on VL knowledge , attitudes and practices were conducted at the beginning ( May 2009 ) and at the end ( February 2011 ) of a VL longitudinal study carried out in rural communities of Libo Kemkem and Fogera , two districts of the Amhara Regional State . Results showed that VL global knowledge was very low in the area , and that it improved substantially in the period studied . Specifically , from 2009 to 2011 , the frequency of proper knowledge regarding VL signs and symptoms increased from 47% to 71% ( p<0 . 0001 ) , knowledge of VL causes increased from 8% to 25% ( p<0 . 0001 ) , and knowledge on VL protection measures from 16% to 55% ( p<0 . 0001 ) . Moreover , the improvement observed in VL knowledge was more marked among the families with no previous history of VL case . Finally , in 2011 more than 90% of the households owned at least an impregnated bed net and had been sprayed , and attitudes towards these and other protective measures were very positive ( over 94% acceptance for all of them ) . In 2009 the level of knowledge regarding VL was very low among the rural population of this area , although it improved substantially in the study period , probably due to the contribution of many actors in the area . VL patients and relatives should be appropriately informed and trained as they may act as successful health community agents . VL risk behavioural patterns are subject to change as attitudes towards protective measures were very positive overall . Visceral leishmaniasis ( VL ) ( also known as kala-azar ) is a vector-borne neglected disease caused by the protozoan parasite Leishmania donovani in East Africa , and transmitted by the bite of female phlebotomine sand fly . Clinical signs and symptoms often include long lasting and irregular fever , weight loss and hepato-splenomegaly; and it is fatal if left untreated [1] . More than 90% of global VL cases occur in six countries: India , Bangladesh , Sudan , South Sudan , Ethiopia and Brazil . Globally , 200 , 000 to 400 , 000 new cases of VL occur every year , and only in Ethiopia it is estimated an annual incidence of 4 , 000 new cases [2] . The principal foci in Ethiopia are the one in the Northwest border with Sudan ( Metema and Humera ) , and the one located in the South , in the Segen and Woito river valleys [3]–[6] . In Libo Kemkem and Fogera ( highland districts in South Gondar , Amhara Regional State ) VL had never been reported until May 2005 when a large VL outbreak was identified , with more than 2 , 500 cases treated . A high mortality rate was reported initially , probably due to the long time required for the recognition of the epidemic [7] . Migration of laborers coming from endemic neighboring areas ( border of Sudan ) is one of the hypotheses for the introduction of VL in the region [8] , [9] that has become a public health concern for the Amhara Regional State Health Bureau ever since . In order to elaborate successful VL control programs it is essential to know the risk factors associated with it , and to understand the disease-related knowledge , attitudes , and practices ( KAP ) of the population [10] . The factors associated with Leishmahia infection in this area have already been described , being related to past history of VL in the household , house conditions or behaviors like sleeping outside , among others [11] . The factors associated with the VL clinical manifestation in this area were sleeping outside or under an acacia tree were among others [8] . However , little is known about how individuals in rural communities of this region perceive the disease and its management . There is a paucity of VL KAP studies in the New World [12] , [13] and in the Old World [14]–[18] in general . And in Ethiopia , to the best of our knowledge , there are no published studies that have focused on these aspects in a rural setting . Only recently it has been published a VL KAP study conducted in Addis Zemen , the urban centre of Libo Kemkem [19] . We expect our study , focused in the rural , to contribute to those urban results , in order to help the Amhara health authorities to promote the involvement of the communities in the control of the disease , a priority for the government of Ethiopia [20] . Health education campaigns should be adapted , in contents , type , and format to the target population [1] , [21] . In other settings it has already been proven that educational strategies with informative materials can contribute to VL control programs [22]–[24] , but written materials in rural communities of Ethiopia with high levels of illiteracy may not be appropriate . In the area of study , since the 2005 outbreak , there have been different actors implementing outreach activities with health education and case screening . And the research study that we conducted included informative and sensitization talks , which may be more appropriate for this population . By carrying out two KAP surveys at the beginning and at the end of the longitudinal study we look forward to assess baseline VL knowledge attitudes and practices , as well as the change in VL knowledge along the study period . Furthermore , as results from other Leishmaniasis KAP studies have suggested that the knowledge of the disease is restricted to those that have suffered from it personally or in a person closely related [25] , we wanted to differentiate the results regarding VL knowledge by households with and without a positive history of VL . Therefore , the aims of this study are 1 ) to assess the knowledge , attitudes and practices of VL in households of a rural endemic area of Amhara Regional State , Ethiopia and 2 ) to evaluate the impact of community interventions in the VL knowledge at household level between 2009 and 2011 , taking into account the previous VL history of the participant households . The study was approved by the ethical advisory boards of the Armauer Hansen Research Institute and the Ethiopian National Ethical Committee in Ethiopia , and the Instituto de Salud Carlos III in Spain . Support letters were obtained from the Amhara Regional State Health Bureau and the different districts' Health Offices . All parents/guardians gave written informed consent prior to responding to the questionnaires directed to them and to the enrolment of their children in the study , and assent was also obtained from children ≥11 years of age . The area of study was located in the Amhara Regional State , South Gondar , Northwest Ethiopia ( See Custodio et al . [11] for geographical location of study site ) , and comprised two districts ( weredas ) mainly rural: Libo Kemkem ( being Addis Zemen town its capital ) and Fogera ( being Woreta town its capital ) . These are adjacent districts most affected by the outbreak of VL occurred in 2004–2005 [26] . According to year 2009 census , the population was 198 , 374 and 226 , 595 for Libo Kemkem and Fogera respectively . The KAP surveys presented in this work were carried out within the framework of a prospective longitudinal study entitled “Visceral Leishmaniasis and Malnutrition in Amhara State , Ethiopia” . The study involved a cross sectional survey conducted in May 2009 to estimate the prevalence and associated factors of VL and malnutrition in school-aged children , and a cohort study that was carried out between May 2009 and February 2011 in order to elucidate the relationships between malnutrition and Leishmania infection in this same age group . The study consisted of four surveys that were carried out in May 2009 , December 2009 , May 2010 and February 2011 . In the first and last surveys questions related to the knowledge , attitudes and practices ( KAP ) towards visceral leishmaniasis and Leishmania transmission were addressed to the care providers ( present at the household at the time of the survey ) of the children participant in the cohorts' study . Population sampling was carried out by multi-staged cluster survey being the primary sampling units the sub-districts ( kebeles ) with high incidence of VL: Bura , Yifag Akababi , and Agita from Libo Kemkem; and Sifatra and Rib Gebreal from Fogera . Secondary sampling units were randomly selected villages ( gotts ) in each of the selected sub-districts , and third sampling units randomly selected households in each of the villages . The sample size was calculated according to the objective of the original project , described in detail elsewhere [9] , [11] , [27] . In May 2009 the care providers of the children recruited for the cohort study were interviewed by trained local personnel using a standardized structured questionnaire that included questions on demographics , household characteristics , child health , VL risk factors and VL KAP . A question regarding if someone in the household had suffered VL in the past was included in order to elaborate the variable Household ( HH ) with positive history of VL . This variable was based on the interviewee's report , but not verified by treatment or medical record . The use and the number of bed nets owned by the household was also reported but not verified by the interviewer . In February 2011 the same households were visited , and the interview consisted of a standardized questionnaire that included questions related to child health and a more extensive VL KAP . However , the question regarding if someone in the household had suffered VL in the past was not included in this last interview . Care providers present in the house at the time of this visit were not necessarily the same who were interviewed in the first survey ( only 40% of them were the same person in the May 2009 and in the February 2011 surveys ) . Therefore , we assess knowledge at household level and not at the individual one . Out of the 276 households visited in May 2009 we were able to collect data on 218 when revisited in February 2011 , which , for the purpose of this study , were the ones to be kept in the analysis . All questionnaires were translated in to Amharic , the main local language . The outcome variables regarding “Awareness” were based on self-perceived knowledge related to VL sings and symptoms , causes , or protective measures respectively ( as an example: Do you know any sign or symptom of VL ? Yes/No//If yes , which ones ? ) . The different answers were thereafter converted into dichotomous variables . And the outcome variables regarding “Proper knowledge” were created as follows: Before the starting of the project , a two days consultation was made with the local community leaders , sub-district ( kebele ) , and district administrators in order to approach why the project was relevant , what was the VL situation in the area , and also to cover VL general information . During the cohorts study four surveys were conducted . The day before visiting the community for the first data collection ( May 2009 survey ) , the supervisor together with the kebele administrator and the community leader conducted a one day sensitization talk to every elder in the community . And before each of the following surveys the supervisor talked to the household head or adult present in the house at the time of the visit . The talks covered VL general signs and symptoms as well as Leishmania infection ways of transmission and protective measures . In addition , in January 2011 a special informative meeting was held with the community leaders of all participant gotts and with kebeles administrator in order to promote leader's encouraging to families to participate in the fourth and last survey of the project . During the data collection process questionnaires were checked on site by the supervisor and , once they were completed , were submitted to the data processing unit of the Armauer Hansen Research Institute ( AHRI ) in Addis Ababa , Ethiopia , where they were double entered in ACCESS and cross checked for consistency . Joint data analysis was conducted in the Spanish National Centre of Tropical Medicine in Madrid , Spain , where data was rechecked and cleaned . Finally , data analysis was performed using STATA version 11 ( Stata Corp . , College Station , TX , USA ) . Descriptive statistics were performed and the Chi square test was used for comparisons between HH with and without positive history of VL , except when the number in any of the categories analysed was below 5 , that the Fisher's exact test was applied . Differences in results pre ( 2009 ) and post ( 2011 ) implementation of the study were examined by the McNemar test for matched data . All p-values were two tailed and a p-value of ≤0 . 05 was taken as significant . A total of 218 households were surveyed , all of them from rural environment with uniform low socioeconomic conditions , described in Table 1 . In 2009 , the majority of the heads of the households were male ( 91 . 3% ) , illiterate ( 78% ) and had a principal occupation related to the cultivation of land ( 99 . 8% ) . Among those who had their own lands ( 97 . 7% ) the mean of acreage owned was 1 . 2 Ha ( SD: 6 . 7 ) and only 11% had more than 3 Ha . The mean household size was 6 . 1 persons ( SD: 1 . 7 ) , with households size ranging from 3 to 10 persons . Radios were present only in 28% of the households . The main source of knowledge regarding VL in 2009 , before the implementation of the study , was the health centre ( n = 57 , 26 . 1% ) followed by knowing someone that had suffered VL ( n = 21 , 9 . 6% ) . In 2009 , 47 . 7% of the population surveyed reported to be aware of VL signs and symptoms , versus an 84 . 7% in 2011 ( p<0 . 0001 ) . The most frequently reported signs and symptoms were abdominal swelling , fever , weight loss , and low appetite . Furthermore , in 2011 a significantly higher frequency of interviewees had “Proper knowledge of VL signs and symptoms” as compared to 2009 ( 71 . 1% to 14 . 6% , p = 0 . 0001 ) . And , when proper knowledge on VL signs and symptoms was stratified by the variable if someone in the household had suffered VL in the past , a higher proportion of respondents living in houses with past history of VL reported correct signs and symptoms , being this difference more marked in 2009 ( 94% to 28% , p<0 . 0001 ) than in 2011 ( 89% to 64% , p = 0 . 001 ) ( Table 2 ) . Regarding self-perceived knowledge of the possible causes of the disease , a 16 . 5% reported to be aware of VL causes in 2009 versus a 58 . 7% in 2011 ( p<0 . 0001 ) . The answer considered appropriate , “Insect” , was the one most frequently reported in 2009 ( 8 . 3% ) and increased to 31 . 8% in 2011 ( p<0 . 0001 ) . Respondents living in houses with past history of VL reported more frequently a proper knowledge on the vector borne disease nature of VL than respondents living in houses with no history of VL , although this difference was only significant in 2009 ( p<0 . 001 ) ( Table 3 ) . In relation to VL protection measures , in 2009 only 21% of the respondents declared to be aware of how to protect from VL , in regard to 58% in 2011 , p<0 . 0001 . The most mentioned protection measures in both years were “Bed Nets” and “Environmental Sanitation” , but the probability of giving a correct answer regarding VL protective measures was almost 8 times higher in 2011 than in 2009 ( p<0 . 0001 ) . When stratified by houses with and without VL history , differences were found only in the responses of the 2009 survey , were proper knowledge on VL protective measures was reported more frequently among respondents of houses with positive history of VL ( 33 . 3% versus 9 . 0% , p<0 . 001 ) ( Table 3 ) . In relation to attitudes and practices , in 2009 57% of the houses reported to own bed nets versus a 98% in 2011 , p<0 . 0001 . The only reason given for not owning bed nets in 2011 was “Because it is difficult to get them” . The number of households owning two or more bed nets increased from 56 ( 25 . 7% ) in 2009 to 177 ( 81 . 2% ) in 2011 ( Table 4 ) . In the majority of the houses , ( n = 181 , 85% ) respondents reported that bed nets were used by all members in the family , followed by the option “Only adults” ( n = 14 , 6% ) and “Mother and children” ( n = 7 , 3% ) . Moreover , 94% of respondents stated that they would accept using impregnated bed nets in the house . In 2011 there were 143 ( 66% ) houses with iron roof , an increase from the 134 ( 62% ) in 2009 but no significant ( p = 0 . 06 ) ( Table 4 ) . The main reasons reported for using iron roof were because it was more solid ( n = 60 , 40% ) and better for weather conditions ( n = 51 , 36% ) . The reason for using straw ( n = 74 , 33% ) , the alternative roof material , was its lower price ( n = 57 , 77% ) . In relation to house conditions , more than 90% of the houses surveyed ( n = 198 ) had cracks in the wall , and when the interviewees were asked about the optimal frequency for repairing them , the responses ranged from “Never” ( n = 15 , 6 . 9% ) , “Every 2 years or more” ( n = 32 , 15% ) , “Once a year” ( n = 56 , 26% ) , “More than once a year” ( n = 79 , 36% ) , to “Every month” ( n = 32 , 15% ) . In 2011 almost every surveyed house ( 96% ) had been sprayed compared to 64% in 2009 ( p<0 . 0001 ) ( Table 4 ) . The acceptance for indoor and outdoor spraying was very high ( 98% to 97% respectively ) . And so it was the acceptance for house surroundings environmental cleaning ( 99% ) . Of the houses surveyed , in 2011 , 63% ( n = 138 ) reported having members of the family sleeping outside , mainly due to far away herding or cattle watching in the house surroundings . However , 29 interviewees ( 21% ) reported that family members sleeping outside made use of bed nets , and 20 ( 15% ) , declared that other protection measures like cloths , blankets or environmental sanitation were used . The reasons reported for not using bed nets while sleeping outside ranged from “Bed nets are difficult to use when sleeping outside” ( 40% ) , “Lack/shortage of bed nets” ( 11% ) to “Bed nets are too expensive” ( 11% ) . Finally , more than 80% ( n = 174 ) of respondents declared that at least one member of the family rested under acacia tree , a risk factor for VL in the area , and the time of the day most frequently reported for doing it was during midday ( n = 144 , 83% ) . In 2011 , the first option for VL treatment was public health facilities ( n = 215 , 99% ) , and only 3 persons ( 1 . 3% ) mentioned home remedies or traditional healer as first choices , based on “Better to try home first” and “Fear of evil eye” respectively . The VL knowledge in the rural communities of Libo Kemkem and Fogera districts is globally poor , and it should be improved through community strategies . Recommendations are: 1 ) to conduct sensitization talks in the affected communities , 2 ) to instruct VL patients and relatives while their stay in the hospital so they can act as health agents in their communities and 3 ) to follow up the maintenance of bed nets and the use of any other prevention measure like household spraying or environmental sanitation as the high level of acceptance perceived suggests that changes in behaviour are possible .	Visceral leishmaniasis ( VL ) is a vector borne disease that can be fatal if left untreated . In northern Ethiopia there was a VL outbreak in 2005 , making the disease a public health challenge ever since . In order to promote the participation of communities in the control of the disease , it is essential to know how they perceive the disease and its management . There is a paucity of studies dealing with the knowledge , attitudes and practices ( KAP ) towards VL in the world in general and in rural Ethiopia in particular . We conducted two KAP studies at the beginning and at the end of a VL longitudinal study carried out between 2009 and 2011 . The project included VL community talks and sensitization , and there were other interventions implemented by different actors in this period . Our results showed that , among the rural communities surveyed , the knowledge regarding signs and symptoms , causes , and protective measures of the disease was very low . However , it improved substantially in the period studied , suggesting that knowledge was subject to change by community interventions . It also showed that VL patients and relatives can act as successful health agents and that the population had positive attitudes towards the implementation of preventive actions .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "medicine", "and", "health", "sciences", "behavioral", "and", "social", "aspects", "of", "health", "tropical", "diseases", "social", "sciences", "anthropology", "health", "care", "preventive", "medicine", "global", "health", "neglected", "tropical", "diseases", "infectious", "disease", "control", "veterinary", "science", "public", "and", "occupational", "health", "infectious", "diseases", "veterinary", "diseases", "zoonoses", "social", "anthropology", "socioeconomic", "aspects", "of", "health", "leishmaniasis", "biology", "and", "life", "sciences" ]	2014	Knowledge, Attitudes and Practices Related to Visceral Leishmaniasis in Rural Communities of Amhara State: A Longitudinal Study in Northwest Ethiopia
The phylum Nematoda is biologically diverse , including parasites of plants and animals as well as free-living taxa . Underpinning this diversity will be commensurate diversity in expressed genes , including gene sets associated specifically with evolution of parasitism . Here we have analyzed the extensive expressed sequence tag data ( available for 37 nematode species , most of which are parasites ) and define over 120 , 000 distinct putative genes from which we have derived robust protein translations . Combined with the complete proteomes of Caenorhabditis elegans and Caenorhabditis briggsae , these proteins have been grouped into 65 , 000 protein families that in turn contain 40 , 000 distinct protein domains . We have mapped the occurrence of domains and families across the Nematoda and compared the nematode data to that available for other phyla . Gene loss is common , and in particular we identify nearly 5 , 000 genes that may have been lost from the lineage leading to the model nematode C . elegans . We find a preponderance of novelty , including 56 , 000 nematode-restricted protein families and 26 , 000 nematode-restricted domains . Mapping of the latest time-of-origin of these new families and domains across the nematode phylogeny revealed ongoing evolution of novelty . A number of genes from parasitic species had signatures of horizontal transfer from their host organisms , and parasitic species had a greater proportion of novel , secreted proteins than did free-living ones . These classes of genes may underpin parasitic phenotypes , and thus may be targets for development of effective control measures . The vast majority of species are unlikely to be selected for whole genome sequencing , whatever their importance in terms of evolution , health and ecology . The few eukaryote species selected for such projects , despite their utility in laboratory investigation , are unlikely to be representative of the genomic diversity of speciose phyla . For example , Arthropoda and Nematoda have over one million species each [1] , [2] and the ∼20 genomes completed [3]–[7] or in sequencing will illuminate only small parts of their diversity . Expressed sequence tags ( ESTs ) have proved to be a cost-effective and rapid method for identification of the genes from a target species [8] . Although the largest EST collections have been generated primarily for the annotation of complete genome sequences ( e . g . human and mouse ) , more than half the sequences in GenBank's EST depository ( dbEST ) [9] are from otherwise neglected genomes . One phylum that has benefited from an EST sequencing approach is the Nematoda [10]–[13] . Nematodes ( or round worms ) are abundant and diverse in terms of biology and ecology [14] . They are ubiquitous members of the meiofauna and play a core role in nutrient recycling . Parasitic species of this phylum are the causative agents of six of the thirteen neglected tropical diseases which afflict around 2 . 7 billion people [15]–[19] . The diseases caused by nematodes are extremely varied , and include anaemia and malnutrition ( caused by hookworms such Ancylostoma ceylanicum ) , African river blindness ( caused by the filarial nematode Onchocerca volvulus ) and elephantiasis ( caused by the filarial nematode Brugia malayi ) . In terms of disability adjusted life years ( DALYS ) , the burden of lymphatic filariasis ( 5 . 8 million DALYs ) , onchocerciasis ( 0 . 5 million DALYs ) and intestinal nematode infections ( 3 million DALYs ) is significant . Among school aged children ( 5–14 ) , the impact of intestinal nematodes is even greater than malaria [20] . Parasites are also responsible for substantial losses in agriculture . Plant-parasitic nematodes , such as the root-knot nematodes ( Meloidogyne spp . ) , are major crop pathogens throughout the world , impacting both the quantity and quality of marketable yields , causing an estimated US$80bn in damage annually [21] , and parasites of livestock are the cause for severe economic losses . The fully sequenced genomes of the free-living nematodes C . elegans and C . briggsae makes the analysis of EST datasets from parasitic nematode species particularly informative , in that both elements of core biology and particular adaptations specific to parasitism can be investigated . Already more than a dozen species- or family-specific analyses of nematode EST datasets have been published , considering parasites of humans [22]–[24] , animals [25]–[27] and plants [28] , [29] . The first whole-phylum meta-analysis was based on 265 , 000 sequences from 30 species , defining 93 , 645 putative genes [11] . Surprisingly , 30–70% of each species' dataset was found to have no significant similarity ( as defined by BLAST searches ) with any other sequence either within or outwith the sampled nematodes . Do these sequences define new genes , with new functions in nematodes ? Or are they transcriptional noise derived from non-coding sequence with no functional significance ? The majority of functional annotations have been assigned through sequence similarity to other proteins [30] , and thus a large number of nematode proteins lack clues as to their importance to the organism's survival . In the absence of annotation , these data are limited in their practical use , for example , in identifying the lead novel targets for anthelminthic drugs . One indication of a gene's significance , in worm survival , is its presence in a number of nematode species . Proteins with essential , conserved functions will tend to be conserved between species , and thus will be members of protein families . Protein families restricted to the Nematoda , but found in a number of species , invite further study to reveal their function . Proteins often share local regions or similarity despite being non-orthologous [31] , with the interplay between these domains underpinning their function . There are a number of widely used protein domain databases [32]–[35] which provide domain models to search . In addition , it is possible to identify new domains through similarity searches [36] , and nematode-restricted novel domains may yield novel insights into avenues for control of parasites . EST datasets have been considered less than ideal for such analyses , due to the occurrence of frame-shifts , ambiguous base calls and untranslated regions [37]–[39] . However , coding regions can be accurately predicted from EST cluster consensuses using a hierarchical approach such as that employed by prot4EST [39] . A great deal of care must be taken when translating sequences that do not have sequence similarity to known proteins . ESTScan , incorporated in the prot4EST pipeline , locates ( and corrects ) coding regions through the identification of frames that have oligonucleotide frequencies resembling those of the training dataset . However , by definition few sequence data are available in the public repositories for neglected species such as parasitic nematodes . Here we have inferred protein translations for over 120 , 000 putative genes from EST data from 37 species of nematodes using both high quality codon usage tables for each species [40] and synthetic training sets . This protein dataset , NemPep3 , is employed here to investigate protein family ( NemFam3 ) and protein domain ( NemDom3 ) composition of nematodes , and presented in an online database NEMBASE3 . Our key findings are: Sequence data were sourced from EMBL/GenBank/DDBJ and from WormBase ( http://www . wormbase . org ) as follows: We used TRIBE-MCL to generate protein families from NemPep3 [50] . In TRIBE-MCL , the Inflation parameter defines the tightness of the clusters . No single Inflation parameter value will correctly return all protein families , just as no single molecular clock exists to describe the evolution of all genes . Therefore we repeated the clustering procedure over a range of values and recorded all the clusters , following a previous study of prokaryote proteins [51] . The input to TRIBE-MCL was an all-against-all BLAST report . The number of families generated varied from 42 , 865 to 71 , 867 . All five sets of protein families are stored in NemBase3 . We used NemFam3 to investigate how sampling from additional species affected the discovery of protein families , generating a “collector's curve” of discovery of novelty . First we took those families for which the only nematode species present was C . elegans . We then added new families identified in each species in turn , adding them in the approximate order of their phylogenetic distance from C . elegans: Rhabditoidea ( CBG; see Figure 1 for three letter species codes ) ; Strongyloidea ( ACP , AYP , HCP , NAP , NBP , OOP , TDP ) ; Diplogasteromorpha ( PPP ) ; Panagrolaimomorpha ( PTP , SRP , SSP ) ; Tylenchomorpha ( GPP , GRP , HGP , HSP , MAP , MCP , MHP , MIP , MPP , PEP , PVP , RSP ) ; Cephalobomorpha ( ZPP ) ; Ascaridomorpha ( ALP , ASP , TCP ) ; Spiruromorpha ( BMP , DIP , LSP , OVP , WBP ) ; Trichinellida ( TMP , TVP , TSP ) ; Dorylaimida ( XIP ) . All EST derived proteins were annotated with matches to the KEGG database [52] with a script developed in house which makes use of BLAST comparisons . We wanted to identify metabolic processes absent in C . elegans but present in other nematodes . To do this we compiled two separate lists of metabolites that are substrates of enzymes in C . elegans and in the other nematodes . This step was important to reduce redundancy , as more than one enzyme ( EC number ) can be assigned to the same step of a pathway . Next we compared the two lists and extracted those substrates missing from C . elegans , highlighting the enzymes that catalyse transformation of these molecules . The Enzyme Commission ( EC ) identifiers of these proteins were obtained through the KEGG database . Assignment of signal peptides was done using the SignalP3 . 0 web-interface [53] with the following parameters: organism group - eukaryotes; method - both neural networks and hidden Markov models; truncation - first 70 residues . We used three Boolean tests provided by SignalP3 . 0 to determine if a signal peptide was present: first ‘D’ must be true; secondly , we considered ‘Cmax’ and ‘Ymax’ , if both were true then we deemed this strong evidence and weaker evidence if only one category was true . Analyses of the secreted proteomes have been carried out previously for Nippostrongylus brasiliensis [54] and H . schachtii [55] . Compared with these studies , and despite using more conservative parameters , we identified a larger number of signal peptide-containing proteins in N . brasiliensis ( 96 were identified , compared with 87 from Harcus et al . [54] ) and H . schachtii ( 105 identified compared with 65 from Vanholme et al . [55] ) . This increase is likely to derive from more robust coding region predictions producing proteins that were more likely to contain the correct N-terminus . NemPep3 proteins were annotated with protein domains using existing domain databases ( PfamA and ProDom ) and by de novo identification of domains in unannotated sequence . NEMBASE3 is a relational database built using the PostgreSQL database manager ( http://postgresql . org ) . It holds all the data types described above , including sequences , clustering information , consensuses derived from EST clusters , peptide predictions , protein families and protein domains . All peptides have been annotated with extensive BLAST-based similarity data , as well as quality scored functional annotation ( GO , EC and KEGG identifiers ) derived from GOtcha [61] and annot8r [62] analyses . The database is available through the www using custom php scripts from http://www . nematodes . org/ . Coding regions for EST cluster consensuses derived from NEMBASE [12] from 37 species from the phylum Nematoda were predicted using prot4EST , yielding a total of 121 , 694 polypeptide sequences ( Figure 1 ) . For each species , specific codon usage tables [40] were used to reverse translate the C . elegans proteome , providing synthetic training-set transcriptomes ( see Methods ) . To assess the accuracy of synthetic transcriptomes , partial datasets built for C . elegans [39] were translated in a similar fashion . Comparison with a complete collection of coding sequences showed only a slight reduction in prediction using synthetic transcriptomes ( data not shown ) . Importantly , for most species the simulated training sets were more accurate than simply using the complete C . elegans or C . briggsae transcriptomes . The mean length of translation for the EST datasets ( excluding the caenorhabditids ) was 137 amino acids ( aa ) ( standard deviation 65 aa ) , and 84% of the bases in the EST cluster consensuses contributed to translations . The regions not covered are likely to be predominantly untranslated regions , as well as regions of low-complexity sequence . Previously , we have shown that the most accurate translations are obtained using similarity to a known protein or the prot4EST implementation of the ESTScan algorithm [38] , [39] . For most nematode species , over 90% of EST cluster consensuses were translated using these two methods ( Figure 1 ) . However , three Spiruromorph species had much lower rates of translation by these methods: Brugia malayi ( 71% translated using similarity or ESTScan methods ) , Onchocerca volvulus ( 78% ) and Wuchereria bancrofti ( 68% ) ( Figure 1; ‘percentage accepted’ ) . These low rates appear to arise from two features of these data . Firstly , a relatively low proportion ( ∼40% ) of these species' EST cluster consensuses had significant similarity to protein sequences in UniRef100 [46] . Secondly , only ∼54% of the novel sequences had compositions that matched models derived from known coding regions , simulated transcriptomes , or , in the case of B . malayi where a first pass annotation of the whole genome sequence is available [63] , an extensive transcriptome dataset . Our inability to derive high quality translations for a significant number of clusters from these taxa could be due to a major biological difference and to the quality of the training set used or to the quality of the sequence data . Other species that had similarly low proportions of sequence similarity matches , had higher rates of compositionally-identified coding regions ( e . g . Trichuris vulpis with 80% of the novel sequences translated by ESTScan and Meloidogyne javanica with 97% ) . The addition of a 12 , 000-transcript , orthologous training set [63] did not improve the proportion of B . malayi cluster consensuses that yielded a translation . For these three problem species , we noted that singleton cluster consensuses were much less likely to be robustly translated , but these species did not have an excess of singletons compared to the other nematodes . The proportions of ESTs lacking detectable coding regions were compared between the source cDNA libraries . Of 25 B . malayi libraries , five were significantly enriched for ESTs not translated ( G-statistic = 682; p≪0 . 001 ) . Two libraries from the eight available for O . volvulus and two for W . bancrofti were also shown to contain an excess of ESTs without a coding region . Strikingly , 93% of the untranslatable sequences from B . malayi came from the highlighted five libraries , while the O . volvulus and W . bancrofti libraries accounted for around 30% of each species suspect contigs . We conclude that some of the unique features of the three species' data derive from the relative quality of some cDNA libraries sampled . To ensure that subsequent analyses were performed on the most accurate collection of polypeptides , we excluded EST cluster consensuses that could not be translated with either the sequence similarity or ESTScan components of prot4EST . Addition of the proteomes from the fully-sequenced C . elegans and C . briggsae yielded a high quality dataset ( NemPep3 ) . The current release of NemPep , version 3 , includes 154 , 501 polypeptide sequences ( Figure 1 ) , with a mean length of 220 amino acids . NemPep3 is available for download from NEMBASE3 ( http://www . nematodes . org/nembase3/ ) . We used TRIBE-MCL [50] to derive putative protein families ( NemFam3 ) from NemPep3 . These families were compared to proteins from the UniProt database [46] to identify overlap with previously defined protein families . The results of the clustering algorithm , MCL , can be tuned with an Inflation parameter . In the context of protein clusters , this value determines how tight , or strict , the clustering is ( see Methods ) . No single parameter set for TRIBE-MCL can be used to accurately identify all ( or even most ) families and so we generated independent estimates at five different Inflation values . To simplify analyses presented here , we have examined in detail the 65 , 179 protein families generated using an Inflation value of 3 . 0 , the default used for the TRIBE-MCL database [64] . Despite having a large sample ( 37 species and over 150 , 000 individual sequences ) we found no evidence of having exhausted the diversity of nematode ‘protein space’ . There was a near-linear increase in the number of protein families identified with addition of sequences and species ( Figure 2 ) . This finding is congruent with that of Parkinson et al . ( 2004b ) but here we have used a rigorous protein family definition schema rather than simply BLAST matches . Analyses of complete prokaryote proteomes also show an increase in the number of novel proteins as further species are sequenced [65] , although as a proportion of all prokaryote proteins the number of novel proteins is decreasing [66] . This trend is not apparent in the nematode dataset ( Figure 2 ) . The distribution of size of the NemFam3 protein families can be described by a power law , matching that of many protein family databases ( Figure 3a ) [67] . We identified protein families that were restricted to all levels of nematode taxonomy , from species-specific to phylum-specific ( Figure 3b ) . By comparing NemFam3 families to proteins from non-nematode species , we divided them into three classes: NemFam3 families that were unique to the Nematoda ( region A of Figure 2 ) ; NemFam3 families that were not found in C . elegans but did have homologues in other phyla ( region B ) ; and NemFam3 families that included C . elegans members and had homologues in other phyla ( region C ) . Region C presumably encompasses proteins with core metabolic functions shared with other phyla . Gene loss is a common feature of genome evolution [68]–[70] . Gene gain by horizontal gene transfer is common in non-eukaryotes , but its role in eukaryotes , and particularly in metazoans , is still controversial [71]–[73] . Gene loss in C . elegans has been reported previously [6] , [74]–[76] . For example , orthologues of the Hox genes Antennapedia and Hox3 are absent in C . elegans but present in B . malayi and other invertebrates [76] . Comparison of C . elegans and C . briggsae [6] identified a large number of proteins in each species that did not have an orthologue in the other . Using NemPep3 and the UniProt database ( release 5 ) reduces the number of orphan proteins in C . elegans from 2 , 108 to 1 , 846 and from 2 , 141 to 1 , 961 in C . briggsae . Comparison with proteomes from additional Caenorhabditis sp . genomes currently being sequenced will clarify the patterns of gene gain and loss in this lineage . We identified 4 , 864 protein families ( containing 6 , 903 proteins ) that had significant sequence similarity to proteins from outside the Nematoda but that contained no C . elegans representatives ( ‘loss/gain candidates’ ) . To investigate the effect of using partial sequences , we compared loss/gain candidate EST cluster consensuses to the C . elegans genome . Thirty-nine loss/gain candidate families ( 92 sequences ) could be aligned to the genome ( using the program BLAT [77] ) and overlapped an annotated coding sequence: the failure of TRIBE-MCL to group the C . elegans proteins with their loss/gain candidate matches was because their BLAST alignments had a low similarity score . Three loss/gain candidate families ( eight proteins ) matched regions of the C . elegans genome that were not part of a coding region: these may correspond to valid but unannotated genes in C . elegans . Thus the majority of loss/gain candidate families are absent from C . elegans . Gene Ontology ( GO ) annotation of the loss/gain candidate families showed that a large number are involved in metabolism . One hundred and fourteen individual Enzyme Commission ( EC ) classifications could be assigned to 240 families ( a full list of these annotations is available in Table S1 ) . Some of these putative functions complemented gaps in the metabolic map of C . elegans . For example , C . elegans lacks a canonical DNA methylation pathway enzyme , cytosine-5′-methyltransferase [78] . Homologues of cytosine-5′-methyltransferase were identified in Ostertagia ostertagi , Teladorsagia . circumcincta and Xiphinema index ( Figure 4 ) , and a homologue has also been identified in Pristionchus pacificus [79] . It will be informative to examine additional nematode genomes for the features of DNA methylation and thus identify when , and perhaps why , this core regulatory mechanism was lost . While the above examples reveal the process of gene loss , we also identified putative gain of genes by horizontal transfer from other organisms ( Table 1 ) . Plant-parasitic nematodes modulate their host's metabolism and induce development of feeding sites ( for example induction of syncytia by cyst nematodes , and of giant cells by root-knot nematodes ) . These modifications involve the secretion by the nematode of exoenzymes such as pectinases , proteinases and cellulases ( reviewed by Vanholme and colleagues [80] ) . Putative effectors have been identified using directed cloning of nematode secretory gland products , including beta-1 , 4-endoglucanases from Globodera rostochiensis [81] , Heterodera schachtii [82] and Meloidogyne incognita [83] . Analyses of plant-parasitic EST data also identified beta-1 , 4-endoglucanase , beta-1 , 4-xylanases [84] and pectate lyases [85] . We identified two Meloidogyne orthologues ( M . javanica and M . hapla ) of a polygalacturonase previously reported from M . incognita [83] . Beta-1 , 4-endoglucanases were identified in seven species , including Pratylenchus vulnus . The enzyme's presence across most Tylenchid genera studied ( missing in the small Rhadopholus similis dataset ) suggests that the acquisition of this endoglucanase gene occurred in an ancient tylench ancestor . We identified seven additional protein families from plant parasitic nematodes that are similar to enzymes found in plants but not previously identified in non-nematode metazoans . The activities that may be carried out by these genes fall into two classes . Four genes , all from Tylenchomorpha , are enzymes that catabolise plant cell wall or starch carbohydrates ( polygalacturonase , beta-amylase and cellulase ) , and may mediate parasite modification or digestion of the root cell walls . Three genes , from the dorylaim X . index and the tylenchomorph M . incognita , encode activities that could modify plant signaling or second metabolites ( flavonol synthase , scopoletin glucosyltransferase and polyneuridine-aldehyde esterase ) , and may represent ‘anti-immunity’ mediators secreted by the parasite in order to subvert the necrotic or other responses of the host . Another mechanism of “gene gain” is de novo evolution of functional proteins . While it is clear that this mechanism has been active on the scale of phyla and kingdoms , its ongoing role in genome evolution is unclear [86] . We identified 56 , 407 protein families ( including 94 , 343 proteins ) restricted to nematodes ( NR families ) . Analyses of novel proteins in other species have shown that they are characterized by a significant reduction in average length compared to proteins with homologues in other taxa [65] . However , the average length of the NR family proteins ( 200 aa ) is only slightly shorter than those with homologues elsewhere ( 220 aa ) . It might be expected that novel genes would be expressed at low levels , and that they might thus be indistinguishable from aberrant transcripts from non-coding regions of the genome . Over 80% of the NR families contained an EST-derived sequence; not restricted to the caenorhabditids . Of these 69% were derived from a single EST ( data not shown ) . For loss-gain candidate protein families , 68% were derived from a single EST , while of families with matches in C . elegans and elsewhere , only 35% were derived from single ESTs . Thus , while the NR family sequences are expressed at low levels compared to core nematode genes , their expression levels are comparable to those of genes with wide phylogenetic distribution . We analyzed further the 2 , 098 NR families with at least five members . The number of NR families restricted to each taxonomic family or species correlated well with the depth of sequencing for each taxon ( Table S2 ) . We note that despite cogent evidence for gene loss in the caenorhabditids [74]–[76] , many NR families with a disjoint distribution in Nematoda are likely to be present in additional species , but as yet unsampled by ESTs . For example , 388 protein families ( 2 , 985 proteins ) were restricted to the complete proteomes of the caenorhabditids ( Family Rhabditoidea ) . The lack of homologues in other nematodes is likely to result in part from the depth of EST sampling , as only 1 , 385 ( 46% ) of these proteins had corresponding C . elegans ESTs ( out of 346 , 064 EST sequences ) . All nine nematode taxonomic families in this study had taxon-restricted protein families . For example , of 35 protein families that were restricted to Spiruromorpha , only three were species-specific ( one restricted to B . malayi and two to O . volvulus; data not shown ) . Fourteen of the spiruromorph protein families occurred in four species and one ( NemFam3 family 3 . 0_3062 ) contained all five species ( whose multiple sequence alignment is shown in Figure S1 ) . Many ( 630 ) NR protein families with at least five members did not contain a protein from the complete proteome of C . elegans . The processes of ‘gene invention’ ( and high rate of protein evolution ) are ongoing in Nematoda . Indeed , the preponderance of apparently species-specific proteins is just what we would predict from this process , given the pull towards new functions , and thus may not be simply due to lack of representation in EST data . However , compared with our previous analysis [11] , many sequences once thought to be species-specific now have inferred nodes of origin deeper in the nematode phylogeny , and we would expect this trend to continue as additional data are collected . It has been hypothesized that the secreted subset of parasitic nematode proteomes may be especially enriched in novel proteins , through rapid evolution to perform novel functions such as interactions with the host and other environmental challenges [54] , [55] . The protein families restricted to the nematodes were significantly enriched for signal peptides ( 19% ) compared to those that had homologues in other phyla ( 12% ) ( Figure 5 ) . Within the class of nematode protein families that did have homologues in other phyla ( non-NR ) , 2 , 490 proteins ( 28% ) were predicted to have signal peptides . Surprisingly , aligning these signal peptide-containing nematode proteins to homologues from other phyla revealed that 1 , 883 nematode proteins ( from 856 NemFam3 families , both NR and non-NR ) appear to have gained an N-terminal signal peptide . For two thirds of these protein families , C . elegans and C . briggsae proteins do not contain a signal peptide , suggesting that the acquisition of a signal peptide did not occur in the caenorhabditid lineage . The T . circumcincta proteome was the most enriched with signal peptides in both nematode-restricted and shared proteins . Mapping these T . circumcincta proteins onto NR families identified 48 strongylomorph-restricted families where signal peptide-containing proteins predominated . Despite the incomplete sampling of nematode protein space it is likely that many of these protein families are involved in specializations of the parasitic mode of life in strongylids . Domains are the basic functional and structural units of proteins and , while primary sequence diversity is expected to be huge , the diversity of domains has been predicted to be rather small [87] , [88] . As novel genes are being evolved in nematodes , we predicted that there might be de novo or accelerated evolution of protein domains . Identification of protein domains typically involves comparing sequences to a library of protein domain alignments [32] , [33] , [35] . These alignments are characterized either as hidden Markov models ( HMM ) or position-specific scoring matrices ( PSSM ) . Such an approach is well suited for full-length sequences , where a match , global ( i . e . full-length ) with respect to the domain , is usually considered necessary . However , proteomes derived from EST projects contain incomplete sequences , where only part of the domain is present making these global searches problematic . In particular it is difficult to robustly recognize domains that extend over the termini of partial translations . We devised a heuristic approach to assigning domain presence , based on different scoring thresholds available for domain models , in order to return a high coverage of domain annotation while keeping number of false positives to a minimum ( see Materials and Methods ) . The resulting nematode domain classification ( NemDom3 ) contained 39 , 944 unique domains ( Table 2 ) of which 2 , 593 were from PfamA and 10 , 684 from ProDom . The majority of these domains were derived from the complete caenorhabditid genomes , but more than half were found in the EST-derived proteome . Previously , 348 PfamA domains had been identified in non-caenorhabditid nematodes . We found 2 , 300 PfamA domain matches in the EST-derived proteomes of which 214 domains ( increased from thirteen ) were absent in C . elegans and C . briggsae . All but eight of these domains were exclusive to protein sequences that we had already identified as loss/gain candidates ( described above ) , including those restricted to plant-parasites: cellulase ( PF00150 ) and pectate lyase ( PF03211 ) . Of the eight domains identified in protein families that include Caenorhabditis sp . members , two of these , domains associated with the ribosomal large subunit protein 6 ( PF03868 ) and NADH:ubiqunione oxidoreductase ( PF08122 ) , have been reported in C . elegans [89] . However their sequences have been so diverged from the domain model as not to be recognized . Seventy-seven PfamA domains were found only in nematodes , with six found in species other than exclusive to C . elegans or C . briggsae ( Table S3 ) . With the exception of the abundant larval transcript ( ALT ) domain ( PF05535 ) , all nematode-restricted ( NR ) domains were first identified in C . elegans [90]–[93] . Surprisingly , we were able to expand the species-distribution in only 24 of the 77 domains . It is possible that the remaining NR domains are restricted to the caenorhabditid lineage . However , it is more likely that many , if not most , are present in other nematode species , but were not yet represented in EST data , or were not recognized by domain models that were too constrained . Inspection of the multiple sequence alignments of caenorhabditid-specific NR domains revealed often extremely high levels of identity . These alignments may generate hidden Markov models ( HMMs ) that cannot identify more divergent members . To illustrate this , we returned to the ALT domain ( PF05535 ) , which was , expectedly , identified in proteins from filarial species , but the searches did not find the known instance in C . elegans [90] , [91] . Using the Pfam alignment for this domain ( based on five filarial sequences ) , we constructed a PSSM and performed a RPS-BLAST search . This identified ALT domains in C . elegans as well as predicted proteins from Ascaris suum and A . lumbricoides . We defined over 23 , 000 protein domains seemingly unique to nematodes . Nearly half of these are found in non-caenorhabditid species . Many of these new domains are found as part of multi-domain architectures , with 15 , 152 ( 65% ) present with at least one different domain ( all classes ) and 6 , 625 associated with a PfamA domain . Profile searches with these novel domains ( see Methods ) identified 3 , 694 domains that matched non-nematode UniProt proteins . The most common distribution of these domains was the 270 domains found throughout the Ecdysozoa . However many domains had disjoint distributions , such as the 56 novel domains apparently exclusive to the nematodes and Viridplantae . Ten of these domains were found in plant-parasite nematode species ( Table 3 ) . The presence of putative homologues for three of these domains in C . elegans confuses of the issue of their origin . The absence of these domains in other metazoans suggests that they were either acquired through horizontal gene transfer or diverged from an ancestral nematode domain . Convergent evolution has been reported previously in nematodes [94] , [95] . Are these domains real , conserved units ? Of the 1 , 652 novel domains that were exclusive to the Spiruromorpha , 824 were found in at least two species of this taxon ( Table S4 ) . Of this latter set , 435 are associated with Pfam or ProDom domains . Being shared across a number of species suggests that these domains are likely to be functional . Hints as to their function may be derived from their associations with previously characterized domains , and from other high-volume datasets such as genome-wide RNAi screens and protein-protein interaction maps . The resources we have generated ( NemPep3 , NemFam3 and NemDom3 ) are presented in an interactive interface in the NEMBASE database at http://www . nematodes . org/ [12] . Release 3 . 1 of NEMBASE3 contains 128 , 709 EST clusters , and 31 , 461 , 090 annotations from 37 nematode species . Data in NEMBASE3 can be searched for individual ESTs , clusters , stage-specific and overall expression levels ( derived from EST counts ) , protein translations , domains , and families . Functional annotations ( Gene Ontology categories , Enzyme Commission numbers , metabolic pathways and best BLAST matches ) are also available . ESTs are typically used to annotate newly assembled genomes or provide snapshots of transcriptomes . Here we have shown that by both clustering ( creating a reference sequence or unigene set ) and careful translation , they can yield high quality partial proteome data . Importantly , the additional effort expended in deriving high quality translations is repaid in the increase in mean lengths of derived proteins , and in the increase in ascribable annotations . This is particularly evident in the correct identification of extended 5′ open reading frames from regions of lower quality EST sequence , and thus an enhanced ability to identify signal peptides ( Figure 5 ) . Issues of lack of relevant training data for model-based identification of open reading frames in neglected species can be overcome by bootstrapping BLAST-identified open reading frames to generate codon usage tables and synthetic proteomes . Comparison to the complete proteomes derived from genome sequence emphasizes the partial nature of EST-derived proteomes . Many genes with core roles in metabolism or signaling pathways are absent from the nematode partial proteomes , but this is likely to be due to lack of evidence rather than true loss . The EST-derived partial genomes systematically lack , or have very reduced , representation of some classes of genes . Thus , while the seven transmembrane helix class of odorant receptor gene is the most abundant gene family in C . elegans , homologues are conspicuously lacking from EST-derived proteomes . Indeed , even within the large C . elegans EST collection , no transcript is assigned to an odorant receptor . However , by comparison to complete genomes , EST-derived proteomes can be used to highlight gene loss events in fully sequenced species . Using this methodology we identified a significant number of gene families ( 4 , 800 ) absent in C . elegans but present in other nematodes and in other phyla . Some of these genes have likely been lost from C . elegans , as they have wide representation in other nematodes , and in non-nematode phyla . The loss of developmental pathway genes such as members of the Hox cluster , and of hedgehog homologues , has been associated with the evolution of a strict , lineage-based developmental control system in C . elegans . We identified additional losses of this type , including the loss of key DNA methylation genes . Other candidates for loss in C . elegans had a distinct pattern of presence in other phyla: they were found in only a restricted subset of nematode species and also in a disjoint group of organisms ( such as plants or bacteria ) . The limited occurrence of these genes is perhaps best explained by horizontal transfer from a host plant or other closely associated genome into the nematode genome . Notably , the proteomes of the plant parasitic Tylenchina contained genes of apparent plant or rhizosphere bacterial origins . Our analysis pushes the event ( s ) of acquisition of these classes of genes deeper into the tylenchine phylogeny , supporting the hypothesis that they may have been a key innovation leading to plant parasitism in the whole group . Another deeply sampled taxon was the medically important Spiruromopha . We have identified 35 protein families that are restricted to this lineage . Importantly , fourteen families had membership in at least four of the five species surveyed . These groups are ideal candidates for functional genomic and reverse genetic technologies that could reveal their function and importance to the survival of these parasitic worms , and thus whether they are possible targets for a next generation of anthelminthic drugs . Cross-comparison of the C . elegans and C . briggsae proteomes identified ∼10% of unique genes in each species . Throwing the draft B . malayi genome into the mix , revealed ∼40% of its proteins did not share homology to C . elegans , C . briggsae nor Drosophila melanogaster [63] . Adding partial proteomes from 37 additional nematode species reduced the number of private genes to ∼8% in each species . While we expect this proportion to decline as nematode EST sequencing continues , along with the release of genomes , we expect that each fully sequenced genomes has a significant complement of novel genes that have arisen since they last shared a common ancestor , less than 100 million years ago [6] , [96] , [97] . If this pattern is true of all the >1 million predicted nematode species , then ‘nematode protein space’ , the portion of possible sequence structures actually occupied by nematode proteins , is likely to be huge . Our analyses suggest that nematode protein space is huge , and that it is likely that our survey has merely scraped its surface . Indeed , some closely-related species , particularly within the Tylenchina , have an even higher proportion of private genes . This pattern is observed in all-against-all BLAST comparisons , in de novo protein family definition , and in derivation of novel domains . Most Nemfam3 families and NemDom3 domains are apparently private to Nematoda , and many have restricted phylogenetic distributions within the phylum . This finding contrasts with that emerging from whole genome analysis within Mammalia , where comparison of the predicted proteomes of eutherian ( human ) and metatherian ( opossum ) identified only 624 genes private to opossum and ∼500 to human ( about 2 . 5% of the predicted gene complement of each species ) , despite ∼180 million years of divergence [98] . However , comparisons of the predicted genes of the osteichthean Oryzias latipes ( medaka ) to those of other fish such as Tetraodon nigroviridis , with which medaka last shared a common ancestor ∼190 million years ago , identified 2936 genes unique to medaka , ∼15% of the total gene count [99] . Similarly , cross-comparison of the D . melanogaster ( fruit fly ) , Anopheles gambiae and Aedes aegypti ( mosquito ) proteomes identified 2924 ( 22% ) A . gambiae and 4181 ( 27% ) A . aegypti genes that were private to each species [100] . The mosquitoes are estimated to have diverged ∼140–200 million years ago . Thus the finding of high rates of novel gene evolution in the Nematoda may reflect a common pattern in Metazoa , with vertebrate taxa having a reduced rate . The identification of this level of protein novelty also challenges estimates of the total number of different protein families , and of the number of different possible domains , in all protein space . Even if our estimates of domain diversity are inflated through difficulties engendered by the use of partial proteome sequences , we have identified as many different domains in Nematoda as have been predicted in the rest of Metazoa to date . Additional meta-analyses of other major non-vertebrate groups , such as Arthropoda and Annelida , are sorely needed to explore the generality of these findings .	The high-throughput sequencing of messenger RNA from parasitic organisms has permitted large-scale sequence analyses typically reserved for complete genome studies . Such expressed sequence tags ( ESTs ) have previously been generated for 37 species from the phylum Nematoda , of which 35 were from parasitic species . These datasets were combined with the complete genomes of Caenorhabditis elegans and C . briggsae . The sequences were assembled into 65 , 000 protein families , and decorated with 40 , 000 distinct protein domains . These annotations were analysed in the context of the nematode phylogeny . We identified massive gene loss in the model nematode , C . elegans , as well as plant-like proteins in nematodes that cause crop damage . Furthermore , many protein families were found in small groups of closely related species and may represent innovations necessary to sustain their parasitic ecologies . All of these data are presented at NemBase ( www . nematodes . org ) and will aid researchers working on this important group of parasites .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results/Discussion" ]	[ "genetics", "and", "genomics/gene", "discovery", "evolutionary", "biology/evolutionary", "and", "comparative", "genetics", "infectious", "diseases/helminth", "infections", "evolutionary", "biology/bioinformatics", "computational", "biology/genomics", "genetics", "and", "genomics/bioinformatics" ]	2008	On the Extent and Origins of Genic Novelty in the Phylum Nematoda
Serological assays for human IgG4 to the Onchocerca volvulus antigen Ov16 have been used to confirm elimination of onchocerciasis in much of the Americas and parts of Africa . A standardized source of positive control antibody ( human anti-Ov16 IgG4 ) will ensure the quality of surveillance data using these tests . A recombinant human IgG4 antibody to Ov16 was identified by screening against a synthetic human Fab phage display library and converted into human IgG4 . This antibody was developed into different positive control formulations for enzyme-linked immunosorbent assay ( ELISA ) and rapid diagnostic test ( RDT ) platforms . Variation in ELISA results and utility as a positive control of the antibody were assessed from multiple laboratories . Temperature and humidity conditions were collected across seven surveillance activities from 2011–2014 to inform stability requirements for RDTs and positive controls . The feasibility of the dried positive control for RDT was evaluated during onchocerciasis surveillance activity in Togo , in 2014 . When the anti-Ov16 IgG4 antibody was used as a standard dilution in horseradish peroxidase ( HRP ) and alkaline phosphatase ( AP ) ELISAs , the detection limits were approximately 1ng/mL by HRP ELISA and 10ng/mL by AP ELISA . Positive control dilutions and spiked dried blood spots ( DBS ) produced similar ELISA results . Used as a simple plate normalization control , the positive control antibody may improve ELISA data comparison in the context of inter-laboratory variation . The aggregate temperature and humidity monitor data informed temperature parameters under which the dried positive control was tested and are applicable inputs for testing of diagnostics tools intended for sub-Saharan Africa . As a packaged positive control for Ov16 RDTs , stability of the antibody was demonstrated for over six months at relevant temperatures in the laboratory and for over 15 weeks under field conditions . The recombinant human anti-Ov16 IgG4 antibody-based positive control will benefit inter-laboratory validation of ELISA assays and serve as quality control ( QC ) reagents for Ov16 RDTs at different points of the supply chain from manufacturer to field use . Onchocerciasis , or “river blindness , ” is a disease caused by the filarial parasite O . volvulus ( Ov ) that affects an estimated 37 million ( 2005 ) in Africa and a few thousand people in the Americas and Yemen [1 , 2] . In recent years , the burden of disease has been reduced significantly through large programs of community-directed treatment with ivermectin ( CDTI ) , an antiparasitic drug donated by Merck . Data from both the Americas and Africa suggest that elimination may be achieved in the most part through mass drug administration ( MDA ) [3–9] . However , in some scenarios , such as where there is Loa loa co-endemicity , other interventions may be required . Active infection is detected by direct observation of the Ov microfilariae ( MF ) through skin snip combined with microscopy . This method is not very sensitive , especially when microfilarial ( MF ) skin densities are low , which are typical in low-transmission settings . Polymerase chain reaction-based assays of skin snips can significantly increase sensitivity [10 , 11] but are not suitable for either surveillance or point of care . Consequently , serological assays have been adopted by onchocerciasis elimination programs such as the Onchocerciasis Elimination Program in the Americas ( OEPA ) to inform whether elimination has been achieved . Specifically , IgG antibodies to the O . volvulus antigen Ov16 are used as a marker for exposure to infection , when applied to a sentinel population of children under ten years of age as a marker of continued transmission [12 , 13] . Currently this test is performed as an enzyme immunoassay ( EIA or ELISA plate format ) , both in the Americas and more recently in Africa [4 , 5 , 14 , 15] . It was previously demonstrated that this same assay could be transferred to the lateral flow platform to generate an RDT for anti-Ov16 IgG4 antibody [16–18] . More recently , PATH and Standard Diagnostics ( Yongin-si , Gyeonggi-do , South Korea ) announced the commercial availability of an Ov16 RDT . This test , called the SD BIOLINE Onchocerciasis IgG4 test , detects anti-Ov16 IgG4 in a finger-prick blood sample . A major challenge in the standardization and inter-assay comparison of serological tests such as ELISA is the lack of a standardized positive control [19 , 20] . Positive controls are typically made by pooling sera from exposed individuals and are distributed either as part of a commercial kit or from a reference laboratory . While this approach is common practice , it requires access to clinical specimens , something that will get increasingly harder as onchocerciasis prevalence continues to decline . Furthermore , each new lot of positive control from pooled sera will need to be equilibrated and validated . The ability to produce clones of human antibody of a desired subtype to specific antigens provides the opportunity to generate a standardized positive control that is inexhaustible [21] . By using the human combinatorial antibody library , HuCAL , such clones can be identified by combining phage display libraries with recombinant protein expression technology [22] . For recombinant antibody generation , the HuCAL PLATINUM library was used , consisting of a collection of 45 billion fully synthetic human Fab genes with diversified complementarity-determining regions ( CDR ) inserted into a phagemid vector [23] . Fifteen different antibodies were identified that are specific for Ov16; the two that showed the best binding properties in ELISA and nitrocellulose platforms were selected and converted into human IgG4 . One of these recombinant human anti-Ov16 IgG4 clones was used to generate standards for the Ov16 serological ELISA . Important to understanding the functionality of the positive control antibody in a serological ELISA , two Ov16 ELISA protocols were compared directly to establish a dynamic range and detection limit of the anti-Ov16 antibody for each ELISA method . Since the use of a common control is paramount to data consolidation , a cross-laboratory comparison of ELISA results generated from the positive control antibody was conducted . During optimization of an Ov16 RDT as an alternative immunoassay platform , the positive control antibody was additionally applied for stepwise monitoring of performance through all aspects of manufacturing and use . Because it is not a limited resource or subject to drift in affinity , as may be observed between clinical positive pools , it could be used more widely across QC checkpoints , including during manufacture , storage stability , and during RDT use by surveillance teams . Furthermore , a standardized positive control would support implementation of a QA program to help monitor commercial RDT quality along the delivery chain from manufacturer to end user . Recombinant antibodies were generated by Bio-Rad AbD Serotec ( Puchheim , Germany ) from the HuCAL PLATINUM collection of human antibody genes [23] by three rounds of selection on immobilized Ov16 antigen , as previously described [24] . Antibodies in the Fab mini-antibody format Fab-dHLX-FSx2 ( bivalent Fab containing a heavy chain C-terminal dHLX-dimerization domain followed by FLAG and Twin-Strep-tag ) were screened as crude extracts of E . coli expression cultures in ELISA [25–27] . More than 200 clones were identified which bind Ov16 and do not bind to an unrelated recombinant protein . From sequence analysis of the 25 clones with the highest signal in ELISA , 15 unique antibodies were identified , which were expressed and purified by streptactin affinity chromatography [28] . Specific binding to Ov16 of the purified mini-antibodies was confirmed by ELISA . Of the 15 clones , two were selected at PATH for conversion into human IgG4 by assessing the binding to Ov16 antigen spotted or striped directly on a nitrocellulose matrix as either a dot blot format or lateral flow format . Binding of the mini-antibodies was detected with the secondary label: peroxidase-conjugated goat anti-human Fab2 ( Jackson Immuno Research Labs , West Grove , PA ) . Absence of observable cross-reactivity to GST and WB123-GST was also confirmed on this platform , and two candidate Fab were converted by AbD Serotec to full-length IgG4 constructs using an expression vector system in mammalian cells designed to add the respective constant regions to the light/heavy chain variable components . Screening of the two different IgG4 against Ov16 was performed using Ov16 HRP ELISA ( described below ) and lateral flow test strips prepared with Ov16 antigen striped onto a nitrocellulose membrane as previously described [16] . The antibody clone AbD19432_hIgG4 was chosen based on both its dynamic range of absorbance values in ELISA and by its greater signal on lateral flow strips , relative to antibody concentration . The concentration of the Ov16 positive control antibody was determined by measuring the absorbance at 280 nm against a 1X PBS blank . The generic immunoglobulin mass extinction coefficient 1 . 37 ( mg/mL ) -1 cm-1 was used to calculate the concentration of the antibody . A 1mg/mL stock of anti-Ov16 recombinant IgG4 in 1X PBS , pH 7 . 4 , is stored at -70°C . When used as a sample in lateral flow tests , the positive control antibody was diluted into either fetal bovine serum , FBS ( Thermo Fisher Scientific , Waltham MA ) or Ov-negative human serum or plasma and applied in the same manner as a test sample . US-sourced negative sera and plasmas were procured from PlasmaLab International ( WA , USA ) and Bioreclamation IVT ( MD , USA ) from donors that had not traveled outside of the United States of America . A clinical Ov-positive specimen pool was made by combining archived plasmas positive for IgG4 antibodies to Ov16 antibody . These archived plasma samples had been collected after written informed consent was obtained from all subjects prior to collection of the samples , and all the subjects consented to having serum or plasma stored for later analysis . All sera or plasmas were stored at -80°C until use . The studies performed in Togo were approved by the PATH research ethics committee and the Togolese Bioethics Committee for Research in Health ( Comité Bioethique pour la Recherche en Santé CBRS ) . Two studies were performed in Togo , one in 2013 ( PATH Study File Number HS 716 ) and one in 2014 ( PATH Study File Number 563009–1 ) . All specimens used in this study were collected from study participants who provided informed written consent . Two studies were performed in Togo to determine Ov16 seroprevalence by ELISA from DBS and to evaluate early prototype versions of an Ov16 RDT . Both studies were performed during routine onchocerciasis surveillance during the early rainy season prior to MDA: June 4 to July 1 , 2013 , and May 11 to June 16 , 2014 . The studies were performed by staff members of the Togo National Program for Onchocerciasis ( PNLO ) and the Laboratory for Onchocerciasis Research . For each study , the staff received a two-day training course for performing study procedures including appropriate consent , data management , running the RDT , and utilizing the dried-down positive control . In addition , a research team member was present for the first several hundred study participants to monitor test use and to provide ongoing support . Briefly , in the 2013 study , 1 , 500 subjects were recruited in 15 villages over a period of 27 days , and in the 2014 study , 1 , 500 subjects were recruited in 20 villages over a period of 38 days . The African Program for Onchocerciasis Control ( APOC ) collected temperature and humidity data during their epidemiological surveillance activities from 2011 to 2012 . USB data loggers ( Monarch Instruments Inc . , Amherst , NH ) that record temperature and % relative humidity readings ( % RH ) were attached to microscopes used by APOC for skin microscopy . The data loggers were set to collect temperature and humidity at a sampling rate of every 15–20 minutes . Since the microscopes were stored at the APOC facilities in Burkina Faso , this dataset allowed collection of transit data as well as operational data . The same temperature and humidity monitors were used during the studies conducted in Togo in 2013 and 2014 . In each study , the monitors were similarly attached to the microscopes that were stored and used in parallel to the RDTs . Blood samples containing the anti-Ov16 positive control antibody were made by diluting the recombinant IgG4 from a stock solution into FBS ( Invitrogen , Grand Island , NY ) and then mixing thoroughly at a 1:1 dilution with packed , washed , red blood cells . 75μL per circle marking of the contrived whole blood sample was spotted on Whatman 903 Protein Saver Cards ( GE Healthcare , Pittsburgh , PA ) . These cards were then dried overnight in ambient laboratory conditions , then stored at -20°C with approximately ten cards per sealed foil pouch containing two-unit clay desiccant packets ( Desiccare , Reno , NV ) . Immulon 2HB ( Thermo Fisher Scientific ) plate wells were coated with 100μL of 2μg/mL Ov16 antigen diluted in 0 . 1M carbonate buffer ( Sigma , St . Louis , Mo ) overnight at 4°C . DBS were eluted at 2 punches ( 6 mm circles ) in 200μL of PBST + 5% BSA ( Sigma ) overnight . The following morning , plates were washed four times with PBST ( Sigma ) and then blocked with PBST + 5% BSA at room temperature . Blocking buffer was removed and DBS eluate samples were added to the plate without dilution at 50μl per well . To prepare dilutions for both the HRP and AP ELISAs , the anti-Ov16 positive control antibody was diluted into PBST + 5% FBS . Samples were incubated at room temperature for two hours , then washed four times with PBST . 50μL of a 1:1 , 000 dilution in PBST of a biotinylated anti-human IgG4 ( clone 6025 , Invitrogen ) was added to each well . The plates were incubated at room temperature for one hour , and washed four times with PBST . 50μL of a 1:2 , 000 dilution in PBST of streptavidin conjugated to alkaline phosphatase ( Invitrogen ) was added to each well . The plates were then incubated at room temperature for one hour , and washed four times with PBST . 50μL of pNPP solution ( Invitrogen ) was added to each well . Plates were incubated at room temperature for 30 minutes and read at 405nm using an ELISA reader . Replicate well ELISA units were averaged and the limit of detection , LOD , was approximated by determining the first concentration value to produce an average absorbance on or above the mean + 2 standard deviations of the background of the assay . Immulon 2HB ( Fisher Scientific ) plate wells were coated with 100μL of 5μg/mL Ov16 antigen diluted in PBS , pH 7 . 4 ( Sigma ) and left overnight at 4°C . DBS , when used , were eluted at one ( 6 mm circle ) punch in 200μL of PBST+2% milk ( Mix’n’Drink , Saco , Middleton , WI ) per 6 mm circle punch , overnight . The following morning , plates were blocked with PBST+5%FBS ( FBS—Invitrogen ) at 37°C . Plates were washed three times with PBST ( Sigma ) and DBS eluate samples were added neat to the plate at 50μl per well . For plasma or serum samples , a 1:50 dilution was made of each sample in PBST+5%FBS and added to the plate at 50μl per well . The anti-Ov16 IgG positive control antibody was diluted into PBST and 5%FBS . Samples were incubated at 37°C for one hour , then washed three times with PBST . A 1:5 , 000 dilution of an anti-human IgG4 ( 6025 clone Hybridoma Reagent Labs , Baltimore , MD ) was added at 50μl per well . Plates were incubated at 37°C for one hour , and washed four times with PBST . A 1:10 , 000 dilution of an HRP-conjugated goat anti-mouse antibody ( Jackson Immuno Research Labs ) was added at 50μl per well . Plates were incubated at 37°C for one hour , and washed four times with PBST . 100μL of TMB ( Sigma ) solution was added to each well . Plates were incubated at room temperature for 15 minutes and then the reaction was stopped by adding 50μl per well of 1N HCl ( Thermo Fisher ) . Plates were read at 450nm . Replicate well ELISA units were averaged and the limit of detection , LOD , was approximated by determining the first concentration value to produce an average absorbance on or above the mean + 2 standard deviations of the background of the assay . The positive control and ELISA reagents were provided to four independent laboratories: 1 ) Laboratory of Parasitic Diseases , National Institute of Allergy and Infectious Diseases , National Institutes of Health , Bethesda , Maryland , USA; 2 ) Global Health Infectious Disease Research Program , Department of Global Health , University of South Florida , Tampa , Florida , USA; 3 ) Department of Internal Medicine , Infectious Diseases Division , Washington University School of Medicine , St . Louis , Missouri , USA; and 4 ) Centers for Disease Control and Prevention , Division of Parasitic Diseases and Malaria , Atlanta , Georgia , USA . Specific dilutions of the positive control were run in duplicate wells per plate , over multiple plates , using the same HRP or AP-based ELISA protocols as described above , but with the respective equipment for each lab . Standard curve data submitted from each laboratory for the Ov16 HRP ELISA were compiled for analysis . The coefficient of variation was expressed as the magnitude of the ratio of the standard deviation/mean of all wells’ results at a given concentration across all laboratories . Two formulations of dried Ov16 positive control were prepared . The first was used in an initial laboratory-based pilot study to assess stability and used to accompany an early-stage RDT prototype tested in Togo in 2013 during surveillance activities . For this formulation , recombinant IgG4 was diluted from a stock solution into FBS at a concentration of 2μg/mL . 30μL was added to 1 . 5mL centrifuge tubes and dried by vacuum centrifugation for 80 minutes at ambient temperature . The tubes were then closed and packaged individually with a ½-gram clay desiccant packet using a heat sealer into pouches in a controlled humidity room with < 20% humidity . To test stability in the laboratory , these pouches were distributed into incubators with the following conditions: 25°C , 45°C , and a variable-temperature incubator with a daily fluctuation between 20°C and 40°C . At specific time points , the dried antibody preparations were rehydrated by adding 60μL of Ov16 RDT running buffer ( provided with the Ov16 RDT kit ) and 2 . 5μL was applied to a bare Ov16 lateral flow test strip to assess test-line intensity . The packaged , dried controls were included during RDT testing in Togo in 2013 at 1 control per 50 RDTs . The dried antibody preparation was rehydrated by adding approximately 60μL of Ov16 buffer 20 minutes prior to use . The second dried formulation was prepared to accompany a late-stage development prototype of a lateral flow-based RDT for Ov16 IgG4 ( Ov16 RDT ) , tested in Togo during MDA and surveillance activities in 2014 . Recombinant IgG4 was diluted from a stock solution into FBS at a concentration of 250ng/mL . 30μL per tube was added to 1 . 5mL centrifuge tubes and left open at room temperature overnight , leaving a film coating the walls of the tube . The tubes were then packaged in the same manner as the first formulation . One control pouch was incorporated with each kit of 25 RDTs . The team was instructed to run a positive control every time a box of 25 RDTs was opened to verify the kit was functional . The positive control was rehydrated by adding four drops , approximately 75μL , of Ov16 buffer , allowing the pellet to dissolve for 20 minutes , and then running the rehydrated control solution using the same protocol as a test sample . To run test blood or plasma samples , 10μL of sample was collected in the sample transfer pipette . The sample was added to the sample port by gently squeezing the pipette to release the sample into the port . Four drops ( approximately 75μL ) of Ov16 buffer was added to the buffer port . The test was read at 20 minutes . The Fab phage display library HuCAL PLATINUM was used for the generation of antibodies binding to recombinant O . volvulus antigen Ov16 . Fifteen antibodies in bivalent Fab mini-antibody format were selected and tested in spot and lateral flow assay as well as in ELISA . Two clones , AbD19432 and AbD19422 , were selected for conversion to full human IgG4 based on performance in the spot and lateral flow test ( signal-to-noise ratio in serial dilutions ) . While both clones produced soluble IgG4 with similar performance on the Ov16 ELISA and lateral flow strips , clone AbD19432_hIgG4 was selected as the final candidate for the study primarily based on a slightly more intense result than the other clone at lower concentrations when used with the RDT . Dose-dependent results were observed both on RDTs and in ELISA ( Figs 1 and 2 ) . A 10 μL sample of 25 ng/mL run on the RDT is visible by eye as a very weak positive test line . Dilutions of the Ov16 positive control were assayed using both the HRP Ov16 ELISA and the AP Ov16 ELISA . Fig 2A shows dose dependence of ELISA signal for both ELISA methods . The analytical limits of detection for the HRP and AP ELISA were approximately 1ng/mL and 10ng/mL antibody , respectively . Fig 2B shows comparison of the HRP and AP ELISA results for positive control-spiked sera and clinical positive pool sera prepared as DBS and eluted with the same samples processed as plasma/serum samples ( 1:50 dilution ) . The DBS specimen type requires the input of approximately 9 μL of plasma or serum per 6mm punch per ELISA evaluation as compared to only 2 μL for plasma serum when diluted 1:50 . However , similar OD were obtained for paired specimen types by ELISA for both HRP and AP ELISA methods . Differences between the ELISA methods were similar to results obtained with the dilution series of positive control; the HRP ELISA had a lower limit of detection than the AP ELISA as well as a broader dynamic range ( Fig 2B ) and the HRP DBS ELISA protocol required only 1 punch instead of the 2 punches used in the AP ELISA protocol . Additionally , true positive patient pools were found to have a higher signal relative to negative sera when run with the HRP ELISA , suggesting the capacity for higher signal-to-noise ratios . Fig 3 shows comparison of standard curves of positive control dilutions , run using the HRP ELISA method , from the participating laboratories . The raw or plate background-subtracted data showed a diverse range of absorbance values with respect to concentration of the positive control , highlighting inter-laboratory variability ( Fig 3A ) . As demonstration that the anti-Ov16 IgG4 antibody can be used as a positive control , the plate absorbance values and standard curves were then normalized to the average value of the 2 . 5ng/mL sample from their respective plates . The resulting normalized data show relatively cohesive absorbance values , particularly in the linear range of the assay ( Fig 3B ) . While the curves disperse significantly at extreme ends of the absorbance range , the normalization permits comparison between multiple plate datasets . For ELISA results produced by 1ng/mL positive control ( near detection limit ) , the magnitude of the coefficient of variation ( CV ) decreased from 0 . 67 , for all points from all laboratories prior to normalization to 0 . 24 , following normalization . For a given concentration , normalization did not always produce a reduction in the magnitude of the CV using the single-point normalization method . However , the average of all inter-laboratory CV values for given concentration points went from 2 . 29 before normalization down to 0 . 79 after normalization . Positive controls for RDTs and the RDTs themselves were exposed for prolonged periods of time to a broad range of temperature and humidity conditions during surveillance activities . In order to capture the range of these conditions , temperature and humidity measurements were collected from a series of APOC surveillance activities and two surveillance activities in Togo . Temperature and humidity monitors were attached to each of two microscopes used for skin snip microscopy , sent from Burkina Faso to five APOC epidemiological surveillance sites in 2011 and 2012 . These included the foci: Logon Occidental , Logon Oriental , and Moyen-Chari in Chad June to August 2012; Kilosa in Tanzania April 2012; Enugu in Nigeria September to November 2011; Adjumain and Moyo in Uganda July to August 2012; and Kasese in Uganda July to August 2012 . The studies included temperature and relative humidity data collected during transit and the period throughout which the surveillance was carried out . The summary statistics are shown in Table 1 . The largest temperature and relative humidity variation was observed during shipment of the microscopes , 8 . 2°C –37 . 9°C and 40 . 2% to 119 . 5% . The infrequent values >100% relative humidity were possibly due to artifacts from brief temperature fluctuations which could affect the relative humidity reported , or possibly due to moisture collection on the monitors . Such values would be interpreted as near 100% humidity . Most critically , the tests and positive controls needed to demonstrate stability over a temperature range of 17 . 6°C–37 . 8°C and a relative humidity of 40 . 2% to 119 . 5% as observed during performance of surveillance activities . Temperature and relative humidity data were also collected during two Togo MDA and surveillance activities in 2013 and 2014 in which the positive control was used to verify functionality of every new RDT kit used as soon as the kit was opened ( Table 1 and Fig 4 ) . This data represents the operational temperature and relative humidity conditions to which the packaged tests and positive controls were exposed during these activities . Generating a dried form of recombinant IgG4 creates an opportunity for a stable antibody available for use in the absence of colder temperature storage . The first formulation was tested for stability in the laboratory using high temperature and daily temperature cycling conditions and was found to produce a clear positive signal after over 1 . 5 years of high-temperature stress ( Fig 5 ) . Despite indication of stability in that all positive controls gave a positive signal by RDT throughout the 2013 study , the vacuum-drying method produced a friable pellet which could break and fragment throughout the tube during transport , as noted during the study . An alternative method of drying-down the positive control , compatible with transport and with the SD BIOLINE Ov16 RDTs , was used to produce a thin film at the bottom of the tube which was less likely to fragment . This type was used in the study conducted in 2014 . Both types of dried positive control were designed so that the user need only add 2 or 4 drops of Ov16 buffer ( depending on dropper bottle type used in study ) to the vial , rehydrate the positive control for 20 minutes , and then run the Ov16 RDT as per a normal sample ( Fig 6 ) . Observation during training of the PNLO surveillance team using the dried-down positive control in the field indicated that this process was simple and easy to perform . This second formulation was used to monitor the performance of RDTs throughout the duration of a surveillance and MDA activity in Togo in 2014 . In brief , 60 kits of RDTs were used over a period of 38 days ( May 11 , 2014 , to June 16 , 2014 ) during which time both the RDTs and the dried positive control , packaged as individual packets similarly to an RDT , were exposed to the environmental conditions shown in Fig 4 . All 60 positive controls gave a positive signal , showing a minimum operational stability at ambient and operational environmental conditions of 90 days ( starting from date of production of the dehydrated positive controls through final use during field study ) . The human IgG4 antibody response to O . volvulus antigen Ov16 has been identified as a specific serologic marker of exposure to Onchocerca volvulus[11] and has been accepted to monitor progress toward elimination of river blindness [29] . To address a need for a consistent , standardized , and widely accessible source of positive control , a recombinant human IgG4 with specific affinity to the Ov16 antigen was generated . Evaluation of the anti-Ov16 IgG4 antibody by ELISA demonstrated a significantly better limit of detection , and favorable signal-to-noise ratios when used with the HRP ELISA method as compared to the AP ELISA method . Differences between the outcomes might be explained by variation of the components in the two different ELISA protocols , such as blocking buffer or antigen-coating concentration and buffer . When such variables were tested as individual changes to the protocol , results did not explain the observed difference in detection limit . Since the AP ELISA uses the same clone of anti-human IgG4 as the HRP ELISA protocol , antibody affinity alone is also insufficient to explain the difference . While data presented here implies the HRP ELISA as having more favorable analytical performance , more studies with clinical specimens would be required to understand its utility for surveillance purposes , and in particular , the relative specificities of the assays . The AP ELISA has already been used in multiple countries as a tool to verify transmission interruption and elimination [4 , 6 , 7 , 14 , 15] . In an elimination context , high specificity of an assay , along with high quality control and reproducibility of the assay , is essential to minimize follow-up on false positives . The positive control described in this study can be used with either ELISA method as either plasma or DBS sample type . This comparison allowed sample type comparison , ELISA type comparison , and showed the feasibility of the positive control for use in multiple specimen formats . Use of such a monoclonal antibody to generate data regarding the relationship between OD of positive control and the assay’s sensitivity and specificity towards clinical samples should aid in ongoing data comparison to help answer research questions that are key to understanding transmission indicators such as antibody persistence and intensity of antigen-specific antibody responses from individual in regions with varying prevalences and stages of elimination of Onchocerca volvulus . In addition to standard curve dilutions or controls for ELISA , the positive control can be used to produce a relevant process control for assays sourcing specimens from DBS . Implemented in ELISAs run in Sokodé , Togo , a DBS control prepared in the lab with a serum concentration of positive control of 250 ng/mL was used in the HRP ELISA as a process and normalization control for each ELISA plate . From a cost perspective , approximately US$1 of positive control antibody can be used to make several milliliters of spiked positive control serum supplying over 200 ELISA assays with standardized positive control in this dried blood spot ( DBS ) form . Prepared more simply as a solution of 2 . 5 ng/mL to use as a plate control , the same amount can supply hundreds of thousands of ELISA plates . As demonstrated from the comparison of ELISAs performed across four laboratories , the anti-Ov16 IgG4 control may facilitate ELISA plate QC and data normalization for data analysis and compilation . A positive value in the linear absorbance range may be less influenced by assay limits and a simple control requiring little handling will be less prone to variation . These parameters should guide choice of form and concentration for normalization . A positive control should be run with every plate , making access to a universal positive all the more critical . Understanding the absorbance result of a given concentration of positive control that optimizes the definition of true positive and true negative from endemic regions requires continued data collection . With the availability of an Ov16 RDT ( SD BIOLINE Onchocerciasis IgG4 test; Alere , Standard Diagnostics , Inc . , South Korea ) , the ability to monitor the quality of an RDT outside the context of a laboratory throughout the lifecycle of the test in shipping , storage , and use is paramount to having reliable results . Point-of-care assays and tools used in surveillance applications must withstand demanding environmental conditions in order to be appropriate and impactful for even remote populations in need , as demonstrated from the data collected for this study across seven onchocerciasis surveillance activities between 2011 and 2014 . Having a portable , stable , positive control that can withstand the same conditions is a key component of this quality monitoring . The simple dry formula , designed for use with the RDT running buffer , demonstrated feasibility as a QA reagent , robust enough as packaged to withstand the stress of temperature and humidity variation without the requirement of cold chain in surveillance activities performed in Togo . This , combined with laboratory stability data that shows little decline in antibody affinity after 19 months at 45°C , suggests that a dried formulation of the recombinant IgG4 antibody is robust enough to be a practical option as an accessible positive control reagent . While in these studies the positive control was run using the Ov16 RDT at a high rate of 1 per 25 tests in order to monitor both the RDT and the positive control itself , ideally the positive control antibody will be used in a mature quality assurance program that will provide recommendations on sampling rates and specific concentrations needed . Use of a single positive control can provide a performance link between ELISA and rapid diagnostic tools , such as the SD BIOLINE Onchocerciasis IgG4 test , to understand limits of detections of clinically-relevant antibody responses . Currently , 10μL of 25 ng/mL positive control in plasma can be detected on the RDT as a weak positive . If such concentration were a test sample , it would by process be diluted 50-fold for use in ELISA . This would produce a final concentration of 0 . 5 ng/mL , which as compared to typical standard curve responses , would barely be detected above baseline . This can be seen in Fig 2B when concentrations of positive control are run as test samples . In this case , The RDT and the HRP ELISA procedure results in similar detection limits of a given concentration . However , caution should be used when making direct comparisons between LOD across the ELISA platform and RDT platform since the LOD of a monoclonal antibody can be assay-specific while important measures of performance such as sensitivity and specificity can only be determined using real clinical samples . Determining the relationship between performance with clinical samples and positive control results must be made for each assay type before performance conclusions can be drawn based on the LOD of the positive control . The approach described in this study should be generalizable to other similar serology-based markers for exposure to pathogens , such as IgG4 specific to the lymphatic filariasis-associated Brugia malayi and Wuchereria bancrofti markers BM14 and WB123 , respectively . Serology marker-based assays represent an opportunity to integrate surveillance activities for multiple pathogens [30 , 31] . Positive control reagents that can function cross-platform as broad-spectrum QA reagents will help ensure these assays can play the intended role in surveillance for transmission of these pathogens . While a monoclonal antibody control minimizes the risk of lot-to-lot heterogeneity and potential drift of clinical positive pools , it cannot be used as a substitute to clinical panels to measure assay performance and caution should be used if applying the positive control antibody to compare performance cross-platform given potential differences in epitope availability between assays . For example , a change in immunoassay platform may greatly affect detection limits of the control antibody but have little effect on clinical detection limits . When used as a QC reagent for the Ov16 antigen itself , the limitations and advantages of a monoclonal antibody should be considered . Limitations may include lack of detection of antigen mutation outside the epitope of the Ov16 protein . Advantages may be the positive control’s sensitivity to epitopic mutation in the antigen and less drift in the affinity of the control itself . Future work may include epitope mapping and comparison of positive control and positive pool samples with different lots of antigen to establish baseline lot-to-lot variability . For quality control , the parameters for positive control use should be assay-specific and well characterized prior to implementation . A comprehensive QC and QA program ensures that a diagnostic test will provide the desired output from its use—a consistent and reliable detection of analyte . QC is a product-based approach implemented at the manufacturing level that is reactive or corrective to identified defects . QA is a process-based approach that comprises procedures and systems to ensure all deliverables are consistently of good quality . Production of the anti-Ov16 antibody-positive control can support QC and QA programs for Ov16-based onchocerciasis diagnostics through incorporation into training aids , QA panels , and reagents to support optimal test use and standard operating procedures ( SOPs ) for ELISA standardization .	Serological markers such as antibody responses to pathogen-specific antigens are used to inform disease epidemiology in many elimination programs . A major challenge with program-scale serological testing , and with any diagnostic test validation , is access to consistent and unlimited control reagents with which to provide assay QC and facilitate data consolidation . In the context of disease elimination , clinical positive sera will be particularly difficult to source and use as routine , inter-laboratory reagents . This study reports on a recombinant antibody specific against a key serological marker for onchocerciasis: its selection , testing , and incorporation into protocols across relevant immunoassay platforms . We have demonstrated it is a viable reagent for integration into QC and QA protocols to support long-term serological testing for onchocerciasis to support disease elimination efforts . This approach should be generalizable to other diagnostic tools supporting programs to achieve the 2020 goals of the London Declaration on Neglected Tropical Diseases .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[]	2016	A Recombinant Positive Control for Serology Diagnostic Tests Supporting Elimination of Onchocerca volvulus
In order to respond reliably to specific features of their environment , sensory neurons need to integrate multiple incoming noisy signals . Crucially , they also need to compete for the interpretation of those signals with other neurons representing similar features . The form that this competition should take depends critically on the noise corrupting these signals . In this study we show that for the type of noise commonly observed in sensory systems , whose variance scales with the mean signal , sensory neurons should selectively divide their input signals by their predictions , suppressing ambiguous cues while amplifying others . Any change in the stimulus context alters which inputs are suppressed , leading to a deep dynamic reshaping of neural receptive fields going far beyond simple surround suppression . Paradoxically , these highly variable receptive fields go alongside and are in fact required for an invariant representation of external sensory features . In addition to offering a normative account of context-dependent changes in sensory responses , perceptual inference in the presence of signal-dependent noise accounts for ubiquitous features of sensory neurons such as divisive normalization , gain control and contrast dependent temporal dynamics . A fundamental goal of any perceptual system is to infer the state of the environment from received sensory signals . These signals are generally noisy and unreliable , so that the same signal can correspond to many different states of the world . For example , the sound of a bell may mean my mobile phone is ringing or there is someone at the door . Contextual cues , such as a vibration in my pocket , can resolve such ambiguities , in this case suggesting that my phone is ringing , and not the doorbell ( Fig 1a ) . Such competition between different explanations of sensory signals is called ‘explaining away’ and is a basic requirement for a perceptual system to discriminate between similar features . Neurally , it implies that groups of neurons which encode different ( but overlapping ) stimuli ( such as the ‘telephone’ and ‘door’ ) should actively compete , via recurrent suppression [1] . The way this competition is implemented has a crucial impact on how neural responses are modulated by stimulus context . In many ‘classical’ models of early visual processing , visual neurons are assumed to integrate inputs from within their receptive field , before undergoing divisive or subtractive inhibition from the surround ( Fig 1b ) [2] . In this case , non-preferred stimuli produce a general suppression of neural responses , but no changes to neural RF and/or tuning curve shapes ( only a general suppression ) . Returning to our previous example , this would predict a general suppression of ‘door selective’ neurons when the phone was vibrating . In other words , the phone vibration would equally suppress the response of these neurons to ringing and knocking sounds ( Fig 1d ) . However , explaining away as described above requires a markedly different form of competition , with inhibition from non-preferred stimuli targeting specific neural inputs , before they are combined ( Fig 1c ) . In this case , suppression would cause neurons to become unresponsive to certain inputs , but not others , resulting in a qualitative modulation of their receptive field ( RF ) shapes and/or tuning curves . For example , if the phone is vibrating , suggesting someone is calling , then the ringing sound ( now explained by another cause ) should not activate the door selective neurons . However , this should not affect how these neurons respond to other cues , such as a knocking sound ( since the phone might be ringing whilst someone is also knocking on the door; Fig 1e ) . Here , we show that the specific form that this input-specific suppression should take depends on how incoming signals are corrupted by noise . In turn , this will deeply affect the predicted dynamics and integrative properties of sensory neurons . For example , if the noise was Gaussian with a fixed variance independent of the signal strength , a sensory neuron should subtract from the other neuron’s inputs its prediction of these inputs . Because this operation is linear , the overall effect is equivalent to a global subtractive suppression by the surround ( i . e . the sum of all subtractive inhibitions from other neurons ) , bringing us back to ‘classical’ models of sensory processing ( Fig 1b ) . However , sensory receptor responses and neural firing rates generally exhibit signal-dependent noise , whose variance scales proportionally with their amplitude [3–5] . We show that in this case , competition should take the specific form of divisive suppression , where each individual neural input is divided by its prediction from other neurons . Since this occurs before these inputs are combined ( Fig 1c ) , it is in no way equivalent to a global surround suppression ( either divisive or subtractive ) . Instead , the receptive fields of individual neurons are dynamically and selectively reshaped by the surround . Divisive inhibition of each input by the surround accounts for experimental evidence showing that neural receptive fields are constantly reshaped by the spatiotemporal context of presented stimuli [6–9] . Importantly , these contextual changes in neural RFs and tuning curves do not imply variations in the stimulus features encoded by each neuron . Rather , variable receptive fields are required in order to maintain an invariant neural code that can be read-out consistently by downstream neurons . Note that this framework is normative and does not depend on how it is implemented at the neuronal level . However , in order to provide more specific predictions , we show that optimal estimation can be performed within a plausible neural circuit in which excitatory neurons undergo divisive inhibition from local interneurons . Neurons in that circuit exhibit general properties of sensory neural responses including response saturation , gain modulation by background stimuli and contrast-dependent temporal dynamics . For a subclass of ‘simple’ stimuli , the responses of excitatory neurons in this network can be phenomenologically described using the canonical divisive normalization model of Heeger et al . [2 , 10–14] . This accounts for why divisive normalization appears so ubiquitously across different sensory areas and organisms . It further suggests avenues for how this canonical model may need to be extended to account for the richness and selectivity of surround suppression and contextual modulation in general . To interact effectively with our environment , we need to know ‘what’s there’ . Thus , perception can be viewed as an inference problem , in which sensory systems infer which combination of stimuli is the most likely , given the noisy signals they receive . Perceptual inference requires basic assumptions about how sensory signals are generated by external stimuli , which can be expressed mathematically using a ‘generative model’ . Here , we consider a simple generative model , in which multiple positive stimulus features , x = ( x1 , x2 , … , xn ) , combine linearly to activate a population of neural inputs , s = ( s1 , s2 , … , sn ) . The mean expected response of the jth input to a stimulus , x , is: 〈 s j x 〉 = ∑ k w j k x k + w 0 , ( 1 ) where wjk describes how strongly the jth input is activated by the kth stimulus feature and w0 describes its mean activity when all stimulus features are zero . The presumed goal of sensory processing is to estimate stimulus features , x ^ , from the received input , s . Consider a population of neurons that encodes stimulus features , x ^ = ( x ^ 1 , x ^ 2 , … , x ^ n ) , via their firing rates , r ( x ^ ) . While the stimulus features cannot usually be estimated directly by pooling the neural inputs , we can set up dynamics of the network so that the encoded features , x ^ , converge to the most likely solution . This will be satisfied if the encoded stimulus features vary in time according to , ∂ x ^ i ∂ t = η ∂ log p s \| x ^ ∂ x ^ i , ( 2 ) where p ( s \| x ^ ) describes the probability that a stimulus , x ^ , would give rise to neural inputs s , and η is a free parameter determining how quickly the estimates vary in time . These dynamics ensure that the encoded stimulus features , x ^ , converge on a local maximum of p ( s \| x ^ ) [15 , 16] . The neural dynamics required to implement the Eq 2 will depend critically on the input statistics , described by p ( s \| x ^ ) . In particular , different assumptions about the reliability of the neural inputs will lead to qualitatively very different predictions . A common experimental observation is that sensory neurons exhibit signal-dependent noise , in which the trial-by-trial variance in single neuron firing rates scales proportionally with their mean firing rate [4 , 5] . When neural inputs are corrupted by independent Poisson noise ( a paradigmatic signal-dependent distribution ) , Eq 2 becomes: ∂ x ^ i ∂ t = η ∑ j w j i s j ∑ k w j k x ^ k + w 0 - 1 . ( 3 ) Thus , the estimate of each stimulus feature varies in time according to a linear sum of ‘fractional prediction errors’ , s j 〈 s j ( x ^ ) 〉 - 1 , equal to the ratio between the received input and the mean predicted input ( given the current estimate ) , minus one ( see section 1 in S1 Text for derivation ) . If the received input is equal to the predicted input , then the fractional prediction error is zero , and the estimate does not change . However , if the received input is larger or smaller than the predicted input , then the estimate is updated to reduce the error . Importantly , dividing the received input by the predicted input is necessary to perform optimal estimation given many different types of signal-dependent noise—as long as the variance in each input is proportional to its mean ( section 1 in S1 Text ) . Poisson input is but one example of such signal-dependent noise statistics . Furthermore , while noise correlations will introduce further terms to Eq 3 , these additional terms also require dividing the received input by the predicted input ( section 2 in S1 Text ) . We note that ‘noise’ in our model refers to trial-by-trial variability of neural inputs , s , given fixed external stimulus features , x . In contrast , the dynamics of the model network , described by Eq 3 , are deterministic ( see Discussion ) . In Eq 3 , each input ( sj ) is divided by a different factor ( 〈 s j ( x ^ ) 〉 ) , before being combined with other inputs . Thus , any neural network implementation of Eq 3 will need to normalize different inputs separately , before they are combined ( Fig 1c ) . For comparison , let us consider an artificial example with the input signal corrupted by constant Gaussian noise , whose magnitude is independent of the signal strength . In such a scenario , the estimate of each feature would evolve as a function of the absolute ( rather than the ‘fractional’ ) prediction errors , s j - s ^ j . Eq 3 could then be separated into two linear terms: a feedforward input and a subtractive lateral inhibition term ( see Methods ) . Moreover , steady neural responses could be described as applying a ‘center-surround’ feedforward receptive field to the stimulus . Thus , if sensory noise was constant Gaussian and not signal dependent , competition between encoded features would result in a global ‘inhibitory surround’ , separable from a static feed-forward ‘center’ ( Fig 1b ) . In the rest of the paper we refer to the network assuming constant Gaussian noise as the ‘subtractive model’ , as opposed to the model assuming signal-dependent noise , which we call the ‘divisive’ model . To relate the estimation algorithm described in the previous section to neural data , we make the basic assumption that each neuron encodes a single stimulus feature , with firing rate proportional to the estimated feature ( r i ∝ x ^ i; see later for neural implementation ) . The divisive model described by Eq 3 requires selective inhibition of specific neural inputs , before they are combined . Thus , if certain inputs are predicted by the stimulus context , they will be inhibited , and the neuron will become differentially less responsive to them . As a result , a neuron’s stimulus selectivity will be reshaped by the context . In contrast , in the subtractive model ( see Methods ) , inhibition acts globally to alter the magnitude of neural responses , but not their stimulus selectivity . To illustrate this , we first consider a simple generative model , where each stimulus feature is assumed to activate two neighbouring sensory inputs . This results in the network shown in Fig 2a , where each neuron receives two equal strength inputs from neighbouring locations in the previous layer . With both subtractive and divisive models , each neuron responds equally strongly to both its inputs ( ‘no context’ condition; Fig 2b ) , while being suppressed by contextual ‘surround’ stimuli , that do not elicit a response when presented alone . However , in the divisive model inhibition selectively targets certain inputs , so that a surround stimulus only suppresses a neuron’s response to nearby inputs ( that are ‘predicted’ by the surround ) . As a result , neurons respond less strongly to stimuli presented in parts of their receptive field that are near the surround ( ‘adjoint context’; Fig 2b ) , than to stimuli presented far from the surround ( ‘disjoint context’ ) . In contrast , the subtractive model predicts the same degree of surround suppression , regardless of the location of stimuli within the cell’s receptive field . A further and related consequence of input-targeted inhibition , is that neural tuning curves are reshaped by contextual stimulation . To illustrate this effect , we considered a generative model in which stimulus features activate nearby sensory inputs , arranged along a single dimension ( e . g . representing the orientation of a presented visual stimulus ) . In the resulting network , neurons responded with bell shaped tuning curves to presented stimuli ( Fig 2c , top left panel; see Methods ) . An overlapping ‘mask’ stimulus , that did not activate a given neuron when presented alone , selectively inhibits inputs to the neuron that overlap with the mask . As a result , the neuron’s tuning curve was reduced in magnitude and shifted away from the mask ( Fig 2c , top left panel ) . This effect is qualitatively similar to contextual shifts in neural tuning curves observed experimentally in cat primary visual cortex ( Fig 2c , top right panel ) [6] . As a control , we considered a ‘linear-nonlinear’ ( LN ) model , with responses obtained by a filter followed by a threshold non-linearity: ri = f ( ∑j vjisj ) . Linear weights were fitted to match , as closely as possible , the responses of the divisive model across all three stimulus conditions ( see Methods ) . As shown in Fig 2c ( lower panel ) an LN model was unable to produce the shifts in neural tuning curves observed with the divisive model . In addition to shifting neural tuning curves , input-targeted divisive inhibition also results in dynamic reshaping of neural receptive fields ( RFs ) . To illustrate this , we extended our previous generative model , to consider the case where presented stimulus features activate sensory inputs , arranged along two spatial dimensions . Neural RFs , estimated using reverse correlation with random sparse stimuli ( see Methods ) , exhibited a ‘centre-surround’ structure , with a central excitatory region surrounded by an inhibitory region ( Fig 2d , above ) . However , simultaneously presenting an overlapping grating stimulus dramatically reshaped the estimated RFs , which were elongated orthogonal to the grating ( Fig 2d , below ) . No such contextual shifts in RFs was observed with an LN model ( Fig 2e ) . Previously , Meister et al . showed that presenting an orientated grating stimulus over a period of several seconds leads to a reshaping of retinal ganglion cell RFs , qualitatively similar to what we observed in our model [17] . ( However , note that to properly model the effects of temporal adaption would require extending our work to consider optimal estimation of temporally dynamic stimuli ) [18] . In early visual areas , where neural RFs are localized within a single region of space , our model predicts simple shifts in neural RFs , as shown in Fig 2d . However , in other sensory modalities ( e . g . olfaction/audition ) , where neural RFs have a more complex structure , contextual reshaping of neural RFs could be more complex [19 , 20] . To illustrate this , we considered a generative model in which individual sensory features ( e . g . presented odors ) produce a distributed and multi-modal activation of sensory receptors , as shown in Fig 3a ( upper panels; see Methods ) . We measured the RFs of neurons in response to a random sparse stimulus plus a contextual mask that activated a small subset of nearby receptors . The contextual mask led to complex changes in neural RFs that could not be characterised as a simple repulsive shift away from the context ( Fig 3a ) . Moreover , the observed reshaping of neural RFs was highly non-local: contextual activation of nearby receptors affected distant regions of a cell’s RF . To explain intuitively this contextual reshaping of neural RFs , we considered a toy generative model consisting of three stimulus features , which produce patterns of sensory activation resembling the letters ‘V’ ‘A’ and ‘I’ , respectively ( Fig 3b ) . We measured the RF of the neuron encoding the letter ‘I’ in response to random sparse stimuli , and in the presence of an overlapping contextual stimulus ( Fig 3c ) . Because of the simplicity of this network , we can understand how the contextual stimuli reshape the neuron’s RF . For example , the first contextual stimulus strongly activated the neuron encoding the letter ‘A’ ( Fig 3c , top left ) leading to targeted inhibition of neural inputs that overlap with the letter ‘A’ . As a result , the recorded neuron became insensitive to these inputs , and they did not form part of its recorded RF ( Fig 3c , top right ) . An analogous effect occurred with a contextual stimulus designed to activate the neuron encoding the letter ‘V’ ( Fig 3c , lower panels ) . Note that this contextual reshaping of neural RFs occurred because inhibition was targeted on a subset of neural inputs ( Fig 1c ) ; it would not occur in a network with global inhibition , that acted directly on neural responses ( Fig 1b ) . The observation that neurons have highly variable RFs could lead one to conclude that the neural code also varies with stimulus context . However , note that each neuron always encodes a fixed stimulus feature , as defined by the generative weights wij . As a result , the neural responses can always be read-out in the same way by the downstream neurons , by interpreting the activity of each neuron as indicating the presence of its preferred feature . For this same reason , our model can be extended to hierarchical frameworks where each layer predicts the responses of the layer below ( section 3 in S1 Text ) . The resulting neural code is thus ‘fixed’ ( as defined by the features wij ) , and the neural representation is ‘invariant’ ( in the sense that sensory neurons always represent the same objects , regardless of context ) . However , in order to maintain this fixed code , neurons in the network need to have variable RFs , that adapt depending on the stimulus context ( Fig 4a ) . To illustrate this idea , we return to our earlier simulation with bell-shaped tuning curves , shown in Fig 2c . This time , however , we plotted neural tuning curves in the presence of three different ‘contexts’ ( Fig 4b; each context was a ‘mask’ , constructed from a random combination of ‘background’ stimulus features; these masks were constantly added to the inputs used to measure the tuning curve and estimate the read-out weights ) . As before , the tuning curves were shifted by the context ( Fig 4c , left panel; tuning curves are rescaled and shifted to have the same magnitude and zero mean ) . Next , we trained ‘readout filters’ to linearly reconstruct the inputs from the neural responses ( see Methods ) . As could be expected , these were similar to the actual read-out weights wij . In particular , and in sharp contrast with the tuning curves , which were shifted by context , readout filters were almost completely invariant to changes in context ( Fig 4c , right panel ) . For comparison , we repeated the same procedure with an LN model ( Fig 4d ) . As seen previously , in this model neural tuning curves are not shifted by context ( only their gain is changed , which does not appear on the re-scaled tuning curves ) . However , readout filters were altered by context , meaning that in each context , downstream neurons would have to integrate responses from the network differently ( depending to the context ) in order to reconstruct the stimulus . As shown in Fig 4e and 4f , the same qualitative effects were observed for the tuning curves and readout filters across the entire neural population , in addition to the example cell shown in Fig 4c and 4d . Finally , we quantified the reconstruction error across all three conditions ( normalized rms error ) , obtained with the ‘correct’ readout filter ( i . e . trained on responses obtained with the same mask; Fig 4g , blue bars ) , compared with a ‘mismatched’ decoder ( trained in different conditions; Fig 4g , red bars ) . In the input-targeted inhibition model , similar performance was achieved in either case , as the readout filter did not change significantly across conditions . In contrast , in the LN model performance was drastically reduced when using a mismatched decoder , learned in a different context . Our results suggest that , rather than trying to describe neural responses using a static ‘encoder model’ ( e . g . tuning curves or RFs ) one may be able to fit a simpler context-invariant ‘decoder model’ , describing how to reconstruct the stimulus from neural responses . Experimental support for this is provided by Marre et al . who were able recover a highly accurate reconstruction of a moving bar stimulus from a simple linear readout of retinal ganglion cell responses [21] . In contrast , neural responses in their experiment were poorly described by an LN model . The advantages of input-targeted divisive inhibition are also seen when discriminating between similar features , presented together . To demonstrate this , we returned to the earlier model with multimodal distributed features , shown in Fig 3a . We considered neural responses to combinations of three similar stimulus features , encoded by different neurons in the network ( Fig 5a ) : feature 1 presented alone , and alongside feature 2 or 3 ( Fig 5b ) . Fig 5c plots the response of five feature-selective neurons . Despite the fact that the three features activated highly overlapping sets of receptors , neural responses were highly specific , with only neurons that encode the presented odors responding on a given trial . In contrast , an LN model could not achieve this degree of specificity ( Fig 5d ) . The estimation algorithm described by Eq 3 could be implemented in more than one way within a neural network . The most direct implementation would be for each neuron to encode a single stimulus feature , with firing rate proportional to the estimated feature ( r i ∝ x ^ i ) . In this case each neuron needs to selectively inhibit the input synapses of neurons encoding different features , as shown in Fig 6a . The response of each neuron evolves in time according to: ∂ r i ∂ t ∝ ∑ j w ˜ j i r s j - const ( 4 ) where w ˜ j i ( r ) is an ‘effective input weight’ , obtained by dividing the feed-forward weight , wji , by the responses of other neurons in the network , according to: w ˜ j i ( r ) = w j i ∑ k w j k r k + w 0 . As a result , feedback connections alter the effective weighting of each input , thereby altering neural stimulus selectivity . There are two reasons why neural dynamics described by Eq 4 may not be biologically plausible , at least in the cortex . First , the network violates Dale’s law: neurons are required to send both excitatory projection to higher sensory layers and inhibitory feedback to other neurons in the same area . Second , it requires a highly selective form of feedback , targeted on individual synapses ( Fig 6a ) . To overcome these issues , we propose an alternative network that consists of two distinct neural populations: excitatory neurons that encode the ratio between the received and predicted input , s j 〈 s j ( x ^ ) 〉 , and inhibitory neurons that encode stimulus features , x ^ i ( Fig 6b ) . Each excitatory neuron receives feed-forward input from one receptor type , and lateral inhibition from interneurons . Its response evolves in time as: a d r j e x c d t = s j - w 0 + ∑ k w j k r k i n h r j e x c . ( 5 ) where a is a constant that ( along with the magnitude of inhibition ) determines the of timescale of excitatory responses . Inhibition acts multiplicatively on the leak term in the firing rate dynamics ( see Discussion for biophysical mechanism ) . These dynamics ensure that in the steady state the response of each excitatory neuron is equal to the ratio of its excitatory and inhibitory input: r j e x c = s j w 0 + ∑ k w j k r k i n h . ( Note that , unlike classical subtractive predictive coding , in the case where sensory inputs are perfectly predicted by the network , excitatory responses are equal to unity , not zero ) . Inhibitory neurons receive lateral input from nearby excitatory neurons . Their responses evolve in time according to: b d r i i n h d t = ∑ j w j i r j e x c - 1 . ( 6 ) where b determines the rate that inhibitory neurons integrate their input . In the steady state ( i . e . when r j = s j w 0 + ∑ k w j k r k i n h ) , this equation is equivalent to the optimal estimation algorithm shown in Eq 3 . Thus , in the steady state , the response of each inhibitory neuron will be proportional to an encoded feature , r i i n h = x ^ i o p t . Both excitatory and inhibitory neural responses are constrained to be positive . Stimulus features can be recovered by neurons in higher-level areas by temporally integrating the responses of the excitatory neurons ( Fig 6c and section 3 in S1 Text ) . Thus , the network implements a form of ‘predictive coding’ , in which the fractional prediction errors , rather than the estimated stimulus features themselves , are communicated to higher level sensory areas [1] . In the following sections we will explore the implications of input-targeted divisive inhibition in the context of this ‘predictive coding’ network . We investigated how divisive inhibition modulates the steady state responses of excitatory neurons , which encode the fractional prediction error . We first considered a very simple model composed of only two sensory receptors , both activated by a single stimulus feature . The corresponding neural network consists of two excitatory neurons that connect with equal strength to one inhibitory neuron ( Fig 6d , left ) . In this network , the sustained response of each excitatory neuron is simply equal to its feed-forward input , divided by the total rectified input to the network ( section 4 in S1 Text ) : r 1 e x c ∝ s 1 max ( s 1 + s 2 , w 0 ) . ( 7 ) This equation bears strong similarity to the canonical divisive normalization equation , developed by Heeger et al . [10 , 22] . Thus , our normative framework parsimoniously predicts the nonlinearities seen in previous phenomenological models of divisive normalization . When the feed-forward input to neuron 1 is very weak ( i . e . s1 ≪ s2 ) , the denominator of Eq 7 is constant , and the neuron’s responses increases linearly with input strength . When the feed-forward input to neuron 1 is very strong ( i . e . s1 ≫ s2 ) , on the other hand , the numerator and denominator of Eq 7 approach equality , and the neuron’s response saturates . Plotted on a logarithmic scale , this gives rise to a sigmoidal input-response curve ( Fig 6d ) [2] . Lateral inhibition from a ‘mask’ stimulus that does not provide direct input to neuron 1 ( i . e . it activates s2 only ) , suppresses the neuron’s response [45] . When s1 ≫ w0 , the effect of the mask is to add an additional constant to the denominator of Eq 7 , shifting the neuron’s input-response curve to the right on a logarithmic scale ( Fig 6d ) . Consequently , a stronger feed-forward input is required to elicit the same neural response . A mask stimulus that provides weak input to neuron 1 and strong input to neuron 2 ( i . e . it weakly activates s1 , and strongly activates s2 , as shown on Fig 6e ) can both suppress or facilitate the response of neuron 1 , depending on the strength of the neuron’s feed-forward input [2] . When the feed-forward input to neuron 1 is very weak , the denominator of Eq 8 is constant ( due to rectification ) , and the neuron linearly sums its feed-forward inputs . As a result , its response is facilitated by the mask ( Fig 6e ) . When the feed-forward input to neuron 1 is strong , the mask increases the size of the denominator , suppressing the neuron’s response ( Fig 6e ) . The results described above also apply to larger networks consisting of many excitatory and inhibitory neurons . Indeed , for “simple” inputs that do not activate multiple overlapping feature detectors , the sustained response of each excitatory neuron is approximately equal to its feed-forward input , divided by the summed input to nearby neurons ( Section 4 in S1 Text ) : r i e x c ∝ s i max ( ∑ j w i j s j , w 0 ) . ( 8 ) Thus , the classical normalization model [10] , that was originally designed to provide a phenomenological description of non-linearities in neural responses , emerges as a special case of our proposed dynamics . We next investigated the temporal dynamics of excitatory and inhibitory neural responses to a constant stimulus in the simple , two neuron network described in the previous section ( Fig 6d , left panel ) . Following stimulus onset , the response of the activated excitatory neuron , encoding the fractional error signal , exhibited a transient peak in activity followed by a decay ( Fig 7a ) . At the same time , the response of the inhibitory neuron , which encoded the sensory estimate , increased continuously towards the steady state ( Fig 7b ) . This qualitative behaviour is a general property of predictive coding , and thus also occurred for the subtractive model , where excitatory neurons encoded the absolute ( rather than the fractional ) error ( Fig 7c and 7d ) . What distinguishes the subtractive and divisive models was the input-depencence of the neural dynamics . For the divisive model , the timescale of excitatory neural responses decreased with the sensory input , resulting in a shorter time to peak response with higher amplitude inputs ( Fig 7a ) . This is because the leak term in the excitatory neural dynamics ( which implements divisive inhibition ) is proportional to its inhibitory input Eq ( 5 ) . Thus , the greater a neuron’s inhibitory input , the quicker its response varies in time . In contrast , the temporal dynamics of the subtractive model were input-invariant . Recent experiments using voltage sensitive dye to measure V1 responses reported contrast-dependent temporal dynamics , consistent with our model [23] . Similarly , Albrecht et al . [24] observed that the time to peak firing rate response decreases with visual contrast . However , Albrecht et al also reported that temporally shifting firing rate responses to compensate for contrast-dependent variations in onset latency resulted in temporal response profiles that were approximately contrast invariant . This discrepancy between voltage data and firing rate data could be accounted for by including a firing threshold into our model . When put into the context of a larger , topographically organized sensory layer , the temporal dynamics of the divisive model could parsimoniously account for the presence of ‘traveling waves’ observed in the visual cortex , where a presented stimulus generates a wave of activity that spreads gradually outwards from a single cortical location ( Fig 8a ) [25] . According to our model , traveling waves will occur when the input generated by a stimulus varies in strength with cortical location [26] ( Fig 8b ) . Neurons that receive strongest feed-forward input will respond quickest , followed by nearby neurons that receive weaker input . The resultant effect is a damped traveling wave that spreads outwards from neurons most strongly activated by the stimulus ( Fig 8c ) . In contrast , with subtractive inhibition , the timecourse of neural responses does not depend on their distance from the input ( Fig 8d ) . It has long been thought that divisive inhibition performs a kind of gain control , that keeps neural firing rates within their available dynamic range [10 , 27 , 28] . Here we provide an alternative interpretation , that divisive inhibition occurs as a consequence of optimal cue combination given sensory noise . When the variance of each input depends on its mean , some signals become more reliable than others . Divisive inhibition insures that each signal is weighted appropriately , before these signals are combined by downstream neurons . In that sense , our work places itself in the more general framework of optimal cue combination [29] where each cue should be weighted according to its reliability before being combined . Human subjects are indeed able to perform such optimal cue combination [30 , 31] , and are also able to implement explaining away , e . g . to resolve ambiguities by assigning inputs to their most likely sources [32] . Our model proposes that optimal cue combination and explaining away are already implemented at a microscopic level by sensory networks , through selective divisive gain modulation of sensory neural responses . In contrast to the gain-control hypothesis , our framework precisely specifies the form of divisive inhibition required for optimal estimation , which should occur before individual inputs are combined ( Fig 1c ) . For simple stimuli , which activate only one feature detector at a time , the predicted neural responses are consistent with the classical divisive normalization model ( Figs 6 and 7 ) . However , for more complex stimuli , which activate multiple overlapping feature detectors , input-targeted divisive inhibition results in dynamic changes in neural tuning properties and receptive field shapes ( Figs 2 and 3 ) , not captured by the classical divisive normalization model . A prediction of our model is that sensory normalisation will vary with changes in neural variability . Thus , future experimental tests of our work could investigate whether divisive normalisation is altered as expected by stimulus-dependent modulations in neural Fano-factor ( see section 1 . 2 in S1 Text ) and noise correlations ( see section 2 in S1 Text ) [33 , 34] . Previously , Schwartz and Simoncelli showed that divisive normalisation can serve to remove statistical redundancies between neural responses , leading to a more efficient code [35] . In a later extension to this work , they showed that divisive normalisation can be interpreted as implementing ‘explaining away’ of global stimulus features ( e . g . global image contrast ) so as to permit optimal inference of local stimulus features ( e . g local reflectance ) [36] . While in both our model and that of Schwartz et al . , divisive normalisation implements explaining away , their underlying assumptions are very different . In Schwartz et al . ’s model , normalisation is predicted because of the assumed high-level structure of sensory signals , as being produced by multiplying local and global stimulus features . In contrast , in our model , divisive normalisation is predicted due to the biophysics of sensory signal transduction , which leads to sensory signals being corrupted by signal-dependent ( and not Gaussian ) noise . Schwartz & Simoncelli’s model also belongs to a broader class of normalisation models in which divisive occurs after sensory inputs have been combined [2] . In contrast , our model predicts that divisive normalisation should act directly on the inputs , before combination . As we showed ( Fig 2 ) , such pre-combination divisive inhibition leads to flexible RFs , which are dynamically shifted by the stimulus context . In contrast , output-targetted divisive normalisation will only lead to such shifts in neural RFs when sensory inputs undergo an additional ( e . g . quadratic ) non-linearity before normalisation . Previously Beck et al . proposed a new role for divisive normalisation in performing a probabilistic compuation known as ‘marginilisation’ [37 , 38] . This computation is required for many different tasks , in which one wants to infer a subset of ‘relevant’ stimulus features , while disregarding ( i . e . marginilising ) other irrelevant features . At some level , this explanation is related to Schwartz et al . ’s work , where normalisation was assumed to factor out ( i . e . marginilise ) global fluctuations in the sensory input , so as to allow inferences about local features . However , Beck et al . ’s model differs from both Schwartz et al . and our work , in that marginilisation is predicted as a result of a particular type of probabilistic neural population code . In previous work , we proposed a model in which input-targeted divisive inhibition implements competition between different stimulus features . However , this model relied on a number of assumptions about sensory stimuli ( e . g . that they were produced by binary stimulus features that had Markov temporal dynamics ) , as well as assumptions about the spiking neural code [18] . Here we show that input-targeted divisive inhibition emerges very generally , and irrespective of additional assumptions about the neural code and signal dynamics , so long as the sensory noise scales with the magnitude of the signal . Recent experimental work suggested that in the ferret auditory cortex , neural responses adapt to the stimulus statistics in such a way as to allow behaviourally relevant signals to be extracted from background noise [39 , 40] . Interestingly , Mesgarini et al . showed that their results could be explained by top-down divisive feed-back . While the details of our model differ from that of Mesgarini et al . ( e . g . they assumed that divisive inhibition acts after inputs are combined ) it gives a suggestion as to why top-down divisive feed-back could result in the noise-invariant neural responses observed in their data . Predictive coding implies that , rather than directly encoding sensory stimuli , feed-forward neurons encode a prediction error that can be used to update the internal representation in higher-level areas [1] . Here we show that , given signal-dependent sensory noise , this error signal should take a fractional form , implying divisive inhibition . Previously , Spratling et al . showed that a predictive coding network that minimizes fractional prediction errors can account for a number of classical and extra-classical response properties of neurons in visual area V1 [41] . We provide a normative interpretation of Spratling’s model , as implementing optimal estimation of presented stimulus features given signal-dependent noise . We find that , for a large family of distributions in which the variance in each input is proportional to its mean ( including , but not limited to Poisson noise ) , the prediction errors take a fractional form , implying divisive predictive coding ( see section 1 in S1 Text ) . With the exception of Spratling’s work , previous predictive coding models have usually assumed that sensory neurons encode the difference between their received and predicted input [1 , 42] . This type of code will be optimal only if the the variance in each sensory input is constant , irrespective of its mean . Subtractive predictive coding results in qualitatively different neural response properties , compared to divisive predictive coding . It predicts that: ( i ) the time course of neural responses is independent of stimulus strength; ( ii ) neural responses vary linearly with their feed-forward input , and thus , do not saturate; ( iii ) neural RFs are largely invariant to changes in stimulus context ( see section 4 in S1 Text ) . In summary , subtractive predictive coding cannot account for many of the non-linear response properties observed in sensory neurons , and that are explained by divisive predictive coding . Here , we show that the optimal form of neural gain control depends on how neural inputs are corrupted by noise . Specifically , signal-dependent noise requires input-targeted divisive inhibition , in contrast to gaussian noise , which requires global subtractive inhibition . ‘Noise’ here refers to the trial-by-trial variability of neural inputs , given a fixed stimulus . Generally , multiple noise sources combine to produce neural variability , including external noise sources ( e . g . random fluctuations in light intensity ) , and internal noise sources ( e . g . spike failure ) . The model is agnostic to these details , as long as the trial-by-trial variability of inputs to the network scales monotonically with their amplitude . In contrast , the model network itself has deterministic dynamics: for a given input , the neural responses are always be the same . However , while this choice was made for simplicity , related work on networks of spiking neurons shows how optimal estimation can be performed in a network of neurons that exhibit Poisson-like spiking statistics [18 , 43] . In these models , internal noise fluctuations that alter the spike times of single neurons are compensated by recurrent connections in the network , such that the read-out from the population response is relatively stable . The effects presented here come about as a result of optimal estimation with signal-dependent noise , and are thus largely independent of the specific neural mechanism that implements divisive inhibition . For example , contextual reshaping of neural RFs ( Figs 2 and 3 ) occurs because ‘explaining away’ takes place at the level of the inputs , before they have been combined , while gain modulation of neural responses ( Fig 6 ) is a property of the fractional prediction error . Nonetheless , in order to make concrete predictions about sensory neural responses we proposed a simple network architecture , in which excitatory neurons encode a fractional prediction error , and receive lateral inhibition from local interneurons that encode individual stimulus features ( Fig 6b ) . However , note that there is more than one way to implement the optimal estimation described by Eq 3 . For example , divisive inhibition could be mediated via top-down feed-back from higher-level areas [1] , or via lateral inhibition of individual synaptic inputs [18] ( Fig 6a ) . However , as they share the same normative structure to our model , these different network architectures result in very similar predictions for the neural responses . In our proposed network , excitatory neurons at the first level of processing each receive input from one type of receptor , and divisive inhibition from lateral interneurons . This closely matches the observed anatomy of both fly and mouse olfactory system , where mitral cells ( or 2nd-order PNs in fly ) receive feed-forward input from one type of olfactory receptor , and lateral inhibitory feed-back that depends on the responses of many receptors [42 , 44] . Furthermore , recent experiments have shown that in the fly , inhibition from lateral neurons is well described by the exact same divisive formula as obtained with our model [45] . Recently , researchers have reported how various interneuron types play different roles in sharpening and/or globally suppressing visual neural responses [46–49] . While generally , our simplified model is not designed to address this level of detail , it is worth noting that when implemented in a hierarchy ( Fig 6c ) , interneurons at different levels of processing will have qualitatively different effects on the tuning curves of excitatory neurons . Specifically , interneurons in the previous layer to a recorded neuron , that target its inputs , will act to sharpen and reshape the neuron’s selectivity , whereas interneurons in the same layer , that provide direct lateral inhibition , will lead to a global suppression ( but no sharpening ) of its responses . In our model , divisive inhibition is implemented via lateral feedback from inhibitory interneurons , which multiplicatively increases the ‘leak’ term in the dynamics of the excitatory neural responses Eq ( 5 ) . A potential candidate for this gain modulation is shunting inhibition [2] ( although see [50–52] ) . More generally however , current experiments suggest that there is not one unique neural mechanism that implements divisive inhibition [22] . Rather a host of different mechanisms , such as synaptic depression [53] , ongoing network dynamics [54] , and neuromodulatory feedback [55] may be responsible for divisive inhibition in different sensory areas and species . This is consistent with our framework , which suggests that it is the computation performed by divisive inhibition , rather than its neural implementation , that is conserved across sensory systems in order to optimally infer the state of the environment . The proposed neural network predicts several qualitative differences between the responses of excitatory neurons , which encode fractional prediction errors , and inhibitory neurons , which encode stimulus features . These differences are: ( i ) long-range ( i . e . between cortical regions ) signals are normalized , while short-range ( i . e . within region ) signals are not; ( ii ) inhibitory neurons respond to more complex non-local features than excitatory neurons in the same area ( they are thus expected to exhibit wider , apparently less selective tuning curves , as indeed observed experimentally [56] ) ; ( iii ) inhibitory responses are less transient than excitatory neural responses . Recent experiments , using optogenetic techniques , have shown that parvalbumin ( PV ) -expressing inhibitory cells can have a divisive effect on excitatory responses to sensory stimuli . Interestingly , PV cells appear to fulfil many qualitative criteria required by inhibitory cells in our model , such as broad stimulus tuning , temporally sustained responses , and minimal contrast normalisation ( relative to layer 2/3 excitatory neurons , to which they provide input ) [48] . Future research will be required to quantify more precisely how the activity of PV cells compares to the predictions of our model . Our model can easily be extended to consider sensory processing in a hierarchy , with neurons at each layer of the network reconstructing stimulus features of increasing complexity based on the inputs they receive from the previous layer ( see Fig 6c and section 3 in S1 Text ) . In this case , optimal estimation also requires using high-level knowledge to constrain and shape the low-level sensory representation . This can be easily incorporated into our framework , in the form of top-down feedback . As well as carrying information about the stimulus features encoded by higher-level areas , this top-down feed-back could also carry information about the organism’s prior experience and task-demands . Future work could investigate whether such top-down feedback is able to account for the experience-dependent and attention-dependent shifts in neural tuning curves that are observed experimentally [57 , 58] . In summary , our model suggests a highly dynamic system , in which neural RFs and tuning curves are continuously reshaped by the spatiotemporal context of presented stimuli , as well as the organism’s prior experience and task-demands . However , the neural code is context invariant: neurons always represent the same external feature , and thus their response can be read the same way by downstream neurons , regardless of the stimulus context . In addition to the model described in the main text , we also considered an artificial example , where the input signal is corrupted by constant Gaussian noise ( whose magnitude is independent of the signal strength ) . In this case , encoded features vary in time according to: ∂ x ^ i ∂ t = η ∑ j w j i s j - ∑ k w j k x ^ k + w 0 ( 9 ) Thus , the estimate of each feature evolves in time according to a sum of ‘absolute prediction errors’ , s j - 〈 s j ( x ^ ) 〉 , equal to the difference between received and predicted inputs . Note that because of the linearity of this equation , the left hand-side can be rewritten as the sum of a feed-forward input term ∑j wjisj and a lateral subtractive inhibition term - ∑ j k w j i w j k x ^ k - w 0 . In the particular case of constant gaussian noise , lateral inhibition is thus separable and can be seen as occurring “after combination” of these input signals . Similarly , in the steady state , the estimated features can be obtained by a weighted linear sum of feed-forward inputs: x ^ i = ∑ j v j i s j , with feed-forward weights vi directly related to the encoded features wi ( i . e . vi= ( WTW ) −1wi ) . In that interpretation , competition between encoded features adds a subtractive component ( an inhibitory surround ) to a static feed-forward filter ( Fig 1b ) . For the initial simulations shown in Figs 2–5 , we sought to investigate the general implications of divisive versus subtractive inhibition Eqs ( 3 ) and ( 9 ) , irrespective of the specific neural implementation . Although we assumed that neurons encode individual stimulus features , with firing rate proportional to the encoded feature ( r i ∝ x ^ i ) , the qualitative results would also be the same for a distributed code , in which each neurons encode a linear combination of stimulus features , according to , r i ∝ ∑ q k i x ^ i . Note that for the simulations used to generate Figs 2–5 the dimensions are essentially arbitrary , and thus all parameters are quoted in unit-less dimensions . Encoded features were initialized at zero , and updated using Eq 3 for the divisive algorithm , and Eq 9 for the subtractive algorithm . The update rate , η , was set to ensure smooth dynamics , while the number of iterations , N , was chosen to allow the estimates to converge on steady state values . The background rate , w0 , was set to 0 . 01 . In our framework , the generative model describing how external stimulus features activate sensory receptors determines the network connectivity . Furthermore , in the case where each neuron encodes a separate stimulus feature , there is a one-to-one correspondence between the structure of the generative model ( parameterized by w ) and the feed-forward connectivity in the network . Specifically , the parameter wji , that determines how strongly the ith feature activates the jth receptor , also determines the connection strength between the ith neuron and the jth receptor . We compared the input-targetted divisive inhibition model to a linear-nonlinear ( LN ) model , with responses obtained by linearly filtering the sensory inputs then applying a static non-linearity: ri = f ( ∑j vjisj + v0 ) . ( Note that this is a simple generalisation of the subtractive model where responses were strictly linear ) . For our simulations we used a threshold non-linearity , while linear weights ( vji ) and offset ( v0 ) , were learned so as to best fit the responses obtained with the input-targetted divisive model . Using a different non-linearity ( e . g . exponential ) had no qualitative effect on the predicted contextual tuning curve changes . In addition , we also considered a ‘global divisive-inhibition’ model ( Section 5 in S1 Text ) . For the simulation shown in Fig 2b there were 30 sensory receptors and 30 neurons . We used a generative model in which each feature activates two neighbouring receptors ( i . e . wii = w ( i+1 ) i = 40 ) . Thus , each neuron received equal strength feed-forward inputs from two neighbouring receptors ( Fig 2a ) . We computed the steady-state response of the kth neuron with both the subtractive or divisive algorithms , in three different stimulus conditions . In the ‘no-context’ condition , only one of the inputs to the recorded neuron was active , with firing rate drawn from a Poisson distribution with mean 50 ( i . e . 〈sk+1〉 = 50 ) . For the ‘adjoint context’ condition , a neighbouring input that did not drive the recorded neuron was also active ( with amplitude 〈sk+2〉 = 20 ) . Finally , for the ‘disjoint context’ condition , an input on the opposite side of the recorded neuron’s receptive field was active ( i . e . 〈sk−1〉 = 20 ) . In each condition , we averaged the neuron’s steady state response over 200 trials . For the simulation shown in Fig 2c there were 30 sensory receptors and 30 neurons . We assumed a generative model in which a stimulus moving in a given direction ( indexed by ‘i’ ) activates multiple neighbouring receptors , described mathematically via the circular basis functions: w j i = w m a x e 4 [ cos ( 2 π 30 ( j - i ) ) - 1 ] ( with wmax = 40 ) . As before , this implies that each neuron receives feed-forward inputs from multiple neighbouring inputs . We first looked at the steady state response of a single neuron to a varying stimulus direction , i . The activation of the jth sensory input was drawn from a Poisson distribution , with mean 〈sj ( i ) 〉 = wji + w0 . We next looked at the response of the same neuron in the presence of a ‘mask’ , which activated a single receptor , shifted 3 receptors to the left or right of the neuron’s preferred input . The activation of this receptor was held constant at 200 . The input to all other receptors was the same as in the previous control condition . The mask was chosen specifically so that it did not elicit any response in the recorded neuron when presented alone . For the simulation shown in Fig 2d and 2e there were 400 neurons , and 900 sensory inputs ( arranged on a 30×30 grid in visual space ) . Each neuron encoded a circular ‘blob-like’ stimulus feature . Specifically , columns of the matrix W specified the feature encoded by each neuron , with elements given by: w j i = w m a x e - 1 2 σ w 2 [ ( x j - x 0 i ) 2 + ( y j - y 0 i ) 2 ] . x0i and y0i specify the preferred region of visual space for the ith neuron , distributed uniformly along the axis spanned by x and y ( 0 → 1 ) . wmax , and σw determine the amplitude and width of the encoded features , and were set to 40 and 0 . 1 respectively . We first performed a simulation with ‘random sparse’ stimuli . Sensory inputs , sj , were either equal to 0 ( with probability 0 . 95 ) or 100 ( with probability 0 . 05 ) . Next , a vertical grating stimulus ( in which each bar spanned 8 pixels ) , of magnitude 20 , was added to the random sparse stimulus ( Fig 2d and 2e , bottom left ) . The phase ( but not the orientation ) of the grating varied randomly on each trial . Thus , on the nth trial , the sensory input was equal to , s n = s g r a t i n g n + s n o i s e n . In each case , neural receptive fields ( RFs ) were quantified using reverse correlation: w ^ j = Q s s - 1 q r s j , where ( Qss ) ij = 〈sisj〉 and ( qrs ) ij=〈sirj〉 , and 〈⋅〉 denotes an average over 104 stimulus presentations . In Figs 3a and 5 we considered an ‘olfactory network’ , with neurons were assumed to have a distributed selectivity , spanning multiple receptor inputs . Mathematically , the network was similar to the network described above . However , for the olfactory simulations , the feature encoded by each neuron consisted of a sum of four ‘blobs’ , distributed randomly across the input space ( see examples shown in upper panels of Fig 3a ) . Neural receptive fields were estimated as before , in response to a random sparse stimulus plus a contextual stimulus . For the plots shown in Fig 3a , the contextual stimulus consisted of a single ‘blob’ ( of magnitude 100 , and width σcntxt = 0 . 1 ) , that activated a set of nearby receptors ( see black and white panels in Fig 3a ) . Finally , we illustrated the principles underlying reshaping of neural receptive field using a simple network of only three neurons , each of which encoded a different letter of the alphabet ( ‘A’ , ‘I’ , and ‘V’ ) . Encoded features ( comprised of 600 sensory inputs , arranged in a 20×30 grid ) , are shown in Fig 3b . As before , neural RFs were estimated using random sparse stimuli , in addition to a contextual mask ( shown in the left panels of Fig 3c ) . We next investigated how divisive inhibition enables the network to maintain an invariant representation of encoded stimulus features . The network used for these simulations was the same as the model with bell-shaped tuning curves , shown in Fig 2c . We measured tuning curves in the same way as before , measuring the mean firing rate of each neuron versus the stimulus orientation . However , in this case we measured tuning curves in the presence of three constant ‘masks’ , constructed from different combinations of encoded features ( Fig 4b ) , added to the varying stimulus . In each stimulus condition , we estimated the linear filters required to reconstruct the stimulus from the neural responses , using linear regression . Thus , readout filters were given by U = 〈 s ¯ r ¯ T 〉 〈 r ¯ r ¯ T 〉 - 1 , where s¯=s−〈 s 〉 and r¯=r−〈 r 〉 . Fig 4d was constructed in the same way using the LN model ( Fig 4e ) . The parameters of the LN model were fitted to minimize the mean squared difference between the responses predicted by the LN model and the input-targeted inhibition model , across all three stimulus conditions . In Fig 5 we demonstrate how a model with input-targeted inhibition is able to discriminate between similar overlapping stimulus features . To illustrate this , we returned to the ‘olfactory network’ used to generate Fig 3a . We compared this input-targeted divisive inhibition model to the output-targeted divisive inhibition model , described previously . Parameters of this model were fitted to minimize the mean squared difference between the responses of the global inhibition model and the input-targeted inhibition models . Stimuli used to fit the model parameters consisted of random linear combinations of the features encoded by the network , corrupted by Poisson noise . We proposed a neurally plausible implementation of the the estimation algorithm described in Eq 3 ( Fig 6b ) . This network consists of two populations of neurons: excitatory neurons with dynamics described by Eq 5 , and inhibitory neurons with dynamics described by Eq 6 . Figs 7 and 8 were generated using discretized version of these equations . For these simulations , the background input was set to w0 = 1 . Parameters , a and b , determining the timescale of excitation and inhibition were set to 0 . 08 and 40 respectively ( see section 6 in S1 Text for the effect of varying the excitatory and inhibitory timescales ) . Input spikes were always counted over a time-window of T = 1s , so that the number of spikes fired by each input is equal to its firing rate . The network connectivity was entirely constrained by the generative model describing how presented stimulus features activate the inputs to the network . That is , the parameter ‘wji’ , that describes how strongly the ith stimulus feature activates the jth receptor , also determined the strength of the lateral connection between the jth excitatory neuron and the ith inhibitory neuron . For the plots shown in Fig 6d and 6e we considered a minimal model with 1 encoded feature and 2 sensory inputs . Within our framework , this corresponds to a network with 1 inhibitory and 2 excitatory neurons ( Fig 6d , left panel ) . The inhibitory neuron received equal strength inputs from both excitatory neurons ( w11 = w21 = 40 ) . Steady-state excitatory responses could be obtained directly from Eq 7 . In Fig 6d , the input to each excitatory neuron was drawn from a Poisson distribution with mean: 〈s1〉 = Itest and 〈s2〉 = Imask respectively . in Fig 6e , the ‘test’ and ‘mask’ stimulus activated both sensory inputs , so that: 〈s1〉 = Itest + 0 . 1Imask and 〈s2〉 = Imask + 0 . 1Itest . For the plots shown in Fig 7 we again considered the minimal network with 1 excitatory and 2 inhibitory neurons ( connection strengths were same as for Fig 6 ) . On each trial , the input to the recorded excitatory neuron was drawn from a Poisson distribution , with mean varying between 0 & 200Hz . The other neuron received zero input . For the plots shown in Fig 8 , we considered a ‘topographic’ network of 30 excitatory and 30 inhibitory neurons . Each inhibitory neuron connected with equal strength to 2 neighbouring excitatory neurons ( wii = wi ( i+1 ) = 40 ) . The input to the jth excitatory neuron was drawn from a Poisson distribution with mean , 〈 s j 〉 = 150 e - \| j - k \| 2 , where k denotes the neuron that receives maximal input .	Perception involves using incoming sensory signals to infer which objects or features are present in the surroundings . To do this , sensory systems must perform two basic operations: ( i ) combination of noisy sensory cues , and ( ii ) competition between different percepts . Here we show that the optimal form of competition depends on how sensory signals are corrupted by noise . Moreover , for the type of noise commonly observed in sensory systems , whose variance scales with the signal amplitude , competition should occur between different sensory cues before they are combined . Implemented neurally , this results in a highly flexible representation , in which neural receptive fields change dynamically depending on the stimulus context . Further we show that competition should take the form of divisive inhibition from the surround , accounting for why divisive normalisation , gain control and contrast dependent temporal dynamics appear so ubiquitous in sensory areas .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "neural", "networks", "social", "sciences", "neuroscience", "signal", "inhibition", "computational", "neuroscience", "neuronal", "tuning", "coding", "mechanisms", "computer", "and", "information", "sciences", "animal", "cells", "sensory", "receptors", "signal", "transduction", "sensory", "neurons", "cellular", "neuroscience", "psychology", "sensory", "cues", "cell", "biology", "neurons", "biology", "and", "life", "sciences", "cellular", "types", "sensory", "perception", "computational", "biology", "cell", "signaling" ]	2017	Sensory noise predicts divisive reshaping of receptive fields
Recent single molecule experiments , using either atomic force microscopy ( AFM ) or Förster resonance energy transfer ( FRET ) have shown that multidomain proteins containing tandem repeats may form stable misfolded structures . Topology-based simulation models have been used successfully to generate models for these structures with domain-swapped features , fully consistent with the available data . However , it is also known that some multidomain protein folds exhibit no evidence for misfolding , even when adjacent domains have identical sequences . Here we pose the question: what factors influence the propensity of a given fold to undergo domain-swapped misfolding ? Using a coarse-grained simulation model , we can reproduce the known propensities of multidomain proteins to form domain-swapped misfolds , where data is available . Contrary to what might be naively expected based on the previously described misfolding mechanism , we find that the extent of misfolding is not determined by the relative folding rates or barrier heights for forming the domains present in the initial intermediates leading to folded or misfolded structures . Instead , it appears that the propensity is more closely related to the relative stability of the domains present in folded and misfolded intermediates . We show that these findings can be rationalized if the folded and misfolded domains are part of the same folding funnel , with commitment to one structure or the other occurring only at a relatively late stage of folding . Nonetheless , the results are still fully consistent with the kinetic models previously proposed to explain misfolding , with a specific interpretation of the observed rate coefficients . Finally , we investigate the relation between interdomain linker length and misfolding , and propose a simple alchemical model to predict the propensity for domain-swapped misfolding of multidomain proteins . Protein misfolding and aggregation are well-known for their association with amyloidosis and other diseases [1 , 2] . Proteins with two or more domains are abundant in higher organisms , accounting for up to 70% of all eukaryotic proteins , and domain-repeat proteins in particular occupy a fraction up to 20% of the proteomes in multicellular organisms [3 , 4] , therefore their folding is of considerable relevance [5] . Since there is often some sequence similarity between domains with the same structure , it is easily possible to imagine that multidomain proteins containing repeats of domains with the same fold might be susceptible to misfolding . Indeed , misfolding of multidomain proteins has been observed in many protein families [6] . Single molecule techniques have been particularly powerful for studying folding/misfolding of such proteins , in particular Förster resonance energy transfer ( FRET ) and atomic force microscopy ( AFM ) . For instance , recent studies using single-molecule FRET , in conjunction with coarse-grained simulations , have revealed the presence of domain-swapped misfolded states in tandem repeats of the immunoglobulin-like domain I27 from the muscle protein Titin [7] ( an example is shown in Fig 1e ) . Domain-swapping [2] involves the exchange of secondary structure elements between two protein domains with the same structure . Remarkably , these misfolded states are stable for days , much longer than the unfolding time of a single Titin domain . The domain-swapped misfolds identified in the Titin I27 domains are also consistent with earlier observations of misfolding in the same protein by AFM , although not given a structural interpretation at the time [8] . In addition , AFM experiments have revealed what appears to be a similar type of misfolding in polyproteins consisting of eight tandem repeats of the same fibronectin type III domain from tenascin ( TNfn3 ) [9] , as well as in native constructs of tenascin [8] , and between the N-terminal domains of human γD-crystallin when linked in a synthetic oligomer [10] . In addition to domain-swapped misfolding , an alternative type of misfolded state is conceivable for polyproteins in which the sequences of adjacent domains are similar , namely the formation of amyloid-like species with parallel β-sheets . Theoretical work in fact made the prediction that such species would be formed in tandem repeats of titin domains [11] . Recently , time-resolved single-molecule FRET experiments on tandem domains of I27 have revealed a surprising number of intermediates formed at short times , which include an unexpected species that appears to be consistent with the previously suggested amyloid-like state [12] . However , since only the domain-swapped species persisted till long times , and therefore are the most likely to be problematic in cells , we focus on their formation in this work . A simplified illustration of the mechanism for folding and misfolding , based on both coarse-grained simulations as well as single-molecule and ensemble kinetics [7 , 12] , is shown in Fig 1 , using the Titin I27 domain as an example . Starting from the completely unfolded state in Fig 1a , correct folding would proceed via an intermediate in which either one of the domains is folded ( Fig 1b ) , and finally to the fully folded state , Fig 1c . The domain-swapped misfolded state , an example of which is shown in Fig 1e , consists of two native-like folds which are in fact assembled by swapping of sequence elements from the N- and C-terminal portions of the protein . The final structure in Fig 1e comprises what we shall refer to as a “central domain” formed by the central regions of the sequence ( on the left in Fig 1e ) and a “terminal domain” formed from the N- and C-termini ( on the right ) . The intermediate structure in Fig 1d , suggested by coarse-grained simulations [7] , and supported by experiment [12] , has only the central domain folded . This central domain can itself be viewed as a circular permutant [13] of the original native Titin I27 structure , as discussed further below . While domain-swapped misfolding of tandem repeats has been identified in a number of proteins to date , there are several other proteins for which it does not occur to a detectable level . For instance , extensive sampling of repeated unfolding and folding of a polyprotein of Protein G ( GB1 ) by AFM revealed no indication of misfolded states , in contrast to Titin [14] . Similarly , early AFM studies on polyUbiquitin also did not suggest misfolded intermediates in constant force unfolding [15–20] , and lock-in AFM studies of refolding [21] were fully consistent with a two-state folding model , without misfolding . More recent AFM [22] studies have suggested the formation of partially folded or misfolded species , which have been attributed to partial domain swapping in simulations [23] , but these are qualitatively different from the fully domain-swapped species considered here . Therefore , it is interesting to ask the general questions: when included in tandem repeats , what types of protein structures are most likely to form domain-swapped misfolded states , and by what mechanism ? In order to investigate the misfolding propensity of different types of domains , we have chosen seven domains , based on ( i ) the superfamilies with the largest abundance of repeats in the human genome [24] , ( ii ) proteins for which some experimental evidence for misfolding ( or lack thereof ) is available and ( iii ) proteins for which data on folding kinetics and stability is available for their circular permutants ( only some of the proteins meet criterion ( iii ) ) . The circular permutant data are relevant because the misfolding intermediates suggested by simulations and experiment [7 , 12] can be viewed as circular permutants of the original structure ( Fig 1d ) . Each of the chosen proteins is illustrated in Fig 2 and described briefly in Materials and Methods . We study the folding and misfolding of the seven protein domains , using the same structure-based model as that successfully employed to treat Titin I27 [7 , 12] . Molecular simulations are carried out to characterize the possible structural topologies of the misfolded intermediates and the mechanism of their formation . Our model is consistent with available experimental information for the systems studied , in terms of which proteins misfold and what misfolded structures they tend to form . We then investigated what factors influence the propensity of multidomain proteins to misfold . The simplest rationalization of the propensity of a multidomain protein for domain-swapped misfolding would seem to be offered by parameterizing a kinetic model based on the scheme shown in Fig 1 , particularly for the steps Fig 1a–1b versus 1a–1d . We hypothesized that the propensity to misfold might be characterized in terms of the folding kinetics of the isolated circular permutants representing the domain-swapped intermediates in Fig 1d . However , contrary to this expectation , we found that the stability of such isolated domains , rather than their folding rate , is the main determinant of misfolding propensity . Although superficially this appears to differ from previously suggested kinetic models [12] , it is completely consistent , with a specific interpretation of the rates . Building on this understanding , we developed a very simplified model which can be used to predict which domains are likely to be susceptible to domain-swapped misfolding . Finally , we have investigated the effect of the composition and length of the linker between the tandem repeats on the misfolding propensity . Tandem Src homology 3 ( SH3 ) domains ( Fig 2a ) are widely found in signal transduction proteins and they share functions such as mediating protein-protein interactions and regulating ligand binding [25] . Kinetic and thermodynamic properties of native and all the possible circular permutations of SH3 single domain have been well characterized [26] . Two different circular permutant constructs of the sequence are known to fold to a circularly permuted native conformation ( PDB accession codes are 1TUC and 1TUD ) that is similar to the wild-tpe ( WT ) protein [26] . With a similar function to the SH3 domains , Src homology 2 ( SH2 ) domains ( Fig 2b ) are also involved in the mediation of intra- and intermolecular interactions that are important in signal transduction [27] . The SH2 domains are well-known from crystallographic analysis to form metastable domain-swapped dimers [28 , 29] . Fibronectin type III ( fn3 ) domains ( Fig 2c ) are highly abundant in multidomain proteins , and often involved in cell adhesion . We have chosen to study the third fn3 domain of human tenascin ( TNfn3 ) , which has been used as a model system to study the mechanical properties of this family . Single-molecule AFM experiments revealed that a small fraction ( ∼ 4% ) of domains in native tenascin ( i . e . the full tenascin protein containing both TNfn3 and other fn3 domains ) [8] , with a similar signature to that observed for I27 . Subsequently , misfolding events have been identified in a polyprotein consisting of repeats of TNfn3 only [9] . Interestingly , a structure has been determined for a domain-swapped dimer of TNfn3 involving a small change of the loop between the second and third strand [30] . PDZ domains ( Fig 2d ) are one of the most common modular protein-interaction domains [31] , recognizing specific-sequence motifs that occur at the C-terminus of target proteins or internal motifs that mimic the C-terminus structurally [32] . Naturally occurring circularly permuted PDZ domains have been well studied [33–35] , and domain-swapped dimers of PDZ domains have been characterized by NMR spectroscopy [36 , 37] . Titin ( Fig 2e ) is a giant protein spanning the entire muscle sarcomere [38] . The majority of titin’s I-band region functions as a molecular spring which maintains the structural arrangement and extensibility of muscle filaments [39] . The misfolding and aggregation properties of selected tandem Ig-like domains from the I-band of human Titin ( I27 , I28 and I32 ) have been extensively studied by FRET experiments [7 , 24] . In the earlier work on tandem repeats of I27 domains , around 2% misfolding events were reported in repeated stretch-release cycles in AFM experiments [8] . A slightly larger fraction ( ∼ 6% ) of misfolded species was identified in single-molecule FRET experiments and rationalized in terms of domain swapped intermediates , captured by coarse-grained simulations [7 , 11] . In contrast , with the above misfolding-prone systems , there are certain polyprotein chains have been shown be resistant to misfolding , according to pulling experiments . For instance little evidence for misfolding was identified in a polyprotein of GB1 [14] ( Fig 2g ) , with more than 99 . 8% of the chains ( GB1 ) 8 folding correctly in repetitive stretching–relaxation cycles [14] . Lastly , we consider polyUbiquitin ( Fig 2f ) , for which there is conflicting experimental evidence on misfolding . Initial force microscopy studies showed only the formation of native folds [15] , with no misfolding . Later work suggested the formation of collapsed intermediates [22] , however the signature change in molecular extension of these was different from that expected for fully domain-swapped misfolds . A separate study using a lock-in AFM [21] found Ubiquitin to conform closely to expectations for a two-state folder , without evidence of misfolding . For this protein , there is a strong imperative to avoid misfolding , since Ubiquitin is initially expressed as a tandem polyUbiquitin chain in which adjacent domains have 100% sequence identity , yet this molecule is critical for maintaining cellular homeostasis [40] . A coarse grained structure-based ( Go-like ) model similar to the earlier work is employed for the study here [7 , 41] . Each residue is represented by one bead , native interactions are attractive and the relative contact energies are set according to the Miyazawa–Jernigan matrix . The model is based on that described by Karanicolas and Brooks [41] , but with native-like interactions allowed to occur between domains as well as within the same domain , as described below [7] . All the simulations are run under a modified version of GROMACS [42] . For the seven species we studied in this work , the native structures of single domains that were used to construct the models for SH3 , SH2 , PDZ , TNfn3 , Titin I27 , GB1 and Ubiquitin correspond to PDB entries 1SHG [43] , 1TZE [44] , 2VWR , 1TEN [45] , 1TIT [46] , 1GB1 [47] and 1UBQ [48] respectively . For the single domains of SH3 ( 1SHG ) , TNfn3 ( 1TEN ) and GB1 ( 1GB1 ) , additional linker sequences of Asp-Glu-Thr-Gly , Gly-Leu and Arg-Ser , respectively , are added between the two domains to mimic the constructs used in the corresponding experiments [9 , 14 , 26] . Construction of the Titin I27 model was described in our previous work [7] . In order to allow for domain-swapped misfolding , the native contact potentials within a single domain are also allowed to occur between corresponding residues in different domains , with equal strength . Specifically , considering each single repeat of the dimeric tandem that has L amino acids , given any pair of residues ( with indices i and j ) that are the native interactions within a single domain , the interaction energy for the intradomain interaction ( Ei , j ( r ) ) is the same as the interdomain interaction between the residue ( i or j ) and the corresponding residue ( j + L or i + L ) in the adjacent domain , i . e . Ei , j ( r ) = Ei+L , j ( r ) = Ei , j+L ( r ) = Ei+L , j+L ( r ) . To investigate the folding kinetics of the dimeric tandem , a total of 1024 independent simulations are performed on each system for a duration of 12 microseconds each . Different misfolding propensities are observed at the end of the simulations . With the exception of Ubiquitin and GB1 , the vast majority of the simulations reached stable native states with separately folded domains . A small fraction of simulations form stable domain-swapped misfolded states . All the simulations are started from a fully extended structure , and run using Langevin dynamics with a friction of 0 . 1 ps−1 and a time step of 10 fs . We note that all the generated domain-swapped misfolding structures , containing the central and terminal domains , can be monitored by a reaction coordinate based on circular permutated native-like contact sets . Each circularly permuted misfold can be characterized according to the loop position K in sequence where the native domain would be cut to form the circular permutant ( K = 0 corresponds to the native fold ) . If a native contact Cnative = ( i , j ) exists between residues i and j in the native fold , the corresponding native-like contacts for the central ( Cin ( K ) ) and terminal domains ( Cout ( K ) ) of the domain swapped conformation are generated as C i n ( K ) = ( i + Θ ( K − i ) L , j + Θ ( K − j ) L ) , C o u t ( K ) = ( i + Θ ( i − K ) L , j + Θ ( j − K ) L ) , where Θ ( x ) is the Heaviside step function and L is the length of each single domain ( plus interdomain linker ) . Sin , K is the set of native-like contacts Cin of the central domain , and Sout , K is the set of all the native-like contacts Cout of the terminal domain . Sin , K and Sout , K can be used to define a contact-based reaction coordinate to analyze the kinetics of the dimeric tandem misfolding . The corresponding fraction of contacts for the central domain could be calculated by: QK ( χ ) =1N∑ ( i , j ) ∈Sin , K11+eβ ( rij ( χ ) −λrij0 ) , ( 1 ) where N is the total number of domain swapped contacts , SK = Sin , K ∪ Sout , K ( equal to the total number of native contacts ) , rij ( χ ) is the distance between residue i and j in the protein configuration χ . r i j 0 is the corresponding distance in the native structure for native-like contacts , β = 50 nm−1 and λ = 1 . 2 is used to account for fluctuations about the native contact distance . The equilibrium properties of a single domain of each system are obtained from umbrella sampling along the native contacts Q as the reaction coordinate . The obtained melting temperature of each system is listed in Table A in S1 Text . A temperature at which the folding barrier ΔGf of approximately ∼ 2 . 5 kBT is chosen for the 2-domain tandem simulations for reasons described below . The stability ΔGs is calculated as Δ G s = - k B T ln ∫ Q ‡ 1 e - F ( Q ) / k B T d Q / ∫ 0 Q ‡ e - F ( Q ) / k B T d Q , ( 2 ) where kB and T are the Boltzmann constant and temperature respectively . Q‡ is the position of the barrier top in F ( Q ) , separating the folded and unfolded states and F ( Q ) represents the free energy profile on Q . Barrier heights ΔGf were simply defined as ΔGf = G ( Q‡ ) − G ( Qu ) , where Qu is the position of the unfolded state free energy minimum on Q . We calculated the relative contact order [49] , RCOK of different circular permutants K via RCO K = 1 L · N ∑ ( i , j ) ∈ S in , K \| i - j \| , ( 3 ) where L is the length of the single domain , and N is the total number of the native like contacts ( the same for different K ) . Sin , K is the contacts set of the circular permutant corresponding to the “central domain” of the misfolded state . Note that the contact order calculation here is using residue-based native contacts ( the same ones defined as attractive in the Gō model ) , instead of all atom native contacts . An Ising-like model was built based on the native contact map , in which each residue is considered either folded or unfolded and so any individual configuration can be specified as a binary sequence , in a similar spirit to earlier work [50–52] . Interactions between residues separated by more than two residues in the sequence are considered . To simplify the analysis , we also consider that native structure grows only in a single stretch of contiguous native residues ( native segment ) , which means the configurations such as …UFFFUUUUU… or …UUUUUFFFU… are allowed , however , …UFFFUUUFFFU… is not allowed ( “single sequence approximation” ) [50] . Each residue which becomes native incurs an entropy penalty ΔS , while all possible native contacts involving residues within the native segment are considered to be formed , each with a favourable energy of contact formation ϵ . The partition function for such a model can be enumerated as: Z = ∑ χ exp [ − G ( χ ) k B T ] = ∑ χ exp [ − n ( χ ) ϵ − N f ( χ ) T Δ s k B T ] where kB and T are the Boltzmann constant and temperature . G ( χ ) is the free energy determined by the number of native contacts n ( χ ) in the configuration χ , and the number of native residues , Nf ( χ ) . The distribution of the microstates ( χ ) can be efficiently generated by the Metropolis-Hastings method with Monte Carlo simulation . In each iteration , the state of one randomly chosen residue ( among the residues at the two ends of the native fragment and their two neighbouring residues ) is perturbed by a flip , from native to unfolded or from unfolded to native , taking the system from a microstate χ1 with energy E1 to a microstate χ2 with energy E2 . The new microstate is subject to an accept/reject step with acceptance probability P acc = min [ 1 , exp ( - E 2 - E 1 k B T ) ] . ( 4 ) To mimic the folding stability difference between native and circular permutant folds , a penalty energy term Ep has been added whenever the native fragment crosses the midpoint of the sequence from either side ( the function θ ( χ ) above is 1 if this is true , otherwise zero ) . That situation corresponds to formation of a domain-swapped structure , in which there is additional strain energy from linking the termini , represented by Ep . We only use the Ising model here to investigate formation of the first domain ( either native or circular permutant ) , by rejecting any proposed Monte Carlo step that would make the native segment longer than the length of single domain , L . In order to characterize the potential misfolding properties of each type of domain , we have used a Gō-type energy function based on the native structure . Such models have successfully captured many aspects of protein folding , including ϕ-values [53 , 54] , dimerization mechanism [55 , 56] , domain-swapping [57–60] , and the response of proteins to a pulling force [61 , 62] . More specifically , a Gō type model was used in conjunction with single-molecule and ensemble FRET data to characterize the misfolded states and misfolding mechanism of engineered tandem repeats of Titin I27 [7 , 12] . We have therefore adopted the same model . Although it is based on native-contacts , it can describe the type of misfolding we consider here , which is also based on native-like structure . Note that this model effectively assumes 100% sequence identity between adjacent domains , the scenario that would most likely lead to domain-swap formation . It is nonetheless a relevant limit for this study , as there are examples in our data set of adjacent domains having identical sequences which do misfold ( e . g . titin I27 ) and those which do not ( e . g . protein G ) . For each of the folds shown in Fig 2 , we ran a large number of simulations , starting from a fully extended , unfolded chain , for sufficiently long ( 12 μs each ) such that the vast majority of them reached either the correctly folded tandem dimer , or a domain-swapped misfolded state similar to that shown in Fig 1e for titin . In fact , for each protein , a number of different misfolded topologies are possible , illustrated for the Src SH3 domain in Fig 3 . Each of these domains , shown in conventional three-dimensional cartoon representation in the right column of Fig 3 and in a simplified two-dimensional topology map in the left column , consists of two native-like folded ( or misfolded ) domains . For convenience , we call the domain formed from the central portion of the sequence the “central domain” and that from the terminal portions the “terminal domain” . We have chosen to characterize each topology in terms of the position , K , in sequence after which the central domain begins . Thus , the native fold has K = 0 , and all the misfolded states have K > 0 . Typically , because of the nature of domain swapping , K must fall within a loop . Of course , there is a range of residues within the loop in question that could be identified as K and we have merely chosen a single K close to the centre of the loop . This position , and the central domain , are indicated for the Src SH3 misfolded structures in Fig 3 . We note that each of these central domains can also be considered as a circular permutant of the native fold , in which the ends of the protein have been joined and the chain has been cut at position K . With this nomenclature in hand , we can more easily describe the outcome of the folding simulations for the seven domain types considered in terms of the fraction of the final frames that belonged to the native fold , versus each of the possible misfolded states . These final populations are shown in Table 1 . We see that for five of the domains ( SH3 , SH2 , PDZ , TNfn3 , Titin I27 ) , misfolded structures are observed , with total populations ranging from 5–10% . For the remaining two domains , Ubiquitin ( UBQ ) and protein G ( GB1 ) , no misfolded population is observed . The ability to capture domain-swapped misfolds with simple coarse-grained simulations potentially allows us to investigate the origin of the misfolding , and its relation , if any , to the topology of the domain in question . However , we also need to benchmark the accuracy of the results against experiment as far as possible , in order to show that they are relevant . There are two main sources of information to validate our results . The first is the overall degree of domain-swapped misfolding for those proteins where it has been characterized , for example by single molecule AFM or FRET experiments . Qualitatively we do observe good agreement , where data is available: in experiment , domains which have been shown to misfold are TNfn3 ( AFM ) and Titin I27 ( AFM , FRET ) , which are both found to misfold here , while there is no detectable misfolded population for protein G ( AFM ) , again consistent with our results . We also do not observe any misfolding for Ubiquitin , consistent with the lack of experimental evidence for fully domain-swapped species for this protein [15–23] . Quantitatively , the fractional misfolded population is also consistent with the available experimental data . For instance , the frequency of misfolded domains in native tenascin is ∼ 4% as shown by previous AFM experiments [8] , the misfolded population of I27 dimers is ∼5% in single-molecule FRET experiments [7] while the misfolded population of GB1 domains in polyproteins ( GB18 ) is extrememly low ( < 0 . 2% ) [14] . Even though the observed misfolding population of the misfolded tandem dimer is low , it is potentially a problem considering that many of the multidomain proteins in nature have large number of tandem repeats , such as Titin which contains twenty-two I27 repeats [63] . Recent FRET experiments on I27 tandem repeats have shown that the fraction of misfolded proteins increases with the number of repeats . For the 3- and 8-domain polyproteins , the fraction of misfolded domains increases by a factor of 1 . 3 and 1 . 8 , respectively , relative to a tandem dimer [12] . The second type of evidence comes from experimental structures of domain-swapped dimers . For several of the proteins , bimolecular domain-swapped structures have been determined experimentally . While no such structures have yet been determined for single-chain tandem dimers , we can compare the misfolded states with the available experimental data . For each experimental example , we are able to find a corresponding misfolded species in our simulation with very similar structure ( related by joining the terminis of the two chains in the experimental structures ) . The domain swapped dimers solved obtained from experiments ( Fig 4a , 4c , 4e and 4g ) are strikingly similar to the domain swapping dimeric tandem from simulations , which are the domain swapped SH3 domains when K ( sequence position after which the central domain begins ) = 37 ( Fig 4b ) , SH2 with K = 72 ( Fig 4d ) , TNfn3 with K = 28 ( Fig 4f ) and PDZ with K = 23 ( Fig 4h ) . Most of these states have relatively high population among all the possible misfolds as observed from the simulations ( “Population” in Table 1 ) . While the coverage of possible domain swaps is by no means exhaustive , the observed correspondence gives us confidence that the misfolded states in the simulations are physically plausible . Having shown that the misfolding propensities we obtain are qualitatively consistent with experimental evidence ( and in the case of Titin I27 , in semi-quantitative agreement with single-molecule FRET ) , we set out to establish some general principles relating the properties of each domain to its propensity to misfold in this way . We can start to formulate a hypothesis based on the alternative folding and misfolding pathways illustrated in Fig 1 . Native folding has as an intermediate a state in which either the N- or the C-terminal domain is folded . In contrast , on the misfolding pathway , the first step is formation of the central domain , followed by that of the terminal domain . This parallel pathway scheme suggests that a descriptor of the overall misfolding propensity may be obtained from the rate of formation of a single correctly folded domain , relative to that of the central domain ( neglecting back reactions , because this are rarely seen in our simulations ) . We can study the central domain formation in isolation , since these structures are just circular permutants of the native fold , i . e . the two proteins have the same sequence as the native , but with the position of the protein termini moved to a different point in the sequence , as is also found in nature [35] . These structures can be thought of as originating from the native by cutting a specific loop connecting secondary structure elements ( the free energy cost of splitting such an element being too high ) , and splicing together the N- and C- termini . In the context of the tandem dimers , the position at which the loop is cut is the same K that defines the start of the central domain in sequence . We investigate the role of the central domain by characterizing the free energy landscape of the single domain of each system , as well as all of its possible circular permutants , using umbrella sampling along the reaction coordinate QK . QK is exactly analogous to the conventional fraction of native contacts coordinate Q [64] , but defined using the corresponding ( frame-shifted ) contacts in the circular permutant pseudo-native structure . The index K indicates the position along the sequence of the WT where the cut is made in order to convert to the circular permutant . The free energy surfaces F ( QK ) of two representative systems , SH3 and Ubiquitin , are shown in Fig 5 , with the data for the remaining proteins given in the Fig A in S1 Text . The free energy barrier height for folding ΔGf and the stability ΔGs are listed in the Table 1 . The free energy plots indicate that the single domains of Ubiquitin and GB1 are stable only for the native sequence order , and not for any of the circular permutants . Based on the type of misfolding mechanism sketched in Fig 1 , one would expect that unstable circular permutants would result in an unstable central domain , and consequently no stable domain-swappping misfolding would occur in the dimer folding simulations , as we indeed observe . This is also consistent with previous studies of polyproteins of GB1 and Ubiquitin using using AFM experiments , which reveal high-fidelity folding and refolding [14 , 65 , 66] . We note that only under very strongly stabilizing conditions is any misfolding observed for ubiquitin dimers: running simulations at a lower temperature ( 260 K ) , we observe a very small ( 1 . 3% ) population of misfolded states from 1024 trial folding simulations . At a higher temperature of 295 K , once again no misfolding is observed . In contrast to the situation for GB1 and Ubiquitin , all of the circular permutants of the SH3 domain in Fig 5 are in fact stable , although less so than the native fold . The destabilization of circular permutants relative to native is in accord with the experimental results for the Src SH3 domain [26] ( rank correlation coefficient stabilities is 0 . 80 ) . The other domains considered also have stable circular permutant structures . This is consistent with the fact that all of these domains do in fact form some fraction of domain-swapped misfolded states . The simplest view of the misfolding mechanism would be as a kinetic competition between the correctly folded intermediates versus the domain-swapped intermediates with a central domain folded ( i . e . a “kinetic partitioning” mechanism [67] ) . In this case one might naively expect that the propensity to misfold would be correlated with the relative folding rates of an isolated native domain and an isolated circular permutant structure . However , the folding barriers ΔGf projected onto Q ( for native ) or QK ( for circular permutants ) show little correlation to the relative frequency of the corresponding folded or misfolded state , when considering all proteins ( Table 1 ) . Since this barrier height may not reflect variations in the folding rate if some of the coordinates are poor ( yielding a low barrier ) or if there are large differences in kinetic prefactors , we have also directly computed the folding rate for the circular permutants of those proteins which misfold , and confirm that the rates of formation of the native fold and circular permutants are similar . We indeed obtain a strong correlation between the folding rate of the isolated circular permutant and the folding barrier ΔGf ( Table B in S1 Text ) , which implies Q is a sufficiently good reaction coordinate here . We have also considered the relative contact order ( RCO ) as a proxy for the folding rate , since it has been found to correlate with folding rates for two-state folding proteins [49 , 68] . However , the RCO calculated based on the native or circularly permuted folds did not correlate with either the barrier height for single domain or circular permutant folding , or with the extent of misfolding in dimeric tandem proteins ( Table 1 ) . Since the folding rates do not explain misfolding propensities by themselves , another possibility is that the reverse reactions have to be considered . However , once they had formed , in most cases we did not observe unfolding of the first native domain , or of the intermediate with central domain folded , indicating that back reactions should not be needed , at least to explain the simulation data . This lack of refolding is a consequence of the significant stability of the native folds , which controls the relative folding and unfolding rates ( and indirectly , those of the circular permutants ) . Under these conditions , given that folding rates are much higher , once a native fold ( or circular permutant misfold ) has formed , it is much more likely that a second domain will fold , rather than the first domain unfolding . Our choice of stabilizing conditions was motivated by the fact that misfolding is observed in experiment under conditions where the folded state is much more stable , and the stabilities ( ΔGs ) of the folded single domains in our simulations are generally comparable to those in experiment ( experiment vs simulation , UBQ: 6 . 1 vs 4 . 2 , GB1: 5 . 3 vs 3 . 1 , PDZ: 7 . 5 vs 4 . 5 , SH3: 4 . 1 vs 9 . 2 , Titin: 7 . 5 vs 8 . 1 , Tnfn3: 5 . 3 vs 8 . 1 kcal/mol ) [26 , 69–72] . On the other hand , we did note that there was a significant , and unexpected , correlation between the population of the final folded or misfolded states and the stability ΔGs of the corresponding intermediate . Spearman rank correlation coefficients between the folded stability ΔGs of the intermediate structure and the frequency of folded/misfolded states were 0 . 63 , 0 . 94 , 0 . 74 , 0 . 81 , 0 . 86 for the SH3 , PDZ , TNfn3 , SH2 and Titin I27 domains respectively . We note that there is also a reasonable correspondence between the relative stabilities of circular permutants in simulation and experiment , where data are available [12 , 26] . How can the correlation with stabilities rather than folding rates of the isolated domains be understood ? The resolution lies in the difference between the folding to either type of intermediate represented in Fig 1 , and folding of the single domain “models” for these species , namely that the intermediates fold in the context of the full sequence . This is important because a large fraction of native ( or native-like ) contacts are shared between the native fold and the various misfolded domains . As such , the native and misfolded states can be considered as belonging to the same folding funnel , with differentiation between the two occurring at a late stage of folding . This scenario is illustrated schematically in Fig 6 , in which folding to either a state with one native domain folded ( on left ) , or one possible domain-swapped misfolded intermediate ( on right ) are considered . The states of the proteins are represented by very coarse-grained contact maps ( e . g . representing contacts between pairs of β-strands [73] , rather than between residues ) . As can be seen , dividing this funnel into the separate funnels by considering only native contacts for the native or circularly permuted fold would be misleading ( green and red funnels respectively ) , since the two funnels share several configurations , and many of their states can be converted to one in the other funnel by flipping a single coarse-grained “residue” between folded and unfolded states . We can see this explicitly by plotting some representative folding transition paths from the Src SH3 dimer simulations . In Fig 7 top row , we show a folding event for a simulation which forms a native fold ( at the N-terminus ) , and in the bottom row , for a simulation which forms a circularly permuted central domain with K = 18 . Each event is projected onto two different reaction coordinates , QK , for K = 0 ( standard native Q ) and K = 18 ( the Q when the circular permutant for K = 18 is considered as “native” ) . As is evident , a large fraction of the transition path looks very similar in Fig 7b and 7d , with contacts that could be considered equally as native-like or central domain-like being formed initially in the lower left part of each plot . Around Q0 ≈ Q18 ≈ 0 . 5 , the first trajectory moves toward the native structure , where it terminates ( Fig 7b ) . The second trajectory also deviates initially more toward the folded structure , but then switches back near the end to form the central domain structure instead ( Fig 7d ) . A similar branching of folding pathways has also been proposed in a recent computational study of domain swapped dimer formation [74] . The common funnel picture helps to explain why the stability of the isolated native or circular permutant domains may be correlated with their frequency of formation in the context of the full length sequence in which either could potentially be formed . Initially , nucleation of folding could occur by formation of native contacts anywhere in the sequence . Indeed , they are most likely to form near the centre of the chain . However , as more native/native-like structure is accumulated , the nascent , partially folded protein will be biased to form the contacts leading to the lower free energy structure , and so the folding nucleus is likely to move towards one of the termini of the protein . We note that while a previous study suggested that the stability of the individual domains might be affected by conjugation to another folded domain [75] , this is unlikely to be relevant because in our case the misfolding is controlled by formation of the first domain , while the second domain is still unfolded . Further insight into how the above free energy bias influences the outcome of the folding kinetics can be obtained by considering the progressive formation of folded structure . In order to characterize the location of nascent folded structures , we define a new order parameter i j ¯ representing the average position of native contacts along the sequence , i j ¯ ( χ ) = 1 \| S ( χ ) \| ∑ ( i , j ) ∈ S ( χ ) i + j 2 , ( 5 ) where ( i , j ) is the native or native-like contact formed by the residues i and j in the configuration χ , and S ( χ ) is the set of all such contacts which are formed in χ . We can locate the position of nascent structure in the sequence by plotting the distributions of i j ¯ ( χ ) for χ drawn from the equilibrium distribution at selected values of the global coordinate Q , defined as the fraction of native contacts in the native dimer structure ( i . e . Q = 0 . 5 corresponds to a single folded or misfolded domain; both native and native-like contacts are counted , and divided by the total number of contacts in the native state ) . Fig 8 shows that early in folding , at low Q values ( shaded histograms in Fig 8 ) , the distribution of i j ¯ is broad , and centered in the middle of the sequence . This implies that folding could potentially begin at many positions along the sequence , with no initial preference for folded or circularly permuted structure . However , as folding proceeds closer to formation of a complete domain , i j ¯ develops two maxima , one in the N-terminal and one in the C-terminal part of the chain , corresponding to native domain formation . The nascent native-like structure thus naturally migrates towards the termini to avoid the free energy penalty of forming a circularly permuted misfolded intermediate . The results from the previous sections show that the misfolding propensity is highly correlated with the the stability of the isolated native domain and its circular permutants . To further explain how this might occur , we investigate the dependence of the misfolding propensity on the stability of the central domain in the context of full sequence ( dimeric tandem ) . We have constructed an even simpler simulation model for formation of the first intermediate ( native , or circularly permuted ) , by using a simulation of a Wako-Saito-Muñoz-Eaton Ising-like model [51 , 76] . In the version we consider here , each residue is considered either to be folded or unfolded , so that each configuration can be described as a binary string . Furthermore , we impose the single-sequence approximation , namely that all native-like structure forms in a single segment of contiguous residues . We also restrict the number of folded residues to be at most one half of the dimer sequence length , so that only a single folded or misfolded domain can form , the aspect we are most interested in . To model the stability difference between native and circularly permuted domains , we introduce an additional energy penalty Ep for any folded segment which crosses the midpoint of the dimer sequence . Such a folded segment must be forming a circular permutant misfold and as such will incur some additional “strain” energy from joining the termini of the original fold . We show results from a typical Monte Carlo trajectory for this model in Fig 9 . We have used two parameters to characterize the results , the fraction of native or native-like contacts , Qres , and i j ¯ ( x ) ( Eq 5 ) . Qres equals to the number of residues which are in the native-like state divided by the total number of residues of one domain ( L ) . The projection of a trajectory for the model onto i j ¯ in Fig 9a shows that the most stable states occur for i j ¯ in the center of either the first of the second natively folded domain . Nonetheless , there are other stable states at intermediate i j ¯ , which correspond to the circular permutant intermediates . These have a lower stability , because a value Ep > 0 was used in this instance . The effect of the stability penalty for the circular permutants is illustrated by the two-dimensional free energy surfaces F ( Q res , i j ¯ ) in Fig 9b–9d . In all cases , there are minima at low Qres corresponding to unfolded structures and at high Qres for folded ( native or circular permutants ) . If the penalty Ep = 0 ( Fig 9b ) , in addition to the stable native folds at i j ¯ ∼ 30 and i j ¯ ∼ 90 , there are a variety of other free energy minima at high Qres corresponding to circular permutants , which have essentially identical free energy to the native fold . However , as Ep is increased , the relative population of these misfolded states decreases ( Fig 9c and 9d ) , as expected . Knowing that misfolding correlates with the relative stability of the native single domain and its circular permutants is useful because it suggests a means to predict the likelihood of misfolding , provided one has an estimate of the circular permutant stability . While one could attempt to determine this experimentally , by synthesizing the circular permutants , or computationally , as we have done , it would be very helpful to have a quick method to estimate this stability a priori [77] . Here , we have developed such a method , based on an alchemical transformation from the native to the domain-swapped misfolded state: the overall conversion between the native and circular permutant can be expressed as the sum of the free energy changes of two steps as shown in Fig 10: firstly , joining the N- and C- termini of WT ( Fig 10a ) to form a cyclic intermediate state ( I ) ( Fig 10b ) , in which procedure the free energy change is ( ΔGJ ) . Note that this step is the same for all circular permutant folds . The second step is to cut different loops on the cyclic configuration I to form a circular permutant CP ( Fig 10c ) with the change of free energy ( ΔGC ) : Assuming the native and circular permutant unfolded states to have the same free energy , the overall change in stability between the circular permutant ( CP ) and the wild-type ( WT ) is: Δ Δ G tot = Δ G CP − Δ G WT = Δ G J + Δ G C . For the first step ( Fig 10a and 10b ) , in order to join the two termini , they must be sufficiently close . In general , bringing them closer together will require peeling off a small part of the native structure , starting from the termini . If we imagine that all of the structure between residues i at the N-terminus and j at the C-terminus remains native , then the change in free energy for linking the termini for cyclization , ΔGJ , can be split into energetic and entropic components: Δ G J = ( E I - E WT ) - T ( S I - S WT ) . Assuming the states of the residues p ∈ {0 . . i − 1} and q ∈ {j + 1 . . L} , which are on the N- and C- termini respectively , change from the native state to non-native state ( joint loop ) , the total energy increase will be ∑x ∈ σ ( p ) ϵpx + ∑x ∈ σ ( p ) ϵqx , which is the summation of all the native contact energy , in the Go model over the sets of residues σ ( p ) , σ ( q ) involving residues p and q respectively . x represents the residues that form the native contacts with either p or q . We approximate the entropy gained per residue as δs , where δs = ∑native ( i , j ) ϵij/ ( TN ) , where N is the number of residues and T is the folding temperature . The gained entropy is set to 0 if residue p or q does not have any contact with other part of the protein except for the neighboring residues , and the number of such residues is denoted by κ . The average length contribution ( r0 ) of peeling off each residue from the native structure is set to 3 . 5 Å here . The topological requirement of joining the two termini by peeling off residues 1 to i − 1 , and j + 1 to L from the native state is that the linear distance between the residues i and j ( d ( Ri , Rj ) ) on the native structure is shorter than the effective length contributed by the joint parts: d ( R i , R j ) < ( i + L - j - M ) r 0 ( 6 ) Note that if N- and C- termini point in opposite directions , such as the TNfn3 , Titin I27 , UBQ and GB1 domains ( Fig 2 ) , around six residues ( three on each side ) of the two termini will form the turn of the joint loop which does not contribute to the the effective length . Therefore , an offset number M = 6 is used in this case . This is justified because turns in proteins are usually defined by four residues ( or 3 residue-residue bonds ) [78] . For SH3 , SH2 and PDZ domains ( Fig 2 ) , whose N- and C- termini align to the same direction , M is set to 0 . With the above condition ( Eq 6 ) , the minimum overall change of ΔGJ by adjusting i and j could be given by: Δ G J = min i , j { ∑ p ∈ 0 , . . , i - 1 ϵ ( p , x ) + ∑ q ∈ j + 1 , . . , N ϵ ( q , x ) - T ( i - 1 + L - j - κ ) δ s } , where i ∈ { 0 , 1 , . . , 9 } and j ∈ { L - 9 , . . , L } Analogously , for the second step ( Fig 10b and 10c ) , assume the loop is cut at the position between residue position K and K + 1 , the states of the residues on each side of the cutting point p′ ∈ {i′ , . . , K} and q′ ∈ {K + 1 , . . j′} , will change from the native state to the non-native state . The gained entropy per residue is δs . ∑x ∈ σ ( p′ ) ϵp′x and ∑x ∈ σ ( q′ ) ϵq′x are the summation of all the energy of the native contacts which are broken due to the cutting . Therefore , by comparing different combinations of i′ and j′ , the minimum change of stabilities ΔGC in this step is: Δ G C = ( E C − E I ) − T ( S C − S I ) = min i ′ , j ′ { ∑ p ′ ∈ i ′ + 1 , . . , K ϵ ( p ′ , x ) + ∑ q ′ ∈ K + 1 , . . , j ′ − 1 ϵ ( q ′ , x ) − T ( j ′ − i ′ − 1 ) δ s } , where i ′ ∈ { K - 3 , . . , K } and j ′ ∈ { K + 1 , . . , K + 4 } . The ΔΔGtot calculated using this alchemical free energy method is very well correlated with the stability of the circular permutant ΔGs obtained by umbrella sampling simulations ( Table 1 ) as shown in Fig 11 . It is consistent with the experimental results that GB1 and Ubiquitin have the most unstable circular permutant folds in general . The main contribution of ΔΔGtot is from ΔGJ , since the enthalpy penalty is large when many native contacts are broken by joining the terminis such as in the case of UBQ and GB1 . The free energy cost of cutting the loop ( ΔGc ) is relatively small and is similar for all the circular permutants during the transformation ( Fig 10 ) . From the alchemical method we can see that the difference of stability ΔΔG largely depends on the native contacts that are broken in the procedure when joining the N- and C- termini . However , ΔΔG could also be lowered by extending the linkers between domains , for instance , by adding extra residues at the two termini . If the loop formed by the linker is long enough , fewer native contacts will need to be disrupted , so that the circular permutant folds would be more stable . Therefore we have investigated the stability of circular permutant folds as a function of the length of C-terminal extension , by adding Gly-Ser repeats ( forming no native contacts ) . This extra peptide corresponds to the linker between the tandem domains . The stability of circular permutants , obtained from simulations using umbrella sampling , as a function of linker length ( ll ) is shown in Fig 12 ( raw potentials of mean force on QK in Figs C and D in S1 Text ) . As expected , longer linkers between the tandem repeats give more stable circular permutants . The relative change from ll = 0 to ll = 20 is roughly the same for all circular permutants of a given protein , as expected since the change in all cases is the same loop extension . Note , however , that the effect is much larger for ubiquitin than for TNfn3 . To investigate the consequences of the change of central domain stability for the misfolding propensity of tandem repeats , we carried out first passage simulations of a tandem dimer of Ubiquitin with linker lengths of 5 , 8 and 10 residues respectively . The setup of the dimer simulations was the same previously . For each linker length , 1024 independent simulations were run from fully extended structures . No domain-swapped misfolding was found for ll = 5 and ll = 10 , however , we indeed obtained three domain-swapped misfolding events for ll = 8 . Two of the misfolds belong to the K = 61 ( Fig 12a ) type and the other one is K = 36 type domain-swapping . As one can see from Fig 12a , the circular permutants K = 36 and K = 61 are the ones which are most stable with ll = 8 . However , they are still somewhat unstable , explaining the small fraction of misfolded states obtained . In this case it is clear that the length of the linker between the termini of ubiquitin is one way in which domain-swapped misfolding is avoided in this protein: since ubiquitin is synthesized initially as an N-C linked polyubiquitin chain [40] , it is essential to avoid such misfolding , given the importance of this protein to cellular homeostasis . It should be noted though that the influence of the linker depends very much on the protein , as might be expected , from the much smaller effect on the stability of circular permutants of TNfn3 than those of ubiquitin in Fig 12 . In experiments on titin I27 , the misfolded population was , within error , the same with and without the addition of a four residue RSEL linker [7] . Lastly , we comment briefly on the effect of linker composition . Although we treat the linkers as structureless chains , not forming native contacts , there may be some effect of linker flexibility , arising from the backbone dihedral potential in our model . To test for this effect , we have carried out an additional 1024 independent simulations with the dimeric tandem repeat of the SH3 domain using a different four residue linker composition , GGGG , rather than the original DETG , as used in the original circular permutant studies by Serrano et al [26] . With the new linker GGGG , the observed misfolded populations are 94 . 3% , 1 . 3% , 2 . 3% and 2 . 1% for K = 0 , 18 , 37 and 46 respectively . The differences are not statistically significant compared to the results with the original linker . We have investigated the factors which favour formation of domain-swapped misfolded states in multidomain proteins , by building on knowledge of the folding/misfolding mechanism . Counter to our original expectations , the misfolding yield does not depend primarily on the relative folding rates of the native single-domain protein and its circular permutants , representing intermediates for correct folding and misfolding respectively . Although the folding rates of wild-type and circular permutants may often be quite similar , the fraction of misfolded protein is much smaller than this comparison would suggest . Instead , it appears that misfolding is correlated with the stability of the native single-domain protein relative to its circular permutants . This can be understood because the rate of formation of the first intermediate ( native-like or misfolded ) occurs in the background of the full-length sequence . In this context , while folding may be initiated at any point in the chain , the nascent structure will tend to migrate towards the N- or C-terminus because of the free energy bias towards the native fold; circular permutants invariably pay a cost in stability for joining the protein termini . Thus the folding rate of isolated circular permutants relative to wild-type protein may not be a good proxy for these rates in the context of the full length sequence , whilst the domain stability is a better guide as to the free energy bias towards a particular structure . This suggests that the rates of formation of these domains inferred from single-molecule experiments [12] should be interpreted as the rates in the context of the full length sequence . In our analysis , we have neglected the effect of back-reactions . Since these occurred rarely in the simulations , they were not needed to explain the results . We have also quantified the effect of linker length on domain swapping , finding that sufficiently long linkers can permit misfolded species to form in cases where they did not for the native spacing . Finally , we have developed a simple model for predicting the stability of misfolded intermediates ( circular permutants of native ) , which should prove useful for determining whether a given protein may be susceptible to this type of misfolding .	Multidomain proteins with tandem repeats are abundant in eukaryotic proteins . Recent studies have shown that such domains may have a propensity for forming domain-swapped misfolded species which are stable for long periods , and therefore a potential hazard in the cell . However , for some types of tandem domains , no detectable misfolding was observed . In this work , we use coarse-grained structure-based folding models to address two central questions regarding misfolding of multidomain proteins . First , what are the possible structural topologies of the misfolds for a given domain , and what determines their relative abundance ? Second , what is the effect of the topology of the domains on their propensity for misfolding ? We show how the propensity of a given domain to misfold can be correlated with the stability of domains present in the intermediates on the folding and misfolding pathways , consistent with the energy landscape view of protein folding . Based on these observations , we propose a simplified model that can be used to predict misfolding propensity for other multidomain proteins .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results" ]	[ "simulation", "and", "modeling", "fluorophotometry", "protein", "structure", "thermodynamics", "research", "and", "analysis", "methods", "fluorescence", "resonance", "energy", "transfer", "proteins", "structural", "proteins", "repeated", "sequences", "molecular", "biology", "spectrophotometry", "free", "energy", "physics", "biochemistry", "biochemical", "simulations", "tandem", "repeats", "protein", "domains", "genetics", "biology", "and", "life", "sciences", "physical", "sciences", "genomics", "computational", "biology", "spectrum", "analysis", "techniques", "macromolecular", "structure", "analysis" ]	2016	Structural Determinants of Misfolding in Multidomain Proteins
Communication between neuronal and glial cells is important for many brain functions . Astrocytes can modulate synaptic strength via Ca2+-stimulated release of various gliotransmitters , including glutamate and ATP . A physiological role of ATP release from astrocytes was suggested by its contribution to glial Ca2+-waves and purinergic modulation of neuronal activity and sleep homeostasis . The mechanisms underlying release of gliotransmitters remain uncertain , and exocytosis is the most intriguing and debated pathway . We investigated release of ATP from acutely dissociated cortical astrocytes using “sniff-cell” approach and demonstrated that release is vesicular in nature and can be triggered by elevation of intracellular Ca2+ via metabotropic and ionotropic receptors or direct UV-uncaging . The exocytosis of ATP from neocortical astrocytes occurred in the millisecond time scale contrasting with much slower nonvesicular release of gliotransmitters via Best1 and TREK-1 channels , reported recently in hippocampus . Furthermore , we discovered that elevation of cytosolic Ca2+ in cortical astrocytes triggered the release of ATP that directly activated quantal purinergic currents in the pyramidal neurons . The glia-driven burst of purinergic currents in neurons was followed by significant attenuation of both synaptic and tonic inhibition . The Ca2+-entry through the neuronal P2X purinoreceptors led to phosphorylation-dependent down-regulation of GABAA receptors . The negative purinergic modulation of postsynaptic GABA receptors was accompanied by small presynaptic enhancement of GABA release . Glia-driven purinergic modulation of inhibitory transmission was not observed in neurons when astrocytes expressed dn-SNARE to impair exocytosis . The astrocyte-driven purinergic currents and glia-driven modulation of GABA receptors were significantly reduced in the P2X4 KO mice . Our data provide a key evidence to support the physiological importance of exocytosis of ATP from astrocytes in the neocortex . ATP acts as neurotransmitter mediating excitatory synaptic transmission and synaptic plasticity in the central nervous system [1] , [2] . There is growing evidence that ATP can also play an important role in signal transfer between neuronal and glial circuits and within glial networks [3]–[6] . ATP can regulate growth and development of neural cells and contribute to various pathological processes [7]–[9] . Action of ATP is mediated by ionotropic P2X and metabotropic P2Y receptors abundantly expressed in many types of neurons and glial cells [1] , [2] . By virtue of the high Ca2+-permeability of P2X receptors and the ability of P2Y receptors to stimulate IP3-dependent Ca2+ release from endoplasmic reticulum , purinergic receptors can transmit robust Ca2+-signals and thereby modulate activity and trafficking of excitatory and inhibitory receptors [10] , [11] . In addition to direct actions mediated by P2 purinoreceptors , ATP can initiate secondary neuromodulation via P1 adenosine receptors after rapid degradation by ecto-nucleotidases to adenosine [2] , [12] . Different mechanisms of ATP release have been identified , such as vesicular release from nerve terminals [13] , [14] and several nonvesicular pathways , including concentration gradient-driven diffusion through gap-junction hemichannels , anion channels , and dilated P2X7 receptors [4] , [7] , [15] , [16] . A physiological role of ATP release from astrocytes has been suggested by the participation of ATP in the propagation of glial Ca2+-waves [17]–[19] and significant contribution of ATP and adenosine to the astroglia-driven modulation of neuronal activity and sleep homeostasis [4] , [12] , [20] . There is growing evidence , albeit obtained mostly in cell cultures , that the release of gliotransmitters may share common mechanisms of vesicular neurotransmitter release such as a dependence on the proton gradient and vesicular transporters , SNARE proteins , andintracellular Ca2+ elevation [9] , [21] , [22] . Importantly , astroglial-driven release of ATP and modulation of synaptic plasticity in the hippocampus were suppressed in transgenic mice expressing a dominant-negative SNARE ( dn-SNARE ) domain selectively in astrocytes [12] . However , the mechanism of gliotransmitter release from astrocytes has been disputed [16] . Ca2+-dependent exocytosis of glutamate and ATP , mainly from cultured hippocampal astrocytes , has been reported [21]–[26] . By contrast , alteration of astroglial InsP3-mediated Ca2+-signaling did not have a significant effect on glutamatergic synaptic transmission in the hippocampal slices [27] , fuelling the debate on the role for glial exocytosis in more intact tissue [16] . Still , more recent in situ and in vivo data demonstrated an effect of astroglial InsP3-mediated Ca2+-signaling on cholinergic modulation of synaptic plasticity in hippocampus and neocortex [28]–[30] . At the same time , two recent studies reported the possibility of nonvesicular release of astroglial glutamate through the TREK-1 and best1 channels [31] and the lack of immunostaining for vesicular glutamate transporters in brain astrocytes [32] , contrasting with the bulk of evidence for glial exocytosis obtained by variety of different techniques [21]–[26] , [33] . Thus , physiological relevance of Ca2+-dependent exocytosis of gliotransmitters remains controversial . In this study , to avoid possible artifacts of cell culture , we investigate release of ATP from acutely isolated cortical astrocytes [34] and astrocytes in the neocortical slices . We provide several lines of evidence for ( 1 ) existence of functional vesicular mechanism of Ca2+-dependent gliotransmitter release in neocortical astrocytes , ( 2 ) quantal P2X receptor-mediated currents directly activated in neocortical neurons by release of ATP from astrocytes , and ( 3 ) glia-driven purinergic modulation of GABAergic transmission that is impaired by astrocytic expression of dn-SNARE or deletion of P2X4 receptors . As a first step in demonstrating the presence of quantal Ca2+-dependent release of ATP from astrocytes , we used a sniffer cell approach where ATP release from astrocyte was detected by HEK293 expressing P2X2 receptors ( see Materials and Methods for details ) . We compared release of ATP from astroglia of somatosensory cortex of wild-type mice and from mice conditionally expressing the SNARE domain of VAMP2 selectively in astrocytes ( dn-SNARE mice ) [12] , [20] . Acutely isolated cortical astrocytes were separately loaded with Ca2+-indicator Fluo-4 and photosensitive Ca2+-chelator NP-EGTA [35] and then distributed into a recording chamber containing preplated HEK293-P2X2 cells ( Figure 1 ) . Whole-cell voltage-clamp recordings were made from a HEK293-P2X2 cell lying in the immediate vicinity to astrocytes ( Figure 1A ) . Identification of astrocytes was confirmed by their functional characterization at the end of experiment , including low input resistance , lack of voltage-gated Na+-conductance , large K+-conductance , large conductance mediated by glutamate transporters , and NMDA receptors lacking Mg2+-block ( Table S1 ) . Thus , influence of nonastrocytic cells in the experiments reported below can be ruled out . Brief flash of UV-light caused the uncoupling of Ca2+ from NP-EGTA ( monitored by Fluo-4 Ca2+-indicator ) . This was accompanied by an asynchronous burst of phasic currents in the adjacent ATP-sensitive sniffer cell in 17 of 20 experiments ( Figure 1B ) . P2 receptor antagonist PPADS ( 10 µM ) prevented the detection of UV-evoked phasic currents ( n = 7 of 7; Figure 1B ) , confirming that they were mediated by ATP acting via P2X receptor . Some phasic currents were observed in the sniffer cells having astrocytes laying on their surface even before triggering the Ca2+-rise in the astrocyte , but at much lower frequency ( Figure 1B ) . These baseline events were not detected in the absence of astrocytes in any of 10 native HEK293-P2X2 cells tested . Application of PPADS ( 10 µM ) reduced the frequency of baseline and UV-evoked events correspondingly by 94%±3% and 97%±3% ( Figure 1C ) , confirming the purinergic nature of all phasic currents . An exocytotic mechanism of ATP release was suggested by the activity-dependent staining of astrocytes with vesicular marker FM1-43 ( Figure 1C ) . To further test the role of an elevation of cytosolic Ca2+ concentration as a trigger for the release of ATP from astrocytes , we applied agonists of astroglial metabotropic PAR-1 receptors ( see also Figure S1 ) [35] , [36] and ionotropic NMDA receptors [6] , [24] . The PAR-1 receptors were chosen as an ample method of astrocyte activation because of the ability to activate IP3 pathway and their predominant expression in glial but not in neuronal cells [35] , [36] . Similar to UV-uncaging , activation of cytosolic Ca2+-transients in the astrocyte either by application of the PAR-1 agonist TFLLR ( 10 µM; n = 10 ) or by application of NMDA ( 20 µM; n = 7 ) elicited bursts of spontaneous currents in the adjacent ATP-sensitive sniffer cells ( Figure 2A ) . TFLLR and NMDA did not evoke any activity in the HEK293-P2X2 cells when applied in the absence of astrocytes ( n = 5 agonist , unpublished data ) . The phasic currents in the sniffer cells had amplitudes of 11 . 2±2 . 4 pA ( n = 34 ) and rise and decay time of 1 . 6±0 . 5 ms and 15 . 2±3 . 9 ms correspondingly ( Figure 2D ) , thus resembling parameters of purinergic synaptic currents [13] , [14] . As a test of a vesicular mechanism of ATP release , we isolated astrocytes from cortical slices pretreated with blocker of vacuolar-type H-ATPase bafilomycin A1 ( 1 µM ) for 2 h . Treatment with bafilomycin caused a decrease in the amplitudes and frequency of phasic currents initiated by Ca2+-elevation in the astrocytes ( Figure 2B ) . The mean amplitude of phasic purinergic currents activated in the sniffer cell by stimulation of bafilomycin-treated astrocytes was only 3 . 1±0 . 4 pA ( n = 19 ) . The overall charge transferred by phasic currents activated after stimulation of bafilomycin-treated astrocytes by UV , TFLLR , and NMDA was 9 . 7%±4 . 2% ( n = 7 ) , 11 . 8%±5 . 6% ( n = 5 ) , and 12 . 6%±5 . 8% ( n = 7 ) of the corresponding control values ( Figure 2E ) . In support of exocytotic mechanism of ATP release , astrocytes obtained from dn-SNARE mice elicited a diminished burst of purinergic currents in sniffer cells regardless of the method used to elevate cytosolic Ca2+-level ( Figure 2C , E ) . We also observed the SNARE-complex-dependent release of ATP from the isolated hippocampal astrocytes ( Figure S2 ) . The amplitude histograms of the phasic P2X-mediated currents activated by elevation of astrocytic Ca2+ exhibited prominent second peak ( Figure 2D; see also Figure S3 ) at amplitude twice that of the primary peak . Fitting of amplitude distribution with simple multiquantal binomial model ( shown in Figure 2D as dotted line ) gave a quantal size of 7 . 9±0 . 13 pA and release probability of 0 . 28±0 . 04 . Similarly , fitting of the distributions of P2X-currents evoked in sniffer cell after application of NMDA and TFLLR gave quantal size of 8 . 14±0 . 15 pA and 8 . 05±0 . 11 pA , respectively . It should be noted that astrocyte-driven purinergic currents observed in our experiments had much faster kinetics ( 10–25 ms ) than nonvesicular release of gliotransmitters from hippocampal astrocytes , which was mediated by TREK-1 potassium channels and best1 chloride channels [31] . To verify the lack of contribution of nonvesicular mechanisms to the quantal purinergic phasic currents in the sniffer cells , we activated astrocytes by TFLLR in presence of TREK-1 channels inhibitor fluoxetine [37] and large conductance chloride channels inhibitors DIDS and NPPB [38] . Application of fluoxetine ( 100 µM ) , DIDS ( 300 µM ) , and NPPB ( 100 µM ) did not have marked effect on the astrocyte-evoked phasic currents in the HEK293 cells in any of five experiments for each inhibitor ( Figure S4 ) . Combined , the above results strongly suggest that activity-dependent release of ATP from cortical astrocytes occurs mainly via quantal exocytotic mechanism , dependent on SNARE protein complex . Vesicular mechanism of ATP release from neocortical astrocytes was also supported by immunostaining of living isolated astrocytes with antibodies to vesicular nucleotide transporter VNUT1 and various vesicular , neuronal , and glial marker proteins ( Figures S5 and S6 ) . Although immunostaining of live cells has certain limitations ( see Text S1 ) and should be interpreted with great caution , our data suggest the good co-localization of VNUT1 and synaptic vesicle ( SV ) markers ( Figure S5A ) , which is in agreement with previous reports of vesicular location of VNUT1 [26] and presence of synaptic-like vesicles in astrocytes [22] , [24] . We observed weaker correlation between VNUT1 and lysosomal markers cathepsin D and LAMP3 ( Figure S5B , C ) , which goes in line with previous observation of astroglial ATP release by the lysosome exocytosis [23] . However , lysosomal exocytosis from astrocytes exhibited much slower kinetics [23] than the purinergic currents measured in the sniffer cells ( Figures 1 and 2 ) . This argues against a major contribution of this mechanism to the present observations . As the kinetics of sniffer cell responses are more consistent with millisecond time-scale of SV exocytosis from astrocytes [24] , we suggest that the astrocyte-driven purinergic currents observed in our experiments could be triggered by exocytosis of ATP from synaptic-like vesicles . We also observed an immunoreactivity for vesicular glutamate transporter ( VGLUT1 ) in the fraction of cortical astrocytes ( Figure S5D , E; Figure S6A ) , which goes in line with data reported previously for hippocampal and cortical astrocytes [22] , [33] . Of course , further investigation of exocytosis of glutamate from neocortical astrocytes is required , which is beyond the scope of this article . Previously we have shown that cortical pyramidal neurons express ionotropic P2X purinoreceptor , which can be activated by synaptic release of ATP [13] . Hence , it might be plausible to detect the glia-driven contribution to purinergic current in neurons . We recorded whole-cell currents in the pyramidal neurons of neocortical layer 2/3 of brain slice at membrane potential of −80 mV in the presence of glutamate receptor antagonists CNQX ( 50 µM ) and D-APV ( 30 µM ) and irreversible blocker of GABA receptors picrotoxin ( 100 µM ) . Like our previous results [13] , we observed residual nonglutamatergic excitatory spontaneous synaptic currents ( Figure 3A , B ) ; neither amplitude nor frequency of residual currents were affected by further increase in concentrations of glutamate and GABA receptors antagonists ( n = 10 cells tested , unpublished data ) . The amplitude of inward spontaneous excitatory currents ( sEPSCs ) was reduced by specific antagonists of P2X receptors PPADS ( 10 µM ) and NF-279 ( 3 µM ) correspondingly by 45%±1 3% ( n = 7 ) and 56%±19% ( n = 16 ) ; the sEPSC frequency was reduced by PPADS and NF-279 by 65%±22% and 69%±27% , respectively ( Figures 3 and S7 ) . At concentrations used , both PPADS and NF279 are selective for P2X receptors [39]–[41] . Based on these data as well as our previous work [11] , [13] , the spontaneous inward currents observed in cortical neurons in the presence of glutamatergic and GABAergic antagonists can be confidently attributed to the ATP receptors . The partial inhibitory action of PPADS and NF-279 on nonglutamatergic sEPSCs could be explained by participation of homomeric P2X4 receptors , which are insensitive to these antagonists [1] , [41] , [42] . Since P2X4 subunit-containing receptors are abundantly expressed in the brain and could potentially contribute to neuronal purinergic signaling [1] , [42]–[44] , we used previously characterized P2X4 receptor knockout mice ( P2X4 KO ) [38] . In the P2X4 KO mice , application of 10 µM PPADS decreased the amplitude and frequency of nonglutamatergic sEPSCs ( Figure 3C ) by 74%±10% and 97%±6% correspondingly ( n = 12 ) ; difference from the wild-type mice was statistically significant with p = 0 . 05 and 0 . 01 . Taking into account that significant attenuation of sEPSCs can put them below the detection threshold , nonglutamatergic sEPSCs can be confidently attributed to neuronal P2X receptors . We triggered the release of gliotransmitters from the astrocytes by rapid application of an agonist of PAR-1 receptor TFLLR to neocortical slices . As in hippocampus [35] , [36] , TFLLR ( 10 µM ) triggered cytosolic Ca2+ rise predominantly in astrocytes ( Figure S1 ) . Application of TFLLR caused a dramatic increase in the frequency of ATP-mediated spontaneous currents ( Figure 3A , D ) . TFLLR also elevated astrocytic Ca2+ in dn-SNARE mice , but the burst of purinergic spontaneous currents was not detected ( Figure 3B , E ) . In the P2X4 knockout mice , the average increase in the purinergic sEPSCs frequency reached only 28%±15% ( n = 12 ) , which was significantly lower ( p<0 . 005 ) than 72%±21% increase observed in wild-type mice ( Figure 3C , F ) . Combined , these results demonstrate that activation of astrocytes can evoke synaptic-like purinergic currents in neurons . In addition to effects of knocking out and inhibiting P2X receptors , the purinergic nature of glia-driven spontaneous currents was corroborated by inhibitory action of ATP-hydrolyzing enzyme apyrase ( Figure S8 ) . The apyrase application significantly decreased the mean amplitude and frequency of sEPSCs and abolished the TFLLR-induced burst . The decrease in the sEPSCs frequency is most likely related to the reduction of their amplitude below threshold of detection . In the wild-type mice , purinergic sEPSCs showed bimodal amplitude distributions ( Figure 3G–I; see also Figure S9 ) with peaks at 3 . 1±0 . 9 pA and 5 . 7±1 . 6 pA ( n = 14 ) ; decay time distributions had peaks at 9 . 1±0 . 9 ms and 15 . 3±1 . 8 ms ( Figure 3G ) . TFLLR selectively increased the probability of detection of smaller and slower sEPSCs in all 14 neurons tested . In contrast , recordings made from cortical neurons of dn-SNARE mice did not show two peaks in the distributions of amplitude or decay time; the amplitude and decay time were not altered after TFLLR application ( Figure 4H ) . In the P2X4 KO mice , the amplitude distribution of purinergic sEPSCs showed peaks as 2 . 5±0 . 7 pA and 4 . 2±1 . 1 pA; decay time distribution had peaks at 9 . 0±0 . 9 and 15 . 7±1 . 7 in control . Activation of astrocytes caused just a moderate increase in the proportion of smaller and slower sEPSCs in the P2X4 KO neurons ( Figure 3I ) . Elimination of the distinct population of smaller and slower currents by astrocytic dn-SNARE expression strongly suggests that this population of purinergic currents was elicited by exocytosis of ATP from astrocytes . The vesicular origin of slower purinergic sEPSCs was also supported by elimination of these events by treatment of the cortical slices with 1 µM bafilomycin A1 ( Table S2 ) . The slower and faster purinergic currents recorded in cortical neurons exhibited different quantal behavior ( Figure S7 ) . The slower purinergic currents evoked by application of TFLLR in the presence of TTX ( Figure S3A ) exhibited multiquantal amplitude distribution , whereas faster currents exhibited monoquantal distribution typical for miniature synaptic currents ( Figure S7B , C ) . Thus , detailed analysis of purinergic sEPSCs in the pyramidal cortical neurons revealed two distinct populations of events , which differ by their amplitude and kinetics . Based on their insensitivity to astrocytic dn-SNARE expression , larger and faster sEPSCs most likely have a neuronal origin . In contrast , sEPSCs of smaller amplitude and slower kinetics can be attributed to the vesicular release of ATP from astrocytes . In the following sections , we provide further experimental support of this notion . We sought to obtain a parallel line of evidence for the vesicular ATP release from cortical astrocytes via an alternative approach: we measured ATP concentration in neocortical slice using microelectrode biosensors ( see Materials and Methods ) , a technique that has been applied previously for evaluation of ATP release in several brain areas [7] , [19] . Selective activation of astrocytic PAR-1 receptor by TFLLR induced a robust increase in the extracellular ATP concentration in the cortical tissues of wild-type mice; this increase was impaired in the dn-SNARE mice and was blocked by bafilomycin , confirming its astroglial origin and vesicular nature ( Figure S10A ) . The increase in the “tonic” concentration of extracellular ATP after activation of astrocytes in the wild-type mice reached 1 . 1±0 . 4 µM ( Figure S10B ) and was inhibited by 84%±5% ( n = 7 ) after incubation with bafilomycin . The TFLLR-evoked elevation of ATP concentration in the dn-SNARE mice was decreased by 56%±12% ( n = 12 ) as compared to wild-type . These results support the significant contribution of vesicular mechanism to the activity-dependent release of ATP from cortical astrocytes . Taking into account that bafilomycin can inhibit only re-charging of ATP-storing vesicles and not all the cortical astrocytes express dn-SNARE protein , one could not expect the full inhibition of vesicular ATP release in these experiments . Thus , the incomplete inhibition of TFLLR-evoked ATP transients in the dn-SNARE mice and in the bafilomycin-treated neocortical slices of wild-type mice could hardly be attributed to a large contribution of nonvesicular release . We have shown previously that stimulation of intracortical afferents is able to significantly elevate cytosolic Ca2+-level in the cortical astrocytes acting via ionotropic and metabotropic receptors to glutamate and ATP [5] , [6] , [34] . We asked whether an episode of high-frequency stimulation ( HFS ) could similarly trigger release of ATP from astrocytes in situ . As before , we monitored spontaneous purinergic currents in the pyramidal cortical neurons at a membrane potential of −80 mV in the presence of CNQX and picrotoxin ( Figure 4 ) . The short HFS train triggered more than a 2-fold elevation of the frequency of purinergic sEPSCs in the pyramidal neurons of wild-type mice ( Figure 4A , C ) . Such an elevation did not occur in the dn-SNARE mice , where HFS train caused only a modest transient increase in the sEPSC frequency ( Figure 4B , D ) . The HFS-induced changes in the amplitude and kinetics of phasic purinergic currents had a complex pattern in the wild-type mice ( Figure 4C , E ) . During the first 30 s after HFS train , the average amplitude of events increased to 12 . 4±4 . 2 pA ( n = 6 ) as compared to 8 . 6±2 . 4 pA in the baseline conditions and their decay time was slightly larger ( 11 . 2±2 . 3 ms , n = 6 ) than in control ( 9 . 9±2 . 7 ms ) . The sEPSCs appeared in the neurons of wild-type mice 1–3 min after HFS train had lower amplitudes ( 6 . 8±1 . 6 pA , n = 6 ) and much larger decay times ( 13 . 4±3 . 6 ms ) than in control . Analysis of the amplitude and decay time distributions ( Figure 4E ) revealed a significant increase in the number of smaller , slower sEPSCs after HFS train that formed the distinct fraction of purinergic spontaneous events . In addition to smaller and slower currents , the number of fast sEPSCs with fast decay times ( 9 . 2±2 . 5 ms ) and large amplitudes ( 19 . 6±2 . 7 pA ) was observed during the first 30 s immediately after stimulation ( Figure 5E , purple lines ) . The amplitudes of these large currents corresponded to the double of unitary amplitude of fast purinergic currents; this explains a short-lived increase in the average amplitude immediately after HFS train . The existence of two functionally distinct populations of purinergic events in the wild-type mice was corroborated by correlation between amplitude and decay time of sEPSCs ( Figure 4G ) . The slower currents ( decay time of 15 . 4±2 . 2 ms ) had smaller amplitudes ( 5 . 5±1 . 3 pA , n = 6 ) , but the faster currents ( decay time of 9 . 2±1 . 3 ms ) had higher amplitude ( 9 . 9±2 . 4 pA , n = 6 ) . The amplitudes of slower currents closely agree with quantal amplitude of TFLLR-evoked slow purinergic currents recorded at −80 mV , whereas quantal amplitude of fast currents is close to the unitary size of TFLLR-insensitive fast purinergic sEPSCs ( Figure S6B ) . The train of HFS significantly increased the number of slower spontaneous currents with smaller amplitude ( Figure 4G ) . In contrast to wild-type mice , the only effect produced by HFS train on purinergic sEPSCs in the dn-SNARE mice was the transient increase in the number of double-quantal fast currents ( Figure 4D , F ) , which led to the brief increase in the average amplitude . The expression of dn-SNARE in astrocytes caused a selective loss of the smaller and slower sEPSCs ( Figure 4H ) . These data strongly support the different origins of fast and slow purinergic sEPSCs , from neuronal terminals and astrocytes correspondingly . To verify that slower purinergic sEPSCs originated from astrocytic ATP release directly , we tested the effect of diadenosine triphosphate ( AP3A ) and UTP , which have been shown presviously to strongly inhibit transport of ATP into astrocytic vesicles [26] . Since these substance are not specific VNUT antagonists and can have an action on purinergic receptors , we applied them intracellulary to minimize side effects . In order to increase the impact of single-cell perfusion , we chose neuron-astrocyte pairs lying in a close proximity ( Figure 5A ) . A similar strategy has been previously used to test the effects of perfusion of Ca2+-chelators into astrocytes [45] . The feasibility of this approach is based on the high probability of synapses of a single neuron falling within functional island enwraped and controlled by single nearby astrocyte [46] . When astrocytes of wild-type mice were perfused only with fluorescent dye , two consecutive HFS episodes caused the burst of slow purinergic sEPSCs in the neighboring neurons ( Figure 5B ) of the magnitude similar to previous experiments with intact astrocytes ( Figure 4 ) . Prolonged intracellular perfusion of astrocyte with 1 mM of AP3A or UTP significantly attenuated the frequency of purinergic sEPSCs ( Figure 5B–E ) in 10 of the 12 experiments . The effect was more prominent after the second HFS episode . This was most likely related to the depletion of the releasable pool of ATP in astrocytes . Analysis of amplitude and decay time distributions showed that inhibitors of vesicular nucleotide transporters selectively affected the fraction of slower sEPSCs ( Figure 5C ) , significantly decreasing their amplitude and frequency ( Figure 5D ) . Although slower purinergic sEPSCs were not abolished completely , this might be explained by incomplete perfusion of distal astrocytic processes or release from other astrocytes . These results strongly suggest that smaller and slower purinergic sEPSCs originated directly from vesicular release of ATP from neighboring astrocytes . In summary , our data provide compelling evidence that quantal release of astrocytic ATP activates a distinct population of purinergic currents in cortical pyramidal neurons . ATP release from astrocytes can activate neuronal P2X and P2Y receptors . The following increase in cytosolic Ca2+-signals may trigger a variety of intracellular cascades implicated in the modulation of synaptic strength [1] , [2] , [11] . In particular , phosphorylation of postsynaptic GABAA receptors might provide an endogenous pathway for Ca2+-dependent regulation of synaptic strength [47] , [48] . To test this hypothesis , we recorded inhibitory synaptic currents ( IPSCs ) in neocortical pyramidal neurons at a membrane potential of −40 mV ( Figure 6 ) in the presence of glutamate receptor antagonists CNQX ( 50 µM ) and D-APV ( 30 µM ) . Under these conditions , we observed inward purinergic currents simultaneously with outward Cl−-currents mediated by GABAA receptors ( Figure 6 ) . Similar to our previous results [13] , outward IPSCs were completely inhibited by bicuculline in all 12 cells tested ( unpublished data ) . The burst of ATP-mediated currents , induced in cortical neurons by activation of astrocytic Ca2+ signaling via PAR1 receptors ( as shown in Figure 3A ) , was accompanied by significant decrease in the amplitude of GABA-mediated synaptic currents ( Figure 6A , B ) . Amplitudes of evoked IPSCs and spontaneous mIPSCs recorded in the wild-type mice were reduced after application of 10 µM TFLLR by 44 . 3%±6 . 1% ( n = 7 ) and 39 . 4%±6 . 3% ( n = 12 ) correspondingly . In the dn-SNARE mice , the inactivation of GABA receptor-mediated synaptic currents was greatly diminished ( Figure 6A , B ) . Application of TFLLR reduced the amplitude of evoked and spontaneous IPSCs in the cortical neurons of dn-SNARE mice just by 6 . 4%±8 . 7% ( n = 8 ) and 4 . 3%±6 . 6% ( n = 8 ) , respectively . The difference in the action of TFLLR on the IPSCs in the wild-type and dn-SNARE was statistically significant with p<0 . 005 for both evoked and spontaneous currents . These results confirm the importance of astroglial exocytosis for the observed inactivation of GABA receptors . It also indicates the lack of unspecific action of PAR-1 agonist on inhibitory synaptic transmission . The effect of TFLLR on inhibitory synaptic currents was mimicked by application of nonhydrolysable ATP analog ATP-γS ( 10 µM ) and was considerably reduced by P2 receptor antagonist PPADS ( 10 µM ) . In the P2X4 KO mice , activation of astrocytes decreased the amplitude of evoked and spontaneous IPSCs ( Figure 6A , B ) only by 14 . 3%±8 . 2% and 15 . 7%±9 . 5% ( n = 12 ) ; the difference between P2X4 KO and wild-type mice was significant with p<0 . 005 . These data strongly support the participation of neuronal ATP receptors in the astrocyte-driven modulation of IPSCs . We found out that exocytosis of gliotransmitters also caused long-term homeostatic modulation of inhibitory synaptic transmission . We observed a marked difference in the amplitude distribution of the baseline ( control ) amplitude of postsynaptic inhibitory currents in the wild-type and dn-SNARE mice ( Figure 6C ) . The amplitude of mIPSCs in the cortical neurons before the application of TFLLR was much higher in the dn-SNARE than in the WT mice . The average baseline amplitude of mIPSCs was 23 . 5±8 . 3 pA in the dn-SNARE mice ( n = 8 ) and 14 . 9±6 . 9 in the WT mice . Application of TFLLR caused the leftward shift of amplitude distribution of mIPSCs only in the WT mice; this shift was significantly reduced by PPADS ( Figure 6C ) . The considerable difference in the baseline amplitude of miniature inhibitory currents in the WT and dn-SNARE mice provides the first evidence that vesicular release of gliotransmitters may be involved in the long-term homeostatic regulation of inhibitory transmission in the neocortex . In order to elucidate the role of post- and presynaptic mechanisms in the modulation of inhibitory transmission , we evaluated the changes in the mean quantal content ( Figure 6A ) , changes in the frequency of spontaneous mIPSCs ( Figure 6B ) , and paired-pulse ratio ( PPR ) of evoked IPSCs ( Figure 7 ) . Neither the mean quantal content of IPSCs ( Figure 6A ) nor the mIPSCs frequency ( Figure 6B ) exhibited marked changes . However , we observed a moderate change in the PPR of IPSCs in the neocortical pyramidal neurons of wild-type mice ( Figure 7A ) . The IPSCs evoked with a 50 ms interval in the control showed paired-pulse depression with mean PPR of 0 . 79±0 . 11 ( n = 12 ) , and application of TFLLR increased PPR by 0 . 16±0 . 07 ( n = 8 ) . The application of TFLLR did not cause a significant change in PPR in the dn-SNARE mice ( Figure 7B , D ) , indicating the involvement of glial exocytosis in the mechanism . The effect of TFLLR was significantly reduced by PPADS ( Figure 7A , D ) and reproduced by application of ATP-γS both in the wild-type and dn-SNARE mice ( Figure 7A , B , D ) . Hence , the presynaptic increase in the PPR of IPSCs was most likely mediated by ATP acting via P2 purinoreceptors . This result is consistent with previous reports of presynaptic facilitation of GABAergic synaptic transmission by ATP and P2 receptors [1] , [2] , [49] . There was no significant difference in the effects of TFLLR and ATP-γS between wild-type and P2X4 KO mice ( Figure 7C , D ) . So the P2X4 receptors hardly make the large contribution in the increase of PPR , contrasting with their prominent role in the astrocyte-induced decrease of IPCS amplitude ( Figure 6 ) . One could speculate that presynaptic facilitation of IPSCs by glia-driven ATP can be mediated by other subtypes of P2X receptors and P2Y receptors , whose role in the presynaptic modulation in the brain was widely reported [1] , [2] . More importantly , the above data clearly demonstrate that the large decrease in the amplitude of evoked and spontaneous IPSCs cannot be attributed to the presynaptic modulation of GABA release , and astrocyte-induced down-regulation of inhibitory transmission operates via postsynaptic mechanism . The postsynaptic mechanism of IPSCs inactivation was corroborated by experiments where the Ca2+-dependent phosphorylation of GABA receptors was impaired by intracellular agents . First , we found that activation of P2X receptors in pyramidal neocortical neurons caused marked reduction of GABA-activated currents via Ca2+- and protein kinase C–dependent mechanism ( Figure S11 ) . Second , application of TFLLR to the cortical slices of wild-type mice did not cause marked reduction in the mIPSCs recorded in neurons perfused with intracellular Ca2+-chelator EGTA ( 10 µM ) or protein kinase C antagonist GF109203X ( 30 nM ) . The mIPSC amplitude was reduced by just 2 . 5%±5 . 9% ( n = 6 ) in the presence of intracellular EGTA and by 11 . 2%±5 . 5% ( n = 4 ) in the presence of the protein kinase C antagonist ( Figure S12 ) . These results agree with previous reports on Ca2+- and PKC-dependent down-regulation of GABA receptors [47] . In addition to the fast IPSCs ( “phasic inhibition” ) , central neurons also receive a diffusional inhibitory signal mediated by the extrasynaptic GABA receptors continuously activated by small concentrations of GABA present in the extrasynaptic space ( “tonic inhibition” ) [50] , [51] . Extrasynaptic GABA receptors , responsible for tonic inhibition , have been reported to undergo Ca2+-dependent phosphorylation [47] , [48] . Thus , one could expect the impact of astrocyte-driven ATP not only on phasic but also on tonic inhibitory signaling in the neocortical pyramidal neurons . To verify this hypothesis , we used a conventional experimental paradigm where tonic inhibition is assessed by the change in the whole-cell holding current under action of GABA receptor blockers [50] , in our case 50 µM bicuculline . In the wild-type mice ( Figure 8A , top ) , layer 2/3 neurons showed the tonic current of 39 . 9±8 . 3 pA ( n = 20 ) at membrane potential of −80 mV . The tonic current in the pyramidal neurons dn-SNARE mice was almost two times higher than in the wild-type ( Figure 8B , top ) , reaching 76 . 9±15 . 1 pA ( n = 10 ) . A significant difference in the amplitude of tonic current between wild-type and dn-SNARE mice suggests that exocytosis of gliotransmitters from cortical astrocytes can modulate the activity of extrasynaptic GABA receptors in the adjacent neurons . Consistent with this notion , activation of gliotransmitter release by TFLLR caused a marked upward shift in the holding current in the pyramidal neurons of wild-type mice ( Figure 8A , middle ) , and the rest of holding current was efficiently diminished by bicuculline . Application of TFLLR did not have a notable effect on tonic current in the dn-SNARE mice ( Figure 8B , middle ) . The amplitude of tonic current recorded under action of TFLLR in the wild-type , dn-SNARE , and P2X4 KO mice was correspondingly 12 . 6±6 . 8 pA ( n = 14 ) , 72 . 2±9 . 1 pA ( n = 7 ) , and 43 . 4±7 . 2 pA ( n = 6 ) . The relative decrease in the amplitude of tonic current caused by TFLLR was , respectively , 68%±14% , 6%±8% , and 25%±11% . The down-regulation of tonic current by PAR-1 agonist was considerably attenuated by antagonist of ATP receptors ( Figure 8A , bottom ) . The effects of TFLLR and PPADS in the wild-type mice as well as a difference between mice strains were statistically significant as indicated in Figure 7C . It has been recently shown that Ca-dependent modulation of astrocytic GABA GAT3 transporters in the hippocampus can alter an extracellular GABA level elevating the tonic current and decreasing the mIPSCs due to GABA receptor desensitization [52] . One should note that this pathway did not contribute significantly to PAR-1 agonist-induced modulation of inhibitory currents in neocortical neurons , since we observed a decrease in both phasic ( Figure 6 ) and tonic currents ( Figure 7 ) . This notion was corroborated by our finding that blocking of the astroglial GAT3 GABA transporter increased the tonic current and decreased the amplitude of mIPSCs both in the wild-type and dn-SNARE mice ( Figure S13 ) . These data also show that dn-SNARE expression does significantly alter the activity of GABA transporters in cortical astrocytes . Taken together , our experiments in brain slices demonstrate that exocytosis of ATP from cortical astrocytes in situ can activate the ATP receptors in the adjacent pyramidal neurons , leading to down-regulation of synaptic and extrasynaptic GABA receptors . Detection of ATP released from acutely isolated cortical astrocytes using sniffer cells demonstrated the SNARE protein-dependent exocytosis of ATP from cortical ( Figures 1 and 2 ) and hippocampal ( Figure S2 ) astrocytes . Our experiments in situ show that release of ATP from astrocytes can be sensed by neighboring neurons where glia-derived ATP can activate neuronal purinoreceptors . Quantal behavior of astrocyte-induced phasic purinergic currents ( Figures 3 , 4 , and S6 ) and their elimination in the bafilomycin-treated cortical slices ( Table S2 ) confirm that they originated from the vesicular release of ATP . Combined with previous observation of astroglial purinergic modulation of neuronal activity in the hippocampus , cortex , and basal forebrain [12] , [20] , these data strongly support the universal character of a vesicular mechanism of ATP release as a gliotransmitter . In the sniffer cell experiment we used P2X2 receptors as a detector for ATP . They have a moderate affinity for ATP ( EC50 around 10 µM ) that allowed us to avoid saturation and resolve the quantal behavior of astrocyte-evoked purinergic currents with a mean quantal size of about 8 pA ( Figure 2D ) . To activate such response , the concentration of ATP released from astrocytes should have reached at least the low micromolar range . Detailed calculations , performed using the approach applied for extrasynaptic release and diffusion of glutamate [16] , [53] , [54] , show that such concentration can be reached after release of about 1 , 000 molecules from a single glial synaptic-like vesicle , and the size of the “active spot” due to diffusion of ATP can reach as far as 1–2 µM ( Figures S14A and S15 ) . This estimation is in line with our data obtained using ATP microelectrode biosensors ( Figure S10 ) as well as with previous evidence that the extracellular ATP level in the brain can reach 1–100 µM depending on the physiological and pathological context [1] , [7] , [55] . Similar calculations , as above , argue against involvement of lysosomal release in generation of purinergic currents observed in our experiments . Indeed , the typical lysosome has a diameter of about 300 nm and can contain more than 1 , 000 , 000 molecules of gliotransmitter [16]—that is , 1 , 000 times more than in the synaptic-like vesicle . Even though lysosome undergoes kiss-and-run exocytosis releasing only 10% of its content [16] , one could expect the peak of a sniffer cell response to reach at least 200–500 pA , with most receptors on its surface saturated with agonist . So the observed quantal size of sniffer cell response ( about 8 pA ) is in much better agreement with release of ATP from synaptic-like vesicles . The low micromolar level of ATP concentration can be sufficient for activation of P2X purinoreceptors abundantly expressed in the central neurons [1] , [42] , [44] . The data of electron microscopy and single molecular imaging [39] , [56] , [57] show both synaptic ( mostly at the periphery of synaptic density ) and extrasynaptic location of P2X receptors . Interestingly , P2X2 and P2X4 receptors are generally located peri-synaptically ( i . e . , at the edges of postsynaptic density ) [56] , which makes them accessible by ATP released from both nerve terminals and extrasynaptic glial release sites . This is consistent with our observation of strong sensitivity of astrocyte-driven sIPSCs to deletion of P2X4 receptors and inhibition of P2X2 receptors ( Figures 3 , 4 , and S7 ) . The distance to the source of ATP would affect the kinetics and amplitude of input mediated by peri-synaptic P2X receptors . Due to diffusion and rapid conversion to ADP , the transient of ATP concentration reaching P2X receptor after vesicular release from the distal ( astroglial ) site will have less magnitude but will decay longer than ATP transient after release from a close intrasynaptic site ( as illustrated in Figure S15 ) . This would lead to smaller quantal amplitude and slower kinetics of purinergic current of the glial origin as compared to neuronally activated synaptic currents . So the diversity in location of the ATP source could be the most plausible explanation for the difference in the amplitude and kinetics of purinergic sEPSCs of neuronal and glial origin observed in the neocortical neurons ( Figures 3 and 4 ) . An alternative explanation might be that EPSCs of smaller amplitude and slow kinetics were generated at neuronal synapses at a much longer electrotonic distance . In this case , however , the attenuation factor would increase gradually with distance , leading to smooth single-peaked distributions of EPSC amplitude and decay time due to the presence of large “intermediate” EPSCs . This contrasts with our observation of two peaks in the amplitude and decay time histograms ( Figures 3–5 ) . These arguments are supported by results of computer simulation using the model of neurotransmitter spillover [53] , [54] adapted for release of ATP from glial and synaptic sites ( Figures S14 and S15 ) . Importantly , the key experimental evidence of the astroglial origin of purinergic currents of slower kinetics and smaller quantal size has been provided by their selective inhibition by disruption of vesicular ATP transport in astrocytes ( Figure 5 ) . Although the possibility of nonvesicular gliotransmitter release from astrocytes has been previously reported in several brain regions [4] , [7] , [15] , [58] , our results do not show a strong contribution of nonvesicular pathways to the release of ATP in the neocortex . On a contrary , both sniffer cell and biosensor data suggest that vesicular mechanism brings major contribution to activity-dependent ATP release . One should note that incomplete inhibition of ATP release in the biosensor experiments could be explained by incomplete inhibitory action of bafilomycin treatment and dn-SNARE expression on vesicular release rather than notable contribution of nonvesicular mechanisms . It is possible that vesicular and nonvesicular ATP release operates in the neocortex in the different time scale , similarly to spinal cord astrocytes [15] , where the fast initial exocytosis of ATP can be followed by slow secondary release of ATP through the pannexin and connexin hemichannels [15] . So it is plausible that vesicular and nonvesicular mechanisms of ATP release coexists in the neocortex , but nonvesicular release was not activated in our experimental context . It is traditionally considered that physiological astroglial Ca2+ signaling is driven mainly by InsP3-mediated release from intracellular stores [16] . Calcium signals arising from activation of metabotropic receptors are believed to be primarily responsible for control of exocytotic release of gliotransmitters [9] , [16] , [17] . Although the role of InsP3-induced Ca2+ signaling in astroglial physiology was questioned [27] , more recent data strongly support the importance of this pathway for glia-derived modulation of synaptic transmission in the hippocampus [29] , [45] and neocortex [28] , [30] . We have shown previously that ionotropic NMDA and P2X receptors can mediate a significant fraction of the synaptically driven Ca2+ rise in cortical astrocytes [5] , [6] , [59] . Our present data suggest that astrocytic NMDA receptors can trigger the release of ATP , acting in parallel to the metabotropic receptors ( Figure 2 ) . The capability of ionotropic receptors to control of exocytosis of gliotransmitters ( Figure 2 ) might account for a lesser than expected impact of modulation of astroglial InsP3 signaling on synaptic transmission and plasticity . Our results shown in Figures 1–4 provide a strong evidence of ATP release by exocytosis and support previous observations of presence of vesicular ATP transporters and synaptic-like vesicles in astrocytes [22] , [24] , [26] . Observation of vesicular location of VNUT1 ( Figures S5 and S6 ) , even with inherent limitations of immunostaining procedure ( see Text S1 ) , agrees with previous reports [22] , [24] , [26] and with our functional data as well . Co-localization of VNUT1 and synaptic vesicle markers ( Figure S5 ) , rather small quantal size and fast kinetics of glia-driven purinergic currents ( Figures 1–4 ) , suggest that release of ATP from synaptic-like vesicles can make a significant contribution into purinergic gliotransmission in cortical astrocytes . Until recently , the action of ATP as gliotransmitter was associated mainly with presynaptic adenosine receptors [4] , [12] , [20] . It was also shown that astrocyte-derived ATP could operate through P2X7 receptors to enhance presynaptic release of glutamate in hypothalamic neurons [10] . Although we observed a moderate presynaptic modulation of inhibitory transmission ( Figure 7 ) , our main finding is that glia-driven ATP can directly activate excitatory currents in neurons acting via postsynaptic P2X receptors ( Figures 3–5 ) . We showed that activation of neuronal purinoreceptors by astrocyte-driven ATP caused a dramatic change in the inhibitory transmission in neocortex inhibiting postsynaptic and extrasynaptic GABAA receptors ( Figures 6 and 7 ) . ATP receptors can act through the Ca2+-dependent phosphorylation by protein kinase C ( Figures S11 and S12 ) ; the latter mechanism is intrinsic for GABA receptors [43] , [60] . The physiological relevance of exocytosis of gliotransmitters was strongly supported by our finding of significant increase in the baseline phasic and tonic inhibitory signaling in the cortical neurons of dn-SNARE mice ( Figures 6 and 7 ) . This result suggests that release of gliotransmitters , presumably ATP , can be involved in long-term homeostatic regulation of neuronal GABA receptors by a mechanism that has yet to be investigated . Our evidence of down-regulation of inhibitory synaptic transmission in the neocortex goes in line with observations that enhanced Ca2+ signaling in cortical astrocytes contributed to neuronal excitotoxicity and epilepsy [18] , [61] . Our results outline a novel mechanism of action of ATP as a gliotransmitter: in addition to catabolism of ATP to ADP and adenosine and modulation of synaptic transmission via presynaptic purinoreceptors [12] , ATP can enhance neuronal excitability by down-regulating the phasic and tonic inhibition acting via postsynaptic P2 receptors . Our present ( Figures 6–8 ) and previous [11] results show that modulation of postsynaptic receptors activated by Ca2+ influx via purinergic P2X receptors can provide an efficient mechanism of regulation of signaling within tripartite synapse . Data on a large contribution of P2X4 receptors to glia-driven modulation of neuronal signaling ( Figures 6–8 ) go in line with a previous observation of a facilitatory role for astroglial release of ATP [12] and P2X4 receptors [43] in the long-term potentiation of synaptic transmission in the hippocampus . Our findings give further insight into the role of P2X receptors in the CNS . There is growing consensus that , despite a clear evidence of participation of P2X receptors in the excitatory synaptic transmission in several brain areas , they bring notable contribution to slow neuromodulation rather than fast excitation [44] . The above results suggest that P2X receptors can also mediate glia-to-neuron signals , which can be activated in the millisecond time scale and have more long-lasting consequences for neuronal excitability . Hippocampal astrocytes have been recently reported to decrease the amplitude of mIPSCs via Ca2+-dependent modulation of GABA transport [52] and increase the frequency of mIPSCs via P2Y receptors of inhibitory interneurons , presumably activated by slow release of ATP from astrocytes through connexin hemichannels [62] . Our data suggest that presynaptic ATP receptors can enhance GABA release in the neocortex ( Figure 7 ) . Down-regulation of both phasic and tonic postsynaptic GABA receptors by astrocyte-driven ATP ( Figures 7 and 8 ) can act downstream of these pathways and significantly affect their impact on inhibitory transmission . Interplay between post- and presynaptic pathways of glial modulation could have diverse effects on neuronal excitability in different physiological contexts . Apparently , the down-regulation of inhibitory transmission provided by postsynaptic P2X receptors prevails in the neocortical pyramidal neurons , at least in our experimental conditions ( Figures 6 and 7 ) . It becomes evident now that even release of one gliotransmitter , ATP ( followed by formation of ADP and adenosine ) , can activate a variety of pre- and postsynaptic regulatory cascades that can affect synaptic efficacy in opposite ways . Combined with diversity of vesicular and nonvesicular pathways of ATP release and the possibility of release of other gliotransmitters , such as glutamate and D-serine [9] , [33] , [63] , this may confer more complex behavior to tripartite synapse than previously assumed [16] , [27] . In summary , our data suggest that Ca2+ elevation in cortical astrocytes , which might occur in response to signals from neurons and/or propagation of glial Ca2+ waves , can trigger exocytosis of ATP from synaptic-like vesicles and activate neuronal P2X receptors that are located at the periphery of synapse . Ca2+ signaling via neuronal ATP receptors can cause phosphorylation-dependent inhibition of postsynaptic GABA receptors acting downstream of astroglial modulation of presynaptic GABA release and GABA uptake . Our results strongly support physiological importance of exocytosis of gliotransmitters , in particular ATP , in communication between astrocytes and neurons and modulation of synaptic efficacy . Mice ( 8–12 wk and 9 mo old ) were anaesthetized by halothane and then decapitated , in accordance with UK legislation . Brains were removed rapidly after decapitation and placed into ice-cold physiological saline containing ( mM ) NaCl 130 , KCl 3 , CaCl2 0 . 5 , MgCl2 2 . 5 , NaH2PO4 1 , NaHCO3 25 , glucose 15 , pH of 7 . 4 gassed with 95% O2 to 5% CO2 . Transverse slices ( 280–300 µm ) were cut at 4°C and then placed in physiological saline containing ( mM ) NaCl 130 , KCl 3 , CaCl2 2 . 5 , MgCl2 1 , NaH2PO4 1 , NaHCO3 22 , glucose 15 , pH of 7 . 4 , and kept for 1–4 h prior to cell isolation and recording . Astrocytes were acutely isolated using the modified “vibrating ball” technique [34] , [59] . The glass ball ( 200 µm diameter ) was moved slowly some 10–50 µm above the slice surface , while vibrating at 100 Hz ( lateral displacements 20–30 µm ) . Isolation protocol was adjusted to provide a high yield of astroglial cells . This technique preserves the function of membrane proteins and therefore is devoid of many artifacts of enzymatic cell isolation and culturing procedures . In particular , vibro-dissociated astrocytes retain many morphological features ( e . g . , GFAP–EGFP fluorescence , size , proximal processes ) and functional properties ( e . g . , high potassium conductance , glutamate transporters , Ca2+ signaling ) while being completely isolated from neuronal somata and nerve terminals . The composition of external solution for all isolated cell experiments was ( mM ) 135 NaCl , 2 . 7 KCl , 2 . 5 CaCl2 , 1 MgCl2 , 10 HEPES , 1 NaH2PO4 , 15 glucose , pH adjusted with NaOH to 7 . 3 . Astrocytes were identified by their morphology under DIC observation , EGFP fluorescence ( astrocytes from dn-SNARE and GFAP-EGFP mice ) , and functional properties as described previously ( see also Table S1 ) [34] , [59] . In the experiments with dn-SNARE mice , administration of Dox [12] , [20] has been removed 4 wk prior to electrophysiological studies . According to our observations , about 70% of astrocytes in the layer 2/3 of somatosensory cortex in situ and freshly isolated cortical astrocytes exhibited EGFP fluorescence . The astrocytic identity of EGFP reporter and dn-SNARE-transgene-expressing cells has previously been confirmed for the dn-SNARE line [12] , [20] . So one can expect that at least 60%–65% of astrocytes in the somatosensory cortex of dn-SNARE mice express the dn-SNARE domain of synaptobrevin II [20] . In the experiments with freshly isolated astrocytes , only the fluorescent cells have been selected to ensure the impairment of SNARE complex function . In the experiments in the somatosensory cortex of dn-SNARE mice in situ , electrophysiological recordings have been performed in the areas with higher density of EGFP-expressing astrocytes to maximize the putative impact of the loss of SNARE function on neighboring neurons . To increase the probability of a neuron lying within a functional island of synapses [46] controlled by dn-SNARE astrocytes , we recorded from the neurons located in the close proximity to at least two fluorescent cells , as illustrated in Figure S16 . After isolation from a brain slice , cortical astrocytes were incubated with 5 µM Ca2+ indicator Fluo4-AM ( or Rhod-2 AM ) and 10 µM of photoliable Ca2+ chelator o-nitrophenyl-EGTA-AM ( NP-EGTA ) for 30 min , re-suspended in a small volume ( 200–300 µl ) of fresh extracellular medium , and placed over cultured HEK293 cells expressing GFP-tagged P2X2 receptors to ATP ( HEK293-P2X2 , gift from Prof . R . Evans , University of Leicester , UK ) . To evaluate release of ATP , the transmembrane currents were recorded in the HEK293-P2X2 cells that had an astrocyte lying on their surface; simultaneously , elevation in cytosolic Ca2+ concentration was induced in the astrocytes by UV uncaging . It has to be noted that contrary to the experiments with astrocytes in culture [21] , the spatial density of acutely isolated cells in our experiments was rather low , as shown previously [34] . So we could easily select HEK cells contacting astrocytes with no other cell lying in the immediate vicinity ( e . g . , as shown in Figures 1 and S2 ) . We performed recordings from the HEK293 cell–astrocyte couples , which were separated from any other cells by at least 15–20 microns of free space to ensure that HEK293-P2X2 cells were activated by ATP released only from contacting astrocyte . Flash photolysis and fluorescent imaging were performed with the aid of a IX51 inverted microscope and epifluorescent illumination via UPLSAPO60XW/NA1 . 2 objective ( Olympus , Tokyo , Japan ) . To monitor the intracellular Ca2+ level , astrocytes were constantly illuminated at 480±10 nm using OptoLED light source ( Cairn Research , Faversham , UK ) , and fluorescence was measured at 535±25 nm . Astrocytes from dn-SNARE and GFAP-EGFP mice were loaded with Ca2+ indicator Rhod-2 and illuminated at 530±10 nm; fluorescence was measured at 590±30 nm . The fluorescent images were recorded using Retiga 2000R enhanced CCD camera ( QImaging , Canada ) ; exposure time was 35 ms at 2X2 binning . Elevation in the intracellular Ca2+ level was evaluated by a ΔF/F0 ratio averaged over the whole cell image after background subtraction . For uncaging of intracellular Ca2+ , cells were illuminated by a brief pulse ( 200 ms ) of UV light ( 365±10 nm ) emitted by high-power LED NCSU033AT ( Nichia , Tokyo , Japan ) , peak power 500 mW , and estimated power at an objective >150 mW . Illumination was delivered via OptoLED dual-port epifluorescence condenser ( Cairn Research , UK ) . In addition to photolysis , Ca2+ rise was induced in the astrocytes by fast application of agonists of PAR-1 receptors ( 10 µM TFLLR ) or NMDA receptors ( 20 µM NMDA ) ; these agonists did not cause any response in HEK293-P2X2 cells when applied without astrocytes placed over HEK cells . Whole-cell voltage clamp recordings from HEK293-P2X2 cells , neurons , and astrocytes were made with patch pipettes ( 4–5 MΩ for neurons and HEK293-P2X2 cells and 6–8 MΩ for astrocytes ) filled with intracellular solution ( in mM ) : 110 KCl , 10 NaCl , 10 HEPES , 5 MgATP , pH 7 . 35 . Intracellular solution for recording in neurons and astrocytes contained 0 . 2 mM EGTA , and the solution for HEK293-P2X2 cells contained 10 mM EGTA and 1 mM CaCl2 . In some experiments ( simultaneous recording of GABA-mediated and ATP-mediated synaptic currents ) , KCl was replaced by KGluconate . Currents were monitored using an AxoPatch200B patch-clamp amplifier ( Axon Instruments , USA ) filtered at 2 kHz and digitized at 4 kHz . Experiments were controlled by the PCI-6229 data acquisition board ( NI , USA ) and WinFluor software ( Strathclyde University , UK ) ; data were analyzed by self-designed software . Liquid junction potentials were compensated with the patch-clamp amplifier . Series and input resistances were , respectively , 5–7 MΩ and 500–1100 MΩ in the HEK cells and neurons and 8–12 MΩ and 50–150 MΩ; both series and input resistance varied by less than 20% in the cells accepted for analysis . For activation of synaptic inputs , axons originating from layer IV–VI neurons were stimulated with a bipolar coaxial electrode ( WPI , USA ) placed in the layer V close to the layer IV border , approximately opposite the site of recording; the stimulus duration was 300 µs . For the recording of EPSCs and IPSCs , the stimulus magnitude was set 3–4 times higher than minimal stimulus adjusted to activate the single-axon response in the layer II pyramidal neurons as described in [13] . In order to trigger synaptically driven astroglial Ca2+ transients , the single episode of theta-burst stimulation ( HFS ) was delivered; the HFS episode consisted of 5 pulses of 100 Hz stimulation , repeated 2 times with 200 ms interval ( total 10 pulses per episode ) . To monitor the cytoplasmic-free Ca2+concentraton [Ca2+]i , cortical astrocytes were loaded by 40 min incubation with Fura-2AM . For fura-2 excitation , cells were alternately illuminated at wavelengths of 340±5 nm and 380±5 nm using the OptoScan monochromator ( Cairn , Faversham , UK ) . Fluorescent images were recorded using Olympus BX51 microscope equipped with UMPLFL20x/NA0 . 95 objective and 2× intermediate magnification and Andor iXon885 EMCCD camera; exposure time was 35 ms at 2×2 binning; experiments were controlled by WinFluor software . The [Ca2+]i levels were expressed as F340/F380 ratio averaged over the whole-cell image . To investigate the Ca2+ signaling activated by PAR-1 receptor agonist , cortical neurons and astrocytes of wild-type and dn-SNARE mice were loaded with 50 µM Fluo-4 . The whole-cell voltage-clamp recordings were used to confirm the identification of neurons and astrocytes and verify the lack of changes in the basic electrophysiological properties of the dn-SNARE astrocytes . The Fluo-4 fluorescence signal was excited at 488±10 nm and measured at 530±20 nm; the fluorescent images were recorded and analyzed as described above . In parts of the experiments ( Figure 1 , Figure 5 , Figure S5 , Figure S16 ) , two-photon imaging of neurons and astrocytes was performed using Zeiss LSM-7MP multiphoton microscope coupled with the SpectraPhysics MaiTai pulsing laser; experiments were controlled by ZEN LSM software ( Carl Zeiss , Germany ) . Images were further analyzed offline using ZEN LSM ( Carl Zeiss ) and ImageJ ( NIH ) software . For investigation of vesicular dynamics ( Figure 1C ) , acutely isolated cortical astrocytes were loaded with FM1-43 fluorescent dye ( 2 . 5 µM ) , and FM1-43 fluorescence was excited at 820 nm and observed at 560±20 nm . For immunolabeling of vesicular transporters , secretory organelles , and astroglial markers , acutely isolated astrocytes were incubated with 0 . 1 µg/ml of antibodies following antibodies: rabbit polyclonal anti-VNUT1 ( T-12 ) , goat polyclonal anti-cathepsin D ( N-19 , Santa Cruz Biotechnology ) , mouse monoclonal anti-VGLUT1 ( McKA1 ) , mouse monoclonal anti-PSD95 ( 6G6-1C9 ) , mouse monoclonal anti-GLT-1 ( 10B7 , Abcam ) , mouse monoclonal anti-NeuN ( A60 ) , rabbit monoclonal anti-S100b ( EP1576Y , Millipore ) , goat monoclonal anti-SV2A , mouse monoclonal anti-NG2 , and rabbit polyclonal anti-GFAP ( Sigma ) . Prior to cell loading , antibodies were conjugated to the green fluorescent dye DyLight488 ( VNUT1 , VGLUT1 , PSD-95 , NeuN , NG2 , and GFAP ) or red fluorescent dye DyLight594 ( VNUT1 , SV2A , cathepsin-D , LAMP3 , GLT-1 , S100β ) using Lighting-Link antibody conjugation system ( Innova Bioscience , Cambridge , UK ) according to the manufacturer's protocol . Antibodies to VNUT1 ( recognizing intraluminal epitopes ) , VGLUT1 ( recognizing cytoplasmic epitope ) , and cathepsin-D ( intraluminal epitope ) were applied to living astrocytes directly; other antibodies were conjugated with BioPORTER protein delivery reagent ( Genlantis , San Diego , CA ) 10 min prior to incubation . Immediately after isolation from the neocortical slice , living astrocytes were pre-incubated with 2% of normal bovine serum ( Sigma ) in the extracellular saline for 30 min to block unspecific binding sites . After them , cells were gently washed two times with clean extracellular saline for 5 min and then incubated with antibodies for 60 min at room temperature . After incubation , cells were washed with laminar flow of extracellular solution in the microscope recording chamber for 30 min prior to image recording . Fluorescence was excited at 820 nm , GFAP-EGFP and DyLight488 signal was observed at 520±10 nm , and DyLight594 signal was observed at 590±20 nm . The photomultiplier gain of the two-photon microscope was adjusted to avoid saturation in both channels but in the same time did not differ more than 10% between red and green channel . Colocalization analysis of images was carried out using ImageJ software and methods described in [65] . The concentration of ATP within cortical slices ( Figure S10 ) was measured using microelectrode biosensors obtained from Sarissa Biomedical Ltd ( Coventry , UK ) . A detailed description of the properties of ATP biosensors and recording procedure has been published previously in [7] , [55] , [58] . Briefly , biosensors consisted of ATP metabolizing enzymes immobilized within a matrix on thin ( 25–50 µM ) Pt/Ir wire . This allowed insertion of the sensors into the cortical slice and minimized the influence of a layer of dead surface tissue . The concentration of ATP has been calculated from the difference in the signals of two sensors: a screened ATP sensor and screened null-sensor , possessing the matrix but no enzymes . This allowed us to compensate for release of any nonspecific electro-active interferents . Biosensors show a linear response to increasing concentration of ATP and have a rise time of less than 10 s [55] . Biosensors were calibrated with known concentrations ( 10 µM ) of ATP before the slice was present in the perfusion chamber and after the slice had been removed . This allowed compensation of any reduction in sensitivity during the experiment . Biosensor signals were acquired at 1 kHz with a 1400 CED interface and analyzed using Spike 6 . 1 software ( Cambridge Electronics Design , Cambridge , UK ) . All data are presented as mean ± SD , and the statistical significance of difference between data groups was tested by two-population t test , unless indicated otherwise . The spontaneous transmembrane currents recorded in HEK293-P2X2 cells and neurons were analyzed off-line using methods described previously [13] , [54] . Briefly , phasic transmembrane currents of an amplitude higher than 2 SD of baseline noise were selected for the initial detection of spontaneous events . Then every single spontaneous event was analyzed within the 140 ms time window , and its amplitude and kinetics were determined by fitting the model curve with single exponential rise and decay phases . As a rule , mean square error of fit amounted to 5%–20% of peak amplitude depending on the background noise . Whenever error of fit exceeded 25% , spontaneous currents were discarded from further analyses . The amplitude distributions of spontaneous and evoked currents were analyzed with the aid of probability density functions and likelihood maximization techniques [54]; all histograms shown were calculated as probability density functions . The amplitude distributions were fitted with either multiquantal binomial model or bimodal function consisting of two Gaussians with variable peak location , width , and amplitude . The decay time distributions were fitted with bimodal functions . Parameters of models were fit using likelihood maximization routine . For each particular distribution , the fit with quantal or bimodal model was accepted only when it has a confidence level α≤0 . 05 . To monitor and analyze the time course of changes in the amplitude and frequency of spontaneous currents , the amplitude and frequency were averaged over the 1 min time window . For the basic analysis of the time course of quantal parameters of the evoked IPSCs , the mean quantal content was evaluated as reciprocal square of coefficient of variation [54] , and the quantal size was calculated as ratio of mean amplitude to the mean quantal content . All animal work has been carried out in accordance with UK legislation and "3R" strategy . Research has not involved non-human primates .	Brain function depends on the interaction between two major types of cells: neurons transmitting electrical signals and glial cells , which control cerebral circulation and neuronal homeostasis . There is a growing evidence of the participation of astrocytes in regulating neuronal excitability and synaptic plasticity via the release of “gliotransmitters , ” which include glutamate and ATP . The importance of ATP release from astrocytes was suggested by studies that demonstrated its contribution to neuronal activity and sleep homeostasis via modulation of known “purinergic” receptors . But the mechanisms underlying gliotransmitter release and the physiological significance of direct glia-to-neuron communication remain unknown and intensively debated . Here , we investigate the release of ATP from astrocytes of brain neocortex and demonstrate that astrocytes can release ATP by Ca2+-dependent exocytosis , most likely from synaptic-like microvesicles . We also find that vesicular release of ATP from astrocytes can directly activate excitatory signaling in the neighboring neurons , operating through purinergic P2X receptors . We saw that activation of these P2X receptors by astrocyte-driven ATP down-regulated the inhibitory synaptic signaling in the neocortical neurons . Our results imply that exocytosis of gliotransmitters is important for the communication between astrocytes and neurons in the neocortex .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "cell", "physiology", "molecular", "neuroscience", "neurochemistry", "anatomy", "and", "physiology", "electrophysiology", "neuroscience", "homeostatic", "mechanisms", "synaptic", "plasticity", "ion", "channels", "neurotransmitters", "signaling", "pathways", "neurological", "system", "nervous", "system", "physiology", "neurochemicals", "biology", "gamma-aminobutyric", "acid", "biochemistry", "nervous", "system", "components", "cellular", "neuroscience", "neuromodulation", "central", "nervous", "system", "synapses", "neurons", "cellular", "types", "molecular", "cell", "biology", "neurophysiology" ]	2014	Exocytosis of ATP From Astrocytes Modulates Phasic and Tonic Inhibition in the Neocortex
Transcriptional activation domains ( ADs ) are generally thought to be intrinsically unstructured , but capable of adopting limited secondary structure upon interaction with a coactivator surface . The indeterminate nature of this interface made it hitherto difficult to study structure/function relationships of such contacts . Here we used atomistic accelerated molecular dynamics ( aMD ) simulations to study the conformational changes of the GCN4 AD and variants thereof , either free in solution , or bound to the GAL11 coactivator surface . We show that the AD-coactivator interactions are highly dynamic while obeying distinct rules . The data provide insights into the constant and variable aspects of orientation of ADs relative to the coactivator , changes in secondary structure and energetic contributions stabilizing the various conformers at different time points . We also demonstrate that a prediction of α-helical propensity correlates directly with the experimentally measured transactivation potential of a large set of mutagenized ADs . The link between α-helical propensity and the stimulatory activity of ADs has fundamental practical and theoretical implications concerning the recruitment of ADs to coactivators . Control of gene expression plays a crucial role throughout all three evolutionary domains of life , allowing cells to establish cellular identity , adapt to environmental challenges and prevent diseases caused by misregulation of transcription [1] . The expression of the genome is controlled predominantly by a network of gene-specific transcription factors ( GSTFs ) that , after binding to target sites on DNA , regulate the rate of expression of nearby genes . GSTFs performing as transcriptional activators usually contain one or multiple activation domains ( ADs; [2] ) that orchestrate localized remodelling of the chromatin structure , enhanced recruitment of components of the basal transcriptional machinery on the core promoter and/or stimulate promoter escape and subsequent elongation events [3–6] . These activities typically require binding of the ADs to coactivators that integrate and convey activation signals to other components of the transcriptional machinery [6 , 7] . The Mediator complex surrounding the basal transcriptional machinery during transcription initiation [8–11] contains coactivators that have been shown experimentally to interact with ADs to regulate gene-specific transcription ( Fig 1; [11–13] ) . While more than 50 common structural motifs have been described for the DNA-binding domains , the available knowledge concerning the structure and function of ADs is comparatively limited [14] . The first ADs described almost three decades ago were shown to be both necessary and sufficient to confer the transcriptional stimulatory properties [2 , 15] . From a structural perspective , ADs are often characterized by their unusual primary amino acid sequence abundant in acidic amino acids , glutamine or proline residues [14–17] . The enrichment for such amino acids is thought to discourage the formation of higher order structures and thus results in an intrinsically disordered structure ( "acid blobs and negative noodles" or "polypeptide lasso" structures [18–20] ) . In turn , the intrinsic disorder allows ADs to interact in a highly adaptable manner with a range of coactivators , culminating in a synergistic regulation of the basal transcriptional machinery by one or multiple activators ( Fig 1; [21 , 22] ) . The affinity of AD-coactivator binding is reasonably high ( low micro- to high nanomolar range [12 , 21 , 23] ) and results in interactions lasting for several milliseconds . NMR-studies provided structural insights into a various aspects of AD-coactivator complexes ( TFIID/Taf40-VP16 [24]; TFIIH/Tfb1-VP16 ( PDB#2K2U [23] ) ; NcoA1-STAT6 ( PDB#1OJ5 [25] ) ; MDM2-p53 ( PDB#1YCQ [26] ) ; CBP-CREB ( PDB#1KDX [27]; MED25/VP16 ( PDB#2XNF [12] and 2KY6 [13]; GAL11-GCN4 ( PDB#2LPB [11] ) . Site-directed mutagenesis and structural studies have shown that evolutionarily highly conserved bulky hydrophobic residues within ADs play a key structural role in mediating interactions with coactivators ( Fig 2A and S1 Text , [16 , 23 , 24 , 26 , 28] ) . When bound to coactivators , ADs form a "fuzzy" family of stochastically related structures ( Fig 2D , [29–31] ) . Many of the yet unanswered questions regarding AD-coactivator interactions are challenging to address experimentally , especially those concerning the dynamic range of AD conformations over time , key interaction points on coactivator surfaces , the energetics of such interactions and the structures of ADs prior to binding coactivators . Computational approaches are highly effective to model such systems on the atomic level , to study their behavior and gain new mechanistic insights that consolidate present knowledge and guide future experimental work . Here we describe the results obtained from a series of long , fully atomistic molecular dynamics simulations focusing on the experimentally well-characterized GCN4-GAL11 system from Saccharomyces cerevisiae . Accelerated molecular dynamics ( aMD ) methods [33 , 34] provide powerful tools for investigating the binding of the ADs to their coactivator targets , as well as for studying the structural properties of ADs in isolation . We describe the structural interplay of AD-coactivator complexes and explore an extensive experimental data set based on synthetic AD variants to demonstrate a high degree of correlation between the α-helix propensity , degree of "fuzziness" and the transactivation potential . The yeast transcriptional activator GCN4 contains two tandemly arranged ADs ( Fig 2A ) that stimulate the expression of more than 70 "downstream" genes . The GCN4 ADs achieve this task by targeting a variety of components of the basal transcriptional machinery , including the coactivator GAL11 ( also known as MED15 ) within the mediator complex [21] ) . GAL11 contains three structurally independent AD-binding domains ( "Activator-Binding Domains" ["ABDs"]; Fig 2A ) . For one of these , ABD-1 , a high-resolution structure shows a stable α-helical structure that includes a groove for interactions with ADs ( PDB#2LPB; Fig 2B–2D; [11] ) . NOE and spin-labeling data of GAL11/ABD-1 complexed with the central AD of GCN4 ( GCN4-cAD ) were used to create several models illustrating the diversity of interaction between this coactivator and the AD . The bound cAD models contain a short helical stretch ( encompassing GCN4 residues 116 to 124 ) that includes three large hydrophobic residues ( W120 , L123 and F124 ) highly conserved during evolution ( Fig 2A and S1 Text ) . The coactivator GAL11/ABD-1 interaction surface displays three computationally detectable "hot spots" ( "Pocket #1" , "Pocket #2" and "Pocket #3" ) [32] that are distinguished by their concave topology and potentially become occupied by these particular GCN4 hydrophobic residues ( Fig 2C ) . We subjected PDB#2LPB-model 1 to extensive aMD simulations to gain deeper insight into various structural aspects , such as variation in AD secondary structure , orientation relative to the coactivator surface and energetic changes underpinning the conformational changes . Simulations were carried out as four independent replica runs with different initial Boltzmann distributions of particle velocities . Each simulation lasted for one microsecond , but the results reflect a period around two or three orders of magnitude longer due to the acceleration protocol used ( that is , hundreds of microsecond- to millisecond-range; Table 1 ) . We were initially curious to see whether the aMD simulations would recreate the different binding states previously proposed by Brzovic et al . [11] . We used distance measurements between GCN4-W120 or GCN4-F124 relative to GAL11/ABD1-A126 ( which forms the floor of Pocket #1; Fig 2B ) to monitor pocket occupancy . The measurements show that the two key hydrophobic residues , in full agreement with the NMR-based models [11] , behave in a switch-like manner and bind to GAL11/ABD1 in the three major binding states via a series of intermediate conformations ( Fig 3 ) . At various stages , the GAL11-ABD1 pocket is occupied by the sidechains of either GCN-4/W120 ( Fig 3A ) or F124 ( Fig 3C ) , respectively . On several occasions , we observe a double occupancy ( Fig 3B ) . A molecular movie illustrates a full time course of the dynamic change , including the changeover between W120 and F124 ( S1 Movie ) . In addition to pocket occupancy state , the NMR-based models also postulate that the GCN4-cAD helical portion takes up several different orientations relative to GAL11-ABD1 . Angular measurements of vectors characterizing the GCN4-cAD helix relative to GAL11-ABD1 α-helix 4 ( Fig 4 ) correspond to orientations directly comparable to the previously described ones , but also suggest the presence of additional states representing transitional conformations . Because W120 and F124 act as pivot points in a comparable manner , the various pocket occupancy states and helix orientations observed do not appear to show any significant correlation . Because we started the aMD simulations from just one of the 13 different models proposed previously , we wondered to what extent he observed motions of the GCN4-cAD on GAL11-ABD1 reflected the conformational space defined by the twelve remaining models . In principle , any extensive simulation of a single member of a family of structural conformers should reveal conformations that encompass the conformations of the majority of the other family members , as these structures are expected to interconvert freely during simulation . Plots of the phase space of the combined trajectories along three coordinates ( helical angle; distances of the two key hydrophobic residues ( W120 and F124 ) relative to Pocket#1 ) demonstrate that approximately 87% of the model coordinates are within highly populated regions ( Fig 5 ) . We conclude that the choice of 2LPB-model#1 as the starting structure for all four aMD simulations did not result in an unusually biased sampling of conformational space . Flexibility and structural adaptability of the cAD thus enables a highly dynamic interplay that accommodates several different combinations of pocket occupancy and helical orientation . This raises intriguing questions regarding the energetics of such a variable interaction . We calculated the molecular mechanics per-residue decomposition of free binding energy ( ΔGBinding ) measurements along the trajectories in one-nanosecond intervals using the Molecular Mechanics Generalized Bourne Surface Area ( MM-GBSA ) method [35] . The van der Waals decomposition data of the GCN4-cAD confirms the dominating contribution of GCN4-W120 and F124 in binding to GAL11-ABD1 ( Fig 6A; electrostatic interactions play a mostly invariant role in the GAL11-ABD1/GCN4 cAD interaction: S1 Fig ) . Despite the major conformational changes of the GCN4 cAD relative to the coactivator surface , the energetic contributions of GCN4-W120 and F124 interactions remain relatively steady throughout all four independent simulations . A more detailed study of these interactions reveal the varying role of at least five residues within the GAL11-ABD1 Pocket #1 in mediating these contacts ( Fig 7 ) . Two hydrophobic residues , GAL11-M173 or Y220 , interact with GCN4-F124 alternatively , depending on whether the F124 sidechain is located within Pocket #1 ( Figs 3A and 7A ) , or has moved out of it and is replaced by GCN4-W120 ( Figs 3C , 7B and 7C ) . While GCN4-F124 occupies Pocket #1 , W120 makes favorable hydrophobic contacts with the sidechains of GAL11-K217 and K221 ( Fig 7A–7C ) . The formation of alternative—but energetically equivalent—contacts thus underpins several alternative modes of binding that are conformationally quite different from each other . Another conserved residue , GAL4-L123 ( Fig 2A ) , provides notable van der Waals contributions , mostly in conjunction with F124 , and occasionally substitutes for F124 in a reversible manner ( particularly obvious in simulation GAL11-ABD1/GCN4-cAD _aMD_no1; Fig 6A ) . This analysis also identifies several additional residues ( GCN4-M107 , F108 , Y110 , L113 , I128 , and V130 ) as making significant additions to ΔGBinding , but in a distinctly non-systematic manner . The residues are nodes in a structurally highly flexible network that facilitates short-lived interactions , but do not a follow recurrent pattern due to substantial and unstable conformational changes in the GCN4-cAD . To exemplify the role of these residues , we investigated the structural interactions of GCN4-M107 in more detail . In simulation GAL11-ABD1/GCN4-cAD _aMD_no2 , this residue is seen as providing a substantial van der Waals contribution lasting throughout most of the second half of the simulation ( timeframe 1 , 400–2000 ns aMD; Fig 6A ) . During this time , GCN4-M107 interacts predominantly with two leucine residues , L169 and L227 , which are located on two different α-helices of GAL11 ( helix 1 and 4 , respectively ) , but are spatially close to each other and interact with each other via hydrophobic interactions in the folded GAL11 structure . GCN4-M107 interacts with either residue on its own , or even with both leucines by bridging them ( Fig 6B and S2 Movie ) . Altogether , we interpret these findings the following way: GCN4-W120 /GCN4-F124 anchor GCN4-cAD to the GAL 11 surface , but provide no preferential stabilization of conformation and/or orientation of the cAD relative to the coactivator surface . Additional hydrophobic residues located on GCN4 contribute notable , but temporary contacts ( estimated by molecular mechanics measurements to contribute only between -3 to -6 kcal . mol-1 each towards ΔGBinding ) . The fuzzy interaction thus results from a variable combination between two relatively strong contributors ( -8 to -10 kcal . mol-1 each ) that are regularly supported by a host of additional minor contributors subjected to continuous change . This molecular interaction pattern predicts that the binding free energy of the GCN4 cAD to GAL11-ABD1 is far from constant and subject to constant fluctuations . The MM-GBSA estimates support the idea of substantial variation in binding affinity ( over a three-fold range on the micro/millisecond time scale [S2 Fig] ) . The half-life of the GCN4-GAL11 interaction is estimated to be in the low millisecond range [11] , which supports this interpretation . Although the constant change in affinity may result in a reasonable average affinity ( and may include very high affinity states ) , it will also drop in affinity with statistical regularity to levels that facilitate immediate dissociation . The aMD simulations , by allowing coactivator-AD domains to be monitored over longer timer frames , convey this message much clearer than the limited number of currently existing static snap-shots of models demonstrating alternative conformations [11] . Having examined the molecular dynamics of the GAL11-ABD1/GCN4-cAD fuzzy complex , we turned our attention to the intrinsic structural properties of these two interaction partners . Specifically , we wanted to investigate to what extent binding of GCN4 affects the structure of GAL11 and , more importantly , how extensively the GCN4-cAD is structured on its own . We started by monitoring the formation/maintenance of secondary structure of the GAL11-ABD1/GCN4-cAD complex simulations described above . The analysis showed formation of a stable helical structure ( "Helix #1" ) typically encompassing GCN4 residues S117 to D125 ( Fig 8A ) . Simulation GAL11-ABD1/GCN4-cAD _aMD_no1 exceptionally shows a mixture of 310-helix and α-helix during the first 750 ns of simulation before settling into a α-helical pattern , whereas all other three simulations ( including the final stages of GAL11-ABD1/GCN4-cAD _aMD_no1 ) display extensive and stable α-helices throughout the entire time course . These results are essentially in agreement with the 13 different models presented in PDB#2LPB [11] , although the simulations suggest that the C-terminal border of Helix#1 routinely extends one residue further than previously proposed to include position D125 . In addition to Helix #1 , the occasional presence of another N-terminally located structure ( "Helix #2" ) is evident . Helix #2 is less stable in GAL11-ABD1/GCN4-cAD _aMD_no1 , no2 and no3 and either takes up a partial 310-helix conformation ( GAL11-ABD1/GCN4-cAD _aMD_no1 and no3 ) , or disappears eventually . In GAL11-ABD1/GCN4-cAD _aMD_no4 , Helix #1 and #2 fuse into a single contiguous α-helix ( spanning from M107 to N126 at its borders ) that remains intact until the end of the simulation . Helix #2 includes the hydrophobic residues ( GCN4-M107 , F108 , Y110 , L113 , I128 , and V130 ) identified above as making occasional energetically favorable contributions to ΔGBinding . We next asked to what extent the observed α-helical propensity of the GCN4-cAD was encoded within its primary structure . ADs are intrinsically disordered and are often thought to only adopt significant secondary structure upon binding to a coactivator target [11 , 29] . This model is , however , controversial . Whereas some NMR and circular dichroism studies of several isolated ADs claim an absence of significant secondary structure elements [11 , 36] , other investigations suggest the presence of a significant fraction of transient α-helices [37–39] or β-sheets [40] in the unbound state of various ADs . In order to eliminate any structural "memory" from the starting structure , we constructed a model of the GCN4-cAD as a completely unfolded polypeptide from its primary amino acid sequence . After aMD simulation , any conformational changes—including the formation of secondary structure elements -will therefore solely be determined by the intrinsic properties of the polypeptide sequence itself . After a short implicit solvation minimization step to fold up the structure in a more compact random coil , we set up four independent microsecond aMD simulations under identical conditions as used previously for the aMD simulations of the GAL11-GCN4 complex ( Table 1 ) . Such simulations sample folding pathways and , especially relevant for disordered structures , reveal shifts in equilibria between short-lived conformations . The formation of α-helices occurs on the nanosecond-microsecond time scale [41] and is therefore well within the scope of the chosen simulation parameters . An investigation of secondary structure elements formed in the GCN4-cAD aMD simulations reveals an unexpectedly high degree of spontaneously formed α-helices ( Fig 8B ) . The formation of α-helices is especially favored in the central portion of the GCN4-cAD that contains the bulky hydrophobic residues that have been experimentally identified as critically important for the transactivation function , as well as making significant contributions to the free energy of binding to coactivators ( Fig 6A ) . Although traces of β-sheet can be seen in GCN4_aMD_no4 ( Fig 8B ) , these structures appear short-lived and do not support the conclusions reached by a previous study [40] . We conclude that the GCN4-cAD has intrinsic potential to form α-helical elements , even in absence of a coactivator , making it likely that these spontaneously preformed secondary structure elements represent key structural features required for coactivator interaction and binding specificity . The absence of substantial random coil elements in the region surrounding GCN4-W120 , L123 and F124 allows us to postulate further that the cAD engages most likely with the coactivator with the necessary α-helices already locally preformed prior to first contact . Although expected to have a less substantial effect , we also attempted to quantitate the effect of AD binding on the conformation of a coactivator . Consequently , we set up four independent one-microsecond aMD simulations of the GAL11-ABD1 in the absence of the GCN4-cAD ( Table 1 ) . A comparison of root mean square fluctuation ( RMSF ) measurements in simulations GAL11-ABD1 _aMD_no1 to no3 in the bound and unbound state shows that ABD1 becomes structurally more restricted upon cAD binding . Especially ABD1 residues involved in either pocket formation or binding of the cAD helix become less mobile ( S3 Fig ) . GAL11-ABD1 _aMD_no4 undergoes a more substantial conformational change that includes a concerted movement of helices 1 and 2 and alters the ABD1 interaction surface . The original pocket for binding GCN4-W120 or F124 is no longer present , suggesting that this conformation of ABD1 may not be able to bind GCN4 . The altered surface , however , develops new pockets , that may potentially offer alternative binding sites for other activators . Up to now , we have focused our attention on a naturally occurring AD/coactivator complex that has been shown to be physiologically relevant [11 , 21] . Extensive mutagenesis experiments have revealed the existence of a cryptic AD within the primary amino acid sequence of GCN4 . This "cAD-like" activation domain , encompassing GCN4 residues 81–100 , partially matches the structural criteria for an AD , but does not display a detectable transactivation potential ( Table 2; [42] ) . Substitutions of hydrophobic residues within the cAD-like motif improve its activity and make it as potent as the GCN4-cAD . This modified cAD-like domain has proven an excellent testing ground for studying the transactivation potential of an array of directly comparable structures created by high-throughput site-directed mutagenesis [42] . We included two examples in our analyses and will refer from here onwards to the members of this collection as "cAD-like xx" ( where xx stands for the transactivation potential that the sequence confers ) . For example , cAD-like07 refers to a 'weak' cAD-like variant that is capable of stimulating ARG3 induction ~7-fold ( which is equivalent to the activation potential of the GCN4-cAD ) . On the other hand , cAD-like96 identifies a strong transactivator motif capable of stimulating ARG3 induction ~96-fold [42] . The system thus offers ideal conditions for further elucidation of the functional necessities of an exceptionally potent AD and its interactions with coactivators . We set up in silico folding aMD simulations for the cAD-like07 and cAD-like96 motifs under identical conditions used earlier for the GCN4-cAD ( Table 1 ) . Taking into account that we previously observed significant α-helical propensity in the isolated and de novo folded GCN4-cAD ( Fig 8B ) , one of the first questions we asked was whether such a propensity could also be detected in the cAD-like variants ( while embedded within the same primary sequence context as used in the experimental work ) . The only ordered secondary structures formed under these conditions are α-helices , albeit with a noticeable difference in effectiveness . In the case of cAD-like07 , contiguous α-helical regions are present fleetingly throughout most of the simulation period , but these fluctuate considerably in length and position relative to the underlying primary amino acid sequence ( Fig 9A ) . In some instances all α-helical structures were lost , but restored in a fully reversible manner shortly afterwards . We conclude that cAD-like07 displays a notable tendency towards α-helical conformations , but these structures undergo a constant equilibrium between conformations of different α-helical content and are consequently unable to adopt a higher-order structure with a degree of stability exceeding the nanosecond range . In contrast , within the first 200 ns of aMD simulation the cAD-like96 variant adopts an extensive α-helical conformation that stably propagates afterwards and encompasses the three key hydrophobic residues ( W94 , L97 and F98 ) that mediate coactivator contact . The substitutions distinguishing cAD-like96 from cAD-like07 are four tryptophan residues ( Table 2; W93 , 95 , 96 and 99; Fig 9B ) . Tryptophan is the strongest known helix conformer in short helices [43] and therefore the extensive helicity in cAD-like96 observed in the aMD simulations is in excellent agreement with expectations . The detected differences in secondary structure content and stability between cAD-like07 and cAD-like96 strongly suggest that pronounced α-helical propensity constitutes a key factor in determining the transactivation potential of an AD , even in absence of additional conformational changes induced by binding to the coactivator surface . We tested this concept further by investigating whether there was a correlation between α-helical propensity predicted by standard bioinformatics tools and the observed effectiveness in mediating transactivation in vivo . A plot of predicted α-helical propensity [44] of 24 different cAD-like variants [42] against experimentally measured transcriptional simulation provides previously undocumented evidence for a strong correlation between these two variables ( Fig 9C ) . The results show that this approach allows a direct prediction of transactivation potentials of cAD-like variants with 95% confidence using only primary amino acid sequence information . The extensive , stable α-helicity , combined with the presence of additional bulky hydrophobic residues next and between residues W94 , L97 and F98 raises some intriguing questions regarding the interaction of cAD-like96 with the GAL11-ABD1 coactivator module . As there is no structural data available for this system , we created a starting structure by in silico substitutions of the orthologous cAD residues in the GAL11-ABD1/GCN4-cAD NMR model ( PDB#2LPB-model 1 ) . Subsequently , four independent aMD simulations were carried out using the conditions described earlier . Just as expected from the results of the simulations of cAD-like96 on its own ( Fig 9B ) , the cAD-like96 adopts a continuously stable α-helical conformation that includes positions W94 , L97 and F98 throughout all four aMD simulations ( Fig 9D ) . Because the cAD motif is surrounded by several additional tryptophan residues , Warfield et al . suggested that these tryptophans might be able to occupy the pocket in a similar manner to the original cAD motif key residues and contribute to increased binding efficiency [42] . A molecular mechanics decomposition of the van der Waals forces of the aMD trajectories of GAL11-ABD1/cAD-like96 ( Fig 10 ) reveals interesting similarities and differences to the previously shown GAL11-ABD1/GCN4-cAD results ( Fig 6A ) . First , the main ΔGBinding contributions are once again centered on two regions ( W94 and L97/F98 ) , in addition to N-terminal contacts ( L84 , P87 , L89 ) that provide fleeting contributions reminiscent of the pattern found for the GCN4-cAD ( Fig 6A and 6B; note that these additional contacts , compared to GCN4-cAD [Fig 8A] , are not in an α-helical conformation [Fig 9D] ) . It is noticeable , however , that the main contributors in cAD-like96 play a less distinct , broader role; in GAL11-ABD1/cAD-like96_aMD_no1 and no4 , L97 makes a major contribution , but is distinctly supported by the flanking residues W96 and F98 . Such a more diffuse energetic contribution is also observable near the W94 position . In GAL11-ABD1/cAD-like96_aMD_no3 , W95 makes the dominant van der Waals contribution instead of W94 ( a state that is briefly and reversibly explored in aMD2 at ~1 , 900 nanoseconds; Fig 10 ) . A situation where W94 and W95 simultaneously occupy the ABD-1 pocket is not observed . As the helix would have to be distorted for these two residues to gain access to the pocket , it is unlikely for this confirmation to occur . In GAL11-ABD1/cAD-like96_aMD_no4 , both W93 and W94 contribute apparently equally and create a stable configuration that remains essentially unchanged throughout one microsecond of aMD simulation conditions . After identification of the possible binding states , angular measurements of the helical domain of cAD-like 96 and ABD1 α-helix 4 were performed to analyse the relative orientations of these structures relative to each other . The measurements show that the main orientations observed for the cAD-like 96 helix range typically between ~60° and ~120° ( Fig 11 ) . In comparison , the GCN4-cAD helix adopts a significantly wider range of orientations ( Fig 4 ) . Consequently , even though rotations are observable for both GAL11-ABD1/GCN4-cAD and GAL11-ABD1/cAD-like96 simulations , the maximal rotation performed by the helical cAD-like 96 domain is only 60° compared to ~180° observed for the GCN4-cAD helix . The orientations also last significantly longer and do not follow the frequent and abrupt changes observed for GCN4-cAD . We conclude that overall the binding of cAD-like96 to GAL11-ABD1 is conformationally significantly more restricted and therefore reduced in "fuzziness" . The increased degree of α-helicity , redundancy of hydrophobic contacts and reduced conformational freedom documented in the aMD simulations provide a quantitative base for understanding the high transactivation potential displayed by cAD-like96 . The molecular mechanisms that GSTFs employ to regulate the expression of their genes are still poorly understood . There is some evidence that the binding of activation domains to basal transcription factors and coactivator complexes induces major conformational changes that could allosterically transmit signals to other components of the transcriptional machinery [56–58] . This hypothesis suggests that transient interactions of activation domains with their targets could trigger the transition between long-lived alternative coactivator conformations . Such mechanisms are , however , exceedingly difficult to study using biochemical or computational tools . In the study reported here , we have found no evidence for any significant conformational change induced in the GAL11-ABD1 structure as a direct consequence of GCN4-AD binding . Even if such changes were observed , it would still be unclear whether ( and how ) such an alternative conformation could be allosterically transmitted to the remainder of the GAL11 subunit ( and beyond ) because currently our structural knowledge of GAL11 is restricted to the ABD1 domain . An alternative—and not necessarily conflicting—view is that ADs exert most of their functions through stabilizing the assembly or position of other functional components of the transcriptional machinery , such as the basal transcriptional initiation complex . Eukaryotic promoters are potentially regulated through dozens of GSTFs bound at nearby enhancer modules , so that a multitude of energetically weak , short-lived interactions between ADs and a variety of targets could provide a significant stabilization effect through synergistic action . The short interaction half-lives and multi-target specificity of the structurally disordered ADs may under such circumstances provide the flexibility to respond to rapidly changing regulatory requirements , or provide the possibility of some components , such as RNA polymerases to "break free" of these complexes after transcription initiation . Our work documents a positive correlation between α-helicity and transactivation potential , suggesting that the overall effectiveness of AD-binding to their targets can be directly controlled through changes in α-helical propensity during evolution . Such changes may , however , have to be counterbalanced with a need for a degree of intrinsic structural disorder to sustain the ability of ADs to interact with multiple target sites . Modelling approaches , including additional AD-coactivator targets , or studying the effect of AD-interactions with larger complexes , offer great opportunities to gain further insights into the dynamical processes of coactivator-activator interactions and open numerous theoretical and applied avenues for the future . Such strategies will most likely be part of synthetic biology approaches that aim at designing artificial transcription factors with a precisely controlled range of specificity and transactivation potential in eukaryotic systems . All structures were prepared to the same specifications to maximize comparability between the simulations . For the ABD1/cAD complex simulation both polypeptide chains of Model 1 of the GCN4-GAL11 complex ( PDB#2LPB ) were capped ( acetyl and N-methylamide groups added to the N- and C-termini , respectively ) using Yasara Structure [59] . For the ABD1/cAD-like96 simulation the GCN4-cAD structure ( PDB 2LPB-Model 1 ) was mutagenized in silico with Yasara Structure [59] to create the cAD-like96 sequence . The coordinates were prepared for simulation in LEaP ( AmberTools 14/15 ) with the Amber 14SB forcefield [60] , neutralized and solvated in a TIP3P [61] solvent box with a minimum distance of 15 Å between solute and border . The final ionic concentration within the water box was adjusted to a final concentration of 150 mM NaCl . Capped structures of the GCN4-cAD , GCN4 cAD-like07 and cAD-like96 were built de novo from their primary amino acid in LEaP , and prefolded using 10 ns of GB implicit MD before solvating them under the same conditions described above . The solvated models were minimized , heated to 300K and relaxed before performing a conventional MD ( cMD ) production run for 100 ns at a target pressure of one atmosphere to obtain values for the total potential and dihedral energy values ( NPT ) . Simulations were carried out using the pmemd . cuda ( Amber 14 ) applying the hybrid single/double/fixed precision model ( SPFP ) GPU support [62 , 63] using 2 fs time steps with a 10 Å cut off under control of a Langevin thermostat [64] and the SHAKE algorithm to restrain hydrogens [65] . Long-range electrostatic interactions were calculated using the Particle Mesh Ewald approximation [66] . The average total potential energies and the average dihedral energies were obtained from the cMD simulations and utilised to calculate the thresholds for dual boost aMD using an α-value of 0 . 2 . All aMD simulations were performed with a target temperature of 300 Kelvin , and a target pressure of one atmosphere ( 101 . 325 kPa ) . Temperature was controlled by the Andersen temperature-coupling scheme and the pressure was controlled by the isotropic position scaling protocol applied in AMBER . Four independent 1000 ns aMD simulations were run for each structure . Details of simulations performed are summarized in Table 1 . The structural models and trajectory data are available as supporting data ( S1 and S2 Data Sets ) Mapping of interaction hotspots was performed using the FTMAP algorithm ( http://ftmap . bu . edu [67] ) . Trajectory visualisation , secondary structure analysis ( based on STRIDE; [68] , imaging and file conversion was performed with VMD v . 1 . 9 . 2 [69] . CPPTRAJ from AmberTools 15 was utilised for distance and angle measurement [70] . Bio3D was implemented for RMSD , RMSF and principal component analysis [71 , 72] . Visualisation of the analytical data was performed with CRAN [73] . The MM-GBSA estimation of binding free energies was performed employing the Amber forcefield ff99 [74] using the MMPBSA . py script [75] . Residue-specific decomposition was based on adding the 1–4 non-bonded interaction energies ( 1–4 EEL and 1–4 VDW ) to the internal potential terms . The cAD sequences were acetylated at the N-terminus and amidated at the C-terminus before predicting their α-helical properties at the residue level at pH 7 . 0 , 150 mM NaCl and 300K [44] .	The regulated transcription of eukaryotic genes is governed by gene-specific transcription factors that contain activation domains to stimulate the expression of nearby genes . Activation domains are unable to take up a defined three-dimensional conformation . Nevertheless , as we demonstrate in our study , molecular dynamics simulations reveal that the key docking point of such domains ( centered around several large hydrophobic amino acid sidechains ) folds into fluctuating α-helical conformations . Analysis of published data shows that this tendency of adopting such local structures correlates directly with stimulation activity . We also investigate the interaction of these structurally unstable domains with a coactivator interaction partner . Computational simulations are ideally suited for analysing the rapidly changing , "fuzzy" interactions occurring between these protein partners . We gained new insights into the competitive nature of the key hydrophobic sidechains in binding to a pocket on the coactivator surface and documented for the first time the rapidly changing movements of an activation domain during these interactions .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "sequencing", "techniques", "molecular", "dynamics", "dna", "transcription", "simulation", "and", "modeling", "protein", "sequencing", "sequence", "motif", "analysis", "transactivation", "molecular", "biology", "techniques", "research", "and", "analysis", "methods", "sequence", "analysis", "gene", "expression", "chemistry", "biophysics", "molecular", "biology", "physics", "biochemistry", "biochemical", "simulations", "genetics", "biology", "and", "life", "sciences", "physical", "sciences", "computational", "chemistry", "computational", "biology", "biophysical", "simulations" ]	2016	Molecular Dynamics of "Fuzzy" Transcriptional Activator-Coactivator Interactions
Alcelaphine herpesvirus 1 ( AlHV-1 ) is a γ-herpesvirus ( γ-HV ) belonging to the macavirus genus that persistently infects its natural host , the wildebeest , without inducing any clinical sign . However , cross-transmission to other ruminant species causes a deadly lymphoproliferative disease named malignant catarrhal fever ( MCF ) . AlHV-1 ORF73 encodes the latency-associated nuclear antigen ( LANA ) -homolog protein ( aLANA ) . Recently , aLANA has been shown to be essential for viral persistence in vivo and induction of MCF , suggesting that aLANA shares key properties of other γ-HV genome maintenance proteins . Here we have investigated the evasion of the immune response by aLANA . We found that a glycin/glutamate ( GE ) -rich repeat domain was sufficient to inhibit in cis the presentation of an epitope linked to aLANA . Although antigen presentation in absence of GE was dependent upon proteasomal degradation of aLANA , a lack of GE did not affect protein turnover . However , protein self-synthesis de novo was downregulated by aLANA GE , a mechanism directly associated with reduced antigen presentation in vitro . Importantly , codon-modification of aLANA GE resulted in increased antigen presentation in vitro and enhanced induction of antigen-specific CD8+ T cell responses in vivo , indicating that mRNA constraints in GE rather than peptidic sequence are responsible for cis-limitation of antigen presentation . Nonetheless , GE-mediated limitation of antigen presentation in cis of aLANA was dispensable during MCF as rabbits developed the disease after virus infection irrespective of the expression of full-length or GE-deficient aLANA . Altogether , we provide evidence that inhibition in cis of protein synthesis through GE is likely involved in long-term immune evasion of AlHV-1 latent persistence in the wildebeest natural host , but dispensable in MCF pathogenesis . Infection with the members of the Herpesviridae family is characterized with lifelong persistence of the viral genomes in the infected hosts . Such latent infection is only possible through the establishment of complex immune evasion mechanisms . Latent infection by γ-herpesviruses ( γ-HV ) is mostly asymptomatic and mainly occurs in lymphoid cells . However , γ-HV latency can induce lymphoproliferative diseases and cancers . As such , Epstein-Barr virus ( EBV ) infection of immunocompromised human individuals has been associated with nasopharyngeal carcinoma , Burkitt’s and Hodgkin’s lymphomas and Kaposi’s sarcoma associated-herpesvirus ( KSHV ) has been associated with primary effusion lymphomas , Castleman’s disease and Kaposi’s sarcoma [1–3] . In addition , lymphoproliferative diseases are caused by other γ-HVs in specific cases of cross-species transmission [4 , 5] . The development of such malignancies has been related to γ-HV latent infection . Latent infection of host cells by many γ-HVs is dependent upon the expression of a viral genome maintenance protein ( GMP ) , which ensures the persistence of the viral episome within actively dividing cells , yet simultaneously evades the immune surveillance . Interestingly , specific EBNA1 and LANA1 peptides have been shown to be presented by cross-priming and specific CD8+ T cells could be readily isolated from EBV- or KSHV-infected individuals [6–8] . In addition , recent reports investigating the immune evasion mechanisms by γ-HV GMPs suggest that latently infected cells evade the detection by host CD8+ cytotoxic T lymphocytes ( CTLs ) rather by a limitation of antigen presentation than an absence of T cell epitopes . Malignant catarrhal fever ( MCF ) is an acute , sporadic and fatal pan-systemic lymphoproliferative disease of a variety of species of the Artiodactyla order , including cattle . The main causative agents of MCF are two γ-HVs that are grouped in the macavirus genus , ovine herpesvirus 2 ( OvHV-2 ) and alcelaphine herpesvirus 1 ( AlHV-1 ) . These viruses cause no apparent disease in their natural host species . Sheep are naturally infected by OvHV-2 , which is responsible for the sporadic sheep-associated form of MCF [5] . Wildebeest are persistently infected with AlHV-1 , the causative agent of the wildebeest-derived form of the disease [9 , 10] . The prevalence of AlHV-1 infection in wildebeest is close to 100% and transmission to MCF-susceptible species mainly occurs during the calving period and in the first months of life [11 , 12] . MCF impact on the local pastoralist populations has largely been underestimated , with recent reports demonstrating that MCF is perceived to be the cattle disease with the highest economic and social impacts in these areas [13–16] . In addition , MCF has been reported throughout the world in game farms or zoological collections where mixed ruminant species including wildebeest are kept [17] . Recent data demonstrated that MCF is caused by the activation and proliferation of latently infected CD8+ T cells [18–20] and that the expression of the AlHV-1 genome maintenance protein , the latency-associated nuclear antigen ( LANA ) -homolog ( aLANA ) encoded by the ORF73 gene of AlHV-1 is essential for both viral persistence in infected hosts and induction of MCF [21] . The expression of aLANA is abundant in the tissues of MCF-developing animals and should potentially induce an anti-viral cytotoxic response . Although such anti-viral response might exist , it is however not protective . Many γ-HVs genome maintenance proteins have cis-acting immune evasion mechanisms mainly , but not only , regulated by their large acidic repeat domains [22–29] . A common feature of γ-HV GMPs is their evasion of specific CTLs through cis-acting but diverse mechanisms [30–34] . Here we have investigated the immune evasion properties of aLANA . We showed that a glycin/glutamate ( GE ) -rich repeat domain in aLANA was necessary and sufficient to inhibit the presentation by MHC-I of an epitope linked to it . We further found that aLANA GE downregulated protein self-synthesis and this mechanism could be associated with reduced antigen presentation in vitro . Importantly , codon modification of the purine bias in GE resulted in heightened antigen presentation of an epitope linked to aLANA and heightened induction of antigen-specific CD8+ T cell responses . These results suggested that inhibition in cis of antigen presentation is controlled by mRNA constraints in GE and reduced protein translation efficiency . Nonetheless , infection of rabbits with an AlHV-1 recombinant strain expressing a GE-deficient aLANA protein did not affect MCF induction , suggesting a dispensable role of cis-acting immune evasion by aLANA in the pathogenesis of MCF . These findings enlarge our understanding of MCF pathogenesis and suggest that cis-acting immune evasion by aLANA is likely involved in long-term viral persistence in latently infected wildebeests . Sequencing of the AlHV-1 genome revealed that its ORF73 gene encodes a 1300 amino acid-long protein [35] , which makes AlHV-1 LANA-homolog protein ( aLANA ) the longest GMP ortholog described to date among rhadinoviruses and macaviruses . Analysis of the primary structure of aLANA revealed the presence of a long central repeat ( CR ) region that could be divided in two main subregions based on the abundance of glycine , proline and glutamate ( GPE-rich region ) or glycine and glutamate residues ( GE-rich region ) . An additional region containing glutamate repeats is enclosed within the GE-rich region , and was termed the E-rich region ( Fig 1A ) . We hypothesized that aLANA possesses mechanisms inhibiting antigen presentation in cis similar to other γ-HV GMPs and therefore renders AlHV-1 latently infected cells difficult to detect by cytotoxic T cells . We used the SIINFEKL peptide of chicken ovalbumin as cognate epitope to be linked in frame to eGFP and transient expression in 293Kb cells for detection of H-2Kb-peptide complexes [36] . In this model , 293Kb cells constitutively express the murine H-2Kb MHC class I haplotype . 293Db cells were used as controls to determine the specificity of the staining ( S1A Fig ) . In order to determine the capability of aLANA to inhibit antigen presentation in cis , we introduced AlHV-1 ORF73 coding sequence into the peGFP-SIIN plasmid to generate vectors expressing aLANA-SIIN and SIIN-aLANA proteins ( Fig 1B ) . 293Kb cells transiently expressing aLANA-SIIN or SIIN-LANA expressed very low amounts of MHC-epitope complexes at the cell surface and were poorly recognized by SIINFEKL-specific B3Z hybridoma cells ( Fig 1C–1F ) . Similar results were also obtained using L929Kb and VeroKb cell lines ( S1B and S1C Fig ) . These results suggested poor antigen processing . There was no evidence for aLANA inhibiting peptide presentation from co-transfected eGFP-SIIN ( Fig 1G and 1H ) . These results suggested an immune evasion mechanism by aLANA acting in cis and not in trans . To identify possible contributions of the CR region in aLANA reduced antigen presentation , we constructed a mutant form of aLANA from which the entire CR domain was removed . The resulting ΔCR aLANA mutant protein was inserted into the peGFP-SIIN expression vector in order to express the ΔCR-SIIN protein ( Fig 2A ) . The expression of MHC-epitope complexes at the cell surface of 293Kb cells transiently expressing ΔCR-SIIN was rescued to levels similar to the SIIN positive control ( Fig 2B and 2C ) . ΔCR-SIIN expression was restricted to the nuclei like the full-length form of aLANA ( S2 Fig ) . Also , 293Kb cells expressing ΔCR-SIIN or SIIN induced similar levels of β-galactosidase activity in B3Z ( Fig 2D ) . These results suggested that antigen processing of aLANA is impaired by the presence of the CR region . Next , we investigated potential contributions of the subregions of the CR domain in limiting antigen presentation in cis . First , we produced two forms of the aLANA protein lacking either the GPE domain ( ΔGPE-SIIN ) or the purine-enriched GE domain ( ΔGE-SIIN ) , respectively ( Fig 3A ) . 293Kb cells expressing ΔGPE-SIIN showed intranuclear localization of eGFP expression ( S2 Fig ) and strongly reduced antigen presentation at the cell surface as well as reduced B3Z activation ( Fig 3B–3D ) . ΔGE-SIIN expression in 293Kb cells was also intranuclear ( S2 Fig ) , but strikingly resulted in rescued peptide presentation and activation of B3Z cells to levels similar to the SIIN control ( Fig 3B–3D ) . Thus , the GE subregion of aLANA impaired antigen processing and presentation of a peptide linked to it . Supporting this conclusion , similar results could be observed in VeroKb cells ( S1C Fig ) and also using plasmid vectors in which protein expression is driven by an eukaryotic EF1-α promoter ( S1D Fig ) . Next , we investigated the role of the E-rich domain within the GE region of aLANA as well as both flanking subregions named GE1 and GE2 , respectively ( Fig 4A ) . We produced four additional constructs expressing different combinations of the GE subregions , namely ΔE- , ΔGE1- , ΔGE2- , and ΔEΔGE2-SIIN . Expression of these mutant forms of aLANA in 293Kb cells did not reach the levels of antigen presentation obtained with ΔGE-SIIN ( Fig 4B–4E ) , suggesting that the entire GE region is necessary for limiting antigen presentation in cis . To examine the possibility that the GE-rich domain is self-sufficient to inhibit the presentation of a linked epitope , we expressed a chimeric protein composed in N-terminus of the eGFP coding sequence fused with the GE domain and a C-terminal SIINFEKL epitope tag ( Fig 5A , GE-SIIN ) . To further investigate if the position of the SIINFEKL peptide could affect its presentation , another construct consisted in the insertion of an additional SIINFEKL epitope tag in between eGFP and GE ( Fig 5A , SIIN-GE-SIIN ) . Following transfection of 293Kb cells , expression of both GE-SIIN and SIIN-GE-SIIN could be detected in both cytoplasm and nucleus and antigen presentation was strongly limited by the presence of the GE domain ( Fig 5B and 5C ) . These results demonstrated that the GE domain is sufficient to significantly reduce antigen presentation . In order to decipher how GE inhibits cis-peptide presentation , we first examined the effect of proteasome inhibition on antigen presentation ( Fig 6A ) . 293Kb cells were transfected to express eGFP , SIIN , aLANA , ΔCR-SIIN , ΔGPE-SIIN or ΔGE-SIIN before treatment with the proteasome inhibitor MG132 and detection of H-2Kb-SIINFEKL complexes at the cell surface by flow cytometry . MG132 treatment resulted in a severe reduction of antigen presentation for all constructs . Although MG132 mainly targets the proteasome , it can also affect autophagy pathways [37] . Thus , treatments using more specific lactacystin or epoxomicin were performed and resulted in significantly reduced detection of the H-2Kb-SIINFEKL complexes ( Fig 6B and S3 Fig ) . However , treatments with rapamycin , chloroquine or 3 methyladenine ( 3-MA ) did not affect antigen presentation . These results confirmed that GE-dependent inhibition of antigen presentation is dependent upon proteasomal degradation , whereas activation or inhibition of autophagy pathways did not affect peptide presentation . Inhibition of γ-HV GMPs epitope presentation by their respective central repeats has been attributed to reduced protein synthesis and reduced protein degradation [22] . To analyze the implication of the CR and GE regions in protein turnover , we first used a cycloheximide ( CHX ) /chase experiment at 24h after transfection ( Fig 6C ) and observed that the levels of proteins up to 30h after CHX treatment were not significantly affected in absence of CR or GE . In an alternative approach , we used the HaloTag technology to generate fusion proteins of aLANA , ΔCR and ΔGE . 293Kb cells were transfected with aLANA- , ΔCR- or ΔGE-HaloTag constructs and pulse-labeled overnight with the HaloTag TMR-Direct ligand . The cells were then washed and chased for 72h and fluorescence intensities were recorded over time ( Fig 6D ) . All constructs showed similar turnover suggesting that the CR or GE regions are not involved in reduced protein degradation . Steady-state levels of protein expression were first investigated using immunoblotting of total cell lysates 48h after transfection with plasmids encoding aLANA- , ΔCR- , ΔGPE- or ΔGE-SIIN ( Fig 7A ) . We observed that the absence of both CR and GE regions resulted in increased protein expression levels . In addition , we observed increased eGFP fluorescence intensities over time in absence of GE and at similar transfection efficiencies ( Fig 7B ) . Thus , GE might regulate the level of protein expression . A pulse-chase labeling experiment was conducted to further address this hypothesis ( Fig 7C ) . At 12h post-transfection of aLANA- , or ΔGE-HaloTag constructs , cells were pulse-labeled overnight with the HaloTag TMR-Direct ligand to label all transfected cells and chased for a further 24h . Then , de novo synthesized proteins were labeled with the HaloTag Oregon-Green ( OG ) ligand . We observed increased proportions of OG-positive cells over time in cells expressing ΔGE-HaloTag , suggesting that GE regulates the efficiency of protein synthesis . Next , we sought to directly investigate translation efficiency of aLANA- , ΔCR- , ΔGPE- or ΔGE-SIIN using an uncoupled in vitro translation assay ( Fig 7D ) . Molar equivalents of capped RNA obtained from in vitro transcription using T7 RNA polymerase were translated in vitro using a rabbit reticulocyte lysate system . Although multiple bands could be observed for some constructs ( S4B Fig ) , analysis of the intensity of the bands at the expected sizes ( S4A Fig ) suggested lower translation levels of aLANA and ΔGPE , while ΔCR and ΔGE showed higher translation efficiencies . Translation of SIIN was used as positive control and the negative control consisted of no RNA . These results suggested that the GE domain downregulates aLANA translation efficiency . In addition , immunoblotting of total cell lysates after transfection with SIIN , aLANA-SIIN or GE-SIIN expression vectors demonstrated that the GE domain was self-sufficient to significantly decrease protein steady-state levels after transfection of 293Kb cells ( Fig 7E ) . To further investigate whether the GE domain could also regulate transcription , we quantified steady-state RNA expression levels in 293Kb cells transfected with aLANA- , ΔCR- , ΔGPE- or ΔGE-SIIN . Interestingly , aLANA and ΔGPE transfected cells had similar low levels of specific transcripts whereas the absence of the CR or GE domains in aLANA resulted in significantly enhanced transcription levels ( Fig 8A ) . Such increase of RNA expression was not due to decreased turnover of ORF73 mRNA as aLANA , ΔCR and ΔGE had similar RNA abundance after actinomycin D treatment following transfection ( Fig 8B ) . The limited presentation efficiency of CD8+ T cell epitopes from EBNA1 has been suggested to be primarily determined by low translation efficiency rather than its intracellular stability [23 , 29] . These observations were in line with the main hypothesis according to which defective ribosomal products ( DRiPs ) generated during protein synthesis rather than mature proteins are the major source of antigens that are processed to MHC class I [38–40] . The absence of the GE domain resulted in increased steady-state levels of aLANA protein expression that was related to increased translation efficiency and increased steady-state RNA levels ( Figs 7 and 8 ) , observations that could be associated with increased presentation of an endogenous epitope ( Fig 4 ) . To explore this possibility , we treated transfected 293Kb cells with citrate buffer to strip MHC-peptide from the cell surface , followed by incubation with CHX to inhibit de novo protein synthesis ( Fig 9A ) . Combined treatments significantly reduced the proportion of cells expressing the MHC-peptide complexes at the cell surface irrespective of the presence or absence of repeat regions ( Fig 9B ) . In contrast , CHX treatment alone had only a minimal effect on antigen presentation . Finally , citrate buffer treatment alone showed a differential effect on 293Kb expressing either aLANA or ΔCR and ΔGE , suggesting a more rapid antigen processing by MHC-I in absence of CR and GE . These results revealed an association between translation efficiency ( and production of DRiPs ) and antigen presentation . Purine-rich GAr mRNA structure of EBNA1 regulates EBNA1 synthesis and presentation of EBNA1 to specific T cells [41–43] . Analysis of the nucleotidic sequence of the aLANA GE domain showed a bias towards usage of codons containing purines , with 100% of glutamate residues and 81% of glycine residues being purine codons ( GAA , GAG and GGA ) ( S5A and S5B Fig ) [27] . Prediction of mRNA secondary structure using Mfold revealed unstable secondary structures ( S5C Fig ) . Thus , a codon-modified GE sequence ( GEm ) was synthesized in order to reduce purine bias , while conserving the identical peptidic sequence ( S5A Fig ) . Modification of the codon usage resulted in increased stabilization of mRNA secondary structure reflected by the significantly more negative Gibbs free energy value ( δG ) of -153 . 91 kcal/mol compared with -43 . 39 kcal/mol for the native GE form ( S5C Fig ) . We hypothesized that purine bias in aLANA GE mRNA sequence might be responsible for reduced antigen presentation . Pairwise mRNA sequence alignment of aLANA or aLANA-GEm with EBV EBNA1 demonstrated that the high sequence homology of aLANA GE with EBNA1 GAr sequence was lost after codon modification in GE ( S5D Fig ) [27] . Synthetic GEm sequence was used to replace the native GE in aLANA and generate the codon-modified aLANA-GEm ( Fig 10A ) . Transfection of 293Kb cells aLANA-GEm-SIIN resulted in nuclear expression and significantly enhanced proportion of H2Kb-SIINFEKL expressing cells compared to native aLANA-SIIN ( Fig 10B and 10C ) . Although the modification of the codon usage to reduce purine bias in GEm did not result in antigen presentation as high as observed in absence of GE ( ΔGE-SIIN ) , our results suggest that constraints in native GE mRNA structure can limit antigen presentation . We further addressed this hypothesis by immunizing C57BL/6J mice using a DNA immunization protocol [21] . DNA immunization with GEm-SIIN expressing plasmid resulted in significantly increased SIIN-specific CD8+ T cells in peripheral blood compared to native aLANA-SIIN construct ( Fig 10D–10G ) . These results demonstrated that mRNA constraints in GE limit antigen presentation and T cell priming . We previously showed that MCF is caused by the lymphoproliferation of latently infected cells expressing high levels of aLANA [21] . Here , we observed that the absence of GE in aLANA resulted in increased presentation of an epitope linked to the protein ( Fig 3 ) . Based on previous reports showing that CTLs specific to EBNA1 or LANA1 could be detected in infected patients [6–8] , that the EBNA1 GAr domain was dispensable for episome maintenance and immortalization of B cells in vitro [23] , and that mLANA immune evasion of CTLs was essential for viral persistence in the host [34] , we hypothesized that the absence of GE in aLANA might result in an enhanced CTL priming and impaired evasion of aLANA-specific CTLs . To address this hypothesis , we generated a recombinant virus expressing a GE-deleted ( ΔGE ) form of aLANA and its revertant ( ΔGE-Rev ) ( S6A and S6B Fig ) . Viral growth in BT fibroblasts was not affected by the absence of aLANA GE domain ( S6C Fig ) . This was expected as viruses impaired for aLANA expression had no growth defect [21] . We then infected rabbits intranasally with the WT , ΔGE or ΔGE-Rev strains ( 105 PFU/rabbit ) . Rabbits infected with the ΔGE virus developed hyperthermia , splenomegaly , lymphadenopathy and expansion of CD8+ T cells similar to the WT and ΔGE-Rev control groups ( Fig 11 ) . Lymphoblastoid cell lines could also be propagated from both ΔGE and ΔGE-Rev-infected animals ( S7 Fig ) , suggesting that a truncated form of aLANA lacking its GE region is able to maintain viral episomes . Together , these results suggest that aLANA GE-mediated inhibition of antigen presentation in cis is dispensable in MCF pathogenesis . Wildebeest infection by AlHV-1 is persistent and nonpathogenic . Such an adaptation of the virus with its host species likely results from a long evolutionary relationship . Understanding the mechanisms acquired during coevolution by AlHV-1 to achieve such an adaptation in wildebeest and evade host immunity is important to develop strategies to target AlHV-1 latency in particular and γ-HV latent infection in general . In addition , AlHV-1 episomal persistence has been shown to be essential in MCF pathogenesis in susceptible species [21] . We have previously shown that a lack of aLANA expression rendered AlHV-1 unable to induce MCF , which positioned latency at the core of MCF pathogenesis . In this context , immune evasion by aLANA during MCF is likely of importance in the pathogenesis of this deadly disease . In the present study , we have brought evidence that aLANA can strongly limit antigen presentation and induction of antigen-specific cytotoxic T cell responses through a mechanism in cis depending on its GE-rich domain . However , AlHV-1 infection of MCF-susceptible rabbits induced MCF irrespective of the presence or absence of GE in aLANA . We used transient expression of chimeric aLANA fusion proteins in a model of MHC class I peptide presentation in order to test the ability of AlHV-1 latency protein to evade immune detection . We observed a strongly reduced presentation of an epitope linked to aLANA . However , aLANA did not reduce antigen presentation in trans in co-transfection experiments . These results demonstrated that aLANA is not efficiently processed to MHC class I . Most viral maintenance proteins contain acidic repeat domains , such as for example the GAr domain of EBNA1 , that have been involved in inhibition of MHC-I antigen presentation in cis [22] . aLANA has a long CR domain and antigen presentation could be restored after deletion of the entire CR sequence . We further investigated the role of subdomains within the CR region and identified a GE-rich region that was essential to inhibit the presentation of a linked peptide . Although subregions could be identified within the GE region , such as the GE1 , E-rich and GE2 domains , only a protein lacking the GE-rich domain as a whole could restore antigen presentation and CTL activation to the levels of control chimeric eGFP-SIIN protein . In addition , we further observed that the transfer of GE to a heterologous eGFP-SIIN protein strongly inhibited antigen presentation . These results were important as they suggest that aLANA antigenic peptides are not effectively processed to MHC-I and identified the GE region as being sufficient to mediate immune evasion . We observed that GE inhibits proteasome-dependent antigen presentation . Increased antigen presentation with ΔCR and ΔGE mutant proteins could have potentially been explained by increased susceptibility to proteasomal degradation . However , it was not the case as a lack of CR or GE did not result in increased protein turnover in pulse-chase experiments . Although the internal repeat region of several GMPs , including LANA1 , EBNA1 , and mLANA significantly decreased protein turnover [33 , 34 , 44] , the protein stability of saimiriine herpesvirus 2 ( SaHV-2 ) ORF73 protein product was not influenced by the CR domain [31] . In addition , the region of LANA1 CR domain that inhibits proteasomal degradation was not involved in the inhibition of antigen presentation [32] . These data were supported by another study revealing that the half-life of a polypeptide did not determine antigen presentation [45] , suggesting that protection from proteasomal degradation may not be sufficient to block antigen presentation . Beside increased susceptibility to proteasomal degradation , several studies have shown that translation efficiency of GMPs such as EBNA1 is an important mechanism to explain the limited cis-antigen presentation [23 , 24 , 27–29] . Such observations resulted from a number of studies on EBNA1 over the last decade that showed that in spite of the inhibitory effect of the internal GAr domain , CTL responses directed towards EBNA1 could be readily detected in EBV seropositive individuals [7 , 46 , 47] , and we could observe SIINFEKL-specific CTLs after DNA immunization with aLANA-SIIN ( Fig 10 ) . This paradox was resolved following extensive in vitro molecular analyses of the endogenous processing of EBNA1 indicating that CTL epitopes from this protein were predominantly generated from newly synthesized DRiPs rather than from the long-lived pool of stable EBNA1 in EBV-infected B cells [23 , 26–28 , 41 , 47] . These observations have subsequently been further extended to demonstrate that the generation of DRiPs is intrinsically linked to the rate at which proteins are synthesized [23] . In the present study , we observed that the GE domain could reduce the efficiency of protein synthesis of aLANA , an observation that could be explained by a combination of increased translation efficiency ( using uncoupled in vitro translation ) as well as increased steady-state RNA expression levels when aLANA is deleted of GE . These results suggested that the GE-rich domain inhibits antigen presentation through regulation of both protein translation and RNA transcription levels , leading to a reduced production of DRiPs . DRiPs have been suggested to be the main source of viral antigens processed to MHC-I [48] . We therefore further hypothesized that targeting the mechanism leading to DRiPs production could be valuable for aLANA to ensure episome persistence during latency while avoiding detection from the immune system . Using citrate buffer and CHX co-treatment to block de novo protein synthesis after MHC-peptide stripping , we observed a significant reduction of detectable MHC-peptide complexes at the cell surface irrespective of the presence or absence of repeat regions . Thus , we can suggest that endogenous processing of CTL epitopes fused to the aLANA-ΔGE protein is not determined by its intracellular stability but rather by the rate at which newly synthesized polypeptides are produced . Comparison of γ-HVs maintenance protein mRNA sequences revealed the presence of highly homologous purine-rich central repeat sequences [27] . Although highly conserved in mRNA sequence , these repeat regions encoded very different peptide sequences in the different viruses . In EBNA1 , frame-shifting of the repetitive region without modifying the functional N- and C-terminal region did not significantly modify the reduced self-synthesis and the associated limitation in cis of antigen presentation . The purine-rich mRNA sequence would therefore be responsible for the reduced expression of these viral mRNAs rather than its encoded protein sequence . Interestingly , whereas the GE region of aLANA shares between 29 . 5 and 50% peptidic sequence homology with EBNA1 GAr region , the nucleotidic mRNA sequence corresponding to the GE region has more than 70% sequence homology with GAr and aLANA GE contains 96 . 6% purines . Codon modification of EBNA1 GAr sequence in order to reduce the purine-bias led to increased protein translation and improved CTL priming [26 , 28] . We hypothesized that reducing the purine bias in the nucleotidic sequence of the GE region in aLANA would result in a better presentation of the epitope linked to it . Whereas both native aLANA and aLANA-GEm encode exactly the same protein , codon modification in GEm mRNA resulted in a significant increase of antigen presentation . Although the increased antigen presentation observed in vitro did not reach the levels obtained with a construct lacking GE , immunization of C57BL/6 mice with aLANA-GEm lead to the activation of significantly more antigen-specific CTLs ( Fig 10 ) . Inhibitory G-quadruplex structures were identified within EBNA1 nucleotidic sequence and are suggested to act as steric blocks to cause a stalling/dissociation of ribosomes , therefore reducing translation efficiency . Although G-quadruplex structures could also be predicted within aLANA central repeat [43] , it is possible that the modification of the codon usage within GE was not sufficient to completely disrupt G-quadruplexes function for effective antigen presentation in vitro but sufficient to significantly increase cross-priming of specific CTLs in vivo . Further analyses are required to determine the presence of such structures in GE and their potential role in the cis-acting immune evasion of aLANA . A rapid priming of an effective adaptive cellular immune response can determine the outcome of primary infection and also the control of persistent viral infections [49] . Infection of rabbits with a virus strain expressing a truncated form of aLANA lacking GE did not impair MCF induction . We previously showed that aLANA expression was essential for viral persistence and episomal maintenance during MCF . The present results suggest that the absence of GE did not impair episomal maintenance , which was further supported by the effective propagation of lymphoblastoid cells from rabbits infected by the ΔGE and ΔGE-Rev virus . The absence of GE did not render the protein unstable in vitro ( Fig 7 ) . We could therefore expect enhanced antigen-presentation by infected cells after infection with the ΔGE virus . Even though such enhanced anti-aLANA specific response could exist , it is not sufficient to clear proliferating infected cells and block the development of MCF lesions . A possible explanation would be the expression during MCF of viral proteins involved in the inhibition of antigen presentation by MHC-I in trans , though such proteins are yet to be identified in AlHV-1 . This observation is important in our understanding of other lymphoproliferative diseases induced by γ-HV latent infection , where evasion mechanisms in trans could also be involved . However , our lack of knowledge of the proteins involved in such mechanisms in AlHV-1 infection renders this hypothesis currently difficult to address . Even though a trans-acting mechanism is possible , we have however previously shown that apart from aLANA-expressing ORF73 , very scarce gene expression could be detected during MCF [19] , which could suggest alternative immune evasion mechanisms developed by aLANA . Interestingly , EBNA1 was shown to be protected from proteasomal degradation through a mechanism depending on its high avidity for cellular DNA through the bipartite Gly-Arg repeat domains [50] . Potentially , similar domains outside GE could similarly protect aLANA against proteasomal degradation and therefore enable immune evasion in vivo . In addition , during KSHV and EBV lytic replication , alternative cytoplasmic isoforms of LANA1 and EBNA1 are expressed [51 , 52] . Interestingly , a LANA1 isoform was shown to antagonize the innate immune DNA sensor cGAS [53] . Whether AlHV-1 also encodes such isoforms of aLANA and whether they can inhibit innate immunity is unknown . Nonetheless , their expression during lytic replication could significantly impair effective priming of aLANA-specific CD8+ T cells , thus potentially explain how a ΔGE virus is able to persist in an immunocompetent host and induce MCF . In this study , we provide evidence that aLANA has evolved mechanisms to evade CTLs by inhibiting its own presentation in MHC-I through its central repeat purine-rich GE region that self-regulates protein synthesis and antigen processing to MHC class I . Importantly , our findings identify a key mechanism to evade immune surveillance by CTLs during AlHV-1 latency . The experiments , maintenance and care of mice and rabbits complied with the guidelines of the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes ( CETS n° 123 ) . The protocol was approved by the Committee on the Ethics of Animal Experiments of the University of Liège , Belgium ( Permits #1127 , 1571 and 1598 ) . All efforts were made to minimize suffering . 293A cells stably expressing murine class I alleles H-2Kb or H-2Db ( referred to as 293Kb and 293Db ) were kindly provided by Dr Jonathan Yewdell ( NIH/NIAID , USA ) [36] . B3Z cells , an H2-Kb-restricted CD8+ T-cell hybridoma , specific for the SIINFEKL ovalbumin peptide ( OVA254-264/Kb ) and L929-Kb cells were obtained from Dr P . G . Stevenson ( University of Queensland , Australia ) . Bovine turbinate fibroblasts ( BT , ATCC Crl-1390 ) , bovine mammary epithelial cells stably expressing the NLS-Cre ( MacT-Cre , ATCC CRL-10274 ) [54] , Vero ( ATCC CCL-81 ) and L929 ( ATCC CCL-1 ) cells were cultured in Dulbecco’s modified essential medium ( DMEM , Sigma ) containing 10% FCS . Madin Darby bovine kidney cells ( MDBK , ATCC CCL-22 ) were cultured in minimum essential medium ( MEM , Sigma ) supplemented with 10% FCS . The pathogenic AlHV-1 C500 strain isolated from an ox with MCF ( initially obtained from Prof . D . Haig , University of Nottingham ) and the AlHV-1 WT BAC clone were used in this study [18 , 55] . Viruses were maintained by a limited number of passages ( <5 ) . Virus was amplified in BT cells before supernatants together with infected cells were concentrated by ultracentrifugation ( 100 , 000×g , 2h , 4°C ) and suspended in PBS before storage at -80°C . Samples were thawed , clarified ( 100×g , 20 min , 4°C ) and supernatants titrated by plaque assay as described previously [19] . The peGFP-C1 ( Clontech ) expression vector was used to clone the chicken ovalbumin epitope SIINFEKL coding sequence into KpnI/BamHI sites using annealed oligonucleotides KpnI/BamHI-SIIN-Fwd and KpnI/BamHI-SIIN-Rev ( S1 Table ) and T4 DNA ligase ( Roche ) to generate the peGFP-SIIN vector . AlHV-1 ORF73 coding sequence was amplified by PCR using primers HindIII-73N-Fwd and KpnI-73CΔstop-Rev ( S1 Table ) and AlHV-1 BAC DNA as template [18] . The obtained amplicon was then cloned into pGEM-T Easy vector ( Promega ) and subcloned by ligation into peGFP-SIIN to generate the peGFP-aLANA-SIIN plasmid . Subsequent plasmid constructs , including those containing the ORF73 coding sequence deleted of different subregions were generated either by ligation ( T4 DNA ligase , Promega ) or by homologous recombination using the In-Fusion HD cloning kit ( Clontech ) and primers listed in S1 Table . Plasmid vector peGFP-SIIN-aLANA was produced from peGFP-aLANA-SIIN after replacement of the SIINFEKL peptide sequence by a stop codon and in-frame insertion of SIINFEKL peptide sequence in N-terminus of ORF73 sequence . Plasmid vectors pT7-eGFP-73Δx-SIIN were generated by sub-cloning the T7 promoter sequence downstream the CMV IE promoter in the NheI site ( S1 Table ) . Plasmid vector pEGFP-GE-SIIN was generated by insertion of ORF73 GE-rich domain sequence into pEGFP-SIIN . An additional SIINFEKL peptide sequence was further inserted in N-terminus of the GE sequence to create peGFP-SIIN-GE-SIIN ( S1 Table ) . Plasmid vector pEGFP-aLANA-GEm-SIIN was generated from pEGFP-ΔGE1-SIIN . A pMK-RQ plasmid containing the synthetic GEm sequence ( obtained by codon-optimization– http://eu . idtdna . com/CodonOpt –and further empirically modified to enrich in pyrimidines , then synthesized by GeneArt , Life technologies ) was digested ( PstI/KpnI ) followed by ligation into pEGFP-ΔGE1-SIIN . Then , the ORF73 N-terminus coding sequence was inserted into KpnI restriction site of pEGFP-ΔGE1-SIIN ( S1 Table ) . HaloTag pHT2 expression vector ( Promega ) was generously provided by Prof . Fransen ( KULeuven , Belgium ) and used to produce the p-aLANA-HT2 , pΔCR-HT2 and pΔGE-HT2 plasmids . To express eGFP-SIIN , aLANA-SIIN and ΔGE-SIIN constructs under control of the promoter EF1-α , pEFIN3 expression vector was digested with XbaI and used for insertion of PCR amplicons generated using the pEGFP-C1-based constructs as templates and primers EGFP-pEFIN3-Fwd and SIIN-pEFIN3-Rev ( S1 Table ) . Plasmid vector pEFIN3-Kb to generate the VeroKb stable cell line was generated by insertion of H-2Kb PCR amplicon from pcDNA3 . 1-H2Kb ( gift from Dr . J . Yewdell ) into the XbaI site of pEFin3 . All construct sequences were verified before being used in transfection experiments . mK3 expression vector was generously obtained from Dr P . G . Stevenson ( University of Queensland , Australia ) . 293Kb , 293Db , VeroKb or L929-Kb cells ( 105 cells/well in 24-well plates ) were transfected with 0 . 25 μg of plasmid using Fugene HD ( Promega ) . For co-transfection assays , cells were co-transfected with 0 . 25 μg of each plasmid DNA . MHC-I-peptide complexes were detected by immunostaining using allophycocyanin ( APC ) -conjugated anti-mouse H-2Kb-SIINFEKL complex ( mouse IgG1 , clone 25-D1 . 16 , eBioscience ) for 30 min on ice . Cells were incubated and washed in FACS buffer ( PBS containing 0 . 1% BSA and 0 . 09% NaN3 ) before acquisition . Where indicated , transfected cells were treated with MG132 ( Z-Leu-Leu-Leu-al ) , lactacystin , epoxomycin ( Sigma ) , rapamycin , chloroquine or 3MA ( InvivoGen ) during 16h before immunostaining . Antigen presentation was assayed using B3Z cells ( 105 cells/well in 24-well plates ) co-cultured during 16h at 37°C with 293Kb cells transfected 48h before . Total cells were then lysed at −80°C during 30min in lysis buffer ( 5mM MgCL2 , 1% Nonidet P-40 , 0 . 15mM Chlorophenol red-β-D-galactopyranoside ( CPRG , Invitrogen ) in PBS ) to quantify β-galactosidase activity . Each cell lysate replicate was then divided in 2 wells of a 96-well flat-bottom plate and incubated at 37°C during 20h . β-galactosidase activity was measured using a spectrophotometric determination of absorbance at 595 nm using an iMark microplate reader ( Biorad ) . 293Kb cells ( 2 . 5x105 cells/well in 12-well plates ) were grown on poly-D-lysine coated glass coverslips and transfected with 1 μg of plasmid using Fugene HD ( Promega ) . Transfected cells were stained 20 min at room temperature with an anti-mouse H-2Kb-SIINFEKL complex ( 25-D1 . 16 , eBioscience ) , washed and then incubated 1h at 37°C with secondary Alexa Fluor 568 goat anti-mouse IgG polyserum ( Life technologies ) . Cells were then washed and fixed in 4% paraformaldehyde in PBS before being mounted in ProLong antifade reagent ( Life technologies ) . Washing and incubation steps were done in PBS containing 10% FCS . Fluorescence was then visualized with a Leica confocal scanner ( TCS ) SP laser scanning microscope . 293Kb cells ( 5×105 cells/well in 6-well plates ) were transfected with 2μg of plasmid and Fugene HD ( Promega ) . Cells were lysed after 24h in 100μL RIPA buffer ( 25mM Tris•HCl pH 7 . 6 , 150mM NaCl , 1% Nonidet P-40 , 1% sodium deoxycholate , 0 . 1% SDS ) . Lysates were heat-denatured ( 95°C , 5 min ) before SDS-PAGE in laemmli buffer ( 31 . 25 mM Tris-Hcl pH 6 . 8 , 1% ( w/v ) SDS , 12 . 5% ( w/v ) glycerol , 0 . 005% ( w/v ) bromophenol blue , Biorad ) containing 10% ( v/v ) β-mercaptoethanol . Proteins were separated by electrophoresis in Mini-PROTEAN TGX precast 4–15% resolving gels ( Biorad ) in Tris-glycin running buffer ( 25mM Tris-base , 192mM glycine , 0 . 1% w/v SDS ) and transferred onto polyvinylidene difluoride membranes ( PVDF , Thermo Fisher Scientific , 0 . 45 μm pore size ) . The membranes were blocked with 3% skimmed milk in PBS/0 . 1% Tween-20 . For detection of eGFP fusion proteins , membranes were blotted with HRP-conjugated anti-GFP antibody ( mouse IgG1 mAb , 5000× diluted , Miltenyi Biotec ) . Mouse anti-β-actin ( Dako ) and anti-GAPDH ( Abcam , GA1R ) monoclonal antibodies were used with secondary HRP-conjugated rabbit anti-mouse polyserum ( Dako ) . Rabbit anti-c-Myc polyserum ( Sigma ) was used with HRP-conjugated goat anti-rabbit polyserum ( Dako ) . Novex Sharp pre-stained protein standard ( Thermo Fischer ) was used to determine band sizes . Chemiluminescent reaction was performed with ECL substrate ( GE Healthcare ) and membranes exposed onto an X-ray film . 293Kb cells ( 2 . 5x105 cells/well in 12-well plates ) were transfected with 1 μg of plasmid using Fugene HD ( Promega ) . Total RNA was extracted 24h after transection using RNeasy Miniprep kit ( Qiagen ) with on-column DNase I treatment . First-strand cDNA was then synthesized using oligo-dT and Superscript III Reverse transcriptase system ( Life technologies ) . Control lacking reverse transcriptase was performed for each RNA sample to monitor DNA contamination . Quantitative real-time PCR reactions ( qPCR ) were performed in iQ-SYBR green supermix using a CFX96 Touch Real-Time PCR Detection System ( Bio-Rad ) . Standard curves were generated using specific PCR amplicons and the obtained copy numbers were normalized to hypoxanthine-guanine phosphoribosyltransferase ( HPRT ) housekeeping gene ( S2 Table ) . Where indicated , transfected cells were treated with actinomycin D ( Sigma ) during 4 . 5h before RNA extraction . Capped RNAs were transcribed in vitro from AvrII-linearized pT7-eGFP-73Δx-SIIN plasmids using T7 RNA polymerase and mMessage mMachine kit ( Ambion ) . Next , molar equivalents were used for uncoupled in vitro translation using a rabbit reticulocyte lysate system ( Promega ) and [S35]-methionine ( Perkin Elmer ) . Translation reaction products were separated by SDS-PAGE before the gel was dried and exposed to an Amersham Hyperfilm MP ( GE Healthcare ) . Pulse-labelling of live cells expressing HaloTag-fusion proteins was performed overnight at 12h post-transfection using cell-permeable HaloTag TMR-Direct Ligand ( 5μM , Promega ) . Cells were then extensively washed with warm PBS and chased in phenol red-free standard growth medium for the indicated period . At various time points , the fluorescence signal was measured using an EnSpire multimode plate reader ( Perkin Elmer ) . In some experiments , pulse-labeled cells were further incubated at various time points of the chase period with cell-permeable HaloTag Oregon Green Ligand ( 5μM , Promega ) during 15min at 37°C followed by analysis by flow cytometry . To strip MHC-epitope complexes from the surface of 293Kb cells , cells were detached 48h post-transfection , washed with PBS and suspended in citrate phosphate ( pH 3 ) buffer ( 0 . 131 M citric acid , 0 . 066 M Na2HPO4 ) for 2 min on ice [56] . The suspension was then neutralized with a 100-fold dilution of growth medium , and cells were washed twice . Then , cells were suspended in growth medium and incubated in the presence or absence of 100μg/ml CHX for 5 h ( Sigma ) . Antigen presentation was assessed by flow cytometry analysis using anti-mouse H-2Kb-SIINFEKL complex monoclonal antibody . The AlHV-1 BAC clone was used to produce the recombinant plasmids using the galactokinase gene ( galk ) as selection marker in SW102 E . coli strain [21] . To generate the BAC ΔGE plasmid , a BAC ΔGE-galK plasmid was first produced by positive selection in presence of an amplicon obtained by PCR using the pgalK vector as a template , the forward chimeric primer H1ΔGE-galK-Fwd and the reverse chimeric primer H2ΔCR-galK-Rev ( S3 Table ) . The amplicon consisted in the galK sequence flanked by 44-bp sequences corresponding to the regions directly flanking the GE region of the AlHV-1 ORF73 ( access no . : Refseq NC_002531 , nt 117121–117170 and 117948–117991 ) in the viral genome . Full deletion of the GE region was further achieved by galK negative selection using annealed oligonucleotides H1H2-ΔGE-Fwd and H1H2-ΔGE-Rev . Finally , the BAC ΔGE-Rev plasmid was produced using two sequential steps consisting in the re-introduction of the galK sequence using the ΔGE-galK amplicon before negative selection performed in presence of the ORF73 full length fragment obtained by SacI digestion of the pEGFP-aLANA-SIIN plasmid ( S6 Fig ) . All deletions and insertions were further verified using restriction endonuclease and Southern blot approaches and sequencing of the recombination sites . All strains were reconstituted in MacT-Cre cells before propagation in BT cells . Southern blotting analysis was performed as described previously [19] . In vitro growth kinetics of recombinant viruses were compared to those of the WT . Cells were infected ( moi = 0 . 05 ) and both supernatants and infected cells were harvested at successive intervals . The total amount of infectious viral particles was determined by plaque assay on MDBK cells as described previously [19] . Four groups of specific pathogen-free New-Zealand white rabbits ( n = 4 ) were used . Animals were inoculated intranasally with 105 PFU of the different AlHV-1 recombinant viruses in PBS or PBS only for the mock-infected group . Rabbits were examined daily for clinical signs . According to bioethical rules , rabbits were euthanized when rectal temperature remained higher than 40°C for two consecutive days . PBMC were isolated from 5-ml of blood collected from the ear central artery before and at different time-points after infection . Immediately after euthanasia , single-cell suspensions were prepared from popliteal lymph node ( pLN ) and spleen as follows . Tissue biopsies were delicately chopped in sterile RPMI media and passed through a 70 μm cell-strainer ( BD Biosciences ) . Mononuclear leukocyte suspensions from peripheral blood and tissue samples were prepared with Ficoll-Paque Premium density gradient media ( GE Healthcare ) . 5-ml single-cell suspension was diluted 1:1 in sterile PBS , overlaid onto 5-ml Ficoll-Paque density cushion and centrifuged ( 1825×g ) during 20-min at room temperature . Mononuclear leukocytes at the interface were collected and washed in ice-cold PBS before further analysis . 8-week-old female C57BL/6 mice were subjected to DNA immunization as described previously [21] . Briefly , 20 μg of plasmid construct in 25 μL of PBS were electroporated in both hind legs at day 0 and 14 in the tibial cranial muscle . Blood sampling were taken at d0 , 7 , 14 and 27 before euthanasia and spleen harvested at d30 . Multi-color flow cytometry analysis of rabbit PBMC was performed as described previously [20] . Briefly , cells were incubated in FACS buffers ( PBS containing 0 . 1% BSA , 0 . 09% NaN3 ) with mAb anti-rabbit CD4 ( IgG2a , KEN-4 ) , CD8 ( IgG1 , 12C . 7 ) , IgM ( IgG1 , NRBM ) antibody cocktail and left on ice for 10min . Cells were washed and further incubated for 10min on ice with isotype-specific PE-conjugated rat anti-mouse IgG1 ( A85-1 , BD ) and biotinylated rat anti-mouse IgG2a ( R19-15 , BD ) antibodies . After a third wash , cells were incubated with FITC-conjugated anti-rabbit T cells ( KEN-5 ) , and APC-conjugated streptavidin ( BD ) before washing , staining with 7-AAD and acquisition . All antibodies were from AbD-Serotec-Biorad or specified otherwise . Following DNA immunization , blood samples were collected from mice at regular intervals , lysed in ACK lysis buffer ( Gibco ) following the manufacturer’s instructions and incubated in FACS buffer with mAb anti-mouse CD3 ( APC , 145-2C11 ) , CD8α ( FITC , 53–6 . 7 ) , CD44 ( PE-Cy7 , IM7 ) . Antibodies were from BD Biosciences . Samples were further incubated with H-2Kb-SIINFEKL tetramer ( Brilliant violet 421nm , NIH Tetramer Core Facility ) . Acquisitions were performed using a LSR Fortessa X-20 ( BD Biosciences ) . A total of 50 to 100 , 000 live events were collected and data were analyzed by Flowjo v10 . 0 . 7 software ( Treestar ) . Statistical analyses were conducted using Graphpad Prism v6 software ( GraphPad , San Diego , CA ) . Unpaired Student’s t test was conducted when comparing statistical difference between two datasets . One or two-way ANOVA were conducted when comparing more than two groups of data as indicated in the figure legends and followed by relevant post-test for mean multiple comparison . Sidak’s method was used when comparing pairs of means within each row , and Dunnett’s or Tukey’s methods were chosen for comparing every mean ( 3 or more columns ) within each row to one control mean or every mean with every other mean , respectively .	The macavirus alcelaphine herpesvirus 1 ( AlHV-1 ) is a γ-herpesvirus ( γ-HV ) that has been first isolated in East-Africa in 1960 from a wildebeest and identified to be the etiological agent of malignant catarrhal fever ( MCF ) in bovine . An interesting aspect of AlHV-1 is that it can persist in wildebeest by remaining latent whereas it induces MCF upon cross-species transmission in several species of ruminants including cattle . MCF is a deadly lymphoproliferative disease developing after a prolonged incubation period . We have recently demonstrated that viral genome maintenance by the latency-associated viral protein aLANA is essential for inducing MCF . In the present study , we have investigated the ability of aLANA to evade antigen-specific cytotoxic T lymphocytes , an important property of γ-HV genome maintenance proteins to enable long term virus persistence . We provide evidence that GE , a specific repeated region in the nucleotidic sequence of aLANA , is directly involved in restraining aLANA protein synthesis . Although the lack of GE in aLANA did not significantly affect MCF induction in the rabbit model , such mechanism resulted in severely reduced presentation of an antigenic peptide linked to aLANA in a model in vitro and ineffective induction of antigen-specific CD8+ T lymphocyte responses in vivo . These results are important as they suggest that immune evasion of aLANA during MCF in susceptible species is not essential in the pathogenesis of the disease but identified GE to be essential for immune evasion during latency , likely in the wildebeest natural host .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "blood", "cells", "transfection", "flow", "cytometry", "innate", "immune", "system", "antigen", "presentation", "medicine", "and", "health", "sciences", "immune", "cells", "immunology", "microbiology", "plasmid", "construction", "protein", "synthesis", "dna", "construction", "molecular", "biology", "techniques", "cytotoxic", "t", "cells", "chemical", "synthesis", "research", "and", "analysis", "methods", "white", "blood", "cells", "animal", "cells", "proteins", "t", "cells", "molecular", "biology", "spectrophotometry", "immune", "system", "biosynthetic", "techniques", "cytophotometry", "biochemistry", "cell", "biology", "viral", "persistence", "and", "latency", "virology", "biology", "and", "life", "sciences", "cellular", "types", "spectrum", "analysis", "techniques" ]	2017	Macavirus latency-associated protein evades immune detection through regulation of protein synthesis in cis depending upon its glycin/glutamate-rich domain
Apoptosis is an evolutionary conserved cell death mechanism , which requires activation of initiator and effector caspases . The Drosophila initiator caspase Dronc , the ortholog of mammalian Caspase-2 and Caspase-9 , has an N-terminal CARD domain that recruits Dronc into the apoptosome for activation . In addition to its role in apoptosis , Dronc also has non-apoptotic functions such as compensatory proliferation . One mechanism to control the activation of Dronc is ubiquitylation . However , the mechanistic details of ubiquitylation of Dronc are less clear . For example , monomeric inactive Dronc is subject to non-degradative ubiquitylation in living cells , while ubiquitylation of active apoptosome-bound Dronc triggers its proteolytic degradation in apoptotic cells . Here , we examined the role of non-degradative ubiquitylation of Dronc in living cells in vivo , i . e . in the context of a multi-cellular organism . Our in vivo data suggest that in living cells Dronc is mono-ubiquitylated on Lys78 ( K78 ) in its CARD domain . This ubiquitylation prevents activation of Dronc in the apoptosome and protects cells from apoptosis . Furthermore , K78 ubiquitylation plays an inhibitory role for non-apoptotic functions of Dronc . We provide evidence that not all of the non-apoptotic functions of Dronc require its catalytic activity . In conclusion , we demonstrate a mechanism whereby Dronc’s apoptotic and non-apoptotic activities can be kept silenced in a non-degradative manner through a single ubiquitylation event in living cells . In multicellular organisms , cells have a turning point in their lives to commit to either living or dying . Cells which are committed to die can employ different forms of cell death , the most common one being a conserved form of programmed cell death , called apoptosis [1 , 2] . Apoptosis plays important roles during development , to maintain tissue homeostasis in adult organisms and in response to stress conditions [3 , 4] . Studies aimed at the elucidation of regulatory pathways of apoptosis are of outstanding importance because dysregulation of apoptosis can lead to many disorders , including neurodegenerative diseases and cancer [5 , 6] . The fruit fly Drosophila melanogaster provides an excellent model system in which to study the molecular mechanisms of apoptosis owing to its genetic conservation with mammals [7] , low genetic redundancy of the apoptotic factors , and a variety of well-established genetic techniques that allow to easily manipulate gene function in specific tissue types and even individual cells . Caspases , a highly conserved family of Cysteine ( Cys ) proteases , play a pivotal role in the regulation and execution of apoptosis . Caspases are produced as inactive monomeric zymogenes that consist of three domains , an N-terminal pro-domain , a large subunit containing the catalytic Cys residue , and a C-terminal small subunit . There are two types of apoptotic caspases: initiator caspases such as Caspase-2 , Caspase-9 and the Drosophila ortholog Dronc; and effector caspases such as the Caspase-3 , Caspase-7 and the Drosophila orthologs Drice and Dcp-1 [8 , 9] . The prodomains of initiator caspases carry protein/protein interaction motifs such as the Caspase Recruitment Domain ( CARD ) [10] . The scaffolding protein Apaf-1 and its Drosophila ortholog Dark also carry an N-terminal CARD domain [11–14] . In apoptotic cells , through CARD/CARD interactions with Dark , Dronc is recruited into and activated by a death-inducing protein complex , termed apoptosome [15 , 16] . Effector caspases which have short prodomains without protein/protein interaction motifs , are activated by the apoptosome through proteolytic cleavages between their subunits . Interestingly , correct stoichiometry between Dronc and Dark molecules is important for execution of apoptosis [17] . There is feedback inhibition between Dronc and Dark . Overexpression of one protein triggers degradation of the other one [17] ensuring that the levels of functional apoptosome units are low under these conditions . Only if both proteins are co-expressed can a significant apoptotic phenotype be recorded . Inhibitor of Apoptosis Proteins ( IAPs ) restrict apoptosis by inhibiting caspases [18 , 19] . IAPs are characterized by the presence of one to three Baculovirus IAP Repeats ( BIR ) and some bear a C-terminal RING domain that provides E3 ligase activity for ubiquitylation [18 , 20 , 21] . In living cells , Drosophila IAP1 ( Diap1 ) interacts with Dronc , Drice and Dcp-1 through the BIR domains [22] . Importantly , binding of Diap1 to caspases is not sufficient for their inhibition; ubiquitylation by the RING domain of Diap1 is required for full inhibition of these caspases [22–24] . In dying cells , the pro-apoptotic proteins Reaper ( Rpr ) , Hid and Grim bind to Diap1 and change the E3 ligase activity of the RING domain which promotes auto-ubiquitylation and degradation of Diap1 [25–32] . This leads to release of Dronc from Diap1 inhibition and free Dronc monomers can be recruited into the Dark apoptosome . Ubiquitylation is a post-translational modification , which results from conjugation of a protein called Ubiquitin to lysine residues of substrates either as a single moiety ( mono-ubiquitylation ) or by conjugation of ubiquitin chains ( poly-ubiquitylation ) [33 , 34] . The fate of a poly-ubiquitylated protein depends on the nature of the ubiquitin linkage . For example , K48 poly-ubiquitylation triggers proteolytic degradation of target proteins , while K63 poly-ubiquitylation regulates non-degradative events such as cell signaling [35–38] . In contrast , mono-ubiquitylation of a protein is usually not associated with protein degradation . Mono-ubiquitylation of target proteins is involved in DNA repair and endocytosis or may regulate translocation and interaction with other proteins [36 , 37] . Both mammalian and Drosophila caspases are subject of regulatory ubiquitylation mediated by IAPs [18 , 20 , 21 , 39–41] . For example , previous studies conducted in vitro and by transfection experiments in cell culture demonstrated that in Drosophila Dronc is ubiquitylated by Diap1 [23 , 24 , 42] . The importance of the RING domain for control of Dronc activity became clear from genetic analysis . diap1 mutants lacking the RING domain are embryonic lethal due to massive apoptosis [25] . Consistently , loss of the RING domain of Diap1 triggers processing and activation of Dronc [24] suggesting that ubiquitylation negatively regulates Dronc processing and activation . Initially , it was proposed that ubiquitylated Dronc is degraded by the proteasome [42–44] . However , we showed recently that the level of Dronc protein does not increase in proteasome mutants [45] suggesting that Dronc is not subject of proteasome-mediated degradation . In fact , the control of Dronc activity by ubiquitylation is much more complex than initially anticipated . In living cells , free monomeric Dronc is subject to non-degradative ubiquitylation , while processed and activated Dronc in the Dark apoptosome is degraded in a Diap1-dependent manner [17 , 24] . That raises the question about the nature and function of non-degradative ubiquitylation of free monomeric Dronc in living cells . Here , we report that in living cells Dronc is mono-ubiquitylated at Lysine 78 ( K78 ) in its CARD domain . To examine the role of K78 mono-ubiquitylation of Dronc , we mutated this residue to non-ubiquitylatable Arginine ( K78R ) . DroncK78R and Droncwt display similar enzymatic activities in vitro . However , DroncK78R is easier incorporated into the Dark apoptosome , is more efficiently processed and thus has higher enzymatic activity there . These data suggest that K78 ubiquitylation inhibits incorporation of Dronc into the Dark apoptosome . Surprisingly , DroncK78R also suppresses some of the phenotypes associated with catalytic inactivity of Dronc such as lethality , loss of compensatory proliferation and defects in male genitalia rotation . These observations provide evidence that K78 mono-ubiquitylation also controls non-apoptotic functions of Dronc and suggest that not all of the non-apoptotic functions of Dronc require its catalytic activity . In summary , this in vivo study provides a mechanistic link of how ubiquitylation of an initiator caspase can control its activity in both apoptotic and non-apoptotic pathways in a non-degradative manner . Because available anti-Dronc antibodies perform poorly in immunoprecipitation ( IP ) experiments , we took advantage of the Gal4/UAS system [46] and expressed Flag-tagged Dronc ( Flag-Dronc ) [47] ubiquitously using the daughterless-Gal4 ( da-Gal4 ) driver ( denoted da>Flag-Dronc ) . Expression of da>Flag-Dronc in whole animals does not cause any significant developmental , apoptotic or lethality phenotypes . To examine the functionality of Flag-Dronc , we tested if it can rescue the lethal phenotype of strong dronc mutants ( droncI24/droncI29 ) [48] . We indeed observed that da>Flag-Dronc is able to rescue the pupal lethality caused by dronc null mutations and can be activated in the apoptosome ( S1 Fig ) . To address the status of Dronc ubiquitylation , we immunoprecipitated Flag-Dronc from embryonic , larval , pupal and adult fly extracts and blotted with FK1 and FK2 antibodies that bind to ubiquitin-conjugated proteins , but not free , unconjugated ubiquitin . FK2 antibody binds to mono- and poly-ubiquitylated proteins , while FK1 antibody detects only poly-ubiquitin-conjugated proteins [49] . Blotting the IPs with FK2 antibody revealed high molecular poly-ubiquitin species; however , these are comparable to the control IPs and may represent unspecific co-immunoprecipitated proteins ( Fig 1A ) . In contrast , in the 60 kDa range , FK2 detected a single band specifically in Dronc IPs ( Fig 1A , arrow ) . This band is found in all developmental stages tested from embryos to adults . The FK1 antibody did not detect this band ( Fig 1A ) . Flag-Dronc has an estimated molecular weight ( MW ) of 51 kDa , and adding one ubiquitin moiety of ~8 . 5 kDa results in a combined MW of about 60 kDa , suggesting that this band may correspond to mono-ubiquitylated Flag-Dronc . To further verify mono-ubiquitylation of Dronc in vivo , we co-expressed da>Flag-Dronc and 6xHis-tagged ubiquitin ( 6xHis-ubiquitin ) and pulled down all ubiquitylated proteins using Ni-NTA agarose beads . Blotting for Flag-Dronc revealed a single band of about 60kDa , that was not present in the control IP in which we only expressed 6xHis-ubiquitin ( Fig 1B ) . This result further confirms that Dronc is ubiquitylated in vivo and the differential detection by FK2 , but not FK1 , suggests that it is—surprisingly—mono-ubiquitylated . As further evidence that this modification of Flag-Dronc corresponds to ubiquitylation , we incubated larval Flag-Dronc immunoprecipitates with a de-ubiquitylating enzyme , USP2 , that removes conjugated ubiquitin from target proteins . Consistently , in immunoblots , the FK2 signal is strongly reduced after USP2 incubation compared to the control ( Fig 1C , upper panel , arrow; quantified in 1C’ ) . Interestingly , although the majority of Flag-Dronc is de-ubiquitylated after USP2 incubation , this does not result in a significant reduction of the molecular weight ( MW ) of non-ubiquitylated Flag-Dronc ( Fig 1C , lower panel ) . Nevertheless , this characterization indicates that Flag-Dronc is mono-ubiquitylated under in vivo conditions . We were also interested to identify the ubiquitin ligase that mediates mono-ubiquitylation of Dronc . One good candidate is Diap1 which has been shown to ubiquitylate Dronc in vitro [23 , 24 , 42] . Ideally , to test if Diap1 ubiquitylates Flag-Dronc in vivo , one should examine homozygous mutant diap1 animals for loss of ubiquitylation of Dronc . However , these animals are early embryonic lethal due to strong apoptosis induction by loss of Diap1 [25] which makes this analysis very difficult . Therefore , we examined Flag-Dronc immunoprecipitates from larvae that were heterozygous for the strong diap15 allele [26 , 27] . Immunoprecipitates of Flag-Dronc from heterozygous diap15 extracts display a significant reduction of FK2 immunoreactivity ( Fig 1D , upper panel; quantified in 1D’ ) suggesting that Diap1 is involved in mono-ubiquitylation of Flag-Dronc . However , as already noted above in the context of the USP2 experiments , the Flag immunoblots do not display a significant size difference between ubiquitylated and non-ubiquitylated Flag-Dronc ( Fig 1D , lower panel ) . The reason for this unusual behavior is not known . To identify the ubiquitylated Lysine ( K ) residue , we submitted the 60kDa band from immunoprecipitated Flag-Dronc samples from both larval and pupal stages to mass-spectrometry ( LC-MS/MS ) analysis . Both analyses showed that Flag-Dronc is ubiquitylated at K78 ( S2A Fig ) . To also examine for poly-ubiquitylation , we submitted higher molecular weight bands of the Flag immunoprecipitates for LC-MS/MS analysis . However , there was no trace of ubiquitylation . In addition to mono-ubiquitylation of K78 , we also observed phosphorylation of Ser130 , an inhibitory modification of Dronc that has previously been reported [47] . Confirmation of a known modification of Dronc validates the LC-MS/MS approach . Importantly , LC-MS/MS analysis of apoptotic extracts ( induced by hs-hid ) revealed that the mono-ubiquitylation at K78 is absent ( S2B Fig ) . This observation suggests that K78 mono-ubiquitylation is a feature of Dronc in living cells and that it may control ( inhibit ) the apoptotic activity of Dronc . To determine whether DIAP1 can ubiquitylate Dronc at K78 , we performed in vitro ubiquitylation assays of Dronc with Diap1 as E3 ubiquitin ligase and analyzed in vitro ubiquitylated Dronc by mass spectrometry . As E2 conjugating enzymes we used either human UBE2D2 or Drosophila UBCD1 . In both cases , Dronc was found to be ubiquitylated at K78 by DIAP1 in vitro ( S2C and S2D Fig ) , suggesting that DIAP1 can mediate K78 ubiquitylation of Dronc . K78 resides in the CARD domain of Dronc ( Fig 1E ) which interacts with the CARD domain of Dark for recruitment of Dronc into the apoptosome . To study the role of K78 ubiquitylation , we mutated K78 to Arginine ( R ) and generated transgenic UAS-Flag-DroncK78R flies by phiC31-based site-specific integration [50 , 51] . In addition , we combined the K78R mutation with a mutation that changes the catalytic Cys ( C ) to Ala ( A ) ( C318A ) , generating transgenic UAS-Flag-DroncK78RC318A flies . As controls , we generated UAS-Flag-Droncwt , a catalytically inactive Dronc ( UAS-Flag-DroncC318A ) and empty vector transgenic flies . All constructs are inserted in the same landing site in the genome ( VK37 on 2nd chromosome ) . To test whether da>Flag-DroncK78R mutant flies lose the mono-ubiquitylation signal , we immunoprecipitated Dronc from larval samples and probed immunoblots with FK2 antibody . da>Flag-DroncK78R larval samples showed significantly reduced levels of mono-ubiquitylation ( Fig 1F , arrow; quantified in Fig 1F’ ) , suggesting that Flag-DroncK78R is less efficiently ubiquitylated compared to Flag-Droncwt . However , because K78 is the only Lys residue being detected by LC-MS/MS , we expected a complete loss of ubiquitylation in the Flag-DroncK78R mutant . Although significantly reduced , the mono-ubiquitylation signal is not completely lost ( Fig 1F’ ) suggesting that in the absence of K78 as major ubiquitin acceptor , another Lys residue may be used as alternative ubiquitylation site ( see Discussion ) . Nevertheless , the K78R mutation revealed that K78 of Dronc is a major ubiquitin acceptor . Interestingly also , as already observed in the USP2 and diap15 experiments , the MW of ubiquitylated and non-ubiquitylated Dronc is not significantly different ( Fig 1F , lower panel ) . Formation of the apoptosome is essential for activation of Dronc . Interestingly , a recent structural report about the Drosophila apoptosome revealed that K78 forms an intramolecular hydrogen bond with a critical residue ( Q81 ) that is required for interaction of the CARD domains of Dronc and Dark for apoptosome formation [16] . Therefore , we hypothesized that mono-ubiquitylation of Dronc at K78 inhibits the interaction with the CARD of Dark , effectively blocking recruitment of Dronc into the apoptosome under surviving conditions . To test this hypothesis in vivo , we used genetic and biochemical approaches . In genetic experiments , we tested whether apoptosis is induced when the K78 mono-ubiquitylation is lost in animals expressing da>Flag-DroncK78R . However , similar to da>Flag-Droncwt , expression of da>Flag-DroncK78R does not induce a significant apoptotic phenotype or even cause lethality . This is most likely due to the feedback inhibition mechanism between Dronc and Dark according to which overexpressed Dronc destabilizes Dark [17] , keeping the number of active apoptosome units low ( see Discussion ) . Nevertheless , combined expression of Flag-Droncwt and Dark ( tagged with GFP ( GFP-Dark ) [17] ) with GMR-GAL4 in the posterior eye imaginal disc induces apoptosis , causing eyes of reduced size with pigment loss ( Fig 2A ) and enhanced pupal lethality . Therefore , we asked whether loss of K78 mono-ubiquitylation causes increased activity of Flag-DroncK78R in the presence of mis-expressed GFP-Dark [17] . Indeed , we found that the adult eyes of GMR>Flag-DroncK78R+GFP-Dark flies are significantly smaller than GMR>Flag-Droncwt+GFP-Dark eyes ( Fig 2A and 2B ) . In addition , the pupal lethality was significantly increased in GMR>Flag-DroncK78R+GFP-Dark compared to GMR>Flag-Droncwt+GFP-Dark ( Fig 2C ) . To understand whether this phenotype is due to increased apoptotic activity of Flag-DroncK78R , we examined 3rd instar larval eye discs for apoptosis using TUNEL labeling . Parallel to the adult eye phenotypes , we observed significantly more apoptosis in the GMR>Flag-DroncK78R+GFP-Dark eye imaginal discs ( Fig 2D and 2E ) . In addition , fluorimetric caspase activity assays with extracts from GMR>Flag-DroncK78R+GFP-Dark heads showed a significantly higher cleavage activity towards the synthetic DEVD substrate than GMR>Flag-Droncwt+GFP-Dark ( Fig 2F ) . These data suggest that loss of K78 mono-ubiquitylation increases the apoptotic activity of DroncK78R in the Dark apoptosome . To examine the role of Diap1 for K78 mono-ubiquitylation of Flag-Dronc , we compared the eye phenotypes of GMR>Flag-Droncwt+GFP-Dark and GMR>Flag-DroncK78R+GFP-Dark in a heterozygous diap15 background . diap1 heterozygosity strongly enhanced the eye phenotype and lethality of GMR>Flag-Droncwt+GFP-Dark animals ( S3 Fig ) . However , loss of one copy of diap1 only weakly enhances the eye phenotype and lethality of GMR>Flag-DroncK78R+GFP-Dark animals ( S3 Fig ) . These genetic interaction data suggest that K78 ubiquitylation depends on Diap1 . Dark has a C-terminal caspase cleavage site that is thought to destabilize Dark , thus reducing its apoptosis-promoting activity [17 , 52] . Consistently , a cleavage resistant version of Dark ( DarkV ) showed a hypermorphic phenotype [52] . Therefore , in theory , DarkV should uncouple the anti-apoptotic feedback of Dronc on Dark . However , experimentally , that was not observed [17] . Co-expression of GMR>Droncwt+GFP-DarkV caused a similar small eye phenotype compared to GMR>Droncwt+GFP-Darkwt [17] . Thus , although DarkV was suggested to be more active than Darkwt , expression of either transgene with Droncwt did not change the equilibrium of the apoptosome activation [17] . Therefore , we examined whether co-expression of Flag-DroncK78R with GFP-DarkV under GMR control is sufficient to shift the equilibrium of apoptosome formation towards higher induction of apoptosis . Indeed , GMR>Flag-DroncK78R+GFP-DarkV executed more apoptosis compared to GMR>Flag-Droncwt+GFP-DarkV ( S4 Fig ) . Both the adult eye phenotype and the pupal lethality are worsened significantly in GMR>DroncK78R+GFP-DarkV flies ( S4 Fig ) . These findings are consistent with the notion that Flag-DroncK78R requires functional Dark for increased activity . To examine if the increased caspase activity of Flag-DroncK78R is due to increased intrinsic catalytic activity , we performed in vitro cleavage assays with bacterially expressed 6xHis-Droncwt , 6xHis-DroncK78R , 6xHis-DroncC318A and 6xHis-DroncK78RC318A . Because bacteria lack an ubiquitin system , 6xHis-Droncwt is not modified by ubiquitin enabling us to directly compare the intrinsic activities of the Dronc variants . In these experiments , we first tested the ability of the Dronc constructs to auto-process [53 , 54] . Both 6xHis-Droncwt and 6xHis-DroncK78R proteins are able to auto-process to a similar extend ( Fig 3A ) . In contrast , the catalytic mutant 6xHis-DroncC318A and double mutant 6xHis-DroncK78RC318A fail to auto-process ( Fig 3A ) , consistent with the expectation . Next , we performed in vitro cleavage assays of these Dronc preparations with its known cleavage target DrICE [53 , 54] which is Myc-tagged and carries a mutation in the catalytic site ( Myc-DriceC211A ) to block auto-processing of DrICE . While the catalytic mutants 6xHis-DroncC318A and 6xHis-DroncK78RC318A failed to cleave Myc-DriceC211A , both 6xHis-Droncwt and 6xHis-DroncK78R processed Myc-DriceC211A in vitro ( Fig 3B ) . However , the cleavage activities of 6xHis-Droncwt and 6xHis-DroncK78R are very similar in these assays suggesting that there are no intrinsic differences in the catalytic activities of 6xHis-Droncwt and 6xHis-DroncK78R . Furthermore , these data imply that the K78R mutation does not cause any structural defect to DroncK78R . However , in vivo , in the presence of Dark , Flag-DroncK78R has a higher catalytic activity than Flag-Droncwt ( Fig 2 ) suggesting that DroncK78R requires Dark for increased catalytic activity . Consistent with the increased catalytic activity of Flag-DroncK78R in the presence of Dark , a significantly higher amount of Flag-DroncK78R is found in the processed form compared to Flag-Droncwt in immunoblots of total extracts from da>Flag-Droncwt + GFP-Dark and da>Flag-DroncK78R + GFP-Dark larvae ( Fig 3C and 3C’ ) . To understand the mechanism of increased processing and catalytic activity of Flag-DroncK78R in the Dark apoptosome , we examined the interaction between DroncK78R and Dark . Because specific antibodies to Dark do not exist , we used the GFP-Dark transgenes [17] to immunoprecipitate GFP-Dark and associated Flag-Dronc . To avoid embryonic lethality of da>Flag-DroncK78R+GFP-Dark , Gal80ts was used to control the expression of UAS-GFP-Dark and UAS-Flag-Dronc transgenes . Using Gal80ts , da>Flag-Droncwt+GFP-Dark , da>Flag-DroncK78R+GFP-Dark and EV ( empty vector ) +GFP-Dark as control were induced for 24 h at 29°C and larval extracts were analyzed for Flag-Dronc and GFP-Dark . Longer induction periods ( e . g . ≥48 h ) also caused lethality . Consistent with a previous report [17] , compared to the EV control , expression of da>Flag-Droncwt+GFP-Dark and da>Flag-DroncK78R+GFP-Dark reduces Dark’s protein stability , as shown for GFP-Dark in Fig 3D ( top panel ) . In co-IP experiments , we detect an increased interaction between Flag-DroncK78R and GFP-Dark compared to Flag-Droncwt and GFP-Dark ( Fig 3D , bottom panel ) . In addition , the ratio between processed versus unprocessed Dronc is significantly increased for Flag-DroncK78R in complex with GFP-Dark compared to Flag-Droncwt ( Fig 3D , bottom panel; quantified in 3D’ ) , consistent with the increased apoptosis in imaginal discs and head extracts ( Fig 2 ) . These results suggest that compared to Flag-Droncwt , Flag-DroncK78R interacts stronger with Dark and is more efficiently processed for apoptosis induction . Taken together , these data suggest that living cells are protected from apoptosis by keeping Dronc at least partially inactive through K78 mono-ubiquitylation which appears to block recruitment into the Dark apoptosome . However , when cells are undergoing apoptosis , K78 mono-ubiquitylation is no longer present , allowing Dronc to interact with Dark in the apoptosome and induce cell death . Next , we examined the physiological role of K78 mono-ubiquitylation of Dronc . For this , we expressed wild-type and mutant Flag-Dronc transgenes using da-Gal4 in a dronc null background and scored for rescue . The null mutants used , droncI24 and droncI29 , have early stop codons at positions 28 and 53 , respectively [48] and do not produce any Dronc protein . droncI24/droncI29 null mutants display a strong semi-lethal phenotype . Less than 10% of the expected dronc homozygous mutant animals survive development ( Fig 4A ) and hatch as adults with wing abnormalities ( S5 Fig ) [48] . Expression of da>Flag-Droncwt rescues the lethality of dronc null mutant flies , but it is only a partial rescue . There is still about a 35% lethality ( Fig 4A ) , suggesting that da>Flag-Droncwt does not reach sufficient Dronc activity for full rescue . Interestingly , however , da>Flag-DroncK78R rescued the lethality of dronc null mutant significantly better than da>Flag-Droncwt . More than 80% of the expected progeny emerges as adults in the presence of Flag-DroncK78R ( Fig 4A ) . Because these transgenes were obtained by phiC31 integration in the same landing site , the expression levels of all Flag-Dronc constructs are comparable ( Fig 4H ) and are not responsible for the observed differences . Therefore , this result further supports the notion that Flag-DroncK78R has more activity than Flag-Droncwt and thus can better substitute for the loss of endogenous dronc . As expected , expression of catalytically inactive da>Flag-DroncC318A failed to rescue the lethality of dronc null mutants ( Fig 4A ) . Surprisingly , however , expression of da>Flag-DroncK78RC318A which lacks the K78 mono-ubiquitylation site and is catalytically inactive ( Fig 3A and 3B ) , did rescue the lethality of dronc null mutants to a significant degree ! About 60% of dronc mutant flies survived when expressing da>Flag-DroncK78RC318A compared to only 10% of dronc mutant flies expressing da>Flag-DroncC318A ( Fig 4A ) . Thus , the K78R mutation behaves as an intragenic suppressor of the lethality associated with loss of catalytic activity of Dronc . This result suggests that loss of K78 ubiquitylation can be advantageous for the survival of dronc mutant flies and can even—at least partially- overcome loss of the catalytic activity of Dronc . Because of the intragenic suppression of the lethality of the catalytic droncC318A mutant by the K78R mutation , we considered—although did not expect—that the K78R mutation would rescue the catalytic activity of DroncC318A and thus the apoptotic phenotype of dronc mutants . To test this possibility , we employed the developing Drosophila retina which consists of individual units called ommatidia . In developing Drosophila retinae , cells produced in excess between ommatidia ( interommatidial cells , IOCs ) are eliminated by apoptosis around 28-30h after puparium formation ( APF ) [55–58] . The retinal lattice is fully differentiated at 42-45h APF . Previous studies showed that droncI24 and dronc I29 mutants fail to remove excess IOCs during development; about six additional IOCs remain per ommatidium in dronc mutants ( Fig 4B and 4C ) [48 , 59] . To understand the relationship between K78 mono-ubiquitylation and catalytic inactivity during developmental apoptosis , we generated droncI29 mutant clones expressing Flag-Droncwt , Flag-DroncK78R , Flag-DroncC318A and Flag-DroncK78RC318A by MARCM and examined the ability of these constructs to restore IOC apoptosis in the pupal retina of dronc mosaics . As expected , while expression of Flag-Droncwt and Flag-DroncK78R rescues IOC apoptosis in droncI29 mutant clones , Flag-DroncC318A does not ( Fig 4D , 4E and 4G; quantified in Fig 4B ) . Importantly , although expression of Flag-DroncK78RC318A rescued the lethality of dronc mutant flies ( Fig 4A ) , it does not restore IOC apoptosis in dronc mutant clones ( Fig 4B and 4F ) . Consistently , da>Flag-DroncK78RC318A expression in dronc null background does not rescue the wing phenotype of dronc mutants ( S5E Fig ) . In addition , Flag-DroncK78RC318A does not have catalytic activity in vitro ( Fig 3A and 3B ) . Therefore , as expected , these findings suggest that the K78R mutation does not restore the catalytic activity of Flag-DroncK78RC318A . They further suggest that the suppression of the pupal lethality of dronc mutants by expression of Flag-DroncK78RC318A occurs independently of the catalytic activity of Dronc which is therefore not absolutely essential for the survival of the flies . These data further imply that K78 mono-ubiquitylation controls additional , non-catalytic ( apoptosis- and effector-caspase-independent ) functions of Dronc whose failure in dronc mutants contributes to lethality . Next , we examined whether K78 mono-ubiquitylation is involved in a non-apoptotic function of Dronc . We and others have shown that Dronc can trigger apoptosis-induced proliferation ( AiP ) of neighboring surviving cells independently of downstream effector caspases and thus apoptosis [44 , 60–62] . Expression of the effector caspase inhibitor P35 is used to uncouple AiP from apoptosis . This treatment blocks apoptosis , but triggers chronic Dronc activity which causes tissue overgrowth due to permanent AiP [44 , 61–65] . It was previously shown that co-expression of p35 with dronc or pro-apoptotic hid using ey-Gal4 ( ey>dronc+p35 or ey>hid+p35 ) in eye imaginal discs causes head overgrowth with pattern duplications , while expression of catalytically inactive ey>droncC318S+p35 did not [61 , 62 , 65] . Consistently , expression of Flag-Droncwt and Flag-DroncK78R in ey>p35 or ey>hid+p35 background induced or enhanced head overgrowth , respectively , while catalytically inactive Flag-DroncC318A displayed wild-type head phenotypes in these assays ( Fig 5A and 5B; S6A Fig ) . Surprisingly , however , expression of Flag-DroncK78RC318A in ey>hid+p35 and ey>p35 assays also showed a similar overgrowth phenotype compared to Flag-Droncwt or Flag-DroncK78R ( Fig 5A and 5B; S6A Fig ) . Thus , similar to the results obtained in the rescue crosses of dronc induced lethality , loss of K78 ubiquitylation can suppress loss of catalytic activity in AiP . As controls , we expressed Flag-Dronc constructs with ey-GAL4 in the absence of p35 . However , simple overexpression of the Flag-Dronc construct did not trigger any overgrowth phenotype in these crosses ( S6B Fig ) . Because we showed in Figs 2 and 3 , that Flag-DroncK78R interacts better with GFP-Dark than Flag-Droncwt , we wondered if the rescue of AiP by Flag-DroncK78RC318A is dependent on the interaction with Dark . Indeed , in the absence of Dark ( by RNAi ) , Flag-DroncK78RC318A is no longer able to restore AiP in ey>hid+p35 background ( Fig 5A and 5B ) . During development , Drosophila male genitalia make a full 360° clockwise rotation [66] . When components of the apoptotic machinery ( hid , dronc , drICE ) are impaired , the rotation fails or is incomplete [67–70] suggesting that it is an apoptosis-driven event . We examined whether expression of da>Flag-Dronc constructs could rescue the genitalia rotation defect in droncI24/droncI29 males . da>Flag-Droncwt and da>Flag-DroncK78R fully rescued the male genitalia rotation phenotype of dronc mutant males ( 100% of males display 360° rotation ) ( Fig 5C; quantified in Fig 5D ) . In addition , these males were fertile . In contrast , da>Flag-DroncC318A was unable to rescue the droncI24/droncI29 rotation defect and had incomplete rotations ranging from 180° to 270° ( Fig 5C and 5D ) . These males were also sterile . Interestingly , da>Flag-DroncK78RC318A partially rescued the rotation defect associated with dronc null mutations ( 62% of males display 360° rotation ) ( Fig 5C and 5D ) . However , sterility caused by dronc null mutations was not suppressed suggesting that other non-apoptotic processes such as sperm maturation are not rescued [71] . The partial rescue of the rotation phenotype by Flag-DroncK78RC318A is potentially interesting because it may suggest that Dronc has two functions for male genitalia rotation: in addition to the previously reported effector caspase-dependent function [69 , 70] , it may also have an effector caspase-independent function . Because effector caspases require catalytic activity of Dronc for activation , only the effector caspase-independent function can be rescued by Flag-DroncK78RC318A , giving rise to the observed partial rescue ( Fig 5C and 5D ) . The rescue of the rotation phenotype by Flag-DroncK78RC318A is also dependent on Dark—at least partially—as dark RNAi reduces the rescue to 38% full rotation ( Fig 5D ) . These data further suggest that K78R mutation is an intrinsic suppressor of loss of Dronc’s catalytic activity . Our in vivo data uncovered an elegant mechanism of how Dronc activation is regulated through mono-ubiquitylation and how this modification affects both catalytic and non-catalytic functions of Dronc . Our MS/LC-MS data from larval and pupal samples demonstrate that in living cells , Dronc is mono-ubiquitylated at K78 . Because mono-ubiquitylation is not a mark for proteasome-mediated degradation , this finding explains why monomeric Dronc is not degraded in living cells [24] . Mono-ubiquitylation of Dronc is not an unprecedented observation in the caspase field . It was previously reported that cIAP2 promotes mono-ubiquitylation of the effector caspases Caspase-3 and Caspase-7 in vitro [72] . However , the significance of this mono-ubiquitylation is not known . Furthermore , the paracaspase MALT1 is subject to mono-ubiquitylation [73 , 74] . Interestingly , this modification leads to MALT1 activation . Here , we add the initiator caspase Dronc in Drosophila to the list of caspases being mono-ubiquitylated . Mono-ubiquitylation of K78 of Dronc does not regulate the intrinsic catalytic activity of Dronc . Purified recombinant Droncwt and DroncK78R have comparable catalytic activities in vitro . However , the location of K78 in the CARD domain suggests a regulatory modification for the interaction with Dark . Consistently , K78 was recently reported to be a critical residue for the interaction between the CARD domains of Dronc and Dark [16] . Indeed , our genetic analysis suggests that DroncK78R increases the physical association with Dark , resulting in increased processing of Dronc and thus higher apoptotic activity . Thus , we propose that in living cells , K78 mono-ubiquitylation of Dronc prevents the interaction with Dark . Because of the increased processing and activation of DroncK78R , we expected a very strong apoptotic phenotype when expressing DroncK78R in flies . However , although we observed increased apoptosis by expression of DroncK78R compared to Droncwt , it was not as severe as expected and depended on the presence of mis-expressed Dark . There are a few possibilities to explain this result . Although K78 was identified as the only ubiquitin acceptor site by LC-MS/MS analyses , we did not see a complete loss of mono-ubiquitylation in Flag-DroncK78R flies . It is possible that when this major ubiquitin acceptor site is mutated , another Lys residue is selected for ubiquitylation . Nevertheless , the partial loss of ubiquitylation in DroncK78R ( Fig 1E ) is sufficient to shift Dronc activity to a higher level . This increased activity depends on the presence of Dark . Another possibility to explain the absence of a significant apoptotic phenotype of da>Flag-DroncK78R is that correct stoichiometry between Dronc and Dark molecules is important for execution of apoptosis [17] . These proteins mutually control their stability . Overexpression of one protein triggers degradation of the other one [17] . This balance ensures that the levels of functional apoptosome units are low and this is most likely the reason why expression of each protein by itself in a tissue or even in the whole animal does not cause a significant apoptotic phenotype or complete lethality [17] . Only if both proteins are co-expressed can a significant apoptotic phenotype be recorded and under those conditions can DroncK78R trigger a stronger apoptotic phenotype compared to Droncwt , as observed in Fig 2 . Nevertheless , it should be pointed out that there are also conditions under which mis-expression of Dronc alone without simultaneous co-expression of Dark is sufficient to induce an ectopic phenotype . The incomplete expansion of the adult wing in response to Dronc-only mis-expression is a prominent example [47] . We also examined the role of K78 ubiquitylation in a catalytically inactive ( C318A ) Dronc background . da>Flag-DroncC318A fails to rescue any of the dronc null mutant phenotypes examined such as lethality , apoptosis and male genitalia rotation , and also fails to induce AiP . However , surprisingly , the ubiquitylation-defective and catalytically inactive double mutant of Dronc ( da>Flag-DroncK78RC318A ) does rescue the lethality and male genitalia rotation phenotypes of dronc null mutants and promotes AiP ( Figs 4 and 5 ) . The rescue of these phenotypes is not the result of restoring the catalytic activity of DroncK78RC318A by the K78R mutation because in vitro cleavage assays demonstrated that the effector caspase DrICE was not processed and in vivo IOC apoptosis was not rescued in dronc null mutants ( Fig 3A and 3B; Fig 4 ) , indicating that DroncK78RC318A has no catalytic and thus no apoptotic activity . Therefore , even though DroncK78R is released from inhibitory ubiquitylation , it still needs its catalytic activity to execute apoptosis . Flag-DroncK78RC318A is an intragenic suppressor of several , but not all , phenotypes associated with loss of the catalytic activity of Dronc . Therefore , the Flag-DroncK78RC318A transgene offers unique opportunities to identify and characterize apoptosis- ( effector caspase- ) independent functions of Dronc and to distinguish them from effector caspase-dependent ones . These results allow making the following important conclusions about Dronc function . Firstly , the pupal lethality ( which is actually a strong semi-lethality ) associated with dronc null mutations is not only due to loss of the catalytic ( enzymatic ) activity . It appears that some non-catalytic functions of Dronc are also very important for survival of the animal . Loss of the catalytic activity may contribute to the pupal lethality , but it may not be the underlying cause . This conclusion may not apply to the embryonic lethality of dronc germline clones [48] . Secondly , because we demonstrated that K78 mono-ubiquitylation controls the interaction of Dronc with Dark , it appears that DroncK78RC318A executes its non-enzymatic functions also through increased interaction with Dark . Thus , increased interaction with Dark is sufficient for induction of several non-apoptotic functions of Dronc such as AiP . Thirdly , it is a hot debate in the caspase field how caspases are restrained from inducing apoptotic death during non-apoptotic processes [75–77] . However , our results imply that at least for the caspase Dronc , its catalytic activity is not strictly required for non-apoptotic processes , although it may contribute to it . Instead , it appears that K78 mono-ubiquitylation controls activation of Dronc for non-apoptotic processes without requiring the catalytic function of Dronc . Dronc is considered to be the Drosophila Caspase-9 ortholog; however it has more protein similarity to mammalian Caspase-2 [78] . Alignment of the CARD domains of Dronc and Caspase 2 showed that K78 is not a conserved residue . However , there are two conserved Lys residues at positions 20 and 65 . It is possible that Caspase-2 may be ubiquitylated at one of these residues and this ubiquitylation may play a role in formation of the PIDDosome , an apoptosome-like protein complex required for Caspase-2 activation [79] . On the other hand , Caspase-9 does not have any Lys residue in its CARD domain . It is possible that the CARD domain of Caspase-9 has not evolved an ubiquitylation control mechanism because the interaction between Caspase-9 and Apaf-1 is not rate limiting for Caspase-9 activation ( Cytochrome c release is ) . Nevertheless , similar to Dronc , mature Caspase-9 ubiquitylation has been shown in vitro [80] , suggesting that Caspase-9 activation may be controlled by ubiquitylation after activation in the Apaf-1 apoptosome . Our work highlights a mechanism where Dronc’s activity is negatively regulated through mono-ubiquitylation that interferes with its interaction partner Dark . This work may help understanding the similarities and differences of caspase activation in mammalian and Drosophila apoptosomes . Embryos , 3rd instar wandering larvae , 1–2 days old pupae and heads of adult flies were lysed in 100 ul of SDS lysis buffer containing 2% SDS , 150 nM NaCl , 10 mM TrisHCl , 20 uM NEM and protease inhibitors ( Promega ) , respectively . The samples were sonicated for 10 seconds twice after they were boiled at 100°C for 10 minutes . 900 ul of dilution buffer ( 10 mM TrisHCl , 150 mM NaCl , 2 mM EDTA and 1% Triton-X ) was added to the samples and samples were rotated at 4°C for 1 hour before centrifugation for 30 minutes . Protein concentrations of supernatants were measured by Bradford Assay . 30 ug and 425 ug of total protein were used for western blots and IPs , respectively . IP was performed with anti-Flag M2 magnetic beads ( Sigma-Aldrich M8823 ) overnight at 4°C with rocking . 100 ul of 150 ng/ul Flag peptide in TBS was used for elution which took place at 4°C for 2 hours . 25 ul of eluted protein was used for western blotting . Dilutions of antibodies used are as follows: anti-Flag M2 antibody ( 1:1000 ) , FK2 and FK1 ( Enzo Life Sciences– 1:200 ) , anti-Actin ( Millipore Mab1501- 1:2000 ) . For ubiquitin pull-down assays , 3rd instar larvae were collected and lysed in urea lysis buffer containing 8 M Urea , 0 . 1 M NaH2PO4 , 0 . 01 M TrisHCl , 0 . 05% Tween 20 , pH 8 . 0 and protease inhibitors . IP was performed with Nickel-NTA magnetic agarose beads ( Qiagen 36111 ) at 4°C overnight with rocking . 60 ul of 250 mM of Imidazole in urea lysis buffer ( pH 4 . 5 ) was used for elution . 30 ul of eluted protein was analyzed by western blot . Anti-His antibody ( Thermo Scientific-Fisher MA1-21315 ) was used at 1:1000 dilution . For co-IPs , 3rd instar larvae were collected and lysed in NP40 buffer ( 20 mM TrisHCl pH 8 . 0 , 137 mM NaCl , 1% NP40 , 2 mM EDTA and protease inhibitors ) . IP was performed with GFP-Trap ( ChromoTek ) magnetic beads at 4°C overnight . GFP-Dark protein was eluted with 50 ul of 0 . 2M Glycine buffer pH 2 . 5 . 25 ul of eluted protein was used for western blot . Anti-GFP antibody ( Thermo Scientific-Fisher MA5-15256 ) was used at 1:200 for IP-western blots , 1:1000 for western blots . Immunoblot band intensities are quantified with GelQuantNET software provided by biochemlabsolution . com . Uncropped immunoblots are presented in S7 and S8 Figs . Immunoprecipitated Flag-Dronc was incubated with 3 ul of USP2 enzyme ( Boston Biochem E-504 ) in deubiquitylation assay solution ( 50 mm EDTA , 100 mm DTT , 50 mm Tris-HCl and 150 mm NaCl ) for 90 min at 37°C . Flag-Dronc was immunoprecipitated from larval and pupal da>Flag-Dronc extracts as described above . 1 mg of protein was used for IPs . Elutions of eight IPs were pooled and concentrated with 0 . 5 ml centrifugal tubes ( Millipore UFC500324 ) . In vitro ubiquitylation assays were performed as described previously [81] . Concentrated IP samples and in vitro ubiquitylated Dronc were loaded to 4–20% gradient SDS-PAGE gels . The gels were stained with Coomassie Blue Solution ( Thermo Scientific-Fisher- 24590 ) and the 60 kDa band as well as higher molecular weight bands ( for in vivo samples ) were excised and submitted to MS Bioworks ( Ann Arbor , MI ) . Samples were digested with Chymotrypsin and analyzed by LC-MS/MS . In the in vitro and in vivo samples , one peptide ( K78ITQRGPTAY ) carried the di-Glycine motif , characteristic for ubiquitylation . The following fly stocks were used: daughterless ( da ) -Gal4; GMR-Gal4; UAS-Flag-Dronc [47]; UAS-Flag-Droncwt; UAS-Flag-DroncK78R , UAS-Flag-DroncC318A and UAS-Flag-DroncK78RC318A ( this work ) ; UAS-6xHis-ubiquitin ( this work ) ; UAS-GFP-Dark and UAS-GFP-DarkV [17]; droncI24 and droncI29 [48]; ey>p35 and ey>hid , p35 [60]; diap15 [26 , 27] . Please note that two UAS-Flag-Droncwt transgenes were used . The first one ( a kind gift of Dr . Sally Kornbluth ) was used in the initial phases of this work and has a random insertion on chromosome 3 [47] . The second one was obtained by phiC31 site-specific integration in the VK37 landing site on chromosome 2 ( see below ) . This line was used in combination with UAS-Flag-DroncK78R , UAS-Flag-DroncC318A and UAS-Flag-DroncK78RC318A . All crosses were carried out at room temperature . 3L MARCM clones were induced by heat shocking L1 larvae at 37°C for 45 minutes as described [82] . Co-expression of UAS-GFP-Dark and UAS-Flag-Dronc transgenes was controlled by GAL80ts [83] . Temperature shift was performed at 29°C for 24 h . 3rd instar larvae were collected for lysis immediately after temperature shift . Wild-type and mutant UAS-Flag-Dronc transgenic flies were generated by the phiC31 site-specific integration system [50 , 51] . Flag-Dronc-pTFW and Flag-DroncC318A-pAFW vectors were kind gifts from Dr . Sally Kornbluth . Flag-Dronc and Flag-DroncC318A were cloned into pENTR3C vector . Point mutations were generated by site-directed mutagenesis . AttB site for site-specific integration was cloned into pTFW vector ( DGRC—1115 ) . Wild-type and mutant Flag-Dronc coding sequences were cloned into attB-pTFW vector by Gateway Cloning Technology ( Gateway LR Clonase II Enzyme Mix ) . Plasmids were sent to Genetivision for injection . VK37 landing site was used for phiC31 integration [84] . UAS-6xHis-Ubiquitin transgenic flies were generated by random integration ( Bestgene ) of a pUAST-6xHis-Ubiquitin construct created by inserting a KpnI-XbaI fragment of N-terminal 6xHis human Ubiquitin pcDNA3 . 1 into pUAST [46] . Expression of 6xHis-Ubiquitin was validated by FK2 Western blotting of urea-based lysis/Ni2+-based purification lysates generated from 20 adult da-GAL4;UAS-6xHis-Ubiquitin flies . 3rd instar larval brain lobes with eye discs were dissected in PBS and fixed in 4% PFA . Samples were blocked with 2% NDS in PBST and stained with c-Dcp-1 ( Cell Signaling 9578–1:100 ) and anti-Flag ( 1:200 ) antibodies [85] . TUNEL was performed as described [86] . For pupal dissections , pupae were aged to 42 h-48 h APF . Pupal discs were dissected , fixed and stained for c-Dcp-1 and Dlg ( DSHB 4F3 anti-disc large -1:100 ) [85] . Imaginal discs were mounted in Vectashield and imaged by confocal microscopy . Caspase activity assays were performed as described [86 , 87] . Briefly , adult heads were lysed in caspase assay buffer ( 50 mM HEPES pH 7 . 5 , 100 mM NaCl , 1 mM EDTA , 0 . 1% CHAPS , 10% sucrose , 5 mM DTT , 0 . 5% TritonX-100 , 4% glycerol and protease inhibitors ) . Protein concentration was measured with Bradford Assay . 40 ug of protein was incubated with 100 uM of DEVD-AMC caspase substrate ( MP Biomedicals 195868 ) in a final volume of 100 ul of caspase assay buffer . Fluorescence was measured with spectrophotometer ( excitation 385 nM emission 460 nM ) at 15 min intervals for 3 hours at 37°C . Each experiment was done at least three times . For in vitro cleavage assays , wild type and mutant Dronc coding sequences were cloned into pET-28a plasmid to yield 6xHis fusion proteins . Generated plasmids were transformed to BL21 ( DE3 ) pLysS competent cells ( Promega L1191 ) . 50 ul of bacterial culture was grown at 37°C . Plasmid expression was induced by 0 . 2 mM IPTG for 3 h at 30°C as described [88] . Bacterial pellets were lysed with 4 ml of CellLytic B Cell Lysis Reagent ( Sigma-Aldrich B7435 ) after adding 0 . 2 mg/ml Lysozyme , 50 units/ml Benzonase and 1X protease inhibitor ( Roche ) . DriceC211A-pET23b plasmid was a kind gift from Dr . Guy Salvesen [53] . DriceC211A coding sequence was cloned into PT7CFE1-Nmyc plasmid ( Thermo Scientific 88863 ) . Myc-DriceC211A protein was generated by using TNT Rabbit Reticulocyte Lysate System ( Promega L4610 ) . 4 ul of Myc-DriceC211A protein was incubated with 100 ug of wild-type and mutant 6xHis-Dronc protein in caspase assay buffer ( 100 mM Hepes pH 7 . 5 , 0 . 1% CHAPs , 10% sucrose , 10 mM DTT , 50 mM Nacl , 0 . 5 mM EDTA , protease inhibitor ) . The reaction was incubated at 30°C for 3 hours [54] and analyzed by western blotting . Anti-Myc antibody ( Santa Cruz SC40 ) was used at 1:200 concentration . Student’s t-test is used in all graphical analyses with parametric statistics . Crosses are repeated at least three times . Numbers of fly eyes used for area calculation and staining intensity are indicated in corresponding figures . The quantification of eye size was done using the Histogram function in Photoshop .	Apoptosis is a programmed cell death mechanism which is conserved from flies to humans . Apoptosis is mediated by proteases , termed caspases that cleave cellular proteins and trigger the death of the cell . Activation of caspases is regulated at various levels such as protein-protein interaction for initiator caspases and ubiquitylation . Caspase 9 in mammals and its Drosophila ortholog Dronc carry a protein-protein interaction domain ( CARD ) in their prodomain which interacts with scaffolding proteins to form the apoptosome , a cell-death platform . Here , we show that Dronc is mono-ubiquitylated at Lysine 78 in its CARD domain . This ubiquitylation interferes with the formation of the apoptosome , causing inhibition of apoptosis . In addition to its apoptotic function , Dronc also participates in events where caspase activity is not required for cell killing , but for regulating other functions , so-called non-apoptotic functions of caspases such as apoptosis-induced proliferation . We found that mono-ubiquitylation of Lysine 78 plays an inhibitory role for these non-apoptotic functions of Dronc . Interestingly , we demonstrate that the catalytic activity of Dronc is not strictly required in these processes . Our in vivo study sheds light on how a single mono-ubiquitylation event could inhibit both apoptotic and non-apoptotic functions of a caspase .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "cell", "death", "invertebrates", "medicine", "and", "health", "sciences", "reproductive", "system", "molecular", "probe", "techniques", "cell", "processes", "immunoblotting", "cloning", "animals", "animal", "models", "drosophila", "melanogaster", "model", "organisms", "immunoprecipitation", "experimental", "organism", "systems", "molecular", "biology", "techniques", "eyes", "drosophila", "research", "and", "analysis", "methods", "head", "molecular", "biology", "insects", "precipitation", "techniques", "arthropoda", "cell", "biology", "anatomy", "phenotypes", "apoptosis", "genetics", "biology", "and", "life", "sciences", "ocular", "system", "organisms", "genital", "anatomy" ]	2017	An inhibitory mono-ubiquitylation of the Drosophila initiator caspase Dronc functions in both apoptotic and non-apoptotic pathways
Homolog pairing and crossing over during meiosis I prophase is required for accurate chromosome segregation to form euploid gametes . The repair of Spo11-induced double-strand breaks ( DSB ) using a homologous chromosome template is a major driver of pairing in many species , including fungi , plants , and mammals . Inappropriate pairing and crossing over at ectopic loci can lead to chromosome rearrangements and aneuploidy . How ( or if ) inappropriate ectopic interactions are disrupted in favor of allelic interactions is not clear . Here we used an in vivo “collision” assay in budding yeast to test the contributions of cohesion and the organization and motion of chromosomes in the nucleus on promoting or antagonizing interactions between allelic and ectopic loci at interstitial chromosome sites . We found that deletion of the cohesin subunit Rec8 , but not other chromosome axis proteins ( e . g . Red1 , Hop1 , or Mek1 ) , caused an increase in homolog-nonspecific chromosome interaction , even in the absence of Spo11 . This effect was partially suppressed by expression of the mitotic cohesin paralog Scc1/Mdc1 , implicating Rec8's role in cohesion rather than axis integrity in preventing nonspecific chromosome interactions . Disruption of telomere-led motion by treating cells with the actin polymerization inhibitor Latrunculin B ( Lat B ) elevated nonspecific collisions in rec8Δ spo11Δ . Next , using a visual homolog-pairing assay , we found that the delay in homolog pairing in mutants defective for telomere-led chromosome motion ( ndj1Δ or csm4Δ ) is enhanced in Lat B–treated cells , implicating actin in more than one process promoting homolog juxtaposition . We suggest that multiple , independent contributions of actin , cohesin , and telomere function are integrated to promote stable homolog-specific interactions and to destabilize weak nonspecific interactions by modulating the elastic spring-like properties of chromosomes . Meiosis is a specialized cell division program that generates haploid gametes from diploid parental cells . A hallmark of the meiosis I division is the reductional segregation of homologous chromosomes while the meiosis II division segregates sister chromatids . The reductional division requires crossing over between homologous chromosomes in combination with sister chromatid cohesion [1] , [2] . Errors preventing normal chromosome segregation are a major cause of birth defects and miscarriage [3] . Crossing over is the outcome of reciprocal exchange of chromosome segments of homologous nonsister chromatids . Typically exchange occurs at allelic positions on homologous chromosomes but can also occur erroneously between ectopic regions of homology located on nonhomologous chromosomes , resulting in deletions , insertions and/or translocations [4]–[7] . Over the past decade , major inroads have been made in understanding mechanisms that promote pairing between homologous chromosomes but little is known about the mechanisms that prevent nonallelic interactions [8] . Several lines of evidence point to the sequestration of repeated elements to “silenced” regions near the nuclear periphery [7] , [9]–[11] or through engagement with allelic DNA sequences by homologous recombination [12] , [13] . The relationship between events that initiate crossing over and mechanisms that promote the side-by-side alignment of homologs varies among species [14] . In a majority of model organisms studied , including mouse , plants and fungi , the repair of Spo11-induced double-strand breaks ( DSBs ) using the homologous chromosome as a repair template is a major driver of pairing [15]–[23] . By contrast , in Caenorhabditis elegans and Drosophila females where recombination is still required for proper disjunction , homologs can pair even in the absence of meiotic-induced DSBs [24]–[26] . In C . elegans , pairing is initiated at pairing centers found at one end of each chromosome [27] . In Drosophila females , achiasmate chromosomes can pair via regions of heterochromatin [24] , [25] , [28] . In Drosophila , and to a lesser extent in budding yeast , an alternative mechanism to segregate achiasmate chromosomes exists that relies on homolog nonspecific interactions between centromere sequences [29]–[31] . While full levels of pairing in budding yeast Saccharomyces cerevisiae requires the formation and repair of DSBs there is also evidence for DSB-independent pairing both in vegetatively dividing cells and during meiosis in ( e . g . in a spo11Δ mutant ) [32]–[36] . The configuration of chromosomes with respect to centromere and/or telomere clustering and chromosome territories contributes in part to associations between homologous chromosomes at these regions [8] , [13] , [37] , [38] . Centromere coupling is an early event during meiotic prophase that involves pairwise associations of centromeres independent of homology [29] , [35] . Examples abound from a wide variety of species for somatic homolog pairing in higher eukaryotes with direct influence on gene expression or DNA repair [7] , [39] . The structural core of the meiotic chromosome axis in budding yeast comprises a conserved group of proteins , including Rec8 , a meiosis-specific α-kleisin subunit of cohesin as well as Red1 , Hop1 and Mek1 [40]–[46] . Inactivation of any of these proteins compromises interhomolog bias and homolog pairing [15] , [18] , [21] , [47]–[53] . Deletion of Rec8 also impacts several events of meiotic prophase not associated with sister chromatid cohesion , including region-specific distribution of Spo11 and DSB formation along chromosomes , meiotic S-phase timing , centromere coupling , synapsis , homolog pairing , transcription , and progression through meiotic prophase [40] , [44] , [49] , [51] , [54]–[58] . Rec8 also plays an important role in chromosome segregation at meiosis I by preventing premature sister chromatid separation prior to anaphase I [40] . In addition to the biogenesis of specialized chromatin architecture , meiotic chromosomes of nearly all species assume a polarized , nonrandom configuration in the nucleus , often with telomeres clustered toward one side of the nucleus [59] . This configuration is associated with vigorous telomere-led movement driven by cytoskeleton structures ( either actin or microtubules depending on the species ) outside the nucleus through protein bridges that span the inner and outer nuclear membranes and attach to telomeres [60] , [61] . In budding yeast , the velocity of telomere-led movement is greatest during late zygotene to pachytene stages when homologs are already paired , however , slower chromosome movement can be observed prior to zygotene during the pairing stage [62]–[65] . Chromosome organization and motion appear to be coupled to events associated with pairing and recombination since nearly every mutation affecting one or both of these aspects also exhibits slow turnover of recombination intermediates and delayed pairing . Meiotic chromosomes are mechanically linked to the cytoskeleton through the intact nucleus by a conserved SUN-KASH protein bridge [66] , [67] . Ndj1 is a fungal-specific telomere-associated protein that promotes telomere/NE associations [63] , [68]–[70] . Ndj1 interacts with the conserved SUN-protein Mps3 that spans the inner nuclear envelope [71] . Mps3 interacts with Csm4 , a putative KASH protein with a single trans-membrane tail domain bridging the outer nuclear envelope; Csm4 is required for telomeres to coalesce into the bouquet configuration and undergo Ndj1-dependent motion [63] , [64] , [72] , [73] . This work focuses on how the structure and organization of chromosomes in the nucleus impacts interactions between allelic and ectopic interstitial chromosomal loci . Here we carried out extensive epistasis analysis using deletion mutations in genes known to be involved in each of the functions described above to define the co-dependent or independent pathways leading to close-stable homolog juxtaposition ( CSHJ ) . We applied an in vivo assay ( Cre/loxP ) that measures the relative spatial proximity/or accessibility of pairs of chromosomal loci [32] . Maximal levels of site-specific recombination between homologous chromosomes indicated close , stable homolog juxtaposition of the assayed interstitial loci [21] . Through the analysis of mutants defective for various processes related to meiotic recombination we found that early steps of homologous recombination , including strand invasion and single end invasion are major determinants of CSHJ , while synapsis plays a relatively minor role [21] , [74] , [75] . To probe the spatial proximity and/or accessibility of pairs of interstitial chromosomal loci in vivo , we measured the frequency of Cre-catalyzed recombination between pairs of loxP sites located at allelic and ectopic chromosomal loci per meiosis . These sites were integrated at positions equidistant from the centromere and the adjacent telomere on the long arms of two average sized chromosomes ( V and VIII; see Experimental Procedures for more details ) . Previously , we measured site-specific recombination events by selecting for prototrophs resulting from the coupling of a promoter region to a selectable reporter gene; prototrophs were recovered from synchronized meiotic cells plated on selective media by “return to growth” ( RTG ) at various time points after the initiation of meiosis ( by transfer to sporulation medium; SPM [21] ) . In this study we measured recombinant DNA products by quantitative PCR using chromosome-specific primers flanking each loxP site ( Figure 1A ) . One key advantage of using qPCR is that Cre/loxP recombination can be assessed in strains that do not survive RTG and processing samples is more efficient . Template DNA for PCR was isolated from cells collected 10 hours after the initiation of meiosis . The recombination events are normalized for each sample by dividing by the copy number of a control locus that does not undergo Cre/loxP recombination ( ACT1 ) . By multiplying the normalized value by the total number of chromatids per chromosome ( four chromatids for most of the strains analyzed in this study ) , we generate the number of Cre-mediated recombination events per meiosis . The output of this assay , for both the RTG and qPCR method , is the frequency of Cre/loxP recombination events per meiosis , which we will refer to here as “collisions” ( Figure 1A ) . Strains used in this analysis carry an ndt80Δ mutation to prevent pachytene exit; blocking this late prophase step provides a control for differences in division timing and/or arrest exhibited by various meiotic mutants [76] . Prolonged arrest at pachytene does not affect the output of the assay since NDT80 and ndt80Δ strains gave similar levels of both allelic and ectopic collisions ( Figure 1B ) . It is important to note that collision levels represent the cumulative events that occur from the time of bulk DNA replication , when Cre recombinase is induced by galactose addition , until a fixed arrest point prior to exit from pachytene . Galactose induction increased allelic and ectopic collision levels 100- and 350-fold compared to those in untreated cells , respectively ( Table S1 ) ; thus , the dynamic range spans two orders of magnitude while background level of events is negligible . Collision levels measured using the qPCR and RTG methods were in agreement: Allelic collision levels were 0 . 13±0 . 03 and 0 . 14±0 . 03 recombinants/4 chromatids for qPCR and RTG values , respectively; ectopic collision levels were 0 . 018±0 . 004 and 0 . 015±0 . 004 , respectively ( Figure S1 ) . From previous studies , we have inferred that the elevated level of allelic versus ectopic collisions is due to homology-dependent interhomolog interactions in sequences outside the reporter locus [21] , [74] , [75] , [77] . A number of mutants defective for meiotic recombination including spo11Δ , sae2Δ , rad51Δ , and rad52Δ exhibited reduced allelic collision levels compared to wild type when measured by qPCR and RTG assays ( Figure S1 ) . Additional mutants defective for biochemical aspects of recombination including zip3Δ , rdh54Δ , and sgs1-mn ( a meiotic null allele of SGS1 ) were analyzed in the course of this study but not discussed here ( Figures S2 , S3 and Table S2; [78]–[83] ) . Rec8 plays dual roles during meiotic prophase; the first is to mediate sister chromatin cohesion and the second as a structural component of the chromosome axis that regulates the position and outcomes of homologous recombination ( Introduction ) . To explore if one or both of these functions influence the proximity and/or accessibility of interstitial chromosomal loci , we measured allelic and ectopic collision levels in a strain expressing the mitotic cohesin SCC1/MCD1 in place of REC8 ( pREC8-SCC1 ) in the rec8Δ mutant . In pREC8-SCC1 , sister chromatid cohesion is maintained while meiosis-specific functions , including chromosome axis and SC assembly are absent [49] , [56] , [84] . In rec8Δ mutants , both aspects of Rec8 function are absent . We found that pREC8-SCC1 reduced the level of allelic collisions 2-fold compared to wild type ( 0 . 064±0 . 009 versus 0 . 127±0 . 031 , P<0 . 00001; Figure 2 ) implicating a role for Rec8 in promoting allelic chromosome interactions independent of its role in sister chromatid cohesion . Brar et al . reached a similar conclusion by analyzing pairing in individual cells using GFP-tagged chromosomes in this mutant background [49] . To our surprise , we found that the level of allelic collisions and ectopic collisions in rec8Δ ( i . e . the absence of cohesion ) was greater than in the pREC8-SCC1 ( P = 6 . 04e−11; Figure 2 ) . In addition , the level of ectopic collisions was elevated 3-fold in the rec8Δ mutant compared to wild type ( 0 . 058±0 . 009 versus 0 . 019±0 . 006 , P = 1 . 04e−13 ) and pREC8-SCC1 ( 0 . 058±0 . 009 versus 0 . 032±0 . 004 , P = 3 . 43e−9 ) . Together these results suggest that Rec8 promotes interhomolog interactions and suppresses ectopic interactions , and perhaps nonspecific allelic interactions ( below ) . These effects may be region-specific since the regions including FLO8 ( V ) and NDT80 ( VIII ) are enriched for Rec8 binding and exhibit decreased levels of DSBs in rec8Δ mutant cells compared to wild type [85]–[87] . We reasoned that the relatively modest reduction in the allelic collision level conferred by rec8Δ compared to pREC8-SCC1 might be due to the inclusion of a significant fraction of “nonspecific” interactions ( i . e . those occurring in the absence of homology-dependent interactions ) . To test this , we measured the level of collisions in a spo11Δ mutant background since homologous recombination is a major driver of allelic collisions in wild type strains [75] . We found that the level of allelic collisions in spo11Δ rec8Δ was 3 . 9-fold greater than in the spo11Δ single mutant ( 0 . 110±0 . 020 versus 0 . 028±0 . 009 , P = 1 . 49e−11 respectively; Figure 2 ) . Allelic collisions in spo11Δ pREC8-SCC1 were also greater than spo11Δ ( 2-fold; 0 . 055±0 . 011 versus 0 . 028±0 . 009 , P = 1 . 72e−5 respectively; Figure 2 ) , but to a lesser extent than spo11Δ rec8Δ . Likewise , ectopic collision levels were elevated in spo11Δ rec8Δ and spo11Δ pREC8-SCC1 compared to spo11Δ ( P = 0 . 0003 and P = 7 . 88e−9 , respectively; Figure 2 ) . This trend was also observed , but to a lesser extent , in a spo11Δ rec8Δ cdc6-mn where cells progress through meiosis without fully duplicating the parental chromosomes ( Figures S2 , S3; [88] ) . These results suggest that the loss of Rec8 cohesion function ( but not loss of sister cohesion per se ) leads to increased interactions between chromosomal loci independent of DSB formation . We next tested if other components of the meiotic chromosome axis ( Red1 , Mek1 and Hop1 ) limit nonspecific collisions similar to Rec8 . Unlike the case for spo11Δ rec8Δ , however , we found that the levels of allelic and ectopic collisions in spo11Δ red1Δ , spo11Δ mek1Δ and spo11Δ hop1Δ mutants were indistinguishable from spo11Δ , with the exception of spo11Δ hop1Δ in which ectopic collision levels were slightly reduced ( P = 0 . 005; Figure 2 ) . These results further indicate that the effect of the rec8Δ on increasing chromosome interactions is due the absence of cohesin and not by disrupting the core axis structure . We next tested the possibility that the relatively high level of nonspecific chromosome interactions in spo11Δ rec8Δ is due to the juxtaposition of loci located at similar chromosomal “latitudes” in the bouquet configuration since the bouquet persists in this mutant background [53] . We reasoned that disrupting the bouquet might reverse the increased levels of nonspecific interactions conferred by rec8Δ . To test this we deleted NDJ1 , encoding a telomere-associated protein that promotes attachment of chromosome ends to the nuclear envelope , and assayed collisions under this condition where the bouquet is absent [70] , [73] , [89] . We found that the levels of both allelic and ectopic collisions were similar in spo11Δ rec8Δ ndj1Δ and spo11Δ rec8Δ ( P = 0 . 3 and P = 0 . 8 respectively ) suggesting that a persistent bouquet is not responsible for increased collision levels in the rec8Δ background ( Figure 3 ) . Interestingly a significant reduction in ectopic collisions was found in the control strain spo11Δ ndj1Δ compared to spo11Δ ( 0 . 012±0 . 004 and 0 . 017±0 . 006 respectively , P = 0 . 003; Figure 3 ) . This finding was further explored as described below . We next speculated that the high levels of nonspecific interactions observed in spo11Δ rec8Δ might be driven by actin-mediated motion . To test this , we added the actin polymerization inhibitor Latrunculin B ( Lat B ) to cell synchronized meiotic cultures and asked if it reduced the levels of allelic and ectopic collisions [90] . We were surprised to find that the level of allelic collisions was instead increased in spo11Δ rec8Δ cells treated with Lat B compared to untreated cells ( 0 . 140±0 . 010 vs . 0 . 110±0 . 020; P = 1 . 9e−5 ) . This trend was also observed for ectopic loci ( 0 . 047±0 . 008 vs . 0 . 043±0 . 012 , albeit above the threshold of significance; Figure 3 ) . This outcome suggests that actin can antagonize nonspecific interactions . There was no measurable effect of Lat B on the spo11Δ single mutant . Moreover , Lat B treatment did not significantly affect the level of allelic collisions in the pREC8-SCC1 spo11Δ mutant where sister chromatid cohesion exists ( Figure 3 ) . These results suggest that the cohesin function of Rec8 acts in opposition to an actin-based mechanism to suppress nonspecific chromosome interactions . If the elevated level of nonspecific interactions in spo11Δ rec8Δ were due to Ndj1-dependent , telomere-led motion we would expect that the addition of Lat B to spo11Δ rec8Δ ndj1Δ would have no effect . Instead , we found that Lat B elevated both allelic collisions ( 0 . 143±0 . 016 versus 0 . 095±0 . 017; P = 0 . 0001; Figure 3 ) and ectopic collisions ( 0 . 051±0 . 002 versus 0 . 037±0 . 007; P = 0 . 0002; Figure 3 ) by approximately 50% even in the absence of Ndj1 . Without the combined constraints of Ndj1-dependent attachment of chromosomes to the nuclear envelope , Rec8-mediated cohesion and an unknown feature of actin ( i . e . Lat B treated spo11Δ rec8Δ ndj1Δ ) , the total level of collisions ( i . e . the sum of allelic and ectopic events ) is 4 . 3 fold greater than when these constraints are intact ( i . e . in a spo11Δ single mutant ) . Thus , even in the absence of DSBs , multiple independent processes appear to impose chromosome order within the 3D space of the nucleus . The addition of Lat B to wild-type cells does not affect overall levels of homolog pairing assayed using FISH , however , it delays the kinetics of pairing compared to untreated cells [53] . When Lat B was added to wild type ( e . g . SPO11 ) cells the level of allelic collisions was modestly reduced to 84% of untreated cells ( P = 0 . 003; Figure 4 ) the level of ectopic collisions remained unchanged ( P = 0 . 2; Figure 4 ) , suggesting that actin can play a positive role in promoting recombination-mediated allelic interactions in addition to antagonizing nonspecific interactions ( above ) . This result is not surprising since the levels of crossover-bound recombination intermediates ( Single End Invasions ( SEIs ) and double Holliday Junctions ( dHJs ) ) are not dramatically reduced in Lat B-treated cells [62] . By contrast , we found that the addition of Lat B increased levels of allelic collisions in rec8Δ , ndj1Δ , csm4Δ , rec8Δ ndj1Δ and ndj1Δ csm4Δ compared to untreated cells ( Figure 4 ) . These results suggest that in the absence of telomere-led movement or dynamic nuclear deformations , a Lat B sensitive process can negatively influence allelic chromosome interactions . Interestingly , only rec8Δ and rec8Δ ndj1Δ mutants significantly increase ectopic collision levels and Lat B further increases these collisions ( P = 0 . 02; P = 0 . 0004 respectively , Figure 4 ) . These results are consistent with the findings shown above ( Figure 3 ) suggesting a role for Rec8 in constraining nonspecific chromosome interactions . Moreover , these data implicate actin in a nuclear process independent of Ndj1-dependent telomere-led motion . We next used an independent visual assay to measure the effect of Lat B on chromosome interactions using strains expressing a TetR-GFP fusion protein that binds integrated tetO arrays at the URA3 locus on homologous chromosomes [91]–[93] . As was also observed by Brar et al . , for wild type , the two loci colocalize forming one focus prior to transfer of cells to SPM . As cells enter meiosis , colocalization is progressively reduced up until about t = 3 hours ( Figure 5A; [49] ) . While untreated cells reach maximum levels of pairing by t = 7 hours ( ∼90% have one GFP spot; Figure 5A ) , pairing in Lat B treated cells was delayed and only ∼55% had one spot at this time point . Thus , the effect of Lat B on pairing using this visual assay was more severe than using the collision assay . Pairing kinetics in the untreated ndj1Δ mutant were delayed compared to wild type , similar to observations made using FISH [70] . This delay was even greater in Lat B-treated cells . Indeed , the kinetics of pairing in both wild type and ndj1Δ cells treated with Lat B were similar ( Figure 5A ) . These results suggest 1 ) Ndj1 does not play a role in promoting pairing other than through its actin-related function and 2 ) that actin promotes pairing of allelic sites , in part , through a process that acts independently of Ndj1 . We did not observe a recapitulation of the collision phenotype where Lat B stimulates allelic collisions . Since the effect of Lat B appears to have an opposite effect in the GFP assay compared to the collision assay in the ndj1Δ , it is apparent that they measure different aspects of meiotic chromosome dynamics . In the absence of Ndj1 , there is a considerable degree of Lat B-sensitive dynamic nuclear deformation [94] . Since dynamic nuclear deformations require the putative KASH protein , Csm4 [94] , we reasoned that addition of Lat B to csm4Δ or csm4Δ ndj1Δ would have little or no effect on the kinetics or absolute levels of pairing . This turned out not to be the case , however , since addition of Lat B reduced and/or delayed pairing levels in both mutants ( Figure 5B ) . Together , these results suggest that an actin-mediated process and/or structure positively influences homolog pairing independent of Ndj1-dependent telomere-led motion or dynamic nuclear deformations . During early meiotic prophase , chromosomes in yeast are loosely organized by centromere coupling and attachment of telomeres to the nuclear envelope . Since this configuration does not require DSB formation , one widely held notion is that this arrangement of chromosomes may precede and set the stage for DSB-mediated pairing , similar to DSB independent pairing of heterochromatin in Drosophila or pairing centers in C . elegans [14] , [35] , . We reasoned that collision levels might reflect this organization and that disruption of telomere attachment ( e . g . ndj1Δ ) and/or centromere coupling ( e . g . zip1Δ ) would reduce ectopic collision levels . To overcome the strong effects of homologous recombination , we carried out the analysis in a spo11Δ mutant background . Interestingly , ectopic collision levels were reduced in both the spo11Δ ndj1Δ and spo11Δ zip1Δ double mutants to 73% ( P = 0 . 003 ) and 75% ( P = 0 . 03 ) of spo11Δ levels , respectively ( Figure 6 ) . Notably , these are the only mutant situations for which we have observed a reduction in ectopic collisions either in the presence or absence of DSBs for dozens of analyzed mutant strains ( Figure S2; [21] , [74] , [75] ) . If the observed decrease in ectopic collision levels was due to the independent contributions of Ndj1 and Zip1 , we expected that the spo11Δ zip1Δ ndj1Δ triple mutant would give even lower collision levels compared to either spo11Δ ndj1Δ or spo11Δ zip1Δ double mutants . Interestingly , however , ectopic collision levels in double and triple-mutant strains were indistinguishable ( P>0 . 99 for both cases; Figure 6 ) . These collision levels ( ∼0 . 013 ) are two orders of magnitude above the lower limit of detection using this assay ( i . e . P<0 . 00005; see Table S1 ) , so a decrease should be detectable , in principle . These data suggest that Ndj1 and Zip1 participate in a single process that facilitates a subset of interactions between ectopic interstitial loci . For example , the simultaneous occurrence of telomere/NE tethering and centromere coupling could promote alignment of similar-sized chromosome arms , irrespective of homology . To further test this model , collisions at additional chromosomal sites must also be analyzed . To our surprise , Lat B added to spo11Δ ndj1Δ and spo11Δ zip1Δ double mutants restored ectopic collision levels to the same as spo11Δ ( 1 . 22-fold , P = 0 . 04 and 1 . 37-fold , P = 0 . 003 , respectively ) . This was also true when Lat B was added to the spo11Δ zip1Δ ndj1 triple mutant ( 1 . 33-fold , P = 0 . 01 ) . By contrast , ectopic collisions in spo11Δ were virtually the same in untreated and treated cells ( 0 . 0173±0 . 006 vs . 0 . 0154±0 . 004; P = 0 . 11 , respectively; Figure 6 light blue bars ) . Thus , there appear to be weak stabilizing forces between ectopic sites maintained in the absence of Ndj1 and/or Zip1 that are sensitive to disruption by an actin mediated process or event . We next tested if dynamic nuclear deformations are responsible for disrupting these weak ectopic interactions by introducing the csm4Δ mutation into these strains . In this case we would expect that the csm4Δ mutation would prevent destabilization and thus phenocopy the effect of Lat B treatment . This indeed turned out to be the case since ectopic collisions in spo11Δ ndj1Δ were increased to spo11Δ levels in the absence of Csm4 ( P = 0 . 94; Figure 6 ) . Importantly , addition of Lat B to spo11Δ csm4Δ and the spo11Δ ndj1Δ csm4Δ triple mutant did not increase ectopic collisions indicating that Csm4 acts in the same pathway as an actin-mediated process ( perhaps by mediating dynamic nuclear deformations ) that destabilizes weak interactions between ectopic loci . Rapid prophase movement of chromosomes is a prominent feature of mid-to-late meiotic prophase , yet little is known about the impact of chromosome motion during early meiotic prophase when chromosomes undergo pairing . We found that by eliminating one or all of three key components required for rapid prophase movement ( Ndj1 , Csm4 and actin polymerization ) the kinetics of homolog pairing was delayed , consistent with findings using FISH and one-spot/two-spot TetR-GFP assays [64] , [70] . In addition , we found that wild type and ndj1Δ cells gave indistinguishable pairing levels in the presence of the actin polymerization inhibitor Lat B , suggesting that the contribution of Ndj1 to pairing occurs entirely through its role in actin-directed chromosome movement . Conversely , we found that Lat B caused a more severe pairing delay in ndj1Δ and csm4Δ mutants compared to untreated cells , also suggesting that actin may play roles in chromosome pairing independent of Ndj1 and Csm4 ( see below ) . We suggest that actin-independent motion ( perhaps diffusion or changes in chromatin compaction ) is sufficient for allowing chromosome pairing , but that the process is accelerated when chromosomes are actively moving . In C . elegans , where pairing does not rely on DSB repair , homolog pairing appears to be driven by a combination of dynein-driven motion and diffusion that initiates at pairing centers [97]; while the loss of active pairing center motion leads to pairing delays , diffusion-based motion is sufficient for pairing [97] , [98] . In S . pombe , mutations that disrupt dynein-driven nuclear movement of chromosomes also decrease the efficiency of the pairing process [20] , [99]–[101] . We suggest an analogous situation occurs in budding yeast except that actin-mediated forces are involved . Some recent studies have drawn similar conclusions [102] , [103] . While the outcomes of our one-spot/two-spot TetR-GFP visual assay indicate a positive role for actin in pairing independent of Ndj1/Csm4 , outcomes from the collision assay indicated that actin might also prevent or destabilize nonspecific interactions . That is , in the absence of Ndj1 and/or Csm4 , we observed that Lat B increased allelic collisions yet slowed the process of pairing . One way to reconcile these two observations is that spurious or nonproductive interactions may be destabilized by actin-mediated mechanism not related to Ndj1/Csm4-dependent motion . While the nature of this mechanism is not clear , one possibility is that interstitial chromosomal sites are subject to motion via the actin cables that surround the nucleus using a KASH protein complex other than Csm4 ( and Ndj1 ) [62] . Alternatively , interaction between interstitial chromosome sites might be prevented by sequestering them in different nuclear compartments and/or the nuclear envelope by association with an actin-associated structure or nuclear localized actin [104] . In interphase cells of yeast and Drosophila , diffusive motion of chromosomes is constrained to a limited subregion of the nucleus and treatment with the microtubule depolymerizing agent nocodazole alleviates this confinement [105] . Perhaps , during yeast meiosis an actin-mediated process acts similarly to constrain interstitial chromosome loci to enhance pairing . Additionally , several nuclear processes including chromatin remodeling have been shown to require actin and/or sensitivity to Lat B [106]–[108] . Chromatin remodeling may be necessary for stable pairing of loci but may not be essential for their collision . We can envision a scenario where interstitial chromosome sites are coupled to one or more actin-based assemblies and their adjacent telomeres are attached to cytoplasmic actin cables via the Ndj1/Mps3/Csm4 protein bridge ( Figure 7; Figure S4 ) . The two systems acting simultaneously could direct discordant movement between chromosomes such that strong interactions persist while weak interactions are taken apart ( Figure 7 ) . Over time , chromosomes would be subject to alternating scrunching and stretching , perhaps analogous to a coupled-spring oscillator . Initially , chromosomes might undergo oscillations independent of one another , increasing the likelihood of productive strand invasion events to promote close , stable homolog juxtaposition . When pairing has been mostly achieved by zygotene [59] , rapid prophase movement could serve to remove chromosome interlocks [60] . One of the most surprising findings in our study was the high level of allelic and ectopic chromosome interactions observed in the absence of Rec8 , even in a spo11Δ background . To better understand this result , mechanistic insight may be gained by considering the function of “SMCs” ( cohesins and condensins ) in distributing spindle-pulling forces across the pericentromeric chromatin loops during mitosis [109] . Bloom and colleagues describe SMCs as having the physical attributes of “slip rings” ( molecular pulleys ) that impart the distribution of tension and regulate elasticity of these pericentromeric loops [110] . Analogously , Rec8 may distribute tension along or within the loop-axis structure of meiotic chromosomes as they are pulled by actin-driven motors , or even subjected to thermal motion ( Figure S4 ) . In the absence of Rec8 , transduction of actin-mediated forces along chromosome segments would be diminished , as would their spring-like properties required for promoting allelic and taking apart ectopic interactions . Indeed , rec8Δ mutants in S . pombe exhibit defects in both chromosome compaction and pairing [111] , [112] . Our observation that addition of Lat B increases nonspecific interactions in rec8Δ , and even more so in a rec8Δ ndj1Δ double mutant , suggests independent contributions of actin , cohesin and telomere function in promoting and limiting chromosome interactions during meiosis . All yeast strains are isogenic derivatives of SK1 ( Table S3 ) [113] . Parental haploid strains SBY1338 ( MAT a ho::hisG lys2 ura3Δ::hisG leu2::hisG ade2Δ::hisG trp1::hisG GAL3 flo8::LEU2-loxP-ura3 ndt80Δ::LEU2-loxP-ade2 ) and SBY1448 ( MAT alpha ho::hisG lys2::GAL1-Cre-LYS2 ura3Δ::hisG leu2::hisG ade2Δ:hisG leu2::hisG ade2re-LYS2 ura3-loxP-ura3 ndt80Δ::LEU2 ) were used for transformation to generate PCR-mediated knockouts [21] . Knockout mutations in SBY1338 and SBY1438 were generated by transformation using PCR-based disruption that replaced the entire open reading frame with the kanMX4 , natMX , or hphMX4 marker [114] , [115] . Integration of the drug-resistant markers into the appropriate genomic location and loss of wild-type markers were confirmed by PCR for every knockout strain created . The cdc6-mn ( meiotic null ) was generated by replacing the endogenous promoter of CDC6 with the promoter of SCC1 , which is down regulated during meiosis . The sgs1-mn allele was generated by placing SGS1 under the control of the CLB2 promoter [116] . The rec8Δ::pREC8-SCC1 construct allows for expression of Scc1 instead of Rec8 by placing SCC1 under the control of the Rec8 promoter [84] . Diploid strains carry an allelic pair of loxP sites on chromosome V ( replacing FLO8; coordinates 377614 to 375215 ) and an ectopic loxP site on chromosome VIII ( replacing NDT80; coordinates 356561 to 358444 ) . Both chromosomes are ∼580 kb in length with centromeres located at ∼110 and ∼150 kb from the right telomere , respectively . The loxP sites are integrated in the left arm at sites roughly equidistant from their adjacent centromeres ( CEN5-FLO8 is ∼230 kb; CEN8-NDT80 is ∼250 kb ) and their adjacent telomeres ( ∼200 kb for both intervals ) . In both cases , the sites are located in a region of the genome that is unremarkable for DSB distribution [85] , [87] . Media preparation and meiotic cell culture synchronization was performed as previously described [117] . Galactose was added to final concentration of 0 . 03% at one hour after transfer to sporulation media ( SPM ) to induce expression of Cre-recombinase . At t = 2 hrs . after transfer to SPM , cells were either treated with 0 . 1% DMSO or 30 uM Latrunculin B dissolved in DMSO . Genomic DNA for qPCR standard curves was isolated from haploids SBY 2576 ( ho::hisG lys2-pGAL1-Cre-LYS2 ura3Δ::hisG leu2::hisG ade2Δ::hisG trp1::hisG GAL3 flo8::LEU2-pGPD1-ura3 ) for the allelic Cre/LoxP recombinant and SBY 2575 ( ho::hisG lys2-pGAL1-Cre-LYS2 ura3Δ::hisG leu2::hisG ade2Δ::hisG trp1::hisG GAL3 flo8::LEU2-pGPD1-loxP-ade2 ) for the ectopic Cre/loxP recombinant . Cells were harvested 10 hours after transfer to SPM ( t = 10 hrs . ) for DNA extraction ( unless otherwise noted ) . DNA purification was performed by vortexing cells in the presence of 0 . 5 mm zirconia/silica beads ( BioSpec Products , Inc . ) and phenol/chloroform , followed by ethanol precipitation of the DNA . DNA from haploid strains containing only the recombinant was serially diluted to make standard curves for the corresponding primer set and ACT1 . The following are the sequence of the primer used: Allelic and Ectopic Forward primer: 5′-CCAAGAACTTAGTTTCGACGGATC-3″ Allelic Reverse primer: 5′-TCGACATGATTTATCTTCGTTTCC-3′ Ectopic Reverse primer: 5′-CAATTGTCCCCCTCCTAATATACCA-3′ ACT1 Forward primer: 5-AATGCAAACCGCTGCTCAAT-3′ ACT1 Reverse primer: 5′-CAAAGCTTCTGGGGCTCTGA-3′ Primers for ACT1 reaction are used at 100 nM concentration . Primers for detection of the ectopic recombinant are used at 500 nM concentration . For detection of the allelic recombinant , the forward primer was used at 500 nM and the reverse primer was used at 900 nM . Quantitative PCR was performed on the ABI 7300 using SYBR Green Power master mix ( ABI ) . The cycling conditions are as follows: 95° for 10 min . Followed by 40 cycles of 95° 15 sec and 60° 1 min . Collisions in all strains except for strains containing the cdc-6mn mutation were calculated as 4 times the recombinant copy number divided by the copy number of ACT1 to yield number of recombinants per four chromatids . Due to the absence of meiotic replication in cdc-6mn mutants , collisions in cdc-6mn mutants were calculated as 2 times the recombinant copy number divided by the copy number of ACT1 to yield number of recombinants for the two chromatids of the unreplicated homolog pair . Cells were synchronized and Lat B was added as described above except that galactose was not added to the cultures . Cells were removed ( 250 ul ) every hour , fixed in 4% paraformaldehyde ( PFA ) in phosphate buffered saline ( PBS; pH 7 . 4 ) for 8 minutes at room temperature . Cells were then washed once with PBS and stored at 4°C until they could be analyzed . Cell morphology and pairing levels were the same in unfixed and in fixed cells for up to one week ( data not shown ) . A monolayer of cells on a slide was prepared according to [118] and cells were immediately imaged at 100× magnification using a hybrid spinning disk confocal microscope ( Intelligent Imaging Innovations ) with a 488 nm laser for 150 msec exposure time per slice . Pairing was assessed visually in projected Z-stacks by determining the fraction of cells containing one GFP spot . Each Z-stack consisted of ∼30 slices with 0 . 25 µm separating each slice . Strains used for visualizing homolog pairing were: SBY4503×SBY4504 ( MAT a/MAT alpha ho::hisG/" LEU2::tetR-GFP/" URA3::tetOx224/" his3::hisG/" ndt80Δ::NAT/" ) SBY4506×SBY4507 ( MAT a/MAT alpha ho::hisG/" LEU2::tetR-GFP/" URA3::tetOx224/" his3::hisG/" ndt80Δ::NAT/" ndj1Δ::Hph/" ) SBY4870×SBY4871 ( MAT a/MAT alpha ho::hisG/" LEU2::tetR-GFP/" URA3::tetOx224/" his3::hisG/" ndt80Δ::NAT/" csm4Δ::Hph/" ) SBY4872×SBY4873 ( MAT a/MAT alpha ho::hisG/" LEU2::tetR-GFP/" URA3::tetOx224/" his3::hisG/" ndt80D::NAT/" ndj1Δ::Hph/" csm4Δ::Hph/" ) . Cells were harvested 10 hours after transfer to SPM ( unless otherwise noted ) . Cell aliquots were pelleted , resuspended in 2% glucose , sonicated 5 seconds at 15% maximum power using the microtip of a 550 Sonic ZD-dismembrator ( Fisher Scientific ) , and diluted appropriately prior to plating on selective ( SC-Ura ) and nonselective media ( YPD-Ade ) . A two-tailed Student's t-test was performed for determining the P-value between treated and untreated cultures . All bar plots signify the mean ± standard deviation of the mean for measured collision levels ( above ) . The total number of independent cultures for all strains is listed in Table S2 . Heatmaps indicating the P-values for comparison of values across multiple strains was obtained from applying a two-tailed Student's t-test followed by Sidak correction where P = 1− ( 1−α ) 1/n .	Meiosis is the key stage of gametogenesis , when the diploid genome complement is reduced by one half to form haploid gametes for sexual reproduction . Accurate chromosome segregation requires that homologous chromosomes pair , recombine by crossing over , and segregate from one another during the first meiotic division . Missegregation of homologs leads to the formation of aneuploid gametes , while erroneous crossing over between ectopic chromosomal loci can lead to chromosomal rearrangements such as translocations and deletions . We found that nonspecific interactions between interstitial chromosomal sites can be enabled or prevented through multiple , independent mechanisms during meiosis in budding yeast . These include organization of chromosomes in the nucleus , integrity of the chromosome axis structure , and actin-led chromosome movement . Acting together , these processes can reinforce strong chromosome interactions that promote pairing , while acting in opposition they can eliminate weak nonspecific interactions . These data provide an integrated view of how homologous chromosome pairing is achieved .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "model", "organisms", "genetics", "biology", "molecular", "cell", "biology", "genetics", "and", "genomics" ]	2013	Multiple Opposing Constraints Govern Chromosome Interactions during Meiosis
Mammalian oocytes are arrested at prophase of the first meiotic division in the primordial follicle pool for months , even years , after birth depending on species , and only a limited number of oocytes resume meiosis , complete maturation , and ovulate with each reproductive cycle . We recently reported that protein phosphatase 6 ( PP6 ) , a member of the PP2A-like subfamily , which accounts for cellular serine/threonine phosphatase activity , functions in completing the second meiosis . Here , we generated mutant mice with a specific deletion of Ppp6c in oocytes from the primordial follicle stage by crossing Ppp6cF/F mice with Gdf9-Cre mice and found that Ppp6cF/F; GCre+ mice are infertile . Depletion of PP6c caused folliculogenesis defects and germ cell loss independent of the traditional AKT/mTOR pathway , but due to persistent phosphorylation of H2AX ( a marker of double strand breaks ) , increased susceptibility to DNA damage and defective DNA repair , which led to massive oocyte elimination and eventually premature ovarian failure ( POF ) . Our findings uncover an important role for PP6 as an indispensable guardian of genomic integrity of the lengthy prophase I oocyte arrest , maintenance of primordial follicle pool , and thus female fertility . In mammals , females are born with a finite number of oocytes contained within primordial follicles that serve as the source of ova for the entire period of reproductive life . To produce mature eggs , dormant primordial follicles are recruited into the growing follicle pool , a process termed as initial follicular recruitment or activation . Activated follicles subsequently develop into primary follicles , secondary follicles , and antral follicles [1] . Throughout this follicular growth process , oocytes grow while being arrested in prophase of meiosis I with homologs held together by chiasmata . Only a few dominant antral follicles reach the preovulatory stage and release a mature egg for fertilization after a gonadotropin surge during each estrus cycle [2] . When the ovarian follicle reserve is exhausted in women , menopause occurs . However , disorders during folliculogenesis could lead to follicle depletion in advance and cause premature ovarian insufficiency ( POI ) or premature ovarian failure ( POF ) , which is a main cause of female infertility in humans and affects nearly 1% women under the age of 40 [3] . Protein phosphorylation , mediated by a conserved cohort of protein kinases and phosphatases , regulate follicular activation and growth , meiotic cell cycle arrest and progression , chromosome dynamics , and ovulation [4] . Numerous studies using genetically modified mice reveal that protein kinases play important roles during folliculogenesis/oogenesis . For example , the PTEN/PI3K/AKT signaling pathway regulates follicular activation and survival [5] . Recently , we reported that LKB1 acts as a gatekeeper of the ovarian primordial follicle pool [6] . In contrast , there is limited information about the roles of protein phosphatases . Among the serine/threonine phosphoprotein phosphatases ( PPPs ) , PP2A , PP4 and PP6 form a subfamily called PP2A-like protein phosphatases , which share high homology in the catalytic subunit and account for the majority of cellular serine/threonine phosphatase activity [7 , 8] . PP2A is involved in regulating chromosome condensation , DNA damage repair , G2/M transition and sister chromatid cohesion [9] . Our recent knockout mouse model revealed that oocyte PP2A is dispensable for folliculogenesis , though PP2A has been reported to dephosphorylate AKT and AMPK , important kinases for folliculogenesis [10] . Although PP6 was discovered almost 20 years ago , progress has been slow regarding its functions in cells , not to mention its specific functions in meiotic cells . The PP6 holoenzyme consists of a catalytic subunit , PP6c , one of the three regulatory subunits including SAPS1 , 2 , 3 ( also known as PP6R1 , PP6R2 and PP6R3 , respectively ) , and one of the three ankyrin repeat subunits including ARS-A , -B , -C [11 , 12] . PP6 is conserved among all eukaryotic species from yeast to humans , attesting to its fundamental importance . Mutations in PP6c are found to exist in 9–12 . 4% melanomas surveyed and may act as drivers for melanoma development [13 , 14] . The PP6 yeast homologue , Sit4/Ppe1 , is required for G1/S progression and equal chromosome segregation [15 , 16] , and plays a role in signaling through the target of rapamycin ( TOR ) , a key nutrient-sensing kinase [17] . Human PP6 has an established role in DNA damage response with its ability to modulate signaling by DNA-dependent protein kinase ( DNA-PK ) , homology recombination-mediated repair of DNA double strand breaks ( DSBs ) [18 , 19] , as well as its interactions with Aurora A kinase [20 , 21] . More recent studies suggest a broader role for PP6 in pre-mRNA splicing [22] , control of apoptosis in immune cells [23] , formation of adherens junctions through interaction with E-cadherin [24] , and modulation of signaling through the Hippo pathway [25] . Overall , these data suggest that PP6 integrates signaling from multiple pathways . Genetically modified mouse models are powerful tools for studying gene function in vivo [26 , 27] . We recently reported that a conditional knockout of PP6 in oocytes from growing follicles ( by crossing Ppp6cF/F mice with Zp3-Cre mice ) causes female subfertility by disrupting MII spindle organization and MII completion after fertilization [28] . Here , we crossed Ppp6cF/F mice with Gdf9-Cre mice to generate mutant mice with specific deletion of Ppp6c in prophase I-arrested oocytes from the primordial follicle stage . We find that PP6 plays a critical role in germ cell survival and follicular development by safeguarding genomic integrity of prophase I-arrested oocytes . To explore the in vivo roles of PP6 during folliculogenesis/oogenesis , we generated mutant mice ( referred to as Ppp6cF/F;GCre+ mice ) , in which exon II-IV of the Ppp6c gene were targeted , by crossing Ppp6cF/F mice [28] with transgenic mice expressing Gdf9 promoter-mediated Cre recombinase [5] ( Fig 1A ) . In Gdf9-Cre mice , Cre is specifically expressed in oocytes of primordial follicles and later stage follicles since postnatal day 3 [27] . By immunoblotting analysis , we confirmed successful depletion of PP6c protein in GV oocytes from Ppp6cF/F;GCre+ females ( Fig 1B ) . To investigate the effect of oocyte-specific knockout of PP6c on female fertility , a breeding assay was carried out by mating Ppp6cF/F or Ppp6cF/F;GCre+ female mice with males of proven fertility for 6 months . As shown in Fig 1C , female Ppp6cF/F;GCre+ mice were completely infertile . The infertility appeared to be due to anovulation in adult mutant mice , whereas control mice ovulated normal numbers of eggs ( 8 . 7±0 . 8 ) in the natural ovulation assays ( Fig 1D ) . To understand the defects of the mutant mice , we first observed the morphology of ovaries from both Ppp6cF/F and Ppp6cF/F;GCre+ mice . At 1 month-of-age , both histological morphology of ovaries and numbers of follicles were similar between Ppp6cF/F and Ppp6cF/F;GCre+ ovaries ( S1A and S1B Fig ) , indicating that comparable numbers of follicles are formed in the wild-type and mutant ovaries . However , after 1 month-of-age , the time of onset of sexual maturity , mutant ovaries started to show differences and became smaller than the controls . In ovaries of 2-month-old Ppp6cF/F;GCre+ mice , there were few growing follicles ( Fig 2B ) in contrast to control ovaries that contained many healthy-looking growing follicles ( Fig 2A ) . However , large clusters of primordial follicles could still be observed on the ovarian surface area of 2-month-old mutant ovaries ( white arrowheads , Fig 2C and 2C’ ) , compared with control ovaries where such clusters barely could be found . Consistently , the number of primordial follicles in 2-month-old Ppp6cF/F;GCre+ ovaries was more than double that of Ppp6cF/F ovaries ( Fig 2M and S1C Fig ) . The numbers of large growing follicles , especially Type 5 and Type 6 follicles , were significantly decreased , corresponding to only 16 . 7% and 36 . 5% of those in control ovaries ( S1C Fig ) . At 3 months-of-age , clusters of primordial follicles were no longer observed on the ovarian surface; instead , many primary follicles appeared in the same location , indicating delayed activation of the arrested primordial follicles ( yellow arrows , Fig 2F and 2F’ ) . Consistent with these observations , quantification of ovarian follicles revealed a significant reduction of primordial follicles and an increase in type 3 and type 4 follicles in 3-month-old mutant ovaries ( S1D Fig ) . Nevertheless , large growing follicles ( including type 5 and type 6 follicles ) were significantly fewer than those in control ovaries ( S1D Fig ) , though both control and mutant ovaries might contain similar numbers of primordial follicles and activated follicles ( Fig 2M and 2N ) . These later activated follicles , however , could not serve as the source of ova for mutant mice , probably because they died soon after activation with only empty follicle-like structures left at the ovarian surface ( yellow arrowheads , Fig 2I , 2I’ , 2L and 2L’ ) . At 4 months-of-age , only a few primary follicles and small secondary follicles were seen at the cortical region of mutant ovaries ( Fig 2H ) , and the other types of follicles ( including primordial follicles , type 5 , 6 and 7 follicles ) were disappearing ( S1E Fig ) . By 6 months-of-age , almost all types of follicles were depleted in Ppp6cF/F;GCre+ ovaries ( Fig 2K and 2L; S1F Fig ) , which is termed POF . In general , from 1 month to 2 months postpartum , more than half of the primordial follicles in the Ppp6cF/F ovaries decreased due to both follicular activation and atresia . In contrast , loss of primordial follicles in Ppp6cF/F;GCre+ ovaries was slower because they failed to be activated upon puberty and stayed arrested until 2 months postpartum , after which time they were rapidly eliminated either through death following delayed activation or degeneration ( Fig 2M ) . Activated follicles in Ppp6cF/F;GCre+ ovaries only survived for a short time and none could develop to the preovulatory stage ( Fig 2N ) . The histological analysis suggested that absence of PP6c in oocytes caused defects in follicular activation and growth . To confirm these observations , we performed immunostaining of the germ cell marker MVH ( mouse VASA homolog ) on 2-month-old ovarian sections . As shown in S2A Fig , in normal control ovaries , primordial follicles were mostly scattered around the cortical region whereas in ovaries of adult Ppp6cF/F;GCre+ mice a significant number of primordial follicles remained in clusters ( S2B Fig ) , indicating abnormal development of primordial follicles . This finding confirmed that the natural incidence of follicular activation after puberty was disrupted by Gdf9-Cre mediated Ppp6c deletion . At 2 months-of-age , although Ppp6cF/F;GCre+ mice still had a large number of growing follicles , these follicles failed to mature and ovulate . As shown by TUNEL assay on ovarian sections , increased granulosa cell apoptosis and follicle atresia ( yellow arrowheads , S2D Fig ) were detected in ovaries of 2-month-old Ppp6cF/F;GCre+ mice compared to ovaries in control mice ( S2C Fig ) . Furthermore , when we tried to stimulate follicle growth with exogenous PMSG , the mutant mice still could not respond normally because almost all the antral follicles initiated atresia by premature luteinization and formed numerous atretic corpora lutea ( CLs ) ( yellow arrowheads , S2F Fig ) instead of developing into preovulatory follicles ( red asterisks , S2E Fig ) . The above data demonstrated that defective follicular development after puberty , including blocked primordial follicle activation and compromised growth of activated follicles , accounted for the infertility of Ppp6cF/F;GCre+ mice . mTOR signaling is essential for oocyte survival and awakening from dormancy within primordial follicles [5 , 29 , 30] . Considering an analogous involvement of PP6 in TOR signaling in yeast and plants [17 , 31] , it is possible that PP6 maintains oocyte survival by regulating the mTOR pathway . Accordingly , we performed immunoblotting analysis with PD35 GV oocytes . Surprisingly , the activity of the AKT/mTORC1/S6K signaling pathway was significantly enhanced , as indicated by elevated levels of phosphorylated AKT ( S473 ) , phosphorylated mTOR ( S2448 ) , phosphorylated S6K ( T389 ) in Ppp6cF/F;GCre+ oocytes ( Fig 3A and S3B Fig ) ; phosphorylated rpS6 ( S240/244 ) did not show obvious changes ( Fig 3A and S3B Fig ) . This finding was not consistent with our phenotypes based on previous reports because enhanced AKT/mTOR signaling is responsible for the over-activation of primordial follicles in Pten and Tsc1/2 mutant mouse models . In contrast , our Ppp6c mutant mice did not show any signs of premature activation of the entire primordial follicle pool , instead showing blockage/delay of follicular activation , although the activity of AKT/mTOR signaling was higher than in controls . Thus , up-regulation of the AKT/mTOR pathway could result from feedback effects to defective oocyte growth or local effects attributed to PP6 in regulating mTOR activity as suggested for other organisms [17 , 31] . Recently , we reported that Lkb1fl/fl; Gdf9-Cre mice exhibit over-activation of primordial follicles starting from the onset of sexual maturity and defective follicle growth at later stages . The phenotypes of Lkb1fl/fl; Gdf9-Cre mice appear opposite to those of Ppp6cF/F;GCre+ mice . Accordingly , examined the activity of AMPK , the main substrate of LKB1 , in Ppp6cF/F;GCre+ oocytes and observed that the level of phosphorylated AMPK ( T172 ) was significantly increased ( Fig 3B and S3B Fig ) ; it is decreased in Lkb1 mutant oocytes . To ascertain whether PP6 interacts with the AMPK pathway we generated double knockout mice for both Lkb1 and Ppp6c ( Lkb1F/F;Ppp6cF/F;GCre+ ) . As expected , Lkb1 deletion within a Ppp6c deletion background rescued the blockage of primordial follicle activation at 2 months-of-age ( Fig 3E ) . Unanticipated was that double knockout ovaries resembled Lkb1 mutant ovaries ( Fig 3C ) by exhibiting large sizes and over-activation of primordial follicles at 2 months-of-age . One difference , however , was that growth of activated follicles in double knockout ovaries was slower with secondary follicles containing unhealthy oocytes and showing apoptotic signals indicating extensive follicle atresia at 2 months-of-age ( yellow arrowheads , Fig 3E ) , which is similar to 2-month-old Ppp6cF/F;GCre+ ovaries ( Fig 3D ) . In contrast , most activated follicles reached the antral follicle stage in Lkb1 mutant ovaries at the same age ( Fig 3C ) . These results showed that knockout of Lkb1 could partially rescue the follicle development phenotype of the PP6c mutant ovaries , which strongly suggested involvement of AMPK in follicle development and PP6c participating in regulating the AMPK pathway . However , knockout of Lkb1 did not rescue Ppp6cF/F;GCre+ oocytes from death , suggesting that PP6 might not only regulate primordial follicle activation but also maintain survival of oocytes within primordial follicles . Therefore , we concluded that there were additional reasons for the PP6c mutant phenotype and pursued this possibility as described below . Because PP6 is involved in the DNA damage response via its ability to dephosphorylate γH2AX and antagonize DNA-dependent protein kinase ( DNA-PK ) [18 , 19] and unrepaired meiotic or induced DNA double-strand breaks ( DSBs ) could cause oocyte elimination and female infertility by triggering DNA damage response pathway [32] , we wondered if loss of PP6c leads to DNA damage in our case . Thus , we collected oocytes from PD35 ovaries and performed western blot analysis . As shown in Fig 4A and S3 Fig , the levels of γH2AX were significantly elevated in mutant oocytes indicating accumulated DSBs . However , the DNA damage response pathway was significantly reduced because the activity of CHK1/2-p53 signaling cascade was much lower than in controls ( Fig 4A and S3 Fig ) . We also confirmed accumulation of γH2AX in small oocytes by immunofluorescence analysis . As indicated in Fig 4B , mutant ovaries contained more and higher nuclear signals of γH2AX within primordial follicles ( yellow arrows ) when compared to controls ( white arrows ) . Oocyte maturation in vitro of mutant oocytes was also compromised . As shown in Fig 4C , the incidence of GVBD ( 56 . 7±7 . 9% ) and PBE ( 48 . 7±14 . 2% ) were lower than controls ( 76 . 6±2 . 0%; 84 . 2±2 . 6% , respectively ) . Moreover , after 8 h of in vitro maturation of Ppp6cF/F;GCre+ oocytes spindles were disorganized with scattered chromosomes , in contrast to the well-organized MI spindles with chromosomes all aligned at the equatorial plate in Ppp6cF/F oocytes . Even after 13 h of in vitro culture , when control oocytes had extruded the first polar body , most mutant oocytes still showed defective spindle organization and aberrant chromosome alignment , and could not complete meiosis I successfully . Taken together , these data demonstrate that loss of PP6c resulted in DSBs accumulation and severely impaired oocyte quality but deactivated the DNA damage response pathway in oocytes until puberty , which could explain why primordial follicle activation is delayed and mutant oocytes are damaged but still survive until 2 months postpartum . Depleting PP6c sensitizes cells to induced DNA damage [19 , 33] . Thus , we speculated that this susceptibility also exists in our PP6c-deficient oocytes and leads to eventual oocyte elimination . Because endogenous DNA damage might be low and long-term , mutant oocytes would wait for repair first and then die slowly within 6 months postpartum . Experimentally increasing DNA damage in mutant oocytes could therefore trigger more rapid apoptosis and accelerate oocyte elimination if PP6c-deficient oocytes are defective in mounting a DNA damage response . To test this proposal , zeocin was used to induce DSBs in vivo by intraperitoneal injection . Forty mg of zeocin was injected per mouse once every day for 5 days [34] after which the mice were allowed 5 days of recovery and then sacrificed around 2 months postpartum . Typically , Ppp6cF/F;GCre+ ovaries were smaller than the Ppp6cF/F ones at 2 months-of-age; however , after zeocin treatment , mutant ovaries were even smaller whereas the size of control ovaries was similar to untreated ones ( Fig 5A ) , suggesting that oocyte elimination was faster after zeocin treatment . This conclusion was confirmed by histological analysis of ovaries and follicle counting . After zeocin treatment , the number of primordial follicles in mutant ovaries ( ~57% ) as well as the number of activated follicles ( ~44% ) decreased dramatically compared to those of untreated mutant ovaries; in control groups , treated ovaries also showed fewer numbers of follicles compared to untreated ones , but these changes were not significant ( Fig 5B ) . Consistently , as shown in Fig 5C , treated mutant ovaries contained more atretic follicles ( yellow arrows ) , with many primordial follicles devoid of oocytes ( yellow arrowheads ) , whereas control ovaries had plenty of healthy-looking growing follicles ( white arrows ) and primordial follicles ( white arrowheads ) . The above observations showed that in response to induced DNA damage , Ppp6cF/F;GCre+ ovaries showed no primordial follicle arrest but oocyte death and follicle depletion , indicating that PP6c-deficient oocytes were more sensitive to DNA damage . To investigate the molecular causes for the results described above , we performed in vitro zeocin treatment in PD35 GV oocytes . GV oocytes were treated with zeocin ( 200 μg/ml for 1 h ) , then washed and cultured in M2 medium containing 2 . 5 μM milrinone overnight for recovery . These GV oocytes were collected for western blot analysis . As shown in Fig 6A and S4 Fig , in comparison to Ppp6cF/F oocytes after treatment , Ppp6cF/F;GCre+ oocytes showed lower levels of γH2AX but a highly active CHK1/2-dependent DNA damage checkpoint response , with p53-induced cell apoptosis . We also performed in vivo zeocin treatment in young mice ( 5 days of zeocin injection and 5 days of recovery ) and collected ovaries for western blot at ~PD35 when mutant ovaries still had similar numbers of follicles as controls ( Fig 6B and S4 Fig ) . The levels of MVH , a marker of germ cells , were similar in both groups indicating mutant ovaries still contained comparable numbers of oocytes to controls . Mutant ovaries , however , showed an enhanced CHK2-p53 DNA damage response pathway activity , suggesting PP6c-deficient oocytes could not repair induced DNA damage and would die eventually . Based on the above results , the main cause for the PP6c depletion phenotype appeared to be an increased susceptibility to DNA damage of PP6c-deficient oocytes . Collectively , these findings support the notion that PP6 is a critical regulator for oocyte survival and follicle development by restraining phosphorylation of H2AX to normal levels and participating in AMPK pathway regulation . In female reproduction , production of high quality eggs requires both successful follicular development and precise completion of oocyte meiosis . Previously , we studied the roles of PP6c in meiosis completion by crossing Ppp6cF/F mice with Zp3-Cre mice . By crossing Ppp6cF/F mice with Gdf9-Cre mice to generate mutant mice with a specific deletion of Ppp6c in oocytes from the primordial follicle stage we were able to investigate the roles of PP6c in follicular development . We find that Ppp6c mutant female mice show defective folliculogenesis and are infertile . Importantly , PP6c depletion caused persistent phosphorylation of H2AX . Thus , susceptibility to DNA damage and defective DNA repair mechanisms turned out to be the main underlying causes for the observed infertility . In addition , PP6c may control follicular activation by regulating the AMPK pathway . During embryonic development , primordial germ cells in female mammals enter meiosis I and finish a crucial process called synapsis that requires homologous recombination ( HR ) , a high-fidelity DNA double-strand break ( DSB ) repair process . Aberrant homolog synapsis or DSB repair triggers checkpoints that eliminate defective meiotic oocytes [35–37] . Loss of oocytes defective in DSB repair occurs soon after birth , which is controlled by the DNA damage checkpoint including the CHK2-p53/p63 pathway [32] . Oocytes are subsequently arrested at the dictyate stage of prophase I in the form of dormant oocytes enclosed in primordial follicles [38] . Such prophase I arrest usually takes weeks or months , or even longer in mice , and after primordial follicular activation , undergo a prolonged period of follicular growth before meiosis resumption and ovulation [39 , 40] . The lengthy dormancy and growth of oocytes makes maintenance of genomic integrity during follicular development more challenging and important for generating healthy gametes . However , the underlying molecular mechanisms to protect genomic DNA after embryonic HR and DSB repair remained undiscovered . The DNA damage checkpoint usually acts around the time oocytes enter meiotic arrest but presumably persists , because resting primordial follicles are highly sensitive to ionizing radiation ( IR ) [41] . In our study , oocyte-specific knockout of PP6c from primordial follicle stages results in increased γH2AX in arrested oocytes and the whole germ cell pool is then progressively eliminated by DNA damage checkpoint pathway within 6 months postpartum . These findings make PP6 a competitive candidate for safeguarding genomic DNA integrity of female germ cells during the long prophase I arrest . As noted above , PP6 is implicated in the cell response to DNA damage . The phosphorylated form of H2AX on S139 ( γH2AX ) is a marker of DSBs . PP6c exhibits phosphatase activity against γH2AX in in vitro phosphatase assays . In human cancer lines , depletion of PP6c or PP6R2 leads to persistent high levels of γH2AX after DNA damage and defective homology-directed repair ( HDR ) [19] . PP6c is recruited to DSB sites by DNA-PK , and PP6 is also required for efficient activation of DNA-PK , which is essential for non-homologous end joining ( NHEJ ) -mediated repair of DSBs [18 , 42] . A recent study also showed that Ppp6c-deficient mouse keratinocytes exhibit a high frequency of both p53- and γH2AX-positive cells , suggestive of DNA damage , as well as up-regulated expression of p53 , PUMA , BAX , and cleaved caspase-3 proteins following UVB irradiation [33] . Our in vivo data show that absence of PP6c also leads to higher levels of γH2AX ( Fig 4 ) and defective DNA repair in oocytes , especially massive oocyte death after induced DNA damage ( Figs 5 and 6 ) , suggesting that PP6 has a conserved role in DNA damage response , which is essential for gamete production and fertility maintenance . As members of the well-known PP2A-like subfamily , PP6 shares common features with PP2A or PP4 . As phosphatases , they all are involved in a diverse set of biological pathways due to their wide range of substrates . Until now , PP6 was implicated in regulation of DNA damage response , cell cycle progression , apoptosis , pre-mRNA splicing , signaling through the mTOR pathway and Hippo pathway , and others [19 , 22 , 23 , 25 , 28 , 33] . Among these multiple functions , mTOR pathway regulation was first considered as the potential cause of the phenotype in our study . mTOR signaling regulates follicular activation and oocyte survival because oocyte-specific deletion of its upstream genes , Pten or Tsc1/2 , lead to premature activation of the entire primordial follicle pool , resulting in POF due to enhanced mTORC1-S6K-rpS6 signaling [5 , 29 , 30] . Although PP6c-deficient oocytes also show similar enhanced AKT/mTOR signaling , the ovarian phenotype of Ppp6c mutant mice is not similar at all , because PP6c mutant ovaries show blocked/delayed follicular activation instead of premature activation , also at later time points . Although the AKT/mTOR pathway is activated in PP6c mutant ovaries , primordial follicles are not activated , perhaps because the downstream effectors of mTOR pathway are not responding . As seen from the western blot results ( Fig 3A ) , the activities of the AKT/mTORC1/S6K signaling are significantly enhanced in Ppp6cF/F;GCre+ oocytes , but as the downstream effector that enhances protein translation , rpS6 does not show an obvious change of activity . Thus , the effects of AKT/mTOR pathway activation are somehow blocked at the execution phase and therefore do not activate primordial follicles in mutant ovaries . In light of these findings , we turned to another important folliculogenesis regulator , the LKB1-AMPK pathway . Our previous study reported that Lkb1 mutant female mice show over-activation of primordial follicles after puberty [6] , at a similar time point as that of Ppp6c mutant mice . Because Western blot results also show up-regulated p-AMPK in PP6c-deficient oocytes , opposite to that in LKB1-deficient oocytes , we generated double knockout of Lkb1 and Ppp6c in oocytes to try to rescue the phenotypes of Ppp6c mutant mice . Indeed , the blocked/delayed follicular activation was rescued , which means mis-regulation of AMPK pathway could be a partial reason for the PP6c mutant phenotypes . Moreover , the double knockout ovaries show over-activation of primordial follicles , more similar to Lkb1 single knockout , but accelerated oocyte death and slower follicle growth , suggesting that absence of PP6c might affect oocyte quality and survival more directly than just control follicular activation . Thus , PP6’s role in DNA damage response could be the main cause . Consistent with this proposal is that PP6c depletion caused increased γH2AX , a marker of DSBs , and defective DNA repair in oocytes , with accelerated oocyte death with induced DNA damage . Interestingly , oocyte defects resulting from PP6c depletion are relatively low in natural circumstances , and oocyte death occurs only when both endogenous and exogenous harm accumulated with time to a certain degree , which could explain why the whole oocyte elimination process took up to 6 months in Ppp6c mutant ovaries . Thus , PP6c could control oocyte quality through its role in DDR pathway as well as regulate follicular activation through participating in the AMPK pathway . Nevertheless , we cannot exclude other possibilities , e . g . , regulation of pre-mRNA splicing and Hippo pathway , that could also contribute to the phenotypes in our mutant mouse model . Female meiosis is error-prone in humans . Our previous study reported that Zp3-Cre mediated PP6c depletion in growing oocytes leads to defective MII spindle function and unfaithful chromatid segregation in meiosis II without affecting folliculogenesis , indicating that PP6 can act as antagonizer to oocyte aneuploidy during the MII exit . Here we demonstrate that Gdf9-Cre mediated PP6c depletion in dormant oocytes causes defective folliculogenesis and massive germ cell elimination at early stages , indicating that PP6 can also safeguard oocyte genomic integrity and regulate folliculogenesis during the long prophase I arrest . Furthermore , isolated GV oocytes from Ppp6cF/F;GCre+ mice before POF occurs show severely impaired in vitro maturation because of DNA damage , in sharp contrast to the unaffected meiotic maturation progress of Ppp6cF/F;ZCre+ oocytes . Although these two knockout mouse models are both oocyte-specific knockouts , they exhibit completely different phenotypes that presumably reflect differences between timing of Zp3-Cre and Gdf9-Cre expression . Both ZP3 and GDF9 are specifically expressed in oocytes . The synthesis of ZP3 starts in primary follicles from PD5 , reaches a maximum in growing follicles , and decreases in full-grown oocytes , which makes Zp3-Cre only suitable for deletion of gene expression in oocytes from primary follicle stages on . However , Gdf9-Cre is expressed in oocytes from primordial follicle stage . This difference in expression is presumably why Ppp6cF/F;GCre+ mice display primordial follicle defects whereas Ppp6cF/F;ZCre+ mice do not . In summary , we provide evidence that PP6 acts as a critical guard of genomic integrity in lengthy prophase I arrest of oocytes and is an indispensable regulator of folliculogenesis , and thus female fertility . Our data may provide valuable information for the design of therapeutics for POF . Animal care and handling were conducted according to the guidelines of the Animal Research Committee of the Institute of Zoology , Chinese Academy of Sciences . The institutional committee which is licensed by Beijing Municipal Experimental Animal Administration approved this study . Mice lacking Ppp6c in oocytes ( referred to as Ppp6cF/F;GCre+ ) were generated by crossing Ppp6cF/F mice [28] with Gdf9-Cre mice . Both transgenic mouse lines have C57BL/6J genomic background . The mice were housed under controlled environmental conditions with free access to water and food . Light was provided between 08:00 and 20:00 . Commercial antibodies were used to detect PPP6C ( rabbit , A300-844A; Bethyl Laboratories , Inc . ) , α-tubulin ( mouse , DM1A; Sigma-Aldrich ) , MVH ( rabbit , ab13840; Abcam ) , γH2AX ( rabbit , 9718; Cell Signaling Technology , Inc . ) , p-CHK1 ( S345 ) ( rabbit , BS4041; Bioworld Technology , Inc . ) , p-CHK2 ( T68 ) ( rabbit , BS4043; Bioworld Technology , Inc . ) , p-p53 ( S15 ) ( rabbit , 12571; Cell Signaling Technology , Inc . ) , CHK1 ( rabbit , BS1052; Bioworld Technology , Inc . ) , p-AKT ( S473 ) ( rabbit , 4060; Cell Signaling Technology , Inc . ) , p-AMPK ( T172 ) ( rabbit , 2535; Cell Signaling Technology , Inc . ) , p-mTOR ( S2448 ) ( rabbit , 5536; Cell Signaling Technology , Inc . ) , p-S6K ( T389 ) ( rabbit , 9234; Cell Signaling Technology , Inc . ) , p-rpS6 ( S240/244 ) ( Rabbit , 5364; Cell Signaling Technology , Inc . ) , GAPDH ( rabbit , 5174; Cell Signaling Technology , Inc . ) and β-actin ( mouse , sc-47778 , Santa Cruz ) . Secondary antibodies were purchased from ZhongShan Golden Bridge Biotechnology Co . , LTD ( Beijing ) . Ovaries used for histological analysis were collected from adult female mice . They were fixed in 4% paraformaldehyde ( pH 7 . 5 ) overnight at 4°C , dehydrated , and embedded in paraffin . Paraffin-embedded ovaries were sectioned at a thickness of 8-μm for hematoxylin and eosin ( H&E ) staining . One or both ovaries from more than three mice of each genotype were used for the analysis . Paraffin-embedded ovarian tissue sections were deparaffinized , immersed in retrieval solution ( 10 mM sodium citrate ) , heated in an autoclave , blocked with 10% normal goat serum , and then incubated overnight with primary antibodies ( anti-MVH and anti-γH2AX at 1:200 dilution ) . For immunofluorescence , localization of the primary antibody was performed by incubation of the sections with the corresponding secondary antibodies ( Invitrogen ) at 1:500 dilution for 1h at room temperature . Finally , nuclei were stained with DAPI . For immunohistochemistry , the Vecta stain ABC kit ( Vector Laboratories , CA , USA ) was used to detect the signal of primary antibody . Analysis of apoptosis in ovarian follicles was carried out by TUNEL assay using the ApopTag Plus in situ apoptosis detection kit ( Chemicon International , Temecula , CA , USA ) . At least three different samples from each genotype were analyzed in parallel . Oocytes for immunofluorescent staining were fixed in 4% paraformaldehyde in PBS for 30 min at room temperature . The fixed oocytes were then transferred to membrane permeabilization solution ( 0 . 5% Triton X-100 ) for 20 min and blocking buffer ( 1% BSA-supplemented PBS ) for 1 h . The oocytes were then incubated overnight at 4°C with FITC conjugated anti-α-tubulin at 1:2000 dilution . Nuclei were stained with DAPI . Finally , oocytes were mounted on glass slides and examined with a laser scanning confocal microscope ( Zeiss LSM 780 META , Germany ) . Quantification of ovarian follicles was performed as previously described [43] . Briefly , to count the numbers of follicles , paraffin-embedded ovaries were serially sectioned at 8-μm thickness and every fifth section was mounted on slides . Then these sections were stained with hematoxylin and eosin for morphological analysis . Ovarian follicles at different developmental stages , including primordial , primary ( type 3 and type 4 ) , secondary ( type 5 ) and antral follicles ( type 6 and type 7 ) were counted in collected sections of an ovary , based on the well-accepted standards established by Peterson and Peters [44] . In each section , only those follicles in which the nucleus of the oocyte was clearly visible were scored and the cumulative follicle counts were multiplied by a correction factor of 5 to represent the estimated number of total follicles in an ovary . For the natural ovulation assay , 2–4 month-old female mice were mated with fertile males overnight . Successful mating was confirmed by the presence of vaginal plugs . Fertilized eggs were harvested from oviducts , counted and analyzed after removal of the cumulus mass with 1mg/ml hyaluronidase ( Sigma-Aldrich ) in M2 medium ( Sigma-Aldrich ) . GV stage oocytes were isolated from ovaries of ~PD35 female mice and cultured in M2 medium under paraffin oil at 37°C , 5% CO2 in air . For in vitro treatment of zeocin , fully-grown GV oocytes were first treated with zeocin ( 200 μg/ml , Invitrogen ) for 1 h in M2 medium supplemented with 2 . 5 μM milrinone and then blocked by the same concentration of milrinone for recovery . Oocytes were collected after 12 hours of recovery for western blot . Ovary lysate was prepared from minced ovaries after removal of suspended granulosa cells by centrifugation for western blot analysis . Thirty μg ovary protein or 200 oocytes were mixed with SDS sample buffer and boiled for 5 min at 100°C for SDS-PAGE . Western blot was performed as described previously [45] , using antibody dilutions as below , antibodies against PPP6C , MVH , γH2AX , p-CHK1 ( S345 ) , p-CHK2 ( T68 ) , CHK1 at 1:500 , antibody against p53 , p-p53 ( S15 ) , p-AKT ( S473 ) , p-AMPK ( T172 ) , p-mTOR ( S2448 ) , p-S6K ( T389 ) , p-rpS6 ( S240/244 ) at 1:1000 , and antibodies against GAPDH and β-actin at 1:2000 . To induce DNA DSBs in vivo , zeocin was injected into the abdominal cavity of female mice once every day for 5 days , and physiological saline ( vehicle ) was injected as control . Zeocin ( 100 mg/ml , Invitrogen ) was diluted in physiological saline to give a final concentration of 400 mg/ml , and 0 . 1 ml ( 40 μg zeocin ) was injected per mouse . At least 3 mice were injected in each group . Mice were sacrificed 5 days after injection and ovaries were fixed for histological analysis or lysed for western blot . In breeding assays , Ppp6cF/F and Ppp6cF/F;GCre+ female mice with sexual maturity were continually mated to Ppp6cF/F male mice with known fertility for 6 months . Cages were checked daily for counting the number of litters and pups . All experiments were repeated at least three times . Student’s t test was used for statistical analysis and performed using SPSS . Data were expressed as mean ± SEM and values are statistically significant at P<0 . 05; *P<0 . 01 .	Formation of haploid gametes from diploid germ cells requires a specialized reductive cell division known as meiosis . In contrast to male meiosis that takes place continuously , a unique feature of female meiosis in mammals is the long arrest in meiosis I , which lasts up to 50 years in humans . Because the size of the germ cell pool determines the reproductive lifespan of females , it is important to discover mechanisms preserving the germ cell pool during the lengthy meiotic arrest . In this study , we examined the physiological role of a member of the PP2A-like serine/threonine phosphatase subfamily , protein phosphatase 6 , in mouse oocytes during ovarian follicular development . This is the first study linking PP6 to the maintenance of the female germ cell pool and fertility . We find PP6 is an indispensable protector of arrested oocytes by safeguarding genomic integrity during their dormancy in the mouse ovary .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "meiosis", "medicine", "and", "health", "sciences", "reproductive", "system", "enzymes", "cell", "cycle", "and", "cell", "division", "cell", "processes", "enzymology", "phosphatases", "germ", "cells", "animal", "models", "dna", "damage", "oocytes", "model", "organisms", "experimental", "organism", "systems", "dna", "research", "and", "analysis", "methods", "prophase", "animal", "cells", "chromosome", "biology", "proteins", "mouse", "models", "ovaries", "biochemistry", "cell", "biology", "ova", "anatomy", "nucleic", "acids", "genetics", "biology", "and", "life", "sciences", "cellular", "types" ]	2016	Protein Phosphatase 6 Protects Prophase I-Arrested Oocytes by Safeguarding Genomic Integrity
Defining the molecular structure and function of synapses is a central theme in brain research . In Drosophila the Bruchpilot ( BRP ) protein is associated with T-shaped ribbons ( “T-bars” ) at presynaptic active zones ( AZs ) . BRP is required for intact AZ structure and normal evoked neurotransmitter release . By screening for mutations that affect the tissue distribution of Bruchpilot , we have identified a P-transposon insertion in gene CG11489 ( location 79D ) which shows high homology to mammalian genes for SR protein kinases ( SRPKs ) . SRPKs phosphorylate serine-arginine rich splicing factors ( SR proteins ) . Since proteins expressed from CG11489 cDNAs phosphorylate a peptide from a human SR protein in vitro , we name CG11489 the Drosophila Srpk79D gene . We have characterized Srpk79D transcripts and generated a null mutant . Mutation of the Srpk79D gene causes conspicuous accumulations of BRP in larval and adult nerves . At the ultrastructural level , these correspond to extensive axonal agglomerates of electron-dense ribbons surrounded by clear vesicles . Basic synaptic structure and function at larval neuromuscular junctions appears normal , whereas life expectancy and locomotor behavior of adult mutants are significantly impaired . All phenotypes of the mutant can be largely or completely rescued by panneural expression of SRPK79D isoforms . Isoform-specific antibodies recognize panneurally overexpressed GFP-tagged SRPK79D-PC isoform co-localized with BRP at presynaptic active zones while the tagged -PB isoform is found in spots within neuronal perikarya . SRPK79D concentrations in wild type apparently are too low to be revealed by these antisera . We propose that the Drosophila Srpk79D gene characterized here may be expressed at low levels throughout the nervous system to prevent the assembly of BRP containing agglomerates in axons and maintain intact brain function . The discovery of an SR protein kinase required for normal BRP distribution calls for the identification of its substrate and the detailed analysis of SRPK function for the maintenance of nervous system integrity . Molecular characterization of synaptic transmission has become a central theme of neuroscience research . A major contribution to the regulation of neurotransmitter release results from properties and interactions of proteins at the presynaptic active zone ( AZ ) . In recent years several AZ components have been identified for vertebrates [1]–[5] . The large AZ specific proteins Piccolo and Bassoon together with RIM form a scaffold at the presynaptic AZ interacting with CAST/ERC , Munc13 and Liprin-α . Together these proteins form the cytomatrix at the active zone ( CAZ ) and are involved in regulation of the synaptic vesicle cycle and plasticity of synaptic transmission . Most presynaptic proteins have been conserved in Drosophila [6] , with the notable exception of Piccolo and Bassoon . Sequence similarity between CAST/ERC and its insect homologue Bruchpilot ( BRP ) is largely confined to two N-terminal domains which in BRP are followed by extensive coiled-coil structures of about 1000 amino acids [7] . A present hypothesis proposes that these regions might convey cytoskeletal interactions that in vertebrates could be mediated by Piccolo and/or Bassoon [7] . Knock-down by RNAi or deletion of the brp gene leads to loss of the presynaptic dense projections ( T-bars , [1] ) , reduction of calcium channel density , severe defects in synaptic transmission , and altered short-term plasticity [7] , [8] . Stimulated emission depletion ( STED ) microscopy using the monoclonal antibody nc82 [9] which recognizes a C-terminal epitope of BRP revealed that BRP forms a donut-shaped ring centered at active zones [8] . No information is as yet available on the molecular mechanisms of AZ assembly in Drosophila while in vertebrates several active zone proteins are transported to the presynaptic terminal in the form of active zone precursor vesicles termed Piccolo transport vesicles ( PTVs ) [10] . Presumably , many more types of transport vesicles will be found , considering the complexity of vesicles observed near nascent active zones [11] , [12] . Posttranslational modification of proteins of the CAZ has been suggested to be an important mechanism of synaptic modulation [13] . For example , mammalian serine/threonine kinase SAD-B is associated with synaptic vesicles and with CAZ , it phosphorylates RIM but not Munc13 , and interference with SAD-B targeting inhibits synaptic transmission [14] . The Drosophila SAD homologue ( encoded by CG6114 ) apparently has not yet been characterized . By screening for mutants with altered tissue distribution of BRP we now have identified a kinase presumably associated specifically with presynaptic active zones of Drosophila . It shows high homology to mammalian SR protein kinases . SR proteins are highly conserved phosphoproteins involved in the regulation of constitutive and alternative splicing [15]–[19] . SR proteins exhibit one or two N-terminal RNA-binding regions termed RRM ( RNA recognition motif ) , as well as a serine/arginine ( SR ) rich C-terminal domain [20] which is required for protein-protein interactions affecting cellular localization [21] and regulation of splicing [22]–[24] . The genome of Drosophila melanogaster ( Dm ) encodes several SR proteins ( SC35 , SF2 ( ASF ) , B52 ( SRp55 ) , RBP1 , RBP1-like , X16 , SRp54 ) [25]–[27] . Members of the highly conserved family of SR protein kinases ( SRPKs ) [28]–[35] have been shown to phosphorylate the RS ( arginine/serine rich ) domain of the SR family of splicing factors [29] , [31] , [36]–[39] . Phosphorylation modulates protein-protein or protein-RNA interactions and is therefore an important mechanism for the regulation of SR proteins . Three mammalian SRPKs have been described . Their characteristic feature is a bipartite highly conserved kinase domain interrupted by a unique spacer region involved in individual regulation [40] . SRPK1 and SRPK2 are widely expressed in mouse embryonic tissues while SRPK3 ( Stk23 ) is a muscle specific protein kinase whose elimination leads to centronuclear myopathy [41] . Four Srpk genes have been detected in the Drosophila genome [25] , [26] , [42] , CG8174 at 52A1 , CG8565 at 13F3-4 , CG11489 ( CG9085 ) at 79D4 , and Doa ( CG1658 ) at 98F6 . Available information on the Drosophila melanogaster ( Dm ) gene CG11489 includes its expression in the embryonic brain [43] and changes in transcript levels in various mutants [44] , [45] . Here we further characterize the gene CG11489 which we propose to name Srpk79D since sequence comparison identifies no clear autology to any of the three mammalian genes . We identify four alternatively initiated and spliced transcripts by RT–PCR , demonstrate that SRPK79D is able to phosphorylate a synthetic SR-substrate in vitro , and generate and characterize hypomorphic and null mutants for the Srpk79D gene . We show that mutation of this kinase gene leads to highly conspicuous accumulations of the active zone protein Bruchpilot ( BRP ) in discrete structures in larval and adult nerves . Functionally , elimination of SRPK79D causes locomotor defects and reduced longevity . Possible links between the lacking kinase activity , the structural abnormalities with altered BRP distribution , the behavioral defects , and the reduced life span are discussed . According to the latest flybase update available from the Berkeley Drosophila Genome Project ( BDGP ) ( http://flybase . org/ , FB2009_01 ) the gene CG11489 generates two transcripts RB and RD which use different transcription and translation start sites , share exons 4 and 5 , but are differently spliced again downstream of intron 5 ( Figure 1A and 1B ) . To reassess this information we sequenced wild-type Canton-S RT–PCR products obtained from adult mRNA by various primer combinations ( Figure 1A ) . We were able to confirm the transcript RB and the transcript RC annotated in an earlier version of flybase . In addition we found a new exon ( exon 7 , 159 bp ) that is alternatively spliced from both transcription starts , resulting in four RNA ( R ) transcripts ( RB , RC , RE , RF ) coding for four protein ( P ) isoforms of 749 ( PB ) and 802 ( PE ) and 816 ( PC ) and 869 ( PF ) amino acids ( aa ) as shown in Figure 1B . RT–PCR experiments designed to detect the RD transcript ( coding for isoform PD of 695 aa ) from adult poly-A+ mRNA were unsuccessful although control PCR from genomic DNA with the same primers generated reliable products . The Drosophila gene CG11489 is predicted to code for serine/threonine protein kinase isoforms whose kinase domains consist of two parts , separated by a spacer of 155 or 208 amino acids ( isoforms without or with the alternatively spliced exon 7 , respectively ) ( http://www . expasy . org/prosite/ ) . Homology searches ( BLAST ) reveal high homologies between these kinase domains and mammalian SR protein kinases ( http://www . ncbi . nlm . nih . gov/ ) . Since sequence comparisons fail to identify a clear autology relationship we propose to name CG11489 the Drosophila Srpk79D gene . Figure 1C shows the homology for the kinase domain of Dm SRPK79D to human SRPK3 ( 65% aa identity ) ( MSSK1 [46] ) and SRPK2 ( 54% ) [47] , [48] . Those parts of Dm SRPK79D that are not included in the kinase domain show no significant homologies to any other known proteins . To investigate if SRPK79D is indeed a newly identified SR protein kinase in Drosophila , we performed in vitro phosphorylation assays with SRPK79D isoforms PC and PB using the synthetic peptide SRPK1tide ( Upstate Chemicon ) as a substrate . To isolate active Dm SRPK79D we transfected HEK293 cells with myc-tagged versions of Srpk79D cDNAs and purified the proteins by immunoprecipitation using an anti-Myc antibody . The substrate SRPK1tide corresponds to the amino acids 204–218 of the SR-rich region of the human SR protein ASF-1/SF-2 [15] , [49] . Both tested isoforms of Dm SRPK79D are able to phosphorylate SRPK1tide significantly in vitro ( Figure 2 ) . Autophosphorylation could be excluded , because control preparations without substrate show no detectable phosphorylation . We therefore conclude that Dm SRPK79D is indeed an SR protein kinase . The P{lacW}CspP2 line was isolated in a screen to identify P-element insertions near the cysteine string protein gene ( Csp ) [50] , [51] . By inverse PCR it was found that in this line the P ( lacW ) transposon inserted 81 bp upstream of the first exon-intron boundary of the Srpk79D- ( RC/RF ) transcripts and destroys the open reading frame of these mRNAs ( Figure 1A ) . We therefore propose to rename this line as Srpk79DP1 , short for P{lacW}Srpk79DP1 . This line is considered a hypomorph because the two transcripts RB and RE remain intact and their expression is maintained at reduced levels as revealed by RT–PCR ( data not shown ) . This P-element was remobilized and white-eyed jump-out lines were characterized by PCR and sequencing . The line Srpk79DVN suffered 3′ to the insertion site a deletion of 3861 bp ( horizontal bar in Figure 1A ) that includes the transcription and translation start of the Srpk79D-RB/RE transcripts and also eliminates the first 85 codons of the highly conserved SR kinase domain ( Figure 1A ) . At the site of the deletion the line contains an insertion of 23 bp ( remnant of the P-element ) . Thus this line is most likely a true null mutant . The line Srpk79DREV shows a precise excision of the P-element and is therefore considered a revertant . Homozygous mutants , both of the hypomorphic Srpk79DP1 allele and the Srpk79DVN null allele , show similar conspicuous accumulations of Bruchpilot in discrete spots in larval segmental and intersegmental nerves ( Figure 3A and 3B ) . These BRP spots are also observed when Srpk79D mRNAs are knocked down by panneural expression of an RNAi construct directed against all four transcripts ( Figure 3C ) . Whether these BRP spots represent aggregates of BRP or are accumulations of protein complexes containing BRP is not known at present . The axonal BRP accumulation phenotype is almost completely rescued by panneural overexpression of the SRPK79D-PF isoform in the hypomorphic mutant Srpk79DP1 ( Figure 3F ) as well as in the null mutant Srpk79DVN ( Figure 3I ) . The BRP accumulation phenotype is also rescued by transgenic expression of the SRPK79D-PB isoform ( C-terminally linked to eGFP ) ( Figure 3L ) or by the SRPK79D-PC isoform ( C-terminally linked to eGFP ) ( data not shown ) in the null mutant . The respective parental controls ( Figure 3D , 3E , 3G , 3H , and 3K ) display the mutant phenotype . Since in an independent study the SRPK97D-PB isoform is reported not to rescue the BRP accumulation phenotype of a different allele ( Srpk79DATC ) of the kinase gene ( Johnson et al . , in press ) , we have repeated the RB rescue experiments and verified the UAS-RB cDNA transgene in the null mutant background by PCR using primer pair 6f/6r of Figure 1A which is specific for the RB/RE cDNAs and includes two introns thus allowing the distinction between genomic DNA and cDNA . Eight of eight larvae of the parental elav-Gal4 driver and 16 of 16 larvae of the parental UAS-RB cDNA line , both in Srpk79DVN null mutant background , showed the BRP accumulations as in Figure 3G and 3K , respectively , and 8 of 8 larvae of the F1 of this cross showed a clear , obvious rescue effect as in Figure 3L . In vertebrates it was shown that proteins which build the cytomatrix at the active zone like Piccolo , Bassoon , RIM , Munc13-1 and N-Cadherin are transported by a special class of vesicles , the precursors of the active zone vesicles [10] , [11] , [52] . It thus seems possible that the axonal transport of either a special class of vesicles , or of the Bruchpilot protein itself is affected by the Srpk79D mutations . We therefore investigated whether the BRP accumulations in the nerve influence synaptic structure or the distribution of Bruchpilot at the active zones of Srpk79D mutant larvae . We analyzed the morphology of synaptic boutons and their active zones in Srpk79DP1 or Srpk79DVN mutants and wild type on larval muscle 6/7 or 12/13 in segment A3 . We could not detect a clear alteration in the gross morphology of the presynaptic boutons of the mutants ( Figure 3M and 3O ) compared to wild type ( Figure 3N ) or the revertant ( Figure 3P ) nor was there a significant difference in the number of boutons or of active zones in the null mutant in comparison to the wild-type controls ( type Ib boutons of muscle 6/7 VN: 43 . 4±3 . 9 , WT: 42 . 4±1 . 8 , p>0 . 1 , n = 8; active zones in type Ib boutons of muscle 12/13 VN: 222±27 . 9 , WT: 275±31 . 8 , p>0 . 1 , n = 6 ) . There was also no obvious difference in the distribution or the qualitative expression level of the Bruchpilot protein at the presynaptic active zones of larval motor neurons ( Figure 3M–3P ) ( cf . Discussion ) . Abnormal accumulations of proteins in the axon are often caused by general defects in axonal transport [53]–[58] . We therefore tested whether the accumulation of the Bruchpilot protein observed in the Srpk79D mutants might be due to a general impairment of the axonal transport machinery . We find that other synaptic proteins are uniformly distributed throughout the larval nerves of both wild-type and Srpk79DVN null mutant . This is illustrated for the synaptic vesicle protein cysteine string protein ( CSP ) [50] in the Srpk79DVN null mutant ( Figure 3Q ) and the w1118 control ( Figure 3R ) . In addition , other proteins of the presynaptic terminal like synaptotagmin or synapsin also do not accumulate abnormally in mutant nerves ( data not shown ) , which allows us to conclude that axonal transport in general is not affected by the mutations of the Srpk79D gene . Finally , we tested whether a similar phenotype is also found in adult nerves . Figure 3T and 3U shows anti-BRP ( nc82 ) staining of horizontal sections of adult null-mutant and wild-type brains , respectively . Neuropil regions of the antennal lobes and central protocerebrum are clearly labelled . In the mutant a large number of spot-like BRP accumulations are seen in both antennal nerves ( arrows ) whereas only a few spots are present in the antennal nerves of the wild type . This clear difference has been observed in four out of four pairs of preparations . In line with the largely normal appearance of the mutant nervous system we find that basic synaptic transmission at the neuromuscular junction of Srpk79DVN null-mutant larvae , as reflected by amplitude and frequency of miniature excitatory junction potentials ( mEJPs ) and amplitude and quantal content of evoked EJPs ( eEJPs ) are not altered in comparison to wild-type controls ( Figure 4 ) . Thus we conclude that the accumulation of the Bruchpilot protein in larval nerves does not lead to major defects in the function of the presynaptic active zone and synaptic transmission at the larval neuromuscular junction . Since at the presynaptic terminal BRP is structurally and functionally associated with electron-dense synaptic ribbons ( “T-bars” ) [7] , [8] we analyzed wild-type and Srpk79D null-mutant larval nerves emerging from the abdominal ganglia by standard electron microscopy in search for conspicuous electron-dense structures specific for the mutant . In each of three Srpk79D null-mutant larval preparations a nerve cross-section area between 6470 and 8717 µm2 was scrutinized , and per animal we found between 8 and 15 large electron-dense complexes of various shapes and varying diameters often surrounded by clear vesicles , as shown in the examples in Figure 5A , 5B , and 5D–5K . In three wild-type larvae an area between 6842 and 9750 µm2 each was analyzed . Here we found per animal between one and three small electron-dense structures of much lower complexity ( not shown , Table 1 ) . On average we found one electron-dense complex in 625 . 4 ( ±105 . 0; n = 3 ) µm2 in the mutant and one in 4904 ( ±1047; n = 3 ) µm2 in the wild type . Mean diameters of electron-dense complexes amounted to 449 . 04 nm ( ±43 . 5 nm; n = 37 ) in the mutant and 155 . 63 nm ( ±17 . 3 nm; n = 6 ) in the wild type . The electron-dense complexes in the mutant axons consist of complex ribbon-like structures resembling multiple T-bars and thus are considerably larger than a typical T-bar of a presynaptic larval neuromuscular bouton ( shown in Figure 5C , arrowhead , for comparison ) . A rough estimate of the volume density of the complexes indicates that it is compatible with the density of accumulations of BRP observed by fluorescence microscopy with an antibody against BRP ( Figure 3B ) . In order to verify that these electron-dense agglomerates correspond to the BRP containing spots seen in fluorescence microscopy we performed pre-embedding immuno-gold labelling of null-mutant axons . Figure 5L–5O shows examples of silver-enhanced gold particles ( unequivocally identified at high image brightness and marked by white circles , cf . Figure S4 ) associated with electron-dense agglomerates similar to those shown in Figure 5A , 5B , and 5D–5K . We noticed that all gold particles observed on 14 agglomerates decorate the periphery of the electron-dense structures rather than their centers . We thus conclude that these agglomerates indeed contain BRP . The density and distribution of silver grains not associated with electron-dense agglomerates ( background ) are similar in wild-type and null-mutant larval nerves ( not shown ) . The finding that the accumulation of Bruchpilot protein in larval nerves of Srpk79D mutants is not due to a general impairment of the axonal transport machinery supports the hypothesis of a specific interaction of BRP and SRPK79D . However , a direct interaction of the two proteins is likely only if they can be shown to co-localize in the same cellular compartment . To investigate this , we generated affinity-purified antisera against the non-overlapping N-terminal parts of the SRPK79D-PB/PE ( anti-PB ) and –PC/PF ( anti-PC ) isoforms . The specificity of the antisera was tested by Western blots of adult heads of wild type , null mutant , and SRPK79D-PB-eGFP or –PC-eGFP overexpressing lines . In the overexpressing lines clear specific Western blot signals were obtained ( Figure S1 ) . From wild-type heads detectable amounts of the PC isoform could only be obtained after enrichment by immuno-precipitation ( Figure S2 ) . To characterize targeting of the kinase isoforms we panneurally overexpressed GFP-tagged SRPK79D-PB and -PC isoforms and performed double immuno-stainings on larval preparations with anti-BRP ( nc82 ) and the anti-PB or anti-PC antisera or antibodies against GFP ( anti-GFP ) . Figure 6A and 6B demonstrates that panneurally overexpressed SRPK79D-PC-eGFP and Bruchpilot co-localize at the presynaptic active zone . The specificity of the anti-PC staining is demonstrated by the parallel staining of the Srpk79DVN ( VN ) null mutant ( Figure 6C ) . Therefore a direct interaction of the PC isoform and BRP at the presynaptic active zone seems possible . The PC isoform is in addition homogeneously distributed in the perikaryon ( Figure 6E ) . The same distribution is observed for the overexpressed PF isoform ( data not shown ) . Overexpressed SRPK79D-PB-eGFP , on the other hand , accumulates in discrete structures in the perikarya of larval neurons ( Figure 6F ) but is not targeted to synaptic terminals . With our antisera against SRPK79D-PB/PE and –PC/PF we could not detect any clear difference in immunohistochemical stainings between wild-type and null-mutant flies . ( Overexpression led to similar observations as in larvae , i . e . synaptic neuropil staining with anti-PC serum and perikaryal spots with anti-PB serum , data not shown ) . However , by two behavioral tests and life span measurements we demonstrate that SRPK79D is required for intact nervous system function also in adults because the mutations in the Srpk79D gene lead to behavioral deficits in adult flies in addition to the BRP accumulation in nerves . Phenotypes of the Srpk79DP1 ( P1 ) and Srpk79DVN ( VN ) mutants include flight impairment ( Figure 7A ) and reduced ability or motivation to walk on a horizontal surface ( Figure 7B ) as well as reduced life span ( Figure 7C ) . Srpk79DREV revertants ( REV ) behave like wild type . All three phenotypes , impaired flight , impaired walking , and reduced longevity were fully or partially rescued by transgenic panneural expression of the SRPK79D-PF isoform in the Srpk79DP1 mutant background when compared to the parental controls w , elav-Gal4;;P1 ( G-P1 ) and w;UAS-RF;P1 ( U-P1 ) ( Figure 7A–7C ) . The large survival difference between the P1 and the null mutant can most likely be assigned to genetic background effects because after extensive outcrossing of the Srpk79DP1 line to wild type w1118 this difference was no longer significant ( data not shown ) . These genetic background effects could also explain why the rescue in Srpk79DP1 mutant background did not extend the 50% survival time beyond the values found for the null mutant ( Figure 7C “VN” and “RES” ) . Using the information provided by the Berkeley Drosophila Genome Project ( BDGP ) [25] and flybase ( http://flybase . org/ , release January 2009 ) we confirmed by RT–PCR and sequencing the structure of the RB transcript and of the transcript RC that had been annotated in an earlier release . We extended this information by the detection of an alternatively spliced exon of 159 bp ( exon 7 in Figure 1A ) . This newly identified exon is differentially included in both RB and RC transcripts generating the transcripts RE and RF , respectively . The 53 amino acids encoded by exon 7 are located in the non-conserved spacer region between the two highly conserved kinase sub-domains of the SRPK79D-PE and -PF isoforms . This spacer region was shown to play a role in the subcellular localization of yeast and human SRPKs [34] . We find that the overexpressed SRPK79D-PC and -PF ( data not shown ) but not the -PB isoform are targeted to synaptic active zones while the overexpressed -PB isoform accumulates in clearly defined perikaryal sub-regions ( Figure 6 ) . This demonstrates that in Drosophila the non-conserved N-terminal domain of SRPK79D isoforms that is generated by the use of alternative promotors contains important protein targeting information while the additional amino acids encoded by exon 7 for the spacer region do not alter the targeting observed here . Recently , an additional transcript ( RD ) has been annotated in flybase . This transcript shares the 5′ region with the RC and RF transcripts , however the entire genomic sequence between exons 6 and 9 is retained as a large exon ( Figure 1B ) . Translation of this transcript would terminate in intron 6 and lead to a truncated protein of 695 amino acids which contains only the N-terminal half of the kinase domain and thus most likely would be non-functional . Our RT–PCR experiments did not detect the RD transcript in adult flies . The RD transcript is based on the existence of a cDNA clone ( GH08190 ) , which was generated in a high-throughput approach to produce cDNA clones for Drosophila melanogaster genes [59] . We propose that this cloned cDNA was reverse-transcribed from an incompletely spliced RNA . This interpretation is supported by the fact that the amino acids of the hypothetical PD isoform encoded by intron 6 show no homology to any known proteins and are not conserved among diptera . BLAST searches revealed that the kinase domain common to all four Dm SRPK79D isoforms shows high homology to the conserved family of SR protein kinases [27]–[34] . These kinases are known to be involved in the regulation of splicing via phosphorylation of SR proteins [28] , [30] , [35]–[38] . The kinase domain of all SR protein kinases consists of two highly conserved sub-domains , divided by a non-conserved spacer . Since homologies to all other kinases are considerably lower and a typical SR protein is readily phosphorylated by both Dm SRPK79D-PB and -PC isoforms we conclude that SRPK79D isoforms are members of the family of SR protein kinases . In Drosophila melanogaster there are three genes ( Srpk79D ( CG11489 ) ; CG8174; CG8565 ) with very high homology to the three human SR protein kinase coding genes ( hs Srpk1-3 ) but little is known about the role of these genes in Drosophila so far . Recently the LAMMER kinase DOA ( gene CG33553 ) which shows intermediate level of homology to SRPKs has been shown to phosphorylate the SR protein DX16 [26] , increasing the number of putative Srpk genes in Drosophila to four . Highest amino acid identity of the kinase domain of Dm SRPK79D is observed with human SRPK3 ( Figure 1C ) which is specifically expressed in muscles [40] . However , the differences in homology to mammalian SRPKs are small and may not be significant . The various phenotypes of the Drosophila Srpk79D mutants and their rescue by panneural RC/RF-cDNA expression demonstrate that in Drosophila SRPK79D-PC/PF function is relevant in the nervous system . Antisera raised against the different N-terminal domains of PB/PE and PC/PF isoforms expressed in E . coli recognize the respective overexpressed proteins both in larvae and adults ( Figure 6 and data not shown ) but fail to detect the endogenous proteins , indicating that the expression level of the Srpk79D gene is very low . It is unlikely that posttranslational processing eliminates recognition of the endogenous kinase ( PC/PF-isoform ) by the anti-PC antiserum since a signal at the expected size can be detected after enrichment of the kinase by immunoprecipitation using the PC antiserum ( Figure S2 ) . In this report we show that mutation of the Srpk79D gene leads to accumulations of the active zone protein Bruchpilot in discrete spots in larval and adult nerves and demonstrate that these accumulations correspond in neuronal axons of larvae to large electron-dense structures surrounded by clear vesicles ( Figure 5 ) . The speculation that these electron-dense structures are molecularly related to T-bars is supported by their ribbon-like appearance , by the observation that the anti-BRP antibody nc82 binds to the periphery of the ribbons similar to what has been shown by STED microscopy for synaptic T-bars [8] , and by the fact that they apparently are able to bind clear vesicles . Photoreceptor terminals of Bassoon knock-out mice contain synaptic ribbons detached from active zones . These “floating” ribbons are associated with vesicles , presumably synaptic vesicles [60] . Also , in axons of young rat hippocampal cultures aggregates of dense-core and clear vesicles are observed which label positively for synaptic vesicle markers like synaptobrevin , SV2 , synaptotagmin and synapsin-I [61] . We excluded that the clear vesicles associated with the axonal agglomerates described here are mature synaptic vesicles because they are not labelled by various antibodies against synaptic vesicle proteins , like cysteine string protein ( CSP ) , synapsin , synaptobrevin , and synaptotagmin . No difference in the staining of larval nerves between wild type and Srpk79D mutants is observed with these antibodies ( Figure 3Q and 3R , and data not shown ) . These experiments also exclude a general impairment of the axonal transport machinery as cause for the BRP accumulation phenotype because synaptotagmin and CSP have been shown to accumulate in the axons of mutants known to affect axonal transport [53]–[57] . Also , light microscopical morphology of the larval neuromuscular junction and the qualitative distribution of BRP as reflected by the number of presynaptic boutons and the number of BRP-positive active zones is not altered in our Srpk79D mutants compared to wild type . We have not attempted to quantify the amount of BRP at the active zones . In a report published simultaneously a different mutant allele Srpk79DATC of the Srpk79D gene is characterized which contains a P-element insertion in intron 8 of the gene and thus disrupts all four transcripts . This mutation causes very similar accumulations of BRP in larval nerves and the authors report a 30% reduction of BRP immuno-fluorescence at the larval neuromuscular active zones in homozygous Srpk79DATC mutants [62] . This observation is interpreted as an impairment of BRP transport to the presynaptic active zone of larval neuromuscular junctions due to a premature assembly of T-bar-like agglomerates in peripheral nerves [62] . Our immunohistochemical studies revealed that transgenically overexpressed GFP-tagged SRPK79D-PC and -PF ( but not –PB ) isoforms co-localize with Bruchpilot at the presynaptic active zone ( Figure 6A–6C and data not shown ) . This observation indicates either that the N-terminus of SRPK79D-PC and -PF isoforms contains targeting signals for active zone localization or that these kinase isoforms can bind to active zone proteins during transport . Thus , a direct interaction of SRPK79D-PC/PF and BRP at the active zone seems possible although co-immuno-precipitation experiments for the two proteins were unsuccessful ( data not shown ) . The obvious question of whether there are SR proteins at active zones and whether RNA splicing can occur at presynaptic active zones has now to be investigated . There is increasing evidence that presynaptic mRNA translation may contribute to synaptic plasticity [63] , [64] . However , larval olfactory conditioning [65] of Srpk79DVN null mutants was not significantly impaired ( Figure S5 ) . Since overexpressed GFP-tagged SRPK79D-PB is not found at active zones but nonetheless rescues the BRP accumulation phenotype in larval nerves of Srpk79DVN null mutants our data do not support the hypothesis that mRNA splicing at active zones might be required to prevent the axonal BRP accumulations . We have not observed a clear functional difference for the different SRPK79D isoforms . The striking axonal BRP accumulation phenotype is seen both in the Srpk79DP1 mutant ( lacking isoforms PC/PF and showing reduced expression of isoforms PB/PE ) and in the Srpk79DVN null mutant ( lacking all isoforms ) . Since it can be rescued in both mutants by all three available rescue cDNA constructs , RB , RC and RF ( Figure 3F , 3I , 3L , and data not shown ) , we conclude that the expression level of the kinase is important , not which N-terminus it contains nor apparently whether it is localized at the active zones . Whether this is also true for the behavioral and survival phenotype must remain open since the corresponding rescue experiments were performed only with Srpk79DP1 mutants overexpressing the RF cDNA ( Figure 7 ) . The reasons why the BRP accumulation phenotype of our deletion mutant Srpk79DVN is rescued by our RB cDNA construct but not the very similar phenotype of the Srpk79DATC allele of Johnson et al . ( in press ) by their RB cDNA construct has now to be investigated . Interestingly , mutation of the serine/threonine kinase Unc-51 that recently has been shown to regulate the localization of Bruchpilot to sites opposing the glutamate receptor fields in the postsynaptic membrane also causes BRP accumulations in larval nerves similar to the ones described here [66] . These authors interpret the BRP accumulations as axonal transport defects . However , a general axon transport defect can be excluded for the Srpk79D mutants . Yet another condition leading to BRP accumulations in larval nerves similar to the ones described here is the overexpression of BRP itself ( Figure 3S ) . We excluded that mutation of the Srpk79D gene influences the expression level of the brp gene by semi-quantitative RT–PCR ( data not shown ) and by immuno-blotting ( Figure S3 ) . We also found no evidence for changes in splicing of brp transcripts in the Srpk79D mutant by RT–PCR ( data not shown ) or by Western blotting ( Figure S3 ) . Thus we speculate that an unknown factor co-transported with BRP along neuronal axons may have to be present at a correct stoichiometric ratio with BRP to prevent the axonal assembly into the large electron-dense structures seen in the electron microscope ( Figure 5 ) . This ratio can be disturbed by BRP overexpression or by reduced expression of the unknown factor . Alternative splicing to regulate transcription factor activities and hence gene expression ( e . g . for the unknown factor ) is a general mechanism known from early Drosophila development or sexual differentiation and from cell cycle regulation [67] . By this hypothesis a link between the available information about SRPK79D function and the larval BRP accumulation phenotype could be proposed . Attempts to prevent the formation of BRP accumulations by simultaneous overexpression of both BRP and SRPK79D-PC failed ( Figure S6 ) . However , the correct wild-type stoichiometric ratio of the unknown factor and BRP is perhaps difficult to restore . The adult Srpk79D mutant flies show general behavioral impairments like locomotor defects and reduced life time ( Figure 7 ) . The effect of the Srpk79D mutation on the distribution of BRP in adults is similar to the larval phenotype in that BRP accumulates in various nerves ( Figure 3T and data not shown ) . The expressivity of this phenotype is however rather variable such that further experiments are required to clarify the cause of this variability . Thus at present we have no evidence that the altered BRP distribution is indeed responsible for the behavioral phenotypes . Altered expression of the unknown factor mentioned above due to defective splicing could of course explain these phenotypes . Since no clear electrophysiological defects are observed in larval nerve-muscle preparations of the Srpk79D null mutants ( Figure 4 ) we cannot offer a more specific hypothesis . On the other hand , subtle synaptic defects which may well play a role in central brain network function cannot be excluded because they may go unnoticed at the robust neuromuscular junction with its huge complement of reserve pool vesicles [68] . Also , since larval nerves contain both sensory and motor axons , we cannot be certain that the BRP accumulations are found in motor axons . The typical large BRP accumulations are not seen in the motor axons just before they branch to form the synaptic boutons . However , the fact that the behavioral defects of the Srpk79DP1 mutant are reverted after precise excision of the P-element insertion and can be fully or partially rescued by panneural expression of the SRPK79D-PF isoform in the mutant ( Figure 7 ) clearly link the phenotype to the Srpk79D gene and rule out genetic background effects . The rescue experiments also demonstrate that the eGFP moiety at the C-terminus does not impair the SRPK79D function required for prevention of the BRP accumulations . The observation that the rescue of the axonal BRP accumulation ( Figure 3F and 3I ) and of the reduced life span ( Figure 7C ) is incomplete , may be due to the fact that the three other isoforms are missing in the rescue animals or to the incorrect amount and/or distribution of the transgenically expressed protein . The present results demonstrate an important role of the kinase SRPK79D for the proper distribution of the active zone protein Bruchpilot . In larval and adult nerves the kinase is required for preventing the formation of conspicuous BRP-containing electron-dense ribbon-like agglomerates observed by electron microscopy in the Srpk79DVN mutant but not in wild-type controls . It is tempting to speculate that these ribbons may be molecularly related to T-bars beyond the association with BRP and that the kinase prevents the premature assembly of T-bars in peripheral axons [62] . Whether BRP is also involved in generating the behavioral and survival defects observed when SRPK79D-PC/PF isoforms or all SRPK79D isoforms are missing is not known . Since BRP does not contain any serine-arginine rich domains it seems unlikely that BRP is a substrate for these kinases . Our in vitro phosphorylation data suggest that in Drosophila SRPK79D isoforms modify SR proteins and thus may be involved in splicing regulation . It will now be necessary to identify the endogenous substrates of the SRPK79D kinase and study the mechanisms by which the formation of the extensive BRP-containing electron-dense agglomerates in wild-type axons is prevented . The characterization of an SR protein kinase that appears to be localized at presynaptic active zones and has dramatic effects on the distribution of an active zone protein is likely to modify current views on vertebrate SRPK function and may initiate new approaches to the study of active zone assembly and function . P{lacW}CspP2 line was generated in this lab [50] , [51] , w1118 , w , elav-Gal4 , w;actin-Gal4/CyO , Δ2–3Ki , p jump starter and balancer lines were obtained from the Bloomington Stock Center , Srpk79D-RNAi line ( ID 8451 ) was obtained from the Vienna Drosophila RNAi Collection . Flies were maintained at 25°C or 18°C under a 14/10 h light/dark cycle at 60–70% relative humidity . Total RNA was isolated using the QIAGEN RNeasy Mini Kit ( Qiagen; Hilden; Germany ) . Before reverse transcription of the messenger RNA into cDNA a 1 hour DNase digestion ( RNase-free DNase , Roche; Mannheim; Germany ) was performed at 37°C . The reverse transcription was performed with SuperScriptII Reverse Transcriptase ( Invitrogen; CA; USA ) following the instruction manual . The oligo-dT-primer used was purchased from MBI Fermentas ( oligo ( dT18 ) , MBI Fermentas; St . Leon-Rot; Germany ) and the dNTPs were from Metabion ( Metabion international AG; Planegg-Martinsried; Germany ) . The cDNA of the Srpk79D gene was amplified using different sets of specific primers ( from Eurofins MWG GmbH; Ebersberg; Germany ) . PCR was performed with Phusion High-Fidelity DNA Polymerase ( Finnzymes; Espoo; Finland ) . The following primer pairs were used for the transcript analysis ( cf . Figure 1A ) : 1f: 5′ACGAGAATTCGATGGCCGGCCTCATC3′ 1r: 5′GTACGGTGTTGGGCTTG3′ 2f: 5′CGGCGAATTCGATGGATGACTTTGGCT3′ 2r: 5′GTACGGTGTTGGGCTTG3′ 3f: 5′AAAAGCTTACCGGTTTCGAG3′ 3r: 5′TCGAAGGCCAAACAGGC3′ 4f:5′GTTGTGGTGTGCATGGAAAG3′ 4r: 5′CAATCATATATGTAGGTGTGGCCA3′ 5f: 5′GCCTGTTTGGCCTTCGA3′ 5r: 5′AAAGCGGCCGCGACGAACTCCT3′ For verification of the parental RB rescue line UAS-RB-cDNA-GFP in Srpk79DVN null mutant background the primer pair 6f: 5′ GCGACTTCAACTTCGTCTCC 3′ 6r: 5′ GCGGATTATGTTACGCACCT 3′ was used which gives a 1240 bp product for the wild-type gene and a 1086 bp product for the cDNA transgene . PCR products were purified with QIAGEN PCR Purification Kit ( Qiagen; Hilden; Germany ) and the DNA-Fragments were sequenced by Eurofins MWG GmbH ( Eurofins; MWG; Ebersdorf; Germany ) . Human embryonic kidney ( HEK ) 293 cells were grown at 37°C in the presence of 7% CO2 and cultured in Dulbecco's modified Eagle's medium ( DMEM , Biowest; Nuaillé; France ) containing 10% fetal calf serum . Transfection with myc-tagged Srpk79D cDNA constructs was performed with PolyFect reagent ( Qiagen; Hilden; Germany ) following the instruction manual . Cells were harvested in PBS , 48 h after transfection , and lysed in 400 µl lysis buffer containing 25 mM Tris pH 7 . 5 , 150 mM NaCl , 2 mM EDTA , 2 mM EGTA , 10% Glycerol , 0 . 1% Nonidet NP-40 supplemented with protease inhibitors ( complete mini EDTA , Roche; Mannheim; Germany ) at 4°C for 40 min . 300 µg of protein lysate were incubated for 10 min with 0 . 8 µg of mouse monoclonal anti-Myc antibody ( 9E10 , Santa Cruz; Heidelberg; Germany ) in a total amount of 500 µl lysis buffer . Protein complexes were precipitated over night at 4°C with 50 µl protein-A-sepharose ( Invitrogen; Carlsbad; CA; USA ) . The beads were washed three times with lysis buffer ( without protease inhibitors ) . The proteins were eluted from the sepharose by incubation with Laemmli buffer and separated by SDS-PAGE , transferred to nitrocellulose membranes and probed with mouse monoclonal anti-Myc antibody ( 9E10 , Santa Cruz; Heidelberg; Germany . The proteins were visualized using HRP-coupled secondary antibodies and the ECL detection reagent ( Amersham Biosciences; Braunschweig; Germany ) . For immunoprecipitation from adult heads using antisera 1000 flies ( volume∼3 ml ) were frozen in liquid nitrogen , heads were isolated by sieving and homogenized in 800 µl of lysis buffer supplemented with protease inhibitor . The supernatant of a 30 min centrifugation at 16 , 000 g was incubated with antiserum ( 1∶1000 f . c . ) for 10 min before 100 µl of protein-G agarose beads were added for over night incubation . For immunoprecipitation using hybridoma supernatant 50 adult flies were frozen in liquid nitrogen , heads were isolated and homogenized in 500 µl of lysis buffer supplemented with protease inhibitor . The supernatant of a 30 min centrifugation at 16 , 000 g was incubated with the hybridoma supernatant ( 1∶100 nc82 ) for 10 min before 100 µl of protein-G agarose beads were added for over night incubation . In both cases the beads were washed three times with lysis buffer and after centrifugation 40 µl of Laemmli buffer was added to the pellet and heated to 96°C for 5 min . The sample was analyzed by Western blotting . Cells were transfected , harvested and lysed as described above . Immunoprecipitated Myc-SRPK79D isoforms were washed five times in lysis buffer ( without protease inhibitor ) , once in 500 µM NaCl and twice in kinase buffer containing 8 mM MOPS-NaOH , pH 7 . 0 , and 0 . 2 mM EDTA . For the kinase reaction , 5 µl kinase buffer , 2 . 5 µl SRPK1tide ( 3 mM , Upstate Chemicon; Lake Placid; NY; USA ) as a substrate , 5 µl H2O , and 10 µl ATP-Mix ( 500 µM ATP , Roche; Mannheim , Germany ) , 0 . 75 µCi γ32ATP ( 10 µCi/µl , 3000 Ci/mmol; GE Healthcare Life Sciences; Braunschweig; Germany ) dissolved in 25 mM Mg ( CH3COO ) 2 were added to the SRPK79D beads , and the mixture was incubated for 30 min at 30°C . The reaction was stopped by spotting the sample onto p81 phosphocellulose paper ( diameter 2 . 5 cm , Whatman International LTD; Maidstone; UK ) . The papers were washed three times for 15 minutes in 175 mM H3PO4 to remove unbound ATP , before transferring them to scintillation vials containing 3 ml H2O . The amount of radioactive 32P incorporated into the substrate was determined in a scintillation counter ( LKB wallac 1214 Rackbeta , Liquid Scintillation counter ) by measuring the Čerenkov radiation . Each experimental condition was tested five times . Samples without kinase and samples without substrate served as negative controls . Significance was calculated performing the Student's t-test with Bonferroni correction . In the line Srpk79DP1 ( P{lacW}CspP2 ) a P{lacW}-element is inserted in the first exon 81 bp upstream of the 3′ exon-intron boundary of the Srpk79D-RC/RF transcripts , as was verified by PCR and sequencing . Flies of this line were crossed to “jump-starter” flies ( Δ2–3Ki , p ) . The offspring was crossed to w;;TM3/TM6 balancer flies . The F2 generation was screened for individuals with white eyes and either TM3 or TM6 . 600 lines were established as balanced stocks from these individuals . Homozygous flies were subjected to PCR to characterize deficiencies produced by the P-element remobilization . The following primers were used: Forward: CGGCCGGCATATGTAGTAGT; reverse: GCGGATTATGTTACGCACCT . The breakpoints of the deletion in the Srpk79D gene were identified by sequencing with the same primers . The SRPK79D null mutant Srpk79DVN suffered a deletion of 3861 bp . To restore the wild-type phenotype in Srpk79D mutants the complete Srpk79D-RF cDNA was cloned into the pP[UAST]-vector [69] and the construct was transformed into the germ line of white ( w− ) Drosophila by standard techniques [70] . Transgene insertions on the 2nd chromosome were recombined with either the Srpk79DP1 or the Srpk79DVN mutation . Rescue experiments were performed with the offspring of these flies crossed to the Gal4 driver line elav-Gal4 ( on X chromosome ) , in the corresponding Srpk79D mutant background . Thus these flies express the SRPK79D-PF isoform in the entire nervous system . To facilitate the subcellular localization of SRPK79D isoforms fusion constructs of the entire Srpk79D-RB , -RC , and RF cDNAs in frame with the complete sequence of the eGFP cDNA were cloned into the pP[UAST]-vector [69] . The eGFP sequence was amplified from the vector pMes-EGFP [71] and fused to the N-terminus for the –PF and to the C-terminus for the PB and PC isoforms . The constructs were transformed into the germ line of Drosophila and expression of the different SRPK79D-eGFP fusion proteins was driven with elav-Gal4 or actin-Gal4 in the nervous system . The cDNAs coding for the respective first exons were amplified by PCR using the primers: for RB/RE: EcoRI sense 5′GGGAGAATTCATGGATGACTTTGGCTC3′ XhoI anti 5′AAAACTCGAGCTCTTCCTTGACCGG3′ for RC/RF: EcoRI sense 5′AAAGAATTCATGGCCGGCCTCATC3′ XhoI-anti: 5′AAAACTCGAGAGACTGACGAATGGGCCG3′ The amplified sequences were cloned into the TOPO-vector ( TOPO TA cloning Kit , Invitrogen; Carlsbad; CA; USA ) . The fragments were excised using EcoRI and XhoI and cloned in frame with a His-tag into the pET-21a+ expression vector ( Novagen; Bad Soden; Germany ) . The protein was expressed in E . coli BL21 and purified by using nickel-chelate affinity chromatography following the protocol ( Qiagen; Hilden; Germany ) . Polyclonal guinea pig antibodies were generated [72] . Whole blood was collected , and the serum was separated for antibody purification . For the antibody purification the protein was expressed in E . coli BL21 and precipitated with chloroform/methanol . Following the protocol for affinity column HiTrap NHS-activated HP 1 m ( GE Healthcare; Buckinghamshire; England ) the column was equilibrated with buffer A and B before loading . Next the serum diluted in PBS was applied to the column . After washing steps the antibodies were eluted with Glycin ( pH 2 . 5 ) and neutralized with Tris/HCl ( pH = 8 . 9 ) . The antibodies were concentrated using CentriPlus 50 Centifugal Filter Units ( Millipore; Schwalbach; Germany ) . For stabilization 4 mg/ml BSA ( Sigma-Aldrich; Schnelldorf; Germany ) and 0 . 1% NaN3 was added . Wandering third instar larvae were dissected in ice-cold calcium-free Drosophila saline containing 128 mM NaCl , 35 mM Sucrose , 2 mM KCl , 4 mM MgCl2 , 3 mM HEPES , pH 7 . 2 , 1 mM EDTA , pH 7 . 0 . The larvae were pinned down , cut open dorsally along the midline and gut and fat body were removed . The preparations were fixed in 4% paraformaldehyde pH 7 . 4 for 1 . 5 hours on ice . Fixative was prepared as follows: 2 g paraformaldehyde was dissolved in 25 ml H2O at 60°C for five minutes . Then 100 µl of 1N NaOH was added . After cooling down to room temperature 25 ml of 2× PEM buffer ( 200 mM PIPES , 4 mM EGTA , 1 mM MgSO4 , pH 7 . 0 ) were added and the pH was checked . After fixation the preparations were washed 3 times for 15 minutes in PBST ( PBS containing 0 . 1% Triton-X 100 ) at room temperature . Non-specific binding was blocked by incubating with a blocking solution ( 2% BSA ( Sigma-Aldrich; Schnelldorf; Germany ) , 5% normal serum ( Vector Laboratories; Burlingame; CA; USA ) in PBST ) for 1 hour at room temperature . Incubation with the primary antibody was performed over night at 4°C . Primary antibodies were nc82 ( anti-Bruchpilot [7] ) , ab49 ( anti-CSP [50] ) and anti-GFP ( rabbit ) ( Dianova; Hamburg; Germany ) . Nc82 and ab49 were diluted 1∶100 , anti-GFP 1∶1000 in blocking solution . Before incubation with the secondary antibody , unbound primary antibody was removed by washing with PBST ( 3 times 45 minutes ) at room temperature . Incubation with the secondary antibody ( goat anti-mouse IgG Cy3 ) or goat anti-rabbit IgG Alexa488 ( Molecular Probes; Karlsruhe; Germany ) was performed at room temperature in the dark for 1 h . Secondary antibodies were diluted 1∶1000 in PBST . Then unbound secondary antibody was removed by washing in PBST for 4 times 1 h in the dark . Finally , preparations were embedded in Vectashield ( Vector Laboratories; Burlingame; CA; USA ) . Scans were performed with a confocal laser scanning microscope and raw data were processed with Image J . The preparation of adult frozen head sections has been described [9] . Briefly , air sacs and proboscis were removed in ice-cold 4% paraformaldehyde to allow quick access of the fixative to the brain . Flies were fixed at 4°C for 3 hours . Next they were incubated over night in 25% sucrose in Drosophila ringer serving as washing solution and freeze protectant . Fly heads were embedded in 16% carboxymethylcellulose ( low viscosity , Sigma-Aldrich; Schnelldorf; Germany ) and frozen in liquid nitrogen . 10 µm thick cryosections were collected on pre-chilled slides ( SuperFrost Plus , Menzel-Glaser GmBH ) and air-dried at RT for 20 minutes . Slides were blocked with normal serum for 2 h at RT and incubated with the first antibody ( nc82 , mouse monoclonal supernatant , dilution 1∶100 ) over night at 4°C . After washing twice with PBST for 10 minutes , the secondary antibody ( goat anti-mouse , Alexa 488 , dilution 1∶1000 , Invitrogen ) was applied for 1 hour at room temperature . After washing the sections twice 10 minutes in PBST they were mounted in Vectashield . Recordings were made from muscle fiber 6 in abdominal segments A3–A5 of third-instar wandering larvae after the peripheral nerves had been cut from the ventral ganglion . Dissections were performed in calcium-free hemolymph-like Ringer's HL3 solution [73]: 70 mM NaCl , 5 mM KCl , 20 mM MgCl2 , 10 mM NaHCO3 , 5 mM trehalose , 115 mM sucrose , and 5 mM HEPES , with a pH of 7 . 2 . Recordings were performed in HL3 , to which calcium was added to a final concentration of 1 mM . Intracellular muscle potential was recorded using a 1600 Neuroprobe amplifier ( A-M Systems Inc . ; Carlsborg; Washington; USA ) . Recording electrodes ( borosilicate glass 1 mm OD/0 . 58 mm ID ) with resistance of 10–20 MΩ were filled with 3 M KCl . Responses were recorded from muscle fibers with resting potentials between −58 and −76 mV . Data was low-pass filtered at 10 kHz , digitized , and acquired with a DAP card ( Microstar Laboratories; Bellevue; Washington; USA ) and recorded with DASYlab ( National Instruments Ireland Resources Ltd . ; Moenchengladbach; Germany ) . Mean evoked EJP amplitude was calculated from 60 consecutive EJPs elicited at 1 Hz . Evoked EJP amplitude and decay kinetics were analyzed using FORTRAN and verified with DASYlab . Miniature EJPs were recorded in 1 minute bins and analyzed with DASYlab . Quantal content was calculated by the ratio of eEJP/mEJP amplitudes after correcting eEJPs for nonlinear summation [74] . Wandering third instar larvae were dissected in Drosophila saline as described for immunohistochemistry . The preparations were fixed with buffered 2 . 5% glutaraldehyde for one hour at 4°C and rinsed five times with 50 mM cacodylate buffer . Post-fixation was accomplished with 4% buffered osmium tetroxide for 90 minutes on ice . Preparations were washed five times with ddH2O on ice and stained en bloc with 0 . 5% aqueous uranyl acetate over night at 4°C . They were washed five times with ddH2O on ice and were dehydrated with graded series of ethanol . ( 50% , 70% , 90% , 96% and two times 100% ethanol at 4°C for 30 minutes each , fresh 100% ethanol and three times propylene oxide at room temperature for 30 minutes each ) . The preparations were then incubated in a 1+1 mixture of Epon ( Serva Electrophoresis GmbH; Heidelberg; Germany ) and propylene oxide over night and twice in pure Epon for two hours at room temperature . Finally , the larvae were embedded in Epon and polymerized at 60°C for 48 hours . Longitudinal ultrathin ( 80 nm ) sections of the larval bundles of axons were cut using a diamond knife . The grids were post-stained with 2% uranyl acetate for 20 minutes and with Reynold's lead citrate for seven minutes . For quantitative analysis sections spaced more than 1 . 5 µm apart were selected to avoid counting the same electron-dense structure multiple times . The nerve sections were analyzed with a Leo 912 AB transmission electron microscope ( Zeiss SMT; Oberkochen; Germany ) at 630× magnification and the cross-section area was measured with the polygon tool of iTEM software ( Soft Imaging System; Muenster; Germany ) . Nerves were screened for conspicuous electron-dense structures at 40000× magnification . These were digitally photographed and their position in the nerve was marked at 16× magnification . Identification and counting of electron-dense structures were done blinded . The diameter of the agglomerates was measured with iTEM as the largest distance in the electron dense field . Mean values and standard error of the mean ( SEM ) were calculated . For ultrastructural localization of Bruchpilot , wandering wild-type and null-mutant ( Srpk79DVN ) larvae were prepared in ice-cold calcium-free saline ( 130 mM NaCl , 36 mM Sucrose , 5 mM KCl , 1 . 9 mM MgCl2 , 5 . 5 mM HEPES , 0 . 5 mM EDTA ) , fixed in 2% paraformaldehyde with 0 . 06% glutaraldehyde in 1× PEM ( 0 . 1 M PIPES , 2 mM EGTA , 1 mM MgSO4×7H2O ) for 90 min on ice , washed twice for 15 min each in 1× PEM , blocked for 1 h in 2% BSA/3% normal horse serum ( NHS ) in PBS containing 0 . 2% Triton-X 100 and incubated overnight at 4°C with the primary monoclonal antibody nc82 diluted 1∶100 in PBST . Larval filets were then washed four times for 1 h in PBST , incubated for 1 h with the secondary antibody Alexa Fluor488 FluoroNanogold-anti-mouse Fab' ( Nanoprobes; Yaphank; NY; USA ) diluted 1∶20 in PBST , washed for 30 min in PBST and then washed overnight at 4°C in PBST . After washing twice for 30 min in PBST , the preparations were post-fixed for 30 min in 2% glutaraldehyde in PBS and washed four times for 10 min in distilled H2O . After the silver enhancement performed for 1 h using the AURION R-GENT SE-EM Kit , the preparations were washed four times for 10 min in distilled H2O , fixed for 30 min in 2% OsO4 in 50 mM cacodylate buffer ( pH 7 . 2 ) and washed overnight at 4°C in distilled H2O . After washing twice for 30 min in distilled H2O and dehydration in ascending ethanol series ( 30 min each in 50% , 70% , 90% , 96% on ice; twice 100% at room temperature ) , the tissue was incubated twice for 30 min in propylene oxide and then in a 1∶1 mixture of propylene oxide and Epon overnight , followed by two 2-h incubation periods in pure Epon . The tissue was then transferred to gelatine capsules with pure Epon and polymerization was allowed to proceed for 3 days at 60°C . Ultrathin ( 70 nm ) sections were cut and transferred to copper grids which were then contrasted with 2% uranyl acetate for 20 min , washed and incubated in Reynold's lead citrate for 10 min and subjected to a final wash . The walking and the flying assays were performed as described [7] . Briefly , for the walking assay individual flies with clipped wings were allowed to walk on a horizontal surface marked with a 2×2 cm grid . The number of lines crossed within 30 s was counted and recorded 3 times for each fly . At least 25 flies of each genotype were tested . In the flight assay groups of 100 flies were tossed through a funnel into a 500 ml cylinder ( of 50 mm diameter ) whose walls had been coated with paraffin oil . Poor fliers drop to the bottom of the cylinder and were counted . 10 groups were tested for each genotype . For life span analyses a total of at least 500 flies were tested . 50 to 100 newly hatched male flies were transferred to fresh food vials every three to four days and the dead flies were counted . Males were chosen because survival of isolated males is less variable than survival of females or mixed populations . Wild-type controls were w1118 since it represents the genetic background of all experimental lines . Values are given as mean and SEM . Statistical significance of the difference to wild type was calculated using the Student's t-test with Bonferroni correction for 6 comparisons in each set of experiments . Larval olfactory conditioning was assayed as described [65] . Briefly , a group of 30 to 35 larvae was exposed for 5 min to vapor from undiluted 1-octanol ( O+ ) while crawling on 1% agarose containing 2 M fructose , transferred to pure 1% agarose and exposed for 5 min to vapor from n-amyl acetate ( A ) ( diluted 1∶50 in paraffin oil ) . This procedure was repeated 3 times , after which they were transferred to pure 1% agarose in the presence of antagonistic gradients of both odors . After 3 min the number of larvae closer to each odor source was counted and the preference index PIO+ = ( # near O−# near A ) /total # was calculated . Next a new group of larvae was treated equivalently , only now A instead of O was presented in the presence of fructose and PIA+ was calculated . The total procedure was repeated 10 times and the learning index LI = ( PIO+−PIA+ ) /2 was calculated . Prior to the experiment the Srpk79DVN mutant had been backcrossed to wild type CS for 6 generations to minimize genetic background effects . The experiments were done blind with respect to the genotype .	Neurons communicate through release of neurotransmitters at specialized contacts called synapses . Modulation of synaptic transmission likely underlies all higher brain function including feature abstraction , learning and memory , and cognition . The complex molecular machinery that regulates neurotransmitter release has been conserved in evolution but is still incompletely understood . Using the genetic model organism Drosophila , we recently discovered a protein of the presynaptic ribbon ( T-bar ) that was called Bruchpilot ( German for crash pilot ) because flies with reduced amounts of this protein cannot fly . We now screened various Drosophila mutants for changes in tissue localization of Bruchpilot and discovered a gene that codes for an enzyme which is similar to mammalian kinases that phosphorylate splicing factors and may co-localize with Bruchpilot at the synapse . Larval nerves of mutants for this gene contain conspicuous accumulations of Bruchpilot that correspond to extensive electron-dense ribbon-like agglomerates surrounded by vesicles . While general axonal transport and basic synaptic transmission at larval nerve-muscle synapses are not affected , adult mutants show reduced life span and impaired flight and walking . The substrate for this kinase and its role in maintaining brain function must now be identified . Its discovery raises important questions about the function of homologous proteins in mammals including humans .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "genetics", "and", "genomics/gene", "discovery", "genetics", "and", "genomics/disease", "models", "physiology/cell", "signaling", "genetics", "and", "genomics/gene", "function", "neuroscience/neuronal", "signaling", "mechanisms", "physiology/membranes", "and", "sorting" ]	2009	Bruchpilot in Ribbon-Like Axonal Agglomerates, Behavioral Defects, and Early Death in SRPK79D Kinase Mutants of Drosophila
CDKs ( cyclin-dependent kinases ) associate with different cyclins to form different CDK-complexes that are fundamental for an ordered cell cycle progression , and the coordination of this progression with different aspects of the cellular physiology . During meiosis programmed DNA double-strand breaks ( DSBs ) initiate recombination that in addition to generating genetic variability are essential for the reductional chromosome segregation during the first meiotic division , and therefore for genome stability and viability of the gametes . However , how meiotic progression and DSB formation are coordinated , and the role CDKs have in the process , is not well understood . We have used single and double cyclin deletion mutants , and chemical inhibition of global CDK activity using the cdc2-asM17 allele , to address the requirement of CDK activity for DSB formation and recombination in fission yeast . We report that several cyclins ( Cig1 , Cig2 , and the meiosis-specific Crs1 ) control DSB formation and recombination , with a major contribution of Crs1 . Moreover , complementation analysis indicates specificity at least for this cyclin , suggesting that different CDK complexes might act in different pathways to promote recombination . Down-regulation of CDK activity impinges on the formation of linear elements ( LinEs , protein complexes required for break formation at most DSB hotspot sites ) . This defect correlates with a reduction in the capability of one structural component ( Rec25 ) to bind chromatin , suggesting a molecular mechanism by which CDK controls break formation . However , reduction in DSB formation in cyclin deletion mutants does not always correspondingly correlate with a proportional reduction in meiotic recombination ( crossovers ) , suggesting that specific CDK complexes might also control downstream events balancing repair pathways . Therefore , our work points to CDK regulation of DSB formation as a key conserved feature in the initiation of meiotic recombination , in addition to provide a view of possible roles CDK might have in other steps of the recombination process . Eukaryotic cell cycle progression is driven by sequentially organized accumulation of different CDK ( cyclin-dependent kinase ) activities formed by a catalytic serine/threonine kinase that binds to a regulatory cyclin subunit [1–3] . In Schizosaccharomyces pombe a single CDK ( Cdc2 ) and six different cyclins have been described . Cig1 , Cig2 and Puc1 cyclins control G1 progression , meanwhile Cdc13 is essential to promote chromosome segregations [4–9] . Though at least in this yeast a single CDK complex ( Cdc2 kinase-Cdc13 cyclin ) can promote both mitotic and meiotic progression , it is not as efficient as in the wild-type situation where additional CDK complexes are present [8 , 9] . This indicates that distinct cyclins have evolved to optimize different aspects of the mitotic and the meiotic divisions and that some kind of specificity is provided by each CDK complex ( Cdc2-Cyclin ) . Indeed , two of the described cyclins ( Rem1 and Crs1 ) are meiosis-specific [10 , 11] , suggesting meiosis-specific functions for these CDK-complexes . Meiosis is a special cell division where a single round of DNA replication is followed by two rounds of chromosome segregation . In the first reductional segregation homologous chromosomes separate apart , and the physical links provided by recombination between the pair of homologs is required for the orientation in the meiotic spindle , and therefore for their successful segregation [12] . Thus , a key feature of meiosis is self-inflicted DNA double-strand break ( DSB ) formation that initiates natural recombination at specific genome locations known as hotspots [13] . Meiotic DSBs are generated by a conserved topoisomerase II-like protein , Spo11 ( Rec12 in fission yeast ) , assisted by a group of accessory proteins forming the conserved pre-recombination complexes SFT and DSBC in fission yeast [13 , 14] . In addition to the DSB machinery , DSB formation requires a meiosis-specific chromosome context , provided by histone variant H2A . Z , meiosis-specific cohesin subunits ( Rec8 and Rec11 in fission yeast ) , and Linear Elements ( structurally related to the axial/lateral elements of the synaptonemal complex of other eukaryotes ) [15–18] . Indeed , meiotic cohesins are required for LinE formation and chromosome loading of LinE components [14–16 , 19–21]; specifically , casein kinase 1-dependent phosphorylation of the meiotic cohesin subunit Rec11 is required for the interaction with the LinE-component Rec10 and LinE formation [22 , 23] . After break formation Spo11 ( Rec12 ) covalently linked to DNA is endonucleolytically removed and resection generates single-stranded DNA ( ssDNA ) tracts that , coated with strand-exchange proteins Rad51/Dmc1 , invades homologous chromosome for homology search and repair [24–30] . This invasion generates by strand displacement and DNA synthesis the so-called D-loop that can be dissolved , and the invading nucleoprotein filament annealed with its sister chromatid in the original chromosome ( synthesis-dependent strand annealing ) , resulting in non-reciprocal exchange between the parental chromosomes ( non-crossovers , NCOs ) . Alternatively , the D-loop can be stabilized and mature into recombination intermediates , Holliday junctions . The way these intermediates are resolved by structure-dependent endonucleases will generate crossovers ( COs , reciprocal exchange between the pair of homologs ) or NCOs [31–33] . Thus , the fate of the D-loop is an important point for CO/NCO regulation . Stabilization of the invading nucleoprotein filament promotes D-loop stabilization and therefore CO formation , meanwhile the counteractions of helicases dissolve the D-loop and therefore hamper CO formation [33–36] . CO homeostasis and CO invariance mechanisms have been proposed to maintain minimal levels of COs and their regular distribution along chromosomes [37–40] . DSB formation occurs after meiotic DNA replication during meiotic prophase . In budding yeast when DNA replication is locally delayed , DSB formation is also locally retarded , pointing to the coordination between both events [41] . Moreover , in fission yeast local changes in origin selection during meiotic DNA replication lead to local changes in the distribution of meiotic recombination [42] . It was proposed that replication-origin firing leads to the recruitment of recombination factors . This hypothesis was recently supported in budding yeast . In this yeast , kinase activities involved in cell cycle regulation are important for DSB formation . S-phase specific CDK and DDK ( Dbf4-dependent kinase ) activities phosphorylate Mer2 ( fission yeast Rec15 ortholog , component of the conserved SFT-complex ) . Mer2 phosphorylation promotes the interaction with other Spo11-accessory proteins , and is essential for association of Spo11 with hotspots and DSB formation [43–46] . A link between Mer2 phosphorylation and DNA replication has not formally been established; however , replication-fork passage correlates with chromosome loading of Rec114 ( a Spo11-accessory protein whose chromatin association depends on Mer2 phosphorylation ) ( fission yeast Rec7 ortholog ) that may promote the local formation of the pre-recombination complexes [30 , 47–49] . However , DNA replication per se is not necessary for DSB formation since inhibition of S-phase initiation by down-regulation of replication factors , both in budding and fission yeast , does not abrogate DSB formation [50–52] . Thus , this coordination depends on an active DNA replication and , indeed , the S-phase checkpoint blocks DSB formation when replication is stalled [53 , 54] . In fission yeast checkpoint inhibition of DSB formation works , at least in part , by repressing the expression of the transciption factor gene mei4 which , in turn , controls the expression of mde2 ( coding for one of the Rec12-accessory proteins ) [14 , 55 , 56] . In budding yeast , among other mechanisms , checkpoint activation down-regulates DDK activity , thereby preventing Mer2 phosphorylation and DSB formation [54] . Unscheduled DSB formation on partially replicated chromosomes generates unrepaired breaks that hamper further replication , and impinges on cell viability [54] . DSB formation is a key conserved feature of meiosis and , apart from the sequence conservation of Spo11 homologs , proteins of the pre-recombination complexes have amino acid similarity and some of them are even structurally conserved among different species [14 , 57–60] , suggesting that regulation by cell cycle kinases could be also a conserved feature . In fission yeast , DDK activity is also required for DSB formation and recombination but the nature of this regulation is currently unknown [61 , 62] . In the case of CDK , DSB formation is still observed when CDK activity is down-regulated [50]; however , DSBs were not quantified in that study and CDK requirement was not addressed . We have studied the role of CDK activity in DSB formation in fission yeast by analyzing the effects of the depletion of CDK complexes normally present in meiotic prophase . We have found that Cig1 , Cig2 , and the meiosis-specific Crs1 cyclin control indeed DSB formation and recombination , with a major contribution of Crs1 , and the stronger reduction of DSBs in the double deletion mutant cig1 crs1 . At least for Crs1 , complementation analysis of the recombination phenotype by increasing copy number of other cyclins and Cdc2 suggests specificity . The absence of these cyclins reduces binding to chromatin of the LinE-component Rec25 , and impairs the maturation of these structures . We have obtained similar results when global CDK activity was down-regulated using an ATP-analog sensitive cdc2-asM17 allele . This study points to CDK regulation of DSB formation as a conserved feature in the initiation of meiotic recombination . Furthermore , comparison of DSB , NCO and CO levels suggests that CDK activity might also control downstream events after DSB formation . Therefore , CDK activity in meiosis may regulate different steps of the recombination process . Meiotic DSB formation was previously analyzed in fission yeast using a temperature- sensitive cdc2 mutant , cdc2-L7 [50] . Using a thermal induction of meiosis in haploid cells , DNA breaks were visualized by separating chromosomes in pulsed-field gel electrophoresis ( PFGE ) and evaluating chromosome fragmentation during meiotic prophase . Rec12-dependent chromosome breakage was observed after meiotic induction at the restrictive temperature , indicating that Cdc2 is not essential to initiate meiotic recombination . However , since DSBs were not quantified in the study , the contribution of CDK to DSB formation was not evaluated . We have revisited this result and , since Cdc2 is essential for meiotic progression , decided to analyze first the recombination efficiency of cyclin deletion mutants . We have focused on cyclins that , with a clear temporal expression pattern ( mRNA , protein , and/or associated kinase activity ) during meiosis , may play a role in recombination . Cig1 , Cig2 and Crs1 cyclins were selected based on this criterion . Cig1 and Cig2 cyclins are expressed around S-phase and prophase , and Cig2 required for meiotic DNA replication ( http://www . pombase . org/spombe/result/SPCC4E9 . 02 and http://www . pombase . org/spombe/result/SPAPB2B4 . 03; expression viewer and pombeTV ) [63 , 64] . The cyclin-related protein Crs1 shows a very high level of expression during meiotic prophase and was identified in our initial screening for mutants affecting meiotic chromosome segregation , compatible with a recombination defect ( http://www . pombase . org/spombe/result/SPBC2G2 . 09c; http://telecic . cicancer . org/pombe/ ) [65] . In the absence of Cig1 , Cig2 , and Puc1 cyclins , Crs1 contributes to meiotic G1 progression [9] . Meanwhile Cig1 and Cig2 are also expressed and contribute to mitotic progression [4 , 5] , Crs1 shows a complex regulation to prevent RNA accumulation in vegetative cells [66 , 67]; indeed , mis-expression of this cyclin in mitotic cycles causes segregation problems and lethality [10 , 68] . Though expressed later in meiosis and involved in meiosis I entry , the other meiosis-specific cyclin Rem1 has an additional role in meiotic recombination . However , this does not depend on the presence of the cyclin-box in the protein , suggesting that it is Cdc2-independent [11 , 69]; therefore , we did not include it in this study . In the case of puc1 , expression is almost undetectable during meiotic prophase , where it is even less abundant than in vegetative cells ( https://www . pombase . org/gene/SPBC19F5 . 01c ) . Meiotic recombination was addressed both in intragenic ( NCOs measured as gene conversion at ade6 on chromosome III ) and intergenic ( COs in the leu1-his5 interval on chromosome II ) recombination assays in single and double deletion mutants . Single deletion mutants of cig1 and cig2 were defective in recombination , with a moderate reduction in gene conversion ( 26% p value 0 . 004 and 27% p value 0 . 016 , respectively ) , and without additive effects in the double cig1 cig2 mutant ( 33% reduction; p value 0 . 025 ) , suggesting that both cyclins might act in the same genetic pathway ( Fig 1A ) . Regarding COs ( Fig 1C ) , cig1 deletion mutants did not show a defect and levels of crossovers were similar to the control . Interestingly , levels of crossovers in the cig2 deletion mutant were even higher than in the control cross ( 1 . 28-fold higher; p value 0 . 034 ) . In the case of the meiosis-specific Crs1 cyclin , crs1 deletion mutants showed stronger defects both in gene conversion , with 53% of the gene conversion level shown in the control ( p value 2 . 1 10−6 ) , and in COs , with 61% of the control levels ( p value 0 . 004 ) ( Fig 1B and 1C ) . Finally , we analyzed the double cig1 crs1 mutant , and found that recombination levels were similar to those in the single crs1 mutant , with 51% of the gene conversion and 56% of the CO level shown in the control ( p value 0 . 006 and 0 . 002 , respectively ) ( Fig 1B and 1C ) , indicating the strong phenotype of the crs1 deletion . This reduction in recombination does not have an impact on spore viability ( S1 Fig ) . Crs1 was recently reported not to have a role in meiotic recombination using different markers for the recombination assays [9] . Since mutants defective in chromosome architecture , such as meiosis-specific cohesin mutants and LinE-mutants , have a regional defect in recombination [15–17 , 70 , 71] , we decided to address recombination in the genetic intervals used in that study . The deletion of crs1 reduced recombination , both gene conversion and crossovers , at ade6 ( ade6-M26 ade6-M210 interval ) and mat1-leu1 interval , to a similar extent as in our previous recombination assays with a 63% of the gene conversion and 45% of the crossover levels shown in the control crosses ( Fig 1D and 1E ) . These genetic data indicate that Cig1 , Cig2 , and Crs1 cyclins control meiotic recombination . The recombination defect led us to analyze DSB formation at the strong natural hotspot mbs1 on chromosome I ( Fig 2A ) [72] . Diploid cyclin-deletion mutants , previously G1-arrested by nitrogen depletion , were induced to enter meiosis synchronously by thermal induction using a temperature-sensitive allele of the meiotic inhibitor Pat1 ( pat1-114 ) [73] . DSB formation was analyzed by Southern blot with DNA samples from cells collected at different time points before and after the induction , including time points when DSBs were expected to form and be repaired ( Fig 2B ) . In the control experiment DSBs at mbs1 were detected from 2 to 4 hr after meiotic induction and then started to disappear due to break repair . Clear levels of breakage were observed from 2 . 5–3 . 5 hr , corresponding to cells in prophase according to flow cytometry and nuclear counting ( S2 Fig ) , with a maximum of 7 . 8% breakage at 3 hr ( Fig 2B ) . The kinetics of DSB appearance and disappearance at mbs1 in cig1 deleted cells was similar to that in the control; however , though not statistically significant , levels of breakage were reduced by 28% compared to the control levels ( p value 0 . 094 ) , with a maximum of 5 . 6% breakage at 3 hr . Similar results were obtained in cig2 deletion mutants with a maximum of breakage of 6 . 2% ( 25% reduction , 8 . 3% maximum control levels; p value 0 . 210 ) , in this case at 3 . 5 hr since this mutant showed a delay in S-phase entry and progression ( S2 Fig ) . The double mutant cig1 cig2 showed a reduction in DSB formation similar to that in the single deletion mutants , with a breakage of 5 . 3% ( 36% reduction , 8 . 3% maximum control levels , p value 0 . 060 ) ( 1–0 . 72x0 . 75 = 0 . 46 expected reduction for an additive effect ) . Reduction in DSB formation was also analyzed genome-wide by entire chromosome visualization . DSB formation appears during prophase as a smear below the intact chromosomes that disappears after repair at later time points . Although this type of analysis is not quantitative , cig1 , cig2 , and cig1 cig2 deletion mutants seemed to reduce the smear intensity suggesting a possible general reduction in DSB formation in these mutants ( Fig 2C ) . DSB levels were similarly analyzed in crs1 deletion mutants ( Fig 3 and S3 Fig ) . crs1 mutants progressed through meiosis with the same timing as the wild-type strain ( S4 Fig ) and DSB formation was clearly detected from 2 . 5–3 . 5 hr after meiotic induction ( Fig 3A ) ; however , levels of breakage were statistically reduced by 45% of the level in the control experiment ( p value 0 . 025 ) , with a maximum of 3 . 7% breakage at 3 hr ( 6 . 9% maximum control levels ) . The double mutant cig1 crs1 showed a stronger reduction in DSB formation and breaks at mbs1 were reduced 58% , with a maximum level of 3 . 1% at 3 hr ( 7 . 5% maximum control levels , p value 0 . 048 ) ( 1–0 . 72x0 . 55 = 0 . 60 expected reduction for an additive effect ) . The defect in DSB formation of crs1 and the double cig1 crs1 mutant were also visible when entire chromosomes were analyzed ( S3 Fig and Fig 3B ) . This physical analysis of the cyclin deletion mutants indicates that Crs1 , and probably Cig1 and Cig2 , control DSB formation . Next , we decided to revisit the requirement of Cdc2 for DSB formation using an inhibition of global CDK activity by means of the ATP-analog sensitive cdc2-asM17 allele [74] . When 1-NM-PP1 ATP-analog was added at the beginning of the time course in cells arrested in G1 , just before thermal induction of entry into meiosis , S-phase entry and chromosome segregation were blocked as expected for the inhibition of Cdc2-kinase activity ( Fig 4A ) ; and more importantly , DSB formation was undetectable at mbs1 and the other break sites in the NotI J fragment ( Fig 4B , top panel ) . The impact on DSB formation after the inhibition of Cdc2 activity was also observed when entire chromosomes were analyzed . No smears below the intact chromosomes were detected compared to the control experiment ( Fig 4B , bottom panel ) . A similar result was obtained when the ATP-analog was added later , at 2 hr after meiotic induction when cells were exiting S-phase and progressing into prophase: DSB formation was significantly reduced and chromosome segregation was blocked ( Fig 5 ) . This was particularly clear at the initial time points of the experiment ( indicated by a double-headed arrow in Fig 5B ) when DSBs were normally detected in the control . DSB formation was activated at later time points probably due to analog inactivation ( see also legend of Fig 4A ) . Since replication defects trigger the S-phase checkpoint to block DSB formation [53 , 56] , we performed the experiment using the rad3 deletion mutant ( coding for the apical sensor kinase in the checkpoint signaling pathway ) [75] to exclude the possibility that the absence of DSB formation results from checkpoint activation due to incomplete replication upon Cdc2 inhibition . Again , a significant reduction in DSB formation at hotspots in the NotI J fragment was observed after inhibition of Cdc2 even when checkpoint activation was abrogated ( S5 Fig ) . These results complement the previous results and support a role of CDK activity in meiotic DSB formation in fission yeast . DSB formation at nearly all hotspots requires LinEs . When these structures are absent , in deletion mutants of their components , DSB formation at the NotI J fragment containing the hotspot mbs1 is extremely impaired and , in the cases genome-wide analyzed , DSB formation at most hotspots is abolished [15 , 17 , 65 , 76] . Therefore , we addressed LinE formation as a possible point of regulation by CDK activity . LinE formation was visualized in intact cells during synchronous diploid pat1-114 meiosis using a Rec25-GFP version , and was first analyzed in double cig1 crs1 deletion mutants that as shown above exhibit the stronger reduction in DSB formation ( Fig 6 ) . Synchrony of the experiment was followed by cytometry and counting of nuclei , and as described above ( S4 Fig ) double cig1 crs1 cells showed a normal timing of meiotic events ( S-phase , MI and MII entry ) ( S6A Fig ) . Rec25-GFP signal was quantified by counting the percentage of cells with no signal , diffuse nuclear signal , diffuse nuclear signal plus foci , and mature signal ( clear foci with no background diffuse signal ) , following the natural dynamics of LinE formation in intact cells ( see Fig 6A and S6B Fig for cells representing each category ) [15] . The kinetics of accumulation of Rec25-GFP signal was similar in control and cig1 crs1 cells , reaching almost 100% of the population during prophase ( Fig 6A ) . Similarly , the earliest transient signal ( diffuse nuclear signal ) appeared and disappeared with the same kinetics during S-phase ( 1 . 5–2 . 5 hr; see S6A Fig for meiotic progression ) . However , although the next transient signal ( diffuse nuclear+foci ) appeared at the same time in cig1 crs1 mutant and control cells , it accumulated in a higher proportion ( 76% of the cells compared to the 39% in the control experiment ) , and disappeared later . Moreover , the latest signal ( mature signal ) reached 67% of the population at 3 hr after meiotic induction in the control experiment , but in the cig1 crs1 deletion mutant was observed in only 22% of the cells ( Fig 6A and 6B , S6B Fig ) . This result indicates that the maturation of LinEs is defective in the absence of Cig1 and Crs1 cyclins . We noticed that at late time points in prophase , 3 . 5 and 4 hr after meiotic induction , cells with a tangled Rec25-GFP signal were frequently present in the cig1 crs1 mutant ( 27% and 38% respectively compared to 4% and 5% in the control cells ) ( S6B Fig ) . The defect in LinE maturation and the frequency of this tangled signal prompted us to perform a cellular fractionation assay to address the ability of Rec25-GFP to bind chromatin in the absence of Cig1 and Crs1 ( Fig 7 ) . Briefly , cells were collected in prophase at 3 hr after meiotic induction when LinEs reach their maximal maturation stage , treated with zymolyase for cell wall digestion , and subjected to a hypotonic lysis . The whole extracts ( WCE ) were centrifuged through a sucrose cushion to collect the cytoplasmic ( SB1 ) and the nuclear ( PP1 ) fractions . Nuclear fractions were treated with detergent to solubilize the nucleoplasmic proteins , and centrifuged to separate into soluble fractions ( SB2 ) and the nuclear-insoluble fractions ( PP2 ) that were gently sonicated for further solubilization generating the final SB3 and PP3 fractions , corresponding to chromatin-bound proteins and highly insoluble nuclear proteins ( Fig 7A ) . In the control experiment , Rec25-GFP protein was detected in the same fractions as the Histone H4 ( PP1 , PP2 , and SB3 ) , indicating its capability to bind chromatin and its resistance to detergent extraction ( Fig 7B , left ) . However , although Histone H4 was detected in the same fractions in the double cig1 crs1 deletion mutant , the amount of Rec25-GFP substantially increased in the nucleoplasmic fraction ( SB2 ) , 2 . 16-fold compared to the control ( Fig 7B , right ) . Thus , in the absence of Cig1 and Crs1 cyclins the binding to chromatin of the LinE-component Rec25 is less efficient . Both LinE maturation and Rec25-GFP chromatin binding were similarly analyzed using the cdc2-asM17 allele to control global CDK inactivation in prophase . Interestingly , maturation defects and reduced chromatin binding were also observed in the cdc2-asM17 mutant without treatment ( Figs 7C and 8 ) , and these defects were more pronounced when CDK activity was depleted by the ATP-analog ( Figs 7C and 9 ) . In the experiment without treatment , the kinetics of appearance and disappearance of the earliest transient signal ( diffuse nuclear signal ) was sharper in the wild-type control . Rec25-GFP signal was observed earlier in the cdc2-asM17 mutant than in the control ( Fig 8B ) , probably due to the faster S-phase progression of the mutant and advanced Rec25-GFP expression ( S7 Fig and Fig 8A ) . At 1 hr after meiotic induction 46% of the population already exhibited this signal , whereas in the control it did not appear until 1 . 5 hr; however , only 61% of the population in the cdc2-asM17 mutant showed the diffuse nuclear signal at 1 . 5 hr after meiotic induction compared to 100% in the wild-type control . Despite this fact , the kinetics of appearance and disappearance of the following transient signal ( diffuse nuclear+foci ) was similar in control and cdc2-asM17 mutant cells , as well as meiosis I entry ( Fig 8A and 8B ) , indicating an extended prophase in the mutant . Both control and mutant cells showed a similar proportion of cells with this signal at 2 . 5 hr after meiotic induction ( 76% and 65% respectively ) . However , at 3 hr only 35% of the mutant population exhibited the latest mature signal compared to the 66% of the wild-type control . This suggests that expression and nuclear localization of Rec25-GFP are normal in the cdc2-asM17 mutant in which lower CDK activity has been reported [74] , but maturation of LinEs into clear nuclear foci is compromised . Next , we studied LinE formation after inhibition of global CDK activity by addition of 1-NM-PP1 when cells were exiting S-phase and progressing into prophase ( 2 hr after meiotic induction; same experimental design as the one described above for the study of DSB formation ) ( Fig 9A ) . ATP-analog addition caused a reproducible delay ( in two independent experiments ) in the disappearance of the transient diffuse nuclear signal; at 3 hr after meiotic induction 20% of the treated population showed this type of signal compared to 6% of the DMSO-treated control cells ( Fig 9B ) . The next transient signal ( diffuse nuclear+foci ) reached a similar proportion in both situations at 3 hr; however , in 1-NM-PP1 treated cells the signal did not progress properly and , instead of decreasing at later times points as observed in the control population , continued to accumulate . This population showed a high proportion ( 76% at 3 . 5 hr ) of cells with tangled Rec25-GFP signal compared to the DMSO treated control ( 7% ) . More importantly , Rec25-GFP mature signal was observed in only 13% of the cells compared to 32% in the control . Regarding chromatin-binding capability of Rec25-GFP protein , we observed an increase of 1 . 36-fold in the nucleoplasmic fraction ( SB2 ) in the cdc2-asM17 untreated mutant compared to the wild-type control , and an increase of 1 . 49-fold in the cdc2-asM17 1-NM-PP1 treated mutant compared with the DMSO treated control ( Fig 7C ) . Similar results , in terms of LinE maturation and Rec25-GFP susceptibility to detergent treatment , were obtained in an independent experiment . Thus , the results using the cdc2-asM17 mutant are consistent with the ones obtained in the double cig1 crs1 deletion mutant , and support a role for CDK activity in the formation/maturation of LinEs . In an effort to identify CDK substrates involved in DSB formation we have done direct mutagenesis of several proteins essential for break formation at hotspots: Rec10 ( LinEs ) , Rec27 ( LinEs ) , Rec7 ( SFT-complex ) and Rec14 ( DSBC-complex ) [14 , 15 , 17 , 65 , 77] ( S8 Fig ) . Given the implication of CDK activity in LinE maturation reported here , we paid special attention to LinE components when generating mutants . Of the four LinE components described only the phosphoprotein Rec10 [78] and Rec27 harbor CDK phosphorylation sites , 8 and 1 respectively . We thought phosphoprotein Rec7 ( SFT-complex ) [14] could be also a good candidate since the S245 in the TSSPFN context is adjacent to T243 and S244 , which are potential sites for DDK activity , resembling the S28-S29-S30 cluster of amino acids in Mer2 that are subjected to CDK-priming phosphorylation , and subsequent DDK phosphorylation [44 , 45] . Finally , we selected the conserved Rec14 protein ( DSBC-complex ) harboring 5 CDK sites . We have changed the putative CDK phosphorylated residues ( in minimal S/T-P and consensus S/T-P-X-K/R/ ( N ) context ) to alanine generating phospho-null mutants at these residues of these proteins . These residues show a good prediction score in PhosphoNet 2 . 0 and/or Phospho Yeast 1 . 0 software , and in some cases ( S347 , T482 , and S529 residues in Rec10 ) are phosphorylated in vivo during meiotic prophase [78] . The mutants were genetically analyzed to score for defects in gene conversion in intragenic recombination assays . None of the mutants impaired recombination rates , except for rec14 ( cdk1 ) mutants where a moderate 20% reduction was observed ( S9 Fig ) . Recombination rates were not further reduced in the rec14 ( cdk total ) mutant harboring mutations in all the putative CDK phosphorylated residues . Furthermore , combinations of some of these mutants ( rec7 , rec14 , and rec27 ) did not reduce gene conversion in qualitative recombination assays ( S10 Fig ) . It is not well established that control of DSB formation by CDK activity is a universal feature of meiosis . Addressing this issue in the fission yeast S . pombe , we have found that cig1 and cig2 cyclin deletion mutants are indeed impaired in meiotic recombination , and NCOs reduced 26% compared to the control levels observed in wild-type strains . Correspondingly , DSB formation is also reduced to a similar extent at the hotspot of reference mbs1 , 28% and 25% respectively . Non-additive defects in NCO and DSB levels were observed in the double cig1 cig2 mutant , indicating these CDK-complexes may act in the same genetic pathway ( Fig 1A and Fig 2 ) . However , CO formation was not correspondingly diminished , and in the case of cig2 mutants CO levels were even increased ( Fig 1C ) . This result suggests that the moderate reduction in DSBs may be real and that in the absence of these CDK-complexes the reduced DSB levels activate homeostatic mechanisms to maintain CO levels , since they are essential to ensure a successful segregation of chromosomes and , therefore , for the viability of the meiotic products . This phenomenon , known as crossover homeostasis , has been described in several organisms , including budding and fission yeast [37 , 38 , 40] . Alternatively , given the increase of COs in cig2 mutants , these CDK-complexes may control downstream events in the recombination process . One possibility is that they negatively control CO formation by regulating the stability of the D-loop after homolog invasion . In fission yeast it has been proposed that Rad51/Dmc1 accessory proteins protect the D-loop from the unwinding action of helicases , promoting in this way the formation of Holliday junctions and COs [33 , 34 , 36] . In this view , Cdc2-Cig2 ( and Cdc2-Cig1 to a minor extent ) could phosphorylate and inhibit accessory proteins required for nucleoprotein filament stabilization and strand-exchange activity [79–82] , impairing D-loop stability . Different phosphoproteomic approaches have identified S/T phosphorylated residues in the Rad51/Dmc1-accesory protein Sfr1 in vegetative cells [83–86]; four of them are putative CDK phosphorylation sites , and at least one of them ( S165 ) is phosphorylated in a Cdc2-dependent manner during mitotic M-phase [86] . Alternatively , Cdc2-Cig2 ( and Cdc2-Cig1 to a minor extent ) could phosphorylate and activate Fml1 ( or other helicases counteracting D-loop formation ) [33 , 36] . As Sfr1 , Fml1 harbors putative phosphorylation sites by Cdc2; however , in this case phosphorylation of these residues has not been reported in phosphoproteomic studies . In addition to Cig1 and Cig2 , the meiosis-specific Crs1 cyclin is also required for DSB formation and recombination , in this case to a greater extent . In the absence of Crs1 both NCOs and COs are similarly reduced in the different tested intervals , 37–47% and 39–55% respectively ( Fig 1B , 1C , 1D and 1E ) . This reduction correlates well with a corresponding 45% reduction in DSB formation both in proficient and deficient DSB-repair conditions ( Fig 3 and S3 Fig ) . The proportional reduction of NCOs and COs indicates that the levels of DSBs observed in this mutant could be under the threshold level to activate CO homeostasis . Interestingly , DSBs in double cig1 crs1 deletion mutants are further reduced ( 58% ) . Although not statistically significant compared to the DSB levels in the single crs1 deletion mutant , this reduction is the expected one for an additive effect ( 1–0 . 72x0 . 55 = 0 . 60 ) , indicating these CDK-complexes might control DSB formation acting in genetically independent pathways ( Fig 3 ) . In spite of this reduction in DSBs , NCOs and COs are not correspondingly affected , and the levels in the double mutants are similar to those observed in the single mutant crs1 ( Fig 1B and 1C ) . A possible explanation for this discrepancy is that in the absence of Cig1 repair is biased towards homologous chromatids and not sister chromatids , which in fission yeast is the most common template used ( 1:3 proportion ) [87] . This observation could imply that Cdc2-Cig1 contributes to crossover invariance , a phenomenon that suggests differential partner choice for repair at hotspots ( with the sister chromatid ) and coldspots ( with a homologous chromatid ) to maintain a constant chromosome CO distribution in spite of different frequency of DSB formation across the genome [39] . Alternatively , the reduction in DSBs may be locus-dependent in the double cig1 crs1 deletion mutant . The requirement of cyclins for meiotic recombination has been recently studied , and Cig1 , Cig2 , and Crs1 reported not to have a role in the process [9] . In the case of Cig1 and Cig2 , recombination was exclusively addressed in intergenic recombination assays , and wild-type levels of COs reported . Accordingly , we have not found reduction in CO levels in these mutants; however , DSBs and NCOs are reduced indicating that indeed these CDK-complexes regulate meiotic recombination . Additional roles of these cyclins in steps downstream of DSB formation may obscure the outcome of the intergenic recombination assays . In the case of Crs1 , both NCOs and COs were genetically analyzed in the published work , and normal levels reported in the crs1 mutant . However , we have consistently found statistically significant DSB , NCO , and CO reductions in this mutant ( complete ORF deletion , see Material and Methods ) , even in recombination assays using the same published genetic intervals ( Figs 1 and 3 ) . We do not have an explanation for this discrepancy apart from possible differences in genetic backgrounds that , although well known in budding yeast to influence both mitotic and meiotic phenotypes [88–91] are not well documented in fission yeast laboratory strains [92] . The role of CDK activity in DSB formation is supported by our results using controlled chemical inhibition of the cdc2-asM17 allele [74] . DSB formation is significantly impaired , both at hotspots in the NotI J fragment and genome-wide , when CDK activity is inhibited prior to meiotic induction or after DNA replication when cells are entering into prophase ( Figs 4 and 5 ) . The fact that DSB formation is not restored when the replication checkpoint is abrogated using a rad3 deletion mutant indicates that this inhibition of DSB formation is not an indirect consequence of checkpoint activation ( S5 Fig ) . Finally , the fact that in crs1 and cig1 crs1 mutants DSBs are reduced meanwhile meiotic progression is normal ( S4 Fig ) , suggests that CDK activity drives DSB formation directly and not indirectly by promoting meiotic progression . The observation of DSB formation when the ATP-analog is partially inactivated in the experiments with the cdc2-asM17 allele strengthens this view; meiotic progression is completely blocked in this situation ( no chromosome segregations ) , however break formation is reactivated ( Fig 5 and S5 Fig ) . The stronger reduction in DSBs in these experiments compared to the levels detected in single and double cyclin deletion mutants indicates that other CDK complexes may contribute to break formation . Redundancy of cyclins has been extensively reported in fission yeast where only Cdc13 is essential [8 , 9] . Therefore , even cyclins normally not abundant during meiotic prophase such as Puc1 and Rem1 could contribute to DSB formation in the absence of Cig1 , Cig2 , and Crs1 . Additionally , Cdc13 is by large the cyclin that contributes the most to the total cellular CDK activity , and the increasing levels of Cdc13 during prophase may also regulate DSB formation . A role of Puc1 in recombination has been suggested since , in contrast to the double cig1 cig2 deletion mutant , a triple cig1 cig2 puc1 mutant reduces crossover levels [9] . In addition , overexpression of a Cdc13-Cdc2 fusion protein , as unique source of CDK activity in the cell , partially sustains recombination; interestingly , the same fusion protein Cdc13-Cdc2 efficiently restores meiotic progression [9] . We have evaluated possible compensatory/redundant effects by increasing copy number of cdc13 and puc1 cyclins in crs1 mutants . Increasing genomic copies of cdc13 or puc1 cyclins ( even two-copy insertion in the case of cdc13 ) does not restore the recombination defect of crs1 deletion mutants . Moreover , neither increasing genomic copies of cdc2 ( which presumably would increase levels of the different CDK complexes ) restores it ( Fig 10 ) . These data suggest Crs1 specificity to promote meiotic recombination . Cdc2 implication in DSB formation was previously studied in synchronous haploid pat1-114 meiosis using a temperature-sensitive cdc2 allele , and DSB formation was detected but not quantified [50] . It is possible that Cdc2 was not completely inhibited in those experiments , and/or the ploidy of the cell could make a difference . Ploidy of the cell has an impact on the requirement of the replication checkpoint for the meiotic arrest upon hydroxyurea ( HU ) treatment , and diploid cells are more dependent than haploid cells on the Rad3/Cds1 pathway to block meiotic progression [93–95] . Similarly , DSB formation could be more sensitive to the inhibition of CDK activity in diploid than haploid meiosis . However , in both cases the inhibition of DSB formation upon HU treatment strongly depends on the Rad3/Cds1 pathway [53 , 56] , indicating that , if the different Cdc2-requirement for DSB formation were due to ploidy of the cell , the Rad3/Cds1 pathway would not account for this difference . Interestingly , untreated cdc2-asM17 mutants progress through meiotic S-phase faster than wild-type cells , although meiosis I entry is not advanced , indicating a prophase extension ( S7 Fig ) . Nevertheless , despite the fact that the cells finish replication earlier , DSB formation is not advanced ( or extended ) compared to the timing of DSB appearance and disappearance in a wild-type strain ( compare control in Fig 2 and Fig 3 to control in Fig 4 and Fig 5 ) . This observation suggests that although S-phase progression and DSB formation are coordinated ( see Introduction ) , S-phase completion is not sufficient for the activation of break formation . We have found that LinE formation is impaired when CDK activity is modulated by depleting Cig1 and Crs1 cyclins ( Fig 6 and S6 Fig ) , or the global CDK activity is reduced by using the cdc2-asM17 allele and an inhibitor ( Fig 9B ) . In both cases , a defect in LinE maturation is observed with Rec25-GFP . Although the kinetics of signal accumulation is normal , mature signal is reduced at least 2 . 5–3 fold compared to the controls ( Figs 6 and 9B ) . Since nuclear foci formation of any of the LinE components depends on each of the others [15 , 16] , we infer that the abnormal maturation of the Rec25-GFP signal reflects a defect in LinE maturation . Moreover , this defect is even observed in the cdc2-asM17 allele without treatment ( Fig 8B ) . This cytological defect may be related to the reduced chromatin-binding capability of Rec25-GFP , since in cig1 crs1 deletion mutants , cdc2-asM17 without treatment , and cdc2-asM17 after CDK inhibition , the protein is more easily recovered from the nuclear fraction by detergent treatment ( Fig 7 ) . A more labile binding of Rec25-GFP ( and the LinE complex ) to the meiotic chromosomes could impair proper LinE organization and subsequent DSB formation , since these complexes are essential for DSB formation in most hotspots [15] . CDK control of synaptonemal complex ( SC ) formation has been reported in budding yeast where the spreading of the central-element protein Zip1 depends on Cdc28 activity; however , no molecular explanation for this phenotype was described [96] . We have generated phospho-null mutants in putative CDK phosphorylation sites of Rec10 and Rec27 ( the only LinE-components harboring CDK sites ) , Rec7 ( SFT-complex ) , and Rec14 ( DSBC-complex ) proteins ( S8 Fig ) . None of the mutants tested ( alone or in different combinations ) reduces levels of gene conversion ( in quantitative or qualitative assays ) , except a moderate 20% reduction in rec14 ( cdk1 ) mutants ( S9 Fig and S10 Fig ) . Although we have not exploited all the possible mutants and combinations , so far our mutational analysis suggests that there is not a clear main CDK substrate to control DSB formation in fission yeast . It is possible that the effect we describe in LinE and DSB formation is indirectly mediated by mis-regulation of cohesins . Meiotic cohesins are required for LinE formation [14–16 , 19–21] , and meiotic cohesin subunits Rec8 and Rec11 harbor putative CDK phosphorylation sites , some of them phosphorylated in vivo [22 , 97 , 98] . Finally , at least Cdc2-Cig2 activity controls Mei4-promotor occupancy to promote timely activation of the middle wave of meiotic transcriptional-induction [99] , including mde2 expression . Mde2 has been proposed to stabilize SFT and DSBC-complex interaction [14] . Therefore , appropriate timing in the formation/loading of pre-recombination complexes may be an additional mechanism to optimize DSB formation . It is possible that in fission yeast several CDK-regulated targets ( in different processes ) equally contribute to DSB formation and recombination , and we may need cumulative defects in this regulation in order to observe a defect in recombination . Our results show that CDK activity regulates the initiation of meiotic recombination , namely DSB formation , in fission yeast . Maturation of LinEs ( essential for DSB formation at hotspots ) seems to be an important point of CDK regulation , modulating the binding to chromatin of one of their structural components , Rec25 . Though CDK activity has been implicated in the biology of the meiotic chromosomes and recombination in different organisms [100–106] , a role in DSB formation was clearly established previously only in Saccharomyces cerevisiae . Given the evolutionary distance between budding and fission yeasts , our work strengths this view and points to CDK regulation of DSB formation as a conserved feature of meiotic recombination . In addition , the comparison between DSB levels and recombination outcomes ( NCOs and COs ) suggests additional points of CDK regulation downstream of break formation , balancing NCOs/COs and intersister/interhomolog repair . Finally , among the cyclins analyzed , meiosis-specific Crs1 shows the major contribution to DSB formation , and complementation analysis suggests specificity for this cyclin to promote recombination . Experimentally required strains were obtained by meiotic crosses . Strains used in this study are listed in S1 Table . Oligos used are listed in S2 Table . Cells were grown in yeast extract medium with supplements ( YES ) or Edinburgh minimal medium ( MM ) with supplements at 32°C or 25°C ( for temperature-sensitive mutants ) . Normal supplements were Adenine , Leucine , Uracil and Histidine ( 225 mg/l ) . YES supplemented with 0 . 1 mg/ml G-418 ( Formedium ) or Hygromycin B ( Formedium ) was used to select and follow deletion mutants and GFP-tagged gene versions . Genetics crosses were done in malt extract plates with supplements ( MEA-4S ) at 25°C . Diploid pat1-114 leu1-32 strains were obtained by protoplast fusion and selection for complementation of ade6-M210 and ade6-M216 alleles [107] . Synchronous meiosis by thermal inactivation at 34°C of the pat1-114 temperature-sensitive allele and cell collection for flow cytometry analysis were done as previously described [16] . When experimentally required 20 μM 1-NM-PP1 ( Toronto Research Chemicals Inc . ) or equal DMSO volume was added to the cultures . A Becton Dickinson FACSCalibur and CellQuest software were used for cell acquisition and data analysis . Cellular fractionations were done as previously described [108] , and equal extract equivalent of each fraction ( 7% of Whole Cell Extract ) analyzed by Western blot with primary anti-GFP ( monoclonal JL-8 , Living colors , Clontech ) , anti-α-Tubulin ( monoclonal Clone B-5-1-2 , Sigma ) and anti-Histone H4 ( rabbit polyclonal ab10158 , Abcam ) antibodies; and secondary anti-mouse light chain-specific ( 115-035-174 Jackson ImmunoResearch ) and anti-rabbit ( A6154 Sigma ) horseradish peroxidase-conjugated antibodies . Rec25-GFP signal was developed with SuperSignal West DURA extended Kit ( Pierce ) , and Tubulin and Histone H4 signals with ECL Western Blotting Kit ( Amersham , GE Healthcare ) . Rec25-GFP quantification was done with Image J 1 . 49b software ( NIH ) and under-saturated scan exposures ( ChemiDoc XRS Imaging System , Bio-Rad ) . Microscopy used to detect Rec25-GFP signal in Methanol/Acetone fixed cells was previously described [15] . Images of whole cells are maximal projections of 11 sections at 0 . 4 μm steps to cover the whole cell ( 4 μm total ) . DNA images are a single focal plane , because out-of-focus DAPI fluorescence obscures the projection . Cytological classification of the different Rec25-GFP categories was done using the raw images and navigating the Z sections . Images were captured with a Nikon Eclipse 90i microscope equipped with a 100x/NA1 . 45/WD0 . 13 Oil Plan APO Lambda lens , a Hamamatsu ORCA-ER camera , and MetaMorph software ( Molecular Devices ) . New cig1 and crs1 deletions were generated . Available cig1 deletion [109] removes 363 bp of adjacent ORF ( rec11 ) , coding for a meiosis-specific cohesin subunit already known to be important for meiotic recombination [15 , 17] . In the case of crs1 , a change in ORF annotation has extended previous ORF designation and the original deletion maintains 20% of the ORF [65] . Complete cig1 and crs1 ORF deletions were generated by PCR-based method [110] using oligos to amplify hphMX6 from plasmid pFA6a-hphMX6 and transformation to Hygromycin B resistance of strain CMC6 ( h90 ura4-D18 ) in the case of cig1 , and CMC66 ( h- pat1-114 ade6-M210 leu1-32 rec25-GFP::KanMX6 ) in the case of crs1 . These oligos were pair cig1-D1/cig1-D2 and pair crs1-D1/crs1-D2 . pJK148 cdc13 ( 4 . 2 kb genomic clone containing 1869 bp upstream and 865 bp downstream of the ORF ) and pIRT22 cdc2 ( 3 . 4 kb genomic clone containing 840 bp upstream and 1372 bp downstream of the ORF ) plasmids were a gift from Sergio Moreno´s laboratory . cdc2 was subcloned in the pJK148 plasmid at the PstI restriction site . puc1 ( 3 . 9 kb fragment containing 2360 bp upstream and 485 bp downstream of the ORF ) and crs1 ( 2 . 5 kb fragment containing 1083 bp upstream and 513 bp downstream of the ORF ) genomic clones were PCR amplified using oligo pairs puc1-SacI/puc1-KpnI and crs1-SalI ( 3 ) /crs1-EcoRI ( 3 ) . All pJK148 plasmids were sequenced . For integration plasmids were digested with NdeI ( pJK148 cdc2 ) , NruI ( pJK148 empty and pJK148 cdc13 ) or Tth111I ( pJK148 crs1 and pJK148 puc1 ) , and the strain CMC1056 ( h- crs1::hphMX6 leu1-32 ) transformed to Leu+ . Single copy and multicopy integrants were selected by PCR using oligos pJK148-1/pJK148-2; and specific plasmid integrations tested using oligo pairs pJK148-upKpnI/pJK148-downBamHI ( pJK148 empty ) , pJK148-upKpnI/cdc13-1 ( pJK148 cdc13 ) , pJK148-upKpnI/cdc2-10 ( pJK148 cdc2 ) , pJK148-upKpnI/puc1-2 ( pJK148 puc1 ) , and pJK148-upKpnI/crs1-3 ( pJK148 crs1 ) . Number of leu1 copies in multicopy integrants was checked by qPCR using the oligo pairs leu1-3/leu1-4 and mde2-3/mde2-4 ( internal control ) ; the parental CMC1056 strain and a single copy integrant were used as controls . pJK148 puc1 and pJK148 cdc2 multicopy integrants were found to be reorganized . rec7 ( cdk1 , cdk2 , cdk1 cdk2 ) , rec14 ( cdk1 , cdk2 , cdk1 cdk2 ) , and rec27 ( cdk ) mutants in putative CDK phosphorylation sites were generated by PCR using plasmids pFA6a-rec7-GFP-hphMX6 ( CMC28 , containing 191 bp upstream ORF + ORF ) , pFA6a-rec14-GFP-kanMX6 ( CMC43 , containing 152 bp upstream ORF + ORF ) , and pFA6a-rec27-GFP-hphMX6 ( CMC34 , containing 184 bp upstream ORF + ORF ) as templates , and the following oligo pairs: rec7-cdk1F/rec7-cdk1R , rec7-cdk2F/rec7-cdk2R , rec14-cdk1F/rec14-cdk1R , rec14-cdk2F/rec14-cdk2R , and rec27-cdkF/rec27-cdkR . PCR products were digested with DpnI and transformed into Escherichia coli , plasmids recovered and sequenced . Double cdk1 cdk2 mutants were similarly generated using the plasmids containing single mutant genes as templates . Cassettes for S . pombe transformation were obtained by PCR using as templates the plasmids containing the different mutants and the following oligo pairs: rec7-3/rec7-STOP ( which amplify unmarked and untagged versions ) , rec14-1/rec14-D2 ( which amplify G-418 resistant GFP-versions ) or rec14-1/rec14-STOP ( which amplify unmarked and untagged versions ) , and rec27-1/rec27-D2 ( which amplify Hygromycin B resistant GFP-versions ) or rec27-1/rec27-STOP ( which amplify unmarked and untagged versions ) . rec7 cassettes were used for transformation of strain CMC945 ( h- rec7::ura4+ ura4-D18 ) to FOA resistance . rec14 cassettes were used for transformation of strain CMC595 ( h90 rec14::hphMX6 ) to G-418 resistance or strain CMC1201 ( h- rec14::ura4+ ura4-D18 ) to FOA resistance . rec27 cassettes were used for transformation of strain CMC952 ( h90 rec27::kanMX6 ) to Hygromycin B resistance or strain CMC966 ( h- rec27::ura4+ ura4-D18 ) to FOA resistance . rec10 ( cdk total ) and rec14 ( cdk total ) mutants were synthetic fragments ( Integrated DNA Technologies ) that were PCR amplified and transformed into CMC1201 ( h- rec14::ura4+ ura4-D18 ) and CMC1218 ( h- rec10::ura4+ ura4-D18 ) strains . rec10 ( cdk total ) harbors an extra mutation ( C597 to A ) changing F199 ( TTC codon ) to L ( TTA codon ) . Deletion strains rec7::ura4+ , rec10::ura4+ , rec14::ura4+ , rec14::hphMX6 , and rec27::ura4+ used for knock-ins were done by PCR-based method [110] using oligos to amplify ura4 gene or hphMX6 from plasmids pFA6a-ura4 and pFA6a-hphMX6 respectively , and transformation to Ura+ prototrophy or Hygromycin B resistance of strains CMC4 ( h- ura4-D18 ) and CMC3 ( h90 968 ) . Oligo pairs used for these deletions were: rec7-D1/rec7-D2 , rec10-D1/rec10-D2 , rec14-D1/rec14-D2 , and rec27-D1/rec27-D2 . Correct deletions and knock-ins were checked by PCR and sequencing . Crosses were done in MEA-4S or SPA at 25°C . After 3–4 days cell masses were treated overnight at 25°C with glucuronidase ( Roche ) , and subsequently incubated 25 minutes at 55°C to kill any remaining vegetative cells . For intragenic recombination assays the ade6-M26 allele was always in the h- parent , and the ade6-3049 or ade6-M210 allele in the h+ parent . For intergenic recombination assays the leu1-32 marker was always in the h- parent , and the his5-303 marker in the h+ parent . For intragenic recombination assays appropriate numbers of viable spores were plated on 10 YE-minus supplement plates ( approx . 600/plate ) or 10 YE+Guanine plates ( 104/plate; Guanine inhibits Adenine uptake and kills Ade- cells [111 , 112] ) , and incubated for 4–5 days at 32°C . Frequency of intragenic recombination was calculated as the number of white colonies ( Ade+ in YE ) per 104 viable spores , pooling the numbers of the 10 plates . Each experiment was plated twice and the final recombination frequency was calculated based on cumulative numbers of the two platings . For frequency of intergenic recombination 300–500 viable spores were plated on 5 YES plates , and after 3 days at 32°C replicated to YES-Phloxin B ( to identify diploid colonies and discard for further analysis ) and MM ( to score for Leu+ His+ colonies ) . Frequency of intergenic recombination was calculated as the number of haploid prototrophic colonies ( Leu+ His+ ) per 100 haploid spore colonies . In the case of mat1-P and leu1-32 markers , spore colonies grown at 32°C in 5 YES plates were replicated to YES-Phloxin B ( to discard diploids ) and MM ( to score for Leu+ ) , and the number of haploid prototrophic Leu+ and total colonies scored . 240 Leu+ colonies were randomly selected and grown as patched in YES plates . These master plates were further replicated to 2 new YES plates . Next day patches in one of the plates were individually mixed with a dense h+ cell suspension , and further incubated at 32°C for one day . Then , both plates were replicated to MEA-4S , and after 6–7 days at 25°C exposed to iodine vapors to determine the presence of spores ( indicative of mating ) . The plate with patches previously not mixed with h+ cells was used to discard h90 ( h+ revertants ) colonies from the scoring . Frequency of intergenic recombination was calculated as the number of haploid h- prototrophic colonies ( iodine-positive Leu+ ) per 100 haploid spore colonies , scaling the number of h- Leu+ colonies in the sample ( 240 selected Leu+ colonies ) to the number of total Leu+ colonies , and to the number of total colonies scored . As for intragenic recombination , each experiment for intergenic recombination was plated twice and the final recombination frequency calculated based on cumulative numbers of the two platings . Recombination assays were repeated 3–10 times and p values were calculated based on Student´s t-test ( unpaired , two tails ) . For detection of DSBs Pulse Field Gel Electrophoresis ( PFGE ) of agarose embedded samples ( plugs ) was used . 30 ml cell samples ( O . D . 0 . 8–1 ) at different times during meiotic time courses were processed as described in [113] with some modifications . Collected cells were washed with 30 ml of cold 50 mM EDTA pH 8 . 0 and resuspended in 300 μl of cold CEPES ( 50 mM EDTA pH 8 . 0 , 40 mM Na2HPO4 , 20 mM citric acid , 1 . 2 M sorbitol , 10 mM sodium azide , 1mg/ml Zymolyase 20T from Arthrobactor luteus ( Seikagaku Biobusiness Corporation ) and 5 mg/ml lysing enzymes from Trichoderma harzianum ( Sigma ) ) . Samples were kept on ice until collection of the last time-point samples , and then all were processed in parallel . Cells were incubated at 37°C for 1 . 5 hr in a thermoblock with gentle agitation . After checking for proper cell wall digestion , all samples were put on ice and agarose plugs then prepared . Samples were warmed at 50°C for 1 min in a thermoblock , mixed with 400 μl of low melting-point agarose ( 1% agarose in 50 mM EDTA pH 8 . 0 , 10 mM Tris-HCl pH 7 . 5 , 1 . 2 M sorbitol ) at 50°C , and divided into the plug molds . Molds were cooled at 4°C for 15 min to solidify , ejected into 2 ml Eppendorf tubes containing 1 . 2 ml of 0 . 25 M EDTA pH 8 . 0 , 50 mM Tris-HCl pH 7 . 5 , 1% SDS , and incubated at 50°C for 90 min . Afterwards , solution was replaced by 1 . 2 ml of Lysis Buffer ( 0 . 5 M EDTA pH 8 . 0 , 10 mM Tris-HCl pH 7 . 5 , 10 mM sodium azide , 1% N-Lauroylsarcorine sodium ) with 1mg/ml Proteinase K ( Roche ) and plugs incubated overnight at 50°C . Next day , Lysis Buffer was replaced with fresh Lysis Buffer with Proteinase K , and plugs incubated at 50°C until next day . Finally , Proteinase K was inactivated washing the plugs in 1 . 2 ml of TE ( 10 mM Tris-HCl pH 8 , 1 mM EDTA ) with 1 mM PMSF for 2 hr at room temperature , and three times in TE for 30 min-1 hr at room temperature with gentle agitation prior to final store at 4°C . For detection of DSBs in intact chromosomes plugs were washed with 500 μl of TAE ( 40 mM Tris-Acetate , 1 mM EDTA pH 8 ) for 1 hr with gentle agitation at room temperature prior to loading in a 0 . 7% agarose ( Pulsed Field Certified Megabase Agarose , Bio-Rad ) TAE gel . Gels were run in a CHEF-DR II system ( Bio-Rad ) for 70h at 2V/cm , 30 min of both initial and final switch time , 120° angle , and 14°C . Finally , gels were stained overnight at room temperature in TAE with 0 . 5 μg/ml of Ethidium Bromide . For detection of DSBs at hotspot mbs1 by Southern blot , plugs were washed twice for 30 min at 4°C in 250 μl of enzyme buffer , buffer replaced by 250 μl of fresh buffer containing 35U of NotI , and incubated at 4°C during 6–7 hr before final incubation at 37°C overnight . Next day , plugs were washed in 1 ml of 0 . 5X TBE ( 90mM Tris , 90 mM Boric Acid , 2 mM EDTA , pH 8 . 3 ) for 1 hr with gentle agitation at room temperature prior to loading in a 1 . 1% agarose ( Pulsed Field Certified Megabase Agarose ) 0 . 5X TBE gel . Gels were run for 24h at 6V/cm , 7 . 9 seconds of initial switch and 54 . 2 seconds of final switch , 120° angle , and 14°C . After electrophoresis , gels were stained with 0 . 5 μg/ml of Ethidium Bromide for 30 min to check proper digestion prior to Southern blotting . Vacuum ( Vacugen , Amersham ) or capillarity alkaline transfer to Nylon membranes ( Amersham Hybond-XL , GE Healthcare ) was performed before standard Southern blot [114] using a 32P radiolabelled probe recognizing the left end of the 501 Kb NotI fragment J [72] . DSB quantification was done with Quantity One software ( Bio-Rad ) and under-saturated phosphorimager exposures ( PMI Personal Molecular Images , Bio-Rad; Fuji imaging BAS-III screens ) . Counts ( CNT x mm ) in the whole lane for each sample and at mbs1 position were obtained , after background lane elimination using the Rolling Disk ( 10 ) function . Whole lane was considered from top gel excluding the wells to the bottom just below the mbs2 site . Correction factor for the difference in total DNA between samples before and after DNA replication was calculated by dividing the total signal of the sample with maximal levels of DSBs ( after DNA replication ) by the total signal of the nitrogen-depleted sample ( before DNA replication ) . The specific signal at the mbs1 band in the nitrogen-depleted sample was multiplied by this factor , and correspondingly subtracted from the mbs1 signal of the rest of the time points of the experiment . Corrected mbs1 signals were then divided by the total lane signal to obtain the percentage of breakage at each time point . Assays were repeated 5–8 times and p values were calculated based on Student´s t-test ( unpaired , two tails ) .	Meiotic division is a cell division process where a single round of DNA replication is followed by two sequential chromosome segregations , the first reductional ( homologous chromosomes separate ) and the second equational ( sister chromatids segregate ) . As a consequence diploid organisms halve ploidy , producing haploid gametes that after fertilization generate a new diploid organism with a complete chromosome complement . At early stages of meiosis physical exchange between homologous chromosomes ensures the accurate following reductional segregation . Physical exchange is provided by recombination that initiates with highly-controlled self-inflicted DNA damage ( DSBs , double strand breaks ) . We have found that the conserved CDK ( cyclin-dependent kinase ) activity controls DSB formation in fission yeast . Available data were uncertain about the conservation of CDK in the process , and thus our work points to a broad evolutionary conservation of this regulation . Regulation is exerted at least by controlling chromatin-binding of one structural component of linear elements , a protein complex related to the synaptonemal complex and required for high levels of DSBs . Correspondingly , depletion of CDK activity impairs formation of these structures . In addition , CDK might control homeostatic mechanisms , critical to maintain efficient levels of recombination across the genome and , therefore , high rates of genetic exchange between parental chromosomes .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "meiosis", "cell", "cycle", "and", "cell", "division", "cell", "processes", "fungi", "model", "organisms", "dna", "replication", "experimental", "organism", "systems", "molecular", "biology", "techniques", "mutagenesis", "and", "gene", "deletion", "techniques", "dna", "synthesis", "phase", "schizosaccharomyces", "homologous", "recombination", "research", "and", "analysis", "methods", "prophase", "chromosome", "biology", "animal", "studies", "schizosaccharomyces", "pombe", "molecular", "biology", "yeast", "biochemistry", "eukaryota", "cell", "biology", "nucleic", "acids", "genetics", "biology", "and", "life", "sciences", "yeast", "and", "fungal", "models", "dna", "recombination", "organisms", "deletion", "mutagenesis", "cyclins" ]	2019	CDK contribution to DSB formation and recombination in fission yeast meiosis
In humans , meiotic chromosome segregation errors increase dramatically as women age , but the molecular defects responsible are largely unknown . Cohesion along the arms of meiotic sister chromatids provides an evolutionarily conserved mechanism to keep recombinant chromosomes associated until anaphase I . One attractive hypothesis to explain age-dependent nondisjunction ( NDJ ) is that loss of cohesion over time causes recombinant homologues to dissociate prematurely and segregate randomly during the first meiotic division . Using Drosophila as a model system , we have tested this hypothesis and observe a significant increase in meiosis I NDJ in experimentally aged Drosophila oocytes when the cohesin protein SMC1 is reduced . Our finding that missegregation of recombinant homologues increases with age supports the model that chiasmata are destabilized by gradual loss of cohesion over time . Moreover , the stage at which Drosophila oocytes are most vulnerable to age-related defects is analogous to that at which human oocytes remain arrested for decades . Our data provide the first demonstration in any organism that , when meiotic cohesion begins intact , the aging process can weaken it sufficiently and cause missegregation of recombinant chromosomes . One major advantage of these studies is that we have reduced but not eliminated the SMC1 subunit . Therefore , we have been able to investigate how aging affects normal meiotic cohesion . Our findings that recombinant chromosomes are at highest risk for loss of chiasmata during diplotene argue that human oocytes are most vulnerable to age-induced loss of meiotic cohesion at the stage at which they remain arrested for several years . In humans , meiotic chromosome segregation errors that give rise to aneuploid gametes are the leading cause of fetal loss and birth defects [1] . Approximately 30% of miscarriages result from aneuploidy and at least 5% of all clinically recognized pregnancies and 0 . 3% of live-borns are aneuploid [1] . Female meiosis in humans is especially error-prone and the majority of segregation errors in oocytes originate during meiosis I [1] , [2] . The link between increased maternal age and meiotic segregation defects in humans is well established . At maternal age 25 , the risk of a trisomic pregnancy is approximately 2% but increases to approximately 35% for a woman at age 42 [3] . Despite its clinical importance , the specific mechanisms that give rise to age-dependent meiotic nondisjunction ( NDJ ) are not understood . The prevailing theory is that segregation errors in older oocytes arise in large part because of the protracted prophase I arrest at which human oocytes remain suspended for decades [1] , [4] . Accurate segregation during meiosis I requires that homologous chromosomes undergo recombination and remain physically attached to one another until they segregate to opposite poles during anaphase I . In the absence of a stable connection , homologues will segregate randomly resulting in meiosis I NDJ . Cohesion between the arms of sister chromatids provides an evolutionarily conserved mechanism for maintaining the association of recombinant homologues ( Figure 1 ) . Normally , the release of arm cohesion at anaphase I allows recombinant homologues to segregrate to opposite poles [5] , [6] . In the absence of cohesion , chiasmata are not maintained and homologous chromosomes missegregate during meiosis I [5] , [7] , [8] . Given that human oocytes undergo meiotic recombination during fetal development and remain suspended in a prolonged dictyate ( diplotene ) arrest until ovulation , the continuous association of homologous chromosomes demands that meiotic sister-chromatid cohesion remain intact for decades . One attractive hypothesis to explain age-dependent NDJ is that deterioration of cohesion with age causes recombinant homologues to dissociate prematurely and segregate randomly during the first meiotic division ( Figure 1 ) . However , testing this hypothesis in humans presents several insurmountable challenges . Using Drosophila as a model system , we have developed an experimental regimen to study the mechanisms that contribute to increased levels of meiotic NDJ in oocytes as a result of aging [9] . The Drosophila ovary is composed of several ovarioles , each of which contains a linear array of oocytes at progressive stages of development ( Figure 2A ) . Throughout the lifetime of the female , germline stem cells at the tip of each ovariole continuously generate a steady stream of newly formed oocytes that enter meiotic prophase and grow and develop as they move posteriorly; as mature oocytes pass through the oviduct they complete meiosis and are fertilized . Under normal conditions ( continuous egg laying ) , Drosophila oogenesis is an uninterrupted process with only a brief arrest at metaphase I before ovulation . However , when egg laying is suppressed , the majority of Drosophila oocytes within each ovariole are halted in developmental progression and “age” within the abdomen of the female ( Figure 2B ) . Such experimentally induced aging of Drosophila oocytes can be used to mimic the normal aging process that human oocytes undergo within the ovary during a female's lifespan [9] . We have used this aging regimen to test the hypothesis that meiotic cohesion deteriorates as the oocyte ages and increases the frequency at which recombinant homologues missegregate . Although maternal age does not dictate the age of the oocyte in fruit flies as it does in humans , for simplicity we will refer to increased NDJ in experimentally aged Drosophila oocytes as “age-dependent NDJ . ” In order to assay loss of chiasma maintenance with age , we needed to verify that chiasma formation was not severely disrupted in smc1+/− oocytes . Cohesion between sister chromatids is required for normal levels of crossovers during meiosis [13]–[17] . Therefore , we measured the frequency and distribution of exchange in smc1+/− heterozygous females ( Figure 3B ) . Although reduction of SMC1 protein has a semi-dominant effect on the number and distribution of meiotic crossovers during female meiosis , high levels of homologous exchange were still observed ( 76% of the wild-type control , Figure 3B ) . The tetrad exchange rank for Drosophila oocyte bivalents can be estimated from analysis of recombinant and non-recombinant meiotic products [18] . Normally , the X chromosome is achiasmate ( non-recombinant ) in 6–12% of Drosophila oocytes [19] . Our recombination data indicates that although crossovers were reduced in smc1+/− oocytes , the majority of tetrads ( 74% ) had at least one exchange event ( Figure 3B ) . In addition , viability and fertility of smc1+/− females is normal and they do not exhibit meiotic segregation defects in the absence of an aging regimen ( Table S1 ) . Therefore , smc1+/− oocytes provide an excellent sensitized system that can be used to assay whether cohesion and chiasma maintenance decline with age . To examine the effect of SMC1 dosage on age-dependent NDJ , we subjected smc1+/− females to the aging regimen and assayed for chromosome missegregation ( see Figure 2B , 2C ) . However , meiotic NDJ did not increase when smc1+/− oocytes were aged ( Table S2 ) . Because Drosophila females harbor a robust mechanism that directs the segregation of achiasmate chromosomes [20] , it is possible that this achiasmate mechanism also ensures accurate disjunction of recombinant chromosomes that fail to maintain chiasmata ( Figure 4 ) . Therefore , we reasoned that in order to detect missegregation of recombinant chromosomes in Drosophila oocytes that have lost chiasmata , the achiasmate system must be compromised ( Figure 4 ) . Hawley and colleagues have shown that P-element disruption of one copy of matrimony ( mtrm[KG08051]/+ ) disrupts the segregation of achiasmate chromosomes in Drosophila oocytes [21] . However , meiotic cohesion , synaptonemal complex assembly and crossover frequency appear to be unaffected in mtrm+/− oocytes [21] , [22] . Therefore , to monitor whether exchange bivalents lose chiasmata with age , we compromised the achiasmate pathway by reducing the dosage of mtrm in smc1+/− females . We observed increased segregation errors in smc1+/− oocytes following the four-day aging regimen when the achiasmate pathway was also compromised ( mtrm+/− ) ( Figure 5A ) . Significant levels of age-dependent nondisjunction were observed for broods 1 and 2 , which represent oocytes that are developmentally most mature during the aging regimen ( see Figure 2A ) . We have performed multiple experiments to compare NDJ in smc1+/− mtrm+/− aged and non-aged oocytes and have also assayed missegregation of different X chromosomes ( data not shown ) . Although the absolute level of NDJ may vary between experiments , we have repeatedly observed higher levels of NDJ in smc1+/− mtrm+/− aged oocytes in the first two 24 hour broods ( see Tables S3A and S3B for examples of raw data from two independent experiments ) . In contrast , mtrm+/− oocytes with normal levels of SMC1 do not exhibit age-dependent NDJ in the first brood ( Table S4 ) ; we did detect a significant increase in brood 2 , but the absolute level of NDJ was relatively low ( Table S4 ) . The finding that increased meiotic chromosome missegregation in aged Drosophila oocytes occurs when the dosage of a cohesin subunit is reduced is consistent with deterioration of cohesion and loss of chiasma maintenance as an underlying cause for age-dependent NDJ . To ask whether deterioration of chiasmata contributed to the increase in age-dependent nondisjuntion in smc1+/− mtrm+/− oocytes , we assayed the recombinational history of the missegregating chromosomes . For this analysis , we collected “Diplo-X” female progeny that received two maternally contributed X chromosomes due to meiotic missegregation ( see Figure 2C ) and performed an additional cross to genotype the two X chromosomes ( Figure 6 ) . Using this strategy , we determined whether missegregating chromosomes had undergone one or more crossovers ( Table S5 ) . In addition , a centromere-proximal marker ( car ) allowed us to assess whether NDJ events involved homologous chromosomes ( MI NDJ ) or sister chromatids ( MII NDJ ) . The genotype of the Diplo-X females suggested that the majority of NDJ events ( 70/72 ) occurred during meiosis I ( Table S5 ) . Approximately one-third of all Diplo-X females arose from nondisjunction of exchange tetrads ( R ) in both aged and non-aged oocytes ( Figure 5B ) . Notably , recombinant bivalents missegregated 1 . 8 times more frequently in the aged oocytes than in non-aged oocytes ( compare R/N for aged and non-aged , Figure 5B ) . In addition , the number of recombinant bivalents that missegregate is under-represented in our assay because it is possible for a Diplo-X female to inherit two non-recombinant chromatids from a recombinant tetrad ( see Figure 6 ) . These data argue that upon reduction of SMC1 , recombinant bivalents become more vulnerable to missegregation when oocytes age . Although the achiasmate pathway is compromised in mtrm+/− mutants , our data indicate that it is not completely disengaged . An estimated 26% of the tetrads are achiasmate in smc1+/− oocytes ( Figure 3B ) but we only observe 4–9% NDJ in non-aged oocytes when the achiasmate pathway is compromised ( smc1+/− mtrm+/− , Table S3 ) . Moreover , segregation of an X chromosome ( In ( 1 ) dl-49 ) harboring an inversion that severely suppresses crossovers is not completely random in mtrm+/− oocytes ( Table S1 ) . The data in Figure 5B indicate that when the level of SMC1 protein is reduced , aging further increases the frequency at which achiasmate chromosomes missegregate in mtrm+/− oocytes ( compare NR/N for aged and non-aged ) . These data implicate the cohesin complex in accurate disjunction of achiasmate as well as chiasmate bivalents . However , the low number of MII NDJ events that we observe ( as determined by the centromere proximal car marker ) argues that the increased missegregation of non-recombinant chromosomes is not due to loss of centromeric cohesion between sister chromatids . Our analysis of age-dependent NDJ in smc1+/− mtrm+/− oocytes indicates that broods 1 and 2 are most susceptible to age effects . Each Drosophila ovariole consists of multiple oocytes at progressive stages of development . At the posterior end of the ovariole ( stages 13/14 ) , the mature oocytes undergo nuclear envelope breakdown and spindle assembly . Stage 14 oocytes remain arrested at metaphase I until passage through the oviduct triggers resumption of the meiotic divisions . After an aging regimen , the most mature oocytes are laid first and therefore will contribute to the progeny in brood 1 ( see Figure 2A ) . In mature oocytes , the chromosomes have already made stable connections with the meiotic spindle microtubules and premature loss of chiasmata during the aging regimen should not affect their segregation . However , if the mature oocytes develop gross defects in the spindle machinery as they age , recombinant chromosomes could missegregate by a mechanism unrelated to chiasma maintenance . Conversely , oocytes that undergo aging at a stage prior to meiotic spindle assembly would depend on chiasma maintenance for proper segregation . Therefore , it was important to determine what fraction of broods 1 and 2 correspond to metaphase I arrested oocytes . To compare the relative distribution of oocytes at different developmental stages in ovaries from smc1+/− mtrm+/− control females and those subjected to the aging regimen , we examined ovaries at 8-hour intervals after completion of the four-day regimen . One striking difference was the increased number of mature oocytes ( stages 13/14 ) following aging ( Figure 7 and Figure S1 ) . Previous reports by other investigators also have documented that mature oocytes accumulate when egg laying is suppressed [23] , [24] . Quantitative analysis indicated that our aging regimen resulted in a ∼2 . 5-fold increase in mature oocytes compared to the ovarioles of control females ( Figure 7 ) . However , after 16 hours of egg laying , this excess of mature oocytes was no longer observed . These results argue that in our NDJ tests , at least some fraction of progeny in brood 1 ( 0–24 hours ) arise from oocytes that have not yet assembled meiotic spindles . Evaluation of the distribution of different developmental stages in fixed ovaries also indicated that following the aging regimen , the increase in mature oocytes occurred primarily at the expense of stage 9–12 oocytes ( Figure 7 ) while oocytes at earlier stages ( stages 1–8 ) remained abundant . Significantly , we found that 16 hours after the aging regimen ceased and egg laying commenced , the distribution of late stages in the aged group resembled that in control ovaries , although the total number of oocytes between stages 8 and 14 was slightly less in the aged ovaries . Consistent with our findings , King [23] reported that stages 9–12 are significantly under-represented in ovaries of 4 day old virgins . Together , these data indicate that oocytes at stages1–8 halt in development and age during the four-day aging regimen , while oocytes at later stages ( stages 9–13 ) continue to mature and do not “age” until they arrest at metaphase I ( stage 14 ) . Given the altered distribution of oocytes at different developmental stages following the aging regimen , we carried out NDJ tests to determine which stages during oocyte development are the most susceptible to age-dependent segregation errors . We subjected smc1+/− mtrm+/− females to our standard four-day aging regimen and set up crosses to assay meiotic NDJ ( see Figure 2 ) . However , this time , we transferred the parents to new vials every 8 hours ( instead of every 24 hours ) to generate 8-hour “sub-broods” of progeny . This allowed us to measure age-related segregation errors in “sub-broods” of progeny that represented snapshots of oocytes at different meiotic stages ( pachytene , diplotene-like and metaphase I ) . Following the aging regimen , age-dependent NDJ was observed in sub-broods 3 and 4 , which arose from oocytes that were fertilized 16–32 hours after the aging regimen was completed ( Figure 8A , Table S6 ) . Our cytological analysis ( Figure 7 and Figure S1 ) and the published time-frame for oogenesis progression [19] , [25] , indicate that these sub-broods correspond to oocytes at stages 7 and 8 which have already disassembled their synaptonemal complex before initiation of the aging regimen [26] , [27] . In contrast , a significant increase in meiotic segregation errors was not observed during the first 16 hours in aged smc1+/− mtrm+/− oocytes ( Figure 8A , Table S6 ) . These results argue that metaphase I arrested smc1+/− mtrm+/− oocytes ( stage 14 ) are not vulnerable to age-related defects . Similarly , oocytes that undergo aging during pachytene ( sub-broods 5 and 6 ) also appear to be refractory to age effects . Therefore , only oocytes that age during a very specific window of oogenesis ( post-pachytene/pre-metaphase I ) exhibit age dependent NDJ . To determine what fraction of missegregating chromosomes were recombinant in the aged versus non-aged oocytes , we genotyped the Diplo-X progeny for a subset of sub-broods obtained from the NDJ analysis shown in Figure 8A . In addition to sub-broods 3 and 4 which exhibited age-dependent nondisjunction , we also analyzed the recombinational history of missegregating-X chromosomes in sub-broods 1 and 6 , which represent opposite ends of the developmental spectrum examined in this study ( Tables S7 , S8 ) . Again , we found that both recombinant and non-recombinant chromosomes missegregated in aged and non-aged oocytes ( Tables S7 , S8 ) . In addition , a striking trend emerged for the aged oocytes when we calculated the frequency at which recombinant ( R ) chromosomes missegregated in each sub-brood ( R/N; Figure 8B ) . NDJ of recombinant chromosomes in sub-brood 3 aged oocytes was significantly higher than that for aged oocytes in sub-brood 1 ( P = 0 . 0029 ) or sub-brood 6 ( P = 0 . 0179 ) . In addition , this trend holds true when the NDJ of recombinant chromosomes in sub-brood 4 aged oocytes is compared to that for sub-broods 1 or 6 aged oocytes , although the difference is not statistically significant . These data contrast strongly with those for non-aged oocytes , in which recombinant chromosomes missegregate at similar frequencies in the four sub-broods analyzed . The data shown in Figure 8 provide compelling evidence that only oocytes at a specific meiotic stage are vulnerable to age-dependent segregation errors even when all oocytes within the ovary are subjected to the aging regimen . The increased nondisjunction observed in specific sub-broods arises in part because segregation of recombinant bivalents becomes more error-prone ( Figure 8B ) . When one takes into account the well-documented time-line for oocyte development while evaluating the sub-brood NDJ results , a striking pattern emerges ( Figure 8C ) . When pachytene oocytes undergo the aging regimen , recombinant chromosomes do not exhibit an age-dependent increase in nondisjunction ( sub-broods 5 and 6 ) . In contrast , sub-broods 3 and 4 , which display the greatest age-related defects , are composed primarily of oocytes at stages 7 and 8 of oocyte development . Both electron microscopy and immunofluorescence analyses have verified that disassembly of the synaptonemal complex in Drosophila oocytes initiates prior to these stages [26] , [27] . Although Drosophila chromosomes do not exhibit typical diplotene/diakinesis morphology in prophase I oocytes , stages 7 and 8 of oogenesis correspond to these classical meiotic stages [19] . Transit through each of these stages takes approximately 5–8 hours during normal oogenesis , but when their developmental progression is halted by our aging regimen , meiotic chromosomes remain suspended in this state up to 19 times longer than usual . Of special note is the finding that recombinant chromosomes are most susceptible to segregation errors if Drosophila oocytes undergo aging during diplotene ( Figure 8 ) because this is the meiotic stage at which human oocytes remain arrested for decades . Our sub-brood analysis also suggests that non-recombinant chromosomes in smc1+/− mtrm+/− oocytes missegregate more frequently when aging is induced in oocytes prior to metaphase I ( Figure S2 ) , although the difference between sub-brood 1 and the other sub-broods is not statistically significant ( 0 . 10<P<0 . 27 ) . However , non-recombinant chromosomes in sub-brood 3 are significantly more vulnerable to missegregation after aging ( P = 0 . 029 ) . These results support the model that the connections that keep achiasmate bivalents together depend at least in part on meiotic cohesion proteins [9] . Together , our data support the hypothesis that weakening of meiotic cohesion with age is a significant determinant of age-dependent nondisjunction in both fly and human oocytes . One major advantage of these studies is that we have lowered but not eliminated the level of the cohesin subunit SMC1 . Although smc1+/− oocytes are sensitized to the effects of aging , meiotic cohesion is initially intact . Therefore our regimen allows us to observe the effect of aging upon the normal cohesin complex that functions during Drosophila meiosis . In this respect our analysis differs markedly from previous studies using knock-out mice in which the meiosis-specific subunit SMC1-β was eliminated [8] . Here we demonstrate that deterioration of normal meiotic cohesion during the aging process causes loss of chiasma maintenance . Loss of arm cohesion in aged oocytes accounts for the increased frequency with which recombinant chromosomes missegregate during meiosis I . In contrast , low levels of MII NDJ indicate that centromeric cohesion remains intact , at least for the duration of our aging regimen . Recombinant chromosomes in Drosophila oocytes that undergo aging after they have disassembled their synaptonemal complex ( diplotene-like ) are most vulnerable to age effects and these oocytes bear a striking resemblance to the stage at which human oocytes remain arrested for decades . Interestingly , Drosophila oocytes that are subjected to aging during pachytene do not exhibit significant age-dependent nondisjunction . Using epifluorescence microscopy to compare smc1+/− mtrm+/− aged and non-aged oocytes , we have not observed any obvious differences in SC morphology , or the timing of its assembly or disassembly ( VVS and SEB , unpublished results ) . Therefore , we speculate that the synaptonemal complex may play a role in protecting sister-chromatid cohesion from age-induced deterioration . The differential response of pachytene and diplotene-like Drosophila oocytes to the aging process has profound implications regarding the vulnerability of human oocytes to age-induced loss of meiotic cohesion . Our data argue that the meiotic stage ( dictyate ) at which human oocytes remain suspended as women age is also the stage during which recombinant chromosomes are at highest risk for loss of chiasmata . Because the achiasmate pathway in Drosophila oocytes will ensure the accurate segregation of recombinant chromosomes that lose chiasmata due to deterioration of arm cohesion during the aging process , we needed to dismantle this system for our analysis . Therefore , we used a P-element insertion in matrimony ( mtrmKG08051 ) as a genetic tool to disrupt the achiasmate pathway . However , we do not think that heterozygosity for mtrm contributes to age-dependent loss of arm cohesion . Recent work has demonstrated that Mtrm protein physically interacts with and inhibits the activity of Polo kinase in Drosophila oocytes from stage 11 until nuclear envelope breakdown ( NEB ) at stage 13 [22] . When Mtrm protein is reduced ( mtrm+/− ) , NEB can occur prematurely ( at stage 12 ) and the spindle assembles around a karyosome that is less compact than in wild type . Although this may account for the eventual missegregation of achiasmate chromosomes in mtrm+/− heterozygotes , sister chromatid cohesion remains intact in these oocytes as evidenced by the striking chiasmata visible in the individualized bivalents [22] . In addition , premature NEB in mtrm+/− oocytes still occurs much later than the stages that exhibit sensitivity to aging in our experiments ( stages 7–8 ) . Therefore , we conclude that reducing the dosage of Mtrm is not causing the increased missegregation of recombinant chormosomes that we observe in aged oocytes . Because the achiasmate segregation pathway is not completely disrupted in mtrmKG08051 heterozygotes , we also have been able to assess the effect of aging on the segregation of non-recombinant chromosomes . Interestingly , our data demonstrate that when the dosage of SMC1 is reduced , MI missegregation of non-recombinant chromosomes also increases with age . This result is reminiscent of our previous findings that NDJ of achiasmate chromosomes increases with age when activity of the meiotic cohesion protein ORD is compromised by a hypomorphic missense mutation [9] . In Drosophila oocytes , accurate segregation of achiasmate chromosomes relies on homologous pairing of centromere proximal heterochromatin [28]–[30] and both ORD and the cohesin complex are highly enriched at the percentric heterochromatin in oocytes [11] , [17] . In addition , our recent data indicate that heterochromatin-mediated pairing is moderately disrupted by ord mutations , even in the absence of aging ( VVS and SEB , unpublished results ) . Together , these data suggest that in addition to holding sister chromatids together , cohesion proteins play an important role in maintaining the heterochromatic pairing of achiasmate homologues in Drosophila oocytes . Whether cohesion proteins function directly to “glue” homologues together or play a more indirect role , such as influencing the structure of heterochromatin , remains to be determined . Regardless of the mechanism , loss of heterochromatin-associated cohesion proteins with age could account for the increased missegregation of non-recombinant chromosomes that we observe . Our data support the hypothesis that age-dependent NDJ in women is caused at least in part by progressive loss of sister-chromatid cohesion over time . One unresolved issue regarding meiotic cohesion in human oocytes is whether the original cohesin molecules used to establish cohesion are maintained for decades or continually replaced as oocytes age . Addressing this question will be essential in understanding the specific mechanisms that lead to loss of meiotic cohesion and chiasma maintenance during the aging process . Early experiments in yeast led to the widely accepted model that cohesion between sister chromatids can only be established during DNA replication [31] , [32] . However , recent evidence in mammalian tissue culture cells indicates that cohesin association with chromatin is much more dynamic than originally predicted [33] . In addition , experiments in yeast have recently demonstrated that cohesion can be re-established on a genome-wide scale during G2/M in a DNA damage-dependent manner [34] , [35] and support the model that non-canonical mechanisms can establish cohesion after S phase is completed . The power of genetic manipulation combined with the ability to age oocytes makes Drosophila an ideal model system to address whether re-establishment of meiotic cohesion occurs during prophase I . These future studies will be pivotal in understanding how cohesion dynamics during meiosis govern chiasma maintenance and ultimately , why and how loss of cohesion occurs as oocytes age . Flies were reared at 25°C on standard cornmeal molasses media . The smc1ex46 allele is a deletion that removes the gene [12] and is denoted as smc1− . The mtrm[KG08051] allele results from a P-element disruption of the gene [21] and is denoted as mtrm− . Descriptions of the other genetic markers and chromosomes used in this study can be found at http://www . flybase . org . Although several proteins ( Axs , Ald/Mps1 , Nod and Mtrm ) are required for the accurate segregation of achiasmate chromosomes in Drosophila oocytes , we considered a mtrm allele to be the ideal choice for our studies [21] , [22] . In contrast to nod , achiasmate segregation is severely affected in the mtrm heterozygote and chromosome loss is not observed [36] . In addition , unlike Axs oocytes , spindle defects and premature onset of anaphase I have not been observed in mtrm heterozygotes [37]–[39] . We thought it unwise to compromise achiasmate segregation using ald mutations given the role of this protein in the spindle assembly checkpoint; ald oocytes also exhibit low levels of NDJ of the exchange bivalents [40]–[42] . The specificity of the mutant phenotype in mtrm+/− oocytes combined with the observation that sister-chromatid cohesion appears to be normal convinced us that this was the best approach to disrupt the achiasmate segregation pathway . To assay X-chromosome crossover frequency and distribution in smc1+/− oocytes , 5–7 females were crossed to 3 yw males per vial . Crossover frequency and distribution was measured by assaying sc , cv , v , f , car markers in the male progeny . The recombination data was used to estimate tetrad exchange ranks [18] . To age Drosophila oocytes , the aging regimen described by Jeffreys et al . was modified [9] such that the glucose agar media was prepared without the addition of fungal inhibitors ( methyl paraben and ethyl acetate ) . The glucose agar media contained 2% agar ( Fisher ) and 5% dextrose ( Fisher ) and was prepared with milli-Q grade water . Yeast paste was prepared by dissolving 30 g of active dry yeast ( Red Star ) in 50 mL milli-Q grade water . A schematic of the aging regimen is shown in Figure 2B . Approximately 200 virgin females of the desired genotype were collected within an 8-hour period and the females were fed yeast overnight in vials with cornmeal molasses media to promote vitellogenesis . Overnight incubation of females with yeast allows yolk deposition and maturation of oocytes so that the ovaries contain a complete complement of oocytes at the different stages . The following day , females were split into two groups and placed in separate plexi-glass laying bottles containing glucose agar plate with a smear of yeast paste . The control group of females was supplied with an equal number of male flies and laid their eggs continuously . Their oocytes were “non-aged” . The experimental group of females were deprived of males . Because oviposition was suppressed in these females and the developmental progression of oogenesis was halted , oocytes “aged” within their abdomens . This experimentally induced aging of oocytes mimics the aging of oocytes in human females . Control and experimental flies were held in the laying bottles for four days with fresh yeast paste/glucose-agar plates supplied each day . At the end of the four-day aging regimen , the experimental females ( with aged oocytes ) and the control females ( with non-aged oocytes ) were crossed to X^Y , v f B males to measure meiotic nondisjunction in the oocytes ( see Figure 2C ) . To generate 24-hour broods , 7 female flies were mated with 3–5 X^Y , v f B males ( per vial ) . The parents were transferred to new vials every 24-hours and three broods of progeny were analyzed for NDJ . For simplicity , we have used the term “sub-broods” to differentiate 8-hour broods of progeny from the 24-hour broods of progeny . To generate sub-broods of progeny at the end of the aging regimen , 21 experimental or control females were crossed to 10 X∧Y , v f B males ( per vial ) . The parents were transferred to new vials every 8-hours for a total of 48 hours and six sub-broods of progeny were analyzed for NDJ . Because Drosophila can tolerate certain sex chromosome aneuploidies , segregation errors during meiosis can be monitored in the viable progeny by using differentially marked sex chromosomes ( see Figure 2C ) . To compensate for the fact that only half of the exceptional progeny survive ( see Figure 2C ) , total NDJ was adjusted according to the following formula: [2Exceptional Progeny/ ( 2Exceptional Progeny+Normal Progeny ) ]*100 . To assess whether recombinant bivalents missegregated in oocytes , the Diplo-X female progeny obtained from NDJ tests were genotyped . Each Diplo-X female was crossed to two yw males and the genotype of the X chromosomes of the Diplo-X female was inferred from the sc , cv , v , f , car markers in her male progeny ( see Figure 6 ) . Because some fraction of Diplo-X females either died before they could be genotyped or were sterile/sub-fertile , the number of Diplo-X females genotyped was lower than the number of Diplo-X females recovered from the NDJ test . After female flies were subjected to the four-day aging regimen , ovaries from experimental and control females were hand-dissected in modified Robb's buffer [36] at 8 hour intervals and fixed in 4% formaldehyde , PBS ( 130 mM NaCl , 7 mM Na2HPO4 and 3 mM NaH2PO4 ) for 10 min . The ovaries were rinsed in PBS containing 0 . 01% Tween-20 . Images were captured using a SMZ1500 stereo-microscope ( Nikon ) equipped with a Pixelink camera ( Diagnostic Instruments ) . To quantify the relative number of ooctyes at specific developmental stages , a single fixed ovary from three females was teased to generate individual ovarioles . Oocyte stages were tabulated and averaged for aged and non-aged ovaries at each time point . Oocytes were staged based on the standard morphological criteria described by Mahowald and Kambysellis [43] . Briefly , oocyte stages were distinguished based on the size of the egg chamber , the relative size of the oocyte and the presence of yolk within the oocyte . Yolk deposition begins during stage 8 and provides a convenient marker . Mature ( stage 13–14 ) oocytes were distinguished by the presence of chorionic appendages . Newly eclosed wild-type and smc1 heterozygous females were fattened with yeast for two days . 30 sets of ovaries from each genotype were dissected in modified Robb's buffer [36] and frozen in liquid N2 . Each frozen tissue sample was homogenized in 240 µL of buffer containing 8 M Urea , 2% SDS , 100 mM Tris HCl pH 6 . 8 and 5% Ficoll containing 4 mM AEBSF ( protease inhibitor , Sigma ) . The extract was cleared by centrifugation at 13 , 000 rpm for 10 min at room temperature and the supernatant was aliquoted and frozen in liquid N2 . The protein concentration for each extract was determined using a Bradford assay ( Biorad ) . For each genotype , 10 and 20 µg of total protein was separated by SDS-PAGE on a 7 . 5% gel and transferred to PVDF membrane . The blot was cut horizontally and anti-SMC1 guinea pig serum [11] was used at 1∶1000 dilution for the top half of the blot and monoclonal DM1A ( Sigma ) was used at 1∶15 , 000 to detect the α-tubulin loading control on the bottom half of the blot . To compare the relative frequency of missegregation occurring in aged and non-aged oocytes , an odds ratio ( OR ) was calculated . The odds ratio provides a method for determining whether the frequency of a certain event ( missegregation ) is equal for two different treatments ( aged and non-aged ) . In our analysis , it is calculated as: where Ae = # of exceptional progeny from aged oocytes , An = # of normal progeny from aged oocytes , NAe = # of exceptional progeny from non-aged oocytes , and NAn = # of normal progeny from non-aged oocytes [44] . If the treatments result in equal frequencies of the event , then the odds ratio will be 1 . In our analysis , an OR >1 implies that the frequency of missegregation is higher when oocytes undergo aging than when they do not undergo aging . Standard errors ( SE ) are calculated on the logarithm of the odds ratio according to: . In Figures 5 and 8 , error bars represent ±1 SElogOR retransformed back into non-logarithmic units . For the NDJ tests , P values were calculated using a 2×2 χ2 Contingency Test ( two-tailed ) . For these calculations , raw data values ( not adjusted values ) for a specific brood or sub-brood were compared for aged versus non-aged treatments ( # of exceptions and # of normal progeny for aged and non-aged ) . A 2×2 χ2 Contingency Test also was used to analyze the recombinational history of missegregating chromosomes recovered in Diplo-X progeny . For all tests , a P value of <0 . 05 was considered statistically significant ( rejection of the null hypothesis that the two groups are the same ) .	In humans , chromosome segregation errors during meiosis are the leading cause of birth defects and miscarriages . Moreover , as women age , these errors increase dramatically . For accurate segregation during the first meiotic division , homologous chromosomes must remain physically associated until anaphase I . Normally , attachments along the arms of sister chromatids keep the recombinant homologues together . Human oocytes complete meiotic recombination during fetal development and arrest until ovulation . Therefore , accurate segregation of homologous chromosomes during the first meiotic division requires that recombinant chromosomes remain associated for decades . One hypothesis to explain why segregation errors increase as women age is that the connections between sister chromatids deteriorate over time and allow recombinant homologues to dissociate prematurely . Here , we address this hypothesis using Drosophila as a model system . We find that when Drosophila oocytes undergo experimentally induced aging , recombinant homologues missegregate during meiosis I . Furthermore , the meiotic stage at which Drosophila oocytes are most vulnerable to age-induced errors is analogous to the stage at which human oocytes remain arrested for decades . Together , our data argue that aging does cause premature loss of the connections between meiotic chromosomes and that this is a major determinant of segregation errors in both Drosophila and human oocytes .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "developmental", "biology/germ", "cells", "developmental", "biology/aging", "genetics", "and", "genomics/chromosome", "biology" ]	2008	Aging Predisposes Oocytes to Meiotic Nondisjunction When the Cohesin Subunit SMC1 Is Reduced
Protein kinase B ( PKB/Akt ) belongs to the AGC superfamily of related serine/threonine protein kinases . It is a key regulator downstream of various growth factors and hormones and is involved in malignant transformation and chemo-resistance . Full-length PKB protein has not been crystallised , thus studying the molecular mechanisms that are involved in its regulation in relation to its structure have not been simple . Recently , the dynamics between the inactive and active conformer at the molecular level have been described . The maintenance of PKB's inactive state via the interaction of the PH and kinase domains prevents its activation loop to be phosphorylated by its upstream activator , phosphoinositide-dependent protein kinase-1 ( PDK1 ) . By using a multidisciplinary approach including molecular modelling , classical biochemical assays , and Förster resonance energy transfer ( FRET ) /two-photon fluorescence lifetime imaging microscopy ( FLIM ) , a detailed model depicting the interaction between the different domains of PKB in its inactive conformation was demonstrated . These findings in turn clarified the molecular mechanism of PKB inhibition by AKT inhibitor VIII ( a specific allosteric inhibitor ) and illustrated at the molecular level its selectivity towards different PKB isoforms . Furthermore , these findings allude to the possible function of the C-terminus in sustaining the inactive conformer of PKB . This study presents essential insights into the quaternary structure of PKB in its inactive conformation . An understanding of PKB structure in relation to its function is critical for elucidating its mode of activation and discovering how to modulate its activity . The molecular mechanism of inhibition of PKB activation by the specific drug AKT inhibitor VIII has critical implications for determining the mechanism of inhibition of other allosteric inhibitors and for opening up opportunities for the design of new generations of modulator drugs . Protein kinase B ( PKB/Akt ) is a key regulator downstream of various growth factors and hormonal signals . It activates a panel of proteins that regulate proliferation , growth , survival , or metabolism and is involved in human cancer [1 , 2] . In particular , its overexpression induces malignant transformation and chemoresistance [3] . PKB belongs to the AGC superfamily of related serine/threonine protein kinases . Three isoforms of PKB exist in mammals ( PKBα/Akt1 , PKBβ/Akt2 , and PKBγ/Akt3 ) that comprise an N-terminal pleckstrin homology ( PH ) domain , a flexible hinge between the PH and the kinase domain , a catalytic ( kinase ) domain , and a C-terminal regulatory part ( containing a hydrophobic motif , or HM ) ( for review [4 , 5] ) . The phosphorylation of Thr 308 in the kinase domain of PKBα/Akt 1 by phosphoinositide-dependent protein kinase-1 ( PDK1 ) [6] and Ser 473 in the hydrophobic motif by mTORC2 complex [7] and/or DNAPK [8] , is central for PKB activation [9] . These phosphorylations were shown to be dependent on the colocalisation of PKB and PDK1 at the plasma membrane through the interaction of their PH domains with PtdIns ( 3 , 4 , 5 ) P3 and PtdIns ( 3 , 4 ) P2 for the former [10–12] and PtdIns ( 3 , 4 , 5 ) P3 for the latter [13] . The dephosphorylation of these residues by okadaic acid-sensitive and -insensitive phosphatases was shown to deactivate PKB [14 , 15] . Recently , the cytoplasmic interaction of inactive PKB with PDK1 was published [14] . The maintenance of PKB in an inactive conformation by the interaction of its PH and kinase domains ( PH-in ) prior to stimulation prevented the phosphorylation of Thr 308 by the associated PDK1 . Upon stimulation , PKB PH-domain interaction with phosphoinositides and its concomitant change in conformation allowed the separation of the PH and kinase domains and the associated co-recruited PDK1 to phosphorylate Thr 308 . Here , the intramolecular interactions of the PKB domains in its inactive conformation were studied in detail using molecular dynamics in conjunction with classical biochemical approaches and in situ Förster resonance energy transfer ( FRET ) /two-photon fluorescence lifetime imaging microscopy ( FLIM ) . A PH domain-induced cavity was discovered in the inactive PKBα kinase domain . The formation of this structure was dependent on the PH domain residue Trp 80 . The position of PKBα C-terminal HM at the apex of the PH domain-induced cavity in the inactive conformation of PKB was also described . Unlike in PKBα , a PH domain-induced cavity did not appear in PKBγ suggesting a potential role of this structure in a differential regulation of these two PKB isoforms . These findings led to the elucidation of the molecular mechanism of inhibition of PKB by the highly specific allosteric inhibitor , AKT inhibitor VIII , and its relationship to the C-terminal HM motif of PKB . The novel findings from this study are summarised in Figure 6 ( see figure legend ) . The molecular mechanism of interaction of the allosteric inhibitor AKT inhibitor VIII to a critical PH domain residue , Trp 80 , was consistent with the model of binding through a PH-induced cavity formed in the kinase domain of PKB . This cavity was present in PKBα , whereas it was negligible in PKBγ . Chimera constructs of the PH and kinase domains suggested that the dimensions of the cavity were strongly correlated with the sensitivity to the inhibitor . The in vivo FRET data showed that in the presence of AKT inhibitor VIII , PKB was maintained in its PH-in conformation . In this conformation , Thr 308 and Ser 473 phosphorylation were impaired . However , a difference in the inhibition of phosphorylation between the two sites was observed . The inhibition of phosphorylation on Thr 308 was linked to PKB being maintained in its PH-in conformation and not being accessible to PDK1 . Recently published in vitro data indicated that the binding of PKB PH domain to phosphoinositides did not seem to be affected by the inhibitor interaction [21] . However , the translocation of the GFP-PKB-mRFP reporter was prevented in presence of inhibitor . These results indicated that the maintenance of PKB at the plasma membrane appeared to require a change in conformation of the protein . The lack of PKB translocation could explain the loss of Ser 473 phosphorylation upon inhibitor binding . However , our results indicated that the inhibitor binding created a steric hindrance preventing the upstream kinase ( s ) from phosphorylating Ser 473 . We have addressed the possible molecular mechanisms involved in the regulation of the PKB inactive conformation by examining intramolecular domain interaction dynamics . The discovery of a PH-kinase domain interface in relation to the C-terminal HM was essential in the understanding of the role of PKB quaternary structure in the control of this critical regulator . Yet , the role played by the flexible linker that joins the PH and the kinase domains close to this interface , might also be important to comprehend fully this regulation . In addition , we have depicted the possible molecular mechanism involved in the association of a new class of selective allosteric inhibitors that to date had remained elusive . We anticipate this will facilitate the optimisation of a new generation of more potent or more selective inhibitors . Okadaic acid sodium salt and AKT inhibitor VIII were from Calbiochem ( Merck KGaA ) . Human PDGF was from R&D Systems . Phospho-Akt Thr-308 ( both rabbit polyclonal and monoclonal ) and Ser-473 , as well as pan Akt , were from Cell Signaling ( New England BioLabs UK ) . Anti-GFP antibodies were in-house monoclonals . Other chemicals were from Sigma-Aldrich . Glass-bottom microwell culture dishes ( 35 mm; MatTek dishes ) were from MatTek Corporation . The Odyssey blocking buffer was from LI-COR UK . The IRdye 800 anti-rabbit and IRdye 700 anti-mouse were used as secondary antibodies for the LI-COR/Odyssey system ( from Rockland Incorporated ) . pEGFP-HA-Akt construct was described previously ( murine AKT1; Genbank accession number NM_009652 ) [23]; to simplify , we called this construct GFP-PKBα . GFP-PKB-mRFP was created by putting PCR-amplified mRFP ( EcoRI-XbaI ) instead of yellow fluorescent protein ( YFP ) in the vector pCDNA3-GFP-PKB-YFP ( described previously [14 , 24] ) , using the oligonucleotides mRFP-Ct sense: AGAGAATTCGCCTCCTCCGAGGA CGTCATCAAGGAGTTCATGCGCTTCAAGG and mRFP-Ct antisense: ATCTCTAGA TTATGCACCGGTGGAGTGGCGGCCCTCGGCGCGCTCG TACTGTTCC . Note that an extra XbaI site had to be removed using the Quick Change mutagenesis kit from Stratagene in the pCDNA3-GFP-PKB-YFP construct before the subcloning of mRFP ( EcoRI-XbaI ) . The W80A mutation on GFP-PKB-mRFP was performed by Maria Deak from Dario Alessi's laboratory ( Dundee , UK ) . GFP-PKBα ΔCt was created by PCR amplification of GFP-PKBα without the C-terminal part , using the oligonucleotides AKT1 ΔCt-sense: GGCATGGACGAGCT GTACAAGGGT and AKT1 ΔCt antisense: TACGGATCCTCACTCCTCATCGAAATACCTGG . The PCR amplification was then placed instead of wild-type PKBα in pEGFP-HA-Akt ( GFP-PKBα ) construct using SalI-BamHI restriction sites . GFP-PKBγ was created by PCR amplification of PKBγ from pCDNA3-Myr HA AKT3 human ( Addgene plasmid 9017 ) . The oligonucleotides used for the amplification were PCR-PKBγ sense: GTCGACATGTACCCATACGATGTTCCAGATTACGCTTCCAGAATGAGCGATGTTACCATTGTGAAAGAAGGTTGG and PCR-PKBγ antisense: GGATCCTTATTCTCGTCCACTTGCAGA GTAGGAAAATTGAGGG . Prior to the PCR amplification of PKBγ , an internal BamHI site was removed in pCDNA3-Myr HA AKT3 by direct site mutagenesis using: PKBγ-BamHI-less sense: CTTTCAGGGCTCTTGATAAAGGACCCAAATAAACGCCTTGG and PKBγ-BamHI-less antisense: CCAAGGCGTTTATTTGGGTCC TTTATCAAGAGCCCTGAAAG . The PCR product was subcloned into the intermediate cloning vector zero blunt TOPO from Invitrogen , then excised using SalI-BamHI restriction enzymes . The excised PCR amplification was then put in phase in C-terminus of GFP ( instead of PKBα in the vector GFP-PKBα cut with SalI-BamHI . The chimera of the PH domain and linker of PKBα in fusion with the kinase domain and C-terminal of PKBγ ( GFP-CHIM PHα-KINγ ) was done first by creating a modified GFP-PKBα construct containing a BamHI site using the oligonucleotides AKT1-BamON-sense: GGGGCTGAAGAGATGGAGGGATCCCTGGCCAAGCCCAAGCAC and AKT1-BamON-antisense: GTGCTTGGGCTTGGCCAGGGAT CCCTCCATCTCTTCAGCCCC . A PCR amplification of PKBγ kinase domain was then performed using the oligonucleotides ChimAKT3-sense: AGTGGATCCCTGGCCAAGCCCAAGCACAGAAAGA CAATGAATGATTTTGACTATTTG and ChimAKT3-antisense: ACTGGATCCTTA TTCTCGTCCACTTGCAGAGTAGG The PCR product was cut with BamHI and inserted in the modified GFP-PKBα vector containing a BamHI site . The mutants of PKB were obtained by direct site mutagenesis using the Quick Change mutagenesis kit from Stratagene using the oligonucleotides PKBα-W80A sense: CATCCGCTGCCTGCAGGCGACCACAGTCATTGAGCGC; PKBα-W80A antisense: GCGCTCAATGACTGTGGTCGCCTGCAGGCAGCGGATG; PKBα-Q218A sense: CGGCCCTCAAGTACTCATTCGCGACCCACGACCGCCTCTGC; PKBα-Q218A antisense: GCAGAGGCGGTCGTGGGTCGCGAATGAGTACTTGAGGGCCG; PKBγ-W79A sense: CAGATGTCTCCAGGCGACTACTGTTATAGAGAGAACATTTCATG; and PKBγ-W79A-antisense: CATGAAATGTTCTCTCTATAACAGTAGTCGCCTGGAGACATCTG . NIH3T3 cells from ATCC were maintained in DMEM 10% Donor Calf Serum ( DCS ) and seeded at 150 , 000 in a well of a six-well plate or in a MatTek dish . The transfection was done with 2 μg of DNA of the different constructs ( 2+2 μg for cotransfections ) using Lipofect AMINE/PLUS reagent ( GIBCO-BRL ) in OPTIMEM medium ( GIBCO-BRL ) as recommended by the manufacturer . The cells were left for 3 h in the transfection mix , and then the medium was removed and replaced with DMEM 10% DCS . The experiments were performed 24 or 48 h after transfection . After stimulation or treatment as indicated , the cells were lysed for 15 min on ice in lysis buffer ( 20 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , 100 mM NaF , 10 mM Na4P2O7 , 10 mM EDTA supplemented with COMPLETE protease inhibitor cocktail tablet [Roche] ) . To terminate the reaction , 4× SDS sample buffer ( 125 mM Tris-HCl [pH 6 . 8] , 6% SDS , 20% glycerol , 0 . 02% bromophenol blue supplemented with 10% β-mercaptoethanol ) was added , and the samples boiled for 5 min . The proteins were separated on a 10% SDS-PAGE gel . After electrophoresis , the polyacrilamide gels were transferred onto PVDF membrane ( Immobilon P; Millipore ) and incubated in blocking buffer TBS-T ( 10 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , 0 . 05% Tween-20 ) supplemented with 3% BSA for 1 h . The membranes were then incubated with the different antibodies: phospho-Akt ( Thr-308 ) antibody ( rabbit polyclonal from Cell Signaling ) , phospho-Akt ( Ser-473 ) ( rabbit polyclonal antibody raised against the C-terminus phospho-peptide HFPQFpSYSASS of Akt1 ) or with the pan AKT at 1:1 , 000 for 1 to 3 h in TBS-T/3% BSA . The membranes were then washed in TBS-T with 0 . 2% Tween-20 and 1% milk for 1 h . The secondary HRP-antibodies were used at 1:5 , 000 in TBS-T with 5% milk for 1 h . Western blots were revealed by incubation with ECL ( Amersham Biosciences ) . Density analysis of bands was done with NIH ImageJ 1 . 33u ( U . S . National Institutes of Health , http://rsb . info . nih . gov/ij/ ) . The analysis was performed by subtracting the background of the autoradiography and normalised as indicated on the graphs . After electrophoresis , the polyacrilamide gels were transferred onto PVDF membrane ( Immobilon FL; Millipore ) , incubated in the Odyssey blocking buffer for 1 h , and then with the different antibodies phospho-Akt ( Thr-308 ) antibody ( rabbit polyclonal or monoclonal from Cell Signaling ) or phospho-Akt ( Ser-473 ) ( rabbit polyclonal antibody raised against the C-terminus phospho-peptide HFPQFpSYSASS of Akt1 ) at 1:1 , 000 together with an anti-GFP in-house monoclonal at 1:2 , 500 for 4 h in Odyssey blocking buffer . The membranes were then washed in PBS 0 . 2% Tween-20 with 1% odyssey buffer for 1 h . The secondary antibodies IRdye 800 anti-rabbit and IRdye 700 anti-mouse ( Rockland ) were used at 1:2 , 500 and 1:5 , 000 , respectively , in Odyssey blocking for 1 h . The blots were scanned with the infrared LI-COR scanner , allowing for simultaneous detection of two targets ( anti-phospho and total protein ) in the same experiment . For each experiment , the values of the phospho bands were normalized for the total amount of protein in each lane . NIH3T3 cells were seeded at 150 , 000 on 35-mm glass-bottom tissue culture dishes ( MatTek ) and transfected as described above . Twenty-four hours after transfection , the cells were treated in DMEM containing 10% DCS . The cells were washed twice with PBS , and then fixed in 4% paraformaldehyde in PBS for 10 min . The dishes were washed twice with PBS , and then 2 ml of PBS supplemented with 2 . 5% ( w/v ) DABCO as an antifade was added to the dishes . The images were acquired straightaway or stored at 4 °C . Details about the method to detect FRET by time domain FLIM was as previously described [14 , 25] . We used a nonparametric Mann-Whitney test to compare the medians of the two datasets for the two-photon FLIM data using GraphPad InStat software ( version 3 . 0 for Mac-2001 ) . To interpret the distribution of data , box and whisker plots were used . The box and whisker plot is a histogram-like method for displaying upper and lower quartiles , and maximum and minimum values in addition to median . In order to check the validity of the models , we performed several molecular dynamics simulations . Vacuum calculations were performed on an SGI Octane workstation using InsightII and Discover ver . 2000 ( Accelrys ) . The hydrated models calculations were performed using NAMD [26] software v2 . 6 and CHARMM 27 force-field . All simulations were performed on a 32-processor IBM PC-cluster using 8 to 16 processors . The simulation boxes were built using VMD v1 . 8 . 6 facilities . All images were captured from VMD [27] using the internal ray-tracing software Tachyon . Two models were built starting from the model of the PH/KIN PKBα complex previously described [14] . The whole complex was immersed in a box of water , fully ionized , with all charges neutralized by counter ions , and salt was added to reach the biological concentration of 150 mM . The wild type was built first , and the W80 was subsequently mutated into A80 . Model 1 ( wild type ) totalled 47 , 114 atoms with 13 , 394 water molecules ( TIP3P ) in a box of 88 . 94 × 73 . 99 × 77 . 87 Å3 , at T = 300 K and P = 1 bar . Time of simulation: 6 . 22 ns . Model 2 ( W80A ) totalled 47 , 103 atoms with 13 , 395 water molecules ( TIP3P ) in a box of 88 . 94 × 73 . 99 × 77 . 87 Å3 , at T = 300 K and P = 1 bar . Time of simulation: 6 . 13 ns . During the two simulations , a water cavity opened quickly ( after about 0 . 2 ns ) just in front of residue 80 going from the outer surface of the protein directly to the surface of residue 80 . For wild-type ( PKBα W80 ) , the cavity was stable and almost as wide as deep . W80 was located about 8–8 . 5 Å from the outer surface of the protein , and the cavity had a maximum opening of around 8–9 Å . The cavity was full of water ( see Figure 2 ) . For the W80A mutant , the cavity appeared more unstable with fluctuations and was deeper than it was wide . A80 was located 14–15 Å from the protein surface ( compared to 8–8 . 5 Å in the wild type ) , and the maximum opening was around 6–7 Å . The cavity was again full of water . The final model of the wild-type PH/KIN PKBα complex after molecular dynamics simulation was superimposed ( kinase domain only ) over the published X-ray structure of AKT-2 ( PKBβ ) [22] ( PDB codes: 1O6K and 1O6L ) . After superimposition , the kinase domain of PKBβ was deleted , leaving the C-terminal ( 466–479 ) of PKBβ lying on the surface of the PH/KIN complex of PKBα . The C-terminal was then mutated to correspond to the sequence of the C-terminal of PKBα , as below: 1O6K:QRTHFPQFDYSASI PKBα:RRPHFPQFSYSASG Underlined residues were mutated during the building ( only three residues ) . The bold and underlined S474D mutation , which was introduced in PKBβ [16] , was also present ( as S473D ) in the first model of PKBα . However , in a second step , the reverse mutation D473S was built in order to get back to the wild-type model . With the D473 mutation , a first model was built in a box of dimensions: 78 . 74 × 91 . 28 × 73 . 66 Å3 , at T = 300 K and P = 1 bar , with a total amount of 48 , 537 atoms of which 13 , 793 were water molecules , simulation time: 3 . 43 ns . During the simulation , an interaction between R465 and D473 was observed due to the fact that the C-terminal was truncated on its N-terminus . Thus the N-terminal of the C-terminal was moved toward the surface of the protein away from D473 , and a second model was built: 79 . 52 × 91 . 79 × 73 . 82 Å3 , at T = 300 K and P = 1 bar , with a total amount of 49 , 619 atoms of which 14 , 153 water molecules , simulation time: 7 . 2 ns . With the reverse D473S mutation , one model was built: 78 . 74 × 91 . 28 × 73 . 66 Å3 , at T = 300 K and P = 1 bar , with a total amount of 48 , 531 atoms of which 13 , 872 were water molecules , simulation time: 11 . 2 ns . Even though the C-terminal is an isolated short peptide strand , it displays a sustainable stability for the duration of the various molecular dynamics runs . This is due to many close contacts and hydrogen bonds with mainly the kinase domain . There are hydrogen bondings of the C-terminal backbone with several polar groups of the kinase domain: F472 with S216 , F469 with Q218 , P467 with T219 , and a strong salt bridge between R465 and E184 . A full p-stacking ( face to face ) between the F462 and F225 residue is observed . Very close contacts between the C-terminal and the following residues in the kinase domain: K183 , V187 , A188 , H220 , D221 , and L223 are also observed . It is of note that when the C-terminal was added to the PH/KIN complex , the number of salt bridges between the PH and kinase domains increased from seven bridges for the PH/KIN complex to ten bridges for the full PH/KIN/C-term complex . This implied a stabilization of the AKT complex upon binding of the C-terminal . The starting point was the previously equilibrated model of the PH/KIN complex of human PKBα with the cavity filled with water molecules . The inhibitor was carefully docked within the water cavity . The complex was submitted to a simulated annealing procedure to get the best docking solution using the Affinity module within Insight software . Initially , random docking was performed with a flexible ligand without electrostatic interactions and with Van der Waals interactions reduced to 10% . At a second stage , the best solutions were submitted to a simulated annealing ( SA ) procedure with all nonbonded interactions on all free protein side-chains and the tethered backbone . Fifty stages of SA of 100 fs each were performed from 500 to 300 K before full minimization ( 1 , 000 steps ) . The best solution was submitted to several 1-ns molecular dynamics without any constraints . In all cases , the benzimidazolone moiety located at the extremity of the inhibitor was associated by hydrogen bonding to the residues Q218 , T219 , or S216 . The imidazolo-quinoxaline moiety remained in close contact with the key residue W80 ( Figure S2 ) . The aromatic part of the inhibitor , which was embedded inside the cavity , was surrounded by a hydrophobic core containing: I186 , F225 , and H194 residues , as well as the W80 .	A critical protein in cell-signalling pathways , called protein kinase B , regulates many aspects of cell biology from metabolism to proliferation and survival , by modifying other proteins with the addition of a phosphate group . Hence , deregulation of its activity has acute consequences on cell function . Increased activity of a tumour-promoting form of protein kinase B or of upstream regulatory proteins has been observed in tumours , while impaired protein kinase B function has been linked to diabetes . Therefore , understanding the molecular mechanism of protein kinase B activation will help reveal how its activity might be regulated to limit disease progression . Toward this end , we studied how protein kinase B structure relates to its function , to identify molecular mechanisms regulating its kinase activity , modifying its cellular localization , and altering its binding to other proteins . By determining the spatial organization of different regions of the protein in inactive protein kinase B , we discovered a cavity at the interface of two distinct functional regions of the inactive form . We also localized the C-terminal end of the protein to the apex of the cavity , suggesting a role of this domain in regulating the inactive form of the protein . This represents a novel example of negative regulation by inhibition across these different regions of the protein . From these findings , we elucidated the mechanism of action of a highly specific protein kinase B inhibitor , AKT inhibitor VIII . We determined that simultaneous binding of the inhibitor to the two different functional regions , through the cavity , “locks” protein kinase B in an inactive conformation and prevents regulatory proteins from accessing the C-terminal domain .	[ "Abstract", "Introduction", "Results", "and", "Discussion", "Materials", "and", "Methods" ]	[ "biochemistry", "cell", "biology", "computational", "biology", "biophysics", "molecular", "biology" ]	2009	Role of a Novel PH-Kinase Domain Interface in PKB/Akt Regulation: Structural Mechanism for Allosteric Inhibition
Various ascomycete fungi possess sex-specific molecular mechanisms , such as repeat-induced point mutations , meiotic silencing by unpaired DNA , and unusual adenosine-to-inosine RNA editing , for genome defense or gene regulation . Using a combined analysis of functional genetics and deep sequencing of small noncoding RNA ( sRNA ) , mRNA , and the degradome , we found that the sex-specifically induced exonic small interference RNA ( ex-siRNA ) -mediated RNA interference ( RNAi ) mechanism has an important role in fine-tuning the transcriptome during ascospore formation in the head blight fungus Fusarium graminearum . Approximately one-third of the total sRNAs were produced from the gene region , and sRNAs with an antisense direction or 5′-U were involved in post-transcriptional gene regulation by reducing the stability of the corresponding gene transcripts . Although both Dicers and Argonautes partially share their functions , the sex-specific RNAi pathway is primarily mediated by FgDicer1 and FgAgo2 , while the constitutively expressed RNAi components FgDicer2 and FgAgo1 are responsible for hairpin-induced RNAi . Based on our results , we concluded that F . graminearum primarily utilizes ex-siRNA-mediated RNAi for ascosporogenesis but not for genome defenses and other developmental stages . Each fungal species appears to have evolved RNAi-based gene regulation for specific developmental stages or stress responses . This study provides new insights into the regulatory role of sRNAs in fungi and other lower eukaryotes . The ascomycete fungus Fusarium graminearum is a major causative agent of Fusarium head blight ( FHB ) in small-grain cereals worldwide [1] . The fungus reproduces using sexual spores ( ascospores ) and asexual spores ( conidia ) . Although both spore types contribute to disease initiation and propagation , ascospores serve as the primary inocula for FHB outbreaks because these spores are forcibly discharged into the air and can move long distances [2] . Additionally , the sexual development process ensures the production of survival structures required for overwintering [3] and the genetic diversity of the population [4]; therefore , understanding the molecular mechanisms underlying F . graminearum sexual reproduction is important for developing novel FHB disease-control strategies . Perithecia ( fruiting bodies ) produced via sexual reproduction have complex multicellular structures with three distinguishable layers [5] . The perithecial outer layer composed of black and hard tissues/cells protects the inner components of the perithecia . Asci are biseriately arranged within perithecia and generate turgor pressure by accumulating osmolytes for forcible ascospore discharge [6] . Inside an ascus , a diploid zygote nucleus undergoes normal meiosis and mitosis to produce eight haploid nuclei , which are then sequestered into eight ascospores [7] . Various genetic and metabolic processes are intricately involved in sexual reproduction; these processes are likely fine-tuned temporally and spatially depending on each step of sexual development [5 , 8] . Many upstream signal transduction pathways and hundreds of signaling mediators , such as transcription factors and kinases , orchestrate sexual reproduction in F . graminearum [9–11] . RNA interference ( RNAi ) is a conserved gene-silencing mechanism that occurs at the post-transcriptional or post-translational level in eukaryotes , including fungi [12–14] . RNAi is triggered by small noncoding RNAs ( sRNAs ) of approximately 20–30 nucleotides , which are generally categorized as short interfering RNAs ( siRNAs ) , microRNAs ( miRNAs ) , and piwi-interacting RNAs ( piRNAs ) ; siRNAs and miRNAs have been extensively studied in fungi , although some piRNA-like sRNAs have been identified in Neurospora crassa [13 , 15] . The siRNA and miRNA duplexes are produced from double-stranded RNA ( dsRNA ) precursors that are cleaved by RNase III-family nuclease Dicers . The resulting siRNA or miRNA duplexes are loaded into an Argonaute protein of the RNA-induced silencing complex ( RISC ) . Subsequent removal of the passenger strands of sRNA duplexes leads to the activation of the RISC , and this RISC-incorporated guide-strand sRNA is then used to identify complementary mRNA for silencing by either mRNA degradation or translational repression . Elucidating the involvement of RNAi in quelling [16] and meiotic silencing by unpaired DNA ( MSUD ) [17 , 18] in N . crassa has been a major recent breakthrough in the genetics of filamentous fungi . Most fungal species possess functional RNAi components except for some single-cell yeasts [19] . Moreover , dsRNA-mediated gene silencing has been widely reported in fungi [13 , 20] , which is promising for the development of an efficient gene-silencing method or the utilization of host-induced gene silencing for disease control [21 , 22] . Advanced next-generation sequencing technologies and bioinformatics tools have enabled researchers of non-model filamentous fungi to identify sRNAs . However , the biological roles of sRNAs in fungal development have rarely been identified , while RNAi pathways have crucial regulatory functions in cellular processes such as development , host defenses , mRNA processing , transcription , and translation in animals and plants [23–25] . Few sRNAs with important biological roles have been characterized in fungi , and sRNAs have been hypothesized to function under specific developmental or environmental conditions [26] . RNAi generally functions in genome defense from viruses and transposable elements in eukaryotes , including fungi [14 , 22] . The possible involvement of sRNAs in fungal development or responses to environmental stimuli has been proposed in Aspergillus flavus ( responses to water activity and temperature ) [27] , Magnaporthe oryzae ( fungal development ) [28] , and Penicillium marneffei ( transition between mycelia and yeast phases ) [29] . Previous studies have proposed biological roles for RNAi based on phenotypic defects of RNAi-deficient mutants . In the ascomycete fungus Trichoderma atroviride , different components of the RNAi machinery are involved in light-dependent asexual reproduction and light-independent hyphal growth [30] . Sexually induced RNAi machinery is required for sexual reproduction but not for virulence in the basidiomycete human pathogen Cryptococcus neoformans [31] . Components of the RNAi machinery involved in MSUD are also important for ascospore formation in N . crassa [18 , 32 , 33] . Null mutants of RNAi components showed various developmental defects , including dysfunction during sexual and asexual reproduction , in Mucor circinelloides [34] . In plant pathogenic fungi , important roles of pathogen-derived sRNAs in host-microbe interactions have been recently reported [35] . Botrytis cinerea microRNA-like RNAs ( milRNAs ) silence genes involved in host defense by inhibiting the host RNAi pathway [36] . However , there is little evidence from genome-wide approaches to characterize transcriptional reprograming by sRNAs during fungal development . Our previous study revealed that the homothallic fungus F . graminearum possesses a functional MSUD mechanism , as is the case with the heterothallic fungus N . crassa [7] . The presence of MSUD led us to investigate the roles of sRNA-mediated RNAi in F . graminearum sexual development . A recent study revealed that hairpin RNA silenced target genes , and specific RNAi components were involved in this silencing process in F . graminearum [37] . Moreover , one of the Argonaute genes ( FgSMS2/FgAGO2 ) was shown to be under the control of a mating-type gene and was required for sexual development in this fungus [9] . Thus , the objectives of the present study were to provide an in-depth characterization of the sexual phenotypes of RNAi component mutants and to determine the genome-wide correlation of sRNAs with the transcriptome and degradome during the sexual development of F . graminearum . This is the first genome-wide characterization of sRNAs involved in balancing transcript levels during fungal sexual development , and it provides insights into the role of sRNAs in fungi . F . graminearum possesses two Dicers and two Argonautes , as previously reported [37] . Both Dicers ( FgDicer1 and FgDicer2 ) contain domain architectures similar to those of other representative orthologs; however , FgDicer1 and Dcl1 of Schizosaccharomyces pombe do not carry the double-stranded RNA-binding ( DSRM ) domain ( Fig 1A ) . Each Argonaute has a conserved catalytic triad ( Asp-Asp-His/Asp , DDH/D ) ; FgAgo2 contains the third histidine residue but is replaced by an aspartate in FgAgo1 . FgAGO2 was also annotated as FgSMS2 in a previous study [9] . We measured the transcript levels of these genes during vegetative growth and sexual development ( Fig 1B ) . The expression levels of FgDICER1 and FgAGO2 were significantly increased after sexual induction , suggesting that they have active roles during perithecia development in F . graminearum . The transcript levels of FgDICER2 were similarly maintained throughout the developmental stages , while FgAGO1 transcripts accumulated to high levels as mycelia aged ( V5 , 5 days after inoculation ) and as perithecia matured ( S3–7 , 3–7 days after sexual induction ) . To characterize the roles of these genes in fungal development , we generated each gene deletion mutant using the homologous recombination method ( S1 Fig and S1 Table ) . Although all of the tested F . graminearum strains produced mature perithecia with similar numbers , the Fgdicer1 and Fgago2 mutants were severely defective in forcible ascospore discharge , while the Fgdicer2 and Fgago1 mutants showed indistinguishable phenotypes compared to those of the wild type ( Fig 1C and 1D ) . Except for defective sexual development , we could not find any defect in vegetative growth , virulence , or stress responses in the mutants , in accordance with a previous report [37] . Double-deletion mutants were generated to verify the redundant roles of Dicers or Argonautes . Fgdicer2 and Fgago1 mutants were first outcrossed with the Δmat2 strain to produce the HK338 ( Δmat2 ΔFgdicer2 ) and HK339 ( Δmat2 ΔFgago1 ) strains , respectively ( S1 Table ) . Then , the double-deletion mutants HK340 ( ΔFgdicer1 ΔFgdicer2 ) and HK341 ( ΔFgago1 ΔFgago2 ) were generated from the outcrosses HK338 × Fgdicer1 and HK339 × Fgago2 , respectively . Deletion of both FgDICER1 and FgDICER2 almost completely abolished ascospore discharge , whereas Fgdicer1 mutants discharged reduced amounts of ascospores compared to those of the wild type , suggesting a synergistic effect of mutant phenotypes between single- and double-deletion mutants ( Fig 1C and 1D ) . Double-deletion Fgago1 Fgago2 mutants showed similar phenotypes to those of Fgago2 mutants with respect to forcible ascospore discharge . Mutant phenotypes and synergisms were confirmed using several complementation assays ( S2 Fig ) . First , reintroduction of the intact FgDICER1 and FgAGO2 genes into each corresponding single deletion mutant restored defective sexual development ( Fig 1C and 1D ) . Next , the roles of FgDICER2 and FgAGO2 were further verified by introducing each gene into the double-deletion mutants using the outcrosses Δmat2 × HK340 ( ΔFgdicer1 ΔFgdicer2 ) and Δmat2 × HK341 ( ΔFgago1 ΔFgago2 ) . The resulting strains ( ΔFgdicer1 FgDICER2 and ΔFgago1 FgAGO2 in Fig 1C and 1D ) showed similar phenotypes to those of the Fgdicer1 and Fgago1 mutants . To determine the mechanism underlying defective ascospore discharge , we dissected mature perithecia to observe rosette asci ( Fig 1E ) . The F . graminearum wild-type strain produced eight normal spindle-shaped ascospores per ascus . However , some asci of the mutants that were defective in ascospore discharge ( ΔFgdicer1 , ΔFgago2 , ΔFgdicer1 ΔFgdicer2 , and ΔFgago1 ΔFgago2 ) contained abnormally shaped ascospores and fewer than eight per ascus . The morphologies of abnormal-shaped ascospores vary , including larger ones , smaller ones , and broken ones . Moreover , small , ghost-like ascospores with incomplete spore delimitation were observed in these mutants ( Fig 1E ) . Thus , we concluded that defective forcible ascospore discharge of Dicer or Argonaute mutants was due to abnormal ascospore production . To investigate the molecular mechanisms underlying defective ascosporogenesis of the RNAi component mutants , we analyzed the transcriptomes of the F . graminearum strains using RNA-seq . Based on a threshold of reads per kilobase of exon per hundred million mapped reads ( RPKHM ) values ( ≥ 10 ) under all tested conditions , 11 , 908 of 13 , 820 genes were selected for further analyses ( S2 Table ) . Differentially expressed genes ( DEGs ) were identified as genes displaying a greater than 3-fold change in transcript levels compared to those of the wild type . Deletion of both FgDICER1 and FgDICER2 induced the differential expression of 380 genes; 241 genes were upregulated , and 139 genes were downregulated ( Fig 2A and 2B ) . While most of the upregulated genes in the Fgdicer2 mutant overlapped with those of the Fgdicer1 Fgdicer2 mutant ( 15/18 ) , 70% of the genes ( 58/84 ) were specifically upregulated in the Fgdicer1 mutant ( Fig 2A ) . Similarly , 80% ( 297/378 ) and 61% of the genes ( 73/119 ) were increased or decreased , respectively , in their expression only in the Fgago2 mutant but not in the Fgago1 Fgago2 mutant ( Fig 2A and 2B ) . In contrast , most of the downregulated genes in each Dicer single mutant were also reduced in transcript levels in the Fgdicer1 Fgdicer2 mutant ( Fig 2B ) . Transcriptomes between the double-deletion mutants , Fgdicer1 Fgdicer2 and Fgago1 Fgago2 , were highly positively correlated ( R2 = 0 . 91; Fig 2C ) , and most DEGs overlapped in these mutants ( Fig 2D ) , suggesting that these two mutants were impaired in the same regulatory pathway . However , each transcriptome of Fgdicer1 and Fgdicer2 showed a mild positive correlation with that of Fgdicer1 Fgdicer2 , supporting the functional synergism of FgDicer1- and FgDicer2-mediated transcriptional regulation during ascospore formation in F . graminearum ( Fig 2E ) . While the transcriptome of the Fgago1 mutant showed a relatively lower positive correlation with that of the Fgago1 Fgago2 mutant , transcriptomes between the Fgago2 and Fgago1 Fgago2 mutants had a much higher correlation ( slope = 0 . 72 and R2 = 0 . 36 ) . Taken together , these results showed that the two Dicers and two Argonautes have redundant roles in the same biological pathway affecting ascospore formation in F . graminearum . To dissect the characteristics of DEGs via disruption of the RNAi pathway , we examined the expression profiles of the upregulated ( 200 ) and downregulated ( 112 ) genes in both the Fgdicer1 Fgdicer2 and Fgago1 Fgago2 mutants compared to those of the wild type . Raw RNA-seq data from F . graminearum sexual developmental stages were obtained from a previous study [8] , and the time-course expression patterns during sexual development were clustered into 10 groups using the R package Mfuzz [39] with the default setting , which performs fuzzy c-means clustering ( S3 Fig and S3 Table ) . Mating-type genes ( MATs ) are important sex-specific upstream regulators that orchestrate sexual reproduction processes in F . graminearum , and the RNAi pathway was proposed to be a downstream regulatory mechanism of MATs [9] . MAT1-1-2 and MAT1-1-3 were included in group 2 , and MAT1-1-1 and MAT1-2-1 were members of group 9 ( Fig 3A and S3 Table ) . More than half of the upregulated genes were included in groups 2 or 9 , in which the transcript levels were decreased at 4 days after sexual induction ( Fig 3A and S4 Table ) . Group 5 possessed approximately 40% of the downregulated genes , and the transcript levels of group 5 genes were generally increased 4 days after sexual induction ( Fig 3A ) . In short , the RNAi pathway affects the transcript abundance of genes closely related to the MAT-mediated regulatory mechanism during the late stages of sexual development in F . graminearum . Our transcriptome analysis revealed that the RNAi pathway affects several important sexual reproduction-related molecular processes in F . graminearum ( S2 Table ) . Pheromone precursor gene PPG1 ( Gene ID: FGSG_05061 ) , two putative transcription factors ( Gene ID: FGSG_01366 and FGSG_16753 ) , and transaldolase ( Gene ID: FGSG_13162 ) , which are known to be important for sexual development , were significantly downregulated in RNAi-deficient mutants [9 , 10 , 42] . Moreover , three of these genes ( FGSG_01366 , FGSG_13162 , and FGSG_16753 ) are under the control of MAT genes in F . graminearum [9] . In addition , 11 fungal-specific transcription factors with a Zn ( II ) 2Cys6 fungal-type DNA-binding domain ( FGSG_01760 , FGSG_03892 , FGSG_03912 , FGSG_04782 , FGSG_04786 , FGSG_09111 , FGSG_11186 , FGSG_11358 , FGSG_11672 , FGSG_12597 , and FGSG_16075 ) were downregulated in RNAi component mutants; null mutants of these factors did not show any mutant phenotype [10] . Gene Ontology ( GO ) enrichment analysis was additionally applied to classify the functions of the predicted DEGs , and the GO terms were statistically analyzed using GOstats [40] and subsequently visualized using REVIGO [41] . REVIGO provides a network structure of non-redundant GO terms ( Fig 3B and 3C ) . Although we did not find any functionally characterized genes from the upregulated DEGs , they were assigned to 62 categories; 27 were included in biological processes , 26 in molecular functions , and 9 in cellular components . Accordingly , the most significant GO terms were related to the transport of organic compounds such as “amino acid transport” and “carbohydrate transport” of biological processes and "transmembrane transporter activity" of molecular functions ( Fig 3B ) . Downregulated genes were assigned to 76 categories ( 36 in biological processes , 36 in molecular functions , and 4 in cellular components ) . Most GO terms corresponding to negatively regulated genes were involved in carbohydrate catabolism , such as “carbohydrate catabolism” and “polysaccharide catabolism” ( Fig 3C ) . Based on our GO enrichment analysis , we demonstrated that one of the important molecular processes regulated by the RNAi pathway is carbon metabolism . Production of triacylglycerol is important for perithecia development , and therefore , dynamic changes in carbon metabolism should occur during sexual reproduction processes in F . graminearum [43] . Accordingly , acetyl coenzyme A production and translocation between cellular organelles are closely involved in various steps of sexual development in F . graminearum [44–47] . In our transcriptome data , transaldolase ( Gene ID: FGSG_13162 ) is involved in the pentose-phosphate pathway , and ascospore formation [9] was markedly downregulated in RNAi-deficient mutants ( S2 Table ) . To determine the relationship between sRNA production and transcriptome alteration , we isolated total low-molecular-weight RNAs from the wild-type and RNAi component mutant strains 5 days after sexual induction and used them for sRNA sequencing ( S5 Table ) . After the adaptor sequences were trimmed , raw reads were normalized using DESeq to adjust for differences in library sizes [48] . Total reads with 18–32 nt perfect matches were used for alignment to the genome of F . graminearum [4 , 49] . The size distributions and 5′ end frequencies of sRNAs were analyzed . Identified sRNAs were 18–32 nt long , with a clear peak at approximately 24 nt in the F . graminearum wild-type strain ( Fig 4A ) . The sRNAs with a 5′ end with U ( 5′-U ) accounted for 70% of the total sRNAs in the wild-type strain and were mostly 21–26 nt long ( Fig 4A and 4B ) . Deletion of either FgDICER1 , FgDICER2 , or FgAGO2 attenuated the biases of sRNAs toward specific sizes ( 21–26 nt ) and 5′-U; these changes were milder in Fgdicer1 than those in Fgdicer2 and Fgago2 . Deletion of FgAGO1 did not result in distinct changes in sRNA characteristics , but results for the Fgago2 mutant were similar to those of Fgdicer2 ( Fig 4B ) . Fgdicer1 Fgdicer2 and Fgago1 Fgago2 mutants typically abolished these biases . Taken together , these findings showed that F . graminearum has distinct FgDicer1- , FgDicer2- , or FgAgo2-dependent sRNA production/accumulation mechanisms and that these sRNAs have strong tendencies toward being 24 nt long and 5′-U . Moreover , both members of Dicers and Argonautes have functional redundancies . To investigate the origin of sRNAs , we mapped sRNA sequences to the F . graminearum genome [4] with a recent annotation [49] to identify their various genomic features , such as coding sequences of rRNA , tRNA , and predicted proteins as well as intergenic regions ( S6 Table ) . In the wild-type strain , sRNAs were in the rRNA region composed of approximately 40% of the total sRNAs ( Fig 4C ) . Approximately 30% and 27% of the sRNAs mapped to protein-coding genes and intergenic regions , respectively . sRNA-producing loci frequently covered both protein-coding sequences and adjacent regions . Because there is little information about the untranslated region ( UTR ) in a recent annotation of F . graminearum [49] , 500 bp upstream and downstream regions were arbitrarily denoted as 5′-UTR and 3′-UTR , respectively . Thereafter , 27% of the intergenic sequences were again separated into 16% of the UTRs and 11% of the intergenic sequences; predicted sRNAs corresponding to putative transcript sequences ( protein-coding regions and UTRs ) accounted for 46% ( Fig 4C ) . Henceforth , we will call these sRNAs exonic siRNAs ( ex-siRNAs ) , as previously reported [50] . When Dicer or Argonaute genes were deleted , the ratios of most genomic features aligned with sRNAs , except that the rRNA , protein-coding gene ( sense strand ) , and UTR ( sense strand ) , were decreased with analogous tendencies shown for size distributions and 5′-U in the corresponding deletion mutants ( Fig 4C ) . In particular , sRNAs mapped to antisense sequences of putative transcripts were markedly reduced in the Fgdicer1 Fgdicer2 and Fgago1 Fgago2 mutants . Various genomic features of sRNA-producing sequences were further assessed in detail with total normalized reads . The sRNAs produced from protein-coding genes converged on exons rather than introns ( Fig 4D ) . The total amounts of sRNAs were decreased in the Fgdicer1 , Fgdicer2 , Fgdicer1 Fgdicer2 , Fgago2 , and Fgago1 Fgago2 mutants , principally because of the reduced number of introns , exons , and UTRs with an antisense direction . Dicer- and Argonaute-dependent sRNAs with 5′-U ( 22–25 nt ) were particularly enriched in antisense sequences of putative transcripts ( Fig 4D and S6 Table ) . Moreover , the results confirmed that Dicers and Argonautes have redundant roles . Interestingly , total counts of sRNAs with antisense sequences of tRNA and rRNA were also markedly reduced in the Fgdicer1 Fgdicer2 and Fgago1 Fgago2 mutants , primarily because of a decreased number of sRNAs with 5′-U . The results suggest that Dicer- and Argonaute-dependent biogenesis/accumulation of sRNAs occurs at a genome-wide level . As most sRNAs were produced from gene regions , except for rRNA-mediated sRNAs , we focused on the characteristics of sRNA-producing genes ( S7 Table ) . The F . graminearum wild-type strain produces sRNAs from 5180 genes , and 80% ( 4155/5180 ) of the genes produced only sense-specific sRNAs , suggesting that sense sRNAs were byproducts of mRNA degradation ( Fig 5A and S8 Table ) . Accordingly , the number of genes producing only sense sRNAs was similar among the wild-type and RNAi component mutant strains . We investigated 5′-U sRNA-producing genes from F . graminearum strains ( Fig 5B and S8 Table ) . The genes were selected when the amount of 5′-U sRNA was more than 70% of the total sRNAs ( S7 and S8 Tables ) . Approximately 1000 genes preferentially produced 5′-U sRNAs in the F . graminearum wild-type strain . Among them , more than half of the genes generated sRNAs with both orientations , and 10% of the genes produced only sense-specific sRNAs . A markedly reduced number of genes produced 5′-U sRNAs in Fgdicer2 and Fgago2 compared to those of the wild type , whereas the Fgdicer1 mutant showed a slight reduction in the number of 5′-U sRNA-producing genes . Double deletion of FgDICER1 and FgDICER2 largely abolished 5′-U sRNA production . Genes producing antisense-specific or sense-specific sRNAs were counted in the Fgago1 Fgago2 mutant . Taken together , these findings show that most sRNAs originating from the gene region with a 5′-U or antisense direction were distinctly produced in a Dicer-dependent manner . We investigated whether sRNAs affected the transcript levels of corresponding genes in F . graminearum . First , all genes were categorized depending on the levels of sense or antisense sRNA counts , and they were applied in correlation analyses of transcript abundance between Fgdicer1 Fgdicer2 and Fgago1 Fgago2 versus the wild-type strain ( Fig 5C ) . As the genes produced more sense or antisense sRNAs , a greater number of genes tended to accumulate more transcripts in Fgdicer1 Fgdicer2 and Fgago1 Fgago2 than those in the wild type . In particular , most genes producing either sense or antisense sRNAs at more than 1000 counts per kilobase ( red dots in Fig 5C ) were upregulated in the RNAi-deficient mutants . To analyze the negative correlation between sRNA production and transcript amounts of genes in detail , we assessed the gene numbers depending on transcript abundance in Fgdicer1 Fgdicer2 or Fgago1 Fgago2 versus the wild-type strain ( Fig 5D ) . Whereas both sense and antisense sRNAs produced below 100 counts per kilobase did not show substantial changes in transcript levels of the corresponding genes , most genes producing antisense sRNAs at more than 1000 counts per kilobase ( red graphs ) were positively regulated in Fgdicer1 Fgdicer2 and Fgago1 Fgago2 compared to those in the wild type ( Fig 5D ) . Both sense and antisense 5′-U sRNA ( 22–25 nt ) -producing genes showed similar results to those obtained from antisense sRNAs ( S4 Fig ) . Taken together , these findings show that a sRNA-mediated gene regulatory mechanism is not a critical factor that determines final transcript levels of most genes , but highly expressed sRNAs with antisense orientations participate in the negative transcriptional gene regulation of corresponding genes at a genome-wide level . Although we found that F . graminearum possesses a global negative transcriptional regulatory mechanism involving antisense sRNAs during sexual development , only 57 genes produced antisense sRNAs among 200 upregulated DEGs ( S2 and S7 Tables ) . Instead , a defective phenotype in ascospore production of RNAi-deficient mutants appears to be derived by the downregulation of sexual reproduction-related genes , such as PPG1 ( Gene ID: FGSG_05061 ) , transcription factors ( Gene ID: FGSG_01366 and FGSG_16753 ) , and transaldolase ( Gene ID: FGSG_13162 ) , as noted above [9 , 10 , 42] . Thus , genes with substantial changes in transcript levels were not directly affected by RNAi-mediated gene regulation . Thirty-two transcription factor genes are possibly modulated by antisense sRNAs , and three of them ( FGSG_01022 , FGSG_09654 , and MAT1-2-1 ) are closely related to perithecia development in F . graminearum [10] . In particular , RNAi participates in the regulation of one of the MATs , MAT1-2-1 , during sexual development . In the case of kinases , another important mediator for signal transduction pathways , four genes ( FGSG_00469 , FGSG_16383 , FGSG_05418 , and FGSG_16493 ) produced antisense sRNAs , and one of them ( FGSG_05418 ) was shown to be important for ascospore formation in F . graminearum [11] . Although most of the mentioned genes involved in sexual development are known to be related to the perithecium development or the early stages of sexual development , our additional transcript analysis during sexual development demonstrated that they also have important functions during the late stages of sexual development ( S5 Fig ) . In particular , most of tested genes were dynamically changed in expression approximately 5 days after sexual induction , and these patterns of several genes were altered in RNAi-deficient mutants compared to those of the wild type . Thus , the RNAi mechanism fine-tunes the transcriptome to modulate various molecular processes , including signal transduction networks , during ascospore production in F . graminearum . We visualized patterns of sequence alignments at the ex-siRNA-producing genes using Integrative Genomics Viewer ( IGV ) [51] . The IGV image of the gene ( FGSG_11711 ) showed that ex-siRNAs are produced over the entire ranges of the gene ( S6 Fig ) , demonstrating that total amount of siRNAs of the gene does not always reflect abundances of specific ex-siRNAs . Therefore , we identified ex-siRNAs with over 500 raw counts ( S9 Table ) . Since repeated trials of the standard Northern blot analysis of six highly expressed ex-siRNAs ( S7 Fig ) were unsuccessful in detecting signals , we quantified the absolute amount of the ex-siRNA with the highest expression ( siRNA of FGSG_09213 ) using stem-loop RT-PCR ( S8 Fig ) . Whereas Northern blot analysis of our system could detect 1–10 fmoles synthetic siRNA of FGSG_09213 , approximately 0 . 05 fmoles of siRNAs were detected in 100 ng of small RNA-enriched RNA samples from the F . graminearum wild type . Thereafter , we analyzed siRNAs using stem-loop RT-PCR ( Fig 6 ) . Abundances of siRNAs in the wild-type and the other mutant strains determined by RT-PCR were mostly consistent with our sRNA-seq results ( S7 Fig and Fig 7A ) . We also verified that transcript levels of FGSG_10502 and FGSG_03222 were negatively correlated with corresponding ex-siRNA production ( Fig 7B ) . We analyzed the degradome of the F . graminearum wild-type strain to elucidate the mechanism for genome-wide sRNA-mediated gene silencing . Total RNAs were isolated from the wild-type strain 5 days after sexual induction , and RNA fragments with a 5′ monophosphate and polyA tail , which were cleaved products of transcripts , were subjected to the degradome library construction and deep sequencing . In our degradome sequencing method , we expected 16 nt reads after trimming the adapter sequences , and enriched reads of 16 nt and 17 nt were further used for alignment ( S10 Table ) . Approximately 95% of the trimmed sequences from two independent degradome libraries perfectly matched one or more positions in the F . graminearum genome ( S11 Table ) . To monitor transcript degradation , we used degradome tags with a sense direction for further analyses . The libraries of degradome tags were markedly biased toward the 3′ ends of protein coding sequences ( CDS ) and 3′-UTR regions ( Fig 8A ) . Specifically , most degradome tags were enriched in the exon and 3′-UTR regions ( Fig 8B and S11 Table ) . The average of the total abundances of degradome tags for the CDS and 3′-UTR regions was calculated as reads per kilobase of sequence per two billion mapped sequence reads with a cutoff value of 10 ( S12 Table ) . Comparison of the degradome abundance of the wild-type strain to the transcript values of the wild-type and RNAi component mutant strains similarly showed a distinct positive correlation ( Fig 8C ) . Accordingly , genes with higher degradome counts tended to accumulate more transcripts in Fgdicer1 Fgdicer2 and Fgago1 Fgago2 than those in the wild type ( Fig 8D ) . Taken together , these findings revealed that degradome tag abundance is primarily determined by the level of the corresponding gene transcript in F . graminearum . The only known mechanism for siRNA-mediated gene silencing in filamentous fungi is siRNA-guided mRNA degradation [34] . As siRNA-guided endonucleolytic cleavage events should lead to the rapid decay of transcripts [52] , we investigated the correlation between the abundance of degradome tags and antisense sRNA-mediated gene regulation . As genes produced more sRNAs with an antisense direction , the absolute number of degraded transcripts increased in the F . graminearum wild-type strain ( Fig 8E ) . As the degradome generally reflects transcript levels , the abundances of degradome tags were normalized with transcript counts , and total genes were listed in the order of their given degradome tags:transcript counts ratios . When a higher portion of transcripts was degraded in the wild-type strain ( higher percentile rank classes ) , corresponding genes tended to be upregulated in Fgdicer1 Fgdicer2 mutants ( Fig 8F ) , suggesting that Dicer-dependent gene regulation globally triggers transcript degradation in this fungus . When sRNA-producing genes were evenly distributed among the percentile rank classes , markedly reduced and increased portions of genes produced sRNAs in the down- ( Fgdicer1 Fgdicer2/Z-3639 < 1 ) and upregulated ( Fgdicer1 Fgdicer2/Z-3639 > 1 ) gene groups , respectively ( Fig 8G ) . More than 70% of the genes were concentrated in the top 20% of the percentile rank class in the genes that were upregulated more than 2-fold in Fgdicer1 Fgdicer2 compared to that of the wild type ( Fgdicer1 Fgdicer2/Z-3639 > 2; Fig 8G ) . Taken together , these results show that RNAi-mediated negative gene regulation occurs post-transcriptionally by degrading corresponding gene transcripts during sexual reproduction in F . graminearum . Fungi have evolved several unique molecular mechanisms that are specifically activated during sexual reproduction processes [53] . Repeat-induced point mutation ( RIP ) and MSUD are well-known genome defense systems induced during meiotic cell division in some ascomycete fungi . In the heterothallic fungus N . crassa , RIP effectively detects and mutates repetitive transposable elements before meiotic prophase , resulting in the generation of nonfunctional transposons [54] . While RIP permanently mutates preexisting transposons , MSUD recognizes and suppresses the expression of repetitive sequences and therefore inhibits the mobilization of transposons during meiotic cell division [18] . Sex-induced silencing by RNAi was reported in the basidiomycete C . neoformans , which silences transgenes and transposons during sexual reproduction to protect the genome [31] . Recently , noncanonical adenosine-to-inosine RNA editing that enhances the diversity of gene products at the post-transcriptional level was shown to specifically occur during perithecium development in approximately half of the expressed genes in F . graminearum [55] . MSUD is also functional in F . graminearum despite its homothallic nature , but its activity was found to be much lower than that of N . crassa [7] . As MSUD is induced by RNAi-mediated gene silencing [26] , F . graminearum was predicted to possess an effective RNAi , and a recent study reported that one of the Dicers ( FgDicer2 ) and one of the Argonautes ( FgAgo1 ) were involved in hairpin-induced gene silencing during vegetative growth [37] . In the present study , we discovered that F . graminearum has a sexual specifically induced RNAi pathway that is important for ascospore formation . RNAi components , which are dispensable for hairpin-induced gene silencing during mycelial growth , FgDicer1 and FgAgo2 primarily participate in the biogenesis of sRNAs with an antisense direction or 5′-U during sexual reproduction . Most sRNAs originated from transcript regions and globally affected expression of the corresponding genes at a post-transcriptional level by degrading corresponding transcripts . Several types of fungal siRNAs have been characterized as functioning in genome defenses against transposable elements [56] or viruses [57] or for endogenous gene regulation to respond to heterochromatin formation [58] or DNA damage [59] . In particular , ex-siRNAs were proposed to regulate the target gene expression involved in fungal developmental processes in M . circinelloides and T . atroviride [30 , 50] , while no correlation was found between ex-siRNAs and corresponding gene expression in M . oryzae [28] . In F . graminearum , RNAi components involved in ex-siRNA-mediated RNAi ( FgAGO2 and FgDICER1 ) were specifically functional during the late stages of sexual development . Similarly , large amounts of ex-siRNAs ( approximately 30% of total sRNAs ) were produced during sexual reproduction in F . graminearum , whereas only 2% , 4% , and 5% of sRNAs were produced from protein-coding regions in T . atroviride [30] , M . oryzae [60] , and F . oxysporum [61] , respectively . RNAi-deficient mutants of M . circinelloides and T . atroviride were defective in asexual development , but those of F . graminearum and M . oryzae were not [37 , 60] . Instead , F . graminearum has evolved a sex-specifically induced ex-siRNA-mediated RNAi for genome-wide post-transcriptional gene regulation , which is important for ascosporogenesis . More than half of the genes with decreasing or increasing tendencies in expression during ascospore formation in the F . graminearum wild-type strain were positively or negatively regulated in RNAi-deficient mutants , respectively ( Fig 3A ) . In particular , 50% of the upregulated genes in the RNAi-deficient mutants showed similar expression patterns to those of MAT genes during perithecia development , demonstrating that RNAi is one of the key pathways under the control of MAT genes , as previously suggested [9] . However , ex-siRNA-mediated RNAi is not a critical process that determines absolute transcript amounts of most genes compared to other transcriptional or post-transcriptional regulatory mechanisms of this fungus . Not all ex-siRNA-producing genes were significantly downregulated in RNAi-deficient mutants , and sex-specific RNAi appears to be used for minute negative post-transcriptional gene regulation . Indeed , sex-specific RNAi orchestrates global gene regulation , which may alter signal transduction networks or carbon metabolism involved in sexual development in F . graminearum ( Fig 3B ) . In addition to a regulatory role for ascosporogenesis , sex-specific ex-siRNAs may not be closely involved in the genome defense mechanism . Few gene duplications and transposons have been found in the F . graminearum genome , although large amounts of siRNAs are produced during sexual reproduction [4] . In addition , ex-siRNAs of F . graminearum are different from sex-specific MSUD-associated siRNAs ( masiRNAs ) . MSUD detects unpaired DNA regions during meiosis and leads to siRNA production for gene silencing in the heterothallic fungus N . crassa [26 , 62] . In a recent study , naturally unpaired regions between two mating partners were predicted to be major sources for ex-siRNAs in the heterothallic fungus N . crassa [63] . However , F . graminearum is homothallic , and therefore , there should be no unpaired DNA regions during meiotic cell division . Two Dicers and two Argonautes have redundant and separate roles for RNAi in F . graminearum . Passenger strand degradation is generally required for RNAi function , and guide strands bound by Argonaute are protected by degradation [64 , 65] . Therefore , the lack of accumulation of sRNAs in Fgago mutants indicated that these sRNAs are functional for RNAi so that in its absence , the siRNAs would be rapidly degraded . Double-deletion mutants of both Dicers and Argonautes similarly abolished sRNA production and altered transcriptomes , suggesting that both of them have redundant functions ( Figs 2 and 4 ) . However , FgDicer2 has a more important role in global sRNA production than that of FgDicer1; FgDicer1 is mostly dispensable for hairpin-mediated sRNA production [37] . Intriguingly , FgAgo2 has a major function in global sRNA production in F . graminearum , whereas FgAgo1 is specifically involved in hairpin-induced sRNA production . Due to the redundant and separate roles of Argonautes , deletion of FgAGO2 may lead to excessive binding of guide strand siRNA to FgAgo1 , resulting in unexpected transcriptional regulation . For this reason , fewer genes were upregulated in the Fgago1 Fgago2 double mutant than those in the Fgago2 single mutant ( Fig 2A ) . Similarly , Ago1 and Ago2 function in a partially redundant manner but generally have roles in miRNA function and siRNA-triggered RNAi , respectively , in Drosophila melanogaster [66] . The specificities of RNAi components may be derived by characteristics of a fungal RNAi system . Sad-1 ( RdRp ) , Sms-3 ( Dicer ) , Sms-2 ( Argonaute ) , and other RNAi components form a silencing complex so that these RNAi components together govern a specific RNAi pathway involved in MSUD in N . crassa [67] . If this is the case in F . graminearum , at least two types of silencing complexes with different specificities may be involved in siRNA- or hairpin-mediated gene silencing; FgAgo1 and FgDicer2 would be components of the same silencing complex for hairpin-mediated RNAi in F . graminearum [37] . The target specificities of Argonautes may also be a reason for the distinct functions of two Argonautes . In A . thaliana , AtAgo2 and AtAgo4 preferentially bind to sRNAs with a 5′ terminal adenosine , while AtAgo1 predominantly recruits miRNAs with 5′-U among ten Ago proteins [68] . In the present study , using combined analysis of functional genetics and deep sequencing of sRNAs and the transcriptome and degradome , we demonstrated that the sex-specifically induced ex-siRNA-mediated RNAi mechanism fine-tunes the transcriptome for ascospore formation in F . graminearum . Ex-siRNA functions are important for various developmental stages and stress responses in the basal fungus M . circinelloides , but F . graminearum has evolved to utilize ex-siRNAs for a specific developmental stage . Therefore , ex-siRNA-mediated RNAi might be involved in various fungal developmental stages and stress responses depending on the fungal species and should be highlighted as an important post-transcriptional regulatory mechanism in fungi . The wild-type strain Z-3639 and transgenic strains derived from this strain were used in this study ( S1 Table ) . Fungal strains were stored as conidia and mycelia in 30% glycerol solution at -80°C . All of the media used in this study were prepared as described in the Fusarium laboratory manual [1] . Genomic DNA was extracted from lyophilized mycelia according to the Fusarium laboratory manual [1] . Restriction endonuclease digestion , agarose gel electrophoresis , Southern and Northern blotting , and hybridization with 32P-labeled probes were performed following standard protocols [69] . PCR and qRT-PCR primers used in this study were synthesized by an oligonucleotide synthesis facility ( Bionics , Seoul , Korea ) ( S13 Table ) . Stem-loop RT-PCR analyses of siRNAs were performed as previously described [70 , 71] . For detection of small RNAs , RNAs that are highly enriched for small RNA species ( less than 200 nt ) were isolated using a mirVanaTM miRNA isolation kit ( Invitrogen ) , separated in a 15% urea-polyacrylamide gel , and transferred to Immobilon-Ny + membranes ( Millipore , Billerica , MA , USA ) . The membranes were probed with 32P-labeled oligonucleotides in PerfectHybTM Plus Hybridization Buffer ( Sigma-Aldrich , St . Louis , MO , USA ) at 37°C . Hybridization and washing steps were performed as previously described [72] . Total RNA was isolated from mycelia that were ground in liquid nitrogen using an Easy-Spin total RNA extraction kit ( iNtRON Biotech , Seongnam , Korea ) , and each first-strand cDNA was synthesized using SuperScript III reverse transcriptase ( Invitrogen , Carlsbad , CA , USA ) . qRT-PCR was performed with SYBR Green Supermix ( Bio-Rad , Hercules , CA , USA ) and a 7500 real-time PCR system ( Applied Biosystems , Foster City , CA , USA ) with corresponding primer pairs ( S13 Table ) . The ubiquitin C-terminal hydrolase gene UBH ( Gene ID: FGSG_01231 ) was used as a reference gene . We compared the cycle threshold ( 2-ΔΔCT ) to measure the transcript levels of target genes in different conditions . PCR was performed three times with three replicates per run . The double-joint ( DJ ) PCR strategy was applied to construct fusion PCR products for targeted gene deletion and complementation [73] . For FgDICER1 deletion , the 5′ and 3′ flanking regions of FgDICER1 were amplified from the genomic DNA of the F . graminearum wild-type strain ( S1 Fig ) . A geneticin resistance gene cassette ( GEN ) was amplified from pII99 . Three amplicons ( 5′ flanking region , GEN , and 3′ flanking region ) were mixed at a 1:3:1 molar ratio and fused by a second round of DJ PCR . Finally , the fusion constructs were amplified with the nested primers using the second-round PCR product as a template . For complementation , the 5′ flanking region that included the FgDICER1 open-reading frame with its own promoter and 3′ flanking region were amplified from genomic DNA of the wild-type strain . The HYG construct was amplified from pBCATPH . Three amplicons were then fused in a second round of DJ PCR . Finally , the fusion constructs for transformation were amplified with the nested primers using the second-round PCR product as a template . Fungal transformation was performed as previously described [44] . The same strategy was used for deletion and complementation of other RNAi component genes ( S2 Fig ) . For self-crosses , mycelia were grown on carrot agar for 5 days and then removed with the back of a surgical blade ( surgical blade #11; Feather Safety Razor , Osaka , Japan ) while applying 2 . 5% sterilized Tween 60 solution [1] . For outcrosses , the heterothallic Δmat2 strain was fertilized with 1 ml of a conidial suspension ( 105 conidia/ml ) from fertilizing parents . All of the sexually induced cultures were incubated under near-UV light ( wavelength: 365 nm; HKiv Import & Export Co . , Ltd . , Xiamen , China ) at 25°C . Transcriptome analysis in F . graminearum was performed as previously described [74] . Total RNA was isolated from each fungal culture at 5 days after sexual induction on carrot agar using an Easy-Spin total RNA extraction kit ( iNtRON Biotech ) . More than five biological replicates of each strain were pooled for RNA-seq library construction . RNA-seq libraries were constructed using the Illumina TruSeqTM RNA sample prep kit with no modifications to the standard low-throughput protocol . Samples were run on an Illumina HiSeq2000 instrument using the reagents provided in the Illumina TruSeq paired-end ( PE ) cluster kit V3-cBot-HS and the TruSeq SBS kit v3-HS ( 200 cycles ) . Similarly , a small RNA library was prepared using a TruSeq small RNA library prep kit following the manufacturer’s instruction . Then , each library was subjected to single-end 100 bp sequencing using a HiSeq2000 instrument . Degradome-seq was performed as previously described with some modifications [75 , 76] . Poly ( A ) + RNA was isolated using the NucleoTraP® mRNA purification kits ( Machery-Nagel , Düren , Germany ) according to the manufacturer’s instructions . A 5′ RNA adapter ( 5′-GUUCAGAGUUCUACAGUCCGACGAUC-3′ ) was ligated to the cleavage products , which contain a 5′ monophosphate . The ligated products were reverse-transcribed into cDNA using an oligo ( dT ) primer ( 5′-CGAGCACAGAATTAATACGACTTTTTTTTTTTTTTTTTT-3′ ) by SuperScript III reverse transcriptase ( Invitrogen ) and amplified by PCR with a pair of cDNA primers ( 5´-GTTCAGAGTTCTACAGTCCGA-3′ and 5´-CGAGCACAGAATTAATACGACT-3′ ) . The resulting product was digested with a MmeI ( NEB , MA , USA ) to obtain short fragments from the 5´ end of double-stranded cDNA . The digested products were ligated with an annealed duplex DNA adapter ( top , 5´-p- TGGAATTCTCGGGTGCCAAGG- 3′ and bottom , 5´-CCTTGGCACCCGAGAATTCCANN-3′ ) using T4 DNA ligase ( NEB ) . The ligated DNA products ( ~62 bp ) were isolated using a 12% polyacrylamide gel ( PAGE ) , and the purified products were amplified by PCR with a set of indexed TruSeq . The final PCR products were purified by running a 6% PAGE gel based on size ( ~128 bp ) . Most enzymes and primer sequences used in this study were obtained from a TruSeq small RNA library prep kit . Libraries were used for throughput sequencing on a HiSeq2000 platform . Alignments were performed with BWA [77] using the F . graminearum genome [49] , and the htseq-count script of the HTSeq package was used to compute the counts per gene [78] . Genome-wide transcript levels were quantified in reads per kilobase of exon per hundred million mapped sequence reads ( RPKHM ) [79] . Genes with maximum RPKHM values below 10 were deleted from the analysis . DEGs were identified based on fold-change values . High-quality small RNA reads were obtained from raw reads by filtering out poor quality reads and removing adaptor sequences using the FASTX toolkit [80] . Adaptor-trimmed unique sequences were aligned to the F . graminearum genome using bowtie [49 , 81] , and structural RNAs , such as tRNA , rRNA , snRNA , and snoRNA , were identified . The perfectly matched reads between 18–32 nt ( for sRNAs ) in length were selected . The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [82] and are accessible through the accession numbers GSE87724 for mRNA-seq and GSE87835 for sRNA-seq . Raw data of degradome-seq have been deposited in NCBI's Sequence Read Archive ( BioProject PRJNA348145 ) .	Control of gene expression by small noncoding RNA ( sRNA ) has recently been highlighted as a significant post-transcriptional regulatory mechanism . To date , researchers have predominantly focused on the identification of microRNA-like RNAs ( milRNAs ) in fungi because microRNAs ( miRNAs ) are key regulators in animals and plants . In this study , we discovered that the sex-induced RNA interference ( RNAi ) mechanism had important roles in sexual reproduction in the head blight fungus Fusarium graminearum . In the late stages of sexual reproduction , small interference RNAs that were produced from gene regions ( ex-siRNAs ) were involved in post-transcriptional gene regulation at a genome-wide level . Based on our results , we concluded that F . graminearum specifically utilizes ex-siRNA-mediated RNAi for sexual development but not for other biological processes . This is the first genome-wide characterization of the sRNAs involved in fungal development .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "rna", "interference", "fungal", "genetics", "gene", "regulation", "developmental", "biology", "fungi", "genome", "analysis", "epigenetics", "morphogenesis", "small", "interfering", "rnas", "mycology", "genetic", "interference", "gene", "expression", "biochemistry", "rna", "fungal", "genomics", "nucleic", "acids", "sexual", "differentiation", "genetics", "transcriptome", "analysis", "biology", "and", "life", "sciences", "genomics", "non-coding", "rna", "computational", "biology", "organisms" ]	2017	Genome-wide exonic small interference RNA-mediated gene silencing regulates sexual reproduction in the homothallic fungus Fusarium graminearum
Animal aging is characterized by progressive , degenerative changes in many organ systems . Because age-related degeneration is a major contributor to disability and death in humans , treatments that delay age-related degeneration are desirable . However , no drugs that delay normal human aging are currently available . To identify drugs that delay age-related degeneration , we used the powerful Caenorhabdtitis elegans model system to screen for FDA-approved drugs that can extend the adult lifespan of worms . Here we show that captopril extended mean lifespan . Captopril is an angiotensin-converting enzyme ( ACE ) inhibitor used to treat high blood pressure in humans . To explore the mechanism of captopril , we analyzed the acn-1 gene that encodes the C . elegans homolog of ACE . Reducing the activity of acn-1 extended the mean life span . Furthermore , reducing the activity of acn-1 delayed age-related degenerative changes and increased stress resistance , indicating that acn-1 influences aging . Captopril could not further extend the lifespan of animals with reduced acn-1 , suggesting they function in the same pathway; we propose that captopril inhibits acn-1 to extend lifespan . To define the relationship with previously characterized longevity pathways , we analyzed mutant animals . The lifespan extension caused by reducing the activity of acn-1 was additive with caloric restriction and mitochondrial insufficiency , and did not require sir-2 . 1 , hsf-1 or rict-1 , suggesting that acn-1 functions by a distinct mechanism . The interactions with the insulin/IGF-1 pathway were complex , since the lifespan extensions caused by captopril and reducing acn-1 activity were additive with daf-2 and age-1 but required daf-16 . Captopril treatment and reducing acn-1 activity caused similar effects in a wide range of genetic backgrounds , consistent with the model that they act by the same mechanism . These results identify a new drug and a new gene that can extend the lifespan of worms and suggest new therapeutic strategies for addressing age-related degenerative changes . Animal aging is characterized by progressive , degenerative changes of tissue structure and function . In humans , these changes have profound negative effects on health by causing morbidity and mortality . An important goal of aging research is to identify interventions that can delay age-related degeneration and promote an extended period of vitality or healthspan . However , no interventions have been demonstrated to delay human aging . By contrast , a growing number of interventions have been demonstrated to delay age-related degeneration and extend lifespan in model animals such as worms , flies and mice [1] . These interventions include dietary changes such as caloric restriction , genetic changes such as reducing the activity of the insulin/insulin-like growth factor-1 ( IGF-1 ) signaling pathway , and drugs such as rapamycin . These studies indicate that pathways that influence aging have been conserved during animal evolution [1] . Thus , model organisms are promising systems to identify and characterize interventions that promote healthy aging and may be beneficial in humans . The terrestrial nematode Caenorhabditis elegans has emerged as an outstanding model organism for studies of aging . The biology of these animals is well suited for studies of aging because they have a rapid life cycle and a relatively short adult lifespan of about 15 days [2 , 3] . A wide variety of age-related degenerative changes have been documented , providing assays of aging and suggesting C . elegans undergoes mechanisms of aging similar to larger animals where progressive degenerative changes are well characterized [4] . Powerful experimental techniques are well established , including forward and reverse genetic approaches and molecular approaches facilitated by a fully sequenced genome [5 , 6] . C . elegans are well suited for pharmacological studies because they ingest compounds that are added to the culture medium . Molecular genetic studies have identified and characterized several pathways that substantially influence the rate of age-related degeneration . The insulin/IGF-1 pathway was first implicated in aging biology in C . elegans and has now been shown to play a conserved role in other animals , including flies and mammals [1] . Mutations that reduce the activity of the daf-2 insulin receptor or the age-1 phosphatidylinositol-3-OH ( PI3 ) kinase substantially extend the adult lifespan , indicating that insulin/IGF-1 pathway activity promotes a rapid lifespan [7 , 8]; these mutant animals also display enhanced resistance to a variety of stresses such as UV light , oxidation , transition metals , and hypoxia [9–12] . A critical effector of the daf-2/age-1 pathway is the forkhead transcription factor DAF-16 , which is activated and localized to the nuclei by low levels of daf-2 signaling [13 , 14] . The activity of daf-16 promotes an extended lifespan , and daf-16 is necessary for the extension of lifespan caused by mutations of daf-2 and age-1 [8 , 15] . Caloric restriction extends the lifespan of a wide range of organisms , including C . elegans , indicating that ad libitum feeding promotes a rapid lifespan . Mutations of genes that are necessary for pharyngeal pumping and food ingestion , such as eat-2 , cause a substantial lifespan extension [16] . Mutations in multiple genes that are necessary for mitochondrial function , such as isp-1 , cause a lifespan extension , indicating that wild-type levels of mitochondrial activity promote a rapid lifespan [17 , 18] . In addition to genetic approaches , C . elegans is emerging as a valuable system for pharmacological approaches that can be used to identify and characterize drugs that influence aging . Compounds that influence C . elegans aging have been identified by screening approaches and by testing candidate drugs based on a known mechanism of action [19–25] . To identify drugs that influence aging , we screened FDA-approved drugs for the ability to extend the lifespan of C . elegans hermaphrodites . Here we report that captopril , an angiotensin converting enzyme ( ACE ) inhibitor used to treat high blood pressure , extended mean lifespan . ACE is a protease that functions in a signaling cascade that is initiated by low blood pressure; in humans , ACE converts angiotensin I to angiotensin II , and angiotensin II binds the AT1 receptor , resulting in increased contractility of endothelial cells and thereby increasing blood pressure [26 , 27] . ACE inhibitors such as captopril are used by a large number of people to control hypertension [27] . The ACE gene has been conserved from bacteria to mammals , indicating it had a primordial function before the evolution of a closed circulatory system that creates blood pressure . The C . elegans homolog of ACE is encoded by the acn-1 gene; acn-1 is necessary for larval molting but has not been previously implicated in adult longevity [28] . We hypothesized that captopril inhibits acn-1 to extend lifespan , and here we present experimental evidence that supports this model . First , inhibition of the acn-1 gene by RNA interference extended lifespan and delayed age-related degenerative changes , indicating that acn-1 activity influences aging and longevity . Second , captopril treatment and reducing the activity of acn-1 caused very similar effects in a wide range of genetic backgrounds , indicating that these interventions have a common mechanism . Third , the lifespan extensions caused by captopril treatment and reducing the activity of acn-1 were not additive , indicating that these interventions may affect the same pathway . These results identify captopril as a new , FDA-approved drug that can extend the lifespan of C . elegans and acn-1 as a new gene that influences C . elegans aging . The findings establish acn-1 as the target of captopril in worms , connecting a pharmacological intervention that extends lifespan to its direct molecular target . In mammals , ACE regulates blood pressure , indicating there is a link between a system that controls aging in worms and physiology in mammals . To identify drugs that influence aging , we selected 15 compounds that are Food and Drug Administration ( FDA ) -approved for human use , have known effects on human physiology , and represent different functional or structural classes ( see Materials and Methods ) . Compounds were added to NGM agar at three different concentrations , and the lifespan of C . elegans hermaphrodites cultured at 20°C with E . coli OP50 as a food source was determined . We previously described a similar screening approach that was used to identify the lifespan extending compounds ethosuximide and valproic acid [19 , 20] . Captopril , an ACE inhibitor , caused a significant extension of lifespan ( Fig 1A and 1C ) . To identify the optimal concentration for lifespan extension , we performed a dose–response analysis . A concentration of 2 . 5mM captopril in the medium caused the greatest lifespan extension , whereas concentrations of 1 . 9mM and 3 . 2mM caused smaller extensions ( Fig 1B; Table 1 , line 1–4 ) . At the optimal concentration of 2 . 5mM , captopril treatment caused a significant 23% extension of mean adult lifespan and a significant 18% extension of maximum adult lifespan ( Fig 1C; Table 1 , line 5–6 ) . We define maximum adult lifespan as the average lifespan of the 10% of the population that are longest lived . To determine the developmental stage when captopril functions to extend lifespan , we administered the drug beginning at the L4 larval stage . The drug was effective with this time of administration suggesting captopril functions in adults to delay age-related degeneration . To determine the temperature dependence of captopril , we analyzed animals cultured at 15°C , 20°C and 25°C . Captopril significantly extended the mean and maximum adult lifespan at all three temperatures , indicating that the effect is not temperature dependent ( Fig 1D; Table 1 , line 7–10; S1A Fig ) . These experiments were conducted with live E . coli as a food source , raising the possibility that captopril may directly affect bacteria and indirectly affect worms . There are precedents for such a mechanism , since antibiotics can extend C . elegans lifespan by reducing the pathogenicity of bacteria [29–31] , and the anti diabetic drug metformin was reported to extend C . elegans lifespan by altering bacterial folate and methionine metabolism [32] . To determine if the effects of captopril are mediated by an effect on live bacteria , we conducted the life span experiment with bacteria killed by exposure to ultraviolet light . Captopril extended the lifespan of C . elegans in these conditions , demonstrating that the mechanism of captopril-mediated lifespan extension is not dependent on live E . coli ( S1C Fig; Table 1 , line 11–12 ) . A large number of age-related degenerative changes have been characterized in C . elegans , including declines of physiological processes , such as body movement , pharyngeal pumping , and egg-laying , and changes in morphology , such as loss of tissue integrity [4 , 33 , 34] . Treatment with captopril caused a small delay in the age-related decline in pharyngeal pumping rate , although the change was not statistically significant with the sample size analyzed ( S2 Fig ) . Several genetic manipulations that extend adult lifespan also affect reproduction . For example , caloric restriction and defects in insulin/IGF-1 signaling reduce total progeny production and increase reproductive span in self-fertile hermaphrodites [35] . To determine how captopril affects reproduction , we monitored progeny production of self-fertile hermaphrodites daily . Captopril did not significantly affect total brood size or reproductive span of self-fertile hermaphrodites ( S3A and S3B Fig ) . Captopril treatment in humans reduces blood pressure by inhibiting the activity of angiotensin converting enzyme ( ACE ) [36] . Therefore , we hypothesized that captopril treatment in C . elegans extends longevity by inhibiting the worm homolog of ACE . To investigate this hypothesis , we analyzed the acn-1 gene because it encodes a predicted protein that is most similar to human ACE [28] . To reduce the activity of acn-1 , we used RNA interference ( RNAi ) [37]; worms were fed bacteria expressing dsRNA from the acn-1 gene , which is predicted to reduce the levels of the acn-1 transcript . Wild-type animals cultured with acn-1 RNAi beginning at the embryonic stage displayed a significant extension of mean and maximum lifespan of 21% and 18% , respectively ( Fig 2A , Table 2 , line 1–2 ) . These results indicate that acn-1 activity is necessary to promote a rapid lifespan . To investigate the time of action of acn-1 , we initiated the exposure to acn-1 RNAi at the L4 larval stage . Exposure only during adulthood caused a similar extension of mean and maximum lifespan of 22% and 20% , respectively , indicating that acn-1 functions in adults to promote a rapid lifespan ( Fig 2B , Table 2 , line 3–4 ) . Several mutations have been identified that increase the sensitivity of worms to feeding RNAi , including mutation of rrf-3 [38] . Feeding acn-1 RNAi bacteria to rrf-3 mutant animals beginning at the embryonic stage caused a significant increase of mean and maximum lifespan of 33% and 24% , respectively ( Fig 2C , Table 2 , line 5–6 ) . Similarly , feeding acn-1 RNAi beginning at the L4 stage caused a significant extension of mean and maximum lifespan of 46% and 33% , respectively ( Fig 2D , Table 2 , line 7–8 ) . The extensions caused by acn-1 RNAi in the rrf-3 background were greater than the extensions in the wild-type background , indicating that rrf-3 mutant animals are indeed more susceptible to the effect of the RNAi treatment . Moreover , acn-1 RNAi also caused a significant extension of mean and maximum lifespan of rrf-3 mutant animals at 25°C ( S1B Fig , Table 2 , line 9–10 ) . To quantify how acn-1 mRNA levels are affected by feeding RNAi , we performed quantitative RT-PCR . acn-1 RNAi reduced mRNA levels about 50% compared to control RNAi in rrf-3 mutant animals ( S4 Fig ) . To characterize how acn-1 influences age-related degeneration , we monitored age-related declines of major physiological processes . Wild-type C . elegans hermaphrodites display coordinated , sinusoidal body movement as young adults , and the frequency and coordination of body movement display age-related declines . To analyze body movement quantitatively , we counted body bends on solid NGM using a dissecting microscope . Hermaphrodites cultured with acn-1 RNAi displayed a significantly higher rate of body movement beginning on day 4 of adulthood and extending to day 26 of adulthood ( Fig 3A ) . To illustrate this difference , we exploited the fact that worms leave tracks in the bacterial lawn as they move . Five animals on day 15 of adulthood were transferred to fresh bacterial lawns , allowed to move for two hours , and the lawns were photographed . Fig 3B shows that hermaphrodites treated with control RNAi left a small number of tracks , and the tracks are suggestive of uncoordinated movement . By contrast , hermaphrodites treated with acn-1 RNAi left abundant tracks that were suggestive of coordinated sinusoidal movement . We monitored the age-related decline in pharyngeal pumping rate quantitatively by direct observation using a dissecting microscope . Hermaphrodites treated with acn-1 RNAi displayed higher rates of pharyngeal pumping on days 12–20 of adulthood ( Fig 3C ) . These results demonstrate that acn-1 is necessary to promote the rapid , age-related decline of body movement and pharyngeal pumping observed in wild-type animals . To analyze the effect on reproduction , we monitored progeny production of self-fertile hermaphrodites . Wild-type animals treated with acn-1 RNAi did not display significant changes in total self fertile brood size or the daily production of progeny ( S2C and S2D Fig ) . Captopril treatment only slightly delayed age-related changes of pharyngeal pumping , whereas acn-1 RNAi significantly delayed age-related changes of pharyngeal pumping and body movement , suggesting acn-1 RNAi may reduce the activity of acn-1 to a greater extent than captopril or the drug may have toxic effects . Several C . elegans mutations that extend longevity also increase stress resistance [39 , 40] . To investigate the function of acn-1 in stress resistance , we analyzed heat and oxidative stress . Embryos were cultured at 20°C with control RNAi or acn-1 RNAi , and after 3 days animals were transferred to stressful conditions and monitored for survival . When exposed to continuous 34°C heat stress , control animals displayed a time dependent decrease in survival with a mean lifespan of 14 . 0 hours; animals treated with acn-1 RNAi displayed a significant , 12% extension of mean lifespan of 15 . 7 hours ( Fig 4A , Table 3 , line 1–2 ) . When exposed to oxidative stress caused by 40mM paraquat , control animals displayed a time dependent decrease in survival with a mean lifespan of 47 . 8 hours; animals treated with acn-1 RNAi displayed a significant , 16% extension of mean lifespan of 55 . 4 hours ( Fig 4B , Table 3 , line 3–4 ) . In addition , we observed a similar result of extended survival in oxidative stress when wild-type animals were treated with acn-1 RNAi ( Fig 4C Table 3 , line 5–6 ) . To determine if the specific conditions or oxidation generating chemical are important for the results , we analyzed oxidative stress in liquid medium using the compound juglone to cause oxidative stress . After nine hours of juglone exposure , animals treated with acn-1 RNAi displayed a significant , 56% increase in survival compared to control animals ( Fig 4D Table 3 , line 7–8 ) . These results indicate that the acn-1 gene is necessary to promote wild-type levels of sensitivity to multiple stresses including heat and oxidation . To investigate the mechanism of action of captopril and acn-1 in lifespan extension , we analyzed how captopril treatment and acn-1 RNAi affects animals with mutations that alter longevity . Caloric restriction extends the lifespan of many organisms , indicating that ad libitum feeding during laboratory culture reduces longevity . Mutations of the eat-2 gene impair pharyngeal pumping , reduce food intake and cause a lifespan extension [16 , 41] . Captopril significantly extended the mean and maximum lifespan of eat-2 ( ad1116 ) mutant animals by 14% and 17% , respectively ( Fig 5A; Table 1 , line 13–14 ) . Similarly , acn-1 RNAi significantly extended mean and maximum lifespan by 12% and 10% , respectively ( Fig 6A , Table 2 , line 11–12 ) . Thus , the lifespan extension caused by caloric restriction was additive with captopril treatment and acn-1 RNAi . Mutations of several genes that are important for mitochondrial function cause a lifespan extension in C . elegans , indicating that normal mitochondrial function promotes a rapid lifespan . The isp-1 gene encodes a iron sulfur cluster containing protein that is important for the function of complex III to catalyze electron transport from ubiquinol to cytochrome c , and isp-1 mutations extend lifespan [17 , 42] . Captopril treatment significantly extended the mean and maximum lifespan of isp-1 ( qm150 ) mutant animals by 23% and 20% , respectively ( Fig 5B; Table 1 , line 15–16 ) . Similarly , acn-1 RNAi significantly extended mean and maximum lifespan by 14% and 13% , respectively ( Fig 6B , Table 2 , line 13–14 ) . Thus , the lifespan extension caused by reducing mitochondrial function was additive with captopril treatment and acn-1 RNAi . Overexpression of SIR2 ( silent information regulator 2 ) is reported to extend the lifespan of several organisms , although this effect is not always observed [43 , 44] . In C . elegans , sir-2 . 1 is predicted to encode a nicotinamide adenine dinucleotide ( NAD ) dependent deacetylase that can extend the lifespan of C . elegans when overexpressed . We examined the null mutation sir-2 . 1 ( ok434 ) [45] . Captopril treatment significantly extended the mean and maximum lifespan of sir-2 . 1 ( ok434 ) mutant animals by 17% and 27% , respectively ( Fig 5C; Table 1 , line 17–18 ) . Similarly , acn-1 RNAi significantly extended the mean and maximum lifespan of sir-2 . 1 mutants by 16% and 18% , respectively ( Fig 6C , Table 2 , line 15–16 ) . Thus , sir-2 . 1 activity was not necessary for the lifespan extension activity of captopril or acn-1 RNAi . The target of rapamycin ( TOR ) signaling network plays an important role in nutrient homeostasis and influences adult lifespan [46] . Loss-of-function mutations in rict-1 affect TOR signaling and cause a shorter lifespan . acn-1 RNAi significantly extended the mean and maximum lifespan of rict-1 mutants by 23% and 25% , respectively ( Fig 6G , Table 2 , line 17–18 ) . hsf-1 encodes a transcription factor that is important for stress response; overexpression of hsf-1 extends lifespan and delays age-related protein miss folding [47] , whereas reducing the activity of hsf-1 causes a shorter lifespan and proteotoxicity [31 , 48 , 49] . acn-1 RNAi significantly extended the mean and maximum lifespan of hsf-1 ( lf ) mutants by 12% and 11% , respectively ( Fig 6H , Table 2 , line 19–20 ) . Thus , the activities of rict-1 and hsf-1 were not necessary for the lifespan extension caused by acn-1 RNAi . Mutations in the insulin/insulin-like growth factor ( IGF ) signaling pathway influence C . elegans lifespan [7 , 8 , 50–53] . Mutations that partially reduce the activity of daf-2 , which encodes a protein homologous to the vertebrate insulin/IGF-1 receptor , or age-1 , which encodes a protein homologous to the vertebrate PI3 kinase , extend lifespan . This signaling pathway controls the activity of a FOXO transcription factor encoded by daf-16 , and daf-16 activity is necessary for the lifespan extension caused by mutations in upstream signaling genes [13 , 14] . Thus , daf-2 and age-1 activity promote a rapid lifespan and inhibit longevity , whereas daf-16 activity promotes longevity . Captopril treatment significantly extended the mean and maximum lifespan of daf-2 ( e1370 ) partial loss-of-function mutant animals by 11% and 9% , respectively ( Fig 5D; Table 1 , line 21–22 ) . Similarly , acn-1 RNAi significantly extended mean and maximum lifespan by 11% and 9% , respectively ( Fig 6D , Table 2 , line 23–24 ) . In combination with an age-1 ( hx546 ) partial loss-of-function mutation that causes an extended lifespan , acn-1 RNAi significantly extended mean and maximum lifespan by 30% and 19% , respectively ( Fig 6F , Table 2 , line 25–26 ) . A similar result was obtained by analyzing age-1 ( am88 ) mutant animals ( Table 2 , line 27–28 , S5 Fig ) [54] . Thus , the lifespan extension caused by reducing daf-2 activity was additive with captopril treatment and acn-1 RNAi , and the lifespan extension caused by reducing age-1 activity was additive with acn-1 RNAi . By contrast , captopril treatment did not extend the lifespan of daf-16 ( mu86 ) loss-of-function mutant animals , but rather significantly shortened the mean and maximum lifespan by 10% and 11% , respectively ( Fig 5E; Table 1 , line 19–20 ) . Similarly , acn-1 RNAi slightly shortened the mean and maximum lifespan by 3% and 2% , respectively ( Fig 6E , Table 2 , line 21–22 ) . These findings indicate that the lifespan extension activity of captopril and acn-1 RNAi require daf-16 activity; however , the reduction of lifespan raises the possibility that the combination of captopril treatment or acn-1 RNAi and the daf-16 mutation causes toxicity . To investigate the possibility that acn-1 functions upstream of daf-16 , we analyzed additional phenotypes associated with the insulin/IGF-1 pathway . Upstream signaling proteins such as DAF-2 control the activity of DAF-16; specifically , daf-2 ( lf ) mutations that cause a lifespan extension also cause DAF-16 protein to localize to the nucleus , where DAF-16 controls the activity of target genes [55] . To examine the nuclear localization of DAF-16 , we used transgenic worms containing a DAF-16::GFP reporter construct [56] . Animals treated with captopril or acn-1 RNAi did not display a substantial nuclear localization of DAF-16::GFP compared to control animals ( S6C Fig , S1 Table ) . Thus , acn-1 RNAi did not cause the same effect as a daf-2 ( lf ) mutation , and acn-1 is not necessary to inhibit nuclear localization of DAF-16 . It has been proposed that daf-16 is regulated by additional mechanisms that do not involve changes in subcellular localization , such as transcript levels [57] , EAK-7 [58] , and phosphorylation [59 , 60] . Our results do not exclude the possibility that captopril or acn-1 RNAi regulate daf-16 in a manner that does not change nuclear localization . daf-2 ( lf ) mutations cause a dauer constitutive ( Daf-c ) phenotype , indicating that daf-2 is necessary to inhibit dauer development . To analyze the role of acn-1 in dauer formation , we cultured worms with acn-1 RNAi bacteria at 20°C , shifted embryos to 27°C to stimulate dauer formation , and scored dauer larvae after 72 hours . acn-1 RNAi did not increase the frequency of dauer formation compared to control RNAi in wild-type animals or rrf-3 mutant animals ( S6A Fig ) . To increase the sensitivity of the assay , we analyzed the function of acn-1 in daf-2 ( lf ) mutants that display a partially penetrant , temperature sensitive Daf-c phenotype [61] . acn-1 RNAi was not different from control RNAi in this assay ( S6B Fig ) . Thus , acn-1 RNAi did not cause the same effect on dauer formation as a daf-2 ( lf ) mutation , and acn-1 is not necessary to inhibit formation of dauer larvae . We hypothesized that captopril inhibits acn-1 to extend lifespan . This hypothesis predicts that the effects of captopril and acn-1 RNAi will not be additive , because our dose-response analysis indicates that we have identified the optimal dose of captopril for lifespan extension . To test this prediction , we combined treatment with captopril and acn-1 RNAi . Captopril treatment alone caused a 20% extension of mean lifespan to 18 . 7 days , whereas acn-1 RNAi alone caused a 38% extension of mean lifespan to 21 . 5 days ( Fig 7A , Table 4 , line 1–3 ) . Combining captopril treatment and acn-1 RNAi resulted in a 18 . 9 day lifespan that was not significantly different from captopril treatment alone and significantly shorter than acn-1 RNAi treatment alone . Thus , captopril and acn-1 RNAi did not have an additive effect on lifespan extension , consistent with the model that both effects are mediated by a similar mechanism . The identification of compounds that can delay age-related degeneration and extend lifespan is an important goal of aging research , because age-related decline is a major cause of disability and death in humans , and so far no compounds have been demonstrated to delay human aging . We reasoned that FDA-approved drugs used to treat human diseases might also influence aging and lifespan . To identify such drugs , we screened examples of different structural and functional drug classes . We previously described the identification of anticonvulsant drugs such as ethosuximide and the neuroactive drug valproic acid [19 , 20] . Here we identified the blood pressure medicine captopril as a way to extend C . elegans lifespan . The effect of captopril was dose dependent; at an optimal dose , captopril significantly extended mean lifespan 22–28% and maximum lifespan 18–32% . Captopril extended lifespan at a variety of temperatures and in a variety of mutant backgrounds , indicating that the effect is robust in the face of environmental and genetic variation . Captopril functioned in adult animals to extend lifespan , suggesting that it affects the rate of age-related decline rather than developmental processes . The first of what is now a large class of ACE inhibitors , captopril is an oligopeptide derivative developed in 1975 based on a peptide found in pit viper venom [62] . ACE inhibitors modulate the renin-angiotensin-aldosterone system , a mechanism by which the body adapts to hypotension [63] . In response to a decline in blood pressure , the kidney releases renin , which cleaves angiotensinogen to angiotensin I . ACE converts angiotensin I to angiotensin II , and angiotensin II acts through a transmembrane receptor to stimulate aldosterone secretion and promote vasoconstriction to increase blood pressure . By blocking ACE and preventing the conversion of angiotensin I to angiotensin II , captopril lowers blood pressure . Two strategies have been used to identify compounds that can extend C . elegans lifespan: screening chemical libraries and testing candidate compounds based on the hypothesis that the target of the drug may influence aging and longevity [64] . Compounds that have been tested included FDA-approved drugs , libraries of chemically defined molecules , and extracts of plants that contain a mixture of chemicals . Library screening resulted in the identification of antidepressant drugs [21 , 25] . Candidate compounds that have been reported to extend worm lifespan include resveratrol [65] , trehalose [24] , lithium [66] and garlic constituent [67] . Extracts of blueberries and ginkgo have been reported to extend worm lifespan [68 , 69] . ACE inhibitors such as captopril have not been previously reported to extend lifespan in worms , so our findings identify a new chemical entity that influences aging in C . elegans . It is well established that ACE is the target that mediates the effect of captopril on blood pressure in humans [62] . ACE genes have been highly conserved during evolution , and acn-1 encodes the C . elegans homolog of ACE [28] . A major issue in aging pharmacology is the identification of the direct target of the drug , and in most cases the targets of drugs that extend C . elegans lifespan remain unknown . We hypothesized that captopril inhibits ACN-1 to extend longevity . This hypothesis makes three important predictions that were verified experimentally . First , it predicts that reducing the activity of acn-1 using genetic techniques can extend longevity . We showed that targeting acn-1 by RNAi increased mean lifespan 20–46% and maximum lifespan 18–33% . Second , it predicts that reducing the activity of acn-1 and treatment with captopril will cause similar effects in a variety of genetic backgrounds . Indeed , captopril treatment and reducing acn-1 activity gave very similar results in five genetic backgrounds ( eat-2 , isp-1 , sir-2 . 1 , daf-16 and daf-2 ) and at two temperatures . In addition , both treatments function in adults to extend longevity . Third , it predicts that the lifespan extension caused by captopril treatment and reducing acn-1 activity will not be additive . This prediction was also verified . While these results are consistent with captopril inhibition of ACN-1 , they do not demonstrate that the drug directly binds ACN-1 protein or inhibits a biochemical activity of ACN-1 . The biochemical activity of ACN-1 has not been established , and ACN-1 may not have protease activity because critical residues in the predicted active site have not been conserved during evolution [28] . Further studies are necessary to establish an assay for the biochemical activity of ACN-1 and directly test the effect of captopril . The expression and function of acn-1 were analyzed by Brooks et al . [28] using a reporter gene encoding ACN-1::GFP and acn-1 RNAi , respectively . acn-1 is expressed in embryonic and larval hypodermis , in the vulva during organogenesis and in the ray papillae of the male tail . RNAi delivered by injection in the gonad caused larvae to arrest at the L2 stage and display evidence of molting defects . RNAi delivered by feeding to L1/L2 larvae caused a cuticle defective phenotype in L3/L4 larvae and adults . The failure to shed cuticle led to secondary defects such as vulva defects and constipation . These results indicate acn-1 is necessary for larval molting , mail tail development and formation of adult alae . Frand et al . [70] identified acn-1 in a genome-wide feeding RNAi screen for molting defects . An ACN-1::GFP transgene was expressed in the hypodermis , including the major body syncytium , hyp7 , and hypodermal cells in the head and tail , the lateral seam cells , and the excretory gland cell . Neither of these studies describe aging phenotypes , so our results establish a new phenotype for acn-1 and a novel link between acn-1 and aging . The previously reported molting defects caused by acn-1 RNAi are partially penetrant [28 , 70]; we did not observe a significant penetrance of molting defects , which may indicate less extreme gene disruption in our studies resulting from differences between the feeding RNAi constructs or the conditions of RNAi delivery . To elucidate the role of captopril and acn-1 in aging , we analyzed interactions with established pathways that influence longevity . Many mutations used in these experiments are not null alleles , and therefore the observation that the effects are additive does not exclude the possibility that two interventions act in the same pathway . Captopril treatment or reducing the activity of acn-1 was additive with the lifespan extensions caused by an eat-2 mutation that causes caloric restriction . Furthermore , these treatments did not reduce self-fertile brood size and reproductive span like caloric restriction , suggesting that captopril and acn-1 do not act by causing caloric restriction . Captopril treatment or reducing the activity of acn-1 was additive with the lifespan extensions caused an isp-1 mutation that reduces mitochondrial activity , suggesting these treatments do not reduce mitochondrial function . The lifespan extension caused by reducing the activity of acn-1 was not abrogated by loss-of-function mutations of sir-2 . 1 , hsf-1 or rict-1 , suggesting that acn-1 does not act by regulating these genes . Captopril treatment and reducing the activity of acn-1 displayed complex interactions with the insulin/IGF-1 pathway . These treatments were additive with the lifespan extensions caused by loss-of-function mutations of daf-2 and age-1 . However , the lifespan extensions caused by both treatments were abrogated by a daf-16 mutation . To further analyze the relationship with daf-16 , we demonstrated that reducing the activity of acn-1 did not cause dauer formation and did not promote nuclear localization of DAF-16 , which are typical of reducing insulin/IGF-1 signaling upstream of daf-16 . Thus , acn-1 does not appear to act upstream and regulate the nuclear localization activity of daf-16 . It is possible that daf-16 is necessary because it functions in parallel to acn-1 or that toxicity develops in the absence of both daf-16 and acn-1 . Overall , acn-1 defines a new gene that influences longevity , and interactions with known longevity pathways suggest that it functions by a mechanism that is distinct from those that have been characterized previously . The ACE inhibitor enalapril and the angiotensin II receptor antagonist losartan have been reported to extend the life span of mice and rats [71–78] . Furthermore , these drugs delay the age-related degeneration of tissue structure and function in the kidney , cardiovascular system , liver and brain . Similarly , Santos et al . , [79] showed that enalapril increased life span in rats . These interesting results indicate that the renin-angiotensin-aldosterone system promotes age-related degeneration , and blocking this system can extend longevity in rodents . The mechanism of these drugs in life span extension is not well defined–the affects are not well correlated with changes in blood pressure but may reflect preservation of mitochondrial number and function . Genetic studies reported by Benigni et al . , [80] provide important support for these pharmacology studies , since disruption of the angiotensin II type I receptor ( AT1 ) promotes longevity in mice . These results may be relevant to humans , since polymorphisms in the angiotensin II type I receptor gene are associated with extreme human longevity [81] . Overall , these studies suggest that the rennin-angiotensin-aldosterone system controls longevity in mammals . Thus , our discoveries in worms are likely to be relevant to mammalian biology . An important issue that has not been established by studies of mammals is the mechanism of action of this pathway in influencing aging and longevity . The results presented here provide new insights into the mechanism of action of captopril in lifespan extension and establish the powerful C . elegans system to investigate critical questions about the conserved activity of the pathway . C . elegans were cultured on 6 cm Petri dishes containing NGM agar and a lawn of Escherichia coli strain OP50 at 20°C unless stated otherwise [2] . The wild-type ( WT ) strain was N2 Bristol . daf-2 ( e1370P1465S ) is a partial loss-of-function mutation that affects the kinase domain of the DAF-2 receptor tyrosine kinase [50] . age-1 ( hx546P806S ) and age-1 ( am88E725K ) are partial loss-of-function mutations that affect the AGE-1 PI3 kinase [53 , 54 , 82] . daf-16 ( mu86 ) is a strong loss-of-function mutation caused by a deletion in the DAF-16 forkhead transcription factor [13 , 14]; eat-2 ( ad1116 ) is a change in a splicing site predicted to decrease the level of mRNA of the EAT-2 non-alpha nicotinic acetylcholine receptor [41]; isp-1 ( qm150P225S ) is a loss-of-function mutation that affects an iron sulfur protein of mitochondrial complex III [42]; sir-2 . 1 ( ok434 ) is a deletion that causes a loss-of-function of the SIR-2 . 1 NAD dependent protein deacetylase [45] . rict-1 ( mg360G1067E ) is a partial loss-of-function mutation of RICT-1 , a component of the target of rapamycin complex 2 ( TORC2 ) that encodes an ortholog of mammalian Rictor [83] . hsf-1 ( sy441W585stop ) is a strong loss-of-function mutation of the HSF-1 transcription factor [84] . DAF-16 nuclear localization was analyzed using strain GR1352 containing the integrated array xrIs87 [DAF-16alpha::GFP::DAF-16B + rol-6 ( su1006 ) ] [85] . The rrf-3 ( pk1426 ) mutation was used for RNAi feeding experiments [38] . Fifteen FDA-approved drugs were screened for extension of C . elegans lifespan using methods described by Evason et al . , [19] ( atropine , yohimbine hydrochloride , captopril , nicotinic acid , phenformin , haloperidol , acetazolam , adenosine , cimetidine , lidocaine , procainamide hydrochloride , caramazepine , 5’-5’-diphenylhydation Sodium , caffeine and imipramine ) . For each drug , we analyzed about 50 hermaphrodites cultured with three concentrations in the NGM medium ( X , 10-100X , 1000X ) . The lowest dose ( X ) was approximately equivalent to the effective dose in humans [63] . Captopril was obtained from Sigma Aldrich ( St . Louis , MO , USA ) , and a 30 mg/ml stock solution was prepared by dissolving the compound in water . Concentrated captopril was diluted to the desired final concentration in liquid NGM that had been autoclaved and cooled to 55°C , and 7–8 ml of medium was dispensed into 6 cm Petri dishes . Petri dishes were allowed to dry 1–2 days at room temperature and then seeded with E . coli OP50 . Lifespan experiments using dishes containing drugs were always conducted in parallel with control dishes containing no drug in the same incubator to control for day-to-day variations in temperature and humidity . Studies of lifespan were begun on day zero by placing approximately 30–40 L4 hermaphrodites on a Petri dish . Each hermaphrodite was transferred to a fresh Petri dish daily during the reproductive period ( approximately the first seven days ) to eliminate self-progeny and every 2–3 days thereafter . Each hermaphrodite was examined every day using a dissecting microscope for survival , determined by spontaneous movement or movement in response to prodding with a pick . Dead worms that displayed matricidal hatching , vulval extrusion or desiccation due to crawling off the agar were excluded from the data analysis . Average mean lifespan was calculated as the number of days from the L4 stage to the last day a worm was observed to be alive . To conduct experiments with dead bacteria , we seeded dishes with live E . coli OP50 , cultured for 24 hours , and exposed the bacteria to ultraviolet light by placing dishes in a UV Stratalinker 2400 for 15 minutes . Death was confirmed by inoculating LB medium with treated bacteria and observing no growth . To analyze progeny production , one L4 hermaphrodite was placed on a Petri dish ( day one ) , transferred to a fresh dish daily until at least 4 days without progeny production , and progeny were counted after two days . Pharyngeal pumping and body movement were determined as described previously [33] . Briefly , we observed pharyngeal pumping using a dissecting microscope for a 10 seconds interval . Body movement was assayed by observation using a dissecting microscope for 20 seconds . Petri-dishes were tapped to stimulate animals to move before scoring . RNAi interference was performed by feeding bacteria that express dsRNA as described by Kammath et al . , [86] . Briefly , E . coli HT115 bacteria with the control plasmid ( L4440 ) or a plasmid encoding acn-1 were obtained from the Ahringer library [37] , and the identity of the clone was confirmed by DNA sequencing . The daf-2 RNAi bacterial strain was provided by M . Crowder . RNAi bacteria were streaked on LB dishes containing 50μg/ml ampicillin and 12 . 5 μg/ml tetracycline . Control and acn-1 RNAi cultures were grown for 6 hours in LB medium containing 50μg/ml ampicillin . Escherichia Coli expressing double-stranded acn-1 RNA did not form thick lawns on RNAi NGM agar dishes containing isopropyl β-d-1-thiogalactopyranoside ( 1 mM ) and 50μg/ml carbenicillin , indicating double stranded acn-1 RNA might inhibit bacterial proliferation . To address this issue , we prepared 3X-concentrated liquid bacterial culture from both control and acn-1 RNAi bacteria , spread this on NGM RNAi dishes , and allowed dishes to incubate overnight . L4 stage larvae were transferred to RNAi dishes and cultured for one day , adults were transferred to a fresh RNAi dish and cultured for one day and then removed . Larva that developed on these plates were analyzed . Dauer formation was assayed as described by Kimura et al . , [50] . Briefly , we collected eggs from wild-type or rrf-3 ( pk1426 ) hermaphrodites cultured at 20° , transferred the eggs to 27°C with ample food , cultured for 72 hr , and examined hatched animals . Animals were classified as non-dauer ( including adults and non-dauer larvae ) or dauer on the basis of morphological criteria [61] . To analyze dauer formation of daf-2 ( e1370 ) mutant animals , we transferred eggs to 15°C , 17 . 5°C , 20°C , 22 . 5°C , or 25°C . For dauer formation experiments , we performed acn-1 RNAi using the feeding protocol described by Kammath et al . , [86] . Briefly , L4 stage hermaphrodites were transferred to dishes with control ( L4440 ) or acn-1 RNAi bacteria at 20°C , and embryos were transferred to fresh dishes with RNAi bacteria at the appropriate temperature and cultured for 3 or 4 days . To analyze DAF-16::GFP localization , we used the strain GR1352 [85] . L4 stage animals were transferred to dishes seeded with control ( L4440 ) and acn-1 RNAi bacteria . Progeny were analyzed at the one day old adult stage using an Olympus SZX12 dissecting microscope ( Tokyo , Japan ) equipped for fluorescence microscopy . To reduce bias , the scoring was done by an observer blind to the RNAi treatment status . We analyzed each worm as having ( 1 ) GFP diffusely localized in the cytosol , ( 2 ) GFP localized in nuclei displaying intensely fluorescing puncta throughout the entire body from head to tail or ( 3 ) intermediate nuclear localization of GFP , defined as puncta observed in at least one or more nuclei but not in most or all nuclei . To perform the data analysis , we combined the nuclear and intermediate nuclear categories . Thermotolerance assays were performed as described by McColl et al . , [87] . Briefly , L4 stage hermaphrodites were cultured at 20°C on control ( L4440 ) and acn-1 RNAi dishes for 3 days . To perform the heat stress assay , we transferred adults to 34°C and scored the percentage of dead and live animals starting at 6 hours and continuing every hour until all animals died . Animals were scored as dead if they did not respond to a mechanical stimulus . To perform oxidative stress assays , we transferred day 3 adult hermaphrodites to NGM dishes containing 40 mM paraquat and scored for survival every 12 hours . For the heat stress and paraquat stress assays , animals that displayed matricidal hatching or vulval extrusion were not included in the data analysis . To perform oxidative stress assays with juglone , we transferred day 3 adult hermaphrodites to 2 ml of liquid M9 medium containing 240 uM juglone in an 18 well dish . Worms were scored for survival after 9 hours . Paraquat and juglone were obtained from Sigma Aldrich ( St . Louis , MO , USA ) . To quantify mRNA levels , we cultured rrf-3 adult worms on control and acn-1 RNAi dishes for 3–4 hours to obtain synchronized eggs , removed adult worms , and continued culture until the eggs developed into two day old adult worms . These adults were washed and collected for RNA isolation . RNA analysis was performed as previously described with modifications [88] . Briefly , RNA was isolated using Trizol ( Invitrogen ) and treated with DNAse 1 enzyme . cDNA was synthesized by using High Capacity cDNA Reverse Transcription kit ( Applied Biosystems ) . Quantitative , realtime PCR was performed using an Applied Biosystems Step One Plus Real-Time PCR system and iTaq Universal SYBR Green Supermix ( BioRad Laboratories , Hercules , CA ) . mRNA fold change was determined by comparing acn-1 mRNA levels with mRNA levels of the reference gene rps-23 . Forward and reverse amplification primers were: rps-23 5′- aaggctcacattggaactcg and 5′- aggctgcttagcttcgacac; acn-1 5′- gtactacgagccactcatcaac and 5′- gaatctcctcgacagtgaatg . All data were analyzed using the two-tailed student t-test for samples with unequal variances by using Excel and http://studentsttest . com . P values less than 0 . 05 were considered statistically significant . To determine if the choice of a statistical test affected the conclusions , we used the log rank ( Mantel-Cox ) method to analyze a subset of the lifespan experiments . Both tests produced similar P values .	Age-related degeneration is a fundamental feature of animal biology and an important contributor to human disability and death . However , no medicines have been shown to delay human aging . To identify drugs that delay age-related degeneration , we screened FDA-approved compounds and discovered that the hypertension drug captopril significantly extended C . elegans lifespan . In humans , captopril inhibits angiotensin converting enzyme ( ACE ) to regulate blood pressure . The C . elegans homolog of ACE is encoded by the acn-1 gene . We discovered that reducing the activity of acn-1 also caused a robust extension of lifespan and delayed age-related changes in C . elegans . Captopril and acn-1 have a similar mechanism of action; both treatments displayed similar interactions with previously characterized pathways , and combining treatment with captopril and reducing the activity of acn-1 did not have an additive effect on life span extension . These results identify a new drug and a new gene that influence aging in C . elegans . They may be relevant to other animals such as humans because the pathway that includes ACE has been conserved during evolution . These findings establish a foundation for possible therapeutic interventions that can delay age-related degeneration .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "bacteriology", "invertebrates", "medicine", "and", "health", "sciences", "rna", "interference", "caenorhabditis", "oxidative", "stress", "microbiology", "animals", "drug", "screening", "animal", "models", "physiological", "processes", "developmental", "biology", "caenorhabditis", "elegans", "age", "groups", "model", "organisms", "adults", "organism", "development", "bacterial", "genetics", "epigenetics", "pharmacology", "microbial", "genetics", "research", "and", "analysis", "methods", "genetic", "interference", "gene", "expression", "aging", "people", "and", "places", "biochemistry", "rna", "cell", "biology", "nucleic", "acids", "physiology", "genetics", "nematoda", "biology", "and", "life", "sciences", "population", "groupings", "drug", "interactions", "organisms" ]	2016	Angiotensin Converting Enzyme (ACE) Inhibitor Extends Caenorhabditis elegans Life Span
Insufficient licensing of DNA replication origins has been shown to result in genome instability , stem cell deficiency , and cancers . However , it is unclear whether the DNA damage resulting from deficient replication licensing occurs generally or if specific sites are preferentially affected . To map locations of ongoing DNA damage in vivo , the DNAs present in red blood cell micronuclei were sequenced . Many micronuclei are the product of DNA breaks that leave acentromeric remnants that failed to segregate during mitosis and should reflect the locations of breaks . To validate the approach we show that micronuclear sequences identify known common fragile sites under conditions that induce breaks at these locations ( hydroxyurea ) . In MCM2 deficient mice a different set of preferred breakage sites is identified that includes the tumor suppressor gene Tcf3 , which is known to contribute to T-lymphocytic leukemias that arise in these mice , and the 45S rRNA gene repeats . Licensing of DNA replication origins begins early during the G1-phase of the cell cycle when ORC and CDC6 recruit CDT1-MCM2-7 to the chromatin [reviewed in 1] . MCM2-7 is a heterotypic hexameric complex that is the replicative helicase and , once loaded , its association with the DNA is stable until it becomes active during S-phase . Licensing is restricted to the G1-phase of the cell cycle to prevent endo-reduplication of the DNA by relicensing already duplicated DNA during S-phase . Many more sites are licensed with MCM2-7 complexes than are typically used in S-phase and these dormant origins are thought to serve as a backup system for recovering replication in the event of replication fork stalling or collapse [2–3] . The importance of this system of backup origins for maintaining genome stability is demonstrated by the observation that mice in which MCM2-7 proteins are deficient or compromised develop normally but exhibit increased genome instability , high rates of cancer and stem cell deficiencies [4–6] . The cancers that arise in MCM deficient mice can be highly specific to a particular genetic background . For example , >80% of hypomorphic MCM4 ( Mcm4Chaos3/Chaos3 ) females succumb to mammary adenocarcinomas when carried on the C3H genetic background on which it was isolated [4] . In contrast when bred into a C57Bl/6J background females are highly prone to histiocytic sarcoma [7] . Similarly , MCM2 deficient ( Mcm2IRES-CreERT2/ IRES-CreERT2 ) mice exhibit near to 100% penetrance of T-lymphocytic leukemia ( TLL ) within 5 months on the 129Sv background on which they were constructed [5] but a broader spectrum of tumor types with more variable latencies in a mixed 129Sv/BALB/c genetic background [8] . The genetic lesions that occur in tumors arising in MCM deficient or hypomorphic mice are largely copy number alterations ( CNAs ) with a preponderance of deletions averaging 400–500 kbp in size [9–10] . Several recurrent deletion sites are found in TLLs from MCM2 deficient mice where , on the 129Sv genetic background , the Tcf3 gene on Chr10 is affected in all tumors examined [9] . Tcf3 is a transcriptional regulator that plays a central role in T-cell differentiation and prior studies have shown that loss of Tcf3 is sufficient to drive TLL in mice [11] . Importantly , the Tcf3 gene lies within a region of Chr10 that shows preferential loss of origin function in MEFs from MCM2 deficient mice [12] . Replication licensing may become limiting due to genetic deletion of Mcm2-7 , which occurs in many tumors [12] , oncogenic stress [1] and during stem cell aging [13] . It is not known if under these conditions genome instability is elevated generally across the genome or whether specific locations are at increased risk; although loss of MCM expression in aging hematopoietic stem cells ( HSCs ) has been correlated with nucleolar associated DNA damage foci [13] . In the present study we use micronuclear DNA sequences to show that the region of Chr10 carrying the Tcf3 gene is also the site of elevated ongoing genome instability even prior to tumor initiation in MCM2 deficient mice on the 129Sv genetic background . The increased instability at this site predicts the high rate of genetic lesions affecting the Tcf3 gene and consequent high rate of TLL in these mice . In addition to loss of Tcf3 , MCM2 deficient mice exhibit genetic lesions within the 45S ribosomal RNA gene repeats clusters on Chrs 12 , 16 , 18 and 19 consistent with the observed nucleolar associated DNA damage foci in aging HSCs . The present study demonstrates that the consequences of reduced MCM expression on local genome instability is reflected in micronuclear DNA sequences allowing prediction of specific genetic lesions in the etiology of cancer and during aging . To map sites of ongoing genomic instability across the genome we take advantage of the fact that in the hematopoietic system of mammals DNA remnants resulting from genetic damage events during differentiation of hematopoietic stem cells ( HSCs ) to erythrocytes are retained in the cells as micronuclei following enucleation [14] . To isolate micronuclear DNA , cells from whole blood were first pelleted and washed to remove serum and serum DNAs . Cells were then fractionated into lymphocyte and red blood cell ( RBC ) plus granulocyte fractions using ficoll-paque . The RBC/granulocyte pellet was then re-suspended , RBCs were lysed using standard RBC lysis protocols and granulocytes were pelleted . DNAs were recovered from three fractions: lymphocyte ( WBC ) , granulocyte ( GRN ) , and the supernatant from the lysed RBCs ( containing the RBC micronuclei , MN ) . Tagged sequencing libraries were prepared from each sample to allow multiplexed high throughput sequencing such that WBC , GRN , and MN from the same animal are run together on the same sequencing lane on an Illumina HiSeq 2500 . The resulting sequences were mapped and wiggle files were generated . The sequence coverage , genome wide , from the MN fraction of a wild type ( wt ) 129Sv mouse is shown in Fig 1a ( for comparison , sequence coverage from the WBC and GRN genomic DNAs are shown in S1 Fig panels a and b respectively ) . Sequence tag density in the WBC and GRN fractions of wt mice are largely uniform both between and across individual chromosomes . In contrast , in the MN fraction , although the minimal sequence tag density is similar between chromosomes , there is a significant deviation from the average sequence tag density as a function of position across each chromosome . This , in part , reflects the frequency with which different portions of each chromosome are present in micronuclei . Similar patterns are seen in the MN fractions of two different wt mice where the correlation between experimental repeats is 0 . 988 . Several parameters are defined to describe the observed changes quantitatively ( Fig 1b and 1c ) . First , a value α is defined as a base line measure of the representation of each whole chromosome . This parameter is specific to each chromosome and is expected to reflect both the whole chromosome loss rate for that chromosome in MN plus any whole genomic DNA contamination from nucleated cells . The value of α is estimated from the minimum of the sequence tag coverage plot across the chromosome . A value β is defined to describe the observation that sequence tag density increases over all chromosomes as a function of distance from the centromere . This observation is consistent with double strand DNA breaks ( DSBs ) resulting in exclusion from nuclei of acentromeric DNA fragments from the breakpoint through the distal end of the chromosome [14] . A third phenomenon , which is the converse of β , is referred to as ρ and describes a small increase in sequence tag density as a function of distance relative to the centromere distal telomere of each chromosome and which is observed primarily on the longest chromosomes ( Chr1 and Chr2 ) . The mechanism leading to the observed ρ effect is unclear but may reflect mis-segregation or breakage of dicentric chromosomes resulting from errors during DNA repair/translocation [14–15] . The contributions of α , β and ρ to MN sequence tag density , as a function of position on the chromosome , are shown schematically in Fig 1b . In addition to chromosome wide changes , there are discrete locations on most chromosomes that show local increases in sequence tag density that are greater than those that would be expected based on the effects of α , β and ρ . These locations are described by γ which is defined as localized changes in sequence tag density that are cumulative distal to the centromere even through the rate of change returns to levels predicted by β and ρ ( Fig 1c ) . In principle , increased γ values are expected to reflect localized regions ( hot spots ) of the genome where chromosome breaks occur at increased frequency . Functions describing β and ρ for wt animals were established using Chr7 and Chr11 since , by inspection , there are few localized increases in sequence tag density on these chromosomes . Both β and ρ are non-linear where the best fit for each is a quadratic function ( S5 Fig panel a ) . It is unlikely that the non-linearity results from a bias in the locations of the initial DSB sites and consistent with this interpretation DSBs that have been repaired by translocation do not exhibit a bias that would lead to β or ρ distributions in micronuclei [15] . One possible explanation is that events occurring during anaphase distort the representation of acentromeric chromosomal fragments in the micronuclear fraction . For example , longer stretches of sister chromatid pairing within the acentromeric region resulting from DSBs more proximal to the centromere may promote non-disjunction and retention of the acentromeric fragment within a nucleus . This could lead to preferential representation of acentromeric chromosome fragments resulting from breaks nearer to the centromere distal ends of the chromosomes consistent with the observed non-linear increase described by β . To identify localized regions of the genome exhibiting elevated instability , γ values were estimated by determining the rate of the sequence tag coverage change within smoothed and normalized 20 kbp windows genome wide . Although the overall contribution of β and ρ to differential representation of centromere distal and , to a lesser extent proximal ends of whole chromosomes is significant over entire chromosomes , over shorter 20 kb intervals these effects are minimal and the slope of the sequence tag density largely reflects the localized effect γ ( e . g . Fig 1d , γ track ) . Similar to the case for sequence independent breaks , the magnitude of peaks identified by γ is biased towards the centromere distal ends of the chromosomes . This result is expected if the same forces that act to skew the distribution of sequences represented in micronuclei following sequence non-specific breaks also act on chromosome fragments resulting from local hot spots . The effect of normalizing γ peaks using β and ρ values is shown in Fig 1d , γ-normalized track . γ plots show 294 peak locations across the genomes of wt mice . Of these 129 occur in early replicating gene rich regions of the genome and 165 occur in gene poor late replicating , regions of the genome . Prior studies have shown that agents that inhibit replication fork progression lead to chromosome breakage at specific locations across the genome referred to as common fragile sites . Characteristics of these sites have been defined where breakage typically occurs at large , late replicating , transcriptionally active genes [16] . To determine if Mic-Seq identifies common fragile sites , Mic-Seq was performed on hydroxyruea ( HU ) treated mice . HU induces replication stress through inhibition of ribonucleotide reductase which leads to reduced nucleotide pools and increased replication fork stalling [17–18] . To establish an informative dose of HU when administered in the drinking water a titration was performed where mice were assayed at both 1 week and 3 weeks of treatment with different HU doses for micronuclear frequency by FACS ( S2 Fig panel a ) and at 3 weeks for effects on the levels of various cell types within the blood by CBCs ( S2 Fig panel b ) . The data shows that there is a narrow HU concentration window at which micronuclear frequency is elevated ( by ~10 fold ) during the interval between 1 and 3 weeks , but which has minimal effects on the frequency of various blood cell types . The short half-life of HU [19] and intermittent dosing resulting from administration in the drinking water makes it likely that only a subset of cells , in various stages of S-phase , is transiently exposed to sufficiently high concentrations of HU to induce damage . The observation that the frequency of micronuclei continues to increase for at least three weeks is consistent with this possibility but also suggests that , once formed , micronuclei are maintained stably in circulating RBCs . Part of the efficacy of Mic-Seq analysis likely depends on the ability to accumulate DNA remnants over a period of time . Mice treated with a dose of 2 mg/ml HU , which resulted in 1–2% micronucleated RBCs , were used for Mic-Seq . Mic-Seq was performed on two mice treated with this dose and an additional untreated wt control . Genome-wide sequence tag distributions are shown in Fig 2 for the MN fraction of both of the HU treated mice and is summarized for all three animals in Fig 3a . The micronuclear fractions from HU treated mice show a more rapid increase in sequence tag density as a function of distance from the centromere relative to wt mice and the value estimated for β ( again using Chr7 and Chr11 ) is increased ~5 fold ( S5 Fig panel a ) . The increase in β is accompanied by a decrease in the value for ρ suggesting that more frequent chromosomal breaks suppress the mechanism that results in preferential retention of centromere proximal sequences ( S5 Fig panel a ) . HU also affects the distribution of MN sequences across whole chromosomes where Chrs 12 , 13 , 16 , 18 , 19 and X show an ~2 fold , increases in α values ( Fig 3b ) . The 45S rRNA gene sequences , which are present on a subset of these chromosomes ( see below ) , also show a modest ( ~5–20% ) overrepresentation in the ratio of rDNA sequence tags to major satellite sequence tags in MN DNA relative to WBC or GRN DNAs from the same mice ( Fig 3c ) . The sequence tag density of major satellite sequences is more strongly enriched ( ~ two-fold ) relative to minor satellite sequences in HU treated mice consistent with preferential induction of breaks in at least a subset of the major satellite sequences by HU ( Fig 3d ) . Inspection of the micronuclear sequence data from HU treated mice reveals sharp increases in sequence tag coverage distal to discrete locations suggesting that specific sites are disproportionately affected by HU treatment ( 5 such locations are marked 1–5 in Figs 2 and 3a ) . Extraction of γ peak values identifies these and additional localized increases in sequence tag densities that include 5 of the 8 molecularly characterized common fragile sites in the mouse ( e . g . site 2 is Wwox , S2 Fig ) and additional large , late replicating , transcriptionally active genes with properties of common fragile sites ( [16]; Fig 4a–4c and S2 Fig ) . In these cases , instability occurs within sub-domains of the genes ( Fig 4a and 4b and S2 Fig , panel c ) consistent with the fragile site core regions observed in prior studies [15 , 20] . To examine the effect of insufficient DNA replication licensing on genome instability , Mic-Seq was performed on mini-chromosome maintenance ( MCM ) protein 2 deficient mice [5] . MCM2 deficient mice on the 129Sv genetic background exhibit early onset T-lymphocytic leukemia , loss and dysfunction of stem cells , and genome instability evidenced by increased γH2AX in nucleated cells and increased micronuclei in reticulocytes/erythrocytes [5] . The elevated frequency of micronuclei is confirmed in S3 Fig , panels 3a-3c , demonstrating that micronuclei are approximately 10 fold more frequent in RBCs of MCM2 deficient relative to wt animals . For Mic-Seq , MCM2 deficient mice between 5–6 weeks of age , well before the onset of overt disease , were used . MN sequence tag density across the genome of an MCM2 deficient 129Sv mouse is shown in Fig 5a . Two biological repeats of the experiment were performed where the genome-wide correlation between experimental repeats was 0 . 975 . The sequence coverage across all chromosomes is show for both experiments and in comparison to two wt mice in Fig 5 panel b . Comparison of the sequence tag densities derived from micronuclei of wt and MCM2 deficient 129Sv mice shows that the additional breaks resulting from MCM2 deficiency suppress ρ values and modestly increase β values relative to wt cells ( S5 Fig panel a ) . However , the most pronounced differences are on the α values associated with chromosomes 6 , 12 , 18 and 19 , and to a lesser extent on chromosomes 15 and 16 , where these chromosomes are over represented by between ~1 . 8–12 fold relative to the average sequence tag density across the genome ( Fig 5c ) . The entire mapped region of each of these chromosomes is affected . Unlike the case for HU treatment , the ratio between minor and major satellite sequences is not affected ( S3 Fig panel g ) suggesting that breaks within the major satellite elements are not responsible . In C57Bl6 mice , chromosomes 12 , 15 , 16 , 18 and 19 carry nucleoli encoding the 45S ribosomal rRNA gene repeats at centromere proximal positions ( NCBI http://www . ncbi . nlm . nih . gov/gene/19791 ) . These are typically composed of 30–40 repeats of an approximately 45 kb repeating unit at each location ( additional low copy number rDNA repeats have also been mapped to Chr1 , Chr6 , and Chr9 in some strains; [21] ) . One potential explanation for the over representation of the subset of chromosomes seen in micronuclei from MCM2 deficient mice is that rDNA repeats are hypersensitive to reduced replication origin licensing resulting in high rates of double strand breaks within these repeats . Since rDNA repeats are adjacent to the centromeres in these chromosomes , a DSB within an rDNA repeat would render nearly all of the affected chromosome acentromeric . To determine the locations of rDNA clusters in 129Sv mice , metaphase chromosomes from wt 129Sv MEFs were assayed by fluorescence in situ hybridization ( FISH ) plus spectral karyotyping ( SKY ) to localize a probe for the 45S ribosomal gene ( containing portions of the 18S , 5 . 8S and 28S ribosomal genes ) to specific chromosomes . Results from these studies ( S3 Fig panels 3d-3e ) localize ribosomal gene repeats to Chrs 12 , 16 , 18 and 19 in most metaphase spreads from 129Sv mice . In addition , a small subset ( 2% ) exhibit signal consistent with recombination events leading to the presence of rDNA repeats on Chr6 ( and in MCM2 deficient MEFS , Chr15 ) . Although these results are consistent with a role for rDNA repeats in the increased representation of specific chromosomes within micronuclei , the relative FISH signal intensity between different chromosomes ( 19 = 12>18 = 16 , S3 Fig panel f ) does not correlate with the representation of these chromosomes in MN of MCM2 deficient mice ( 19>>12 = 18>>16 , Fig 5c ) . However , if the increased representation of the mapped regions of these chromosomes results from DSBs in the rDNA repeats , representation of the acentric regions is expected to be affected by β . Following normalization for β , the representation of different chromosomes in MN is similar to that expected based on rDNA copy number as estimated from FISH ( i . e . 19 = 12>18 = 16 ) . rDNA sequences are over represented by a factor of 2–3 in the MN fraction , relative to the GRN or WBC fractions , of MCM2 deficient but not wt mice ( Fig 5d ) . Further , rDNA sequences are enriched by a factor of 2–5 fold relative to peri-centromeric ( major satellite , Fig 5e ) and centromeric ( minor satellite ) repeat sequences in MN of MCM2 deficient mice in comparison with the WBC or GRN fractions of the same mice or all fractions from wt animals . These results support a large increase in the rate of DSBs in at least a subset of rDNA repeats of MCM2 deficient relative to wt mice . Consistent with this interpretation , many of the additional γH2AX foci observed in MCM2 deficient , relative to wt , MEFs are located over nucleoli ( S4 Fig panels a-g ) . Further , short nascent stand analysis shows that a subset of DNA replication origins within the 45S rRNA gene repeats are preferentially affected by MCM2 deficiency ( S4 Fig panels h-j ) . To examine the effect of MCM2 deficiency on localized increases in genetic damage in regions other than the 45S rRNA gene repeats , γ values were compared between wt and MCM2 deficient mice genome-wide ( e . g . Chr10 , Fig 6a ) . Most locations where peaks are present in the γ plot of MN from wt mice are also represented in MN from MCM2 deficient mice ( 214 common peaks across the autosomes ) but at increased values where the average increase was 2 . 4 fold ( Fig 6b , blue diamonds ) . An additional 63 peaks are present only in MN from MCM2 deficient mice ( Fig 6b , green circles ) . Prior studies [9] have identified locations of recurrent deletions in T-lymphocytic leukemias ( TLLs ) arising in MCM2 deficient mice . One location that undergoes deletion in all TLLs of MCM2 deficient mice on the 129Sv genetic background is the Tcf3 gene on Chr10 and this location is highlighted in Fig 6a . Tcf3 is required for T-cell differentiation and loss of Tcf3 results in TLL [11] . This site is also a location at which MCM2 deficiency has a disproportionately strong effect in reducing origin usage as measured by short nascent strand analysis [12] . Increased genome instability is detected by Mic Seq at this site and , of the early replicating regions of the genome , the region containing the Tcf3 gene shows the highest level of instability genome wide ( Fig 6c ) . It is likely that instability at this site drives a high rate of loss of Tcf3 resulting in the near 100% penetrance of early onset TLLs in these mice . Smaller local increases in γ values are also found at many of the additional recurrent deletion sites found in these tumors ( Fig 6c ) . However , locations where MCM2 deficiency has the largest effects on representation of sequences in MN , and results in the greatest increases in γ peak values , occur in preferentially in late replicating gene poor regions of the genome that are not the sites of recurrent deletions in TLLs . Chromosomal fragile sites are genomic locations that are hotspots for genome instability leading to translocations , amplifications and deletions . Such locations were first defined cytogenetically following treatment of cells in culture with agents that impede DNA polymerase ( including aphidicolin , hydroxyurea , 5-azacytidine and bromodeoxyuridine ) and mapping breaks in banded metaphase chromosomes . Numerous studies have characterized sites that are frequently affected in human and mouse cells and led to identification of a set of locations termed common fragile sites that are affected under conditions of chemically induced replication stress in a high proportion of individuals . These sites have been extensively characterized and are significantly associated with the presence of large , transcriptionally active , and late replicating genes over 300 kbp in size [16] . The frequency of chromosome breaks at these locations is dependent on cell type and the specific agent used to induce replication stress . Here we have used the representation of different genomic regions in micronuclear DNA sequences to infer the frequency of chromosome breaks in erythroid cells in vivo . Many micronuclear sequences result from the presence of double strand DNA breaks that lead to failure of acentromeric portions of chromosomes to segregate to the nucleus during mitosis . Further we take advantage of the fact that micronuclei are retained in maturing RBCs following enucleation in mammals . In contrast to defining fragile sites cytogenetically , harvesting micronuclei from RBCs allows recovery of tens of millions chromosomal remnants each of which defines a breakpoint that can be queried at nucleotide level resolution by high throughput sequencing . Although there is a concern that erythroblasts ( just before enucleation ) may not reflect the DNA repair and checkpoint responses typical of other cells , application of the Mic-Seq method to define fragile sites in mice treated with hydryoxyurea shows that the method identifies 5 of the 8 molecularly characterized fragile sites previously defined in mouse lymphocytes [22] including those occurring at the Wwox and Immp2 genes . Further sites detected in this study support that the majority of the most sensitive HU induced fragile sites occur within subdomains of the transcribed regions of large ( >300 kbp ) , transcriptionally active , genes similar to prior studies and consistent with the possibility of interference between the transcription and replication machinery under conditions of DNA polymerase inhibition [15–16 , 20] . Similar to HU treatment , reduced replication licensing results in genome instability , increased DSBs , and chromosomal deletions and rearrangements [2–5 , 9] . The mechanism resulting in this damage is likely to differ from that mediating HU induced damage; however , it has not been previously determined whether similar or different locations across the genome are preferentially affected . In this study we demonstrate that MCM2 deficient mice exhibit a distinct sequence tag distribution profile in Mic-Seq relative to HU treated mice . HU induced fragile sites are not preferentially sensitive to MCM2 deficiency ( e . g . Fig 4 , panels a and b ) and there is little overlap with early replicating fragile sites induced by higher concentrations of HU [23] . Unlike HU treated mice , the locations at which MCM2 deficient mice exhibit localized damage by Mic-Seq analysis are largely locations that are already sensitive in wt cells but become more prone to breakage in MCM2 deficient mice . Differences in the genome instability profiles are likely to reflect differences in the mechanisms by which HU and MCM2 deficiency affect genome stability . Prior studies have shown that , unlike HU or aphidicolin , reducing MCM levels does not affect the rate of DNA polymerization since similar tract lengths of CldU or IdU incorporation are found between wt and MCM deficient cells by DNA fiber analysis [2–3 , 8] . However , DNA fiber analysis also shows that MCM deficient cells are unable to initiate DNA synthesis from dormant origins under conditions ( HU treatment ) that lead to replication fork stalling [2–3 , 8] consistent with a reduction in the frequency of licensed origins . Short nascent strand analysis has shown that origin usage does not decline uniformly across the genome , but rather specific locations lose function preferentially , in MCM2 deficient MEFs [12] . These locations tend to occur in gene rich , early replicating , regions of the genome although they are not exclusive to transcribed regions of active genes and include origins within inactive genes and intergenic regions [12] . In contrast , Mic-Seq localizes sites that are preferentially prone to breakage to subsets of both early and late replicating regions . This result suggests that factors in addition to the degree of reduction in replication licensing affect the rate of chromosome breaks under conditions of MCM2-7 deficiency . However , even in late replicating regions of the genome these factors are not the same as those determining fragile site locations following HU treatment since the locations that become sensitized to breakage by MCM2 deficiency are not known common fragile sites , or large genes generally , and many contain few or no transcribed regions . Even within early replicating regions of the genome chromosomal breaks detected by Mic-Seq show only a modest correlation with locations where SNS analysis demonstrates that origin usage is most affected . Nonetheless , it is important that many of the locations showing recurrent deletions in the TLLs that arise in MCM2 deficient mice [9] are sites at which there is both a preferential loss of origin function [12] and an increase in DSBs detected by Mic-Seq . In particular , loss of Tcf3 is sufficient to drive TLL formation [11] and the region carrying this gene shows a strong differential signal between wt and MCM2 deficient mice in both SNS and Mic-Seq analyses . These differences are apparent in samples taken well before tumorigenesis is observed , likely before tumors are initiated , and support that an increased DSB rate detected by Mic-Seq can in some instances predict a high probability of tumor occurrence . The strongest signals observed by Mic-Seq in MCM2 deficient mice are associated with chromosomes carrying nucleoli and implicate a high rate of DSBs within rDNA clusters as the location of much of the DNA damage , and the source of a large proportion of the additional micronuclei , found in MCM2 deficient mice . This observation confirms and extends prior studies showing that , as mice age , MCM levels are suppressed in HSCs and , coincident with the loss of MCM expression , nucleolar associated γ-H2AX foci accumulate [13] . The presence of unidirectional replication fork barriers [reviewed in 24] may sensitize rRNA gene repeats to loss of licensed replication origins . These results demonstrate that experimental reduction of MCM proteins in young asymptomatic mice is sufficient to cause a profile of genome instability that predicts at least a subset of chromosomal locations where genetic damage is found in both cancers and during aging . The observation that , unlike HU , many locations where MCM2 deficiency causes instability can already be detected in young wt mice suggests that even under normal conditions the distribution of licensed DNA replication origins contributes to base line levels of chromosomal instability . Although replication stress is widely recognized as a potent cause of genomic instability [1 , 25] , the present study emphasizes that different mechanisms leading to replication stress have very different consequences for instability at various locations across the genome and can result in very different phenotypic outcomes . Animal husbandry programs and protocol reviews are in compliance with NIH , USDA , and New York State Standards . Mice were maintained in facilities covered under NIH assurance #A-3143-01 , certified by New York State for the use of living animals , and the USDA APIHS registration as research facility #21–124 . The studies were approved by the Roswell Park Cancer Institute Animal Care and Use Committee under Protocols 817M and 876M . Five to six week old wild type 129Sv and Mcm2 IRES-CreERT2/IRES-CreERT2 ( MCM2 deficient ) mice were used in studies addressing the effects of MCM2 deficiency . For studies addressing the effects of hydroxyurea ( HU ) , 3 month old wild type 129Sv mice were administered HU continuously in the drinking water at the concentrations indicated in the text . Blood samples were taken by retro-orbital bleed or cardiac puncture . For flow cytometric analysis of micronuclei [26] blood samples were fixed in methanol on the day of sample collection and processed for flow cytometry using the Litron MicroFlow plus kit for mouse blood as per the manufacturer’s instructions ( Cat . No . 552730 , BD Biosciences ) . Combined SKY/FISH was performed on wt or MCM2 deficient mouse embryonic fibroblast ( passage 3 ) by the Roswell Park Cancer Institute SKY/FISH core facility . The rDNA probe for FISH analysis was prepared using Nick Translation Reagent Kit 07J00-001 ( Abbott Molecular Inc . ) Green-dUTP 02N32-050 ( Abbott Molecular Inc . ) to fluorescently label a 7109 bp EcoRI fragment from human genomic ribosomal gene DNA containing a portion of the 18S ribosomal RNA gene , the intergenic spacer , the 5 . 8S ribosomal RNA gene and a portion of the 28S ribosomal RNA gene . Between 400–500 μl of whole blood was washed with 10 ml of phosphate buffered saline ( PBS ) 3 times and re-suspended in 3 ml PBS . The sample was then layered over 2 ml of Lymphocyte Separation Medium ( density = 1 . 077–1 . 080 g/ml; Mediatech Inc . ) in a 15 ml centrifuge tube and spun for 15 min at 800 RPM . Lymphocytes ( WBCs ) at the PBS-media interphases were collected and placed in a 15 ml tube and washed 3 times with 10 ml PBS prior to pelleting for DNA isolation . Separation media was removed from the red blood cell ( RBC ) /granulocyte ( GRN ) pellet in the original tube and cells were washed 3 times with 10 ml PBS and pelleted . The cell pellet was then resuspended in 4 ml of RLF lyse buffer and incubated at room temperature for 5 minutes prior to spinning at 800 RPM for 5 minutes . The supernatant was collected as the RBC micronuclear fraction ( MN ) . The pellet is the granulocyte fraction and was washed with 10 ml PBS and pelleted for DNA isolation . WBC and GRN pellets were re-suspended in 2 ml 1X lysis buffer with 1 ug/ml proteinase K and the RBC fraction was brought to 1X lysis buffer and 1 μg/ml proteinase K using 4X lysis buffer . Samples were incubated at 37°C overnight and extracted with an equal volume of phenol-chloroform-isoamyl alcohol . Nucleic acid was precipitated with isopropanol , washed with 70% ethanol and re-suspended in 100 ul TE buffer . Samples were treated with 2000 units ribonuclease T1 ( Invitrogen ) and 3 ug ribonuclease A ( Invitrogen ) for 30 minutes at 37°C followed by phenol-chloroform-isoamyl alcohol extraction and ethanol precipitation of the DNA . Sequence data from this study have been submitted to the NCBI short reads archive ( SRA ) , ( https://www . ncbi . nlm . nih . gov/sra/ ? term=SRP091564 ) under accession number SRP091564 .	Many RBC micronuclei result from double strand DNA breaks that give rise to acentromeric chromosomal fragments that fail to incorporate into nuclei during mitosis and consequently remain in the cell following enucleation . Here , RBC micronuclear DNA is sequenced ( Mic-Seq ) to define the locations of breaks genome-wide and this assay is used to study ongoing genome instability resulting from insufficient DNA replication origin licensing . Using a mouse model , we show that there is increased instability at discrete sites across the genome , which include genes that are recurrently deleted in the T-lymphocytic leukemias that eventually arise in these mice . Mic-Seq may provide an effective means of predicting locations that are susceptible to genetic damage and these predictions may have prognostic value .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods", "Data", "Access" ]	[ "genetic", "networks", "micronuclei", "microbiology", "sequence", "tagged", "site", "analysis", "dna", "replication", "network", "analysis", "protein", "structure", "mammalian", "genomics", "dna", "research", "and", "analysis", "methods", "sequence", "analysis", "computer", "and", "information", "sciences", "chromosome", "biology", "bioinformatics", "proteins", "repeated", "sequences", "protein", "structure", "networks", "molecular", "biology", "animal", "genomics", "biochemistry", "macromolecular", "structure", "analysis", "cell", "biology", "nucleic", "acids", "database", "and", "informatics", "methods", "gene", "identification", "and", "analysis", "genetics", "biology", "and", "life", "sciences", "protozoology", "genomics", "chromosomes" ]	2017	A Signature of Genomic Instability Resulting from Deficient Replication Licensing
AIP56 ( apoptosis-inducing protein of 56 kDa ) is a major virulence factor of Photobacterium damselae piscicida ( Phdp ) , a Gram-negative pathogen that causes septicemic infections , which are among the most threatening diseases in mariculture . The toxin triggers apoptosis of host macrophages and neutrophils through a process that , in vivo , culminates with secondary necrosis of the apoptotic cells contributing to the necrotic lesions observed in the diseased animals . Here , we show that AIP56 is a NF-κB p65-cleaving zinc-metalloprotease whose catalytic activity is required for the apoptogenic effect . Most of the bacterial effectors known to target NF-κB are type III secreted effectors . In contrast , we demonstrate that AIP56 is an A-B toxin capable of acting at distance , without requiring contact of the bacteria with the target cell . We also show that the N-terminal domain cleaves NF-κB at the Cys39-Glu40 peptide bond and that the C-terminal domain is involved in binding and internalization into the cytosol . The NF-κB family of transcription factors is evolutionarily conserved and comprises five members ( NF-κB 1 ( p50 ) , NF-κB 2 ( p52 ) , RelA ( p65 ) , RelB and cRel ) , which form different combinations of homo- or hetero-dimers [1] . Under normal physiological conditions , NF-κB complexes remain inactive in the cytosol through association with the IκB proteins that mask the nuclear localization domains on NF-κB subunits . A variety of stimuli , including bacterial and viral products and cytokines , acting via cellular receptors such as Toll-like receptors ( TLRs ) , Interleukin-1 receptor ( IL-1R ) and TNF receptors ( TNFRs ) , trigger a signalling cascade that leads to phosphorylation and degradation of the inhibitory IκB proteins with rapid activation and transport of the NF-κB complexes to the nucleus , resulting in the up-regulation of inflammatory and anti-apoptotic genes [2] . NF-κB activation is considered to be the central initiating event of host responses to microbial pathogen invasion [2] . Therefore , it is not surprising that successful microbial pathogens have evolved complex strategies to interfere with NF-κB signalling . A number of pathogenic bacteria were recently found to interfere with this pathway by targeting different intermediates of the NF-κB activation cascade [2]–[4] . Photobacterium damselae piscicida ( Phdp ) is a Gram-negative bacterium that infects several warm water fish species worldwide and is recognized as one of the most threatening pathogens in mariculture [5]–[8] . In acute infections , a rapid septicemia develops and causes very high mortality [5] , [6] , [9] . Early descriptions of the histopathology of Phdp infection recognized the occurrence of cytotoxic alterations [5] , [10]–[16] that we found to result from pathogen-induced macrophage and neutrophil apoptotic death [17] , [18] by a process that uses mechanisms of the intrinsic and extrinsic apoptotic pathways [19] . The phagocyte destruction observed in Phdp infections occurs systemically and culminates in a secondary necrotic process with lysis of the apoptosing cells [17] , [18] , [20] . This leads to the impairment of host immune defences and to the release of the cytotoxic contents of the phagocytes , contributing to the formation of the necrotic lesions observed in the diseased animals . We have previously shown that phagocyte apoptosis observed in Phdp infections results from the activity of AIP56 , a plasmid-encoded exotoxin secreted by virulent strains , and that passive immunization with anti-AIP56 rabbit serum protects against Phdp infection [17] , [21] . These results implicated AIP56 as a key virulence factor of Phdp . However , the molecular target ( s ) of the toxin remained unidentified and nothing was known about its structure-function relationship . AIP56 is synthesized as a precursor protein with a cleavable N-terminal signal peptide that is removed during secretion , originating a 497-amino acid mature toxin with the conserved HEIVH zinc-binding motif within its N-terminal region [21] , similarly to tetanus neurotoxin [22] . The N-terminal region of AIP56 is homologous to NleC [21] , [23] , a type III secreted effector present in several enteric pathogenic bacteria , while the C-terminal region is highly similar to an uncharacterized hypothetical protein of Acrythosiphon pisum bacteriophage APSE-2 [24] and to the C-terminal portion of a hypothetical protein of the monarch butterfly Danaus plexippus ( Figure S1 ) . This suggested that AIP56 is a two domain protein , belonging to the group of A-B type toxins that includes diphtheria and tetanus toxins [23] , [25] , [26] . Recently , it was shown that NleC inhibits NF-κB activation and represses NF-κB-dependent transcription by cleaving NF-κB p65 within its N-terminal region [27]–[31] . Here , we show that AIP56 is a zinc-metalloprotease that cleaves NF-κB p65 and that its enzymatic and apoptogenic activities are correlated . In contrast to NleC , which is delivered into the host cell's cytosol through a type III secretion system , AIP56 is an A-B-type exotoxin with an N-terminal domain responsible for the proteolytic activity and a C-terminal domain involved in binding and internalisation into target cells . In order to clarify the role played by the zinc metalloprotease activity of AIP56 , a mutant ( AIP56AAIVAA ) containing a disrupted putative zinc-binding motif was produced . The oligomerization state and secondary structure content of the toxin were undisturbed by the introduced mutations ( Figure S2 ) and atomic absorption spectroscopy did not detect zinc in AIP56AAIVAA , while in AIP56 equimolar amounts of zinc ( 0 . 93±0 . 04 mol zinc/mol protein ) were present . When tested ex vivo , AIP56AAIVAA failed to induce apoptosis of sea bass phagocytes , whilst a large number of cells with apoptotic morphology were observed after treatment with AIP56 ( Figure 1A ) . These results indicate that an intact metalloprotease domain is essential for the apoptogenic activity of AIP56 . It is worth noting that the AIP56 concentrations used in the present work are biologically relevant , since they are similar to those detected in the plasma of infected fish ( Figure S3 ) . When incubated with sea bass cell lysates , AIP56 cleaved p65 with the appearance of a lower MW fragment ( Figure 1B ) . Proteolysis of p65 did not occur in cell lysates incubated with AIP56AAIVAA or with AIP56 in the presence of the metalloprotease inhibitor 1 , 10-phenanthroline ( Figure 1B ) . The p65 fragment was recognised by an antibody specific for a peptide located at the C-terminal region of p65 indicating that the AIP56-mediated p65 cleavage occurred within the N-terminal region , where the Rel-homology domain is located . To map the cleavage site , recombinant sea bass p65Rel domain ( sbp65Rel ) was incubated with the toxin . SDS-PAGE analysis showed that AIP56 cleaved recombinant sbp65Rel in vitro ( Figure 1C ) , and N-terminal sequencing of the cleaved fragment revealed that the cleavage occurred at the Cys39-Glu40 peptide bond , similar to what was described for NleC [27] . Experiments using in vitro synthesised 35S-labeled sea bass p65Rel domain ( sbp65Rel ) and three sbp65Rel mutants ( sbp65RelC39A , sbp65RelE40A and sbp65CE39-40AA ) showed that mutation of the evolutionarily conserved Cys39 had no effect on p65 cleavage by either AIP56 or NleC ( Figure S4 ) . However , mutation of the following Glu40 inhibited cleavage and double mutation of Cys39 and Glu40 completely abolished p65 proteolysis by AIP56 and NleC ( Figure S4 ) . To determine if cellular intoxication by AIP56 involves cleavage of NF-κB p65 , sea bass peritoneal leukocytes were incubated with wild type toxin or with AIP56AAIVAA mutant and p65 proteolysis assessed by Western blotting . Wild type AIP56 caused NF-κB p65 depletion , whilst AIP56AAIVAA was inactive ( Figure 1D ) . It has been reported that caspase-3 can cleave p65 [32] , [33] . To investigate whether caspases are involved in AIP56-dependent cleavage of p65 , cells were incubated with the toxin in the presence or absence of the pan-caspase inhibitor ZVAD-FMK ( Figure 1E ) , previously shown to block AIP56-induced apoptosis [19] . In these experiments , ZVAD-FMK was effective in protecting cells from AIP56-induced apoptosis ( data not shown ) , but did not affect NF-κB p65 cleavage ( Figure 1E ) , indicating that AIP56-mediated p65 depletion is a caspase-independent event . Taken together , the above results demonstrate that the metalloprotease activity of AIP56 is responsible for the cleavage of NF-κB p65 at the Cys39-Glu40 peptide bond . The primary structure of AIP56 suggests that this toxin comprises two functional domains and could be an A-B toxin with its two moieties linked by a single disulphide bond ( Figure S1 ) [23] . Therefore , in order to define domain boundaries within the toxin , limited proteolysis experiments were performed . SDS-PAGE analysis of AIP56 digested with chymotrypsin , trypsin or proteinase K revealed that the toxin is highly resistant to trypsin digestion , whereas chymotrypsin and proteinase K cleaved AIP56 into two major fragments with approximately 32 and 24 kDa ( Figure 2A ) . These two fragments were only detected upon treatment with the reducing agent DTT , suggesting that they are linked by a disulphide bridge ( Figure 2B ) . N-terminal Edman sequencing revealed that chymotrypsin cleavage occurred between Phe285 and Phe286 , in the amino-acid stretch flanked by the two unique cysteine residues ( Cys262 and Cys298 ) of AIP56 ( Figure 2C ) . Altogether , these results indicate that AIP56 is composed of two domains linked by a disulphide bridge . To better understand the function of the two AIP56 domains , constructs corresponding to the N- and C-terminal portions of the toxin ( AIP561–285 and AIP56286–497 , respectively ) were designed , taking into account the boundary defined by the chymotrypsin cleavage site ( Figure 2C ) . Purification of these two recombinant proteins using the experimental conditions used for the full-length toxin revealed that they display a major propensity to oxidize leading to the formation of DTT-sensitive dimers ( Figure S5A ) , a phenomenon that could have a functional impact and complicate subsequent analyses . Therefore , versions of the constructs with the single cysteine replaced by serines ( AIP561–285C262S and AIP56286–497C298S , respectively ) were produced . The mutants are undistinguishable from the non-mutant proteins , as assessed by CD ( Figure S5B and S5C ) with the N-terminal domain composed mainly of α-helices , whereas β–sheet is the predominant secondary structure of the C-terminal moiety ( Figure 2D ) . Furthermore , the weighted sum of the CD spectra of the N- and C-terminal domains reproduces the spectrum of the entire protein ( Figure 2D ) , indicating conservation of the native structure . To test the catalytic activities of the AIP56 N- and C-terminal domains , AIP561–285C262S and AIP56286–497C298S were incubated with fish leukocyte lysates ( Figure 1B ) or with in vitro translated 35S-labeled sbp65Rel ( Figure S5D ) . The C-terminal construct did not display catalytic activity , whereas the N-terminal domain cleaved p65 , similarly to the full-length toxin . However , neither changes in cellular p65 levels ( Figure 1D ) nor apoptosis ( Figure 3A ) were observed in sea bass leukocytes incubated with the N- or C-terminal truncate or with a mixture of both . This indicates that the two AIP56 domains are non-toxic and suggests that they need to be part of the same molecule to elicit a biological effect . The cytosolic location of NF-κB p65 could mean that the lack of toxicity of the N-terminal domain was related to its inability to enter the cells and reach its target . Hence , a strategy to deliver the N-terminal domain into the cell cytosol was designed . Chimeric proteins consisting of the N-terminal portion of Bacillus anthracis LF fused to the AIP56 protease domain ( LF11–263•AIP561–261 ) or to the C-terminal domain ( LF11–263•AIP56299–497 ) were produced . Intoxication assays were performed in the presence of PA , the receptor-binding subunit for LF [34] . In cells incubated with LF11–263•AIP561–261 the p65 levels were significantly reduced , confirming that LF11–263•AIP561–261 was successfully delivered into the cell cytosol , while no changes in p65 levels were observed in cells incubated with LF11–263•AIP56299–497 ( Figure 3B ) . Accordingly , LF11–263•AIP56299–497 did not display apoptogenic activity , while incubation with LF11–263•AIP561–261 resulted in an increased number of cells with apoptotic morphology ( Figure 3B ) , similar to what was observed in cells incubated with AIP56 . Thus , delivery of the AIP56 N-terminal domain into the cytosol reproduces the toxic effect of the full length toxin , confirming that this domain is responsible for the toxin's catalytic and apoptogenic activities . These results also suggest that the C-terminal domain of AIP56 is involved/required for entrance of the toxin into cells . To investigate this possibility , AIP56AAIVAA , AIP561–285C262S or AIP56286–497C298S were used in competition experiments with AIP56 . Both p65 cleavage and apoptosis were monitored in these experiments . AIP56286–497C298S and AIP56AAIVAA , but not AIP561–285C262S , were able to inhibit the apoptogenic activity of wild type AIP56 in a dose-dependent manner ( Figure 3C , left panel ) . Furthermore , AIP56AAIVAA and AIP56286–497C298S inhibited AIP56-mediated p65 degradation , whereas no effect could be observed when AIP561–285C262S was used as competitor ( Figure 3C , right panel ) . These results indicate that the C-terminal domain mediates binding of the toxin to the cell surface and entry into the cells . Results obtained in experiments using N- and C-terminal truncates of AIP56 suggested that the two domains must be part of the same molecule to display toxicity . In order to investigate if the two domains of the toxin bound by a disulphide bridge are able to intoxicate cells , we nicked the toxin with chymotrypsin . Nicking of the toxin and integrity of the disulphide bridge linking the two fragments were confirmed by reducing and non-reducing SDS-PAGE ( Figure S6A ) . Surprisingly , no changes in p65 cellular levels ( Figure 1D ) and no apoptosis ( Figure 4A ) were observed upon incubation of sea bass peritoneal cells with nicked toxin . Similar results were obtained using a reconstituted version of the toxin ( Figure 1D and 4A ) consisting of disulphide-bound AIP561–285/AIP56286–497 along with trace amounts of AIP561–285 and of AIP56286–497 homodimers and monomers ( Figure S6A ) . Although nicking abolished cellular toxicity , it did not induce major structural changes ( Figure S6B ) and only a 1°C decrease in Tm ( 39±0 . 13°C for AIP56 and 38±0 . 25°C for nicked AIP56; mean±SD of 16 measurements in four independent experiments ) was measured by DSF . More importantly , nicked AIP56 retained both proteolytic activity against p65 in vitro ( Figure 1B and 4B ) and cell binding ability , as indicated by the partial inhibition of the AIP56-mediated p65 cleavage and apoptosis in competition experiments ( Figure 4C ) . These results suggest that the integrity of the linker region between the two cysteine residues is needed for toxin internalization , in contrast to what is known for the diphtheria , tetanus and botulinum toxins , where nicking of the inter-cysteine loop is required for toxicity [26] , [35] . In tetanus and botulinum neurotoxin type A , it has been shown that the disulphide bridge is essential for neurotoxicity [36] , [37] . We found that disruption of the disulphide bridge linking Cys262 and Cys298 of AIP56 by alkylation ( Figure S6D ) did not affect the catalytic activity of the toxin in vitro ( Figure 4B ) , but partially compromised its toxicity ( Figure 4D ) , suggesting that in AIP56 the disulphide bridge plays a role in the intoxication process but is not an absolute requirement for toxicity . In this study , we report the structural and functional characterization of AIP56 as an A-B type bacterial exotoxin that cleaves NF-κB p65 . Considering the anti-apoptotic functions of NF-κB , and in particular , of its p65 subunit [38]–[40] , AIP56-mediated depletion of NF-κB p65 likely explains the disseminated phagocyte apoptosis observed in Phdp infections that contributes to subvert the host immune response and determines the outcome of the infection [17] , [19] , [21] . This adds to a general theme of host-pathogen interaction that has recently emerged , consisting in the induction of apoptosis of the host immune cells to the pathogen advantage [41]–[43] . Here we demonstrate that , similar to the anthrax lethal factor and to the clostridial neurotoxins [44] , [45] , AIP56 is a zinc-endopeptidase but with a catalytic activity towards NF-κB p65 . Furthermore , we show that AIP56 is organized into two distinct domains linked by a single disulphide bond . The N-terminal domain of AIP56 harbours the catalytic activity of the toxin and cleaves NF-κB p65 at the Cys39-Glu40 peptide bond , within the p65 N-terminal Rel homology domain , where several key residues of p65 known to be involved in DNA interaction are located [46]–[48] . In the last decade , several reports revealed that Cys38 of human p65 ( Cys39 in sea bass p65 ) interacts with the phosphate backbone of NF-κB binding sites [46] , that its oxidation and nitrosylation inhibit DNA binding [49] and that it is targeted by several inhibitors of NF-κB with anti-inflammatory and/or anticancer properties [50]–[60] . More recently , it was shown that hydrogen sulphide-linked sulfhydration of Cys38 of human p65 plays a key role in regulating the anti-apoptotic actions of NF-κB [40] . Therefore , cleavage of sea bass p65 by AIP56 disrupts a segment crucial for DNA interaction . Considering that the proteolytic activity of AIP56 towards p65 is similar to the one previously described for NleC ( both proteases cleave p65 at the same peptide bond ) , and based on the observation that p65 cleavage by NleC compromises NF-κB dependent transcription [27] , [28] , [30] , it is likely that AIP56 also affects NF-κB transcriptional activity . Although in many cell types down-regulation of NF-κB is not sufficient to trigger apoptosis , it is widely recognised that cells with inactivated NF-κB are more prone to commit suicide in response to different stimuli , including TNF-α and TLRs ligands [38] , [61]–[63] , and that the anti-apoptotic actions of NF-κB can be largely attributed to its p65 subunit [38]–[40] . In the context of bacterial infections , inhibition of NF-κB function usually leads to impairment of the inflammatory responses [2] . The induction of apoptosis by bacterial effectors through interference with NF-κB activity has also been described , but is a far less common scenario . Examples are Yersinia YopP/J [64]–[66] and Aeromonas salmonicida Aop [67] , both inhibiting the degradation of the inhibitory IκB proteins [64] , [68]–[71] , and V . parahaemolyticus protein VP1686 that interacts with and suppresses DNA binding activity of NF-κB [72] . It remains to be determined whether AIP56-mediated depletion of p65 is sufficient to induce apoptosis , in resemblance to what has been suggested for the macrophage apoptosis induced by V . parahaemolyticus type III secreted effector VP1686 [72] , or if it requires an additional stimulus . Almost all bacterial effectors that have been described to target NF-κB signalling are injected directly into the host cell cytosol by type III or type IV secretion systems ( see reviews by [2] , [4] ) . In contrast , we have found that the AIP56 N-terminal metalloprotease can only act when linked to a C-terminal binding domain that , by analogy with other A-B toxins , may assist the protease domain in its membrane translocation into the cytosol [34] , [73] , [74] . Bacterial A-B toxins are often secreted as a single polypeptide chain that is cleaved into the disulphide-bound A and B domains [26] , [74] . In these toxins , proteolytic nicking and integrity of the disulphide bond linking the A and B domains are essential for toxicity [19] , [35] , [36] , [75]–[77] . In contrast , AIP56 toxicity is abolished by proteolytic nicking and only mildly compromised by disruption of the disulphide bridge by alkylation . Considering that nicked AIP56 retains the ability to interact with the cell membrane , these observations suggest that the linker region ( between Cys262 and Cys298 ) is involved in translocating the toxin into the host cell cytosol . The decreased toxicity resulting from alkylation suggests that the integrity of the disulphide bond is important , although not absolutely required , for AIP56 intoxication . The disulphide bond may be involved in stabilizing the spatial relationship between the domains . In addition , that bond is hydrophobic and polarizable and its alkylation can have implications in membrane insertion , as reported for tetanus and botulinum neurotoxins [78] . AIP56 is synthesised as a single polypeptide and , contrary to what has been reported for most A-B toxins , there is no evidence of proteolytically processed toxin in the bacterial culture supernatants or in the serum of infected fish [21] . Furthermore , despite several attempts , we were unable to detect proteolytic processing of AIP56 upon its interaction with host cells . If AIP56 needs to be processed in order to exert its effect , the lack of detection of processed toxin may result from a very small amount of processed toxin ( not detectable in our experiments ) being sufficient to intoxicate the cells , similarly to what was described for other toxins [79] . Alternatively , after endocytosis , unprocessed AIP56 may be translocated into the cytosol as described for Pseudomonas exotoxin A [80] , [81] or may localize in an endomembrane ( e . g . endosomal membrane ) with the catalytic domain facing the cytosolic compartment where it can interact with and cleave p65 . Studies aiming at discriminating between these hypotheses will be developed in the future . It is now recognised that horizontal transfer of entire genes or portion of genes plays a key role in generating diversity in pathogens by allowing them to acquire novel phenotypic characteristics . Indeed , there are several examples of bacterial genes with a mosaic structure , composed of diverse segments with different origins [82]–[87] . The structure of AIP56 suggests that the toxin has a chimeric structure , having an N-terminal catalytic domain highly identical to the type III effector NleC and a C-terminal domain homologous to a hypothetical protein of the bacteriophage APSE-2 . The actual transfer events that gave rise to such a chimeric protein toxin remain to be disclosed . The AIP56 catalytic domain and NleC have the same NF-κB p65 cleavage activity . However , NleC requires a type III secretion machinery for activity , while AIP56 has an intrinsic ability to reach the cytosol , due to the presence of the additional C-terminal domain that functions as a “delivery module” . This difference may have relevant implications when considering the use of both pathogen-derived molecules as therapeutic agents in situations associated with uncontrolled activation of NF-κB such as inflammatory diseases and cancer . This study was carried out in accordance with European and Portuguese legislation for the use of animals for scientific purposes ( Directive 86/609/EEC; Decreto-Lei 129/92; Portaria 1005/92 ) . The work was approved by Direcção Geral de Veterinária , the Portuguese authority for animal protection ( ref . 004933 , 2011-02-22 ) . Sea bass ( Dicentrarchus labrax ) , were kept in a recirculating , ozone-treated salt-water ( 25–30‰ ) system at 20±1°C , and fed at a ratio of 2% body weight per day . Fish were euthanized with 2-phenoxyethanol ( Panreac; >5 ml/10 L ) . DNA coding sequences were cloned into NcoI/XhoI restriction sites of pET-28a ( + ) ( Novagen ) as described in Supporting Information . Mutants were generated by site directed mutagenesis using QuickChange Site-Directed Mutagenesis Kit ( Stratagene ) following manufacturer's instructions . Recombinant His-tagged proteins were expressed in E . coli BL21 ( DE3 ) cells . AIP56 , AIP561–285 , AIP56286–497 , AIP56286–497C298S , NleC , LF11–263•AIP561–261 and LF11–263•AIP56299–497 were purified from the soluble fraction of induced bacteria by metal-affinity chromatography . After this step , AIP56 , AIP561–285 , AIP56286–497 were subjected to anion exchange chromatography , whereas AIP56286–497C298S and NleC were subjected to size exclusion chromatography . AIP56AAIVAA and AIP561–285C262S were purified from inclusion bodies by metal-affinity chromatography under denaturing conditions , refolded by dialysis against sea bass PBS ( sbPBS; phosphate buffer saline with osmotic strength adjusted to 322 mOsm ) with 10% ( v/v ) glycerol and purified by size exclusion chromatography . For reconstitution of AIP56 , AIP561–285 and AIP56286–497 were mixed in equimolar amounts in 8 M urea , 1 mM DTT and refolded by extensive dialysis against sbPBS . Nicked AIP56 was obtained by limited proteolysis with 25 µg/ml chymotrypsin , as described below , followed by metal-affinity chromatography purification . To prepare alkylated toxin , 63 µM AIP56 in 20 mM Tris-HCl pH 8 . 0 , 200 mM NaCl , 10 mM DTT was incubated with 5 mM iodoacetamide ( Sigma ) for 30 min at RT and dialysed against 20 mM Tris-HCl pH 8 . 0 , 200 mM NaCl . Sea bass NF-κB p65 REL homology domain ( sbp65Rel ) was purified from the soluble fraction of induced bacteria by metal-affinity chromatography . Untagged 35S-labeled sbp65Rel and sbp65Rel mutants ( sbp65RelC39A , sbp65RelE40A , and sbp65C39E40AA ) were produced using the TNT T7 Quick Coupled transcription/Translation kit ( Promega ) , in the presence of Redivue™ L-[35S] methionine ( specific activity of 1000 Ci/mmol ) . AIP56 at 0 . 6 mg/ml in 10 mM Tris-HCl pH 8 . 0 , 200 mM NaCl was incubated with 0 . 25 to 25 µg/ml trypsin , chymotrypsin or proteinase K ( molar ratios of protease:AIP56 of approximately 1∶10 to 1∶1000 ) for 30 min on ice . Proteases were inactivated by addition of PMSF to a final concentration of 250 µg/ml . Digests were analysed by reducing and non-reducing SDS-PAGE . The two major chymotrypsin digestion fragments were subjected to N-terminal sequencing . Far UV CD spectra were acquired on an Olis DSM 20 circular dichroism spectropolarimeter controlled by the Globalworks software . Each spectrum is the average of three scans collected at 20°C with a 0 . 2 mm path length cuvette and with an integration time of 4 seconds . Proteins were dissolved in 10 mM Tris-HCl , 50 mM NaCl , pH 8 . 0 and concentrations were determined by absorbance measurements . Analysis of the protein secondary structure was performed using the Globalworks software algorithm . Sea bass peritoneal leukocytes were obtained as previously described [19] and used at a density of 2×106 cells/ml . The peritoneal population of cells consists of approximately 70% macrophages and 20% neutrophils with the presence of small numbers of eosinophilic granular cells , lymphocytes and erythrocytes [19] . Cells were incubated for 4 h at 22°C with AIP56 or AIP56 derived proteins at the indicated doses . Where indicated , the cells were pre-treated for 30 min at 22°C with 25 µM of the pan-caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp ( O-Me ) fluoromethyl ketone ( Z-VAD-FMK ) . In experiments using LF chimeric proteins , cells were incubated with the indicated concentrations of LF11–263•AIP561–261 or LF11–263•AIP56299–497 with or without 10 nM of anthrax protective antigen ( PA ) obtained as described [88] . Mock- and AIP56-treated cells were used as controls . Apoptosis was assessed as described [18] , by light microscopy morphological analysis of cytospin preparations stained with Hemacolor ( Merck ) after labelling neutrophils using Antonow's technique [89] , [90] . AIP56AAIVAA , AIP561–285C262S and AIP56286–497C298S were tested for their ability to inhibit AIP56's apoptogenic activity and AIP56-mediated p65 cleavage . Cells were pre-incubated for 15 min on ice with different concentrations ( 350 nM to 3 . 5 µM ) of AIP56AAIVAA , AIP561–285C262S , AIP56286–497C298S or nicked AIP56 , followed by incubation for further 15 min on ice with 8 . 75 nM of AIP56 in the presence of the competitors . Unbound proteins were removed by washing with ice cold supplemented L-15 medium [19] and the cells incubated at 22°C for 4 h . Statistical analysis was performed using a randomized block design , where fish are treated as blocks and the concentration of the treatments/competitors as a factor . The data , percentage of apoptotic cells , have been transformed using the arcsine transformation . Post-hoc comparisons were performed using the Tukey's Honest Significant Difference test . Significance was defined for p<0 . 05 .	The apoptosis inducing protein of 56 kDa ( AIP56 ) is a key virulence factor secreted by Photobacterium damselae piscicida ( Phdp ) , a Gram-negative bacterium that causes septicaemic infections in economically important marine fish species . It is known that AIP56 induces massive destruction of the phagocytic cells of the infected host , allowing the extracellular multiplication of the bacteria and contributing to the genesis of the pathology . Here we show that AIP56 acts by cleaving NF-κB p65 . The NF-κB family of transcription factors is evolutionarily conserved and plays a central role in the host responses to microbial pathogen invasion , regulating the expression of inflammatory and anti-apoptotic genes . Pathogenic bacteria have evolved complex strategies to interfere with NF-κB signalling , usually by injecting protein effectors directly into the cell's cytosol through bacterial secretion machineries that require contact with host cells . In contrast , AIP56 acts at distance and has an intrinsic ability to reach the cytosol due to the presence of a C-terminal domain that functions as “delivery module . ”	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "gram", "negative", "biology", "microbiology", "host-pathogen", "interaction", "bacterial", "pathogens", "pathogenesis" ]	2013	The Apoptogenic Toxin AIP56 Is a Metalloprotease A-B Toxin that Cleaves NF-κb P65
Fungal infection has become one of the leading causes of hospital-acquired infections with high mortality rates . Furthermore , drug resistance is common for fungus-causing diseases . Synergistic drug combinations could provide an effective strategy to overcome drug resistance . Meanwhile , synergistic drug combinations can increase treatment efficacy and decrease drug dosage to avoid toxicity . Therefore , computational prediction of synergistic drug combinations for fungus-causing diseases becomes attractive . In this study , we proposed similar nature of drug combinations: principal drugs which obtain synergistic effect with similar adjuvant drugs are often similar and vice versa . Furthermore , we developed a novel algorithm termed Network-based Laplacian regularized Least Square Synergistic drug combination prediction ( NLLSS ) to predict potential synergistic drug combinations by integrating different kinds of information such as known synergistic drug combinations , drug-target interactions , and drug chemical structures . We applied NLLSS to predict antifungal synergistic drug combinations and showed that it achieved excellent performance both in terms of cross validation and independent prediction . Finally , we performed biological experiments for fungal pathogen Candida albicans to confirm 7 out of 13 predicted antifungal synergistic drug combinations . NLLSS provides an efficient strategy to identify potential synergistic antifungal combinations . In recent years , fungal infection has become one of the leading causes of hospital-acquired infections with high mortality rates due to growing populations of patients with weakened immune systems , for example due to cancer , organ transplant or Acquired Immune Deficiency Syndrome ( AIDS ) . In these patients , infections caused by Candida , Aspergillus and Cryptococcus neoformans fungi strains may take the form of potentially lethal blood stream infections , lung infections and other infections . For example , Candida causes candidiasis , which becomes the fourth most common fungal blood stream infection among hospitalized patients in the United States according to the Centers for Disease Control & Prevention . Unfortunately , fungal infections that include Candida albicans have become resistant to current drug treatments . Therefore , there is an urgent need to develop new therapies to overcome the drug resistance and kill C . albicans . Drug combinations have been widely used to overcome drug resistance and treat complex disease such as cancer and infectious diseases [1–4] . Drug combinational treatment could inhibit new multiple targets and thus provide the opportunity for overcoming drug resistances of infectious fungi [5–7] . The potential molecular mechanism underlying this is that biological systems are less able to compensate for the simultaneous activity of two or more drugs [1 , 5 , 8 , 9] . Indeed , we have seen growing enthusiasm over the development of synergistic drug combinations in academia , as well as the pharmaceutical industry . For example , CRx-102 is a novel synergistic drug candidate combination comprised of dipyridamole and low-dose prednisolone . This drug combination can be used for the treatment of osteoarthritis ( OA ) and has already completed Phase II study in Knee OA [10] . Also , moduretic , a combination of Amiloride and Hydrochlorothiazide , is used to treat patients with hypertension [11 , 12] . The use of synergistic drug combinations can increase treatment efficacy and decrease drug dosage to avoid toxicity . It also has been pointed out that off-target effects could be overcome by drug combinations [13] . These advantages have increasingly driven researchers towards the search for safe and effective combinatorial drugs [5–7 , 14] . Traditionally , effective drug combinations have been identified through experimentally screening all possible combinations of a pre-defined set of drugs [5 , 15] . Given the large number of drugs , experimental screens of pairwise combinations of drugs will be cost expensive , time consuming and labor intensive . For example , given n drugs , there will be n ( n−1 ) /2 pairwise drug combinations and many more higher-order combinations . Furthermore , new drugs will be produced every year , therefore , the number of possible drug combinations will exponentially increase [15] . Since a comparatively small number of compounds will provide a very large number of combinations [6] , experimentally testing all the possible drug combinations would pose a formidable challenge in terms of cost and time . Even when high-throughput screens are adopted , limited drug combination experiments would only sample a small fraction of so many candidate drug combinations . Thus , it is not easy to identify optimal drug combinations using the experimental screen approach [16] . To overcome this problem , we intend to develop a method that computationally ‘screens’ synergistic drug combinations and identifies optimal drug pairs for treating drug resistance of fungus-infected diseases . Our computational methods can select the most promising drug combinations for rigorous validation through biological experimentation , thus saving time and money . In this sense , this method could guide the drug combinations experiments and also benefit the understanding of mechanisms underlying synergistic drug combinations . Previous research was mostly focused on defining the concept of synergy , quantitatively measuring dose-effect curves , and determining whether or not a given drug combination could achieve synergistic effect according to the definitions of the synergy and experiment results [1] . Ever since Loewe proposed the Loewe additive model to describe synergy drug combination in 1928 , numerous researchers have devoted to drug combination analysis [1 , 14 , 17–22] . Loewe defined Loewe additive equation as follows to determine whether or not the given drug combination would result in a synergistic effect [17 , 18]: ( D ) 1 ( Dx ) 1+ ( D ) 2 ( Dx ) 2=1 Variables in the numerator are the dosage of each drug ( drug ‘1’ and drug ‘2’ ) when these two drugs are combined and x% is the inhibition rate with this concentration combination . Variables in the denominator are the dosage of each drug that can inhibit the system by x% . The left-hand side of this equation is less than 1 and more than 1 mean Loewe synergism and Loewe antagonism , respectively . Then , Bliss defined the expected combination effect as IMult = IX + IY − IXIY , where IX and IY are single drug inhibition at concentrations X and Y [19] . Berenbaum proposed the highest single agent ( HSA ) model , which defined the expected response as IHSA = max{IX , IY} , where IX and IY are defined in a manner similar to that of the Bliss model [23] . Chou and Talalay proposed the median-effect equation [21 , 24 , 25] , the Combination Index ( CI ) -Isobologram equation [20 , 21] , and the dose-reduction index equation [21 , 26] for quantitative determination of drug combination interactions . In their scheme , CI<1 , = 1 , and >1 indicate synergism , additive effect , and antagonism , respectively [1] . Greco also established a new method , termed universal response surface approach ( URSA ) , for the quantitative assessment of drug interactions [27] . However , all aforementioned models only determine whether or not a given drug combination could achieve synergistic effect and can’t be used to predict potential synergistic drug combinations . In recent years , some methods have been developed to decrease the number of drug combination experiments . Jansen et al . [14] used chemogenomic profiles to identify potential combinatorial drugs . Firstly , sensitivity-based chemogenomic profile data generated from the literature and profiling experiments were analyzed . Then , any given compound pair that had chemogenomic profiles similar to the known synergy pairs was considered as potential antifungal synergy candidates . Chen et al . [28] combined fractional factorial design and stepwise regression to dramatically reduce the time of experiments required to identify synergistic drug combinations . However , these two methods both strongly rely on biological experimental results . Li et al . [29] defined the parameters of topology score and agent score to evaluate the synergistic relationship for given drug combinations and further established the algorithm termed NIMS to uncover potential synergistic drug combinations on a large scale . Zhao et al . [30] represented drugs based on a set of their properties and further developed a novel computational method to prioritize candidate drug combinations by integrating molecular and pharmacological data . Huang et al . [31] integrated clinical side-effect information and the drug label to predict drug combination and demonstrated that three FDA black-box warned serious side-effects contributed mostly to the prediction performance . Huang et al . [32] developed a computational synergistic drug combination prioritization tool ( DrugComboRanker ) based on drug functional network construction and partition . Yin et al . [33] shown drug synergy or antagonism to be a property of target-related network topology and analyzed several basic synergistic and antagonistic motifs to indicate that designing novel synergistic drug combinations based on network topology could be promising . Iwata et al [34] integrated known synergistic drug combinations from the Orange Book and KEGG DRUG database , drug-target interactions , and drug Anatomical Therapeutic Chemical Classification System codes to construct a sparsity-induced classifier for the potential synergistic drug combination inference . Recently , considering the important fact that synergistic drug combination may act on the same pathway through different drug targets , Chen et al . [35] included the information of systematic pathway-pathway interactions and further developed a novel network-based synergistic drug combination prediction model . However , only computational models have been developed and no experimental validation could be found in aforementioned seven studies . Thus , in this study , we developed a novel algorithm , called Network-based Laplacian regularized Least Square Synergistic drug combination prediction method ( NLLSS ) , to conduct computational ‘screens’ by integrating several types of information such as known synergistic drug combinations , unlabeled combinations ( all the drug combinations without known synergistic evidences ) , drug-target interactions , and drug chemical structures . NLLSS obtained excellent performance in both cross validation and independent antifungal drug combinations prediction . Furthermore , we experimentally confirmed 7 out of 13 predicted antifungal synergistic drug combinations for fungal pathogen Candida albicans . These combinations could provide new treatments for overcoming fungal drug resistance . Finally , NLLSS provides an efficient strategy to find potential synergistic antifungal combinations by exploring new indications of existing antifungal drugs . Further , NLLSS could also be used for predicting synergistic drug combinations for treating other diseases . First , we investigated hundreds of studies on drug combinations and selected 69 compounds involved in antifungal drug combination experiments ( see S1 Table ) . Then we searched literatures with the keywords ‘synergy’ , ‘synergic’ , ‘synergistic’ , ‘synergism’ , ‘interaction’ and ‘combination’ in the PubMed , Google Scholar and Web of Knowledge and collected 75 experimentally confirmed synergistic antifungal drug combinations ( dataset 1 , see S2 Table ) . Therefore , all the antifungal compounds involved in antifungal drug combination experiments are considered in this study . We do not require they must have known synergistic partners . We also classified these compounds into principal drugs and adjuvant drugs according to the following rules . If one compound in the synergistic combination shows activity in the antifungal assay , but the other does not , as reported before , then the former compound is considered as the principal drug , and the latter is considered as the adjuvant drug . If both compounds in the synergistic combination show activity in the antifungal assay , or neither one shows activity , as reported before , then these two compounds are considered as both principal and adjuvant drug . If one compound does not have antifungal synergistic effect with any other compound , then this compound is classified according to its antifungal activity . To further confirm the predictive ability of NLLSS , we further obtained two other antifungal drug combination datasets from the dataset 1: synergistic antifungal drug combinations against Candida albicans ( dataset 2 , the drugs are the same as the ones in dataset 1 and the combinations are the ones in dataset 1 against Candida albicans ) and synergistic azole drug combinations ( dataset 3 , the drugs are the ones in dataset 1 which are azole drug and the combinations are the azole drug combinations in dataset 1 ) . Related information on these two synergistic drug combination datasets can be obtained from the supplementary materials ( See S3–S6 Tables: drugs and synergistic drug combinations in dataset 2 and 3 , respectively ) . Statistics of three drug combination datasets were listed in Table 1 , including the number of drugs , principal drugs , adjuvant drugs , known synergistic combinations ( A ) , and drug pairs without known synergistic relationship ( B ) and the ratio A/B . In this study , NLLSS was developed based on the framework of Laplacian Regularized Least Square ( LapRLS ) , which required drugs in the same combination must be divided into principal drug and adjuvant drug . However , according to the classification rules mentioned before , many drugs have been considered as both principal and adjuvant drugs . This fact ensure that we still can obtain plenty of potential synergistic drug combinations composed of two principal drugs or two adjuvant drugs . Chemical structure similarities between compounds were calculated by SIMCOMP [36] based on chemical structure information from the DRUG and COMPOUND Sections in the KEGG LIGAND database [37] . The similarity calculated from SIMCOMP is a global score based on the ratio between the size of the common substructures and the size of the union structures [36] . Applying this operation to all compound pairs , chemical structure similarity scores between principal ( adjuvant ) drugs can be obtained Target proteins of all the drugs in three drug combination datasets were obtained from the Drug Bank database [38] and related literatures . The drug-protein interactions in three datasets were shown in S7–S9 Tables , respectively . Ketoconazole , Fluconazole , Voriconazole , Posaconazole , Itraconazole , Terbinafine , Flucytosine , Radicicol , Disulfiram , Lovastatin , Geldanamycin , Caspofungin and Micafungin were purchased from a local pharmaceutical company . Amphotericin B , Beauvericin and FK506 were purchased from Sigma . NLLSS assumes that principal drugs which obtain synergistic effect with similar adjuvant drugs are often similar and vice versa , which is referred to as the similar nature of drug combinations ( see the examples in Fig 1 and S1 Fig ) . Based on this assumption , the following conclusion can be deduced: principal drugs which obtain synergistic effect with the same adjuvant drug are often similar and vice versa . In this paper , similarity between two drugs is established on the basis of drug chemical structure , drug-target interactions , and known synergistic drug combinations . The similar nature of drug combination was further formulated into two classifiers based on the framework of LapRLS in the principal and adjuvant drug space , respectively . The classifiers in the principal drug space and adjuvant drug space both considered all possible drug pairs . They both used the information of known synergistic drug combinations and unlabeled drug combinations . The difference between these two classifiers was that they adopted the different drug similarity . Classifier in the principal drug space only used principal drug similarity . Correspondingly , classifier in the adjuvant drug space only used the information of adjuvant drug similarity . Finally , two classifiers were combined into a single classifier to give a final predictive result . Based on the model , a score to assess how likely two drugs will obtain synergistic effect can be obtained . Drug combination pairs with high scores can be expected to have a high probability of obtaining synergistic effect when combined , thus having priority in subsequent biological experiments and , in turn , reducing the cost of identifying potential synergistic drug combinations . The flow chart of NLLSS is shown in Fig 2 . NLLSS first calculates the similarity between drugs . In this model , the similarity between two drugs depends on three factors: drug chemical structure similarity , drug target similarity , and drug synergistic similarity . Drug chemical structure similarity can be obtained by SIMCOMP as noted before . We defined principal ( adjuvant ) drug chemical structure similarity matrix as SCP ( SCA ) . Next we want to extract the information from drug-target interactions for the measurement of drug similarity . The underlying assumption made here was that two drugs are similar if they share more common target proteins ( see Fig 3 ) . Based on this assumption , the principal ( adjuvant ) drug target similarity matrix STP ( STA ) was defined . The entity of the matrix was the number of target proteins shared by two drugs . Third , we extracted the information from known drug synergistic combinations , assuming that if two principal ( adjuvant ) drugs obtain synergistic effect with more common adjuvant ( principal ) drugs , they will have greater similarity ( see S2 Fig ) . The principal ( adjuvant ) drug synergistic similarity matrix was defined as SSP ( SSA ) . The entity of the matrix was the number of common adjuvant ( principal ) drugs which have synergistic effect with two principal ( adjuvant ) drugs . Drug-target similarity matrix and drug synergistic similarity matrix must be normalized . For STP , we defined a diagonal matrix DTP such that DTP ( i , i ) was the sum of row i of STP . We set STP'= ( DTP ) −1/2STP ( DTP ) −1/2 which yielded a symmetric matrix where STP' ( i , j ) =STP ( i , j ) /DTP ( i , i ) DTP ( j , j ) . A similar operation was applied to another three matrices . Now , the principal drug similarity matrix SP can be obtained by linear combination as follows: SP=αPSCP+βPSTP'+γPSSP'αP+βP+γP where combinatorial coefficient means the weight of various similarity measures for the final integrated principal drug similarity . Similarly , the adjuvant drug similarity matrix SA can be obtained by the following form . Here , we have adopted the method of weighted averaging for the drug similarity integration , which means all the drug similar measures have equal weight ( i . e . αP = βP = γP = 1/3 , αA = βA = γA = 1/3 ) for the final drug similarity matrix . For the employment of the LapRLS , Laplacian operation must be applied to the similarity matrix . The diagonal matrices DP and DA were defined such that DP ( i , i ) and DA ( i , i ) were the sum of row i of SP and SASA , respectively . The normalized Laplacian matrices were defined as follows: LP= ( DP ) −1/2 ( DP−SP ) ( DP ) −1/2LA= ( DA ) −1/2 ( DA−SA ) ( DA ) −1/2 Let matrix Y represents prior synergistic drug combination information . If principal drug i and adjuvant drug j were known to produce synergistic effect , then Y ( i , j ) = 1; otherwise Y ( i , j ) = 0 . The aim was to obtain a continuous classification function , which reflected the probability that two drugs could obtain synergistic effect when combined . Intuitively , it is anticipated that when similar principal ( adjuvant ) drugs are combined with the same adjuvant ( principal ) drug , these combinations can obtain similar synergistic probability scores . Also this classification function should comply with prior synergistic information . LapRLS defines a cost function and wants to minimize this cost function in order to obtain an optimal classification function . The classification function was composed of optimal functions in the principal drug space and adjuvant drug space . We first address how to obtain optimal classification function in the principal drug space . Cost function was defined as follows: Fp=arg minFP[‖Y−FP‖F2+ηP‖FpTLPFP‖F2] Where ‖ . ‖F is Frobenius norm and ηP is the trade-off parameter in the principal drug space . Then , we can get the optimal classification function [39 , 40] as follows: FP=SP ( SP+ηPLPSP ) −1Y We also can get the optimal classification function in adjuvant drug space in a similar manner: FA=SA ( SA+ηALASA ) −1YT where ηA is the trade-off parameter in the adjuvant drug space . We set these two trade-off parameters as 0 . 3 in this study according to previous literatures [40–43] . Hence , the classification function can be obtained by combining the prediction results in both principal and adjuvant drug space , as follows: F=FP+ ( FA ) T2 We converted the probability of candidate drug combinations to the Rank Probability ( RP ) . The probabilities of combinations were ranked in ascending order , and each candidate combination obtained Rank ( R ) . Rank Probability ( RP ) of a drug combination was calculated by Rank ( R ) divided by the total number of candidate drug combinations ( N ) . In this case , the most probable synergistic drug combination will get the RP of 1 . Candida albicans SC5314 was used as a test strain for antifungal and synergistic antifungal bioassay . All procedures were described previously [16] . The experiments were carried out in flat bottom , 96-well microtiter plates ( Greiner ) , using a broth microdilution protocol modified from Clinical and Laboratory Standards Institute M-27A methods [44] . Overnight cultures were selected to prepare the strain suspension with RPMI 1640 medium ( Gibco ) at the concentration of 1×104 cells/mL counted by hemocytometer . To the test wells in 96-well plates , 2 μL of the samples were added , followed by an additional 80 μL of the strain suspension . The test plates were incubated at 35°C . The antifungal MICs were determined by measuring and comparing the optical densities of the positive control and test wells at different time points . For the synergistic antifungal assay , checkerboard assay was used , and beauvericin combined with ketoconazole served as positive control [16] . The MICs were determined by measuring and comparing the optical densities of the positive control and test wells at different time points . We evaluated the predictive performance of NLLSS using leave-one-out cross validation ( LOOCV ) . To do so , each known synergistic drug combination was treated as a test dataset in turn , while the remaining known synergistic drug combinations were used as the training dataset . First , we calculated the enrichment score to measure the performance of NLLSS . When LOOCV is implemented , if there are n candidate drug combinations without known synergistic evidences , the enrichment score is calculated by dividing n/2 by the rank of the left-out drug combination among candidate drug combinations . For example , if NLLSS gives the left-out known synergistic drug combination the highest ranking ( ranked 1st in the candidate drug combinations ) , there would be an enrichment score of n/2 . Furthermore , if the left-out known synergistic drug combination is ranked by random , it would have the rank of n/2 and therefore have an enrichment score of 1 . Therefore , enrichment score could represent the difference between prediction accuracy obtained by NLLSS and random . Here , the average of enrichment scores for all the left-out known combinations is calculated for the final evaluation . Next , receiver-operating characteristic ( ROC ) curve was used as another evaluative measure . The ROC curve plots the true-positive rate ( TPR ) versus the false-positive rate ( FPR ) . The area under the ROC curve ( AUC ) was calculated to reflect predictive accuracy . Here , the ROC curves of NLLSS based on the combination of two classifiers and only based on a single classifier in three drug combination datasets were compared ( Fig 4 , S3 Fig , S4 Fig ) . The AUCs for the combination of two classifiers in three dataset were 0 . 9054 , 0 . 8963 , and 0 . 8819 , respectively , which shows reliable ability to predict potential synergistic drug combinations . The AUCs for the classifier in the principal and adjuvant drug space were significantly inferior to the combined classifier , which shows the reasonableness of combining the classifiers in the principal and adjuvant drug space . Also , the comparisons of combined and single classifiers in terms of fold enrichment score in the three drug combination datasets were shown in Fig 5 , still illustrating the prefect performance of NLLSS . Performance comparison between NLLSS with current state-of-the-art computational models can’t be reasonably implemented . Different synergistic drug combination benchmark datasets and drug data sources for drug similarity calculation have been used in the different studies . For example , we only paid attention to antifungal synergistic drug combinations and integrated drug chemical structure , drug-target interactions , and known synergistic drug combinations to calculate drug similarity . In order to compare different computational models based on the same benchmark dataset , we must obtain different data sources of all the drugs in the benchmark dataset , such as drug-target interactions , drug side-effect information , and drug chemical structure . It is difficult to obtain all the datasets . Furthermore , some studies didn’t use known synergistic drug combinations to predict potential ones . It is unreasonable to directly compare them with NLLSS . The AUCs of NLLSS was 0 . 9054 , which has been better than AUCs reported in the previous studies . Furthermore , in order to confirm NLLSS is robust to the training sample selection , we implemented 10-fold , 5-fold , and 3-fold cross validation ( CV ) in all the three datasets , respectively ( See Table 2 ) . Here , all the known synergistic combinations were randomly divided into 10-fold , 5-fold , and 3-fold , which means 90% , 80% , and 66 . 67% of the known synergistic combinations were regarded as the training samples for model learning and the other 10% , 20% , and 33 . 3% were used as test samples for performance validation , respectively . Considering the potential influence caused by sample division , we implemented 100 different random divisions and calculated the mean and the standard deviation of all the obtained AUCs . As the results listed in Table 2 , NLLSS has a reliable and robust performance in all the validation schemas . To experimentally validate the predicted combinations which have potential synergistic antifungal activities , we tested all combinations in vitro on the leading human pathogen Candida albicans . Here , we implemented experiments for the top 10 potential drug combinations in all the three datasets ( See Table 3 ) . The synergis tic activities were judged by fractional inhibitory concentration index ( FICI ) values , which were calculated by comparing MICs in combinations with MICs of the each drug used alone ( See S13 Table ) at different time points . Biological experimental results indicated that 6 , 5 , and 6 out of the top 10 potential combinations in three datasets did indeed obtain antifungal synergistic effect ( See Table 3 ) . After the removal of duplicate combinations , we found 7 synergistic combinations ( Fig 6 ) and proved that 6 groups were nonsynergistic combinations ( S5 Fig ) in total . Considering the validated 7 synergistic combinations are totally new combinations , which have not been reported in any publicly published literatures , this prediction accuracy could be considered high . To identify 7 synergistic combinations , we only need implement experiments for 13 candidate combinations , which have greatly reduced the cost and time of pure experimental research . Most of previous computational studies for synergistic drug combination prediction didn’t implement any experimental validation . Only Jansen et al ( 2009 ) implemented biological experiments for predicted synergistic combination . However , training samples and criterion of selecting potential combinations for experimental validation were totally different between their studies and NLLSS . Therefore , it is also difficult to compare these two models based on the accuracy of independent prediction . Among the predictions , 7 pairs showed synergistic effects . For the Group 1 in Fig 6 , FK506 is a novel immunosuppressant isolated from Streptomyces [45] and has been demonstrated to bind to FKBP12 to inhibit calcineurin [46] , which is the key pathway for the cells responding to different stresses [47–50] . Radicicol and geldeanamycin from Groups 5 and 13 , respectively , can bind to Hsp90 and alter its function . Also , Hsp90 can act as the molecular chaperone to calcineurin [51 , 52] . Synergistic activity between FK506 and ketoconazole occurred at 16 h ( Fig 6 and S14 Table ) , but only at 48 h for radicicol with fluconazole and geldeanamycin with fluconazole ( Fig 6 , S15 and S16 Tables ) . Synergism between beauvericin and ketoconazole from Group 7 was identified based on our high-throughput synergistic screening platform [16] . Beauvericin can inhibit drug efflux pumps to reduce the accumulation of azoles and present synergistic activity [53] , and the synergistic activity was observed at 48 h ( Fig 6 and S17 Table ) . In Group 8 , caspofungin is a lipopeptide antifungal drug that inhibits the enzyme β ( 1 , 3 ) -D-Glucan synthase , thus interfering with the integrity of the fungal cell wall [54] . Our results proved that it could synergize with voriconazole and that synergistic activity started at 24 h ( Fig 6 and S18 Table ) . In Group 10 , posaconazole is a triazole antifungal drug that inhibits lanosterol 14α-demethylase ( ERG11 ) to block ergosterol biosynthesis [55 , 56] . Terbinifine also inhibits ergosterol biosynthesis by inhibiting squa lene epoxidase ( ERG1 ) [57] . These two drugs , which have the same pathway , showed synergistic activity that occurred at 24 h ( Fig 6 and S19 Table ) . Lovastatin from Group 12 is a member of the drug class of statins , used for lowering cholesterol by inhibiting the 3-hydroxy-3methylglutaryl-coenzyme A reductase ( HMG-CoA reductase ) , an enzyme that catalyzes the conversion of HMG-CoA to mevalonate [58] . Lovastatin could synergize with itraconazole in this study , and the synergistic activity started at 16 h ( Fig 6 and S20 Table ) . We found that 6 pairs of the predictions showed non-synergistic synergistic effects . Amphotericin B in Group 2 is a kind of polyene antifungal drug by binding to ergosterol to destroy the cell membrane [59 , 60] . Some researchers observed that a combination of azoles ( such as posaconazole and itraconazole ) can decrease the fungal infection burden [61–63] . However , others have observed no difference using a combination of ketoconazole or fluconazole and amphotericin B [64–66] . Ayse et al . reported that some isolates even showed antagonistic activities between amphotericin B and fluconazole [67] . Our results proved the absence of synergistic activity , even after 48 h ( S5 Fig and S21 Table ) . In Group 3 , no synergistic activity was observed between terbinafine and ketoconnazole on C . albicans , which is in agreement with a previous report [68] . However , it does show synergism with itraconazole [16 , 69 , 70] , fluconazole , voriconazole [71] and posaconazole ( S5 Fig and S22 Table ) . Flucytosine from Group 4 is a fluorinated pyrimidine analogue , and it inhibits fungal RNA and DNA synthesis [72] . The combination of ketoconazole and flucytosine can increase the antifungal effect , but no synergistic activity was observed [73] . Our result demonstrated that this combination only had “additive effect” ( 0 . 5<FICI<1 ) , but no synergistic activity ( S5 Fig and S23 Table ) . Disulfiram in Group 6 is an antifungal and can inhibit the drug efflux pump from C . albicans [74] . Ann et al . found antagonistic activities in clinical isolates when combining disulfiram with fluconazole [75] . Our results show that the combination of fluconazole and disulfiram had no synergistic activity during two days of incubation ( S5 Fig and S24 Table ) . Micafungin , the second available agent in the echinocandins class , is a potent inhibitor of 1 , 3-β-D-glucan synthase . It is used against fungal pathogens , such as Candida spp . and Aspergillus spp [76] . Our result showed that the combination of voriconazole and caspofungin had synergistic activity ( Group 8 ) , but micafungin did not show synergy with ketoconazole ( Group 11 , S5 Fig and S25 Table ) during two days of incubation , indicating that the interactions between echinocandins and triazoles may be related to the chemical structures of individual agents . For the same reason , in this study , we confirmed that caspofungin has synergistic activity with voricanazole ( Group 8 ) . However , the combination of ketoconazole and caspofungin had no synergistic activity ( Group 9 , S5 Fig and S26 Table ) , just like the nonsynergistic combination of ketoconazole and micafungin ( Group 11 , S5 Fig and S25 Table ) , suggesting that the synergistic activity may be related to chemical structures . Drug combinations represent a promising strategy for overcoming fungal drug resistance and treating complex diseases . In this work , NLLSS was developed to predict potential synergistic drug combinations by integrating known synergistic drug combinations , unlabeled drug combinations , drug-target interactions , and drug chemical structures on a large scale . NLLSS was motivated based on the observation that principal drugs which obtain synergistic effect with similar adjuvant drugs are often similar and vice versa . Both cross validations and experimental validations indicated that NLLSS has an excellent performance of identifying potential synergistic drug combinations . Out of 13 predicted antifungal synergistic drug combinations , 7 candidates were experimentally confirmed . NLLSS could provide a new strategy to identify potential synergistic antifungal combinations , explore new indications of existing drugs , and provide useful insights into the underlying molecular mechanisms of synergistic drug combinations . Previous research about synergistic drug combinations could be divided into the following three categories: only give the definition of synergy to determine whether or not a given drug combination is synergistic , only implemented experiments to screen synergistic combinations , and only implemented computational predictions to provide potential synergistic combinations . For example , methods such as combination index equation , Loewe additive model , HAS model , and universal response surface approach , only defined the concept of synergy , measured dose-effect curves , and determined whether or not a given drug combination is synergistic . Further , plenty of previous research has been devoted to implementing drug combination screening to search synergistic combinations and certain previous methods [14 , 28 , 77] strongly rely on experimental results , our method predicts synergistic drug combinations based only on the information from databases and the scientific literatures . Finally , although some computational methods has been developed , such as the model developed by Li et al . [29] , Zhao et al . [30] , Yin et al . [33] , Huang et al . [31] , Huang et al . [32] , Iwata et al [34] , and Chen et al . [35] , no experimental validation could be seen in their studies . Therefore , NLLSS differs significantly from previous methods based on the following four aspects , which also constitute the success factors of NLLSS . First , chemical structure information , drug-target interactions , and known synergistic drug combinations were integrated to capture potential synergistic associations . Second , known experimentally verified synergistic drug combinations were used as a seed dataset for predicting potential candidate combinations . Furthermore , a semi-supervised technique was adopted , whose advantage over supervised methods has been shown in many previous studies . More importantly , we not only developed computational models to quantitatively identify potential synergistic drug combination candidates , but also implemented experimental validations . These features highlight that NLLSS is essentially different from most of previous methods for drug combination predictions . Furthermore , the drug combination exploration space of NLLSS could include any drugs even if they do not have any known synergetic partners . Based on the procedure of NLLSS and drug combination similar nature proposed in this study , one drug could have predicted synergistic drug partners so long as it has at least one similar drug and this similar drug has been involved in known synergistic drug combinations . As mentioned , we implemented 10-fold , 5-fold , and 3-fold CV in all the three datasets , respectively . In this case , many drugs would do not have any known synergistic partners in the training samples . As the results listed in Table 2 , NLLSS has a reliable and robust performance in all the validation schemas , which could indicated NLLSS could be effectively applied to the drugs without any known synergistic drug combination partners . However , some limitations of NLLSS should be mentioned . First , the performance of NLLSS could be further improved by more available known synergistic drug combinations and drug-target interactions . Second , a more reliable measure of drug similarity would improve NLLSS . To do this , more biological information should be integrated to measure drug similarity . We also plan to develop new similarity integration methods to integrate different similarity measures for the further performance improvement . Furthermore , drugs must be classified into principal drug and adjuvant drug before predicting potential combinations in our method . Currently , there is still no acknowledged dividing standard for principal drugs and adjuvant drugs . Furthermore , when two drugs in the combination were classified as both principal drug and adjuvant drug , the same drug combination will obtain two different synergistic probability scores based on NLLSS . In this paper , we only chose the greater score as the final synergistic probability score of such drug combinations . In this study , NLLSS was developed based on the framework of LapRLS , which used different drug similarity matrix to construct different classifiers . Therefore , we can’t directly obtain a single classifier from the start . If we introduced the information of principal drug similarity matrix and adjuvant drug similarity matrix into the same cost function , we can’t obtain the analytical solution of the corresponding optimization problem . In the future , we would develop new computational tool , which could construct a single classifier in the beginning . Finally , some drug combinations are composed of more than 2 drugs . The current version of NLLSS only can predict drug combinations consisting of 2 drugs . In future work , we will develop new tools and methods to overcome the limitations of NLLSS . Finally , predictions based on NLLSS benefits the understanding of the mechanisms underlying synergistic drug combinations . For example , we predicted that the inhibitors from the calcineurin pathway and ergosterol biosynthesis pathway are the most popular synergistic combinations ( Groups 1 , 5 and 13 , Fig 6 ) . Calcineurin is a Ca 2+ / calmodulin-dependent serine/threonine phosphatase , and its structure and activation pathways are highly conserved from yeast to high eukaryotes [78] . In C . albicans , some reports revealed the involvement of calcineurin in antifungal tolerance , cell morphogenesis and virulence . The deletion of genes from the calcineurin pathway resulted in loss of tolerance to several antifungal agents , such as fluconazole , terbinafine , inhibitors for ergosterol biosynthesis , and caspofungin , an inhibitor of cell wall biosynthesis , and other growth-inhibiting agents ( e . g . , fluphenazine , caffeine ) [78] . When fungal cells are treated by triazole drugs , it is possible that cells incur membrane damage , as well as the accumulation of toxic sterols , and , at this time , the calcineurin pathway is activated to respond to these stresses [47–50 , 79] . The use of calcineurin pathway inhibitors makes fungal cells vulnerable to triazole drugs ( Fig 6 ) . These results indicate that many more synergistic antifungal combinations can be discovered from these two pathways . It is not clear that how lovastatin affects C . albicans , but it was reported that fluvastatin , the analog of lovastatin , has synergistic effect with itraconazole [51 , 80 , 81] . We proved that lovastatin can also synergize with itraconazole ( Group 12 , Fig 6 and S20 Table ) and that lovastatin may act in a similar manner with fluvastatin . It is interesting that two inhibitors ( Group 10: posaconazole and terbinafine ) from the ergosterol biosynthesis pathway have synergistic antifungal activity ( Fig 6 and S19 Table ) . Terbinafine targets ERG1 , which is the upstream gene for ERG11 , the target for posaconazole . Based on our predictions , it is the only synergistic combination that targets the same pathway . However , ketoconazole does not show synergistic effect with terbinafine ( Group 3 , S5 Fig and S22 Table ) , indicating either that the synergistic activity results from the different chemical structure of triazoles or that triazoles have some other effects on fungal cells , such as mitochondria [82] . Synergistic antifungal activities from groups 1 and 12 can be observed at 16 h , and they maintain their activity , even after 48 h , but group 8 showed synergistic activity after 24 h incubation . The synergistic activities of groups 5 , 7 and 13 can only be observed at 48h . The synergistic activity from group 8 only can be observed at 24 h ( Fig 6 ) . These time-course studies provide important information for the application of these synergistic combinations .	Drug combinations represent a promising strategy for overcoming fungal drug resistance and treating complex diseases . There is an urgent need to establish powerful computational methods for systematic prediction of synergistic drug combination on a large scale . Based on the assumption that principal drugs which obtain synergistic effect with similar adjuvant drugs are often similar and vice versa , NLLSS was developed to predict potential synergistic drug combinations by integrating known synergistic drug combinations , unlabeled drug combinations , drug-target interactions , and drug chemical structures . NLLSS has obtained the reliable performance in the cross validation and experimental validations , which indicated that NLLSS has an excellent performance of identifying potential synergistic drug combinations . Out of 13 predicted antifungal synergistic drug combinations , 7 candidates were experimentally confirmed . It is anticipated that NLLSS would be an important and useful resource by providing a new strategy to identify potential synergistic antifungal combinations , explore new indications of existing drugs , and provide useful insights into the underlying molecular mechanisms of synergistic drug combinations .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "antimicrobials", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "pathogens", "drugs", "microbiology", "antifungals", "drug", "screening", "fungi", "model", "organisms", "pharmacology", "fungal", "pathogens", "research", "and", "analysis", "methods", "mycology", "antimicrobial", "resistance", "medical", "microbiology", "microbial", "pathogens", "yeast", "biochemistry", "candida", "drug", "research", "and", "development", "microbial", "control", "biology", "and", "life", "sciences", "biosynthesis", "yeast", "and", "fungal", "models", "drug", "interactions", "drug", "information", "organisms", "candida", "albicans" ]	2016	NLLSS: Predicting Synergistic Drug Combinations Based on Semi-supervised Learning
Vibrio cholerae-specific bacteriophages are common features of the microbial community during cholera infection in humans . Phages impose strong selective pressure that favors the expansion of phage-resistant strains over their vulnerable counterparts . The mechanisms allowing virulent V . cholerae strains to defend against the ubiquitous threat of predatory phages have not been established . Here , we show that V . cholerae PLEs ( phage-inducible chromosomal island-like elements ) are widespread genomic islands dedicated to phage defense . Analysis of V . cholerae isolates spanning a 60-year collection period identified five unique PLEs . Remarkably , we found that all PLEs ( regardless of geographic or temporal origin ) respond to infection by a myovirus called ICP1 , the most prominent V . cholerae phage found in cholera patient stool samples from Bangladesh . We found that PLE activity reduces phage genome replication and accelerates cell lysis following ICP1 infection , killing infected host cells and preventing the production of progeny phage . PLEs are mobilized by ICP1 infection and can spread to neighboring cells such that protection from phage predation can be horizontally acquired . Our results reveal that PLEs are a persistent feature of the V . cholerae mobilome that are adapted to providing protection from a single predatory phage and advance our understanding of how phages influence pathogen evolution . A chief determinant of microbial survival is protection from predation . Phages are viral predators that act with exquisite specificity to kill their perpetually evolving bacterial targets . The overall success of epidemic Vibrio cholerae , the causative agent of the diarrheal disease cholera , is partly due to its ability to defend against predatory phages . Such phages are found in the aquatic environment [1] and are co-ingested with V . cholerae , permitting continued phage predation of V . cholerae within the human intestinal tract [2] . Recent molecular characterization of lytic phages associated with epidemic cholera has revealed that phage diversity is strikingly low over significant time periods , indicating that a surprisingly limited number of phage types place a significant predatory burden on V . cholerae in the context of human infection [2 , 3] . The most prominent phage found with V . cholerae in cholera patient stool in the endemic region of Bangladesh are the ICP1-related virulent ( lytic ) myoviruses [3] . ICP1 uses the lipopolysaccharide O1 antigen of V . cholerae to bind to cells and initiate infection [3] . The O1 antigen is required for V . cholerae to efficiently colonize the small intestine [4] , which places mutational constraints on V . cholerae in the human host and ensures ICP1 has access to susceptible V . cholerae in order to propagate [5] . Bacteria have evolved diverse antiviral resistance strategies to defend against the ubiquitous threat of predatory phages [6] . As obligate bacterial parasites , phages counter-adapt to overcome these resistance barriers , resulting in a dynamic co-evolutionary arms race [7] . The pervasiveness of ICP1 in Bangladesh with continued cholera epidemics suggests that V . cholerae has strategies to limit ICP1 predation that do not compromise virulence , and that ICP1 can evolve to overcome such defenses . Comparisons between sequenced ICP1 isolates revealed that roughly half of all ICP1 isolates encode a functional CRISPR–Cas ( clustered regularly interspaced short palindromic repeats–CRISPR-associated proteins ) system [8] . CRISPR–Cas systems function as adaptive immune systems that utilize small effector RNAs in complex with Cas proteins to direct the sequence specific degradation of invading DNA [9 , 10] . Typically , bacteria employ CRISPR–Cas to target invading phage DNA , therefore the ICP1 phage-encoded CRISPR–Cas system is a unique example of the unexpected genetic novelty found in studying phage-host coevolution . The ICP1 phage-encoded CRISPR–Cas system is utilized to mediate the degradation of a phage-inhibitory chromosomal island encoded by V . cholerae referred to as a phage-inducible chromosomal island-like element ( PLE ) [8] . The nature of how the PLE protects V . cholerae from infection by CRISPR–Cas deficient phage has not been described . PLEs have no sequence similarity to other known anti-phage systems; however , PLE 1’s designation was based on evidence that this island functionally resembles phage-inducible chromosomal islands ( PICIs ) of Gram-positive bacteria [8] . The staphylococcal pathogenicity islands ( SaPIs ) are well studied PICIs that take advantage of helper phages to enable their own replication and spread [11 , 12] . SaPIs are named for their role in pathogenesis , as they carry genes encoding for toxic shock syndrome toxin and other superantigens [13] . SaPIs exist quiescently in their host’s chromosome and are induced to excise and replicate upon initiation of their temperate helper phage’s lytic cycle . The SaPI life cycle results in the packaging of the SaPI genome into infectious phage-like transducing particles that permit horizontal spread of the SaPI . SaPIs use structural gene products encoded by the helper phage for their encapsidation [14 , 15] . SaPI mobilization , however , interferes with helper phage replication , a phenotype typified by their ability to inhibit helper phage plaque formation [16–18] . Like the SaPIs , V . cholerae PLE 1 inhibits plaque formation by ICP1 in the absence of ICP1 phage-encoded CRISPR targeting , and PLE 1 excises in response to ICP1 infection [8] . Here , we used bioinformatic approaches to identify PLEs in a geographically and temporally diverse collection of V . cholerae isolates . We discovered a total of five PLEs and found that a conserved feature of these islands is their ability to interfere with ICP1 phages . PLE activity abolishes ICP1 proliferation , and while we were unable to recover phage mutants that escape PLE-mediated interference in experimental evolution experiments , we found that ICP1 isolates recovered from patient samples display unique susceptibility patterns to different PLEs . We show that PLEs , like SaPIs , are mobilized in response to phage infection and can spread to neighboring cells such that protection from phage predation can be horizontally acquired . We demonstrate that phage genome replication is inhibited by PLE activity and cell lysis is accelerated following ICP1 infection when PLE is active , indicating a multi-faceted mode of phage interference . Together , our results reveal the significance of a specific predatory phage in the evolutionary history of epidemic V . cholerae and provide new insight into mechanisms underpinning phage-host coevolution . By analyzing the genomes of >200 V . cholerae isolates with known geographic and temporal origins [19 , 20] we identified five unique PLEs , each predicted to encode up to 29 open reading frames ( ORFs ) ( Fig 1A ) . A nucleotide alignment of V . cholerae PLEs shows that PLEs are void of genomic rearrangements . At the protein level , PLEs encode a conserved set of eleven predicted proteins ( protein translations for all predicted PLE-encoded ORFs are found in S1 Dataset ) . In silico analyses ( using CDD [21] , Pfam [22] and BLASTp ) of PLE proteins revealed that only five proteins have conserved domains shared with known proteins ( e-value < 1 × 10−2 ) . All PLEs encode an integrase with a serine recombinase domain ( cl02788 ) and all PLEs except PLE 3 encode predicted proteins with helix-turn-helix ( HTH ) DNA-binding domains ( including those in the MarR family ( COG1846 ) and general HTH superfamily members ( cl21459 ) ) . PLE 2 encodes a protein with an InsE domain ( COG2963 ) , typical of a transposase or inactivated derivative . PLE 1 also encodes a protein with a domain found in the PSK transcription factor superfamily ( cl01834 ) ( Fig 1A ) . In total , 51 out of the 208 V . cholerae isolates analyzed ( ~25% ) harbor a PLE . PLE+ V . cholerae have been isolated between 1949–2011 ( spanning the entire collection period in these studies [19 , 20] ) from disparate locations including Egypt , Mozambique , Bangladesh and Thailand ( S1 Table ) . PLEs are present in both classical and El Tor biotype strains , associated with the previous sixth and current seventh pandemics , respectively [23] , with PLE 5 restricted to classical isolates and PLEs 1 , 2 , 3 and 4 present in El Tor strains . The temporal distribution of each PLE is such that previously prevalent PLEs disappear when new variants emerge ( Fig 1B ) . All PLEs were located in chromosome II of V . cholerae ( Fig 1C and S2 Table ) , and all but PLE 2 were integrated within the superintegron , a gene capture system with hundreds of gene cassettes of mostly unknown function [24] . PLE 1 was previously shown to excise upon phage ICP1_2011_A infection and block plaque formation by that phage [8] . Here , we evaluated the specificity of anti-phage activity for all five PLEs . We constructed PLE+ derivatives of V . cholerae E7946 ( see Materials and methods ) to compare PLE+/ PLE- in the same strain background for these and all subsequent experiments . ICP1 isolates were assessed for their ability to form plaques on V . cholerae E7946 PLE+ derivatives . CRISPR-Cas+ ICP1 isolates were engineered to prevent CRISPR-mediated anti-PLE activity by deleting spacers in the CRISPR array or by deleting cas2-3 [25] , which possesses the nuclease activity required for target DNA degradation [10] . As shown in Fig 2A , all PLEs excised in response to ICP1 infection . Importantly , in the absence of ICP1 infection , PLE circularization was not detected ( Fig 2A ) , nor could it be detected following infection with unrelated phages ICP2 or ICP3 ( S1A Fig ) or following treatment with mitomycin C ( S1B Fig ) . For the mitomycin C treatment , we tested the PLE+ V . cholerae E7946 derivatives constructed in this study , as well as at least one PLE+ clinical isolate , since they may carry other prophages or mobile elements; however , we were unable to detect excised PLE in the absence of ICP1 infection . These results suggest that in contrast to SaPIs [11] , resident prophages activated by the SOS response do not activate PLEs . Our data demonstrate that PLEs do not block all phages , but that they do block ICP1 phages; however , until the molecular determinants of PLE activity are deciphered , it remains possible that other phages not tested here may also stimulate PLEs in V . cholerae . All PLEs blocked plaque formation by at least one ICP1 isolate ( Fig 2B ) and did not block plaque formation by ICP2 or ICP3 ( S1C Fig ) , demonstrating that ICP1 interference is a conserved feature of these elements . The ability of ICP1 isolates to form plaques on a given PLE+ strain was an all or nothing phenotype: ICP1 isolates that formed plaques on a PLE+ host strain did so at the same efficiency as on a PLE- strain , and when plaque formation was blocked , plaques could not be detected even when 108 plaque forming units were added to a PLE+ host strain ( Fig 2B ) . Interestingly , some ICP1 isolates recovered from cholera patient samples form plaques in the presence of certain PLEs independent of CRISPR activity ( Fig 2B ) . This finding suggests that ICP1 isolates have evolved to prevent triggering PLE activity or that they have CRISPR-Cas independent mechanisms to perturb PLE activity once it has been triggered . We quantified PLE-mediated ICP1 interference using one-step phage growth analysis . In a permissive V . cholerae PLE- host , ICP1 infection culminated in the release of approximately 90 infectious virions per cell within 25 minutes ( Fig 3A and 3B ) . Phage production was undetectable in PLE+ V . cholerae ( Fig 3B ) . All ICP1 isolates use the O1 antigen receptor to initiate infection and the CRISPR-Cas+ wild-type phage isolates form plaques on all PLE+ strains ( S2 Fig ) . Therefore , PLE activity does not block the phage genome from entering the cell , so we next quantified phage genome replication in the face of PLE activity . Interestingly , PLE 1 does not appear to perturb the kinetics of ICP1 replication in the first 10 minutes of infection , however , we found that PLE activity significantly reduces phage genome replication by approximately 4-fold by the end of the infection cycle ( p < 0 . 005 , Student’s t-test ) ( Fig 3C ) . Since phage genome replication is reduced but not eliminated , our results suggest that at least one additional mechanism of ICP1 interference is necessary to achieve the complete elimination of progeny virus production seen in the one-step phage growth analysis . To investigate whether PLE activity protects phage infected V . cholerae cells from cell death , we quantified cell survival following infection with ICP1 . Although PLEs block phage production , approximately equivalent levels of bacterial cell death were observed for PLE+ ( 97–98% ) and PLE- ( 98% ) V . cholerae after infection ( Fig 4A ) . In these analyses , we found that at a multiplicity of infection ( MOI ) of 5 , PLE activity accelerates the lysis of V . cholerae following phage infection . Upon infection of PLE- V . cholerae , we saw a slow increase in lysis of the bacterial culture as measured by OD600 ( Fig 4B ) . In stark contrast , infection of the PLE+ strains resulted in both an accelerated decline and more complete clearance of the bacterial culture . Since monitoring of OD600 after phage infection for PLE- at high MOI did not match the expected lysis timing obtained from one-step growth curves ( Fig 3A ) , we performed time-lapse fluorescence microscopy to more precisely determine how PLE activity impacts bacterial cell lysis dynamics . In these experiments , we imaged PLE+ and PLE- V . cholerae infected with phage at an MOI = 5 in the presence of the membrane stain FM 4–64 and the nucleic acid stain , Sytox Green , which brightly stains cells only when the membrane barrier is compromised . At the first several time points the two strains appeared identical , however , quantification of the loss of membrane integrity over time showed that cell lysis is accelerated in V . cholerae harboring PLE 1 compared to PLE- ( Fig 4C and 4D ) . Of note , the timing of the onset of lysis is the same in both strains , however , V . cholerae PLE 1 cells lysed in a more synchronized manner , and more PLE 1 cells were lysed at intermediate time points ( for example , 40% percent of the PLE+ population lysed between 45–65 minutes post-infection , while 10% of PLE- cells lysed during the same time period ( Fig 4C ) ) . The timing of lysis under the static conditions used for microscopy is delayed compared to in liquid culture ( Fig 4B vs 4C ) , but collectively these results indicate that PLE activity results in accelerated cell lysis after phage infection . The mechanism of accelerated lysis , which could be mediated directly through a PLE-encoded product ( s ) , through manipulation of the ICP1 lysis program , or even involve V . cholerae chromosomal product ( s ) , and the relative contribution of PLE-mediated accelerated lysis to phage inhibition are not known . Having established that PLE excision and cell lysis occurs in response to ICP1 infection , we next wanted to determine whether PLEs replicate and are packaged into infectious virions during ICP1 infection . We quantified PLE DNA before and after phage infection and observed that PLEs replicate to high copy number ( Fig 5A ) . Sequential sampling of PLE 1 copy number after phage infection showed that PLE replication is low 10 minutes post-infection , but increases substantially 15 and 20 minutes post-infection ( Fig 5B ) , which may indicate a switch from ICP1 replication ( which occurs unperturbed early in infection ( Fig 3C ) ) to PLE replication in infected cells . After replication , SaPI DNA is packaged into infectious phage-like transducing particles composed exclusively of helper phage virion proteins [14 , 15]; on entry into new cells , SaPI DNA integrates in a site-specific manner into the chromosome [11] . To investigate whether PLEs are similarly mobilized by ICP1 infection , we inserted a kanamycin resistance marker downstream of the last ORF in each PLE and measured PLE transduction frequency with ICP1 . We first confirmed that introduction of the kanamycin resistance cassette did not alter PLE replication ( S3 Fig ) . We then added cell-free supernatants from ICP1-infected PLE::kan cultures to recipient V . cholerae ( ΔlacZ:: spec ) and plated on agar plates supplemented with both antibiotics to select for cells that acquired PLE . PLE transducing units were detected at a frequency of ~104−105 per 108 infected cells , indicating that the overall efficiency of PLE packaging into infectious virions is low ( fewer than 1 transducing unit produced per 100 infected cells [PLE 2] or per 1000 infected cells [other PLEs] ( Fig 6A ) ) . As a control , when the same marker was inserted elsewhere in the chromosome of PLE+ strains ( shown as PLE 1 ( chr::kan ) ) , transduction was below the limit of detection , indicating that the packaging of PLE is not random ( Fig 6A ) . We hypothesized that PLE transduction would be dependent on the PLE encoded integrase . To test this , we constructed a PLE 1 Δint mutant and found that transduction was below the limit of detection , consistent with its predicted role in mediating PLE integration in recipient cells . To begin to address if PLEs are packaged into particles composed of ICP1 proteins , we evaluated if PLE transduction requires the V . cholerae lipopolysaccharide O1 antigen ( which is the ICP1 receptor [3] ) . Indeed , we found that PLE 1 could not be transduced to O1-antigen deficient V . cholerae ( ΔwbeL ) ( Fig 6A ) . Our results show that PLE transduction has the same receptor requirements as ICP1 infection and are consistent with the hypothesis that PLE DNA is packaged into virions composed of ICP1 proteins , although further analysis is required to evaluate the molecular nature of PLE transduction . We determined the site of PLE integration in recipient cells by amplifying and sequencing PLE chromosomal junctions with arbitrary primed PCR . For all PLEs , integration occurred in a site-specific manner ( Fig 6B and S3 Table ) . PLEs 1 , 3 , 4 and 5 integrated into a V . cholerae repeat ( VCR ) . VCRs are ~124 bp sequences found flanking gene cassettes in the V . cholerae superintegron [26] . VCRs are present in >100 copies , therefore ICP1-mediated PLE transduction yielded recipients in which the PLE integrated into a VCR and was consequently surrounded by unique flanking genes ( Fig 6B ) . PLE transductants showed phenotypic conversion to ICP1 resistance ( Figs 2B and 6C ) , and the position of the newly acquired PLE within the superintegron did not appear to impact phenotypic conversion to ICP1 resistance ( Fig 6C ) . In contrast to the other PLEs , PLE 2 integrated into VCA0581 ( encoding a hypothetical protein ) , a finding that is consistent with the observation that of the PLE encoded integrases , the PLE 2 integrase is the most divergent ( Fig 1A ) . We also determined the site of PLE integration in naturally occurring PLE+ V . cholerae isolates and found that the site of integration was the same as in experimental transductants ( that is , PLE 2 integrated into VCA0581 and PLEs 1 , 3 , 4 and 5 were found integrated in a VCR ( S2 Table ) ) . For natural V . cholerae isolates harboring PLEs integrated within a VCR , all PLE+ V . cholerae isolates were identical with respect to PLE flanking genes , indicating vertical transmission of PLEs in nature . Therefore , we found no evidence of ICP1-mediated PLE transduction ( horizontal acquisition ) in the natural strains we tested ( S2 Table ) , however it is possible that those strains are not representative of the breadth of PLE+ V . cholerae in nature . CRISPR activity is necessary for phage ICP1_2011_A replication on a PLE 1 host [8] , but unexpectedly , we found that PLE 1 transduction efficiency was unchanged when CRISPR was active ( S4 Fig ) . This indicates that the extent to which PLEs are packaged , potentially in ICP1 structural components , is not responsible for ICP1 interference . Such a result also implies that PLE DNA copy number is not the component that limits PLE transduction , but potentially that phage components required for packaging PLE particles may be limited during PLE-mediated phage inhibition . We have shown that PLEs are persistent genomic islands in geographically disparate V . cholerae isolates that provide highly efficient protection from a predatory phage . Our data demonstrate that all PLEs provide protection from ICP1 , the dominant V . cholerae phage found in cholera patient stool samples from Bangladesh [3] . The dominance of this phage and our current finding of a dedicated ICP1-defense system in V . cholerae isolates collected over a 60-year sampling period serve to further validate that interactions with ICP1-related phages have been a significant driver in the long-term evolution and selection of V . cholerae . Accessory genetic elements , like the PLEs , confer a fitness advantage in the face of ICP1 predation without the costs associated with compromising core functions through mutation . PLEs have no sequence similarity to other known anti-phage systems , and thus bioinformatic-based predictions to understand how PLEs block phage replication are largely uninformative . PLEs do , however , show some functional similarities to SaPIs , which are well known for their ability to parasitize helper phages to permit their own packaging and spread . We have provided evidence that like SaPIs , PLE transmission is facilitated by phage infection , and we have identified features of the PLE life cycle that provide insight into understanding how these evolutionarily conserved elements function . In contrast to SaPIs , PLEs do not encode identifiable regulatory , replication or packaging modules [12] . Nonetheless , our data demonstrate that ICP1 infection of PLE+ V . cholerae leads to PLE excision , replication and packaging . PLE activity is characterized by accelerated cell lysis ( Fig 4B ) and a complete block in progeny phage production ( Fig 3B ) , phenotypes that to our knowledge have not been reported for SaPIs . PLE activity is therefore similar to abortive infection systems , which act at the expense of the infected cell to eliminate phage production and protect the surrounding clonal population from infection . Although PLEs do capitalize on ICP1 infection to spread to neighboring cells , it appears to be relatively inefficient , raising the possibility that PLEs are ancient phage parasites that have evolved into specialized phage defense systems at the cost of their own horizontal transfer . In support of this idea , we found that PLE packaging following ICP1 infection is not responsible for ICP1 interference , since PLE transduction still occurs when the phage’s CRISPR-Cas system is active and PLEs are not inhibitory ( S4 Fig ) . The robust anti-phage activity of PLEs may be mediated in part by accelerated host cell lysis . Since cell lysis is not premature for all infected cells per se , we do not expect that PLE-mediated accelerated lysis is sufficient to explain the complete block in phage production . However , even slight deviations from the precisely controlled expression of the genetic information needed to amplify the phage genome , assemble viral particles and package phage DNA could have dramatic effects on phage viability , and there are likely some PLE+ cells in which the phage’s developmental program is incomplete prior to lysis . Similarly , PLE transduction may be limited by accelerated host cell lysis if phage components required for particle formation do not reach optimal levels prior to lysis . In addition , PLE activity interferes with phage genome replication , which may act in concert with accelerated host cell lysis and/or other yet to be identified mechanisms to efficiently block phage production . As a phage parasite , PLE packaging is likely completely dependent on phage-encoded structural proteins , and thus favoring accelerated cell lysis may come at a cost for the PLE . Since the requirements of PLE-mediated accelerated cell lysis have not been elucidated , it remains to be seen whether relieving accelerated cell lysis both restores some progeny phage production and enhances PLE transduction , as our model would predict . The evolution of a phage-encoded CRISPR-Cas system [8] to overcome PLE activity is remarkable and may speak to the relative strength of PLEs as defensive barriers in comparison to SaPIs . Some SaPIs decrease phage titer by only ~3x and still prevent plaque formation [16] , however , PLEs eliminate progeny phage production entirely . The mechanisms allowing phage to coevolve and overcome these genomic islands also differ . Helper phages that fail to induce SaPI activity can be readily selected for under laboratory conditions because SaPI induction depends on a single , dispensable helper phage-encoded protein [27 , 28] . Characterization of such mutants has led to the identification of SaPI inducing proteins , which function as phage-encoded antirepressors that induce SaPI excision , replication and packaging . In contrast to the SaPI-helper phage paradigm , we have been unable to use experimental evolution experiments to select for ICP1 mutants that escape PLE-mediated interference . This indicates that there may be an insurmountable fitness cost to altering or losing the PLE inducing cue and/or that multiple ICP1 products induce PLE activity to permit redundancy and ensure an adequate response by the PLE+ host . By deleting CRISPR-Cas in our collection of ICP1 isolates , we have identified certain phage isolates that can escape PLE mediated interference , highlighting the need to study naturally evolved bacterial and viral populations . ICP1 isolates differ by thousands of single nucleotide polymorphisms and by the presence of accessory modules like CRISPR-Cas [3 , 8] , making bioinformatic approaches to identify the defining feature ( s ) mediating PLE escape ineffective . As we strive towards a more comprehensive understanding of the role of phage in shaping bacterial communities in health and disease , it is imperative that we consider the vast gene pool enabling the acquisition of novel traits and continued coevolution that inherently cannot be replicated in laboratory evolution experiments . The long-term interactions between V . cholerae and ICP1 serve as a useful paradigm to understand the evolution of phage-resistance and counter-resistance in the context of human disease , and may allow for the potential manipulation of these systems for therapeutic or prophylactic benefit . Strains utilized in this study are listed in S4 Table . PLEs were transduced into V . cholerae E7946 [29] ( described below ) to generate PLE+ derivatives in the same strain background for comparisons in these studies . Bacteria were routinely grown at 37°C on lysogeny broth ( LB ) agar or in LB broth with aeration . Media was supplemented with kanamycin ( 75 μg/ml ) , spectinomycin ( 100 μg/ml ) , and/or streptomycin ( 100 μg/ml ) when appropriate . Antibiotic resistance markers were introduced into V . cholerae strains by natural transformation as described [30] . Splicing by overlap extension ( SOE ) PCR was used to generate all PCR constructs . Primer sequences are available upon request . In order to generate PLE+ derivatives in the same strain background , PLEs were marked with a kanamycin resistance cassette downstream of the last ORF . PLEs 1–3 were mobilized by transduction with an ICP1 isolate into V . cholerae E7946 . Natural transformation and transduction were used to generate V . cholerae E7946 harboring PLE 4 or PLE5 in the following manner: V . cholerae E7946 was made competent by growth on chitin [30] and ~2μg purified genomic DNA from the kanamycin resistant PLE 4 or PLE 5 derivative strain was added and the mixture was incubated at 30°C overnight and then plated onto LB kanamycin plates . Kanamycin resistant colonies were screened by PCR to ensure the desired incorporation of the entire PLE , and then to ensure a clean genetic background , these derivatives were used as donors in ICP1_2011_A-mediated transduction assays into V . cholerae E7946 . For all PLE+ strains , the kanamycin resistance cassette was removed using cotransformation [31] of the wild-type locus with a selected product to replace lacZ with a spectinomycin resistance marker . The spectinomycin resistance marker was subsequently replaced by the wild-type lacZ locus and screening for desired transformants on plates containing 40 μg/mL 5- bromo-4-chloro-3-indolyl-β-D-galactopyranoside . The PLE 1 integrase deletion construct was constructed using FLP-FRT recombination as described [32] . Mutations in ICP1-related phages were generated using CRISPR-Cas mediated genome engineering as described [25] . V . cholerae E7946 PLE+ were grown to OD600 = 0 . 3 and infected with phage at an MOI of 5 . Samples were taken 20 minutes post-infection , boiled and used as template for PCR to detect the circularized PLE using outward facing primers as depicted in Fig 2A . In order to determine if PLEs circularize in response to ICP2 or ICP3 [3] , boiled plaques on V . cholerae E7946 PLE+ served as a template for circularization PCR . Positive controls using plaques on PLE+ strains infected with ICP1-related phages were used in all assays . In order to test if induction of the SOS response could stimulate PLE circularization , PLE+ strains ( both V . cholerae E7946 PLE+ transductants and clinical isolates naturally found to harbor each PLE ) were grown to OD600 = 0 . 3 and treated with mitomycin C ( at 20 ng/mL and 100 ng/mL ) for 30 minutes . Treated samples were boiled and used as a template for PCR as above . All PCR reactions were carried out under identical conditions for 30 cycles with positive controls in all assays . Circularization products were confirmed by sequencing . Phage susceptibility was determined using the soft agar overlay method as described [25] . One-step growth curves were used to determine the average phage burst size [33] . One-step growth curves were performed in triplicate and the phage burst is reported as the means ± SD ( Standard Deviation ) in Fig 3B . Bacterial survival was determined following infection of V . cholerae E7946 and its PLE+ derivatives with phage as follows: strains were grown to an OD600 = 0 . 3 and infected with ICP1_2006_E ΔCRISPR ( MOI = 5 ) . After 15 minutes of incubation at 37°C with aeration , serial dilutions of each infected culture were plated on LB streptomycin plates . Uninfected cultures were plated for CFUs immediately prior to infection and the percent survival was calculated as ( CFU ( phage treatment ) /CFU ( uninfected ) ) x100 . The average percent survival was determined from three biological replicates and is reported as the means ± SD in Fig 4A . The kinetics of phage infection of V . cholerae E7946 and its PLE+ derivatives with ICP1_2006_E ΔCRISPR were performed at the MOI indicated at 37°C with aeration . V . cholerae strains were grown to an OD600 = 0 . 3 and then concentrated 5-fold before being infected with ICP1_2006_E ΔCRISPR ( MOI = 5 ) in a 200 μL volume . 1 μL each of 0 . 05mM Sytox Green nucleic acid stain ( Thermo Fisher Scientific ) and 1 μg/μL FM 4–64 ( Thermo Fisher Scientific ) were added and the mixture was incubated for 5 minutes at room temperature . 10 μL of the cell suspension was then placed on an agarose pad ( 1 . 5% diluted in LB ) made using a gene frame seal ( Thermo Scientific ) . Images were taken at 5-min intervals with the stage set to 37°C with an Olympus FV1000 confocal microscope with a 60X objective . The average percent intact cells were determined from three biological replicates and are reported as the means ± SD in Fig 4C . For transduction assays , phage ( MOI = 5 ) were added to V . cholerae strains at an OD600 = 0 . 3 for 5 minutes at 37°C with aeration . The mixture was centrifuged and washed to remove unabsorbed phage , resuspended in fresh LB broth and incubated for 30 minutes at 37°C with aeration . The lysate was treated with chloroform and centrifuged to remove bacterial debris . 100 μL lysate was mixed with 100 μL overnight culture of recipient V . cholerae ( ΔlacZ:: spec as wild-type recipient , or ΔwbeL ΔlacZ:: spec as indicated ) at 37°C for 1 hour . This mixture was plated on LB agar plates supplemented with kanamycin and spectinomycin to enumerate transducing units . PLE transducing units were calculated from three biological replicates and are reported as the means ± SD of each donor/recipient pair indicated in Fig 6A . A kanamycin cassette inserted into the neutral gene VC1807 served as donor strains for detecting transduction of non-PLE associated sequence from PLE+ strains . The site of PLE integration in clinical isolates and experimental transductants was determined by arbitrary-primed PCR [34] . qPCR reactions were performed with iQ SYBR Green Supermix ( Bio-Rad ) using a CFX Connect Real-Time PCR Detection system ( Bio-Rad ) . For all assays , at least three independent samples were tested for each condition and each template sample was tested in technical duplicate . In order to quantify phage genome replication , V . cholerae was grown to OD600 = 0 . 3 . Phage ( at an MOI = 0 . 1 ) were added and incubated at 37°C with aeration . At the times indicated , 20 μL samples were taken , boiled and diluted 1:50 and used as template for qPCR , which was compared to the input sampled immediately after adding phage . Phage-specific primers zac68 ( 5’-CTGAATCGCCCTACCCGTAC-3’ ) and zac69 ( 5’-GTGAACCAACCTTTGTCGCC-3’ ) were used in this analysis . For PLE replication following phage infection V . cholerae was grown to OD600 = 0 . 3 . Phage ( at an MOI = 5 ) were added and incubated at 37°C with aeration . Samples were taken as above , boiled and diluted 1:1000 and used as template for qPCR for comparison to the input that was sampled immediately before adding phage . Primers universal for all PLEs were used for qPCR: zac14 ( 5’-AGGGTTTGAGTGCGATTACG-3’ ) and zac15 ( 5’-TGAGGTTTTACCACCTTTTGC-3’ ) .	Vibrio cholerae is the causative agent of the severe diarrheal disease cholera . V . cholerae is commonly recovered from patient samples with predatory bacteriophages ( phages ) , which impose strong selective pressure favoring phage resistant strains over their vulnerable counterparts . Here , we investigated the activity of PLEs ( phage-inducible chromosomal island-like elements ) , a novel group of mobile genetic elements that have contributed to phage resistance in V . cholerae over the last 60 years . Surprisingly , we found that PLEs are protective against a single , prevalent phage type . We found that PLE activity reduces phage genome replication and accelerates the kinetics of bacterial cell lysis . Our study shows that mobile genetic elements play a key role in phage resistance in successful epidemic V . cholerae .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "biotechnology", "genome", "engineering", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "engineering", "and", "technology", "bacteriophages", "pathogens", "vibrio", "synthetic", "biology", "tropical", "diseases", "microbiology", "synthetic", "bioengineering", "crispr", "viruses", "physiological", "processes", "bacterial", "diseases", "vibrio", "cholerae", "dna", "replication", "neglected", "tropical", "diseases", "molecular", "biology", "techniques", "dna", "bacteria", "bacterial", "pathogens", "microbial", "genomics", "synthetic", "genomics", "research", "and", "analysis", "methods", "bioengineering", "synthetic", "genome", "editing", "viral", "genomics", "infectious", "diseases", "cholera", "tissue", "repair", "artificial", "gene", "amplification", "and", "extension", "medical", "microbiology", "microbial", "pathogens", "molecular", "biology", "lysis", "(medicine)", "biochemistry", "nucleic", "acids", "polymerase", "chain", "reaction", "virology", "physiology", "genetics", "biology", "and", "life", "sciences", "genomics", "organisms" ]	2017	A highly specific phage defense system is a conserved feature of the Vibrio cholerae mobilome
Schistosomiasis is a debilitating neglected tropical disease that infects over 200 million people worldwide . To combat this disease , in 2012 , the World Health Organization announced a goal of reducing and eliminating transmission of schistosomes . Current control focuses primarily on mass drug administration ( MDA ) . Therefore , we monitored transmission of Schistosoma mansoni via fecal egg counts and genetic markers in a typical school based MDA setting to ascertain the actual impacts of MDA on the targeted schistosome population . For 4 years , we followed 67 children enrolled in a MDA program in Kenya . Infection status and egg counts were measured each year prior to treatment . For 15 of these children , for which there was no evidence of acquired resistance , meaning they became re-infected following each treatment , we collected microsatellite genotype data from schistosomes passed in fecal samples as a representation of the force of transmission between drug treatments . We genotyped a total of 4938 parasites from these children , with an average of 329 . 2 parasites per child for the entire study , and an average of 82 . 3 parasites per child per annual examination . We compared prevalence , egg counts , and genetic measures including allelic richness , gene diversity ( expected heterozygosity ) , adult worm burdens and effective number of breeders among time points to search for evidence for a change in transmission or schistosome populations during the MDA program . We found no evidence of reduced transmission or schistosome population decline over the course of the program . Although prevalence declined in the 67 children as it did in the overall program , reinfection rates were high , and for the 15 children studied in detail , schistosome egg counts and estimated adult worm burdens did not decline between years 1 and 4 , and genetic diversity increased over the course of drug treatment . School based control programs undoubtedly improve the health of individuals; however , our data show that in an endemic area , such a program has had no obvious effect on reducing transmission or of significantly impacting the schistosome population as sampled by the children we studied in depth . Results like these , in combination with other sources of information , suggest more integrated approaches for interrupting transmission and significantly diminishing schistosome populations will be required to achieve sustainable control . Recently , there has been increased awareness of the massive global health burden of schistosomiasis and other neglected tropical diseases ( NTDs ) . Notably , in 2001 , Resolution 19 of the World Health Assembly called for increased drug treatment of NTDs to reach a minimum coverage of 75% of school aged children in endemic areas ( WHA 54 . 19 ) . Thus , the initiation of multiple mass drug administration ( MDA ) programs was prompted [1] . School aged children have been specifically targeted because they are suspected to harbor the heaviest worm burdens and thus experience a high degree of morbidity from infection . Additionally , school systems provide infrastructure within which such programs can successfully operate , a key consideration for lowering the cost of drug administration . Traditionally , the primary goal of MDA programs has been to reduce worm burdens in individuals and thus reduce the morbidity caused by NTDs . However , a revised roadmap published by the World Health Organization in 2012 reaches further: toward the elimination of NTD transmission in some regions [2] . To a large part , this goal is being facilitated by a major increase of pharmaceutical donations to expand coverage of MDA . With this goal in mind , it is important to determine if current deworming programs are moving toward elimination by reducing schistosome transmission in endemic regions where MDA programs are ongoing . This knowledge is critical to inform control protocols to achieve the WHO goals . It is challenging to determine if drug treatment of a focal group of patients is reducing the entire schistosome population in a region ( especially so for worms in untreated people ) and thus the force of infection . For schistosomes , the force of infection has been defined as the rate of establishment of patent infections . Ideally , one would treat all infected people; however , this is often not logistically possible . Population genetic estimates indicate that schistosome populations are large and widely distributed , not only geographically but also among snails , human hosts of a wide range of ages and reservoir hosts [3]–[6] . Thus , most MDA programs treat a portion of people infected with schistosomiasis , exposing only a portion of the schistosome population to drug treatment while the rest of the population remains in “refugia” , isolated from drug exposure . Because the goal of most deworming programs is to reduce worm burdens in individuals and thus morbidity ( rather than elimination or reducing the force of infection ) , success is typically assessed by comparing infection intensity ( worm burden estimated by fecal egg counts ) and prevalence of infection in the treated portion of the population from before treatment to some time point afterwards . It is difficult to use these data to assess the reduction of the larger schistosome population for two reasons . First , infection starts in infancy and schistosomes are long lived ( 5–10 years ) [7]–[9] , thus pre-mass drug treatment levels of infection reflect long term accumulation of worms and should be higher than recolonization levels after treatment clearance [10] . Second , it has been shown that a large portion of individuals treated for schistosomiasis acquires partial resistance to reinfection that is measurable within 2–3 drug treatments [11]–[15] . Thus , a decrease in schistosome burdens in the treated portion of the population is expected , and this decrease may not reflect an actual decrease in the entire schistosome population or in the force of infection . In fact , it is possible that the force of infection could actually be increasing even when worm burdens of the treated individuals are declining due to acquired resistance in the treated population . One way to monitor the change in schistosome populations is to monitor those treated individuals who do not acquire resistance to reinfection after repeated treatments ( i . e . those that remain susceptible ) . Such individuals were termed “phenotypically susceptible” by Black et al . 2010 [11] because they failed to produce protective immunity after repeated drug treatment . Assuming comparisons are not made with pre-control baselines , treatment is successful , and prior worm burdens are cleared , the worm populations acquired by patients after treatment can be measured by the number of schistosome eggs released in a fecal sample via the Kato-Katz methodology [16] , [17] , especially when coupled with genetic data on miracidia from these eggs . Genetic data can provide an additional and powerful perspective because they can be used to estimate worm burdens within an individual ( [i . e . number of breeding pairs represented by sibling groups 18] ) , to detect overall population declines ( i . e . population bottlenecks [19] , [20] ) , and measure genetic diversity which is a measure of the parasite's ability to adapt to environmental pressures [21] . Some attempts have been made to assess changes in schistosome populations following drug treatment . Norton et al . [21] . monitored changes in microsatellite populations of Schistosoma mansoni miracidia ( n = 20 per child ) derived from Tanzanian school children following a single round of treatment and found a significant decline in genetic diversity [21] . They noted this was true even for young children entering the school who had not been treated previously . Following the fate of infections in individual hosts as an important indicator of efficacy of control was also suggested . Using cytochrome c oxidase I as a marker , Betson et al . [22] found substantial genetic diversity of S . mansoni within both children less than six years old and their mothers ( they examined 1347 parasites from 35 mothers and 45 children ) [22] . They did not observe any change in schistosome genetic diversity before or six months after treatment in samples collected from the same individual or from the same family . The authors noted that even exposure times of 1 . 5 years were sufficient to result in genetically diverse infections in young children . Within the context of a typical school-based MDA program in central Kenya targeting S . mansoni for control on an annual basis , we took an approach that emphasized a deeper sampling of miracidia within individual children , and that began with examination of children for four years following treatment . Deeper sampling enables the calculation of novel measurements of worm burdens within a patient and also enables genetic diversity to be more accurately measured because the bias that is introduced by including related parasites in the sample can be reduced ( see Steinauer et al . [23] ) . We used both fecal egg counts and genetic measures to assess changes in phenotypically susceptible children over a 4 year period ( 2008–2011 ) . This protocol was approved by both the Kenya Medical Research Institute ( KEMRI ) Scientific committee and the KEMRI Ethical committee . Parents/guardians of all children involved provided written informed consent on behalf of all child participants . Informed consent was obtained by first holding a meeting with the parents/guardian to explain the purpose of study followed by a question and answer session . Thereafter , they were requested to sign the written consent forms on behalf of the children . A school-based schistosomiasis and soil-transmitted helminth ( STH ) control project in Mwea , Kenya was established in 2004 through a collaboration between the Japan International Cooperation Agency ( JICA ) and the Kenya Medical Research Institute ( KEMRI ) . Mwea is a large rice growing region in the Kirinyaga County , central Kenya . The Thiba and Nyamindi Rivers that pass through this area serve as source of water for the irrigation schemes , and schistosomiasis transmission is believed to take place primarily from the irrigation canals that supply water to the rice schemes . A pilot study in 2004 measured prevalence of helminths in school aged children followed by MDA ( praziquantel and albendazole ) [24] . At this time , the human population of the area was estimated to be 125 , 000 and the prevalence of Schistosoma mansoni in school aged children was 47 . 4% . Starting in 2004 , the KEMRI-JICA Project administered annual doses of anthelmintics to all school aged children ( >40 , 000 ) in the region regardless of their infection status . This treatment included a single dose of 40 mg/kg of praziquantel using the tablet dose pole to determine the number of tablets [25] and albendazole in a 400 mg single dose . Prior to the beginning of the program a baseline determination of prevalence and intensity of parasitic infections through examination of stool samples of class three children ( age range 5–14 years ) was undertaken [26] . Our study recruited the subjects by randomly sampling among the children who had previously tested positive within this cohort . We focused on infected individuals because we were seeking those that were not acquiring resistance to reinfection with repeated drug treatment . Starting in 2008 , we followed 67 students previously enrolled in the KEMRI-JICA Project . These students were between 5 and 14 years of age and were from four primary schools: Kirogo ( 00°39S/37°23E ) , Nyamindi ( 00°40 S/37°24E ) , Mukou ( 00°40S/37°20E ) and Ngurubani ( 00°41S/37°21E ) . The schools are less than 8 km apart ( [figure in 3] ) . Prior to each treatment , fecal samples were obtained from three consecutive days and infection was assessed using the Kato Katz methodology [16] , [17] . Following each treatment , there was a three month follow-up stool exam for each child , and each child was found to be negative ( treatment was successful in every case ) . For the Kato-Katz procedure , two slides per patient were examined for schistosome eggs . To assess population level reduction in schistosome transmission , we collected genetic data from the schistosome infrapopulations of 15 of the 67 children . These 15 children were deemed phenotypically susceptible throughout the four year time period because they became reinfected at every time point with large worm burdens ( as determined by egg counts ) indicating high susceptibility of these children . More of the 67 children also fell into the phenotypically susceptible category , but we only compiled genetic data from the 15 children from whom we were able to collect samples at each sampling period . Miracidia were collected from their fecal samples and used for microsatellite genotyping . Schistosome miracidia were hatched from fecal samples [27] and individuals were genotyped at 12 microsatellite loci [28] and included GenBank accession numbers: AF325695 , AF325698 , AF202965 , AF202966 , AF202968 , L46951 , M85305 , R95529 , AF202968 , AI395184 , AI067617 and BF936409 . We compared reductions in fecal egg counts in the 15 phenotypically susceptible children and in the other 52 children over the 4 time points ( annual samples prior to MDA 2008–2011 ) using a repeated measures ANOVA with Systat 11 ( Systat Software , Inc . ) . Data were natural log transformed to meet the assumptions of parametric statistical tests . Single degree of freedom polynomial contrasts were used to determine the significance of trends in the data over time . The hypothesis was that a reduction in the force of infection of schistosomiasis would be matched with a reduction in fecal egg counts in all children and in the 15 phenotypically susceptible children . We also compared population genetic parameters across the four time points at two levels: populations within each patient ( infrapopulations ) ( [e . g . 21] ) and all patients combined ( component population ) ( [e . g . 10] ) . We hypothesized that significant reductions in the schistosome population should be accompanied by reductions in all of these genetic parameters over the four year time period . For the infrapopulation level analysis , we first compared two estimators of the worm burdens within a patient: the number of full sibling families , and the effective number of breeders ( Nb ) . These parameters were compared across all time points using repeated measures ANOVA for each dependent variable . The number of full sibling families is a measure of the number of breeding worms within a patient ( worm burden ) because the miracidia collected in a fecal sample are offspring of the adult worms inside the patient . The offspring were partitioned into their families based on shared alleles using COLONY v . 2 . 0 [29] , [30] . COLONY has been shown to accurately reproduce schistosome families using genotype data at the same microsatellite loci [18] . Nb was estimated using the sibling assignment method implemented in COLONY [31] . Nb is the effective population size Ne measured from a single breeding cohort ( miracidia derived from a single patient ) . We also compared two measures of population genetic diversity: allelic richness and gene diversity ( expected heterozygosity ) . We corrected our datasets for the bias induced by related miracidia in a fecal sample by inferring the family structure present in a sample , and then resampling the dataset to include only one member of each family [18] , [32] . Ten resampled datasets were generated and both parameters were calculated using FSTAT 2 . 9 . 3 [33] . Parameters were natural log transformed and compared across patients between the first and last sample using repeated measures ANOVA ( Systat 11 , Systat Software , Inc . ) . One patient was removed from the analysis due to high family structure and resulting low sample size in the resampled datasets ( n = 6 unrelated individuals remaining after correction , thus all individuals sampled fell into one of 6 families ) . At the component population level ( all patients combined within a year ) , we used permutation tests on the 10 corrected datasets to determine significant differences in gene diversity and allelic richness between year 1 and year 4 of the MDA program . These tests were performed with FSTAT and 10 , 000 permutations were used to determine significance . Table 1 provides a summary of the 67 children investigated , and includes their gender , age at baseline , and pre-treatment egg counts ( average eggs per gram of feces determined by Kato-Katz ) at each of the four annual treatment periods . Table 2 indicates for the 15 phenotypically susceptible children investigated in depth , for each annual observation , the number of miracidia genotyped ( N ) , the effective number of breeders ( Nb ) , and the schistosome census number ( Nc ) . We genotyped a total of 4938 parasites from these children , with an average of 329 . 2 parasites per child for the entire study , and an average of 82 . 3 parasites per child per annual examination . Repeated measures ANOVA of the 15 phenotypically susceptible children indicated that egg counts were significantly different among years ( F3 , 42 = 5·662 , P = 0·002 ) ( Fig . 1A ) . Prevalence of infection in all 67 patients declined significantly from 97% at year one level , to 77 . 6% in year 2 , and 68 . 7% in years 3 and 4 ( Fisher exact test Yr1 v . Yr 4 , P = 0·0001 ) ( Fig . 1B ) . After the first year , three individuals were not found infected in any subsequent years , and three additional children remained negative for the remainder of the study after treatment in the second year . Polynomial contrasts indicated significant second and third degree trends ( quadratic and cubic ) ( F1 , 14 = 8·012 , P = 0·012 , F1 , 14 = 7·014 , P = 0·014 ) , but not a significant linear trend ( first degree ) ( F1 , 14 = 1·183 , P = 0·183 ) ( Fig . 2A ) . The trend was a decline in egg counts through years 1 , 2 , and 3 and then an incline back to original levels in year 4 . Egg counts in the remaining 52 children also changed significantly over time ( F3 , 153 = 9·89 , P<0 . 001 ) ( Fig . 2B ) with a significant linear trend ( F1 , 51 = 39·562 , P<0 . 001 ) , but not quadratic or cubic trends ( F1 , 51 = 1·8 , P = 0·176; F1 , 51 = 0·005 , P = 0·946 ) . The linear trend was a decline over the four years . This change appeared to be driven by an increase in uninfected individuals ( egg counts of 0 ) rather than a decrease in egg counts of infected individuals , as indicated by the decline in prevalence ) because a one-way ANOVA using only values from infected individuals indicated no significant differences between years ( F3 , 145 = 1·019 , P = 0·604 ) . Thus , fewer children were infected; however , those that became reinfected did not have significantly lower burdens . The number of full sibling families and effective number of breeders did not show statistically significant change over time ( full sibling families: F3 , 42 = 0·397 , P = 0·756; Nb: 0·757 , P = 0·757 ) ( Figure 3 ) . Allelic richness and gene diversity significantly increased over time ( allelic richness: F1 , 126 = 416·2 , P<0 . 0001; gene diversity: F1 , 126 = 123·5; P<0 . 0001 ) . Significant interactions were detected between patient and time for both parameters ( allelic richness: F13 , 126 = 159·5; P<0 . 0001; gene diversity: F13 , 126 = 159·5; P<0 . 0001 ) indicating that for some patients , these values decreased over time . The findings of our study highlight the ability of schistosomes to remain stubbornly entrenched in endemic areas despite school based MDA . Over a period of four years , we found a reduction in the number of children infected with schistosomiasis , a primary goal of the JICA/KEMRI program . This program has undoubtedly reduced or prevented severe morbidity caused by schistosome infection in children . However , we have found no evidence of an overall reduction of schistosome transmission in the region either by monitoring worm burdens via either egg counts or using genetic parameters to monitor population changes . We saw changes in egg counts in our 15 phenotypically susceptible patients over time , but this change was a decline in years 1–3 , but one that was followed by an increase back to initial levels in year 4 . Also , in the other 52 children , although some did not become reinfected at some of the sampling times which is certainly a desired outcome of control , we saw no change in the egg burdens over the four years of study for these individuals when they did reacquire infections . Furthermore , among our 15 phenotypically susceptible children , we detected an increase in genetic diversity over the course of treatment , which would not be expected had a population bottleneck or significant decline actually occurred due to treatment . Increased diversity is unlikely to be explained by selection due to the MDA program because these microsatellite markers are presumably neutral with regard to drug resistance or immune evasion . The mechanism driving the increase in genetic diversity is unknown and could be due to population increase ( more individuals means more chance of novel mutations ) , or immigration from other surrounding schistosome populations ( outbreeding ) . Although annual treatments of all school aged children in the Mwea region has been successful in reducing infection in children , it appears to have had no effect on the overall transmission of schistosomes in the region . This observation is consistent with that of Kihara et al . [26] who observed that following each round of treatment , prevalence increased , albeit not to baseline levels . Note that our sampling did not include a baseline measure . Interestingly , annual prevalence in the 67 children was seen to decline over time , but this decline was not matched in the egg counts of infected individuals . Following MDA , prevalence is predicted to rapidly rise to pre-control levels , while mean worm burden rises much more slowly due to the highly non-linear relationship between prevalence and intensity ( large changes in intensity result in very small changes in prevalence ) [34] . This pattern may be reflecting acquired resistance after repeated drug treatment or behavioral changes as the children age and due to education . Another explanation is a lack of reliability of the Kato Katz technique to accurately measure worm burdens [18] , [35] , [36] . Understanding the reasons behind these changes is important for successful monitoring and also for successful modeling of MDA strategies and should be further investigated . Our findings differ from recent studies that detected a reduction in genetic diversity ( allelic richness and gene diversity ) of schistosomes collected from fecal samples of children enrolled in a school based treatment program near Lake Victoria in Tanzania [10] , [21] . In these studies , a reduction in allelic richness and gene diversity ( heterozygosity ) was found between baseline infection levels and a single year following treatment . The differences between these studies and ours may indicate important differences in local epidemiology , MDA programs , or monitoring protocols . For instance , sampling schemes differed greatly between our studies . We sampled over four years rather than between baseline and year 1 which gives a longer perspective on population changes . Our modest sample size of patients did not appear to limit statistical power to detect deviations from the null hypothesis of no change among years as we were able to reject the null hypothesis in many of our analyses . The patterns we found did not fit the hypothesized decline over time . Also , in a previous study of schistosome genetic diversity , resampling simulations indicated that data from only 10 hosts are enough to achieve sufficient statistical power to detect differences in genetic diversity [10] . Here we note that caution may be required in extrapolating our results for these15 children to the schistosome population in the entire Mwea scheme area . However , these children come from different schools within the scheme which varied in their baseline prevalence and level of transmission . From our results , we are confident that large declines in the schistosome population , such as those nearing elimination , were not occurring in the Mwea population . Due to costs , a tradeoff between in depth sampling , number of individuals , and number of years to sample is inevitable . By choosing an increased depth of sampling over four years , we were able to look at longer term patterns of change , use novel measures of population change ( Nb and FSF ) , and more accurately measure genetic diversity after correcting for bias that is present when collecting miracidia from a patient . Our findings coincide with recent modeling results that indicate deworming programs targeted solely at school aged children are likely to be limited in terms of their impact on community-wide parasite transmission even at high levels of drug efficacy ( 95% ) and coverage of school aged children ( 85% ) because of the relatively small portion of the parasite population exposed to treatment [37] . Increasing the demographic that is included in the treatment program is likely to reduce parasite transmission; however , it must also be noted that increasing the proportion of parasites that are exposed to drug treatment is also predicted to increase the selection pressure on drug resistance [38] . This is a particular concern for schistosomiasis given that there is only one drug effective against all schistosomes and that schistosomes with reduced susceptibility to praziquantel have been reported from Kenya [39] . Together , these findings underscore the need to include alternative control approaches [34] , [40] . In general , in our view , control efforts should take cognizance of the specific local environmental and epidemiological circumstances , try to use the inherent biodiversity present to limit transmission , and use an integrated multi-pronged approach as a more sustainable way forward . In summary , after four years of school based MDA , we did not detect a reduction in schistosome population . Our data set is unique in that it combines both egg counts and novel genetic parameters to measure population changes . Also , our study follows a portion of the population that has remained phenotypically susceptible to monitor changes in the schistosome population over a four year period with annual school-based mass treatments . This design allows changes to be monitored without interference of acquired immunity and accumulation of worms over time , two road blocks to measuring the efficacy of school-based treatment programs on a community wide scale .	Schistosomiasis is a chronic and debilitating disease . Current control focuses primarily on mass drug administration ( MDA ) . We monitored schistosome transmission via fecal egg counts and genetic markers in a school based MDA setting to learn how the intervention was influencing transmission or the targeted schistosome population . For 4 years , we followed 67 children enrolled in a MDA program in Kenya , and for 15 of these , we repeatedly acquired in depth genetic data regarding the schistosome populations they harbored . Although prevalence declined in the 67 children , we found no evidence of reduced egg counts/worm burdens in those that reacquired infections and genetic diversity of schistosomes increased over the course of treatment . Our data indicate that this school based MDA program had a strong benefit to individual health as fewer children were infected over time . However , this decline does not appear to be due to schistosome population reduction , and may be caused by either acquired resistance or behavioral changes of the children . In conclusion , control programs based on chemotherapy alone or based on only a subset of the population will need to be supplemented with additional approaches if we are to achieve the WHO goal of eliminating human schistosomiasis by 2025 .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "genetics", "biology", "and", "life", "sciences", "population", "biology", "genomics" ]	2014	No Apparent Reduction in Schistosome Burden or Genetic Diversity Following Four Years of School-Based Mass Drug Administration in Mwea, Central Kenya, a Heavy Transmission Area
The coordination of cell proliferation and cell fate determination is critical during development but the mechanisms through which this is accomplished are unclear . We present evidence that the Snail-related transcription factor CES-1 of Caenorhabditis elegans coordinates these processes in a specific cell lineage . CES-1 can cause loss of cell polarity in the NSM neuroblast . By repressing the transcription of the BH3-only gene egl-1 , CES-1 can also suppress apoptosis in the daughters of the NSM neuroblasts . We now demonstrate that CES-1 also affects cell cycle progression in this lineage . Specifically , we found that CES-1 can repress the transcription of the cdc-25 . 2 gene , which encodes a Cdc25-like phosphatase , thereby enhancing the block in NSM neuroblast division caused by the partial loss of cya-1 , which encodes Cyclin A . Our results indicate that CDC-25 . 2 and CYA-1 control specific cell divisions and that the over-expression of the ces-1 gene leads to incorrect regulation of this functional ‘module’ . Finally , we provide evidence that dnj-11 MIDA1 not only regulate CES-1 activity in the context of cell polarity and apoptosis but also in the context of cell cycle progression . In mammals , the over-expression of Snail-related genes has been implicated in tumorigenesis . Our findings support the notion that the oncogenic potential of Snail-related transcription factors lies in their capability to , simultaneously , affect cell cycle progression , cell polarity and apoptosis and , hence , the coordination of cell proliferation and cell fate determination . Members of the Snail superfamily of zinc-finger transcription factors are essential during development and their deregulation has been implicated in various malignancies including tumorigenesis [1]–[4] . One of the best known functions of Snail-related proteins is their role in the induction of epithelial-mesenchymal transitions ( EMTs ) [1] , [2] , [4] , [5] . EMTs are fundamentally important for normal development and , in particular , for processes such as mesoderm formation , gastrulation and neural tube formation . EMTs are also important for tumorigenesis since they are responsible for the invasive behavior of certain types of tumor cells [1] , [2] , [5] . Hallmarks of EMTs are the loss of apico-basal polarity and adhesive properties , which is critical for the ability of epithelial cells to become migratory . Snail-related proteins contribute to these cellular changes by repressing the transcription of genes that encode factors required for apico-basal polarity and cell adhesion , such as Crumbs and E-cadherin , respectively [6]–[8] . Snail-related proteins have additional cellular functions that can operate independently of the induction of EMT . In Drosophila melanogaster , for example , the Snail family members Snail , Worniu and Escargot are important for both the cell polarity of neuroblasts and the ability of these cells to divide [9] , [10] . Snail , Worniu and Escargot are required for the polarity of embryonic neuroblasts because they promote the expression of the gene inscuteable , which encodes an adaptor protein that , by forming a physical link between the proteins Par3 and Pins , is thought to connect cell polarity to spindle position [10]–[12] . In the case of the division of neuroblasts , Snail , Worniu and Escargot are thought to enhance cell cycle progression by promoting the expression of the gene cdc25string , which encodes a Cdc25 phosphatase homolog required for the removal of inhibitory phosphates on Cyclin-dependent kinases ( CDKs ) and , hence , CDK activation [10] , [13] , [14] . However , whether the effect of Snail , Worniu and Escargot on cdc25string is direct or indirect remains to be determined . In mammals , Snail-related proteins have also been shown to regulate cell proliferation [4] . Specifically , a reduced rate of cell proliferation is observed in cultured epithelial cells transfected with Snail1 ( formerly referred to as ‘Snail’ ) and in regions of the mouse embryo that express endogenous Snail1 [15] . The inhibitory effect of Snail1 expression on cell proliferation is due to the ability of the Snail1 protein to directly repress the transcription of the cyclin D2 gene , which is required for the G1 to S phase transition [15] . In the same study , an inverse correlation was also found between Snail1 expression and apoptosis in the mouse embryo , suggesting that Snail1 can repress apoptosis . Additional evidence that Snail-related transcription factors can repress apoptosis in mammals comes from studies on radiation-induced apoptosis in hematopoietic precursor cells . Snail2 ( formerly referred to as ‘Slug’ ) was found to block apoptosis by repressing the transcription of the pro-apoptotic BH3-only gene puma [16] . The ability of Snail-related transcription factors to block apoptosis was initially discovered in Caenorhabditis elegans and during the analysis of the NSM ( NSM , neuro-secretory motoneuron ) lineages ( Two bilaterally symmetric NSM lineages exist , the left and right NSM lineage ) . About 410 min after the 1st division of the embryo ( referred throughout the manuscript as “1st round of division” ) , the two NSM neuroblasts ( which are generated about 280 min after the 1st division ) divide asymmetrically along the ventral-lateral dorsal-medial axis to each generate two daughter cells of different sizes and different cell fates , the larger NSM , which survives and differentiates into a serotonergic neuro-secretory motorneuron , and the smaller NSM sister cell , which undergoes apoptosis and forms a cell corpse about 30 min after the completion of the NSM neuroblast division [17] , [18] . A dominant gain-of-function ( gf ) mutation of the ces-1 ( ces , cell-death specification ) gene , which encodes a Snail-related protein , was found to block the death of the NSM sister cells and the I2 sister cells [19] , [20] . Otherwise , ces-1 ( n703gf ) animals are indistinguishable from wild-type animals at least at the dissecting microscope level . This ces-1 gf mutation affects a regulatory region of the ces-1 locus , which is likely to results in the over-expression of the ces-1 gene in specific lineages , including the NSM lineage [20] . In ces-1 gf mutants , the CES-1 protein blocks the death of the NSM sister cells by binding to a cis-acting element of the BH3-only gene egl-1 ( egl , egg-laying defective ) , thereby preventing the HLH-2/HLH-3- ( HLH , basic helix-loop-helix transcription factor ) dependent activation of egl-1 transcription [21] . Interestingly , the ces-1 gf mutation also affects the cell polarity of the NSM neuroblast . Specifically , in ces-1 gf mutant animals , the NSM neuroblast divides symmetrically along randomly selected axes rather than dividing asymmetrically along the ventral-lateral dorsal-medial axis [17] . The same polarity defect is observed in animals that lack a functional ces-2 or dnj-11 ( dnj , DnaJ domain ) gene , which encode a HLF-like bZIP transcription and a MIDA1-like chaperone , respectively , and which act upstream of ces-1 to repress ces-1 transcription in the NSM neuroblast [17] , [19] , [20] , [22] . Furthermore , in a wild-type background , expression from a functional Pces-1ces-1::yfp construct is detected only in the larger NSM daughter that is destined to survive ( the NSM ) ; however , in a ces-2 or dnj-11 mutant background expression of this construct is detected in the NSM neuroblast as well as both daughter cells [17] . Based on these findings it has been proposed that in the NSM neuroblast , CES-2 and DNJ-11 maintain ces-1 expression below a certain level and that this is important for the establishment and/or maintenance of NSM neuroblast polarity and the ability of the NSM neuroblast to divide asymmetrically . After the NSM neuroblast divides , CES-1 protein is restricted to the NSM , where it acts as a direct repressor of egl-1 transcription and hence , apoptosis . In ces-2 , dnj-11 or ces-1 gf mutant animals , the level of CES-1 protein in the NSM neuroblast is elevated and this leads to the symmetric , random division of the NSM neuroblast . This results in the formation of two daughters of similar sizes , both of which contain CES-1 protein and , consequently , survive [17] , [23] . We now demonstrate that CES-1 has an additional function in the NSM lineage . Specifically , we present evidence that CES-1 can also regulate cell cycle progression in the NSM neuroblast by functionally interacting with core components of the cell cycle machinery . By simultaneously controlling cell cycle progression , cell polarity and apoptosis , the Snail-related transcription factor CES-1 plays a crucial role in the coordination of cell proliferation and cell fate specification in the NSM lineage . Wild-type larvae carrying the Ptph-1his-24::gfp reporter ( tph , tryptophane hydroxylase; his , histone structural gene ) , which is specifically expressed in serotonergic neurons ( and labels the nuclei of these neurons ) [24] , have two GFP-positive neurons in the anterior pharynx , the left and right NSM ( Figure 1A , +/+ ) . In animals carrying a gf mutation of ces-1 , n703 , the NSM neuroblast divides symmetrically , resulting in two daughter cells of similar sizes , both of which survive [17] , [19] . Therefore , ces-1 ( n703gf ) larvae carrying the Ptph-1his-24::gfp reporter have four GFP-positive neurons in the head region , the left and right NSM and the left and right ‘undead’ NSM sister cell ( Figure 1A , ces-1 ( n703gf ) ) . To identify targets of the CES-1 protein involved in the asymmetric division of the NSM neuroblast , we performed a ces-1 ( n703gf ) suppressor screen using the Ptph-1his-24::gfp reporter as a tool . Specifically , we screened mutagenized ces-1 ( n703gf ) animals for mutations that cause a reduction in the number of GFP-positive NSMs and undead NSM sister cells ( i . e . less than four GFP-positive cells in the anterior pharynx ) . Using this approach , we isolated the mutation bc416 . At 15°C , 100% of ces-1 ( n703gf ) larvae have four GFP-positive cells . In contrast , only 5% of ces-1 ( n703gf ) larvae homozygous for bc416 ( ces-1 ( n703gf ) ; bc416 ) have four GFP-positive cells ( Table 1 ) . The bc416 mutation is recessive and does not show maternal rescue ( data not shown; Table S1 ) . At 15°C , 33% of ces-1 ( n703gf ) ; bc416 larvae have three GFP-positive cells ( Table 1 ) . Interestingly , based on Ptph-1his-24::gfp labeling , in these animals , one nucleus is larger than the other two nuclei ( Figure 1A , ces-1 ( n703gf ) ; bc416 ) . This phenomenon is observed at a high frequency . Based on this observation , we hypothesized that instead of suppressing the inappropriate survival of NSM sister cells in ces-1 ( n703gf ) animals , the bc416 mutation might affect the division of the NSM neuroblasts . To test this , we directly analyzed the division of the NSM neuroblasts in ces-1 ( n703gf ) ; bc416 embryos . To that end , we used a transgene that expresses a plasma membrane-targeted mCherry fusion protein as a tool [25] . We identified the NSM neuroblasts based on their positions during the comma stage of embryogenesis and tracked their fates until the 2-fold stage , which is the stage during which in wild-type animals , the NSM neuroblasts complete their division and the NSM sister cells undergo apoptosis [17] , [18] . We found that at 15°C , only 13% of the NSM neuroblasts divide in ces-1 ( n703gf ) ; bc416 embryos ( Figure 1B , C ) . To rule out the possibility that the division of the NSM neuroblasts in ces-1 ( n703gf ) ; bc416 animals is delayed rather than blocked , we scored ces-1 ( n703gf ) ; bc416 animals in the background of the ced-3 loss-of-function mutation n717 , which causes a general block in apoptosis [26] . If the NSM neuroblasts divided in late embryos or in larvae and the resulting NSM sister cells underwent apoptosis , using Ptph-1his-24::gfp as a tool , we should observed an increased number of animals with four GFP-positive cells in the ced-3 ( n717 ) background . However , we found that the percentage of animals with two , three , or four GFP-positive cells is not affected by ced-3 ( n717 ) ( Table 1 , ces-1 ( n703gf ) ; bc416; ced-3 ( n717 ) ) . Therefore , the reduction in GFP-positive cells observed in ces-1 ( n703gf ) ; bc416 larvae is the result of a failure of the NSM neuroblasts to divide . For this reason , we are presenting the data acquired in larvae using Ptph-1his-24::gfp not only in the form of ‘% animals with two , three , or four GFP-positive cells’ but also as ‘% NSM neuroblasts dividing’ , which is defined as the percentage of the NSM neuroblasts that divide ( Table 1 ) . To further examine the cell cycle defect in ces-1 ( n703gf ) ; bc416 animals , we determined the relative DNA content in non-dividing NSM neuroblasts . We labeled DNA in ces-1 ( n703gf ) ; bc416 animals with the fluorescent dye DAPI and measured fluorescence intensity in non-dividing NSM neuroblasts [27] , [28] , in NSMs and in undead NSM sister cells . We found that in ces-1 ( n703gf ) ; bc416 mutants with three GFP-positive cells , the average DNA content of the cells with the larger nuclei ( presumably the non-dividing NSM neuroblasts ) is 1 . 8 times greater than the average DNA content of the cells with the smaller nuclei ( presumably the NSMs and undead NSM sister cells ) ( Figure S1 ) . Furthermore , in ces-1 ( n703gf ) ; bc416 animals with two GFP-positive cells , the average DNA content of both cells is about 2-fold higher than that of control pharyngeal muscle cells ( data not shown ) . Taken together , these observations suggest that in ces-1 ( n703gf ) ; bc416 animals , the NSM neuroblasts complete DNA replication but fail to undergo mitosis . Hence , in this mutant background , the NSM neuroblasts arrest between S phase and M phase . Interestingly , just like NSMs , the non-dividing , tetraploid NSM neuroblasts express the Ptph-1his-24::gfp reporter during larval stages , which suggests that they differentiate into serotonergic neurons . Besides exhibiting a defect in NSM neuroblast division , ces-1 ( n703gf ) ; bc416 animals have additional defects . When raised at 15°C or 25°C , 8% or 76% of ces-1 ( n703gf ) ; bc416 animals , respectively , exhibit an embryonic lethal ( Emb ) phenotype and arrest at the elongation stage during embryogenesis ( Figure 2A , B ) . Arrested embryos have multiple defects in hypodermal morphogenesis ( Figure 2B ) . Since the Emb phenotype is temperature sensitive , we performed temperature-shift experiments to define the temperature-sensitive period ( TSP ) of ces-1 ( n703gf ) ; bc416 animals . Embryos were shifted from 25°C to 15°C or vice versa at different stages during embryonic development , and viability was assessed 24 h to 48 h later . As shown in Figure S2 , the TSP of ces-1 ( n703gf ) ; bc416 animals lies between the 50-cell stage and the comma stage of embryogenesis . Therefore , at least in the ces-1 ( n703gf ) mutant background , the gene defined by the bc416 mutation is essential and its activity required for embryonic development between the 50-cell stage and the comma stage . Furthermore , animals that escape embryonic lethality and hatch also display morphological abnormalities ( Figure 2C ) . Finally , when raised at 25°C , about 30% of ces-1 ( n703gf ) ; bc416 animals that develop into adults are sterile ( Ste phenotype ) ( data not shown ) and the average brood size of the fertile ces-1 ( n703gf ) ; bc416 adults is smaller than that of wild-type adults ( Table S2 ) . Since a highly penetrant Emb phenotype was observed in ces-1 ( n703gf ) ; bc416 animals raised at 25°C , we investigated whether cell divisions other than the divisions of the NSM neuroblasts are affected in these animals . A systematic analysis of all cell lineages using 4D lineage analysis [29] showed that cell division defects are not restricted to the NSM lineage . We found that at 25°C , the ABarp , C and E lineages are also affected in ces-1 ( n703gf ) ; bc416 animals ( Figure 3 ) . All other lineages were not affected . ABarp is a major hypodermal precursor and the C founder cell generates additional posterior and dorsal hypodermal cells [18] . In the ABarp lineage , most cell divisions that give rise to ventrolateral ectoblasts ( V1 to V6 ) are blocked ( Figure 3 ) . Furthermore , in the C lineage , many cell divisions that generate the embryonic large hypodermal syncytium ( hyp7 ) fail to occur ( Figure 3 ) . The defects in the ABarp and C lineage most likely cause or contribute to the observed hypodermal abnormalities ( Figure 2B , C ) . A failure in the formation of the hypodermis has previously been shown to cause embryonic lethality [30] . In addition , the phenotype of arrested ces-1 ( n703gf ) ; bc416 embryos is similar to the phenotype of mutants with hypodermal defects [30] . Therefore , the cell division defects in the ABarp and C lineages observed in ces-1 ( n703gf ) ; bc416 animals most probably cause the Emb phenotype exhibited by these animals . In addition , we identified variable defects in the E lineage . Specifically , some cell divisions of the 7th round of division during C . elegans embryogenesis do not occur in the E lineage ( Figure 3 ) . Based on these observations , we conclude that in the ces-1 ( n703gf ) background and at 25°C , bc416 affects the divisions of cells other than the NSM neuroblasts . ( The defects caused by bc416 in an otherwise wild-type background will be discussed below . ) The bc416 mutation was mapped genetically to a 900 kb region ( between SNPs F22B7:15755 and ZK1098:19075 ) on LGIII using linkage analysis , three-factor mapping and SNP mapping ( Figure S3A ) . In addition , we used Illumina deep sequencing technology to sequence the entire genome of ces-1 ( n703gf ) ; bc416 animals . In the F22B7:15755 - ZK1098:19075 region , we found a G to A transition at the conserved 5′ splice-donor site of the first intron of the gene cya-1 ( ZK507 . 6 ) , which encodes one of two C . elegans Cyclin A homologs , CYA-1 ( Figure S3B ) [31] . A 4 . 3 kb genomic DNA fragment that contains the entire coding region of cya-1 rescues the NSM neuroblast division defect ( Table 1; ces-1 ( n703gf ) ; bc416 plus ‘cya-1’ transgene ) and the Emb phenotype observed in ces-1 ( n703gf ) ; bc416 animals ( data not shown ) . Similar to bc416 , partially reducing cya-1 function by RNA-mediated interference ( RNAi ) blocks 90% of the NSM neuroblast divisions in the ces-1 ( n703gf ) mutant background ( Table 1; ces-1 ( n703gf ) ; cya-1 ( RNAi ) ) . In addition , cya-1 ( RNAi ) leads to embryonic lethality and the terminal phenotype of arrested embryos is similar to the terminal phenotype of arrested ces-1 ( n703gf ) ; bc416 embryos ( Figure 2D ) . ( For the cya-1 RNAi experiment , sequences of exons 4 and 5 of cya-1 were used . Sequence alignments reveal that these two exons are highly homologous [≥60%] to cya-2 , the second C . elegans cyclin A gene . For this reason , it is possible that cya-1 ( RNAi ) also causes a decrease in cya-2 function . ) Finally , a null mutation of the cya-1 gene , he153 , which causes embryonic lethality , fails to complement bc416 in the ces-1 ( n703gf ) mutant background ( data not shown; S . van der Heuvel , personal communication ) . In conclusion , the gene defined by bc416 is identical to the cya-1 gene . Since the accuracy of the 5′ splice-donor site is important for the recognition and removal of introns , we determined whether the bc416 mutation influences the splicing of the primary cya-1 transcript . Using reverse transcriptase PCR ( RT-PCR ) , we found that , at both 15°C and 25°C , bc416 affects the splicing of the cya-1 gene and results in aberrantly spliced messages , in which parts of the first intron are retained ( Figure S3D ) . The translation of these aberrant messages is predicted to result in the synthesis of a truncated , non-functional CYA-1 protein that includes only the first 12 amino acids of the full-length protein . Using quantitative real-time PCR ( qPCR ) , next , we determined the level of correctly spliced , wild-type cya-1 transcript . We found that at both 15°C and 25°C , compared to wild-type ( cya-1 ( +/+ ) ) animals , the level of correctly spliced cya-1 transcript is reduced by about 50% in cya-1 ( bc416 ) animals ( Figure S3E ) . These results suggest that bc416 affects the pre-mRNA splicing of the cya-1 gene , resulting in a reduction of correctly spliced mRNA and hence , presumably , full-length CYA-1 protein . Therefore , bc416 most likely represents a partial loss-of-function ( lf ) mutation of cya-1 . While performing cya-1 RNAi experiments , we noticed that cya-1 RNAi by injection results in a more penetrant Emb phenotype when performed in the ces-1 ( n703gf ) background ( 64% embryonic lethality at 25°C ) compared to the wild-type background ( 40% embryonic lethality at 25°C ) . To test whether the defects observed in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals are dependent on the presence of the ces-1 ( n703gf ) mutation , we analyzed bc416 in an otherwise wild-type background . Using the plasma membrane-targeted mCherry fusion protein as a tool , we found that at 15°C , 80% of the NSM neuroblasts divide in cya-1 ( bc416 ) embryos ( Figure 1C ) . For comparison , only 13% of the NSM neuroblasts divide in ces-1 ( n703gf ) ; cya-1 ( bc416 ) embryos . Similarly , using the Ptph-1his-24::gfp reporter as a tool , we found that 92% of NSM neuroblasts divide in cya-1 ( bc416 ) animals in the ced-3 mutant background ( Table 1 , cya-1 ( bc416 ) ; ced-3 ( n717 ) ) . For comparison , in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals in the ced-3 mutant background only 25% of the NSM neuroblasts divide . Finally , we found that at 25°C , 40% of cya-1 ( bc416 ) animals arrest at the elongation stage during embryogenesis and therefore exhibit an Emb phenotype ( Figure 2A ) . For comparison , 76% of ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals exhibited an Emb phenotype when raised at 25°C . Temperature-shift experiments revealed that the TSP of cya-1 ( bc416 ) animals lies between the 50-cell stage and the comma stage of embryogenesis and therefore is identical to the TSP of ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals ( Figure S2 ) . Lineage analysis revealed that cya-1 ( bc416 ) embryos have cell division defects in the ABarp , C and E lineages ( Figure S4 ) . However , cell division defects in the ABarp and C lineages were only observed in one out of three cya-1 ( bc416 ) embryos analyzed . For comparison , cell division defects in the ABarp and C lineages were observed in all three ces-1 ( n703gf ) ; cya-1 ( bc416 ) embryos analyzed . Lineage analysis also revealed that the ces-1 ( n703gf ) embryos have no cell division defects in the ABarp , C or E lineages ( Figure S4 ) . Therefore , ces-1 ( n703gf ) increases the penetrance of the cell division defects caused by cya-1 ( bc416 ) , especially in the ABarp and C lineages . This is consistent with the more penetrant Emb phenotype observed in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals ( 76% embryonic lethality at 25°C , Figure 2 ) . These findings demonstrate that ces-1 ( n703gf ) enhances the NSM neuroblast division defect and the Emb phenotype caused by cya-1 ( bc416 ) . Interestingly , ces-1 ( n703gf ) does not enhance the defect in brood size caused by cya-1 ( bc416 ) . When raised at 25°C , the brood size of both cya-1 ( bc416 ) animals and ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals is reduced by about 2 . 5-fold when compared to the brood size of wild-type animals ( Table S2 ) . Snail-related transcription factors are thought to predominantly act as repressors of transcription [1] . To determine the mechanism through which ces-1 ( n703gf ) enhances cya-1 ( bc416 ) , we therefore analyzed the expression of candidate target genes . The Drosophila melanogaster Snail family has been implicated in the control of the expression of the gene cdc25string , which encodes the D . melanogaster ortholog of Cdc25 [10] , [13] , [14] . For this reason , we analyzed the level of expression of the four C . elegans cdc25 homologs ( cdc-25 . 1 , cdc-25 . 2 , cdc-25 . 3 and cdc-25 . 4 ) [32] in wild-type animals and in animals over-expressing the ces-1 gene using qPCR . To that end , using a heat-inducible promoter , ces-1 expression was induced for 1 h in embryos and embryos were collected after a 1 . 5 h recovery period . This induction scheme resulted in a 3-fold increase in the relative expression level of ces-1 ( Figure 4 ) . Using this experimental set-up , we found that the relative expression levels of cdc-25 . 1 and cdc-25 . 4 are not significantly changed in embryos over-expressing ces-1 . In contrast , the relative expression level of cdc-25 . 2 is significantly decreased , indicating that ces-1 over-expression can repress the expression of cdc-25 . 2 ( Figure 4 ) . We also found that the relative expression level of cdc-25 . 3 is significantly increased . Since CES-1 is thought to predominantly act as repressor of transcription and since it has been suggested that mammalian Cdc25A and Cdc25B may compensate for the loss of Cdc25C [33] , [34] , we analyzed whether the increase in the relative expression level of cdc-25 . 3 in embryos over-expressing ces-1 is an indirect effect that is triggered by decreased cdc-25 . 2 expression . We found that the relative expression level of cdc-25 . 3 is not increased in embryos in which cdc-25 . 2 function is knocked-down by RNAi ( Figure S5 ) . This suggests that , independently of decreasing cdc-25 . 2 expression , ces-1 over-expression increases cdc-25 . 3 expression . Finally , the relative expression level of cya-1 expression is not significantly changed in embryos over-expressing ces-1 ( Figure 4 ) . In summary , our data indicate that CES-1 directly or indirectly represses the expression of cdc-25 . 2 . To determine whether CES-1 directly controls the transcription of the cdc-25 . 2 gene , we analyzed CES-1 ChIP-seq ( ChIP-seq , chromatin immunoprecipitation with massively parallel DNA sequencing ) data acquired by the modENCODE Project ( http://www . modencode . org/ ) [35] , [36] ( M . Snyder , S . Kim , T . Kawli , personal communication ) . This data was acquired using , as starting material , embryos that harbor multiple copies of an engineered , stably integrated ces-1 fosmid , which expresses a ces-1::gfp transgene under the endogenous ces-1 promoter [37] , [38] . ( We have previously shown that a Pces-1ces-1::yfp transgene can rescue the ces-1 loss-of-function mutant phenotype and , hence , is generating a fusion protein that is functional [17] . ) The ChIP-seq data obtained indicate that CES-1 binds to a 1 . 7 kb region that is located 4 . 8 kb to 6 . 5 kb upstream of the predicted transcriptional start site of cdc-25 . 2 ( Figure 5 ) ( Integrated Genome Browser [39] ) . In contrast , we did not identify peaks indicative of CES-1 binding sites in the immediate regions 5′ or 3′ of the predicted cdc-25 . 1 , cdc-25 . 3 or cdc-25 . 4 transcription units , nor within their introns ( Figure 5 ) . Based on these findings we conclude that cdc-25 . 2 most likely is a direct target of CES-1 and , hence , that the effect of ces-1 over-expression on the relative expression level of cdc-25 . 2 is a direct effect . The over-expression of ces-1 results in a reduction of the relative expression level of cdc-25 . 2 by about 30% ( Figure 4 ) . To determine whether a decrease in cdc-25 . 2 dosage by 50% is sufficient to enhance the NSM neuroblast division defect caused by cya-1 ( bc416 ) , we analyzed cya-1 ( bc416 ) animals heterozygous for cdc-25 . 2 ( ok597 ) , a deletion of the cdc-25 . 2 gene that removes 2 . 7 kb of the cdc-25 . 2 locus , including four of its six exons [40] . We found that cdc-25 . 2 ( ok597 ) /+; cya-1 ( bc416 ) animals exhibit a NSM neuroblast division defect similar to the defect observed in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals . Specifically , at 15°C , 25% and 33% of the NSM neuroblasts divide in ces-1 ( n703gf ) ; cya-1 ( bc416 ) or cdc-25 . 2 ( ok597 ) /+; cya-1 ( bc416 ) animals , respectively ( in the ced-3 mutant background ) ( Table 1 ) . Conversely , we tested whether the transgenic expression of the cdc-25 . 2 transcription unit under the control of the endogenous cdc-25 . 2 promoter can rescue the NSM neuroblast division defect observed in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals . We found that the expression of cdc-25 . 2 significantly reduces the NSM neuroblast division defect in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals . Specifically , at 15°C , the expression of cdc-25 . 2 increases the percentage of NSM neuroblasts dividing from 22% to 82% ( Table 1; ces-1 ( n703gf ) ; cya-1 ( bc416 ) plus ‘cdc-25 . 2’ transgene ) . Together , these findings support the notion that ces-1 ( n703gf ) enhances the NSM neuroblast division defect caused by cya-1 ( bc416 ) by decreasing cdc-25 . 2 expression . Next , we analyzed the phenotypes caused by the downregulation of the four cdc-25 genes . We found that the downregulation by RNAi of cdc-25 . 1 or cdc-25 . 2 results in embryonic lethality . In contrast , the downregulation by RNAi of cdc-25 . 3 or cdc-25 . 4 does not cause any obvious abnormalities , which is consistent with previous observations [32] . While cdc-25 . 1 ( RNAi ) embryos arrest during early embryonic stages ( as early as the 4-cell stage ) ( data not shown ) , cdc-25 . 2 ( RNAi ) embryos arrest at the elongation stage during embryogenesis ( Figure 2E , cdc-25 . 2 ( RNAi ) ) . Using lineage analyses , we determined the phenotype of cdc-25 . 2 ( RNAi ) embryos in more detail . We found that the inactivation of cdc-25 . 2 causes an increase in cell cycle length in all lineages ( Table S4 ) . For example , in wild-type animals , the average time between the 6th and 7th and between the 7th and 8th round of division in the ABala lineage is 28 min and 36 min , respectively . In cdc-25 . 2 ( RNAi ) animals , the average time is 36 min and 52 min , respectively . In general , many cell divisions of the last three rounds of division during embryonic development are blocked ( Figure 3 and Figure S6 ) . For example , the divisions of the NSM neuroblasts are blocked in cdc-25 . 2 ( RNAi ) embryos ( Figure 1C ) . In addition , the ABarp lineage is particularly sensitive to reduced levels of cdc-25 . 2 function . Most of the cell divisions during the 9th round of division are blocked in the ABarp lineage ( Table S3 , Figure 3 and Figure S6 ) . For comparison , in the case of other AB descendants , only half or less than half of the cell divisions during the 9th round of division are blocked ( Table S3 ) . We also observed severe cell division defects in the C and E lineages in cdc-25 . 2 ( RNAi ) animals ( Figure 3 and Figure S6 ) . Interestingly , the cell lineages that exhibit increased sensitivity to reduced levels of cdc-25 . 2 activity are identical to the cell lineages that are affected in ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals . In summary , these findings support the notion that , in a specific set of lineages , ces-1 , cdc-25 . 2 and cya-1 act together to control cell cycle progression . These lineages and cells include the ABarp , C and E lineages as well as the NSM neuroblasts . Finally , to determine whether the loss of ces-1 function affects the phenotype caused by cya-1 ( bc416 ) , we analyzed the NSM neuroblast division in animals homozygous for cya-1 ( bc416 ) and the ces-1 deletion allele tm1036 . tm1036 is a 1 . 3 kb deletion that removes exons 2 , 3 and 4 of the ces-1 transcription unit and that is predicted to result in the synthesis of a truncated protein lacking two of the five zinc-finger domains of the CES-1 protein [20] . ces-1 ( tm1036 ) animals are indistinguishable from wild-type animals at the dissection microscope level , and in an otherwise wild-type background , the loss of ces-1 function causes no obvious phenotype in the NSM lineage [17] , [19] , [20] . We found that , in the ced-3 mutant background , ces-1 ( tm1036 ) does not suppress the NSM neuroblast division defect caused by cya-1 ( bc416 ) ( Table 1 ) . However , ces-1 ( tm1036 ) reduces the embryonic lethality caused by cya-1 ( bc416 ) . While 40% of cya-1 ( bc416 ) animals exhibit an Emb phenotype , only 26% of ces-1 ( tm1036 ) ; cya-1 ( bc416 ) animals exhibit an Emb phenotype ( Figure 2A ) . Based on these observations , we suggest that ces-1 may play a role in the control of cell cycle progression at least in certain cell lineages . Like ces-1 ( n703gf ) , the loss of dnj-11 function causes symmetric NSM neuroblast division and inappropriate NSM sister cell survival [17] , [19] . Therefore , we determined whether the loss of dnj-11 function also enhances the NSM neuroblast division defect caused by cya-1 ( bc416 ) . We found that animals homozygous for cya-1 ( bc416 ) and dnj-11 ( tm2859 ) , a deletion allele of dnj-11 that removes 614 base pairs of the coding region [17] , exhibit a partially penetrant Emb phenotype ( data not shown ) . Whereas in viable larvae 61% of the NSM neuroblasts had divided , only 30% of the NSM neuroblasts divide in dnj-11 ( tm2859 ) ; cya-1 ( bc416 ) animals that arrest during embyogenesis ( at 15°C , Table 1 ) . For comparison , 92% of NSM neuroblasts divide in cya-1 ( bc416 ) animals raised at 15°C . These findings demonstrate that , like ces-1 ( n703gf ) , the loss of dnj-11 function enhances the NSM neuroblast division defect caused by cya-1 ( bc416 ) . Based on these findings , we conclude that dnj-11 regulates ces-1 function also in the context of cell cycle progression . We have isolated and characterized a hypomorphic allele of the cya-1 gene , one of two C . elegans cyclin A genes [31] . This cya-1 mutation , bc416 , presumably results in a reduction in the level of CYA-1 protein thereby causing cell division defects in specific lineages ( ABarp , C , E and NSM lineages ) and partially penetrant embryonic lethality . Given that animals homozygous for a cya-1 deletion allele are not viable ( S . van der Heuvel , personal communication ) , we conclude that cya-1 is essential for embryogenesis . We also present evidence that cya-1 ( bc416 ) causes a block in cell cycle progression between S phase and M phase , which is consistent with the proposed function of the CYA-1 protein , as predicted based on the function of Cyclin A in other organisms [31] , [41] . Based on the function of Cyclin A in other organisms , we also speculate that it is the C . elegans CDKs CDK-1 and/or CDK-2 that CYA-1 binds to and activates [31] , [42] . Furthermore , we provide evidence that in the ABarp , C , E and NSM lineages , cya-1 acts with cdc-25 . 2 to cause CDK activation and , hence , cell cycle progression . Interestingly , in cya-1 ( bc416 ) animals , at 15°C around 10% of the NSM neuroblast divisions are blocked , while no block in cell division was observed in the ABarp , C or E lineage ( Table 1 and Figure 1C ) . However , at 25°C , a cell division defect was not observed for the NSM neuroblasts , while cell division defects were observed in the ABarp , C and E lineages in around 30% of the cya-1 ( bc416 ) embryos ( Figure 1C and Figure S4 ) . Therefore , the cell division defect in the NSM lineage is more severe at 15°C , whereas the cell division defects in the ABarp , C and E lineages are more severe at 25°C . We found that compared to wild-type animals , the reduction in the level of correctly spliced cya-1 transcript is similar at both 15°C and 25°C in cya-1 ( bc416 ) animals ( Figure S3E ) . This suggests that the splicing defect is not temperature sensitive . Why different cell lineages ( NSM neuroblast , ABarp , C and E lineages ) respond differently to a reduction in cya-1 function and why their responses differ in their temperature sensitivity is unclear . Lineage or tissue-specific control of cell cycle progression has previously been observed in C . elegans [42]–[46] . One determining factor could be cell cycle length . Since cell cycle length is influenced by temperature and diverges greatly in different lineages , this behavior might reflect different CDK-activity thresholds for different lineages and/or differential regulation of CDKs within specific lineages at different temperatures [42]–[46] . We propose that the dnj-11 MIDA1 , ces-2 HLF , ces-1 Snail pathway , which has previously been shown to control asymmetric cell division and apoptosis in the NSM lineage [17] , [19] , [20] , also controls cell cycle progression in this lineage ( Figure 6 ) . Specifically , we demonstrate that the dnj-11 loss-of-function mutation tm2859 or the ces-1 gain-of-function mutation n703 enhances a defect in NSM neuroblast division caused by cya-1 ( bc416 ) . Furthermore , we have uncovered the molecular mechanism through which this pathway controls cell cycle progression in this lineage . We provide evidence in support of the notion that the Snail-related transcriptional repressor CES-1 directly represses the transcription of the cdc-25 . 2 gene thereby decreasing the level of CDC-25 . 2 . While in an otherwise wild-type background , this does not lead to a block in NSM neuroblast division , it does cause a block in NSM neuroblast division in a cya-1 ( bc416 ) mutant background , in which the level of CYA-1 presumably is reduced . The observation that the loss of dnj-11 function or the ces-1 gain-of-function mutation n703 are synthetic lethal with cya-1 ( bc416 ) furthermore suggests that the dnj-11 MIDA1 , ces-2 HLF , ces-1 Snail pathway may act to control cya-1- and cdc-25 . 2-dependent cell cycle progression in lineages other than the NSM lineage . Unlike dnj-11 ( tm2859 ) and ces-1 ( n703gf ) , the ces-1 deletion allele tm1036 did not affect the NSM neuroblast division defect caused by cya-1 ( bc416 ) . This finding suggests that ces-1 may not have a physiological role in cell cycle progression in the NSM neuroblast . Alternatively , the function of ces-1 in cell cycle progression in the NSM neuroblast may be redundant with that of another gene or genes . Interestingly , the functions of the D . melanogaster Snail-related genes snail , escargot and worniu in cell cycle progression and polarity in embryonic neuroblasts are redundant and defects are only observed in animals in which all three genes are inactivated [9] , [10] . Apart from ces-1 , the C . elegans genome contains at least two additional genes that encode Snail-related transcription factors , scrt-1 and K02D7 . 2 ( http://www . wormbase . org ) [1] . Therefore , we speculate that the functions of ces-1 in cell cycle progression , polarity and apoptosis in the NSM lineage are redundant with the functions in these processes of scrt-1 and K02D7 . 2 . Finally , the observation that ces-1 ( n703gf ) enhances but ces-1 ( tm1036 ) partially suppresses the embryonic lethality caused by cya-1 ( bc416 ) supports the notion that ces-1 has a physiological role in cell cycle progression in some cell lineages , such as the ABarp and C lineages . Members of the Snail superfamily have previously been shown to affect cell cycle progression . Mammalian Snail1 has been shown to block cell cycle progression in cultured epithelial cells or in mouse embryos , and this effect appears to be mediated through the direct repression of cyclin D2 transcription [15] . In contrast , over-expression of the Snail1 gene in mouse epidermis causes hyperproliferation [47] . In D . melanogaster , the Snail-related proteins Snail , Escargot and Worniu have been shown to promote cell cycle progression in embryonic neuroblasts in part by , directly or indirectly , promoting cdc25string expression [10] . Cdc25string is a critical regulator of M phase during D . melanogaster development , whose activity is regulated at the transcriptional level [13] , [14] . In support of the model that cdc25string acts as an integrator of signals that regulate cell division during D . melanogaster development , the cdc25string locus is subject to complex transcriptional regulation . Interestingly , it has been shown that in D . melanogaster larval neuroblasts , the level of Worniu has to be precisely regulated as well . A low level of Worniu in larval neuroblasts leads to a delay in cell cycle progression and premature differentiation , whereas an elevated level of Worniu results in cell cycle arrest due to increased Prospero expression [48] . These findings suggest that the roles of Snail-related proteins in cell cycle progression are complex and might be cell- or tissue-type specific . However , so far , no Snail-related transcription factor has been implicated in the cdc25-mediated block of cell cycle progression . Here we have identified a new mechanism through which Snail-related proteins can block cell cycle progression . Specifically , we present evidence that , by binding to a region 4 . 8 kb to 6 . 5 kb upstream of the cdc-25 . 2 transcription unit , CES-1 most likely directly represses cdc-25 . 2 transcription thereby causing a block in cell cycle progression in a sensitized background ( the cya-1 ( bc416 ) background ) . Interestingly , the region 4 . 8 kb to 6 . 5 kb upstream of the cdc-25 . 2 transcription unit is at least partially conserved in other Caenorhabditis species such as Caenorhabditis remanei or Caenorhabditis briggsae ( UCSC genome browser http://genome . ucsc . edu/ [49] ) . In analogy to D . melanogaster cdc25string , this finding suggests that the transcriptional regulation of cdc-25 . 2 might be an important aspect of the developmental control of cell division in C . elegans , a notion that is supported by the observation that the cdc-25 . 2 transcription unit is flanked by extensive intergenic regions that are conserved ( UCSC genome browser [49] ) . As mentioned above , while D . melanogaster Snail , Escargot und Worniu promote cell cycle progression and cdc25string expression , C . elegans CES-1 blocks cell cycle progression and represses cdc-25 . 2 expression . Interestingly , we found that besides repressing cdc-25 . 2 transcription , the over-expression of ces-1 directly or indirectly increases the relative level of cdc-25 . 3 , which encodes another member of the Cdc25 phosphatase family of C . elegans . Hence , depending on the cell lineage and cellular context , members of the Snail superfamily may enhance or repress the expression of Cdc25 phosphatases . The over-expression of Snail-related transcription factors has been implicated in the formation and progression of metastatic cancers , in part due to the ability of Snail-related transcription factors to induce EMTs [1]–[4] , [50] . Their potency as proto-oncogenes is thought to lie in their capability to cause loss of cell polarity and adhesive functions on the one hand and acquisition of migratory properties on the other . We argue that their ability to , within the same cell lineage , also block cell cycle progression and apoptosis is similarly important for the formation of metastases . Cells undergoing EMT have high invasive potential and are primarily found at the margins of tumors . Whether EMT has to be accompanied by a reduction in proliferation is a question still under debate . It has previously been shown that increased expression of the gene encoding the transcription factor YB-1 ( Y-box binding protein ) , which is frequently observed in human cancers and which results in increased Snail1 expression , induces EMT accompanied by enhanced metastatic potential and reduced cellular proliferation [51] , [52] . Here we demonstrate that the ces-1 gf mutation not only affects cell polarity and apoptosis in the NSM lineage but also enhances the defect in cell cycle progression caused by a partial cya-1 loss-of-function mutation . These findings support the notion that a block in cell cycle progression and , hence , cell proliferation may be important for EMT . A block in cell cycle progression could , for example , provide the time necessary for cytoskeletal reorganizations or cell polarity transitions . Finally , Snail-related transcription factors have recently been implicated in the acquisition and maintenance of the stem cell fate in mammals [4] , [53]–[55] . We speculate that it is the ability of Snail-like transcription factors to coordinately influence cell cycle progression , cell polarity and apoptosis that allows specific cells to adopt and maintain the cancer stem cell fate . Models of metastatic cancers have mainly focused on the analysis of the starting and end points of the cellular transformations that cells undergo during the formation of metastases ( i . e . the epithelial and mesenchymal phenotype ) [5] . There is a need for in vivo models that allow the analysis of intermediate stages of this process . We suggest that the over-expression of the Snail-related gene ces-1 in C . elegans ( i . e . the ces-1 gf phenotype ) may serve as such a model at least for certain stages of this process . The ability to combine systems biology approaches ( such as ChIP-seq analyses ) with cell biological and genetic dissection at single cell resolution will allow us to further dissect the complex role of Snail-like transcription factors during normal development and tumorigenesis . C . elegans strains were maintained and cultured as described [56] . Bristol N2 was used as the wild-type strain , unless noted otherwise . CB4856 ( Hawaii ) was used for SNP mapping . Mutations and transgenes used in this study are listed below and are described by Riddle et al . unless noted otherwise [57]: LGI: ces-1 ( n703gf ) , ces-1 ( tm1036 ) ( National BioResource Project ) ( 3 times backcrossed ) . LGII: rrf-3 ( pk1426 ) [58] . LGIII: dpy-17 ( e164 ) , cya-1 ( bc416 ) ( this study ) ( 5 times backcrossed ) , bcIs66 ( Ptph-1his-24::gfp ) ( this study ) , unc-69 ( e587 ) , ltIs38 ( Ppie-1gfp::phPLC1δ ) [25] . LGIV: ced-3 ( n717 ) , dnj-11 ( tm2859 ) [17] . LGV: ltIs44 ( Ppie-1mCherry::phPLC1δ ) [25] , cdc-25 . 2 ( ok597 ) [40] . LGX: lin-15 ( n765ts ) . The plasmids pBC1153 ( cdc-25 . 2 ) and pBC1282 ( cdc-25 . 3 ) used for in vitro transcription of double-strand ( ds ) RNA were generated by cloning PCR fragments containing exons of the targeted genes into the EcoRV site of pBluescript II KS+ . The plasmid pBC1098 ( cya-1 ) , which was used for RNAi by feeding as well as in vitro transcription of dsRNA , was generated by cloning a PCR fragment containing genomic DNA of the cya-1 locus into the NcoI and XmaI sites of pPD129 . 36 [59] . The plasmids pMM#47 ( PHSces-1; pPD49 . 78 based ) and pMM#48 ( PHSces-1; pPD49 . 83 based ) were generated using a full-length ces-1 cDNA ( R . H . Horvitz and M . M . Metzstein , personal communication ) . For rescue experiments , DNA fragments containing the gene of interests ( including regulatory regions ) were amplified by PCR ( NEB LongAmp Taq ) and purified . The sequences of oligonucleotides used for PCR are provided in Table S5 . RNAi by feeding was performed as described by Fire and co-workers [59] using 6 mM IPTG . For RNAi experiments by microinjection [60] , pBC1153 ( cdc-25 . 2 ) and pBC1282 ( cdc-25 . 3 ) were used as templates and oligonucleotides 5′-ttgtaaaacgacggccag-3′ and 5′-catgattacgccaagcgc-3′ as primers to generate PCR products containing at their ends , either the T3 or T7 promoter . pBC1098 ( cya-1 ) was used as template and oligonucleotides 5′- taatacgactcactataggg-3′ as primer to generate PCR products containing T7 promoter at both ends . These PCR products were used to synthesize dsRNA in vitro using T3 and T7 polymerase ( Ambion ) . RNAi was performed by microinjection of dsRNA into young adults . Injected animals were incubated at 25°C for at least 20 h and the phenotype of their progeny was determined . Germline transformation was performed as described [61] . For rescue experiments , ces-1 ( n703gf ) ; cya-1 ( bc416 ) animals were injected with purified PCR products ( 0 . 5–6 ng/µl ) using pRF4 ( 100 ng/µl ) as coinjection marker , which confers a dominant Rol phenotype . For ces-1 over-expression experiment , pMM#47 ( 5 ng/µl ) and pMM#48 ( 5 ng/µl ) were injected into lin-15 ( n765ts ) animals using pL15EK ( 80 ng/µl ) , which rescues the Muv phenotype caused by lin-15 ( n765ts ) , as coinjection marker . The NSMs and the surviving NSM sister cells were identified in L3 or L4 larvae carrying the Ptph-1his-24::gfp reporter using fluorescence microscopy essentially as described for Ptph-1gfp [21] . The division of the NSM neuroblast was analyzed in embryos using a plasma membrane-targeted GFP fusion protein ( Ppie-1gfp::phPLC1δ ) or mCherry fusion protein ( Ppie-1mCherry::phPLC1δ ) as described [17] , [25] . Embryos were imaged using 4D microscopy and cell lineage analysis was performed using a Zeiss Imager microscope and SIMIBioCell software ( Simi Reality Motion Systems GmbH , Unterschleissheim , Germany ) as described [29] . L4 larvae were fixed and stained with DAPI as described [62] , with the exception that slides were mounted in 1 . 0 µg/ml DAPI in PBS diluted 1∶1 with VectaShield ( Vector Laboratories ) . Fluorescence intensities were measured using Metamorph software . Fluorescence intensities were normalized by comparing the intensities of the nuclei of interest with the intensities of nuclei of neighboring , pharyngeal muscle nuclei . The GFP signal from the Ptph-1his-24::gfp reporter was still observed after fixation and DAPI staining , and was used to identify the NSMs , NSM sister cells and non-dividing NSM neuroblasts . Embryos were collected and frozen at −80°C in TRIzol ( Invitrogen ) . Frozen embryo pellets were disrupted using a 7-ml tight Dounce tissue grinder ( Fisher Scientific ) and total RNA was prepared using the RNeasy Mini Kit ( Qiagen ) . The first strand cDNA synthesis reaction was performed using the SuperScript III system ( Invitrogen ) . For cDNA synthesis , an oligo ( dT ) primer was used . Transgenic animals carrying an extra-chromosomal array of pMM#47 , pMM#48 and the coinjection marker pL15EK were used as the sample group . Early stage embryos were isolated by bleaching synchronized hermaphrodites that contain four to six embryos . Isolated embryos were allowed to develop on large NGM plates at 20°C for 70 min . ces-1 expression was induced by heat-shocking the embryos at 32°C for 1 h . After a 1 . 5 h recovery period at 20°C embryos were collected . RNA extraction and cDNA synthesis were performed as described above . Transgenic animals carrying an extra-chromosomal array of only coinjection marker pL15EK were used as control and were treated the same way . Fast SYBR green master mix ( Applied Biosystems ) was used to amplify cDNA templates by real-time PCR . Each sample was performed in triplicate on a Biorad CFX96 real-time PCR machine . The sequences of the primers used are provided in Table S5 . The ‘housekeeping’ gene act-1 served as endogenous control [63] . Results were analyzed using the relative standard curve method . To produce the standard curve for each target sequence including act-1 , first , 5-fold dilution series of standard N2 cDNA were prepared and subjected to real-time PCR to determine Ct values for each dilution . Second , average Ct values were plotted versus the logarithm of the concentration ( base 5 ) of the template . To determine the amounts of the target sequences in the starting samples , average Ct values for each sample were compared to the standard curve . To normalize the samples using act-1 , the result of a particular target sequence was divided by the act-1 control result . A ces-1::gfp fosmid reporter was generated as described [37] , [38] . Briefly , using recombineering , a GFP::3×FLAG tag was inserted in-frame to the C-terminus of the ces-1 transcription unit in a fosmid that contains the entire locus of ces-1 . This fosmid reporter was integrated into the worm genome using the method of bombardment , which produces transgenic animals with low-copy integrated arrays . The in vivo binding sites of the CES-1::GFP::3×FLAG fusion protein synthesized in these animals were then determined using the method of ChIP-seq as described [64] .	Animal development is a complex process and requires the coordination in space and time of various processes . These processes include the controlled production of cells , also referred to as ‘cell proliferation’ , and the adoption by cells of specific fates , also referred to as ‘cell fate determination’ . The observation that uncontrolled cell proliferation and cell fate determination contribute to conditions such as cancer , demonstrates that a precise coordination of these processes is not only important for development but for the prevention of disease throughout life . Snail-related transcription factors have previously been shown to be involved in the regulation of cell proliferation and cell fate determination . For example , the Caenorhabditis elegans Snail-related protein CES-1 affects cell fate determination in a specific cell lineage , the NSM ( neurosecretory motorneuron ) lineage . We now present evidence that CES-1 also controls cell proliferation in this lineage . Within a short period of time , CES-1 therefore coordinates cell proliferation and cell fate determination in one and the same lineage . Based on this finding , we propose that CES-1 is an important coordinator that is involved in the precise control - in space ( NSM lineage ) and time ( <150 min ) - of processes that are critical for animal development .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2013	Coordination of Cell Proliferation and Cell Fate Determination by CES-1 Snail
We present a novel methodology to construct a Boolean dynamic model from time series metagenomic information and integrate this modeling with genome-scale metabolic network reconstructions to identify metabolic underpinnings for microbial interactions . We apply this in the context of a critical health issue: clindamycin antibiotic treatment and opportunistic Clostridium difficile infection . Our model recapitulates known dynamics of clindamycin antibiotic treatment and C . difficile infection and predicts therapeutic probiotic interventions to suppress C . difficile infection . Genome-scale metabolic network reconstructions reveal metabolic differences between community members and are used to explore the role of metabolism in the observed microbial interactions . In vitro experimental data validate a key result of our computational model , that B . intestinihominis can in fact slow C . difficile growth . Human health is inseparably connected to the billions of microbes that live in and on us . Current research shows that our associations with microbes are , more often than not , essential for our health [1] . The microbes that live in and on us ( collectively our “microbiome” ) help us to digest our food , train our immune systems , and protect us from pathogens [2 , 3] . The gut microbiome is an enormous community , consisting of hundreds of species and trillions of individual interacting bacteria [4] . Microbial community composition often persists for years without significant change [5] . When change comes , however , it can have unpredictable and sometimes fatal consequences . Acute and recurring infections by Clostridium difficile have been strongly linked to changes in gut microbiota [6] . The generally accepted paradigm is that antibiotic treatment ( or some other perturbation ) significantly disrupts the microbial community structure in the gut , which creates a void that C . difficile will subsequently fill [7–10] . Such infections occur in roughly 600 , 000 people in the United States each year ( this number is on the rise ) , with an associated mortality rate of 2 . 3% [11] . Each year , healthcare costs associated with C . difficile infection are in excess of $3 . 2 billion [11] . An altered gut flora has further been identified as a causal factor in obesity , diabetes , some cancers and behavioral disorders [12-17] . What promotes the stability of a microbial community , or causes its collapse , is poorly understood . Until we know what promotes stability , we cannot design targeted treatments that prevent microbiome disruption , nor can we rebuild a disrupted microbiome . Studying the system level properties and dynamics of a large community is impossible using traditional microbiology approaches . However , network science is an emerging field which provides a powerful framework for the study of complex systems like the gut microbiome [18–23] . Previous efforts to capture the essential dynamics of the gut have made heavy use of ordinary differential equation ( ODE ) models [24 , 25] . Such models require the estimation of many parameters . With so many degrees of freedom , it is possible to overfit the underlying data , and it is difficult to scale up to larger communities [26 , 27] . Boolean dynamic models , conversely , require far less parameterization . Such models capture the essential dynamics of a system , and scale to larger systems . Boolean models have been successfully applied at the molecular [28 , 29] , cellular [20] , and community levels [30] . Here we present the first Boolean dynamic model constructed from metagenomic sequence information and the first application of Boolean modeling to microbial community analysis . We analyze the dynamic nature of the gut microbiome , focusing on the effect of clindamycin antibiotic treatment and C . difficile infection on gut microbial community structure . We generate a microbial interaction network and dynamical model based on time-series data from metagenome data from a population of mice . We present the results of a dynamic network analysis , including steady-state conditions , how those steady states are reached and maintained , how they relate to the health or disease status of the mice , and how targeted changes in the network can transition the community from a disease state to a healthy state . Furthermore , knowing how microbes positively or negatively impact each other—particularly for key microbes in the community—increases the therapeutic utility of the inferred interaction network . We produced genome-scale metabolic reconstructions of the taxa represented in this community [31] , and probe how metabolism could—and could not—contribute to the mechanistic underpinnings of the observed interactions . We present validating experimental evidence consistent with our computational results , indicating that a member of the normal gut flora , Barnesiella , can in fact slow C . difficile growth . Buffie et al . reported treating mice with clindamycin and tracking microbial abundance by 16S sequencing [32] . Mice treated with clindamycin were more susceptible to C . difficile infection than controls . The collection of 16S sequences corresponding to these experiments was analyzed by Stein et al . [24] . First , Stein et al . aggregated the data by quantifying microbial abundance at the genus level . Abundances of the ten most abundant genera and an “other” group were presented as operational taxonomic unit ( OTU ) counts per sample . We use the aggregated abundances from Stein et al . as the starting point for our modeling pipeline ( Fig 1 ) . This processed dataset consisted of nine samples and three treatment groups ( n = 3 replicates per treatment group ) . The first treatment group ( here called “Healthy” ) received spores of C . difficile at t = 0 days , and was used to determine the susceptibility of the native microbiota to invasion . The second treatment group ( here called “clindamycin treated” ) received a single dose of clindamycin at t = -1 days to assess the effect of the antibiotic alone , and the third treatment group ( here called “clindamycin+ C . difficile treated” ) received a single dose of clindamycin ( at t = -1 days ) and , on the following day , was inoculated with C . difficile spores ( S1A Fig ) . Under the clindamycin+ C . difficile treatment group conditions , C . difficile could colonize the mice and produce colitis; however this was not possible under the first two treatment group conditions . The gut bacterial genus abundance dataset included some variation in terms of time points in which genera were sampled . That is , genus abundances were measured between 0 to 23 days; however , not all samples had measurements at all the time points ( S1A Fig ) . Particularly , the healthy population only included time points at 0 , 2 , 6 , and 13 days and Sample 1 of clindamycin+ C . difficile treated population was missing the 9 day time point . Missing abundance values for these 4 points were estimated using an interpolation approach ( S1B Fig ) . For healthy samples , the 16 and 23 day time points could not be interpolated as the last experimentally identified time point for these samples is at 13 days . The assumption of the approximated polynomial for these samples is that extrapolated data points are linear using the slope of the interpolating curve at the nearest data point . Because genera abundances are fairly stable across time in this treatment group ( i . e . the slope of most of the genera abundances is approximately zero ) , extrapolating two time points was deemed reasonable . A principal component analysis was completed on the interpolated data ( Fig 2A ) and shows that the interpolated time series bacterial genus abundance data clusters by experimental treatment group in the first two principal components . Furthermore , the results of the binarization for the healthy population suggest that interpolation did not have any concerning effects on the 16 and 23 day time points ( S2 Fig ) . Natural cubic spline interpolation was used to estimate genus abundances at missing time points in some samples . A cubic spline is constructed of piecewise third order ( cubic ) polynomials which pass through the known data points and has continuous first and second derivatives across all points in the dataset . Natural cubic spline is a cubic spline that has a second derivative equal to zero at the end points of the dataset [33] . Natural splines were interpolated such that all datasets had time points at single day intervals through the 23 day time point ( S1B Fig ) . We use a Boolean framework in which each network node is described by one of two qualitative states: ON or OFF . We chose this framework because of its computational feasibility and capacity to be constructed with minimal and qualitative biological data [34] . The ON ( logical 1 ) state means an above threshold abundance of a bacterial genus whereas the OFF ( logical 0 ) state means below-threshold genus absence . The putative biological relationships among genera are expressed as mathematical equations using Boolean operators [29 , 34] . We inferred putative Boolean regulatory functions for each node , which are able to best capture the trends in the bacterial abundances . These rules , ( edges in the interaction network ) can be assigned a direction , representing information flow , i . e . effect from the source ( upstream ) node to the target ( downstream ) node . Furthermore , edges can be characterized as positive ( growth promoting ) or negative ( growth suppressing ) . An additional layer of network analysis is the dynamic model , which is used to express the behavior of a system over time by characterizing each node by a state variable ( e . g . , abundance ) and a function that describes its regulation . Dynamic models can be categorized as continuous or discrete , according to the type of node state variable used . Continuous models use a set of differential equations; however , the paucity of known kinetic details for inter-genus and/or inter-species interactions makes these models difficult to implement . Genus abundance data was binarized ( converted to a presence-absence dataset ) to enable inference of Boolean relationships for modeling applications . We adapted a previously developed approach called iterative k-means binarization with a clustering depth of 3 ( KM3 ) for this purpose [35] . This approach was employed because binarized data is able to maintain complex oscillatory behavior in Boolean models constructed from this data , whereas other binarization approaches fail to maintain these features [35] . Briefly , this approach uses k-means clustering with a depth of clustering d and an initial number of clusters k = 2d . In each iteration , data for a specific genus G are clustered into k unique clusters C1G , … , CkG , then for each cluster , CnG , all the values are replaced by the mean value of CnG . For the next iteration , the value of d is decreased and clustering is repeated . This methodology is repeated until d = 1 . This approach , with d = 3 ( called here as KM3 binarization ) has previously been demonstrated as a superior binarization methodology to other binarization approaches for Boolean model construction because it conserves oscillatory behavior [35] . These analyses were performed using custom Python code based on a previously written algorithm [35] and is available in the supplemental materials . Because KM3 binarization has a stochastic component ( the initial grouping of binarization clusters ) , we employed KM3 binarization on the entire bacterial genus abundance time series dataset 1000 times . The average binarization for each sample ( S2 Fig ) was used to determine the most probable binarized state of each genus in each sample at each time point ( S3 Fig ) . A principal component analysis of the most probable binarized genus abundances for each sample demonstrates that as with the continuous time series abundances ( Fig 2A ) , binarized bacterial genus abundance data cluster by experimental treatment group ( Fig 2B ) . For inference of Boolean rules from the binarized genus abundances ( S3 Fig ) , the consensus of two of three samples for each treatment population was used as the binarized state of each genus at each time point in each sample ( Fig 2C ) . The Best-fit extension was applied to learn Boolean rules from the binarized time series genus abundance information [36] . For each variable ( genus ) Xi in the binarized time series genus abundance data , Best-fit identifies the set of Boolean rules with k variables ( regulators ) that explains the variable’s time pattern with the least error size . The algorithm uses partially defined Boolean functions pdBf ( T , F ) , where the set of true ( T ) and false vectors ( F ) are defined as T = {X′ ∈ {0 , 1}k: Xi ( t + 1 ) = 1} and F = {X′ ∈ {0 , 1}k: Xi ( t + 1 ) = 0} . Intuitively , the partial Boolean function summarizes the states of the putative regulators that correspond to a turning ON ( T ) or turning OFF ( F ) of the target variable . The error size ε of pdBf ( T , F ) is defined as the minimum number of inconsistencies within X′ that best classifies the T and F values of the dataset . The Best-Fit extension works by identifying smallest size X′ for Xi . For more detailed information refer to [36] . In line with this , we considered the most parsimonious representation of the rules with the smallest ε . If the most parsimonious rule was self-regulation , we also considered rules with the same ε that included another regulator . If multiple rules fit these criteria for a given Xi , it implied that they can independently represent the inferred regulatory relationships . In cases where the alternatives had the same value of ( non-zero ) ε , we explored combinations ( such as appending them by an OR rule ) and used the combination that best described the experimentally observed final ( steady state ) outcomes . For example , we combined the two alternative rules for Blautia with an OR relationship . In the case of Barnesiella , we chained three rules ( "Other" , "Lachnospiraceae_other" , "Lachnospiraceae" ) by an OR relationship , and "not Clindamycin" by an AND relationship to incorporate the loss of Barnesiella in the presence of clindamycin ( Fig 2C ) . This was also done for rules for “Lachnospiraceae” , “Lachnospiraceae_other” and “Other” and all four nodes attained the same rule . There are six nodes with multiple inferred ( alternative ) rules: “Barnesiella” , ”Blautia” , ”Enterococcus” , ”Lachnospiraceae” , ”Lachnospiraceae_other” , and”Other” had 4 , 2 , 5 , 4 , 4 , and 4 rules , respectively . The six other nodes had a single inferred rule . The network in Fig 2C represents the union of all of the alternative rules produced by Best-Fit , or in other words , –it is a super-network of all alternative rules . Any alternative networks would be a sub-network of what we show . A strongly connected component between the nodes inhibited by clindamycin is a feature of the vast majority of these sub-networks . We used the implementation of Best-Fit in the R package BoolNet [37] . Dynamic analysis is performed by applying the inferred Boolean functions in succession until a steady state is reached . Boolean models and discrete dynamic models in general focus on state transitions instead of following the system in continuous time . Thus , time is an implicit variable in these models . The network transitions from an initial condition ( initial state of the bacterial community ) until an attractor is reached . An attractor can be a fixed point ( steady state ) or a set of states that repeat indefinitely ( a complex attractor ) . The basin of attraction refers to the set of initial conditions that lead the system to a specific attractor . For the network under consideration , the complete state space can be traversed by enumerating every possible combination of node states ( 212 ) and applying the inferred Boolean functions ( or “update rules” ) to determine paths linking those states . The state transition network describes all possible community trajectories from initial conditions to steady states , given the observed interactions between bacteria in the community . We made use of two update schemes to simulate network dynamics: synchronous ( deterministic ) and asynchronous ( stochastic ) . Synchronous models are the simplest update method: all nodes are updated at multiples of a common time step based on the previous state of the system . The synchronous model is deterministic in that the sequence of state transitions is definite for identical initial conditions of a model . In asynchronous models , the nodes are updated individually , depending on the timing information , or lack thereof , of individual biological events . In the general asynchronous model used here , a single node is randomly updated at each time step [38] . The general asynchronous model is useful when there is heterogeneity in the timing of network events but when the specific timing is unknown . Due to the heterogeneous mechanisms by which bacteria interact , we made the assumption of time heterogeneity without specifically known time relationships . Synchronous and asynchronous Boolean models have the same fixed points , because fixed points are independent of the implementation of time . However , the basin of attraction of each fixed point ( i . e . the initial conditions that lead to each fixed point ) may differ between synchronous and asynchronous models ( S2 Table ) . For identification of all of the fixed points in the network ( the attractor landscape ) , the synchronous updating scheme was used . However , for the perturbation analysis , the asynchronous updating scheme was used because it more realistically models the possible trajectories in a stochastic and/or time-heterogeneous system . The simulations of the gut microbiome model were performed using custom Python code built on top of the BooleanNet Python library , which facilitates Boolean simulations [39] . Our custom Python code is available in the supplemental materials . To capture the effect of removal ( knockout ) or addition ( probiotic; forced over abundance ) of genera , modification of the states/rules to describe removal or addition states were performed . These modifications were implemented in BooleanNet by setting the corresponding nodes to either OFF ( removal ) or ON ( addition ) and then removing the corresponding updating rules for these nodes for the simulations . By examining many such forced perturbations , we can identify potential therapeutic strategies , many of which may not be obvious or intuitive , particularly as network complexity increases . We used asynchronous update when simulating the effect of perturbations on the microbial communities . In each case we performed 1000 simulations and report the percentage of simulations that achieve a certain outcome . To generate draft metabolic network reconstructions for each of the ten genera in the paper , we first obtained genome sequences for representative species by searching the “Genomes” database of the National Center for Biotechnology Information ( NCBI ) . Complete genomes for the first ten ( or if less than ten , all ) species within the appropriate genus were downloaded . During the process of reconstructing genus-level metabolic reconstructions , some genera were underrepresented ( fewer than 10 species genomes ) in the NCBI Genome database , including Akkermansia , Barnesiella and Coprobacillus ( S3 Table ) . The search result order is based on record update time , and so it is quasi-random . Genomes were uploaded to the rapid annotations using subsystems technology ( RAST ) server for annotation [40] . Draft metabolic network reconstructions were generated by providing the RAST annotations to the Model SEED service [41] . Metabolic network reconstructions were downloaded in “ . xls” format . Genus-level metabolic reconstructions were produced by taking the union of all species-level reconstructions corresponding to each genus , as has been done previously [42] . The one exception was C . difficile , which was produced by taking the union of three strain-level reconstructions . Subsystems were defined as the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) map with which each reaction was associated [43 , 44] . These associations were determined based on annotations in the Model SEED database [41] . To quantify enrichment , the complete set of unique reactions from all genus-level reconstructions was pooled , and the subsystem annotations corresponding to those reactions were counted . To determine enrichment for a given subset of the community ( either a single genus-level reconstruction , or a set of reconstructions corresponding to a subnetwork ) , the subsystem occurrences were counted within the subset . The probability of a reconstruction containing N total subsystem annotations , with M or more occurrences of subsystem I , was determined by taking the sum of a hypergeometric probability distribution function ( PDF ) from M to the total occurrences of subsystem I in the overall population . Enrichment analysis was performed in Matlab [45] . To quantify metabolic interactions , we started by utilizing the seed set detection algorithm developed by Borenstein et al . [46 , 47] . The algorithm follows three steps: The rationale is that metabolites that feed into the network , but cannot be produced by any reactions within the network , must be obtained from the environment . Competition metrics were generated following the process of Levy and Borenstein [46] . For a given pair of genera , the competition score is defined as: Here SeedSeti is the set of obligatory input metabolites to the metabolic network reconstruction for genus i , and \|SeedSeti\| is the number of metabolites contained in SeedSeti . The competition score indicates the fractional overlap of inputs that genus i shares with genus j , and so ranges between zero and one . The higher the score , the more similar the metabolic inputs to the two networks , making competition more likely . For a given pair of genera , the mutualism score is defined as: Here ¬SeedSetj is the set of metabolites that can be produced by the metabolic network for species j ( i . e . all non-seed metabolites ) . The mutualism score indicates the fractional overlap of inputs that genus i consumes which genus j can potentially provide . The mutualism score ranges between zero and one . The higher the score , the more potential there is for nutrient sharing between species . While the score does not measure “mutualism” per se ( it cannot necessarily distinguish between other interactions such as commensalism or amenalism [48] ) , for simplicity , we will refer to these scores as the competition and mutualism scores . All metabolic reconstructions , seed sets , competition scores and mutualism scores are available in the supplemental materials . Seed set generation was performed using custom Matlab scripts , which are available in the supplement . [45] . Statistical tests were performed in R [49] . Barnesiella intestinihominis DSM 21032 and Clostridium difficile VPI 10463 were grown anaerobically in PRAS chopped meat medium ( CMB ) ( Anaerobe Systems , Morgan Hill , CA ) at 37 C . To prepare B . intestinihominis spent medium , B . intestinihominis was grown in CMB until stationary phase ( 44 hours ) . The saturated culture was centrifuged , and the supernatant was filter sterilized ( 0 . 22 μM pore size ) . Growth curves were obtained by inoculating batch cultures in 96-well plates and gathering optical density measurements ( 870 nm ) using a small plate reader that fits in the anaerobic chamber [50] . Single cultures were inoculated from overnight liquid culture to a starting density of 0 . 01 . The co-cultures were started at a 1:1 ratio , for a total starting density of 0 . 02 . Optical density was measured every 2 minutes for 24 hours , and the resulting growth curves were analyzed in Matlab [45] . Maximum growth rates were calculated by fitting a smooth line to each growth curve , and finding the maximum growth rate from among the instantaneous growth rates over the whole time course: [log ( ODt+1 ) —log ( ODt ) ] / [t+1-t] . The achieved bacterial density—area under the growth curve ( AUC ) —in a culture was calculated by integrating over the growth curve in each experiment using the “trapz ( ) ” function in Matlab . It can be thought of as representing the total biomass produced over time . The simply additive null model was calculated by fitting a Lotka-Volterra model [24] to the single cultures for both B . intestihominis and C . difficile . The null model of co-culture ( assuming zero interaction between species ) was simulated by using the parameters from single culture , and summing the predicted OD870 values . All scripts used to analyze the data are available at https://bitbucket . org/gutmicrobiomepaper/microbiomenetworkmodelpaper/wiki/Home . To capture the dynamics of inter-genus interactions in the intestinal tract we employed a pipeline ( Fig 1 ) which translates metagenomic genus abundance information into a dynamic Boolean model . This approach involves three steps: 1 ) discretization ( binarization ) of genus abundances , 2 ) learning Boolean relationships among genera , and 3 ) translation of genus associations into a Boolean ( discrete ) dynamic model . Boolean rules ( S1 Table ) were inferred from the time series binarized genus abundances using an implementation of the Best-fit extension [36] in the R Boolean network inference package BoolNet [37] ( see Methods ) . A network of 12 nodes and 33 edges was inferred ( Fig 2D ) . The inferred interaction network has a clustered structure: the cluster ( subnetwork ) containing the two Lachnospiraceae nodes and Barnesiella is strongly influenced by clindamycin whereas the other subnetwork is largely independent of the first , except for the single edge between Barnesiella and C . difficile ( Fig 2D ) . In fact , Lachnospiraceae nodes , Barnesiella and the group of “Other” genera form a strongly connected component; that is , every node is reachable from every other node . Most nodes of the second subnetwork are positively influenced by C . difficile , with the exception of Coprobacillus , for which no regulation by other nodes was inferred , and Akkermansia , which is inferred to be regulated only by Coprobacillus . These latter two genera are transiently present ( around day 5 ) in the clindamycin treatment group , but they do not appear in the final states of any of the treatment groups ( see S1 Fig ) . This network structure is consistent with published data in which the dominant Firmicutes ( Lachnospiraceae ) and Bacteroidetes ( Barnesiella ) are devastated by antibiotic administration [51 , 52] . Furthermore , the clustered structure ( Fig 2D ) supports the established mechanism of C . difficile colitis: loss of normal gut flora , which normally suppresses opportunistic infection ( clindamycin cluster ) , and the presence of C . difficile at a minimum inoculum ( C . difficile cluster ) [10 , 53] . The network clusters have a single route of interaction between Barnesiella and C . difficile . The negative influence of Barnesiella on C . difficile is in agreement with recently published findings in which Barnesiella was strongly correlated with C . difficile clearance [54] . The role of Barnesiella as an inhibitor of another pathogen ( vancomycin-resistant Enterococci ( VRE ) ) has been shown in mice [55] , which is also visible in the network model as an indirect relationship between Barnesiella and Enterococcus ( Fig 2D ) . Related species of Bacteroidetes have been shown to play vital roles in protection from C . difficile infection in mice [56] . Furthermore , the network structure shows that Lachnospiraceae positively interacts with Barnesiella , leading to an indirect suppression of C . difficile . Interestingly , the two Lachnospiraceae nodes and the “Other” node form a strongly connected component , suggesting a similar role in the network , particularly in promoting growth of Barnesiella , which directly suppresses C . difficile . In support of this finding , Lachnospiraceae has been shown to protect mice against C . difficile colonization [52 , 57] . Therefore , the structure of the network is both a parsimonious representation of the current data set , and is supported by literature evidence . We applied dynamic analysis using the synchronous updating scheme ( see Methods ) to determine all the possible steady states of the microbiome network model . In a 12 node network , there are 212 possible network states . We employed model simulations using the synchronous updating scheme to visit all possible network states and identify all fixed points of the model . Exploration of the steady states of this network reveals 23 possible fixed point attractors ( S4 Fig ) . Three of the identified attractors ( Fig 3A ) are in exact agreement with the experimentally identified terminal time points of binarized genus abundances ( Fig 2C ) . These attractors make up a small subset of the entire microbiome network state space ( S2 Table ) . The attractor landscape can be divided into six groups based on abundance patterns they share ( S4 Fig ) . Group 1 is made up of a single attractor wherein all genera are absent ( OFF ) . The second group attractor consists of the experimentally defined healthy state ( Attractor 2 ) and genera in the C . difficile subnetwork which can be abundant ( ON ) independent of the clindamycin subnetwork . The third grouping has the clindamycin treated steady state ( Attractor 7 ) and genera in the C . difficile subnetwork that can survive in the presence of the clindamycin . Group 4 contains the clindamycin plus C . difficile steady state ( Attractor 12 ) and its subsets in which one or both of the source nodes Mollicutes and Enterobacteriaceae are absent . Group 5 contains attractors in which clindamycin is absent and C . difficile is present . Even if clindamycin is absent , our model suggests that C . difficile can thrive if Lachnospiraceae and Barnesiella are absent , i . e . these states represent a clindamycin-independent loss of Lachnospiraceae and Barnesiella . Lastly , group 6 attractors have both clindamycin and C . difficile as OFF . Blautia and Enterococcus are always abundant in these attractors . Indeed , because of the mutual activation between Blautia and Enterococcus they always appear together . Attractors in this group may also include the abundance ( ON state ) of the source nodes Mollicutes and Enterobacteriaceae . We next explored the perturbation of genera in the gut microbiome network model . We considered the clinically relevant question of which perturbations might alter the microbiome steady states produced by clindamycin or clindamycin+C . difficile treatment after clindamycin treatment was removed . Thus , we considered the clindamycin-treated steady state ( Attractor 7 in S3 Fig ) and the clindamycin+C . difficile treated steady state ( Attractor 12 ) as initial conditions and assumed that clindamycin treatment was stopped . Our simulations , employing asynchronous update ( see Methods ) , indicate that for both initial conditions , only the state of clindamycin changes after the treatment is stopped; these steady states become Attractor 1 and Attractor 19 , respectively ( S4 Fig ) . In other words , the steady states remain identical in the absence of clindamycin . We next explored the effect of addition ( overabundance; Fig 3B , left column ) and removal ( knockout; Fig 3B , right column ) of individual genera , simultaneously with the stopping of clindamycin treatment , on the model predicted steady states . For the perturbation analysis , the model was initialized from the clindamycin treated steady state ( Fig 3B , top row ) or the clindamycin+C . difficile steady state ( Fig 3B , bottom row ) . From the clindamycin treated state , addition of Lachnospiraceae or “Other” nodes restores the healthy steady state; however , no removal restore the healthy steady state ( Fig 3B ) . From the clindamycin+C . difficile state , addition of Barnesiella , Lachnospiraceae , or “Other” nodes lead to a shift toward the healthy steady state ( suppression of C . difficile ) . Species-level reconstructions from the genus Enterobacteriaceae contained the most reactions on average ( 1335 ) , while those from Mollicutes contained the least ( 485 ) ( S3 Table ) . The Barnesiella and Enterococcus reconstructions contained the most unique reactions ( S4 Table ) and , interestingly , also displayed more overlap in reaction content between each other ( 503 reactions ) than was observed between any other pair of reconstructions ( S5 Table ) . Lachnospiraceae and Barnesiella had the next highest degree of overlap ( 424 reactions ) . Mollicutes and Coprobacillus had the least degree of overlap ( 363 reactions ) ( S5 Table ) . Note that the metabolic reconstructions produced by the SEED framework are draft quality , and as such , may lack the predictive power of well-curated metabolic reconstructions . Enrichment analysis was performed for the 99 unique subsystem annotations that were observed in the community . 22 subsystems displayed interesting enrichment patterns with respect to the structure of the interaction network ( Fig 4 ) . The subsystems for glycolysis/gluconeogenesis and nucleotide sugars metabolism are enriched in all taxa , highlighting the fact that all taxa contain relatively full complements of reactions within those subsystems . Interestingly , C . difficile is highly enriched for reactions in cyanamino acid metabolism compared to all other genera . Lipopolysaccharide ( LPS ) biosynthesis and cyanoamino acid metabolism subsystems are differentially enriched between C . difficile and both Barnesiella and Lachnospiraceae . Between Barnesiella and Enterococcus , Barnesiella is more highly enriched for d-glutamine and d-glutamate metabolism , pantothenate and CoA biosynthesis , LPS biosynthesis . With respect to Enterococcus , Barnesiella is less highly enriched in pyrimidine metabolism , and phenylalanine , tyrosine , and tryptophan biosynthesis . The metabolic reconstructions were used to explore the potential metabolic underpinnings of the inferred interaction network . Competition scores were generated for all pairwise relationships between the genera considered in the model ( self-edges were excluded ) . The two Lachnospiraceae genera were treated as metabolically identical , and the “Other” group was excluded . We grouped pairs of genera into five groups based on being connected by a positive or negative edge , a negative or positive path ( meaning an indirect relationship ) , or no path . A positive relationship was found between competition score and edge type in the interaction network ( i . e . positive edges tend to have a higher competition score ) , which was not statistically significant , perhaps due to the small sample size ( p-value = 0 . 058 by one-sided Wilcoxon rank sum test ) ( S5A Fig ) . The mutualism score did not display any obvious trends with respect to the network structure ( S5B Fig ) . All pairs with inferred edges exhibited relatively high competition scores and low mutualism scores ( S5C Fig ) . Barnesiella , a key inhibitor of C . difficile in the interaction network , holds the second smallest competition score with C . difficile ( see Fig 5A ) . Barnesiella and C . difficile also have the highest mutualism score among all interacting pairs in the network ( S5C Fig ) . The positive relationship between edge type and competition score suggests that more metabolic similarity between genera tends to foster positive interaction . The converse is also true , where less metabolic similarity tends to foster negative interactions ( S5A Fig ) . Here , “positive/negative interaction” is derived from the Boolean model , where a positive edge between species A and B indicates that if A is ON at time t , then B is likely to turn ON at t+1 . Barnesiella intestinihominis was chosen as a representative species for the genus Barnesiella for the in vitro experiments . C . difficile grew more slowly in B . intestinihominis spent media ( n = 16 , p-value < 0 . 005 , by one-sided Wilcoxon rank sum test ) ( Fig 5B ) . The co-culture with both B . intestinihominis and C . difficile grew more slowly than C . difficile alone ( n = 16 , p-value < 0 . 05 , by one-sided Wilcoxon rank sum test ) ( Fig 5B ) . C . difficile area under the growth curve ( AUC ) , a measure of the achieved bacterial density over the experiment , was not statistically different between growth in fresh media and B . intestinihominis spent media ( n = 16 , p-value = 0 . 22 by one-sided Wilcoxon rank sum test ) . However , the co-culture displayed a much lower AUC than expected under a null model of interaction ( in which the two species do not interact ) ( Fig 5C ) . Examining the co-culture growth curve , it maintained a consistently lower density than a null model ( Fig 5D ) . The first feature that stands out in the inferred interaction network is its clustered structure . Clindamycin has a strong influence on the subnetwork containing the two Lachnospiraceae nodes and Barnesiella . The other subnetwork contains C . difficile and other genera that become abundant during C . difficile infection ( Fig 2D ) . Also worth noticing are the two contradicting edges in the network , between Coprobacillus and Blautia , and the self-edges for Blautia ( Fig 2D ) . These arise from rules in the Boolean model that are context-dependent . Such context-dependent rules can manifest as opposite edge types , depending on the state of other nodes in the network . Context-dependent interactions have been demonstrated in many microbial pairings , and nutritional environments can even be designed to induce specific interaction types [58] . It is possible that subtle environmental changes over the course of the experiment altered conditions in a way that flipped the Coprobacillus-Blautia interaction . Because the interaction network is derived from time-series data , it is possible to estimate causality , and therefore , derive a directed graph . A directed network with clear , causative interactions can be used to study community dynamics . This is in contrast with association networks , which are often derived from independent samples , and cannot determine direction of causality [48 , 59–61] . Such networks are more limited in utility because they cannot be used to predict system behavior over time , or system responses to perturbations [24 , 62] . Note that the inferred network structure represents a set of hypotheses as to potential interactions among genera . Determining whether or not the interactions truly occur requires further experimentation , similar to the experimentation completed to validate the edge between Barnesiella and C . difficile . We experimentally validated a key edge in the interaction network , and showed that Barnesiella can in fact slow C . difficile growth . C . difficile was grown alone , in co-culture with B . intestinihominis , and in B . intestinihominis spentmedia . C . difficile grew more slowly in both co-culture and spent-media conditions . Though moderate , the effect was statistically significant ( Fig 5B ) . The fact that C . difficile growth rate was inhibited under spent-media conditions indicates that B . intestinihominis-mediated inhibition does not require B . intestinihominis to “sense” the presence of C . difficile . Further , C . difficile growth on B . intestinihominis spent media demonstrates that the two species have different nutrient requirements . Whether the reduction in growth rate is a result of nutritional limitations ( e . g . C . difficile resorts to a less preferred carbon source ) is unknown , but unlikely given the AUC data . The AUC—a summation of the OD over the entire time course—is a measure of the total bacterial density achieved over the course of the experiment . It can be thought of as a single metric combining growth rate and biomass production over time . Examining the AUC for all conditions showed that C . difficile AUC did not significantly change between fresh media and spent media ( Fig 5C ) . Thus , C . difficile was able to produce comparable overall biomass despite a reduction in growth rate , further demonstrating that nutrient availability was sufficient in the spent media condition . The AUC for the co-culture was much lower than expected in a simulated null model ( Fig 5C ) . Apparently , in co-culture , the total community biomass production capacity is reduced from what would be expected in a scenario without species interaction . Thus , there is a measurable negative interaction between B . intestinihominis and C . difficile in co-culture that impacts biomass production . This can be observed over the full time-course of the co-culture , where the overall density is consistently lower than what would be expected in a null model ( Fig 5D ) . Computational perturbation analysis showed that forced overabundance of Barnesiella led to a shift from the “disease” state ( clindamycin+ C . difficile treatment group ) to a state highly similar to the original healthy state ( loss of C . difficile ) . This result is particularly interesting from a therapeutic design standpoint . In this case , the model results indicate that Barnesiella may serve as an effective probiotic . Model-driven analysis can be used to identify candidate organisms for probiotic treatments . Recent work by Buffie et al . performed a proof-of-concept study in which they used statistical models to identify candidate probiotic organisms , which were then tested on a murine model of C . difficile infection [54] . This model-driven approach can be favorably contrasted with the brute-force experimental approach in which successive combinations of microbes are tested until a curative set is found [56] . The model-driven approach requires far fewer experiments , and saves time and resources . While the computational model presented here differs from that used by Buffie et al . , the integration of computational models in probiotic design has been shown to be a feasible , effective approach . Improved tools , such as the perturbation analysis presented here , will surely accelerate the probiotic design process and shorten the path to the clinic . Genome-scale metabolic network reconstructions can be used to estimate the interactions between microbes in a complex community based purely on genome sequence data . Our use of genus-level metabolic network reconstructions ( a union of several species-level reconstructions ) may not reflect the unique , species-level interactions and heterogeneity within a community . This higher-level model will only capture broad trends and the possible extent of metabolic capacity within a genus . Furthermore , the draft status of these models precludes the effective application of flux balance analysis ( FBA ) to estimate interactions among genera . This is due to the established lack of precision in draft reconstructions in predictions of growth rates and substrate utilization patterns [63] , and the sensitivity of interaction models to metabolic environment and model structure [58 , 64] . Future efforts to infer metabolic interactions using FBA and well-curated metabolic networks could provide deeper insights into specific metabolites that are shared ( or competed for ) between specific microbial pairs . The application of competition scores demonstrated here ( S5A Fig ) could potentially be used to quickly establish a rough expectation ( notice the spread of competition scores for the species pairs not connected by a path through the network ) for community structure—based solely on genomic information—that can then be tested experimentally . Interestingly , the fact that higher competition score is associated with more positive interactions inferred from the Boolean model relates to previous work that demonstrates that higher competition scores were associated with habitat co-occurrence [46] . In this same work , the authors suggest that this effect is due to habitat filtering; that is , microbes with similar metabolic capabilities tend to thrive in similar environments . It has been shown experimentally that microorganisms from the same environment tend to lose net productivity in batch co-culture , indicating similar metabolic requirements [65] . Thus , it appears that metabolically similar organisms tend to co-locate to similar niches , and over evolutionary time , co-localized organisms tend to develop positive relationships with each other . Understanding this relationship between competition score and interaction type leads to the conclusion that negative interactions are probably not caused by metabolic competition . Of all the genus competition scores with C . difficile , Barnesiella showed the second lowest ( Fig 5A ) . In other words , Barnesiella is among the least likely to share a similar metabolic niche with C . difficile , which fits with the broad trend mentioned above . The fact that the competition score between C . difficile and Barnesiella is so low strongly suggests that the negative interaction between them is due , not to competition for scarce resources ( although it does not completely exclude the possibility ) , but rather to some non-metabolic mechanism . The similarity in reaction content between Barnesiella and Enterococcus indicates similar network structure ( S5 Table ) , and yet , Enterococcus does not inhibit C . difficile in the inferred interaction network ( Fig 2D ) . Either the differences that are present between Barnesiella ( 65 unique reactions ) and Enterococcus ( 36 unique reactions ) are hints at the mechanism of interaction , or metabolism does not play a significant role in C . difficile inhibition in the environment of the gut . For example , enrichment analysis showed that that , with respect to Enterococcus , Barnesiella is more highly enriched for d-glutamine and d-glutamate metabolism , pantothenate and CoA biosynthesis and LPS biosynthesis . With respect to Enterococcus , Barnesiella is less enriched in pyrimidine metabolism , and phenylalanine , tyrosine , and tryptophan biosynthesis . The possible role of LPS is discussed further on . The possible roles of these other metabolic pathways in C . difficile inhibition is unclear . There is experimental evidence that Barnesiella ( and other normal flora ) may combat pathogen overgrowth through non-metabolic mechanisms . As a first step , it has been shown that VRE can grow in sterile murine cecal contents—indicating the presence of sufficient nutrition to support VRE—but is inhibited in saline-treated cecal contents—indicating that live flora are needed to suppress VRE growth , and that this suppression is not through nutrient sequestration [66] . Further , the presence of B . intestinihominis has been demonstrated to prevent and cure VRE infection in mice [55] , and is strongly correlated with resistance to C . difficile infection in mice [54] . Clearly , Barnesiella plays a key role in pathogen inhibition , and pathogen inhibition can be caused by mechanisms other than nutrient competition . This non-metabolic mechanism may be direct or indirect ( Fig 6A ) . We demonstrated in vitro that B . intestinihominis can inhibit C . difficile growth rate ( Fig 5C and 5D ) . The fact that C . difficile grows on B . intestinihominis spent media at all indicates that the metabolic requirements of the two species are different , which is consistent with our computational results supporting the hypothesis that C . difficile and Barnesiella do not compete metabolically ( Fig 5B ) . Further , C . difficile is moderately inhibited both in co-culture with B . intestinihominis and in B . intestinihominis-spent media , indicating a direct mechanism of inhibition . In further support of a direct mechanism , it has been shown that Clostridium scindens inhibits growth of C . difficile through the production of secondary bile acids [54] . Perhaps Barnesiella works through an analogous mechanism in vivo , enhancing the moderate inhibition observed in vitro . In support of an additional indirect mechanism of bacterial interaction , Buffie and Pamer , in a recent review , hypothesized that the normal flora ( of which Barnesiella is a member ) may prevent pathogen overgrowth by stimulation of a host antimicrobial response [67] ( Fig 6A ) . Specifically , they point out that Barnesiella can activate host toll-like receptor TLR signaling , which activates host antimicrobial peptide production . For example , LPS and flagellin have been shown to stimulate the host innate immune response through toll-like receptor ( TLR ) signaling and production of bactericidal lectins [68 , 69] . Barnesiella shows enrichment for LPS biosynthesis pathways ( Fig 4 ) . However , this mechanism did not seem to be responsible for inhibition of VRE by Barnesiella [55] . An indirect , host-mediated mechanism is further supported by the fact that members of the normal gut flora can interact differently with pathogens depending on the host organism [54] . Regardless , any indirect mechanism is in addition to the direct inhibitory mechanism observed in vitro . Both direct and indirect mechanisms may play a role in vivo , and further work is needed to clearly discern the underlying process that allows Barnesiella to play this protective role . We demonstrate that dynamic Boolean models capture key microbial interactions and dynamics from time-series abundance data in a murine microbiome . We show that this computational approach enables exhaustive in silico perturbation , which leads to fast candidate selection for probiotic design . We further describe the use of genome-scale metabolic network reconstructions to explore the metabolic potential attributed to community members , and to estimate metabolic competition and cooperation between members of the microbiome community . Analysis of genome-scale metabolic network reconstructions indicates that Barnesiella likely inhibits C . difficile through some non-metabolic mechanism . We present empirical in vitro evidence that B . intestinihominis does in fact inhibit C . difficile growth , likely by a non-metabolic mechanism , and our findings are in good agreement with published results . We present this work as a demonstration of the use of dynamic Boolean models and genome-scale metabolic reconstructions to explore the structure , dynamics , and mechanistic underpinnings of complex microbial communities .	The community of bacteria that live in our intestines ( called the “gut microbiome” ) is important to normal intestinal function , and destruction of this community has a causative role in diseases including obesity , diabetes , and even neurological disorders . Clostridum difficile is an opportunistic pathogenic bacterium that causes potentially life-threatening intestinal inflammation and diarrhea and frequently occurs after antibiotic treatment , which wipes out the normal intestinal bacterial community . We use a mathematical model to identify how the normal bacterial community interacts and how this community changes with antibiotic treatment and C . difficile infection . We use this model to identify bacteria that may inhibit C . difficile growth . Our model and subsequent experiments indicate that Barnesiella intestinihominis inhibits C . difficile growth . This result suggests that B . intestinihominis could potentially be used as a probiotic to treat or prevent C . difficile infection .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[]	2015	Inference of Network Dynamics and Metabolic Interactions in the Gut Microbiome
Human African Trypanosomiasis is a devastating disease caused by the parasite Trypanosoma brucei . Trypanosomes live extracellularly in both the tsetse fly and the mammal . Trypanosome surface proteins can directly interact with the host environment , allowing parasites to effectively establish and maintain infections . Glycosylphosphatidylinositol ( GPI ) anchoring is a common posttranslational modification associated with eukaryotic surface proteins . In T . brucei , three GPI-anchored major surface proteins have been identified: variant surface glycoproteins ( VSGs ) , procyclic acidic repetitive protein ( PARP or procyclins ) , and brucei alanine rich proteins ( BARP ) . The objective of this study was to select genes encoding predicted GPI-anchored proteins with unknown function ( s ) from the T . brucei genome and characterize the expression profile of a subset during cyclical development in the tsetse and mammalian hosts . An initial in silico screen of putative T . brucei proteins by Big PI algorithm identified 163 predicted GPI-anchored proteins , 106 of which had no known functions . Application of a second GPI-anchor prediction algorithm ( FragAnchor ) , signal peptide and trans-membrane domain prediction software resulted in the identification of 25 putative hypothetical proteins . Eighty-one gene products with hypothetical functions were analyzed for stage-regulated expression using semi-quantitative RT-PCR . The expression of most of these genes were found to be upregulated in trypanosomes infecting tsetse salivary gland and proventriculus tissues , and 38% were specifically expressed only by parasites infecting salivary gland tissues . Transcripts for all of the genes specifically expressed in salivary glands were also detected in mammalian infective metacyclic trypomastigotes , suggesting a possible role for these putative proteins in invasion and/or establishment processes in the mammalian host . These results represent the first large-scale report of the differential expression of unknown genes encoding predicted T . brucei surface proteins during the complete developmental cycle . This knowledge may form the foundation for the development of future novel transmission blocking strategies against metacyclic parasites . Sleeping Sickness , or Human African Trypanosomiasis ( HAT ) , is a fatal parasitic disease transmitted by the bite of an infected tsetse ( Glossina spp . ) fly . The disease agents are the extracellular protozoan parasites belonging to the Trypanosoma brucei species complex . It is estimated that 60 million people in 36 African nations are at risk for HAT . The same parasite species complex also infects animals causing nagana , an economically important disease of livestock in Africa . There are no mammalian vaccines for disease control and the drugs used for chemotherapy have major adverse effects , are difficult to administer and have decreased efficacy in light of the emergence of parasite drug resistance . A number of disease control strategies , mainly focused on vector control and treatment of infections , have been applied . These are often successful in the short term , although a sustainable long-term solution remains unidentified . African trypanosomes undergo multiple differentiation steps as they complete their life cycle in the challenging environments of the mammalian and invertebrate hosts . Trypanosomes circulating in the mammalian bloodstream ( bloodstream form , BSF ) are found as either long slender forms that perpetuate the infection in the mammal , or as short stumpy forms that are infective to the tsetse fly . In the mammalian host , BSF parasites evade the adaptive immune system by changing their surface coat molecules in a process known as antigenic variation [1] . Antigenic variation has effectively prevented the development of mammalian vaccines to date . In the tsetse flies , a strong immune response apparently clears the parasites in the majority ( over 95% ) of challenged tsetse [2] but those parasitized flies remain infected for their lifetime and contribute to disease transmission . Once acquired in an infected bloodmeal , trypanosomes undergo several stages of differentiation in the fly before they are transmissible to the mammalian host [3] . In the midgut , the stumpy BSF parasites differentiate to the procyclic form ( PF ) parasites . Although the majority of flies can clear parasite infections at this stage [2] , in flies where the PF cells survive , trypanosomes migrate anteriorly to the proventriculus , and differentiate initially into the mesocyclic trypomastigote , then long and short epimastigotes . It is thought that only the short epimastigotes can invade the salivary glands , where they attach and differentiate ultimately giving rise to the free mammalian infective metacyclic trypomastigotes ( MCF ) , which are transferred to the mammalian host in saliva as the infected fly blood feeds . Only the BSF and PF developmental stages of T . brucei can be maintained in culture in vitro . The remaining developmental stages of the parasite can only be maintained in tsetse , making the access to and evaluation of these life stages difficult . The genomes of several related kinetoplastid parasites have been published , including T . brucei brucei and Trypanosoma brucei gambiense [4]–[8] . Improved technologies such as RNA sequencing have identified over 1 , 000 new transcripts in T . b . brucei [9] . Particularly relevant for disease control tools are surface expressed proteins that interact with the host environment , and specifically with the host immune system . Protein features that are suggestive of surface expression are associated signal peptides , trans-membrane domains , and glycosylphosphatidylinositol ( GPI ) anchor attachment domains . Many GPI-anchored proteins in mammalian systems have been shown or predicted to have hydrolytic activity , or serve as receptors or adhesion molecules , while some are suggested to be involved in trans-membrane signaling or membrane trafficking [10] , [11] . The two well-studied GPI-anchored surface coat proteins of T . brucei are the variant surface glycoproteins ( VSGs ) and procyclins , expressed by the BSF and PF cells , respectively . The VSG coat of the BSF trypanosome allows the parasite to evade the adaptive immune response and therefore persist in the mammalian host . The procyclins were initially thought to shield PF parasites from the digestive enzymes of the fly midgut [12] , but procyclin-null mutant trypanosomes were subsequently found to be capable of infecting tsetse [13] . BARP , a third GPI-anchored surface protein family identified in T . brucei [14] is expressed by immature salivary gland stages [15] . Functional assessment of the BARP proteins have not yet been described , so it is unknown if trypanosome survival or maturation in the salivary gland environment would be influenced in their absence . The serum resistance associated protein ( SRA ) , which allows the survival of Trypanosoma brucei rhodesiense in the human host was recently determined to be GPI-anchored [16] , demonstrating the role of GPI-anchored proteins in the host-range of this pathogen . Additionally , a sub-unit of the transferrin receptor ( ESAG6 ) was also demonstrated to be GPI-anchored [17] , and work continues on the characterization of this molecule . Here , we report on the differential expression of transcripts corresponding to putative GPI anchored proteins with unknown functions in T . brucei . The selected genes were initially identified in silico using the Big PI and subsequently by the FragAnchor GPI-prediction algorithms . The signal peptide and trans-membrane domains of the putative proteins were also analyzed in silico . Gene expression data was obtained from parasites infecting the tsetse and mammalian hosts . We discuss the implications of the observed transcript expression profiles with regard to parasite survival and transmission processes , with consideration of the mammalian infective metacyclic trypomastigote . This experiment was carried out in strict accordance with the recommendations in the Office of Laboratory Animal Welfare at the National Institutes of Health and the Yale University Institutional Animal Care and Use Committee . The experimental protocol was reviewed and approved by the Yale University Institutional Animal Care and Use Committee ( Protocol 2011-07266 ) . Genes encoding putative GPI anchor attachment domains were identified in silico and manually curated during the annotation of the first publicly available T . brucei brucei strain 927 genome sequence [4] . GPI anchor predictions were made by the consortium using the publicly available software Big-PI Predictor ( http://mendel . imp . ac . at/gpi/gpi_server . html ) [18] . The known GPI anchored protein families , such as VSG and procyclin , were removed from the resulting list . A second program , FragAnchor ( http://navet . ics . hawaii . edu/~fraganchor/NNHMM/NNHMM . html ) , was applied to genome annotation data from version 4 of the T . brucei genome [19] . Non-VSG , non-procyclin genes were categorized as hypothetical , hypothetical conserved , or annotated with known functions , according to the parameters set by the T . brucei genome consortium . Interpro domains associated to these genes were retrieved from TriTrypDB ( http://tritrypdb . org/tritrypdb/ ) . Gene products with less than 36% identity to a match in the public databases were considered hypothetical proteins , having no known function . When protein identity levels of 36% and higher to other hypothetical proteins were detected , the protein was considered hypothetical conserved . Hypothetical conserved genes , which were predicted to have homologs within the T . brucei genome , were further classified as hypothetical gene family members . Homology to other kinetoplastid species was determined using either existing data on the Sanger T . brucei website ( for Leishmania major and Trypanosoma cruzi , ( TriTryp ) ) , or by using the omniBLAST protein search function on the Sanger GeneDB website ( http://www . genedb . org ) . Signal peptide and cleavage site predictions were determined by SignalP ( http://www . cbs . dtu . dk/services/SignalP/ ) [20] . Trans-membrane predictions were made using DAS Software ( http://www . sbc . su . se/~miklos/DAS/ ) [21] . Predictions of glycosylation sites were performed using the NetNGlyc 1 . 0 ( http://www . cbs . dtu . dk/services/NetNGlyc/ ) and NetOGlyc Servers ( http://www . cbs . dtu . dk/services/NetOGlyc/ ) [22] . All prediction software is publically available on the internet . The parasite strains used were T . b . brucei RUMP 503 and T . b . rhodesiense YTAT 1 . 1 . For gene expression analysis , RNA was prepared from BSF T . b . rhodesiense expanded in rats . Trypanosomes were harvested from infected blood at peak parasitemia using DEAE cellulose chromatography [23] , [24] . For fly infections , BSF T . b . brucei expanded in rats were cryopreserved for subsequent use . Newly emerged male flies from the Glossina morsitans morsitans colony maintained in the Yale insectary received 2×106–2×107/mL T . b . brucei parasites in defibrinated bovine blood meal diet using an artificial membrane system [25] . After a single parasite challenge , flies were maintained on defibrinated bovine blood provided every other day . Flies were dissected after a minimum of 40 days post infection ( dpi ) and 72 hrs after their last blood meal . Salivary gland ( SG ) infection status was microscopically determined on a Zeiss Axiostar Plus light microscope at 400× . Infected SG , proventriculus ( PV ) and midgut ( MG ) tissues from the same flies were collected in Trizol , vortexed , and midguts were homogenized immediately . Metacyclic form ( MCF ) parasites were obtained by collecting the blood remaining on the feeding apparatus after flies with mature SG infections were fed . Blood was collected in PSG buffer ( 0 . 04 M Na2HPO4 2 H20 , 0 . 006 M NaH2PO4 2 H2O , 0 . 07 M NaCl , to pH 8 . 0 with 1 M H2PO4 ) , centrifuged 5 min . at 3000 rpm , and the pellet was resuspended in Trizol and stored at −20°C until RNA isolation . Total RNA was isolated from fly tissues and infected blood using Trizol extraction , according to manufacturer's instructions ( catalog no . 15596-026 , Invitrogen , California ) . Genomic DNA was removed by incubation with DNAse I , according to manufacturer's protocol ( catalog no . 04716728001 , Roche , Indiana ) . Reverse transcription was performed according to manufacturers instructions for oligo d ( T ) primed reactions ( SuperScript II Reverse Transcriptase , catalog no . 18064-014; RNaseOUT , catalog no . 10777-019 , Invitrogen , California ) . Nucleotide sequences for all experimental genes were obtained from the publicly available genome reference at the Sanger Institute ( http://www . genedb . org/Homepage/Tbruceibrucei927 ) . Primer sequences were identified by using the OligoPerfect™ Designer primer design tool ( http://tools . invitrogen . com/content . cfm ? pageid=9716 ) ( see Table S1 ) . All primer sets were used in a PCR amplification reaction with gDNA to confirm that they amplified a single gene fragment of the expected size . PCR amplification conditions were: 2 minutes hot start at 95°C , 32 cycles at ( 95°C for 45 s , 53°C for 45 s , 74°C for 1 min ) and 74°C for 6 min . Primers used to amplify procyclin transcripts were designed to recognize both EP and GPEET procyclin . The trypanosome structural gene alpha-tubulin was used for normalization of experimental cDNAs: trypanosome infected tsetse SG , PV , and MG , as well as BSF obtained from infected rats . Five and ten-fold serial dilutions of each cDNA pool were analyzed by PCR for the presence of alpha-tubulin transcripts . Cycling conditions were: 2 min at 95°C , 28 cycles at ( 95°C for 45 s , 53°C for 45 s , 74°C for 1 min ) and 74°C for 6 min . The PCR amplification products from the different cDNA dilutions were resolved on a 1% agarose gel , visualized on a KODAK Image Station 2000R and gel images were captured using the IS2000R Image Aquire Software ( Eastman Kodak Co , Rochester New York ) . The cDNA dilutions that resulted in PCR products of equal intensity from the different tissue samples were identified and all subsequent PCR reactions were performed using these cDNA template dilutions . All experimental reactions were performed using the cDNA templates prepared as described above at 32 and 36 cycles for each sample in duplicate . As controls , alpha-tubulin and BARP sequences were amplified at 32 and 36 cycle reactions , respectively . Primer sequences can be found in Table S1 . All amplification products were analyzed by electrophoresis and imaging as described above . Genes that resulted in no amplification products or that yielded multiple bands after amplification were excluded from further analysis ( Table S2 ) . Expression analysis was repeated for genes that yielded a product in only one tissue cDNA or for genes with unclear results due to low levels of expression . Gel images obtained from the 36 cycle reactions were used to obtain a semi-quantitative measurement of expression variation between different developmental samples . The values were normalized to the trypanosome alpha-tubulin control to account for variation between the four experimental tissue samples ( SG , PV , MG , and BSF ) and experimental runs . The adjusted expression values based on alpha-tubulin levels were used to categorize the expression profile of experimental genes . Based on these adjusted values , the fold change was calculated for the four developmental samples tested , for each gene yielding expression data . Where no expression could be detected , that transcript was classified as not detected ( nd ) . If expression in one tissue was at least 2-fold higher than any other tissue , that gene was classified as being specific to that tissue . Parasite gene expression was classified as preferential for a tissue ( or tissues ) when gene expression was detected but the levels were less than 2 fold higher than that detected in other tissues . Genes with expression levels too low to be confidently categorized , or with expression profiles not corresponding to any other category were classified as miscellaneous . Expression levels were classified as high , medium or low based on the adjusted net intensities of the most prominent band for the experimental gene . Net intensity values ≥501 were classified as high , 101–500 as medium , and 0–100 as low . All expression data are being submitted to TriTrypdb . org . To validate the expression profiles observed with semi-quantitative analysis , 5 genes were selected for quantitative RT-PCR ( qRT-PCR ) . Standard curves were developed for each gene using serial dilutions of plasmids containing cloned inserts . Each standard was used to calculate transcript numbers in the experimental cDNAs tested . qRT-PCR primers and cycling conditions are listed in Table S3 . All reactions were performed on an icycler iQ real time RT-PCR detection system ( Bio-Rad ) . Three independent biological replicates of infected SG , PV and MG tissues were used , with 2 technical replicates per sample . For comparison to the quantitative data , the semi-quantitative fold change data was evaluated based on the SG , PV , and MG data points . As no BSF samples were evaluated by qRT-PCR , the semi-quantitative data for the BSF parasites was excluded from this comparison . Alpha-tubulin levels were used for expression normalization . Values are represented as the mean fold change ( ±SEM ) and statistical significance was determined using a Student's t test implemented in Microsoft Excel software . An in silico analysis of the T . b . brucei strain 927 genome data using the BigPI GPI-anchor prediction software identified 163 putative proteins with GPI anchor attachment motifs . Fifty-seven of these gene products had known or predicted functions such as BARP , GP63 , trans-sialidase and the procyclin-associated genes , and were excluded from further analysis ( Table S4 ) . The remaining 106 putative proteins were evaluated for the presence of conserved domains ( Table S5 ) . These putative products were further searched for glycosylation , signal peptide , and trans-membrane domains , and a second predictive algorithm for GPI-anchor attachment domain ( FragAnchor ) was applied ( Table S6 ) . Typical GPI anchored proteins are expected to have a signal peptide and no trans-membrane domains [26] . Our analysis reduced the initial 106 genes down to 25 genes , which were predicted to encode products with GPI-anchor attachment motifs ( Table 1 ) . Of the 25 highly probable GPI-anchored gene products with unknown functions , only Tb09 . 142 . 0410 was considered to be hypothetical , having no identified homologs . Two genes ( Tb927 . 4 . 3290 and Tb927 . 10 . 990 ) were shared only with Trypanosoma congolense , while five others were conserved at the level of the TriTryp genomes . Seventeen genes were identified to be members of larger gene families . Interestingly , these were not widely shared between related kinetoplastids . Only one family ( Tb927 . 8 . 930 and Tb927 . 8 . 950 ) had homologs outside of the T . brucei complex , and these were found in Trypanosoma vivax . The remaining gene family members were either detected only as repeated genes in the genome of T . b . brucei ( 9 ) , or as having homologs in the genome of T . b . gambiense ( 6 ) . Here we report on the identification of T . brucei genes encoding predicted unknown surface proteins obtained via in silico GPI-anchor attachment signal sequence prediction analysis . Expression profiling analysis from mammalian and tsetse developmental stages indicate that transcripts for the majority of the hypothetical and hypothetical conserved proteins are expressed in parasites during their development in the tsetse salivary glands and proventriculus . Most notably , we identified 8 trypanosome genes specifically expressed in parasitized salivary glands , expression for all of which was also detected from mammalian infective MCF trypanosomes present in fly saliva . The results of this analysis give the first large-scale insight into stage-regulated expression of genes encoding putative hypothetical surface proteins during key developmental processes in the tsetse fly , and support the established paradigm of differential expression through development . Functional characterization of these unknown proteins , particularly expressed by metacyclics in saliva , ay lead the way to novel transmission blocking strategies in the mammalian host . Proteins with GPI posttranslational modification are typically expressed on the surface of eukaryotic parasites and have the potential to participate in important biological processes such as cell–cell interactions , signal transduction , endocytosis , complement regulation , and antigenic presentation [27] . In protozoan parasites , GPI anchored glycoconjugates extensively coat the plasma membrane and are involved in many aspects of host–parasite interactions , such as adhesion and invasion of host cells , modulation and evasion from host immune response [26] . As such , there is interest in identifying the surface proteins of the medically important kinetoplastids , as reported in L . ( V . ) braziliensis and T . cruzi where proteomic techniques were applied to capture this class of proteins [28]–[30] . Current knowledge of the VSGs and procyclins , two of the best characterized GPI-anchored surface proteins of T . brucei has demonstrated the importance of these proteins in trypanosome developmental processes . Further , GPI biosynthesis has also been implicated as a molecular target for development of new drugs against African sleeping sickness [31] , [32] . The availability of the T . brucei genome allows for postgenomic discoveries including screens for hallmark motifs such as GPI anchor attachment signals associated with surface proteins [26] . Several publically available programs can be used to predict post-translational modifications ( PTM ) such as glycosylation and GPI-anchor attachment , although a gold standard for prediction software remains to be found [33] . As a result , experimental validation of predicted features is always warranted . The quality of predictive algorithm outputs vary in response to several factors . In the case of GPI-anchor prediction , variables include the size of the motif recognized , quality of the underlying data used to test the algorithm , and correct application of learning procedures such as neural networks [34] , [35] . The ideal tool would have high sensitivity to detect true positives , with a low false prediction rate [33]–[35] . Also relevant is the biological context being considered , as a result there are algorithms specifically for protozoa , fungi , plants , etc [34] . As seen with our dataset , two algorithms can generate different results from the same dataset . In this work , FragAnchor agreed with most , but not all of those genes previously identified by a BigPI search specific for protozoa GPI anchor attachment domains . A similar outcome with these two programs was reported after testing both against known positive and negative control GPI-anchored protein datasets [34] , and against a dataset from the protozoan pathogen Plasmodium falciparum [19] . In both of these cases , although correct identification of true GPI-anchored proteins was high , the false positive rate was high as well . Conversely , another group found FragAnchor to be more accurate than BigPI , while maintaining the same false positive rate [35] , although limitations associated with the algorithm they employed for comparison make it difficult to draw clear conclusions [34] . With these challenges in mind , we opted for a conservative approach in the identification of putative GPI-anchored proteins by selecting only those genes encoding products that showed agreement between the two predictive programs . As the absence of predicted trans-membrane domains is necessary to support a prediction of GPI-anchoring [26] , we further excluded putative proteins bearing any predicted trans-membrane domains from expression analysis despite predictions of GPI-anchoring . While the presence of a GPI anchor attachment signal suggests cell surface membrane expression as mentioned earlier , there is evidence that both N- and O- glycosylation status directs nascent proteins to the apical region [35]–[37] . Like GPI anchor attachment sites , glycosylation sites can be predicted using in silico methodology . Importantly , while the presence of predicted glycosylation sites support the expectation of surface expression , the absence of glycosylation does not imply a lack of surface expression of a protein [38] . Fifty-six of the in silico-identified genes in the T . b . brucei genome had known or predicted functions in other closely related kinetoplastid parasites and were not pursued for further expression analysis . These included all members of the BARP family , and many genes with putative functions , such as GP-63 surface protease ( 5 copies ) , trans-sialidase ( 4 copies ) , procyclin associated gene 4 ( 2 copies ) , and numerous carrier or transporter proteins . Our aim was to identify unknown SG stage-regulated genes for downstream characterization and investigation as novel transmission blocking targets . Of the 163 non-procyclin , non-VSG coding genes that were identified as encoding GPI-anchor proteins using the BigPI prediction software , 104 were confirmed with FragAnchor . With regard to possible function of these gene products , 106/163 had no known functions . A search of the available whole genome sequence information from T . b . gambiense , L . major , T . cruzi , T . congolense and T . vivax indicated that about 21% ( 22/106 ) of the identified genes were unique to T . b . brucei . With regard to the 25 genes that met our criteria to be considered likely to encode predicted GPI-anchored proteins , 5 were conserved at the level of the TriTryp genomes , 10 were shared with other species of Trypanosoma , and 10 were unique to T . b . brucei . It is possible that the lack of homologs in these genomes reflects the different biology of the parasite species , although it is also possible that as genome annotations improve homologs may be revealed . While T . b . gambiense is more closely related to T . b . brucei than the other trypanosomatid species analyzed , its biology differs from T . b . brucei . It remains to be seen if the unique genes in T . b . brucei genome contribute to its differing epidemiology . The annotated whole genome sequence of T . b . rhodesiense is not yet available , however , the status of T . b . brucei specific genes in T . b . rhodesiense is of interest both from an evolutionary and epidemiological point of view . Gene expression profiling analysis showed that the majority of the 21 genes for which we detected transcripts , are expressed by trypanosome developmental stages present in the tsetse fly PV and SG tissues , while comparatively fewer are expressed by mammalian bloodstream forms and none in the MG . A similar trend was found in genes encoding proteins with less likelihood of GPI anchoring . Similarly , a proteomic analysis that identified GPI-anchored molecules in T . cruzi insect-stage epimastigote cultures also found the majority of the identified proteins to be novel [30] . In the case of T . brucei , obtaining sufficient epimastigote and metacyclic parasites from infected tsetse flies for functional analysis is difficult since these stages are unculturable in vitro . Confirmation of the corresponding stage-regulated protein expression is a necessary next step , and the resulting data may shed light on the roles of these products in parasite biology . Complex gene expression profiles for putative surface proteins in the proventricular and salivary gland stages of T . brucei may reflect the multiple discrete trypanosome developmental stages infecting these tissues , or heightened sensitivity of these trypanosomes to the tsetse or mammalian bite-site host environment . Unlike the SG and PV , far fewer unknown putative surface proteins were associated with the BSF and MG stages . This minimal detection of unknown transcripts in PF and BSF samples may be related to the abundant expression of known GPI-anchored major surface proteins in these stages- specifically the procyclins and VSGs , respectively . Interestingly , genes encoding 8 of the 21 putative GPI-anchored proteins were specifically upregulated by parasites infecting tsetse SG . Although trypanosomes undergo four distinct developmental steps in this tissue , only two GPI-anchored protein families have been demonstrated on the surface of any SG stages to date . The alanine-rich BARP proteins are expressed on epimastigotes attached to the salivary gland epithelium . Free metacyclics in saliva no longer express BARP , but have upregulated the metacyclic variant surface glycoproteins ( M-VSGs ) in advance of inoculation into the mammalian host [17] , [39] . The data presented here suggest a more complex series of events may be involved in the maturation of the SG-inhabiting trypanosome stages . Proteins specifically expressed on the immature SG stages might be involved in host-parasite interactions and as such could be targeted to prevent parasite maturation in the fly using genetic modification strategies in the tsetse host [40] . On the other hand , proteins expressed on the mature metacyclics may present novel vaccine targets for use in the vertebrate hosts . Importantly , transcripts corresponding to the SG stage-regulated genes were not detected in the bloodstream form stages . Since the mammalian infective metacyclic trypomastigote is suggested to be “pre-adapted” to life in the vertebrate host , one could expect these samples to share proteins . There are two potential explanations for this observation . First , many gene products associated with adaptation to the vertebrate environment are likely to be intracellular i . e . related to energy metabolism , and therefore not bearing GPI-anchor attachment domains . As a result , these genes are expected to have been excluded from the in silico screen applied here . Second , when an infective fly bites the vertebrate host , metacyclic parasites are detected for several days with the bloodstream forms being not apparent until nearly a week after the infective bite [41] , [42] . Thus it is possible that transitional metacyclics ( t-MCFs ) , i . e . those detected in vertebrate blood in the days immediately after an infective tsetse bite , but before differentiation to the BSF , may have a transcriptome that reflects the parasite adaptation process from the environment of invertebrate saliva to vertebrate blood . MCF trypanosomes , like malarial sporozoites , are the critical developmental stage of the parasite which gives rise to infection in the vertebrate host . While considerable effort has been mounted towards development of a sporozoite vaccine for the prevention of malaria , this has not been the case with the MCF of T . brucei . To date , VSGs have effectively thwarted all attempts at developing a vaccine against the mature BSF . It is thought that MCF parasites also express variable proteins ( M-VSGs ) , which would hamper vaccine development efforts targeting MCF . Our results suggest however that GPI-anchored surface protein repertoire of MCF may be more complex and different from the BSF forms than originally thought . The expression of the genes encoding putative surface proteins on the mammalian-infective stage suggests a complex interface of MCF and mammalian bite-site . In summary , the in silico and semi-quantitative gene expression analyses approach used here has allowed an important first look at the stage-regulated expression of genes encoding putative GPI-anchored proteins with no known functions in the human and animal pathogen T . brucei . The findings presented here suggest that the tsetse host-parasite interplay during differentiation may be quite complex . Most importantly , these results greatly increase our understanding of trypanosome biology at the point of transmission to the vertebrate host , and identify a number of putative invariant surface proteins , which could be investigated further for novel transmission blocking strategies .	Human African Trypanosomiasis ( HAT ) is a fatal disease caused by African trypanosomes and transmitted by an infected tsetse fly . Presently , there are no vaccines to prevent mammalian infections . Proteins expressed on the trypanosome surface can influence the host environment and allow for their transmission . Potentially accessible to the adaptive immune systems of vertebrate hosts , these proteins could serve as future vaccine targets . Identification and characterization of these currently unknown proteins can help us develop strategies to alter the host environment , making it inhospitable for the parasite , thereby reducing disease transmission . While there is extensive knowledge about trypanosome development in the mammalian host , less is known about the molecular events in the tsetse fly , particularly the salivary gland stages . We used an in silico approach to identify putative surface proteins from the known genome sequence of Trypanosoma brucei , and we describe the stage specific expression of these genes during development in the tsetse fly and mammalian host . Our findings show that a majority of unknown transcripts encoding predicted surface proteins are expressed by the parasites infecting tsetse salivary glands . These data will help focus future investigations into transmission-blocking approaches targeting the expressed antigens of trypanosomes infecting tsetse salivary glands .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "veterinary", "diseases", "biology", "microbiology", "molecular", "cell", "biology", "veterinary", "science" ]	2012	Transcript Expression Analysis of Putative Trypanosoma brucei GPI-Anchored Surface Proteins during Development in the Tsetse and Mammalian Hosts
Meiotic recombination plays a critical role in sexual reproduction by creating crossovers between homologous chromosomes . These crossovers , along with sister chromatid cohesion , connect homologs to enable proper segregation at Meiosis I . Recombination is initiated by programmed double strand breaks ( DSBs ) at particular regions of the genome . The meiotic recombination checkpoint uses meiosis-specific modifications to the DSB-induced DNA damage response to provide time to convert these breaks into interhomolog crossovers by delaying entry into Meiosis I until the DSBs have been repaired . The meiosis-specific kinase , Mek1 , is a key regulator of meiotic recombination pathway choice , as well as being required for the meiotic recombination checkpoint . The major target of this checkpoint is the meiosis-specific transcription factor , Ndt80 , which is essential to express genes necessary for completion of recombination and meiotic progression . The molecular mechanism by which cells monitor meiotic DSB repair to allow entry into Meiosis I with unbroken chromosomes was unknown . Using genetic and biochemical approaches , this work demonstrates that in the presence of DSBs , activated Mek1 binds to Ndt80 and phosphorylates the transcription factor , thus inhibiting DNA binding and preventing Ndt80’s function as a transcriptional activator . Repair of DSBs by recombination reduces Mek1 activity , resulting in removal of the inhibitory Mek1 phosphates . Phosphorylation of Ndt80 by the meiosis-specific kinase , Ime2 , then results in fully activated Ndt80 . Ndt80 upregulates transcription of its own gene , as well as target genes , resulting in prophase exit and progression through meiosis . One of the most dangerous things for a cell is the occurrence of DNA double strand breaks ( DSBs ) in its chromosomes . Failure to repair a DSB may result in a loss of genetic material and lethality . DSBs arise due to exogenous damage such as radiation , or endogenous errors such as stalled replication forks . Repair of DSBs by non-homologous end joining may lead to deletions , translocations or inversions , which can have adverse consequences such as cancer [1] . The most conservative way to repair a DSB is by homologous recombination , using the sister chromatid as the template . Indeed , in mitotically dividing cells , homologous recombination mediated by the evolutionarily conserved recombinase , Rad51 , is biased towards using sister chromatids [2 , 3] . DSBs trigger an evolutionarily conserved DNA damage checkpoint , which delays or arrests cell cycle progression to provide time for repair [4] . The DNA damage checkpoint is mediated by two kinases , Tel1 ( ATM in mammals ) , which responds to blunt ends , and Mec1 ( ATR in mammals ) which is activated by single stranded DNA generated by resection of the 5’ ends of the breaks . In yeast , these kinases phosphorylate the adaptor protein , Rad9 , which in turn recruits the Forkhead-associated ( FHA ) -domain containing effector kinase , Rad53 , ( related to Chk2 in mammals ) , resulting in Rad53 autophosphorylation and activation . Rad53 phosphorylation of various proteins then prevents cohesin destruction and mitotic exit . While the purpose of mitosis is to produce genetically identical daughter cells , the specialized cell division of meiosis divides the chromosome number in half to produce gametes for sexual reproduction . After premeiotic chromosome duplication , meiotic chromosomes segregate twice without an intervening round of DNA synthesis: homologous pairs of sister chromatids go to opposite poles at Meiosis I ( MI ) , while sister chromatids separate at Meiosis II ( MII ) . Proper alignment at Metaphase I requires tension that is generated when sister kinetochores from one homolog attach to the spindle pole opposite that of the other homolog . This tension occurs because homologs are physically connected by a combination of crossovers ( COs ) and sister chromatid cohesion [5] . COs are initiated by programmed DSBs generated by Spo11 , a meiosis-specific , evolutionarily conserved topoisomerase-like protein that cuts in preferred regions of the genome called “hotspots” [6] . Unlike mitotic cells , meiotic DSB repair is biased to use the homolog as the repair template [7] . It is key that every pair of homologs contains at least one crossover . Towards this end , many more DSBs are generated during meiotic prophase than the number of necessary COs ( e . g . , yeast , 160 DSBs/16 homolog pairs; mouse , 250–300 DSBs for 20 homolog pairs ) [6] . The repair of these breaks must be carefully regulated to ensure not only the requisite number of COs , but also that no breaks remain when Anaphase I begins . The meiotic recombination checkpoint delays meiotic prophase while interhomolog recombination is occurring . This checkpoint uses meiosis-specific modifications to the DNA damage checkpoint and is dependent upon protein components of a specialized chromosomal structure called the synaptonemal complex ( SC ) [8–10] . After chromosome duplication in yeast , cohesin complexes containing the meiosis-specific Rec8 kleisin subunit hold sister chromatids together [11] . Sister chromatids condense along protein cores containing Rec8 , as well as the meiosis-specific Hop1 and Red1 proteins , to form axial elements ( AEs ) . Hop1 contains the evolutionarily conserved HORMA domain which mediates homo-oligomerization , as well as interaction with Red1 [12–15] . Chromosome condensation occurs by the formation of chromatin loops , with axis proteins at their bases [16 , 17] . Spo11 is indirectly recruited to the axes by phosphorylation of the DSB protein , Mer2 [17–19] . In addition , Mer2 interacts with Spp1 , which binds to trimethylated histones flanking hotspot sequences to bring the hotspots to the axis [17 , 20 , 21] . DSB formation on the loops therefore occurs in the vicinity of Hop1 and Red1 on the axis . COs created by DSB repair are primarily generated using a functionally diverse set of proteins collectively called the ZMM proteins ( Zip1-3 , Zip4/Spo22 , Msh4 , Mer3 , Msh5 , and Spo16 ) [22 , 23] . Holliday junctions formed by the ZMM pathway exhibit biased resolution to form COs that are distributed throughout the genome [22 , 24 , 25] . The ZMM pathway is also necessary to form stable associations between homologs , leading to the insertion of the transverse filament protein , Zip1 , between the AEs to create the tripartite SC [22 , 26 , 27] . At the pachytene stage of meiotic prophase , all the homolog pairs are fully synapsed . Similar to vegetative cells , meiotic DSBs result in the recruitment and activation of the Tel1 and Mec1 checkpoint kinases . These kinases phosphorylate Hop1 , which replaces Rad9 as the adaptor [28] . Phosphorylated Hop1 is bound by the FHA domain of the meiosis-specific paralog of Rad53 and Chk2 , Mek1 ( also known as Mre4 ) , resulting in Mek1 oligomerization and activation by autophosphorylation in trans [28–31] . Chromatin-immunoprecipitation experiments using phosphorylation of Histone H3-T11 as a marker for Mek1 activity revealed that this activity is highest at axis sites that correlate with the presence of Hop1 and Red1 and can spread for several kilobasepairs ( kb ) surrounding a DSB [32] . Mek1 is a key regulator of meiotic DSB repair . It promotes interhomolog bias by inhibiting Rad51 from interacting with its accessory factor , Rad54 in two ways: ( 1 ) phosphorylating and stabilizing Hed1 , a meiosis-specific protein that binds to Rad51 , thereby excluding Rad54 and ( 2 ) phosphorylating Rad54 which reduces its affinity for Rad51 [33–36] . These mechanisms prevent Rad51 from competing with the meiosis-specific recombinase , Dmc1 , which mediates the bulk of meiotic recombination [37 , 38] . Mek1 antagonizes sister chromatid cohesion locally at DSBs to facilitate strand invasion of homologs and regulates whether interhomolog recombination intermediates are repaired as either COs or noncrossovers by enabling phosphorylation of Zip1 by the Cdc7-Dbf4 ( DDK ) cell cycle kinase [39 , 40] . Finally , MEK1 is required for the meiotic recombination checkpoint delay that prevents cells from entering into the meiotic divisions with unrepaired DSBs [41–44] . Checkpoint delay is part of the normal meiotic program , but this delay can be exacerbated in mutants that initiate , but fail to complete , DSB repair . An extreme case occurs in dmc1Δ diploids in the SK1 strain background , where strand invasion does not occur because Dmc1 is absent and Rad51 activity is inhibited by Mek1 [34 , 45–47] . The high number of DSBs generates high levels of activated Mek1 , resulting in meiotic prophase arrest due to a lack of Cdc28-Clb1 ( CDK-Clb1 ) activity [8 , 41 , 45 , 48 , 49] . The checkpoint inhibits CDK-Clb1 by two separate mechanisms: ( 1 ) activation and stabilization of the Swe1 kinase which places an inhibitory phosphate on tyrosine 19 of Cdc28 [50] and ( 2 ) inactivation of the meiosis-specific transcription factor , Ndt80 , thereby preventing CLB1 transcription [51–53] . During early meiotic prophase , a meiosis-specific E3 ligase targets mitotic regulators such as polo-like kinase ( Cdc5 ) and Clb1 cyclin for degradation [54] . As a result , their production is dependent upon the transcriptional activity of Ndt80 . Ndt80 is a sequence-specific DNA binding protein that recognizes a nine-base pair sequence called the middle sporulation element ( MSE ) in the promoters of >300 target genes ( called “middle” and “late genes” ) [51 , 55 , 56] . NDT80 transcription occurs in two stages [57] . In the first stage , expression of NDT80 requires the transcriptional regulator Ime1 , which is also responsible for transcribing early genes such as HOP1 , MEK1 , SPO11 and DMC1 [58] . NDT80 transcription is delayed relative to the early genes , however , because of the Sum1 repressor , which binds to MSEs in the NDT80 promoter and the promoters of Ndt80 target genes [59] . Sum1 removal requires phosphorylation by the meiosis-specific Ime2 kinase , in combination with CDK and DDK [60–62] . Since IME2 is an early gene , it must be transcribed and translated before Sum1 repression can be relieved , hence the delay in Ime1-mediated NDT80 transcription . The relatively low level of Ndt80 protein generated by Ime1 is inhibited by the meiotic recombination checkpoint until sufficient DSB repair has occurred to lower Mek1 kinase levels below the amount necessary to inactivate Ndt80 [41 , 51–53] . The second stage of NDT80 transcription is marked by phosphorylation of Ndt80 by Ime2 that facilitates Ndt80’s ability to activate transcription [48 , 63 , 64] . Ndt80 then activates transcription of its own gene to initiate a positive feedback loop , as well as promoting transcription of target genes such as CDC5 . Expression of CDC5 triggers to resolution of Holliday junction intermediates into COs and degradation of Red1 to dissemble the SC [41 , 54 , 65] . Removal of Red1 leads to inactivation of the remaining Mek1 , allowing residual DSBs to be repaired prior to CLB1-promoted entry into Meiosis I [41] . Exit from pachynema and entry into Meiosis I has been proposed to be controlled by a switch between two stable states [54] . In the first state , CDK-Clb1 levels are low due to the meiotic recombination checkpoint , thereby preventing meiotic progression . In the second state , CDK-Clb1 levels are high because DSBs have been repaired , leading to a decrease in the checkpoint signal and activation of Ndt80 , thereby allowing CLB1 transcription and progression into the meiotic divisions . What was unknown was how this switch is controlled . This work shows that Mek1 , after being activated by DSBs , directly binds and phosphorylates Ndt80 , thereby inhibiting Ndt80 from activating transcription . As DSBs are repaired , Mek1 activity decreases , and inhibitory Mek1 phosphosites are removed . Ime2 phosphorylation then promotes Ndt80 activity , resulting in expression of genes necessary for completing recombination and exiting prophase . Mek1 phosphorylation of Ndt80 therefore provides an elegant way for cells to know when it is safe to enter the first meiotic division . A two-hybrid screen using lexA-MEK1 revealed an interaction with a fragment of NDT80 ( amino acids 287–627 ) fused to the Gal4 activation domain ( GAD ) . This fusion is hereafter referred to as GAD-NDT80 . The strain contained HIS3 and lacZ reporter genes under the control of promoters containing lexA operator sites [66] . Two-hybrid interactions were therefore manifested either by growth on medium lacking histidine or production of ß -galactosidase . The GAD-NDT80 fragment begins near the end of the Ndt80 DNA binding domain ( DBD ) and goes to the end of the protein . In addition to the activation domain in the C terminus , this fragment includes a 57 amino acid sequence in the middle of Ndt80 that is required for meiotic recombination checkpoint arrest ( Fig 1A and 1B , row 2 ) [64 , 67–69] . The NDT80-bc allele , which encodes an Ndt80 protein deleted for this 57 amino acid sequence , no longer responds to the checkpoint triggered by unrepaired breaks in both zip1Δ and dmc1Δ mutants ( bc stands for “bypass checkpoint” ) [69] . We have therefore named this 57 amino acid sequence the “bc” domain . Disruption of the Mek1 FHA domain using the R51A mutation had no effect on the Ndt80 interaction , indicating that the FHA domain does not mediate binding ( Fig 1C ) [47 , 70] . In contrast , deletion of the bc domain from GAD-NDT80 eliminated interaction with lexA-MEK1 , even though the GAD-Ndt80-Δbc protein was more abundant than GAD-Ndt80 , ruling out protein instability as the reason for the loss of the two-hybrid signal ( Fig 1B , row 3 and 1D , lanes 2 and 3 ) . In addition to being necessary for lexA-MEK1 interaction , the bc domain was also sufficient , as the GAD-bc fusion produced a positive two-hybrid signal in combination with lexA-MEK1 ( Fig 1B , row 4 ) . A 60 amino acid sequence containing the bc domain was used to probe Ndt80 proteins from other fungi for homology . A small region containing amino acids 371–375 , RPSKR , is conserved in several yeast species ( Fig 1E ) . A consensus motif generated from these alignments showed that lysine ( K ) 374 and arginine ( R ) 375 from S . cerevisiae Ndt80 are completely conserved ( Fig 1F ) . Deletion of the sequence encoding RPSKR from GAD-NDT80 abolished the two-hybrid interaction with lexA-MEK1 , as did substituting the KR sequence with aspartic acids ( DD ) ( Fig 1B , rows 5 and 7 ) . The KR to alanine ( AA ) mutant still interacted with lexA-MEK1 , although not quite as well as GAD-NDT80 ( Fig 1B , compare rows 6 and 2 ) . The RPSKR sequence is therefore required for interaction between lexA-Mek1 and Ndt80 . The Ndt80-Mek1 interaction has thus far not been confirmed by co-immunoprecipitation experiments from meiotic extracts due to technical problems obtaining soluble Ndt80 . A functional genetic approach was therefore used to test the importance of this interaction in vivo . Overexpression of NDT80 can partially bypass the meiotic recombination checkpoint arrest triggered by the unrepaired DSBs that accumulate when the DMC1 recombinase is absent [53 , 69 , 72] . One explanation for this result is that during meiosis in wild-type ( WT ) cells there is sufficient Mek1 to bind and inactivate all of the Ime1-dependent Ndt80 protein . However , when Ndt80 is in excess of Mek1 , some Ndt80 escapes phosphorylation , resulting in transcription of the NDT80 gene to start the positive feedback loop leading to meiotic progression . If this model is correct , and if the bc domain recruits Mek1 to Ndt80 in dmc1Δ-arrested cells , then over-expressing the bc domain by itself could titrate Mek1 away from endogenous Ndt80 , resulting in activation of the transcription factor and sporulation . To limit expression of the GAD-bc fusion to meiotic cells , the hybrid gene was placed under the control of the MEK1 promoter . A dmc1Δ diploid transformed with a DMC1 CEN ARS plasmid only partially complemented the sporulation defect , perhaps due to plasmid loss during growth on the Spo plate ( Fig 1G ) . The GAD-bc transformants partially bypassed the dmc1Δ checkpoint arrest , exhibiting increased sporulation after three days on Spo medium compared to GAD alone ( Fig 1G ) . Further support for the titration model is that this partial checkpoint bypass was decreased when counteracted by overexpression of GST-MEK1 from a high copy number plasmid ( Fig 1G ) . Previous work has shown that the presence of Cdc5 is sufficient to trigger SC disassembly and Red1 degradation [41 , 54 , 65 , 75] . These experiments showed that induction of CDC5 in the ndt80Δ background ( where Mek1 levels are low ) resulted in the disappearance of Red1 and elimination of the SC [41 , 65] . Interfering with the Mek1-Ndt80 interaction in the dmc1Δ background allowed activation of Ndt80 ( indicated by increased phosphorylation ) and production of Cdc5 in the presence of high levels of Mek1 activity ( Fig 3C , dmc1Δ ndt80-ΔRPSKR and dmc1Δ ndt80-KR>DD ) . In these cases , Red1 persisted for at least two hours after Cdc5 was first detected . In contrast , Red1 was eliminated within two hours after the appearance of Cdc5 in the dmc1Δ mek1Δ diploid ( Fig 3C ) . These results suggest that Cdc5 is not as efficient in targeting the degradation of Red1 in the presence of high levels of Mek1 activity ( Fig 3B ) . Red1 disappears more rapidly in the dmc1Δ ndt80-ΔRPSKR mutant than dmc1Δ ndt80-KR>DD and persisted for the length of the time course in the dmc1Δ ndt80-KR>AA strain ( Fig 3C ) . These differences reflect the larger defect in Mek1 interaction resulting from the deletion of the RPSKR sequence compared to the aspartic acid substitution mutations . Ndt80 was activated more quickly in ndt80-ΔRPSKR ( increased phosphorylation at 8 hours ) ( Fig 3C ) so Cdc5 was produced earlier as well . While Mek1 kinase activity delayed Red1 degradation in the presence of Cdc5 , it did not prevent it completely . As a result , Mek1 was gradually decreased due to loss of Red1 , allowing DSB repair by Rad51 ( indicated by loss of Hop1 phosphorylation ) , leading to a further reduction in Mek1 activity and more efficient Cdc5-dependent degradation of Red1 ( Fig 3C ) . These results have therefore uncovered yet another mechanism to ensure that cells do not enter MI prematurely , i . e . , the prevention of Red1 degradation and therefore , SC disassembly , when Mek1 levels are high . The mechanism by which Mek1 inhibits Red1 degradation remains to be determined . Meiotic time courses were performed with NDT80 , NDT80-6A , ndt80-6D and ndt80-R177A diploids . The ndt80-R177A diploid was used as a negative control as the Ndt80-R177A protein is defective in binding to MSEs and therefore is unable to activate transcription either of itself or other NDT80 targets [61 , 67 , 68 , 77] . The ndt80-6D diploid was phenotypically identical to ndt80-R177A , indicating that it also is defective in activating transcription . Both mutants arrested in meiotic prophase , while NDT80 and NDT80-6A exhibited similar kinetics for meiotic progression ( Fig 4A ) . The R177A and 6D diploids entered the meiotic program efficiently , as evidenced by similar levels of phosphorylated Hed1 protein compared to NDT80 and NDT80-6A at the 4-hour time point ( Fig 4B ) [34] . Whereas the Ndt80 and Ndt80-6A protein levels peaked at six hours and then decreased until they were nearly gone by 10 hours , the R177A and 6D proteins exhibited reduced levels that slowly accumulated throughout the length of the time course ( Fig 4B and 4C ) . This result is consistent with the occurrence of Ime1-driven transcription of the ndt80-R177A and ndt80-6D genes , followed by a failure of the mutant proteins to activate transcription of their own genes . In addition , the R177A and 6D mutants failed to express CLB1 and CDC5 , although both proteins were observed for NDT80 and NDT80-6A ( Fig 4B ) . An alternative explanation for the ndt80-6D phenotypes is that Ndt80-6D is transcriptionally active , but the aspartic acid substitutions destabilize the protein so that there is insufficient Ndt80-6D protein to promote transcription of CDC5 , CLB1 , etc . This hypothesis was tested using ndt80-6D under control of the GAL1 promoter in a strain containing a GAL4-estrogen receptor fusion ( GAL4-ER ) . The resulting allele ( indicated as ndt80-6D-IN ) can be induced ectopically by addition of estradiol to the Spo medium [41 , 48 , 78] . If the aspartic acid residues destabilize Ndt80 , then the induced Ndt80-6D levels should be lower compared to Ndt80 and Ndt80-6A . In contrast , if the reduced level of the endogenous Ndt80-6D protein is due to a failure in Ndt80-activated transcription of the ndt80-6D gene , Ndt80-6D levels should be equivalent to Ndt80 and Ndt80-6A , since transcription is now under the control of a heterologous promoter . Induction of the NDT80-IN alleles after five hours in Spo medium resulted in meiotic progression of the NDT80 and NDT80-6A diploids , while ndt80-6D remained arrested in prophase ( Fig 4D ) . Both the Ndt80-6A and Ndt80-6D proteins were present in greater abundance than Ndt80 throughout the timecourse . Importantly , the Ndt80-6A and Ndt80-6D proteins exhibited similar kinetics of induction and peaked at the same level . However , while the Ndt80-6A protein was nearly gone by 9 hours , the Ndt80-6D protein persisted and exhibited reduced phosphorylation ( Fig 4E and 4F ) . Cdc5 and Clb1 were generated in the WT and NDT80-6A strains but not in ndt80-6D , confirming that constitutive negative charges at Mek1 consensus sites in the DBD impede the ability of Ndt80 to activate transcription ( Fig 4E ) . Phosphorylation of Ndt80 by Ime2 results in multiple mobility shifts that enhance Ndt80 transcriptional activity [48 , 63 , 64] . One report found that inactive Ndt80 derived from checkpoint arrested cells was not phosphorylated [53] , and suggested that Ndt80 phosphorylation is solely used for activation of the transcription factor . A different group detected a phosphorylation-dependent mobility shift in a dmc1Δ diploid which was not as slow as the Ndt80 mobility shifts observed from WT cells [63] , consistent with our hypothesis that Mek1 phosphorylation of Ndt80 is inhibitory . One difficulty with interpreting these experiments is that the checkpoint prevents Ndt80 from activating transcription of itself , and therefore Ndt80 protein levels are low , making the protein more difficult to detect [51–53] . The estradiol-inducible NDT80 system was therefore used to determine whether inactive Ndt80 is phosphorylated . A dmc1Δ mek1-as NDT80-IN diploid was incubated in Spo medium for 5 hours to arrest cells with unrepaired DSBs . The mek1-as allele encodes an analog-sensitive ( as ) kinase with an enlarged ATP binding pocket that allows for inhibition of the kinase by addition of the 1-NA-PP1 inhibitor to the Spo medium [47] . NDT80 transcription was induced by addition of estradiol in the presence or absence of Mek1-as inhibitor . Inactivation of Mek1 resulted in loss of phosphorylated Hop1 at the 6-hour time point , consistent with repair of DSBs , loss of Hed1 phosphorylation and efficient meiotic progression ( Fig 5A and 5B , +1-NA-PP1 ) . Ndt80 was highly phosphorylated , resulting in production of Cdc5 and destruction of Red1 ( Fig 5B , +1-NA-PP1 ) . That this high level of phosphorylation occurs only after Mek1 inactivation suggests that Mek1 activity somehow inhibits phosphorylation of Ndt80 by Ime2 . In the absence of inhibitor , Ndt80 was inactive . Only a small fraction of cells entered the meiotic divisions ( Fig 5A ) , phospho-Hop1 , phospho-Hed1 and Red1 persisted , and Cdc5 was not detected three hours after induction ( Fig 5B , -1-NA-PP1 ) . The activation state of Ndt80 at the 6-hour timepoint was therefore determined by whether Mek1 was active ( inactive Ndt80 ) or inhibited ( active Ndt80 ) . Phosphatase treatment of inactive Ndt80 resulted in the loss of the slower migrating species , producing two predominant bands ( Fig 5C ) . The band indicated as “pr-Ndt80” has previously been interpreted to be unphosphorylated Ndt80 , while the fastest migrating band ( “Ndt80” in Fig 5C ) was said to be a “degradation fragment” [53 , 63] . Instead , the latter band more likely represents completely unphosphorylated Ndt80 because ( 1 ) it runs close to the molecular weight for unmodified Ndt80 ( 69 kD ) ; ( 2 ) the extracts used for these experiments were fixed with trichloroacetic acid prior to lysis and protease inhibitors were included during lysis , making proteolysis unlikely; and ( 3 ) this band was not observed when phosphatase inhibitors were included in the reactions ( Fig 5C ) [53 , 63] . We propose that “pr-Ndt80” represents phosphorylated Ndt80 that is more refractile to phosphatase treatment than the phosphorylated forms exhibiting slower mobility . The critical point is that unphosphorylated Ndt80 appeared when inactive Ndt80 was treated with phosphatase , indicating the presence of phosphates . Determining whether phosphorylation of inactive Ndt80 is dependent upon MEK1 is difficult because inhibition of Mek1 eliminates the checkpoint by allowing intersister DSB repair [79] , resulting in activated Ndt80 and the Ime2-dependent shift [53] . Instead we tested whether phosphorylation of inactive Ndt80 by Ime2 could be ruled out . This goal was accomplished by phosphatase treatment of extracts from a dmc1Δ NDT80-IN diploid containing an analog sensitive version of IME2 , IME2ΔC241-as . This allele encodes a truncation of the C-terminus of Ime2 that results in stable , constitutively active kinase [80] . Ime2ΔC241-as activity can be abolished using the 3-MB-PPI inhibitor [48 , 81] . Induction of NDT80 in the dmc1Δ IME2ΔC241-as diploid resulted in significantly more meiotic progression than the dmc1Δ mek1-as NDT80-IN diploid , although it was much slower than the isogenic DMC1 diploid , indicating that the checkpoint was active ( Fig 5A and 5D ) . The IME2ΔC241-as allele makes hyperactive Ime2 due to the removal of a C-terminal negative regulatory domain [80] . Constitutively high levels of Ime2 activity may be able to counteract the inhibitory phosphorylation of the induced Ndt80 protein better than the endogenous Ime2 activity in the mek1-as dmc1Δ diploid , resulting in more progression . Addition of the Ime2ΔC241-as inhibitor decreased the amount of meiotic progression , indicating that the inhibitor was working ( Fig 5D ) . In addition , the slowest migrating Ndt80 species disappeared , Hed1 and Hop1 phosphorylation were stabilized and only a very low level of Cdc5 was detected at the 8-hour time point ( Fig 5D and 5E ) . Therefore , the 7-hour time point in the presence of inhibitor represents inactive Ndt80 that lacks Ime2 phosphorylation . Phosphatase treatment resulted in faster migrating bands , demonstrating that checkpoint inactivated Ndt80 contains Ime2-independent phosphates which we propose are mediated by Mek1 ( Fig 5F ) . Crystal structures of the Ndt80 DBD ( amino acids 1–340 or 59–330 ) bound to an MSE show that the Mek1 consensus sites at S205 , T211 , S327 and S329 are juxtaposed to the sugar-phosphate backbone of the DNA ( Fig 6A ) [67 , 68] . ( No structural information is available for S24 ) . Phosphorylation of these sites therefore places negatively charged phosphates in positions where they could repel the negatively charged DNA , thereby preventing DNA binding . This idea was tested using electrophoretic mobility shift assays ( EMSA ) with recombinant Ndt80 DBD ( aa 1–340 ) and 29-mer duplexes containing either the MSE from SPS4 or a non-specific sequence designated as Scr ( Fig 6D ) . Note that these DBDs contain the first five Mek1 consensus sites , but not S343 . The SPS4 MSE was previously used for in vitro DNA binding assays and structural studies [67] . The Ndt80 WT , 5A and 5D DBDs were fused to a six-histidine tag and purified following the protocol of [82] . All three purified proteins exhibited similar yields and elution profiles ( Fig 6B ) ( S1 Fig ) . Differential scanning fluorimetry ( DSF ) was used to determine whether any of the proteins were unfolded . This assay involves incubating proteins with a fluorescent dye and then slowly increasing the temperature to denature the proteins . As the proteins unfold , hydrophobic regions bind the dye , resulting in an increase in fluorescence [83] . The fluorescence values were then normalized to generate melting curves ( Fig 6C ) . Melting temperatures ( Tm ) were calculated as temperatures with fluorescence values midway between the two extremes . The Ndt80 WT , 5A and 5D DBDs exhibited similar melting curves with Tms of 45 . 2°C , 45 . 5°C , and 47°C , respectively . These values indicate that all three proteins were similarly folded , while the 5D protein was even more stable than the WT or 5A protein ( Fig 6C ) . DNA binding was assayed by incubating Ndt80 WT DBD with a Cy3 fluorescently labeled DNA duplex containing an MSE . All of the duplex was bound , resulting in decreased mobility of the fluorescent DNA ( Fig 6D , lane 2 ) . More than 90% of the binding was specific for the MSE sequence , since unlabeled MSE duplex was an effective competitor , decreasing the amount of shifted fluorescent duplex to ~5% ( Fig 6D , lanes 3–5 ) . In contrast , equivalent molar amounts of unlabeled Scr duplex did not compete for binding ( Fig 6D , lanes 7–9 ) . The Ndt80 5A DBD also bound specifically to the MSE duplex , although less efficiently than WT , while no binding was observed for the 5D protein ( Fig 6E , lanes 2–4 ) . Non-specific DNA binding was assessed in a separate reaction using fluorescently labeled Scr duplex . A similar pattern was observed: the WT DBD exhibited the highest level of non-specific binding , followed by the 5A DBD and no binding for the 5D protein ( Fig 6E , lanes , 6–8 ) . We conclude that negative charges on the DBD inhibit Ndt80’s ability to interact even non-specifically with DNA . Having recombinant WT and 5A DBDs in hand allowed us to test whether Mek1 directly phosphorylates the Ndt80 DBD in vitro . Phosphorylation was detected using the semi-synthetic epitope system [84 , 85] . Kinase assays contained active GST-Mek1-as isolated from meiotic yeast cells and the ATP analog , 6-Fufuryl-ATPγS . This ATP analog can fit into the enlarged ATP pocket present in the GST-Mek1-as kinase , but not in the ATP binding pockets of other kinases that may have co-purified with GST-Mek1-as . Phosphorylation by GST-Mek1-as transfers a thio-phosphate onto its substrates which are then chemically alkylated to generate an epitope that is recognized by a commercially available thio-ester antibody . GST-Mek1-as auto-phosphorylation was used as an internal control to show that the kinase reaction worked ( Fig 6F , lane 1 ) [35 , 85] . The Ndt80 WT DBD was phosphorylated by GST-Mek1-as ( Fig 6F , lane 3 ) . Both GST-Mek1-as autophosphorylation and DBD phosphorylation were eliminated by addition of 1-NA-PP1 , confirming that GST-Mek1-as kinase activity was responsible for the signal ( Fig 6F , lane 2 ) . The 5A DBD was also phosphorylated by GST-Mek1-as , but less efficiently ( Fig 6F , lane 9 and 6G ) . Decreasing the amount of kinase reduced 5A phosphorylation more rapidly than WT DBD phosphorylation . We conclude that: ( 1 ) Mek1 phosphorylates at least one of the Ndt80 DBD consensus sites in vitro and ( 2 ) Mek1 can also phosphorylate non-consensus sites within the DBD . It has been known for several years that the meiotic recombination checkpoint in yeast requires MEK1 and that a key target of the checkpoint was Ndt80 , but how the two were connected was unclear . The simplest idea , that Mek1 inactivates Ndt80 by directly phosphorylating it , was not considered for two reasons . First , Ndt80 phosphorylation was proposed to promote , not inhibit , Ndt80 activity [48 , 63 , 64] . Second , deletion of MEK1 has no effect on the mobility shift of Ndt80 , leading to the conclusion that Ndt80 is not a substrate of Mek1 [53] . The latter result is misleading , however , because absence of MEK1 results in efficient DSB repair using sister chromatids and therefore removes the signal to the checkpoint . As a result , Ndt80 is activated and phosphorylated by Ime2 . Therefore , it is impossible to determine whether Ndt80 is phosphorylated by Mek1 under checkpoint arrested conditions simply by comparing Ndt80 mobility shifts in diploids with or without Mek1 activity . Using a combination of different approaches , we have demonstrated that Mek1 phosphorylation of Ndt80 is responsible for the meiotic recombination checkpoint delay/arrest . First , Ndt80 is phosphorylated when it is inactivated by the meiotic recombination checkpoint and this phosphorylation is independent of IME2 . Second , Ndt80 contains ten Mek1 consensus phosphorylation sites , eight of which are located either within the DNA binding domain or the “middle region” that is required to inhibit Ndt80 in response to the checkpoint . Preventing phosphorylation at these sites using alanine mutations results in partial bypass of the checkpoint triggered by unrepaired DSBs in the dmc1Δ background . Third , Ndt80 contains a conserved five amino acid sequence within the middle region that is required both for checkpoint arrest and for interaction with Mek1 . Mutation of this site results in checkpoint bypass without directly affecting Mek1 kinase activity . Fourth , substitution of negatively charged amino acids at Mek1 consensus sites within the Ndt80 DBD constitutively inactivates the transcription factor . Several of these putative Mek1 sites are located immediately adjacent to the negatively charged DNA sugar-phosphate backbone of the MSE . Recombinant Ndt80 DBD containing negative charges at these sites does not bind DNA , even non-specifically . These observations suggest that phosphorylation of the DBD by Mek1 prevents Ndt80 from binding to MSEs and explains how Mek1 phosphorylation can inhibit Ndt80 activity . Finally , Mek1 directly phosphorylates at least one of the Mek1 consensus sites in the Ndt80 DBD in vitro , with less efficient phosphorylation of at least one non-consensus amino acid that has not yet been identified . While disrupting DNA binding may be the major mechanism by which Ndt80 is inactivated by Mek1 , it is unlikely to be the only one . Mek1 phosphorylation of Ndt80 at multiple sites appears to inhibit Ndt80 activity in additional ways because preventing phosphorylation of all of the Mek1 consensus sites in the Ndt80 DBD only weakly bypassed the dmc1Δ checkpoint arrest . This bypass was increased when the DBD alanine substitutions were combined with alanine mutations within the bc domain ( ndt80-10AMS ) . Furthermore , the ndt80-10AMS checkpoint bypass was less efficient than that observed for the deletion of the RPSKR sequence within the bc domain . Since RPSKR is necessary for Mek1 interaction in the two-hybrid system , we propose that deletion of RPSKR disrupts the Mek1-Ndt80 interaction in meiotic cells , preventing any Mek1 phosphorylation of Ndt80 from occurring . In contrast , Mek1 can bind to the Ndt80-10AMS protein via the RPSKR motif and may then phosphorylate non-consensus sites within the DBD and/or the middle region . Ndt80 that is inactivated by the meiotic recombination checkpoint preferentially localizes to the cytoplasm [69] . It has been proposed that this localization is due to a checkpoint activated cytoplasmic tether but how this tether would work is not clear . An alternative possibility is that Ndt80 constantly shuttles in and out of the nucleus and only when it binds to DNA does Ndt80 remain stably inside the nucleus . Ndt80 is larger than 40 kD , meaning that it is too big to diffuse freely through nuclear pores and must be actively transported [86] . Therefore cytoplasmic localization of inactive Ndt80 could also occur if phosphorylation either promotes nuclear export or inhibits nuclear import . Finally , Mek1 phosphorylation of Ndt80 could inhibit Ime2 phosphorylation at different sites on the transcription factor . The following model describes how meiotic gene transcription is integrated with meiotic chromosome structure and DSB repair to promote entry into the meiotic divisions only after DSB repair is complete . In vegetative cells , homologs are not associated and transcription of NDT80 and its target genes are repressed by Sum1 bound to MSEs [59] ( Fig 7A , 7D and 7G ) . Early meiotic gene expression is prevented by the Ume6 repressor complex bound to a specific Upstream Repression Sequence called URS1 [90 , 91] ( Fig 7D ) . Transfer to Spo medium results in the removal of Ume6 and binding of the Ime1 transcriptional activator at URS1 sites , resulting in expression of early genes such as REC8 , HOP1 , RED1 , MEK1 and SPO11 [92 , 93] ( Fig 7E ) . These gene products ( and others ) function to assemble AEs , make DSBs and activate Mek1 ( Fig 7B ) . Early in meiosis , when DSBs are first occurring , they are repaired primarily using sister chromatids , indicating that Mek1 activity has not reached the threshold necessary to impose interhomolog bias [94] . Similarly , a threshold amount of Mek1 is necessary to inactivate Ndt80 . The cell provides time for Mek1 activation by delaying Ime1-driven NDT80 transcription through the additional step of removing the Sum1 repressor . This removal requires that Sum1 be phosphorylated by Ime2 , along with CDK and DDK ( Fig 7E ) [57 , 60–62] . Since IME2 is an early gene , it must be transcribed and translated after induction of meiosis . By the time that Ime1-driven transcription of NDT80 occurs , there is sufficient activated Mek1 ( Fig 7B , yellow stars ) to phosphorylate Ndt80 ( Fig 7H , red stars ) , thereby preventing Ndt80 from binding to DNA ( Fig 7H ) . Mek1 phosphorylation somehow interferes with Ime2 phosphorylation of Ndt80 , which also contributes to keeping the transcription factor inactive . Deletion of SUM1 from dmc1Δ diploids results in bypass of the meiotic recombination checkpoint [72] . We propose that when Ndt80 is prematurely expressed in the sum1Δ , there is not enough time to make DSBs and activate Mek1 , thereby allowing Ndt80 to activate transcription of its own gene and start the positive feedback loop . A stable interaction between Mek1 and Ndt80 may be necessary to ensure that the kinase is able to counteract removal of phosphates by a phosphatase such as Glc7 , which has a role in promoting pachytene exit [95] . As DSBs are processed into double Holliday junctions , their repair promotes chromosome synapsis , resulting in the elimination of most of the Mek1 from chromosomes and a reduction in overall Mek1 kinase activity ( Fig 7C ) [41 , 96] . Without sufficient Mek1 activity , the phosphatase wins out and removes the Mek1-dependent phosphorylation . As a result , Ndt80 becomes activated ( which is enhanced by phosphorylation due to Ime2 ) , binds to an MSE in its own promoter to become stably localized to the nucleus and activates a second wave of NDT80 transcription in a positive feedback loop ( Fig 7F and 7I ) [48 , 57 , 63 , 64] . In addition , Ndt80 target genes are expressed , including CDC5 and CLB1 ( Fig 7F ) . Cdc5 promotes Holliday junction resolution into COs , degradation of Red1 and SC disassembly , thereby eliminating any remaining Mek1 activity [41 , 54 , 65 , 75] . As a result , Rad51 can bind to Rad54 and repair any remaining DSBs prior to entry into MI [41] . Finally , prophase exit resulting from Ndt80-mediated transcription shuts down Spo11 so that no further DSBs are made [97 , 98] . While the Ndt80 protein is not conserved outside of fungi , the structure of the DBD is conserved . Ndt80 is a member of the Ig-fold family of transcription factors that includes p53 , RUNX , NFAT and NF-kb from mammals [67 , 68 , 77 , 99] . This domain contains a series of loops extending out from several β-sheets that contact DNA to mediate site-specific binding [99] . Interestingly , Ndt80 has more extensive contacts with DNA than other Ig-fold transcription factors , perhaps because Ndt80 binds DNA as a monomer , in contrast to other the proteins which bind as dimers [67 , 77] . A common feature of Ig-fold transcription factors is the inhibition of DNA binding by phosphorylation , leading to cytoplasmic localization . For example , NFAT is required for the transcription of cytokine genes involved in T cell activation . In unstimulated cells , phosphorylation of the NFAT nuclear localization signal ( NLS ) blocks import of the protein into the nucleus [100] . Stimulation of a human T cell lymphoma cell line with phorbol ester activates a phosphatase that removes the phosphorylation at the NLS , allowing translocation into the nucleus where NFAT binds to a specific DNA sequence in its target genes [101] . This binding is inhibited by Cyclosporin A , which results in phosphorylation of the NFAT DBD and cytoplasmic localization of the protein [101] . In another example , phosphorylation of threonine 173 in the DBD of Runx3 by Aurora kinase prevents DNA binding [102] . Similar to the Ndt80 phosphosites , T173 is present at the Runx3-DNA interface . Dissociation from the DNA results in relocalization of Runx3 to the cytoplasm and centrosome during early mitosis . Finally , the p53 protein is a transcription factor that functions in tumor suppression by transcribing genes that promote cell cycle arrest and apoptosis in response to DNA damage [103] . Aurora-A phosphorylates serine 215 in the p53 DBD in vivo . A p53-S215D , but not S215A , mutant prevents DNA binding , resulting in down regulation of target genes necessary for tumor suppression [104] . Although these transcription factors and Ndt80 regulate vastly different processes , there is clearly conservation of the regulatory mechanism that controls them . Many components of the meiotic recombination checkpoint are conserved between yeast and mammals , even though mammalian meiosis is more complicated than yeast , due to the presence of the X and Y sex chromosomes in males and the dictyate arrest that occurs in oocytes after pachytene exit [10] . DSB-dependent checkpoints have been observed in both mouse oocytes and spermatocytes [105–108] . To study the role of the meiotic recombination checkpoint in mice , a hypomorphic allele of the Trip13 gene called Trip13mod has been used . This mutant has the advantage that chromosomes synapse but many DSBs remain unrepaired , thereby eliminating signals that might arise from a synapsis checkpoint . Trip13mod triggers a DSB-dependent arrest in early pachynema that can be distinguished from later arrest points by the absence of a testis-specific histone variant called H1t . Using this assay , the DSB-dependent arrest has been shown to be dependent on Atm , Chk2 , and HORMAD1/2 , similar to the requirements for the orthologous yeast genes , TEL1 , MEK1 and HOP1 , respectively in the meiotic recombination checkpoint [10] . A key target of the mammalian checkpoint is p53 and its paralog , TAp63 . Deletion of p53 or TAp63 in Trip13mod/mod mice allows both oocytes and spermatocytes to progress beyond the early pachytene arrest [105 , 106] . Using radiation induced DSBs in oocytes , Bolcun-Filas et al ( 2104 ) [105] showed that TAp63 is phosphorylated in a Chk2-dependent manner that requires the Chk2 consensus phosphorylation site ( LXRXXS ) [109] . The mammalian checkpoint response therefore resembles that of yeast: DSBs indirectly activate an FHA-domain containing effector kinase , Chk2 or Mek1 , in the context of the AE to regulate an Ig-fold transcription factor , p53/TAp53 or Ndt80 , thereby creating an arrest . The main difference is that in mice the checkpoint activates the p53/TAp63 transcription factors while in yeast phosphorylation of Ndt80 prevents transcription . S1 Table contains a list of plasmids used in this work with the relevant yeast genotypes . S2 Table lists oligonucleotides and their sequences that were used to construct plasmids . Relevant genes in all of the plasmids were sequenced in their entirety by the Stony Brook University DNA Sequencing Facility to ensure that no unexpected mutations were present . Site directed mutagenesis of NDT80 was carried out using the URA3 NDT80 integrating plasmid , pHL8 [61] and the protocol in the Quikchange kit ( Stratagene ) to generate the 2A , 4AMS , 5AMS , 7AMS , 9AMS , 10AMS 10DMS , 6N , S24D , S343D , S205D T211D , S327D S329D , K374A R375A and K374D R375D mutations . The ndt80-6A , 6D and 8D alleles , in pNH400 , pNH401 and pNH405 , respectively , were constructed using three fragment Gibson Assembly ( GA ) reactions ( New England BioLabs ) . One fragment was pRS306 digested with EcoRI and ClaI [110] . The second fragment was 3 . 3 kb , with overlapping homology with the EcoRI side of the vector and the NDT80 gene between codons 379 and 385 . It was amplified using the polymerase chain reaction ( PCR ) with the primers NDT80-WT-EcoRI-F1 and NDT80-R-385 . The third fragment was 1 . 1 kb and contained NDT80 sequence between codons 379 and 385 and overlapping homology with the ClaI side of the vector . Amplification of this fragment used primers NDT80-WT-ClaI-R1 and NDT80-F-379 . For pNH400 and pNH405 , pHL8-10AMS was used as the template for the 3 . 3 kb fragment containing the S24A , S205A , T211A , S327A , S329A and S343A mutations . For pNH401 , pHL8-10DMS was the template for the 3 . 3 kb fragment containing the S24D , S205D , T211D , S327D , S329D and S343D mutations . The 1 . 1 kb fragment for pNH400 and pNH401 was amplified from the NDT80 gene in pHL8 , while pHL8-2A ( T399A T420A ) was the template for 1 . 1kb fragment used to make pNH405 . The three fragment GA reaction used to make NDT80-Δbc ( pHL8-Δbc ) used pHL8 to generate two fragments , one using primers NDT80-WT-EcoRI-F1/NDT80-bc-Cla-R1 and the other using NDT80-bc-Cla-F2/NDT80-WT-ClaI-R1 , which were then assembled into EcoRI/ClaI digested pRS306 . Estradiol inducible alleles of NDT80 ( NDT80-IN ) were created using three fragment GA reactions . One fragment was EcoRI/ClaI digested pRS306 . The second fragment , containing the GAL1 promoter with homology on one end to sequences flanking the EcoRI site of pRS306 , was amplified using pFA6a-HIS3MX6-PGAL1-GFP as the template and PGAL1-EcoRI-F1 and PGAL1-R1 as primers . The third fragment contained the NDT80 open reading frame ( ORF ) and 3’ flanking sequence . One end had homology to the 3’ end of the GAL1 promoter and the other end to sequences flanking the ClaI site in pRS306 . For pBG4 , this fragment was amplified using pHL8 as template and the primers , NDT80-ORF-GAL1-F1 and NDT80-WT-ClaI-R1 . For pXC11 and pXC12 , the template for Fragment 3 was pNH400 and pNH401 , respectively . The E . coli expression plasmids , pNH407-WT , -5A , and -5D were also constructed using GA . The plasmids contain the NDT80 DBD ( codons 1–340 ) followed by a stop codon , fused in frame to six histidines in the pET-28a vector ( Novagen ) . 1 . 1 kb fragments containing the DBD with overlapping homology flanking the Nde1 and XhoI sites of pET-28a were amplified using either pHL8 ( WT ) , pHL8-10AMS ( 5A ) or pHL8-10DMS ( 5D ) and the primers pET28a-NDT80-F and pET28-NDT80-340-R . These fragments were then incubated with pET-28a digested with NdeI and XhoI and the GA reagent . The lexA-MEK1 plasmid , pTS3 , was constructed using PCR and the primers MEK1-lexA-5/MEK1-lexA-3 to amplify a 1 . 5 kb fragment containing MEK1 with BamHI sites engineered onto either end . This fragment was ligated into BamHI-digested pSTT91 , resulting in an in-frame fusion of the MEK1 ORF with lexA . The R51A mutation in the FHA domain was introduced into pTS3 by site-directed mutagenesis to make pTS3-R51A [47] . The GAD-ndt80284-627 fusion ( plasmid A32 ) was isolated from a two-hybrid screen using lexA-MEK1 as bait . This allele was then re-created de novo in pXC13 using GA so that direct comparison to various deletion alleles could be made . All ndt80 sequences were fused in-frame with GAD and had the same transcriptional terminator and NDT80 3’ untranslated region ( UTR ) . For pXC13 , PCR was used to amplify a fragment containing the GAD-ndt80 fusion from A32 using the primers , NDT80-GAD-F/NDT80-GAD-R . The resulting 1 . 8 kb fragment was then cloned into pACTII digested with NcoI and XhoI . The K374A R375A and K374D R375D mutations were separately introduced into pXC13 by site-directed mutagenesis to make pXC13-KR>AA and pXC13-KR>DD , respectively . The GAD-ndt80-Δbc allele contains an internal , in-frame deletion of the 57 codons of the “bc” domain and was created using a three fragment GA reaction . The first fragment contained GAD fused to NDT80 codons 284–345 and was generated using the primers , NDT80-GAD-F/NDT80-bc-Cla-R1 . The second fragment contained NDT80 codons 403–627 along with overlap with the 3’ end of fragment 1 . In this case the primers were NDT80-bc-Cla-F2/NDT80-GAD-R . These fragments were then cloned into NcoI/XhoI digested pACTII to generate pXC14 . The pXC18 plasmid contains the 57 amino acid “bc” domain directly fused to GAD . Fragment 1 was generated using NDT80-GAD-bc-F3 and NDT80-N1-R1 . This fragment was reacted with the NDT80 3’UTR fragment and NcoI/Xho1-digested pACTII to make GAD-bc . The RPSKR sequence was deleted from GAD-ndt80 to make GAD-ndt80-ΔRPSKR in the following way . Fragment 1 was created using NDT80-GAD-F and NDT80-370-R as primers and pXC13 as template to generate a fragment with homology on one end to the 3’ end of GAD and on the other end to NDT80 ending at codon 370 . The second fragment was amplified using NDT80-RPSKRΔ-F and NDT80-GAD-R with the pXC13 template . This fragment overlaps on one end with NDT80 codons 360–370 , then deletes codons 371–375 and continues to end of NDT80 and homology to the XhoI digested end of pACTII . These two fragments were joined with NcoI/XhoI-digested pACTII by GA to make pNH318 . A similar strategy was used to delete the RPSKR codons from the NDT80 ORF to make pNH317 . The three fragment GA reaction consisted of ( 1 ) EcoRI/ClaI digested pRS306; ( 2 ) a fragment amplified from pHL8 using NDT80-WT-EcoRI-F1 and NDT80-370-R and ( 3 ) a fragment amplified from pHL8 using NDT80-RPSKRΔ-F and NDT80-WT-Cla-RI . To put GAD-bc under control of the MEK1 promoter , a 1 . 2 kb fragment containing GAD-bc was amplifed using YEp-GADbc-F and YEp-GADbc-R with pXC18 as the template . This fragment has one end homologous to the MEK1 promoter and the other end homologous to sequences downstream of the NdeI site in pDW14 . GA was used to introduce the GAD-bc fragment into Nde1-digested pDW14 to make pLB1 . All strains were derived from the SK1 background unless otherwise noted and their genotypes are listed in S3 Table . Liquid and solid media used for growing cells vegetatively or for sporulation are described in [85] . PCR-based methods were used to delete genes with the drug resistance markers , kanMX6 , natMX4 and hphMX4 [111–113] . All deletions were confirmed by PCR . The presence of the deletion allele was confirmed using a forward primer upstream of the ORF and a reverse primer in the drug resistance gene . The absence of the WT allele was also tested using the same forward primer with a reverse primer internal to the gene’s ORF . To make NH2081 , NDT80 was deleted from the haploid parents of NH144 , which were then mated . The second exon of DMC1 was then deleted from the ndt80Δ::hphMX4 haploids and mated to make NH2402 . The NH144 diploid is heteroallelic for leu2 , making it impractical to transform with LEU2 plasmids since transformants cannot be distinguished from mitotic recombinants . The LEU2 gene was therefore deleted in one of the NH144 parents and then mated to the other to make the leu2ΔhisG/leu2Δ::kanMX6 diploid , NH2444 . The NH2426:pEP1052::pX2 ( where the “2” indicates a homozygous plasmid ) series of diploids was constructed by first deleting NDT80 with hphMX4 from SKY370 and SKY371 . The TRP1 GAL4-ER integrating plasmid , pEP105 , was digested with Nhe1 and integrated at the trp1::hisG locus in the resulting haploids [41] . URA3-integrating plasmids containing different alleles of NDT80 were then digested with NsiI to target integration to ura3 . Integration of the plasmids was confirmed by PCR . The haploids were mated to form homozygous diploids . The phosphatase experiments were performed NH2437::pEP1052::pBG42 , which was derived from the haploid parents of NH2092 [114] . These haploids , NH2091-2-4::pJR2 and NH2091-8-2::pJR2 contain pJR2 , a mek1-as URA3 plasmid integrated just upstream of mek1Δ::kanMX6 [114 , 115] . Cells that lost the pJR2 plasmid were selected for using 5-fluororotic acid ( 5-FOA ) [116] . To determine whether the mek1-as or mek1Δ::kanMX6 allele remained in the chromosome , FOAR colonies were screened for sensitivity to G418 . The first 222 bp of the TRP1 gene were then deleted from the resulting mek1-as haploids using natMX4 [39] and NDT80 was deleted using kanMX6 . The GAL4-ER fusion was integrated into the 3’end of TRP1 using NheI-digested pEP105 . The PGAL1-NDT80 plasmid , pBG4 , was integrated at the ura3 locus using NsiI . The haploids were mated to make NH2437::pEP1052::pBG42 . NH2451 was created by deleting the second exon of DMC1 with kanMX6 from the haploid parents of yLJ92 and mating to make the diploid . Two-hybrid experiments were carried out using the L40 strain that contains lexA operator sequences upstream of the HIS3 and lacZ genes [66] . HIS3 expression was assayed on selective medium ( SD-leu–trp–his ) , while lacZ was assessed using a colorimetric enzyme assay that produces blue color when ß-galactosidase is present [117] . For the two-hybrid screen , L40 containing pTS3 ( 2μ lexA-MEK1 TRP1 ) was transformed with a genomic 2μ LEU2 GAD fusion library [118] and 1 . 1 X 106 transformants were screened for growth on SD -leu , -trp , -his medium . Fifteen His+ transformants also expressed lacZ . The GAD plasmids were isolated from the transformants and the fusion junctions sequenced using GAD-AD-5’ . One of these transformants contained GAD fused in-frame to codons specifying amino acids 284–627 of NDT80 . Transformants containing different GAD plasmids and lexA-MEK1 were grown overnight at 30°C on a roller in SD-Leu-Trp . The cells were diluted 1:10 in water and the optical density at 660 nm ( OD660 ) was determined using a spectrophotometer . Culture volumes equivalent to two ODs were pelleted in microfuge tubes and resuspended in 100 μl sterile water . The cells were transferred to a 96-well plate and ten-fold serial dilutions were made . Ten μl cells were spotted onto SD-Leu-Trp and SD-Leu-Trp-His plates . In addition , four μl of each dilution were plated on a paper filter placed onto an SD-Leu-Trp plate . After growth overnight at 30°C , ß-galactosidase assays were performed . The remaining plates were incubated for three days prior to being photographed . Both the 6xHis-Ndt80-WT-DBD ( called WT DBD ) , 6xHis-Ndt80-DBD-5D ( called 5D DBD ) and 6xHis-Ndt80-DBD-5A ( called 5A DBD ) proteins were purified from two different 250 ml cell pellets and used for DNA binding assays . Similar results were obtained with both protein preparations . The protein purification protocol was based on the one described in [82] . The E . coli expression plasmids , pNH407-WT , -5D and 5A were each transformed into BL21 ( DE3 ) Codon Plus RIL bacterial cells ( Agilent Genomics ) . Transformants were selected on LB + kanamycin ( 50 μg/ml ) and chloramphenicol ( 30 μg/ml ) ( LB +KC ) plates . For each plasmid , multiple transformants from a single plate were scraped together and used to inoculate 10 ml LB +KC liquid medium . The cultures were incubated with shaking at 37°C overnight . The next day each culture was diluted to an OD600 of 0 . 02 in 1 L LB + KC in a 4 L flask and the cultures were grown with shaking at 37°C to an OD600 of 0 . 4 . Imidazole was added to a final concentration of 1 mM to induce transcription of the tagged 6xHis-ndt80 DBD alleles and the cells remained shaking at 37°C for 4 hours . Each liter of culture was divided into 250 ml aliquots and the cells were pelleted by centrifugation , resuspended in wash buffer ( 50 mM Tris-HCl , pH 7 . 5 , 100 mM NaCl and 1 mM EDTA ) , transferred to 50 ml conical tubes and pelleted again . The supernatants were discarded and the cell pellets stored at -20°C . To lyse the cells , 250 ml cell pellets were thawed on ice and resuspended in 17 . 5 ml Vershon Lysis Buffer ( VLB ) ( 50 mM NaH2PO4/Na2HPO4 , pH 7 . 8 , 1 M NaCl ) containing 5 mM imidazole and 0 . 2 mM phenylmethylsulfonyl fluoride ( PMSF ) . The imidazole was made fresh and the PMSF added immediately before sonication . Cell suspensions were transferred to pre-chilled 50 mL glass beakers on ice and the cells lysed by sonication using a Qsonica Q500 ultrasonic processor with a 12 . 7 mm probe ( 6 pulses of 15 sec , with 30 sec rests ) at 70% power . The lysates were then transferred to pre-chilled polyallomer Beckman centrifuge tubes ( 25 X 80 mm ) and centrifuged in a JA-25 . 50 rotor at 19 , 647 X g for 30 min at 4°C . The cleared lysates were then transferred to 15 ml conical tubes , flash frozen in liquid nitrogen and stored at -80°C . To purify the recombinant proteins , the lysates were thawed on ice , distributed between microfuge tubes and spun at 13 , 000 X g in a microfuge for 10 min to remove any precipitated material . Lysates were pooled and loaded onto a column containing 0 . 5 ml bed volume of Ni-NTA Superflow agarose beads ( Qiagen ) equilibrated in VLB+ 5 mM imidazole . Protein bound beads were washed twice with 2 . 5 ml VLB+ 5 mM imidazole and then subjected to increasing concentrations of imidazole in the following steps: 10 mM , 50 mM , 100 mM , 200 mM and 250 mM . All of the DBDs eluted with the 50 mM and 100 mM steps ( S1 Fig ) . The second 2 . 5 ml 50 mM imidazole fraction was mixed with the first 2 . 5 ml 100 mM imidazole fraction and the proteins were concentrated by centrifugation using Amicon Ultra filters ( UFC50124 ) . The molar concentrations of the proteins were determined based on the OD280 absorbance measured by a NanoDrop spectrophotometer ( Thermo Scientific ) and the calculated molecular weight of the DBD which is 40 , 169 g/mole . To visualize the proteins , an appropriate volume of 5 X protein sample buffer was added to each sample and the samples were heated at 95°C for 5 min . Proteins were then fractionated on 12 . 0% SDS-polyacrylamide gels ( 1 . 0 mm spacers ) , using 250 volts for 25 minutes , and then stained with GelCode Blue ( Thermo Scientific ) . DSF was performed on an Applied Biosystems 7500 Fast Real-Time PCR System using the protocol outlined in the “Protein Thermal Shift Studies” User Guide ( Applied Biosystems ) with minor modifications . DSF experiments used purified Nt80 DBDs at a final concentration of 5 μM in 96-well PCR plates for fast thermocyclers ( VWR , Cat . No . 892180296 ) . Each well ( 50 μL ) contained SyproOrange Dye ( Sigma Aldrich ) diluted to a final concentration of 5x . The plate temperature was ramped from 25°C to 95°C with a linear gradient ( 1% ramp rate ) . The fluorescence of the SyproOrange Dye was detected by selecting ROX as the reporter ( filter 4 , emission range between 600–625 nm ) . The fluorescence values were normalized to a range between 0 . 0 and 1 . 0 using the equation Ynormalized = ( Yraw data−Ymin ) / ( Ymax−Ymin ) , where Ymin and Ymax refer to the minimum and maximum values of fluorescence , respectively . Tm values were calculated using the Boltzmann sigmoidal equation in the program GraphPad Prism 4 . The Boltzmann sigmoid equation is Yfluorescence = Bottom + ( Top—Bottom ) / [1 + exp ( ( V50—Xtemp ) / Slope ) ] . V50 refers to the temperature at which fluorescence is halfway between the bottom and top fluorescence values . Tm values are equal to the calculated V50 value . Fluorescently labeled 29-mer oligonucleotides ( oligos ) containing either the SPS4 MSE sequence ( 5’ Cy3-ATTGACGCGCGCCACAAAAACGTATCATT ) or the Scr sequence ( 5’ Cy5-ATTGACGCGGCTTCATCTCACGTATCATT ) ( indicated in bold , respectively ) were synthesized by Integrated DNA Technologies with high performance liquid chromatography ( HPLC ) -grade purity . Unlabeled complementary strands were ordered through the Stony Brook Oligonucleotide Facility . Oligos were resuspended at a concentration of 100 μM in water . Complementary strands were annealed by combining equal amounts of each oligo , adding NaCl to a final concentration of 100 mM , incubating the oligos at 95°C for 5 min and then turning off the hot block to allow the strands to slowly anneal overnight . The resulting Cy3- and Cy5-labeled duplex molecules were purified by size exclusion chromatography using a Superdex 200 Increase 10/300 GL column in eluent A ( 10 mM Tris-HCl , pH 7 . 5 , 1 mM EDTA , 0 . 01% NP-40 substitute , 50 mM NaCl ) at a flow rate of 0 . 5 mL/min . The molar concentration of DNA in the peak fractions was quantified based on ultraviolet absorbance at 260 nm using a NanoDrop spectrophotometer . To visualize the DNA , the appropriate volume of 6 X sucrose buffer ( 7 . 2 g sucrose in 10 . 2 mL 1x TE pH 7 . 8 ) was added to each sample , and the DNA was resolved on a 6% polyacrylamide gel ( 1 . 5 mm thickness ) in 0 . 5 X TBE at 110V for 45 minutes . In-gel Cy3 or Cy5 fluorescence was detected by a Typhoon 9500 scanner ( GE Healthcare ) . Unlabeled duplexes were constructed similarly , except the Cy3 and Cy5 oligos were replaced with unlabeled oligos and the DNA was visualized by in-gel Sybr Gold staining ( Thermo Scientific ) . Protein dilution buffer ( 20 mM Tris-HCl , pH 8 , 50 mM NaCl , 1 mM EDTA , 1 mg/ml bovine serum albumin ( BSA ) , 5 mM 2-mercaptoethanol ) and EMSA reaction buffer ( 10 mM Tris-HCl pH 7 . 5 , 40 mM NaCl , 4 mM MgCl2 , 6% ( w/v ) glycerol , 10 mg/ml BSA , 10 μg/ml sonicated salmon sperm DNA ) were taken from [82] . DNA binding assays were carried out in 20 μl reactions . Reactions were started by addition of the DBD and were incubated at room temperature for 30 min . Reactions and gels were covered with aluminum foil to minimize exposure of the fluorescently labeled to DNA to light . Competition experiments contained 10 nM ( 1X ) Cy3-MSE and 50 nM WT DBD . Unlabeled MSE or Scr duplex 29-mers were added at 2 . 5X , 10X or 40X the amount of the labeled duplex . Four μl 6 X sucrose buffer were added to each reaction , which were then immediately loaded onto a 6% 0 . 5 X TBE gel . The DNA was visualized as described above . Specific and . non-specific DNA binding for the different DBDs were compared using 50 nM of each DBD and 50 nM Cy3-MSE or Cy5-Scr . Protein extracts were generated using the tri-chloroacetic acid method described in [119] . A list of primary and secondary antibodies , sources and dilutions can be found in S4 Table . Calf intestinal alkaline phosphatase ( AP ) treatment of TCA extracts was based on a protocol described in [120] with the following modifications . Sixty μL of extract in 100 mM Tris-HCl , pH 6 . 8 , 4% SDS , 200 mM dithiothreitol ( DTT ) and 20% glycerol were diluted with 408 μL PMP buffer ( 50 mM HEPES , pH 7 . 5 , 100 mM NaCL , 2 mM DTT and 0 . 01% Brij 35 ) . One PhosSTOP tablet ( Sigma , Cat . #4906845001 ) containing phosphatase inhibitors was dissolved in 0 . 5 ml PMP buffer . For each extract , 4 μL AP ( 80 units ) ( Sigma , 11097075001 ) were added to 40 μL PMP buffer ( AP alone ) , as well as 40 μL PMP buffer plus PhosSTOP inhibitors ( AP + Inhibitors ) and incubated at room temperature for 30 min . This preincubation step was necessary to get more complete inhibition of the AP . Equal amounts of the diluted extracts ( 156 μL ) were aliquoted into separate microfuge tubes: ( 1 ) no AP , ( 2 ) AP , and ( 3 ) AP plus phosphatase inhibitors . To the “no AP” tube 1 , 40 μL PMP and 4 μL AP buffer [25 mM Tris-HCl , pH 7 . 5 , 1 mM MgCl2 , 0 . 1 mM ZnCl2 , 50% glycerol ( v/v ) ] were added; 44 μL AP in PMP buffer was added to tube 2 and 44 μL AP in PMP buffer plus inhibitors was added to tube 3 . The final reactions therefore contained 10-fold less protein than the TCA extracts . The reactions were incubated at 30°C for two hours and then stopped by the addition of 5 X Protein sample buffer . The proteins were fractionated on a 7 . 5% SDS-polyacrylamide gel , transferred to a filter and probed with α-Ndt80 antibodies .	Sexual reproduction requires that cells deliberately introduce large numbers of double strand breaks into their chromosomes . Repair of these breaks creates physical connections between homologs that promote proper segregation during meiosis . It is critical that segregation not proceed until all the breaks have been fixed . How does the cell determine when sufficient double strand break repair has occurred ? Our work provides a mechanistic explanation to this question . The meiosis-specific Mek1 kinase is activated by double strand breaks . High numbers of breaks result in high Mek1 activity , resulting in phosphorylation of the meiosis-specific Ndt80 transcription factor . Negative charges conferred by phosphorylation prevent Ndt80 from binding the promoters of its target genes , including genes necessary for completing recombination and meiotic progression , thereby preventing their transcription . As breaks are repaired , Mek1 kinase activity decreases and the inhibitory phosphorylation on Ndt80 is lost , allowing Ndt80 to activate transcription of its target genes . As a result , crossover formation is completed and intact chromosomes proceed properly through the meiotic divisions .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "phosphorylation", "meiosis", "gene", "regulation", "cell", "cycle", "and", "cell", "division", "regulatory", "proteins", "cell", "processes", "enzymes", "dna-binding", "proteins", "enzymology", "chromatids", "dna", "transcription", "phosphatases", "transcription", "factors", "dna", "homologous", "recombination", "chromosome", "biology", "proteins", "gene", "expression", "biochemistry", "cell", "biology", "post-translational", "modification", "nucleic", "acids", "genetics", "biology", "and", "life", "sciences", "dna", "recombination", "chromosomes" ]	2018	Mek1 coordinates meiotic progression with DNA break repair by directly phosphorylating and inhibiting the yeast pachytene exit regulator Ndt80
Because there is considerable variation in gene expression even between closely related species , it is clear that gene regulatory mechanisms evolve relatively rapidly . Because primary sequence conservation is an unreliable proxy for functional conservation of cis-regulatory elements , their assessment must be carried out in vivo . We conducted a survey of cis-regulatory conservation between C . elegans and closely related species C . briggsae , C . remanei , C . brenneri , and C . japonica . We tested enhancers of eight genes from these species by introducing them into C . elegans and analyzing the expression patterns they drove . Our results support several notable conclusions . Most exogenous cis elements direct expression in the same cells as their C . elegans orthologs , confirming gross conservation of regulatory mechanisms . However , the majority of exogenous elements , when placed in C . elegans , also directed expression in cells outside endogenous patterns , suggesting functional divergence . Recurrent ectopic expression of different promoters in the same C . elegans cells may reflect biases in the directions in which expression patterns can evolve due to shared regulatory logic of coexpressed genes . The fact that , despite differences between individual genes , several patterns repeatedly emerged from our survey , encourages us to think that general rules governing regulatory evolution may exist and be discoverable . A complex network of molecular interactions that orchestrates gene expression provides multiple sources for regulatory variation between species [1] . Changes in transcriptional regulation can occur in two fundamentally different ways: in trans regulators [2] , [3] , for example through changes in protein sequences or expression patterns of transcription factors , or in cis elements via changes in identity or location of transcription factor binding sites [4] , [5] . Although the importance of variation in gene regulation for evolution is well appreciated [6]–[8] , many details remain to be elucidated . For example , do mutations in cis arise and go to fixation more frequently than changes in trans [9] , [10] ? Are regulatory mutations pleiotropic and , if so , what are their effects [11] ? Our research has focused on cis-regulatory elements ( CREs ) . These sequences consist of multiple transcription factor binding sites and a core promoter , but these motifs tend to be short , diffuse , and flexible in their locations [12] . Traditional sequence alignments may not therefore be reliable indicators of functional conservation [13] . Because cis elements integrate signals from multiple trans-acting factors in the context of an intact cell , their functions have to be assessed in vivo [14] . The study of functional evolution of cis-regulatory elements has relied on two approaches . One typically starts with the knowledge of the location of binding sites in a regulatory sequence of one species and is followed up by the functional tests of these binding sites in the original and other species [15] , [16] . This approach is labor-intensive and is more difficult to scale . An alternative consists of assessing the functions of orthologous regulatory sequences , without detailed knowledge of identity and location of binding sites , from multiple species in the same trans-regulatory environment ( reviewed in [17] ) . This approach has the advantage of being applicable to less well-studied regulatory regions and can be scaled up to multiple genes , allowing researchers to infer general rules of regulatory sequence evolution . Because they often use different methodologies and criteria for comparisons , studies that investigate the regulatory evolution of individual genes are not easily comparable . It has therefore been difficult to generalize results and infer common features of cis-regulatory evolution . Still , several trends are evident . Multiple studies documented divergence [18]–[22] and constraint [23]–[25] in cis-regulatory mechanisms between species . While functionally equivalent enhancers in different species are often found in similar locations [26] , [27] , this is not always the case [28]–[30] . In some cases , differences in cis-regulatory mechanisms reflect divergence in endogenous expression patterns [30] , [31] . In others , divergent regulatory mechanisms underlie overtly conserved endogenous expression patterns [32]–[34] , suggesting compensatory changes in cis and in trans [17] , [22] , [35] . In this study , we aimed to survey the amount of functional variation that exists in gene regulatory elements of closely related species . C . elegans offers an attractive model system for this work because of its simple and invariant anatomy [36] , [37] , which is conserved with close relatives [38] . The ease of describing gene expression with a single-cell resolution permits more precise comparisons than those possible in other multicellular model systems . Cis-regulatory sequences are often located within 1 kb upstream of the translation start site [39] . Several species from the Caenorhabditis genus that are approximately as divergent as human and mouse [40] are routinely used for comparisons with C . elegans . We selected eight genes from five Caenorhabditis species that have available genome sequences: C . elegans , C . briggsae , C . remanei , and C . brenneri , the latter three equidistant to C . elegans; and C . japonica , a more distantly related species . In all cases , orthologous regulatory sequences were cloned , and the expression patterns they drove were evaluated in the C . elegans trans-regulatory environment . We report several general trends of cis-regulatory divergence gleaned from these observations . The goal of this study is essentially comparative , that is , to test whether orthologous cis elements are functionally equivalent . Our work is part of a broader research program aiming to investigate functional divergence of gene regulatory systems [41] . In this study we introduced cis-regulatory sequences ( fused to GFP reporters ) from several Caenorhabditis species into C . elegans and compared their expression patterns to those of their C . elegans orthologs . This approach can be seen as an extension of a fruitful paradigm that analyzes gene expression in hybrid organisms [21] , [42]–[44] . In our experiments the “hybrid” portions of the genome range from a few hundred to a few thousand nucleotides directing gene expression . While it is certainly desirable to document endogenous gene expression patterns and uncover all regulatory elements required to direct them , these questions remain outside the scope of our experimental program . Instead , our goal is to assess functional conservation of cis-regulatory sequences . To do so , we need only to ascertain whether cis elements from different species direct the same or different expression patterns . To ensure comparability , only the sequences from the immediately upstream regions were considered; consequently , if some regulatory sequences are located in introns , transgenes may not recapitulate the entire endogenous expression patterns . Movements of cis elements between the upstream intergenic regions in one species and introns in another , dubbed “nomadic” enhancers [30] , illustrate one type of regulatory divergence our approach can uncover . Due to the persistence of the GFP protein , we are unlikely to detect minor dynamic differences in expression patterns . Testing all cis-regulatory elements in the common trans-regulatory environment of C . elegans simplifies the interpretation of these comparative data – any difference in expression patterns , whether gain or loss , reveals functional divergence between orthologous cis-regulatory elements , regardless of the expression patterns driven by these sequences in their endogenous trans-regulatory environments . In addition to C . elegans , we selected for our study four species with sequenced genomes: C . briggsae , C . remanei , C brenneri , and C . japonica [45] , [46] . We decided to focus on these species because , based on previous experience [19] , [22] , [47]–[51] , we anticipated many cis-regulatory functions to be substantially conserved . Given the established phylogenetic relationships between these five species [52] , our experiments interrogated the extent of functional divergence accumulated over two time scales – one between C . elegans and the equidistant C . briggsae/C . remanei/C brenneri , and another between C . elegans and a more distant C . japonica ( Figure 1 ) . Estimates suggest that the phylogenetic distance between the latter pair of species is comparable to that within the Sophophora subgenus of Drosophila [40] , [52] or vertebrate classes [53] . While the phylogeny is well-resolved , the paucity of fossil Rhabditidae nematodes [54] precludes a reliable estimate of the age of species divergence . We focused on eight genes expressed in relatively small groups of easily identifiable cells . Three genes are terminal effectors of the GABAergic fate: unc-25 [55] , unc-46 [56] , unc-47 [57] , and are thus expressed in all GABAergic neurons . Two other genes , oig-1 and acr-14 [58] , are thought to be expressed in subsets of GABAergic neurons . We chose these five coexpressed and partially coregulated [58] genes to test whether shared regulation imposes particular constraints on their evolution . To offset this bias to a particular class of neurons , we added two genes expressed in other neuronal types – one expressed in amphid ( chemosensory ) neurons , gpa-5 [59] , and one expressed in serotonergic neurons , mod-5 [60] . The pattern of serotonergic neurons is conserved between C . elegans , C . briggsae and C . remanei [61]; the pattern of GABAergic neurons is conserved between C . elegans and C . briggsae [22] , as well as with C . remanei and C . brenneri ( AB & IR , unpublished data ) . Finally , we included one gene expressed outside the nervous system , kat-1 , which encodes a conserved thiolase [62] involved in a fat storage pathway [63] . The protein-coding sequences of all eight genes are highly conserved ( Figure 1 ) . Moreover , the synteny with the immediate upstream genes is conserved among all five species ( Figure S1 ) , making us confident that all of them are single-copy , one-to-one orthologs of the C . elegans genes . We tested the entire intergenic regions containing putative cis elements to ensure that comparisons indeed included orthologous regulatory sequences . In contrast with the high conservation of coding sequences , the noncoding upstream regions ( which we assume to contain the majority of CREs [39] ) are much more variable . We aligned orthologous intergenic sequences upstream of C . briggsae , C . remanei , C . brenneri , and C . japonica to their C . elegans counterparts and visualized the results using software package VISTA [64] . The CREs of unc-46 , acr-14 , and unc-47 showed relatively high levels of conservation , spanning ∼150 to 300 nucleotides in most or all species ( Figures 2A–4A ) . The CREs of kat-1 and unc-25 displayed somewhat lower conservation , although some blocks of high similarity could still be clearly identified ( Figures 5A , 6A ) . The CREs of gpa-5 , oig-1 , and mod-5 had little obvious evidence of conservation in the proximity of the translation start site ( Figures 7A–9A ) , although some regions of putative conservation were present substantially upstream of these genes . Sequence comparisons within non-coding regions are notoriously challenging because we do not understand the “rules” by which these sequences evolve [1] . Therefore , we considered two additional measures of sequence divergence , namely the length of the longest contiguous sequence that is perfectly conserved between orthologs and the number of nucleotides contained within blocks of perfect conservation of 7 bp and longer . By both of these measures , cis elements of unc-46 and acr-14 , and to some extent unc-47 , appear to be more conserved than those of the rest of the genes included in this study ( Table S1 ) . Next we tested functional conservation of these regulatory elements . In all experiments we used sequences upstream of translation start sites , thus making translational fusion genes , to ensure that the tested regions encompass basal promoters and more distal regulatory sequences . Expression patterns directed in C . elegans by the orthologous cis elements of the eight studied genes were largely similar ( Figures 2–9; detailed descriptions of the observed patterns are presented in Text S1 ) . However , patterns driven by heterologous CREs were indistinguishable from those directed by their C . elegans orthologs in only three instances: C . brenneri unc-25 ( Figure 6B ) , C . remanei gpa-5 ( Figure 7B ) , and C . brenneri mod-5 ( Figure 9B ) . In the rest of the cases , the expression patterns of heterologous CREs differed from their C . elegans counterparts . Some failed to direct expression in some of the cells in which C . elegans cis elements were active , others drove expression in additional cells . For reasons of brevity , in the following we will refer to the former as “losses” and to the latter as “gains” or ectopic expression , without the implication that these reflect differences in endogenous expression patterns . They do , however , reveal instances of divergence of the regulatory mechanisms controlling expression of orthologous genes in the examined species . “Losses” of expression in the endogenous pattern typically affected single cell types . In two cases ( unc-46 and unc-25; Figures 2B and 6B ) , the expression patterns driven by the C . elegans CREs were completely recapitulated by all heterologous CREs . In three instances ( unc-47 , gpa-5 , and oig-1; Figures 4B , 7B , and 8B ) , while the patterns were qualitatively conserved , portions directed by one or more heterologous CREs were markedly decreased , in frequency or intensity . For example , the C . remanei cis element of unc-47 drives weak and inconsistent expression in the neuron RIS ( Figure 4B ) , the C . briggsae and C . brenneri CREs of gpa-5 direct weak and inconsistent expression in AWAL/R ( Figure 7B ) , and the C . remanei , C . brenneri and C . japonica CREs of oig-1 are expressed inconsistently in DVB ( Figure 8B ) . The C . japonica CRE of acr-14 fails to direct expression in several cell types in the ventral nerve cord , only maintaining expression in D-type neurons , while expression in AVAL/R is much weaker than with other species' CREs ( Figure 3B ) . The C . remanei and C . brenneri CREs of kat-1 fail to drive expression in the gonadal sheath ( Figure 5B ) , the somatic tissue enveloping the proximal gonad . In the most severe case , mod-5 , the CREs from C . briggsae and C . remanei only support expression in ADFL/R ( Figure 9B ) . In addition to “losses” of expression in subsets of endogenous patterns , most heterologous cis elements also drove ectopic expression . Indeed , only six tested CREs did not show any evidence of “gain” of expression: C . remanei unc-47 ( Figure 4B ) , C . brenneri unc-25 ( Figure 6B ) , C . remanei gpa-5 ( Figure 7B ) , and all three heterologous cis elements of mod-5 ( Figure 9B ) . Ectopic expression was seen in as few as one and as many as five different cell types , depending on the gene . In some cases , this expression was driven in the same cells or tissues by all heterologous CREs of a given gene: unidentified lateral ganglion neurons in the head ( unc-46 , Figure 2B ) , AVnL/R neurons in the lateral ganglion ( acr-14 , Figure 3B ) , and head muscles ( kat-1 , Figure 5B ) . In other instances , only some of the orthologous elements directed co-occurring expression: HSNL/R for unc-46 ( Figure 2B ) , hypodermis for kat-1 ( Figure 5B ) , DVB for gpa-5 ( Figure 7B ) , and ADEL/R , PDEL/R , HSNL/R with oig-1 ( Figure 8B ) . The results described above reveal pervasive divergence in cis-regulatory function . However , divergence can also stem from changes in trans regulators [2] , [3] . To test whether the trans environments were functionally equivalent between species , we compared spatial expression patterns driven by four C . briggsae CREs in C . elegans and C . briggsae . Although expression patterns generated by these sequences were qualitatively similar between the two species , in every instance there were reproducible differences as well ( Figure S2 ) . These results further reinforce the notion that divergence has taken place in both cis- and trans-regulatory mechanisms . Most of the orthologous cis elements we analyzed directed patterns of expression in C . elegans that either substantially or completely matched the expression patterns of the orthologous C . elegans CREs ( Figures 2–8; with the possible exception of mod-5 , Figure 9 ) . This result , supported by 30 transgenes , suggests that the mechanisms controlling orthologous gene expression are largely conserved among the studied species . Yet , in the vast majority of these cases ( 27/30 ) , orthologous CREs directed expression patterns that differed from their C . elegans counterparts . These differences were fairly subtle , typically affecting only a few cells , as previously reported in other species [65]–[67] highlighting the value of detailed , focused , multi-gene analyses to reveal trends . Differences in the lengths of tested cis elements did not appear to correlate with the observed differences in expression patterns ( Figure S3 ) . We observed “losses” , as well as “gains” of expression , as compared to the patterns generated by the C . elegans CREs . Even cis elements from two closely related species , C . briggsae and C . remanei , often differed in the expression patterns they directed , indicating that divergence could accumulate relatively quickly . Because in most instances it is difficult to establish the precise endogenous expression patterns of the genes , the observed differences either reflect lineage-specific changes in gene expression or divergence in the mechanisms that regulate conserved expression . In several cases , however , compelling indirect evidence points to the latter scenario . Three of the eight genes in this study , unc-25 , unc-46 , and unc-47 , are terminal effectors of the GABAergic neuronal fate . Immunostaining for GABA in C . elegans [68] , Ascaris suum [69] , and C . briggsae and C . remanei ( AB & IR , unpublished data ) revealed very similar patterns . Furthermore , the expression driven by the C . briggsae unc-47 CRE in its endogenous trans-regulatory environment is identical to that driven by the C . elegans unc-47 CRE in C . elegans [22] . Similarly , patterns of immunostaining for serotonin in C . elegans , C . briggsae , and C . remanei were identical [61] , [70] . These results suggest that the number and relative position of GABAergic and serotonergic neurons , and thus the expression patterns of key genes defining these neuronal fates ( the three GABA genes above and mod-5 ) , are conserved among these Caenorhabditis nematodes . Thus , differences in cis regulatory elements of these four genes ( Figures 2B , 4B , 6B , 9B ) likely reveal changes in the specific ways in which these conserved expression patterns are encoded . This interpretation stresses noticeable divergence in gene regulation even between closely related lineages , consistent with what has been seen in others species [71] , [72] . This view suggests that changes in trans-regulatory mechanisms and cis-regulatory elements accumulate in a somewhat compensatory fashion to ensure that the overall expression patterns of genes remain conserved [22] , [35] , [42] , [73] . The different expression patterns of four C . briggsae CREs in C . elegans and C . briggsae ( Figure S2 ) support the idea that trans-regulatory divergence is prevalent . Consistent with previous reports [73]–[75] , we saw no obvious correspondence between the extent of large-scale sequence conservation and functional conservation . For example , while the CREs of unc-25 and oig-1 show relatively scant primary sequence conservation , their functions appear to be conserved no less well ( Figures 6 , 8 ) than those of genes with apparently greater sequence conservation ( e . g . unc-46 , Figure 2 ) . Sequence comparisons in noncoding regions , particularly when these are of different length , are notoriously challenging . Other metrics of sequence similarity , like the portion of the CRE that is conserved , also failed to reveal a discernible relationship to functional conservation ( Figure S3 , Table S1 ) . We also tested shorter cis elements of mod-5 and unc-25 that excluded the majority of conserved sequence blocks; their expression patterns were qualitatively similar to those of their longer counterparts ( data not shown ) . These findings are consistent with previous reports that conserved expression patterns can be driven by highly divergent regulatory elements [76]–[83] . Previous research suggested that at least in some instances , long tracts of conserved sequences in cis elements may reflect particular features of regulatory organization , rather than unusually stringent selection for the maintenance of expression patterns [84] . Collectively , these results suggest that we may need to reevaluate a common reliance on large-scale sequence conservation when using comparative sequence data to identify cis-regulatory elements . Presence or absence of transcription factors binding sites , their arrangement and spacing may be more informative , although harder to detect [73] , [85]–[89] . We did not detect greater functional divergence of CREs from the more distant C . japonica compared to C . briggsae , C . remanei , and C . brenneri . Among the six genes that have been tested from all four of these species , C . japonica cis elements show approximately the same number of “gains” and “losses” as their orthologs from other species ( Table S2 ) . It is possible that the ∼2-fold difference in the phylogenetic distance [40] separating , on the one hand , C . elegans and C . japonica and , on the other hand , C . elegans and C . briggsae/C . remanei/C . brenneri , does not offer enough power to test this hypothesis . Examining more distantly related pairs of species may be required . Finally , the complexity of the expression pattern of a gene does not seem to be correlated with the amount of functional divergence in its cis element ( Figure S3 ) . One striking pattern evident in our results is that a substantial majority of functional differences between orthologous cis elements is due to “gain” , rather than “loss” or reduction , of expression relative to the pattern directed by the C . elegans CREs . Put another way , when tested in C . elegans , heterologous regulatory elements more commonly directed expression in more rather than fewer cells , compared to the C . elegans-driven patterns . When all experiments reported here are considered together , the total number of “gains” was nearly three-fold higher than the number of “losses” ( 51 vs . 18 ) . Even when minor differences in patterns are counted as “losses” , their number ( 23 ) is still less than half than that of “gains” ( 51 ) . This phenomenon does not appear to be due to greater power to detect “gains” compared to “losses” ( Figure S4 ) . Restricting comparisons only to those genes for which all four non-C . elegans species were tested , does not substantially alter this conclusion ( 12 vs . 44 or 16 vs . 44 , if “losses” are counted more liberally ) . Therefore , our results suggest that the two regulatory modalities , namely one directing expression in certain cells and another repressing inappropriate expression , evolve at different rates . The molecular mechanisms and evolutionary forces that could account for this observation remain to be investigated . It is possible , however , that the positive and negative regulatory aspects of gene regulation evolve under different regimes , because of the difference in the ways in which they are encoded within cis elements . The relatively large number of cases in which heterologous cis elements directed ectopic expression when in C . elegans , allowed us to investigate whether these “gains” followed a pattern . Notably , for the neuronal genes unc-46 , acr-14 , unc-47 , unc-25 , oig-1 , and gpa-5 , nearly all “gains” occurred in neurons ( Figures 2–4 , 6–8 ) . This tropism suggests that the regulatory architecture of neuronal CREs – some transcriptional inputs are pan-neuronal in nature [90] , [91] – may restrict ectopic expression to neurons . We further noted that in several instances , CREs of different genes or from different species directed ectopic expression in the same cells ( Figure 10 ) . The cells “gaining” expression do not appear to be transcriptionally promiscuous , because ectopic expression is seen in several different cells not previously noted for indiscriminate expression ( Text S1 ) . Furthermore , the “gain” of expression is not likely to be due to effects of vector sequences . We used a standard vector utilized by us and others thousands of times . Previous studies using this vector documented ectopic expression in the intestine and pharynx [49] , [92] , not specific subsets of neurons , as we reported here . Instead , we favor a hypothesis that the cis elements themselves could be sharing certain characteristics that make them more likely to direct expression in particular cells . The recurrent “gains” of expression were seen for unc-46 , acr-14 , unc-47 , and oig-1 , which are coexpressed in a subset of GABAergic neurons and are know to be coregulated by at least one transcription factor , UNC-30 [58] . It is therefore plausible that these cis elements share some features , for example transcription factor binding sites or general organization , and that this similarity may bias the trajectories that evolution could follow [15] . This may in part account for the commonly observed instances of parallel evolution [33] , [93]–[95] . With this survey , we established several trends of functional conservation and divergence of cis-regulatory elements . We found pervasive functional divergence in transcriptional regulatory mechanisms , both in cis and in trans . More strikingly , we identified inherent biases in the nature and functional consequences of this divergence , hinting at possible mechanisms underlying repeated evolution . Putative cis-regulatory elements ( extending from the first exon to the nearest upstream gene ) were PCR amplified from genomic DNA using Phusion polymerase and cloned upstream of GFP into the pPD95 . 75 plasmid , routinely used for analysis of gene expression in C . elegans [96] . Cloned fragments were sequenced to ensure accuracy . C . elegans CREs of unc-46 , acr-14 , kat-1 , unc-25 , and oig-1 , were also cloned into the plasmid HYM153 ( kind gift of H . -Y . Mak ) upstream of the mCherry reporter gene as controls . C . elegans transgenic lines were established by injecting into pha-1 ( e2123 ) worms cocktails consisting of 5 ng/µL CRE::GFP reporter constructs with 5 ng/µL rescue plasmid [97] and 100 ng/µL salmon sperm DNA; this is thought to facilitate the formation of complex transgenic constructs as extrachromosomal arrays [98] . For five genes ( unc-46 , acr-14 , kat-1 , unc-25 , and oig-1 ) , plasmids carrying C . elegans CREs fused to mCherry were coinjected with the plasmids carrying orthologous CREs from each of the five species fused to GFP . C . briggsae transgenic lines were established by injecting cocktails consisting of 5 ng/µL CRE::GFP reporter constructs with 5 ng/µL rescue plasmid and 100 ng/µL salmon sperm DNA into Cbr-unc-119 ( nm67 ) worms [99] . Mixed-stage populations of transgenic worms were grown with abundant food and L4-stage larvae or young adults were selected . These were immobilized on agar slides with 10 mM sodium azide in M9 buffer . The slides were examined on a Leica DM5000B compound microscope under 400-fold magnification . Worms without any visible GFP expression were assumed to have lost the transgene . Each photograph showing worms in figures is composed of several images of the same individual capturing anterior , middle , and posterior sections . At least fifty individuals from no fewer than two independent strains were analyzed for each transgene . The plasmid pPD95 . 75 has been used extensively by the C . elegans community over the last two decades . It has been reported to direct low-level background expression in the pharynx and anterior and posterior intestine [49] , [92] , [96] . We have previously reported that extrachromosomal arrays direct expression patterns that are concordant with those of integrated and single-copy transgenes [22] , [100] . Still , to obtain conservative estimates of expression differences between CREs from C . elegans and other species , we only counted discrepancies ( missing or extra expression ) observed in two or more strains . Data on consistency of expression patterns between strains and individuals are presented in Tables S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 and S10 .	Given the importance of gene expression changes in evolution , a better understanding of how they accumulate is desirable . However , gene regulation is a complex biochemical process and it is not a priori clear whether general trends even exist . We systematically addressed this question by testing , in C . elegans , the functions of regulatory elements of eight different genes from four other nematodes . We saw rampant variation in gene regulatory mechanisms , even between closely related species . While the differences were usually seen in a relatively small number of cells , there was a discernible trend – there were many more instances of gain , rather than loss of expression , compared to patterns directed by the C . elegans cis elements . Finally , the recurrence of ectopic expression in the same cells suggests that the paths open to evolution may be constrained by the composition of regulatory elements . We view these patterns as a reflection of general mechanisms of gene regulatory evolution and suggest that these can be refined , and others discovered , using systematic functional tests .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "invertebrates", "caenorhabditis", "gene", "regulation", "animals", "dna", "transcription", "animal", "models", "developmental", "biology", "caenorhabditis", "elegans", "model", "organisms", "molecular", "genetics", "morphogenesis", "pattern", "formation", "research", "and", "analysis", "methods", "gene", "expression", "evolutionary", "genetics", "gene", "regulatory", "networks", "genetics", "nematoda", "biology", "and", "life", "sciences", "computational", "biology", "evolutionary", "biology", "organisms", "evolutionary", "developmental", "biology" ]	2014	Pervasive Divergence of Transcriptional Gene Regulation in Caenorhabditis Nematodes
Type III secretion system 1 ( T3SS1 ) is used by the enteropathogen Salmonella enterica serovar Typhimurium to establish infection in the gut . Effector proteins translocated by this system across the plasma membrane facilitate invasion of intestinal epithelial cells . One such effector , the inositol phosphatase SopB , contributes to invasion and mediates activation of the pro-survival kinase Akt . Following internalization , some bacteria escape from the Salmonella-containing vacuole into the cytosol and there is evidence suggesting that T3SS1 is expressed in this subpopulation . Here , we investigated the post-invasion role of T3SS1 , using SopB as a model effector . In cultured epithelial cells , SopB-dependent Akt phosphorylation was observed at two distinct stages of infection: during and immediately after invasion , and later during peak cytosolic replication . Single cell analysis revealed that cytosolic Salmonella deliver SopB via T3SS1 . Although intracellular replication was unaffected in a SopB deletion mutant , cells infected with ΔsopB demonstrated a lack of Akt phosphorylation , earlier time to death , and increased lysis . When SopB expression was induced specifically in cytosolic Salmonella , these effects were restored to levels observed in WT infected cells , indicating that the second wave of SopB protects this infected population against cell death via Akt activation . Thus , T3SS1 has two , temporally distinct roles during epithelial cell colonization . Additionally , we found that delivery of SopB by cytosolic bacteria was translocon-independent , in contrast to canonical effector translocation across eukaryotic membranes , which requires formation of a translocon pore . This mechanism was also observed for another T3SS1 effector , SipA . These findings reveal the functional and mechanistic adaptability of a T3SS that can be harnessed in different microenvironments . Type III Secretion Systems ( T3SSs ) are used by a variety of Gram-negative bacteria for interkingdom delivery of proteins ( known as effectors ) from the bacterial cytosol into eukaryotic cells [1] . For bacterial pathogens , such as Salmonella enterica , Yersinia spp , and pathogenic Escherichia coli , these molecular syringes are key virulence determinants essential for a variety of processes including: adherence; invasion; intracellular survival and cytotoxicity . This broad repertoire is due to the diversified nature of effectors rather than the mechanism of delivery , which is highly conserved [2 , 3] . T3SS delivery is a contact-dependent process characterized by the formation of a pore , or translocon , at the point of contact with the eukaryotic membrane and through which effectors are delivered into the host cell [4] . Salmonella enterica serovar Typhimurium ( hereafter Salmonella ) , a leading cause of gastroenteritis , possesses two functionally distinct T3SSs , encoded on Salmonella Pathogenicity Islands 1 and 2 ( SPI1 and SPI2 ) [5] . The SPI1-encoded T3SS1 triggers invasion of non-phagocytic cells , such as intestinal epithelial cells , following contact with the plasma membrane . A cohort of translocated effectors targets the actin network and membrane phospholipids to direct formation of membrane ruffles , leading to uptake of the bacterium into a modified phagosome known as the Salmonella Containing Vacuole ( SCV ) [6] . Within the SCV , the SPI1 regulon is rapidly down-regulated whereas the SPI2 regulon is induced . Consequently , SCV biogenesis is primarily determined by effectors translocated via T3SS2 [7] . In epithelial cells , some Salmonella escape from the SCV and can survive and replicate in the cytosol resulting in two distinct populations of intracellular bacteria [8 , 9] . Cytosolic Salmonella replicate faster than vacuolar bacteria [9 , 10] and this “hyper-replication” results in a subpopulation of infected cells that are filled with Salmonella , both in vitro and in vivo [8] . In cell culture models , such as HeLa and C2BBe1 cells , cytosolic replication occurs in a largely synchronous fashion starting at ~4 h post-infection ( hpi ) and continuing for several hours until the inflammasome mediated death of the host cell at ~8–10 hpi [8 , 9 , 11 , 12] . In contrast , the contribution from vacuolar bacteria to intracellular replication is primarily seen from 12 hpi onwards [9] . Thus , within a time frame of 4–10 hpi , cells containing cytosolic Salmonella can be distinguished from those containing only vacuolar Salmonella due to the higher bacterial numbers . Additionally , these populations can be differentiated using fluorescent transcriptional reporters for SPI1 and SPI2 genes during this time period . SPI2-induced bacteria are only observed in the vacuole whereas SPI1 induction has only been observed in cytosolic hyper-replicating bacteria [8 , 12] . Several lines of evidence suggest that the SPI1-encoded T3SS1 has post-invasion activities in addition to its well characterized role in invasion . For example , the T3SS1 effector SopB ( SigD ) , contributes to actin remodeling and induces phosphorylation of the pro-survival kinase Akt during invasion of epithelial cells [13–16] , but SopB-dependent Akt phosphorylation can be detected for several hours following invasion [15] and it has been implicated in intracellular replication of Salmonella [17] . It is not clear whether the late activity of SopB can be attributed solely to effector translocated by T3SS1 during or post-invasion , or if it can be translocated by T3SS2 . An alternative possibility is that the SP1-induced cytosolic subpopulation of intracellular Salmonella delivers de novo synthesized SopB . However , it remains to be determined whether T3SS1 is functionally active in this intracellular population of bacteria . Here , we investigated the post-invasion intracellular role of T3SS1 using SopB as a model effector and Akt phosphorylation as an indicator of effector activity . Our study reveals widespread expression and activity of T3SS1 in cytosolic Salmonella . This activity results in delivery of a second wave of SopB , leading to resurgence in Akt phosphorylation and ultimately prolonging the lifespan of this subpopulation of infected cells . Unexpectedly , effector delivery was not inhibited in the absence of the translocon , indicating that effector delivery by cytosolic Salmonella can occur via a non-canonical translocon-independent mechanism . We previously reported that the T3SS1 effector SopB induces sustained Akt phosphorylation following T3SS1-mediated invasion of epithelial cells by Salmonella Typhimurium [14 , 15] . Akt phosphorylation is readily detected in infected HeLa cell lysates using antibodies that specifically recognize phosphorylated Akt . Although Akt phosphorylation peaks at 0 . 5–1 hpi , a low level of phosphorylation is sustained for at least 7 hpi ( Fig 1A ) . We hypothesized that this persistent Akt phosphorylation could be due to either: a low level of phosphorylation in all infected cells; or a high level of phosphorylation in a subpopulation of infected cells . In particular , we wondered if Akt phosphorylation levels were different in cells containing cytosolic vs vacuolar Salmonella . These two distinct intracellular populations can be resolved in infected HeLa cells by a modified gentamicin protection assay , in which chloroquine is used to selectively kill vacuolar bacteria , or by fluorescence microscopy ( Fig 1B ) . We focused these studies on the time period from 2–8 hpi , which encompasses the peak of cytosolic replication [9 , 12] . In order to synchronize invasion , and the subsequent intracellular processes , we grew the Salmonella under SPI1-inducing conditions and then used a high moi ( approximately 50 bacteria/cell ) and a relatively short infection time ( 10 min ) as previously described [18 , 19] . Under these conditions , the number of bacteria internalized into each cell is controlled so that over 95% of infected cells contain 1–10 bacteria at 2 hpi ( Fig 1B ) . Following the onset of intracellular replication , the number of bacteria per cell increases and by 8 hpi approximately 40% of infected cells contain >10 bacteria . Approximately 25% of these cells , or 10% of total infected cells , contain hyper-replicating cytosolic bacteria and can be readily identified due to the high number ( >50 per cell ) of intracellular bacteria [8 , 9 , 12] . To determine whether Akt phosphorylation is increased in cells containing cytosolic bacteria , we used fluorescence microscopy combined with in situ proximity ligation amplification ( PLA ) , a sensitive and specific method to detect low levels of proteins [20] . In cells infected with WT Salmonella , confocal microscopy revealed high levels of phosphorylated Akt ( pAkt ) compared to uninfected cells ( Fig 1C and 1D ) . At 1 hpi the pAkt signal was 20- to 30-fold higher in cells containing WT Salmonella compared to uninfected cells in the same monolayer ( Fig 1D ) . Akt phosphorylation was similarly increased in cells infected with a SopB deletion mutant ( ΔsopB ) complemented with plasmid borne SopB ( psopBWT ) but not a catalytically inactive mutant ( psopBC460S ) [14 , 15] . This SopB-dependent Akt phosphorylation was transient since , by 3 hpi , pAkt levels in infected cells decreased to near background levels and were not significantly affected by the presence or absence of active SopB ( Fig 1D ) . A resurgence of Akt phosphorylation was observed at 6 hpi , although only in a subset of infected cells . We observed a clear correlation between the number of intracellular bacteria and levels of pAkt , which was 20- to 30-fold higher than background in cells containing ≥20 bacteria but was at background levels in cells containing <20 bacteria ( Fig 1C and 1D ) . This cut off was selected as we have previously shown that , within this time frame , cells with 20 or more bacteria contained predominantly cytosolic Salmonella [8] . As at 1 hpi the increase in pAkt was dependent on the catalytic activity of SopB ( Fig 1C and 1D ) . Similar results were seen in C2BBe1 colorectal epithelial cells ( Fig 1E ) . These results show that , at a single cell level , there are two distinct periods of SopB-dependent Akt phosphorylation . The first phase is widely induced during invasion and is largely depleted by 3 hpi , whereas the second phase is induced later ( observed at 6 hpi ) in the subpopulation of infected cells that contain greater than 20 , likely cytosolic , bacteria . Since SopB-dependent Akt phosphorylation was detected in cells containing >20 bacteria at 6 hpi we next wanted to: confirm the presence of SopB in this subpopulation; determine its subcellular localization; and show whether it correlates with the presence of cytosolic or vacuolar bacteria . Detection of T3SS effectors , such as SopB , in host cells is difficult due to the low amounts delivered as well as a lack of good antibodies . For these reasons , we used epitope tagged SopB for these experiments . Specifically , we used WT Salmonella expressing 3xFLAG-tagged SopB ( SopB3xFLAG ) from the sopB chromosomal locus and under the control of its native promoter ( Salmonella sopB3xFLAG ) . Infected HeLa cells were fixed and stained with an anti-FLAG antibody . Under the conditions used here ( paraformaldehyde fixation and permeabilization with saponin ) antibodies cannot cross the bacterial cell wall so that intrabacterial SopB3xFLAG is not detected . At 1 hpi , ~20–30% of infected cells contained detectable levels of extrabacterial SopB3xFLAG ( SopB+ cells ) ( Fig 2A and 2B ) . Thereafter , the number of SopB+ cells rapidly decreased so that by 2 hpi , less than 10% of infected cells were SopB+ and this did not change significantly up to 8 hpi . At 6 hpi , similar to pAkt , SopB3xFLAG was predominantly detected in cells containing >20 bacteria . In cells infected with the ΔsopB strain , no significant FLAG staining was observed at any time point even in cells containing hyper-replicating bacteria . Interestingly , SopB3xFLAG staining was predominantly detected in close proximity to individual bacteria whereas pAkt was distributed throughout the cytosol ( compare Figs 1C and 2B ) . In our experience , cytosolic bacteria are readily distinguished from vacuolar bacteria at later time points ( 4–8 hpi ) due to their greater intracellular numbers . However , individual cells are often infected with more than one bacteria and can harbor both vacuolar and cytosolic bacteria . Therefore , in order to definitively determine whether SopB3xFLAG staining is associated with cytosolic or vacuolar bacteria , and how this changes with time , we used digitonin-based differential permeabilization ( a . k . a . phagosomal protection assay ) [8] . Selective permeabilization of the plasma membrane , while leaving the phagosome membrane intact , allows cytosolic bacteria to be selectively stained with antibodies . Cells were infected with Salmonella sopB3xFLAG , or ΔsopB strain as a control . Cells containing one or more bacteria and staining positive for SopB3xFLAG were categorized into two populations: 1 ) Infected cells containing only vacuolar bacteria , and 2 ) Infected cells containing any cytosolic bacteria . At 1 hpi , the T3SS1 effector was found in both populations ( Fig 2C and 2D ) . In contrast , at 6 hpi , SopB3xFLAG staining was almost exclusively detected in cells containing cytosolic hyper-replicating bacteria . To determine whether this is specific to SopB we performed similar experiments with another T3SS1 effector , SipA , and obtained almost identical results ( S1 Fig ) . Thus , at 6 hpi , the presence of T3SS1 effectors in infected cells strongly correlates with the presence of hyper-replicating cytosolic Salmonella reinforcing that it is this population , rather than vacuolar bacteria , which delivers the effectors into the host cell . The above experiments revealed that , in HeLa cells , T3SS1 effectors are present in the subpopulation of cells that contain cytosolic hyper-replicating Salmonella . To investigate whether this phenotype occurs in vivo , we utilized a C57BL/6J mouse model in which Salmonella disseminates to various tissues , including the gallbladder . Cytosolic hyper-replication has been observed in gallbladder epithelial cells which are extruded into the lumen where they can be readily identified [8] . In order to detect the T3SS1 effectors , mice were infected with Salmonella expressing either SopB3xFLAG or SipA3xFLAG from their chromosomal loci . Gallbladders harvested 5 days pi were stained for Salmonella and FLAG-tagged effectors . Extruded epithelial cells containing cytosolic , hyper-replicating Salmonella were observed in the lumen and many of these cells stained for SopB3xFLAG or SipA3xFLAG ( Fig 3A , 3B and S2 Fig ) . In contrast , there was no significant FLAG staining in cells containing hyper-replicating WT Salmonella ( no FLAG-tagged effectors , Fig 3C ) . Thus , in vivo , SopB and SipA are present in extruded epithelial cells containing cytosolic Salmonella . To separate the post-invasion role of SopB in the cytosolic population of Salmonella from its role in invasion we used an inducible gene expression system in the ΔsopB background . The hexose phosphate transporter promoter , PuhpT , responds to exogenous glucose-6-phosphate , a glucose metabolite that is exclusively found in the cytosol [21 , 22] . To first verify the fidelity of PuhpT in our system , we used a plasmid borne GFP transcriptional reporter , pPuhpT-gfp , in WT Salmonella . These bacteria were internalized into HeLa cells and then the differential permeabilization assay and confocal microscopy used to determine their intracellular localization and GFP status ( S3 Fig ) . At all time points GFP+ bacteria were found exclusively in the cytosol confirming that PuhpT is specifically induced in this subpopulation of intracellular Salmonella . Next , we used a transcriptional fusion of this cytosol inducible promoter with HA-tagged sopB , PuhpT-sopB2XHA , in a ΔsopB background to assess the role of SopB in cytosolic bacteria . As expected , immunostaining for SopB2xHA in infected cells showed an expression pattern similar to GFP under PuhpT , confirming that the effector was expressed only in cytosolic Salmonella ( Fig 4A ) . Using PLA to detect pAkt revealed a dramatic increase in pAkt in SopB2xHA-positive cells , particularly in cells with cytosolic hyper-replicating bacteria at 6 hpi . In contrast , infected cells lacking bacterial SopB2xHA expression were devoid of pAkt . To confirm that there was no SopB activity at early time points in this system , lysates prepared from cells infected with ΔsopB/pPuhpT-sopB2xHA were assessed by Western blotting ( Fig 4B ) . At 0 . 5 hpi , the pAkt levels were similar to those infected with the ΔsopB strain and uninfected cells whereas at 3 and 6 hpi , pAkt levels were 2 . 5- and 2-fold higher than ΔsopB infected cells , respectively , and were comparable to WT infected cells . Increased Akt phosphorylation was SopB-dependent since there was no increase in cells infected with ΔsopB bearing a promoterless control plasmid ( pPNULL-sopB2xHA ) ( Fig 4A and 4B ) . These results illustrate that post-invasion cytosolic induction of SopB expression is sufficient for late Akt phosphorylation in infected cells . The above experiments showed that , in the absence of SopB during invasion , de novo production of SopB by cytosolic bacteria could account for SopB-dependent Akt phosphorylation at 6 hpi . However , we also wanted to know whether , in the absence of cytosolic SopB , persistence following invasion could also possibly contribute to Akt phosphorylation . For this , we used a strain ( ΔsopB/pPBAD-sopB2xHA ) with a plasmid borne arabinose inducible construct . Expression of the tagged effector ( SopB2xHA ) was induced immediately prior to invasion by addition of arabinose during growth of Salmonella under SPI1-inducing conditions . Following internalization in HeLa cells , SopB2xHA should be rapidly depleted due to the absence of arabinose in the intracellular environment . Immunofluorescence microscopy and Western blot analysis of HeLa cells infected with ΔsopB/pPBAD-sopB2xHA showed SopB2xHA expression and elevated pAkt only at 0 . 5 hpi , but not at 3 and 6 hpi ( Fig 4A and 4C ) . Thus , SopB delivered during invasion does not persist and , therefore , does not sustain Akt phosphorylation . Altogether , these results indicate that SopB delivered by cytosolic bacteria post-invasion is both essential and sufficient for sustained Akt phosphorylation in infected HeLa cells . We have previously found that SopB could protect infected HeLa cells from apoptosis [15] , although at that time we assumed that intracellular replication of Salmonella occurred within the SCV . The data presented above suggests that the anti-apoptotic effect of SopB could be limited to cells containing cytosolic hyper-replicating Salmonella . To address this question , we measured release of the cytosolic enzyme lactate dehydrogenase ( LDH ) from monolayers of HeLa cells infected with WT or ΔsopB Salmonella at 6 hpi ( Fig 5A and 5B ) . At this time point , cells infected with the WT strain had similar levels of lysis ( 3 . 5±0 . 9% ) as uninfected cells whereas the cells infected with the ΔsopB strain showed a two-fold increase in cell lysis ( 8 . 6±0 . 6% ) , despite lower numbers of intracellular of ΔsopB relative to WT ( Fig 5A and 5B ) . In contrast , no increase in lysis was observed in cells infected with the ΔsopB strain complemented with plasmid borne SopB , either under the control of its own promoter ( psopBWT ) or the uhpT promoter ( pPuhpT-sopB-2xHA ) . A control “promoter-less” plasmid ( pPNULL-sopB-2xHA ) did not rescue this phenotype . Considering that only 10–15% of infected cells contain hyper-replicating cytosolic Salmonella ( Fig 1 [8] ) and that not all HeLa cells in the monolayer are infected , the relatively small total amount of lysis caused by strains lacking SopB is not unreasonable . The lower numbers of ΔsopB bacteria observed at 6 hpi ( Fig 5B ) as well as other studies suggested that SopB may contribute to intracellular replication of Salmonella [17] . However , we were unable to detect any defect in intracellular replication of the ΔsopB strain when the results were normalized to the number of intracellular bacteria at 2 hpi ( Fig 5C and 5D ) . In order to gain a better understanding of the role of SopB in cells containing hyper-replicating cytosolic bacteria we focused our analysis on this specific sub-population of cells by using a live cell imaging approach previously used to compare the replication rates of cytosolic and vacuolar Salmonella [9] . In order to assess cytosolic replication we took advantage of the plasmid borne GFP transcriptional reporter , pPuhpT-gfp , described above ( Fig 4 ) . Infected HeLa cells were imaged on a spinning disc confocal system for up to 10 h pi ( Fig 5E , 5F and 5G ) . As a rapid readout for cell death , propidium iodide was used . This red fluorescent nuclear and chromosome counterstain is not permeant to live cells and is commonly used to detect dead cells . Post acquisition analysis of time lapse movies showed that cells containing the ΔsopB strain died earlier ( 412 min pi ) than those containing WT Salmonella ( 482 min pi ) ( Fig 5E and 5G ) . In contrast , there was no detectable difference in the doubling rates of the two strains ( Fig 5F ) . As a control for doubling rate , we included a strain in which constitutive expression of red fluorescent protein ( RFP ) causes a defect in intracellular replication . Altogether these results show that SopB does not affect intracellular replication of Salmonella but rather promotes the survival of HeLa cells containing hyper-replicating cytosolic Salmonella . Previous studies have shown that some , but not all , cytosolic hyper-replicating bacteria are SPI1 induced [8 , 12] . Since this would have significant implications for our findings that SopB-dependent Akt phosphorylation is widespread in this population of cells , we considered whether this was an accurate estimation of SPI1 induction . One possibility is that limitations in the sensitivity of the experimental systems could result in an underestimation of SPI1 induction . Specifically , HeLa cells were infected with Salmonella containing a GFP-prgH transcriptional reporter with destabilized GFP , GFP[LVA] ( PprgH-gfp[LVA] ) [8 , 12] . As shown in Fig 6A , this system reveals many SPI1-induced ( GFP+ ) bacteria at 0 . 5 hpi and , thanks to the short half-life of the destabilized GFP[LVA] , down-regulation of SPI1 results in a dramatic decrease in the number of GFP+ bacteria by 3 hpi . By 6 hpi , GFP fluorescence reappears , although only a fraction of hyper-replicating cytosolic bacteria are GFP+ , suggesting that SPI1 induction is not universal in this population ( [8] and Fig 6A ) . Nevertheless , we considered that , while the use of destabilized GFP[LVA] is critical for following down-regulation of the prgH promoter , it could also result in the under-detection of SPI1-induced bacteria . Therefore , to increase the sensitivity of detection without losing temporal fidelity , we used an anti-GFP antibody to amplify the fluorescent signal ( Fig 6B ) . One caveat of this system is that , in order to get the antibodies into the bacteria , we had to permeabilize with methanol fixation , which resulted in denaturation of GFP and , consequently , loss of fluorescence . However , antibody recognition was not affected and thus the antibody-mediated amplification revealed a dramatic increase in the proportion of cytosolic hyper-replicating bacteria that were GFP+ at 6 hpi ( Fig 6 , compare A and B ) . The fidelity of the amplified system was confirmed by the lack of any detectable increase in GFP staining in intracellular populations of bacteria that are not SPI1 induced ( see 3 h time point and vacuolar bacteria at 6 h , Fig 5B and 5C ) . Thus , by 6 hpi , T3SS1 is widely expressed in cytosolic hyper-replicating , but not vacuolar , Salmonella . During invasion of epithelial cells , SPI1 effectors , including SopB and SipA , are translocated across the plasma membrane by the T3SS1 . In the absence of a functional T3SS1 , bacterial internalization into HeLa cells is almost completely abrogated , making it technically challenging to study the role of T3SS1 in post invasion events [23] . To circumvent this problem , we developed an inducible strain , which has a functional T3SS1 for invasion , but is essentially a T3SS1 defective mutant in the intracellular environment . To do this we used the arabinose-inducible promotor PBAD to control expression of the SPI1 gene invA , which encodes a conserved structural component of T3SS1 ( Fig 7A ) [24] . This T3SS1 arabinose-inducible strain , T3SS1IND+-sopB3xFLAG , was invasion competent when grown under SPI1-inducing conditions in the presence of arabinose ( S4A Fig ) but was unable to deliver detectable amounts of either of the T3SS1 effectors ( SopB3xFLAG , SipA ) in cells containing cytosolic hyper-replicating Salmonella ( >50 bacteria/cell ) ( Fig 7 ) . In these , and subsequent , experiments we used a higher stringency cut-off for cytosolic bacteria ( >50 bacteria/cell ) so as to avoid any possibility of contamination by cells containing only vacuolar bacteria . To confirm that the defect in effector delivery by T3SS1IND was not due to a lack of effector expression , we immunostained intracellular bacteria following methanol-permeabilization . Levels of intrabacterial effector staining for the T3SS1IND and WT strains were comparable ( S4B–S4E Fig ) . We also tested whether the SPI2 encoded T3SS2 could be involved in SopB or SipA delivery since some T3SS1 effectors , including SopB , have the potential to be delivered by T3SS2 as well as T3SS1 [25 , 26] . However , a T3SS2 deficient strain ( ΔT3SS2 ) delivered SopB3xFLAG and SipA at levels indistinguishable from Salmonella WT ( Fig 7C and 7E ) . Similar results , showing a requirement for T3SS1 but not T3SS2 , were obtained in C2BBe1 cells ( Fig 7D and 7F ) . Thus , T3SS1 , but not T3SS2 , is required for delivery of the effector proteins SopB and SipA by cytosolic hyper-replicating Salmonella . Canonical delivery of effectors into host cells via T3SSs requires contact with host membrane and subsequent formation of a translocon pore through which the effectors are delivered . For T3SS1 , the translocator proteins , SipB and SipC , are required to establish the translocon pore and translocation does not occur in their absence ( see illustration in Fig 7A ) [27–29] . Immunostaining for SipB in WT Salmonella infected cells revealed this translocon component in association with cytosolic but not vacuolar bacteria at 6 hpi , corroborating the induction of T3SS1 in cytosolic Salmonella ( S5 Fig ) . To examine whether delivery of SopB and/or SipA by cytosolic Salmonella requires the T3SS1 translocon , we took advantage of a translocation defective mutant ( ΔsipB ) , which is unable to invade epithelial cells ( S4F Fig ) but retains the ability to secrete effectors in broth culture ( S4G and S4H Fig ) . To generate an invasion-competent sipB mutant strain , we again used the PBAD inducible system so that sipB expression was induced when bacteria were grown under SPI1-inducing conditions in the presence of arabinose ( SipBIND ) ( S4I Fig ) . Confocal microscopy was used to assess the ability of the SipBIND strain to deliver SopB3xFLAG and/or SipA3xFLAG in epithelial cells . At 6 hpi , effector staining in infected HeLa cells infected with WT or SipBIND bacteria revealed no difference in the numbers of cells containing cytosolic hyper-replicating Salmonella ( >50 bacteria/cell ) that were positive for effectors ( ~40–50% , Fig 8B and 8C ) . Immunostaining for SipB in SipBIND infected HeLa cells confirm expression at 0 . 5 hpi but not at 3 and 6 hpi , excluding the contribution of residual SipB following invasion ( S5 Fig ) . Similar results were obtained in C2BBe1 cells ( Fig 8D and 8E ) . Thus , the T3SS1 translocon pore is not required for SopB or SipA delivery by cytosolic bacteria . The ability of Salmonella to actively invade and colonize epithelial cells is critical for pathogenesis . Following invasion , the bacteria can survive and replicate within the SCV , a modified phagosome , or in the cytosol . Effectors translocated by the SPI1-encoded T3SS1 play critical roles in invasion . Here , we show a novel role for T3SS1 in the cytosolic subpopulation of intracellular bacteria ( See model in Fig 9 ) . Delivery of the effector SopB , via a translocon-independent mechanism , leads to Akt phosphorylation and prolonged survival of epithelial cells containing cytosolic Salmonella . In enteric infection , environmental signals in the lumen of the small intestine—though still not well understood—are believed to be essential for SPI1 induction [31 , 32] . Our findings show that induction can also occur in the intracellular environment , specifically the cytosol of infected epithelial cells . This must confer an advantage to the bacteria since expression of the SPI1-encoded T3SS1 , and its associated effectors , comes at a cost [33] . The T3SS1 needle/apparatus is recognized in the cytosol of mammalian cells resulting in inflammasome activation and host cell death [34] . In epithelial cells , the SPI1-encoded T3SS1 is specifically expressed in the cytosolic subpopulation of bacteria [8] , leading to cell death via activation of a non-canonical inflammasome [11 , 35] . Whether this is an advantage to the host or pathogen may depend on the context . Is SPI1-induction in these bacteria priming them for the extracellular environment , i . e . lumen of small intestine , or is there a specific function for T3SS1 in the cytosolic niche ? Here , we show that the latter may be true since delivery of T3SS effectors delays the onset of cell death thus conferring an advantage to the intracellular pathogen . The activity of Akt , a central regulator of eukaryotic cell death and survival , is regulated by phosphorylation . Salmonella uses SopB to target this pro-survival kinase suggesting an important role for T3SS1 in regulating host cell survival [14 , 15 , 36] . However , since SopB is not essential for intracellular replication , the role of SopB-dependent Akt phosphorylation in epithelial cells was still unclear . In order to resolve this disparity , we used a single-cell approach to re-examine the temporal relationship between SopB and Akt phosphorylation . We found that , at later time points , Akt phosphorylation was highly induced in the subpopulation of infected cells containing hyper-replicating cytosolic Salmonella . De novo T3SS1-dependent delivery of SopB by cytosolic bacteria , as opposed to persistence of the effector , was key to this second wave of Akt phosphorylation . As a master regulator , Akt affects diverse cellular processes and lies at the crossroads between cell survival and death [37] . Dysregulation has been implicated in many human cancers and recently , an interesting link between Salmonella , Akt and human cancer was described [38] . The incidence of gallbladder cancers is increased in areas where there is high incidence of Salmonella Typhi infection , and experiments in chronically infected mice showed that tumors are induced in a SopB-dependent manner , suggesting a causal relationship [39] . This is consistent with our finding showing that SopB-dependent Akt activitation does not directly affect the ability of Salmonella to replicate but rather delays host cell death [15 , 40–43] . This may be a more widespread strategy since the PI3K-Akt signaling cascade is also targeted by other bacterial pathogens , including Chlamydia trachomatis , Chlamydia pneumonia , Coxiella burnettii and Mycobacterium tuberculosis , to delay or prevent apoptosis in infected cells [44–47] . Previously , the continued presence of T3SS1 effectors post-invasion has been explained by effector persistence in the intracellular environment [48] or delivery by the T3SS2 [25 , 26 , 49] . The main reason that the role of T3SS1 post-invasion is not well understood is because mutants lacking T3SS1 activity are unable to infect host cells . To overcome this obstacle we developed a set of inducible mutants , which allowed us to specifically address the post-invasion roles of the T3SS1 and SopB . Using this approach yielded a more complete picture of the differences between the two intracellular populations of Salmonella and particularly the activity of T3SS1 in the cytosolic population . SopB delivery by cytosolic bacteria requires continued transcription and translation of the molecule by the bacterium post-invasion [50] . Furthermore , T3SS2 has no role in delivery of SopB or SipA to infected host cells , although that has previously been suggested as a mechanism for sustained delivery both in vitro and in vivo [25 , 26 , 49] . Using a murine systemic infection model , we confirmed that SPI1 effectors are present in cells containing large numbers of cytosolic bacteria in vivo . Altogether , our data show that T3SS1 can account for the delivery of T3SS1 effectors at later stages of infection . To the best of our knowledge this is the first time that T3SS1 has been shown to use a translocon-independent mechanism of delivery . Cytosolic Salmonella are not contained within a membrane-bound compartment , have no obvious contact with a membrane and can deliver effectors in the absence of the translocator protein SipB . For both Salmonella T3SS1 and the Shigella flexneri Mxi-Spa T3SS , the translocon proteins are gatekeepers that effectively prevent secretion of effectors in the absence of a specific signal . Mutants lacking translocon proteins constitutively secrete effectors but are unable to invade host cells [51 , 52] . Interestingly , Shigella , which also replicates in the cytosol of epithelial cells , can uncouple translocation and secretion depending on environmental stimuli [53] . Whether the same is true for Salmonella is still to be determined . The Salmonella SPI1 , Shigella Mxi-Spa and Yersinia Ysa T3SSs are members of the Inv/Mxi-Spa subfamily , and other members are found in a number of pathogens including , Burkholderia pseudomallei ( Bsa ) and Chromobacterium violaceum ( Cpi-1/-1a ) [1 , 54] . While these T3SSs generally have key roles in the invasion of non-phagocytic cells [55–58] , several of them also have post-invasion activities that contribute to pathogenesis [34 , 59 , 60] . SPI1 is a paradigm for bacterial adaptation to the host environment [61] . Here , we identified a novel activity of the SPI1-encoded T3SS1 in the cytosol of mammalian epithelial cells . Thus , this highly regulated T3SS mediates host cell interactions in at least two vastly different environments: the lumen of the gut , and the cytosol of epithelial cells . Given the central role of T3SSs at the host-pathogen interface , our findings highlight the need to reassess the contributions of the Salmonella T3SS1 and its effectors in post-invasion events to fully understand their impact on bacterial pathogenesis . Bacterial strains , plasmids and oligonucleotides used in this study can be found in the supporting tables document . All mutants are derivatives of the parental Salmonella Typhimurium strain SL1344 [62] and were constructed using the bacteriophage λ recombinase system [63] . Bacteria were cultured in Miller formulation lysogeny broth ( LB-M ) supplemented with appropriate antibiotic , where necessary . HeLa cells ( human cervical adenocarcinoma , ATCC CCL-2 ) and the Caco-2 subclone C2BBe1 ( human colorectal adenocarcinoma , ATCC CRL-2102 ) were grown at 37°C in 5% CO2 in complete growth medium ( CGM ) : HeLa—MEM supplemented with 10% ( v/v ) heat-inactivated fetal bovine serum ( Invitrogen ) , 2 mM L-Glutamine and 1 mM sodium pyruvate; C2BBe1—DMEM supplemented with 10% ( v/v ) heat-inactivated fetal bovine serum , 4 mM L-Glutamine and human transferrin ( 10 μg/mL ) . Both cell lines were passaged as recommended by ATCC and used within 15 passages of receipt . Plasmids are listed in S1 Table . Oligonucleotides are found in S2 Table . A low copy arabinose-inducible expression plasmid was constructed by excising the araC gene and promoter from pBAD18-Cm using ClaI and HindIII , and cloning into the same restriction sites of pMPMA3ΔPlac . A transcriptional terminator ( Part Bba_B0015 , parts . igem . org ) was subsequently cloned in using the SphI and HindIII sites , resulting in the plasmid pMPMA3ΔPlac PBAD TT . invA or sipB was amplified from SL1344 genomic DNA using oligonucleotides containing a ribosomal binding site and cloned into pMPMA3ΔPlac PBAD TT with NheI and SphI , resulting in the plasmids pPBAD-invA , and pPBAD-sipB , respectively . The sopB-sigE operon was amplified from the plasmid pSopB2xHA in the same manner , for the resulting plasmid pPBAD-sopB-2HA . A cytosolically induced gfp reporter was constructed by transcriptionally fusing the promoter of uhpT to gfp as follows . First , gfpmut3 . 1 was amplified from pFPV25 . 1 and cloned into the XhoI and KpnI sites of pMPMA3ΔPlac , resulting in plasmid pMPMA3ΔPlac-gfp . The uhpT promoter region was amplified from SL1344 genomic DNA and cloned into the NotI and BamHI sites of pMPMA3ΔPlac-gfp . Finally , part Bba_B0015 ( see above ) was cloned into ClaI and XhoI , resulting in the final plasmid pPuhpT-gfp . To construct pPuhpT-sopB-2xHA and the promoterless control plasmid pPNULL-sopB-2xHA , the sopB-2xHA sigE operon was amplified from psopB2xHA and cloned into SphI and HindIII sites of pPuhpT-gfp or into the BamHI and XhoI sites of pMPMA3ΔPlac-gfp , thus replacing the gfp gene in each plasmid . Cells were passaged 16–18 h prior to infection into 24-well tissue culture treated plates . For C2BBe1 cells , coverslips were pre-coated with collagen type I ( BD Biosciences ) . Bacteria were grown under conditions optimizing T3SS1-dependent invasion [18 , 19]: a 2 mL overnight culture was subcultured in 10 mL of LB-Miller broth ( 1:33 dilution , no antibiotics ) with shaking at 37°C for 3 . 5 h . For strains harboring pPBAD-invA , 0 . 02% arabinose was added for the duration of the subculture; for pPBAD-sipB , 0 . 2% arabinose was added for the duration of the subculture; for ΔsopB harboring pPBAD-sopB-2xHA , 0 . 2% arabinose was added for the final 1 h of the subculture . Bacteria were pelleted ( 1 mL , 8 , 000 x g , 2 min ) at room-temperature ( RT ) and re-suspended in an equal volume of Hanks’ Balanced Salt Solution ( HBSS , Mediatech ) . Monolayers were infected at an MOI of ~50 for 10 min at 37°C in 5% CO2 . Extracellular bacteria were removed by washing with HBSS ( x2 ) and cells were incubated in antibiotic-free CGM until 30 min pi . Cells were then incubated for 1 h in CGM supplemented with L-Histidine ( 500 μg/mL ) and gentamicin ( 50 μg/mL ) , followed by CGM supplemented with L-Histidine ( 500 μg/mL ) and gentamicin ( 10 μg/mL ) for the remainder of the infection . HeLa cells were infected according to the ‘Bacterial Infection of Mammalian cells’ section . At the desired time points , wells were lysed in 0 . 2% ( w/v ) sodium deoxycholate to enumerate viable intracellular bacteria . HeLa cells were infected according to the ‘Bacterial Infection of Mammalian cells’ section with the following modification: 1 h prior to sample collection , media in replicate wells were replaced with growth media containing gentamicin and histidine , with or without chloroquine ( 400 μM ) . Monolayers were lysed in 0 . 2% ( w/v ) sodium deoxycholate and serial dilutions plated . Proximity Ligation Assays ( PLA ) were performed using the Red DuoLink In Situ PLA Kit ( Sigma-Aldrich ) . Cells were serum starved ( 0% FBS ) 3h prior to and for the duration of the infection . Infected cells were fixed in 2 . 5% w/v paraformaldehyde ( PFA ) for10 min at 37°C , washed with PBS and stained with Alexa Fluor 647-conjugated wheat germ agglutinin for 10 min at 37°C ( Life Technologies ) . Cells were washed in PBS , fixed in PFA for 5 min at RT , and blocked and permeabilized in 0 . 2% ( w/v ) saponin and 10% ( v/v ) normal donkey serum in PBS ( SS-PBS ) for 30 min at RT . The PLA assay was subsequently performed according to the ‘custom solutions’ protocol of the DuoLink In Situ PLA kit . Antibodies ( goat anti-CSA1 , 1:300; KPL and rabbit phospho-Akt Ser473 ( D9E ) , 1:400; Cell Signaling ) and PLA probes were diluted in SS-PBS . For HeLa cells , phosphorylated Akt PLA signals were manually quantified from maximum intensity projections assembled from 20 slice stacks . For C2BBe1 cells , signals were quantified using CellProfiler ( www . cellprofiler . org ) [64] from maximum intensity projections assembled from 15 slice stacks . Our analysis pipeline involved image thresholding followed by nuclei detection in the DAPI channel . A perinuclear area defined by a 50-pixel extension of the identified nucleus object was used to define this region for PLA signal quantification . PLA signals were identified following image thresholding and related to their respective parent nucleus . At the indicated time points , infected HeLa cells were washed three times with KHM buffer ( 110 mM potassium acetate , 20 mM HEPES , 2 mM MgCl2 , pH 7 . 3 ) , and the plasma membrane selectively permeabilized by incubation with digitonin ( 40 μg/mL in KHM buffer ) for 1 min at RT , followed by three washes with KHM buffer . Cells were then incubated for 12 min at RT with rabbit anti-Calnexin ( Stressgen , 1:250 in KHM ) , to label the cytosolic face of the endoplasmic reticulum in permeabilized cells , and goat anti-Salmonella CSA1 antibodies ( KPL , 1:100 in KHM ) , to detect cytosolic bacteria . Cells were washed in PBS , fixed in PFA and all host cell membranes were permeabilized with SS-PBS . Antibodies delivered post-digitonin permeabilization were detected with Alexa Fluor 647-conjugated anti-rabbit and Alexa Fluor 568-conjugated anti-goat antibodies . After washes with PBS , SopB3xFLAG and SipA effectors were detected using mouse anti-FLAG M2 ( Sigma-Aldrich , 1:250 ) and mouse anti-SipA ( 1:50 ) , respectively . Cells were washed again in PBS and incubated with Alexa Fluor 488-conjugated anti-mouse to detect effector bound antibodies and Pacific Blue-conjugated goat anti-Salmonella CSA1 to label all intracellular bacteria . Coverslips were then washed sequentially with PBS and distilled water , and mounted on glass slides in a Mowiol solution . Infected cells on coverslips were washed with PBS and fixed in either 2 . 5% PFA for 10 min at 37°C or 100% ice-cold methanol for 1 min . After three washes with PBS , cells were blocked and permeabilized with SS-PBS for 30 min at RT , incubated in primary antibodies diluted in SS-PBS for 1 . 5 h at RT , washed sequentially in PBS and saponin-PBS , incubated in Alexa Fluor-conjugated secondary antibodies diluted in SS-PBS for 45 min at RT and washed sequentially in PBS and distilled water . Coverslips were mounted onto glass slides in a Mowiol solution supplemented with 2 . 5% ( w/v ) DABCO . Samples were imaged on a Carl Zeiss LSM 710 confocal laser-scanning microscope equipped with a Plan APOCHROMAT 63X/1 . 4 N . A . objective and assembled into flat maximum-intensity projections using FIJI ( NIH ) or Zen 2012 SP1 software . Figs were assembled using Adobe Photoshop CC . HeLa cells were plated onto 6-well tissue culture treated plates and infected as described above except the monolayers were infected at an MOI of ~100 and serum starved 3h prior to and for the duration of the infection . At the indicated times , monolayers were solubilized in boiling 1 . 5X Laemmli sample buffer and kept at -20°C until ready for analysis . Protein samples were boiled for 10 min at 95°C , separated by SDS-PAGE ( 10% v/v bis-acrylamide gels ) and transferred to nitrocellulose ( 0 . 45μm; Bio-Rad Laboratories ) at 100V for 1 h . Membranes were incubated in blocking buffer ( 5% w/v BSA in Tris-buffered saline/0 . 1% Tween 20; TBST ) for 1 h at RT and incubated overnight at 4°C with primary antibody diluted as recommended by the manufacturer . Membranes were washed three times in TBST and incubated at room temperature for 1 h with appropriate HRP-linked IgG secondary antibody ( Cell Signaling Technology; 1:20 , 000 diluted in blocking buffer ) . Membranes were washed three times in TBST and developed with SuperSignal West Femto Chemiluminescent Substrate ( Thermo Fisher Scientific ) . Seven-week old C57BL/6J mice ( Jackson Laboratories ) were infected by retro orbital i . v . injection with ~500 CFU of Salmonella Typhimurium SL1344 SopB3xFLAG or SipA3xFLAG ( 4 mice per strain ) . Mice were monitored daily for signs of clinical illness . On days 4 and 5 , animals were anesthetized with isofluorane prior to transcardial perfusion with pharmaceutical grade heparin/saline ( 100 U/mL ) , followed by perfusion with 4% ( w/v ) PFA . Gallbladders were harvested and post-fixed in 4% ( w/v ) PFA for 4 h at RT , washed three times with PBS ( 30 min each ) and cryopreserved overnight in 30% ( w/v ) sucrose/PBS at 4°C . The following day , gallbladders were incubated in ‘optimal cutting temperature’ compound ( OCT; Sakura Finetek ) for 15–30 min ( RT ) and 5 μm sections were prepared on Shanodon Positively Charged Superfrost slides ( Thermo Scientific ) using a Leica CM3050S Kryostat . Slides were air-dried and stored at -20°C until staining . Cryosectioned slides were equilibrated to room temperature , the OCT layer removed and washed twice with PBS ( 5–10 min each ) . To quench free aldehyde groups , samples were incubated for 1 h with 0 . 3 M Glycine/PBS , rinsed once with PBS , incubated for 10 min with 50 mM NH4Cl/PBS and washed again in PBS . Samples were incubated in a humidity chamber for 30 min in block solution ( 2% donkey serum , 1% BSA , 0 . 5% Triton X-100 , 0 . 95% Tween-20 , PBS pH 7 . 2 ) prior to overnight incubation at 4°C with primary antibodies . Slides were washed twice in PBS ( 10 min each with rocking ) and incubated with secondary antibody for 1 h at room temperature . Slides were finally washed twice in PBS and mounted in Prolong Gold Anti-fade reagent with DAPI ( Life Technologies ) . Tile scans of murine gallbladder images were acquired at frame size of 3073 x 3072 pixels with 40X EC Plan-Neofluar 40x/1 . 30 N . A . , 0 . 6X digital zoom . Antibodies used were as follows: rat anti-LAMP1 1D4B ( 1:100; Abcam ) , Alexa Fluor 488-conjugated goat anti-CSA1 , Alexa Fluor 568-conjugated mouse anti-FLAG M2 ( concentrations determined by user post-conjugation ) and Alexa Fluor 647 goat anti-rat ( 1:400 ) . SPI1-induced cultures ( 10 mL ) were prepared as detailed in the main Experimental Procedures . 1 mL of culture was pelleted by centrifugation ( 8 , 000 x g , 2 min , RT ) , supernatant removed and the pellet resuspended in 250 μL boiling 1 . 5x Laemmli SDS-PAGE sample buffer , boiled at 95°C for 10 min , and snap-frozen . In parallel , the remaining culture ( ~ 9 mL ) was split between two pre-chilled thick-wall , polycarbonate ultra-centrifuge tubes ( Beckman Coulter #355647 ) on ice and centrifuged at 30 , 000 x g in a MLA-80 rotor in an Optima Max Ultracentrifuge ( Beckman Coulter; 20 min , 4°C ) . Supernatant was carefully removed , filtered through a 0 . 2 μm PES low-protein binding filter ( GE Healthcare ) and proteins were precipitated overnight at 4°C in 10% ( v/v ) trichloroacetic acid . Precipitated sample was divided into pre-chilled 2 mL snap-cap tubes and centrifuged for 20 min at 16 , 000 x g ( 4°C ) . Pellets were washed once with ice-cold acetone , centrifuged as above and allowed to dry for 5–10 min in a fume hood . Pellets were resuspended in a total volume of 200 μL pre-heated 1 . 5X Laemmli sample buffer ( 95°C ) , snap frozen and stored at -80°C until use . Aliquots of 10 μL were analyzed by SDS PAGE to determine secretion of relevant effector proteins . Antibodies used for protein detection were as follows: mouse anti-FLAG M2 ( 1:1000; Sigma-Aldrich ) and mouse anti-DnaK ( 1:20 , 000; Enzo Lifesciences ) . Cytotoxicity assays were performed using the colorimetric CytoTox 96 Non-Radioactive Cytotoxicity assay kit ( Promega ) . HeLa cells were infected as described above except that cells were serum starved ( 0% FBS ) 3 h prior to and for the duration of the infection . Incubations from 0 . 5 hpi were carried out with the addition of 0 . 2% BSA to minimize the cytotoxic effects of serum starvation . At 6 hpi , supernatants from sample wells were collected and assayed for lactate dehydrogenase release following manufacturer’s instructions . Absorbance at 492 nm was measured with a Tecan Infinite M200 Pro . HeLa cells were plated and infected in black glass-bottom 24-well plates ( Grenier BioOne ) according to the ‘Bacterial Infection of Mammalian cells’ protocol . When appropriate , cells were treated with either 50 μg/mL AF 488-dextran ( Invitrogen ) 20h prior to infection or 250 ng/mL propidium iodide ( ImmunoChemistry Technologies ) 1 . 5h post-infection . Between 3 and 10 hpi , images were acquired every 5 minutes using a Nikon TiE spinning disc confocal microscope ( CSU10 Yokogawa ) with Perfect Focus , Cascade II CCD camera ( Photometrics ) , and custom laser launch ( Prairie Technologies ) . All imaging was preformed within a stage-top incubation chamber ( Pathology Devices ) at 37°C , 75% humidity , and 5% CO2 . Wells were imaged using a Plan Fluor 40X 0 . 75 N . A . Ph2 air objective . All post-acquisition image analysis was done using ImageJ software ( W . S . Rasband , National Institutes of Health , Bethesda , MD version 2 . 0 . 0 ) and Adobe Photoshop ( CS5 v12 . 1 Adobe ) . Unless otherwise stated in the figure legend , data were analyzed for statistical significance by a one-way analysis of variance ( ANOVA ) with Bonferroni’s post hoc test . A P-value of ≤ 0 . 05 was considered significant . * = P ≤ 0 . 05 , = P ≤0 . 01 , * = P ≤0 . 001 , ns = not-significant/ P > 0 . 05 . All animal work at Rocky Mountain Laboratories adhered to the U . S . Government Principles and applicable humane and ethical policies in accordance with the Public Health Service ( PHS ) policy , the Guide for Care and Use of Laboratory Animals and the Animal Welfare Regulations . The Rocky Mountain Laboratories ACUC reviewed and approved this research ( ASP# 2014–028 ) .	Non-Typhoidal Salmonella are important agents of food borne disease worldwide . These facultative intracellular bacteria use a specialized Type III Secretion ( T3SS1 ) system to invade intestinal epithelial cells . Effector proteins translocated by this system across the eukaryotic plasma membrane induce actin rearrangements and target signaling pathways . One such effector is SopB , which contributes to invasion and mediates activation of the pro-survival kinase Akt . Within epithelial cells , Salmonella survive and replicate within a modified phagosome , known as the Salmonella-containing vacuole , or the host cell cytosol . Here , we investigated the post-invasion role of T3SS1 in epithelial cells , using SopB as a model effector . SopB-dependent Akt phosphorylation was observed at two distinct stages of infection: during and immediately after invasion , and later during peak cytosolic replication . SopB delivery by cytosolic Salmonella required T3SS1 but was translocon-independent . This was also observed for another T3SS1 effector , SipA , indicating that T3SS1 effectors may be secreted directly into the cytosol . Infection with a SopB deletion mutant eliminated the induction of Akt phosphorylation and decreased the lifespan of infected cells . These effects were reversed by expressing SopB specifically in cytosolic bacteria , confirming a role for SopB and T3SS1 during the cytosolic stage of infection . Thus , T3SS1 has two temporally distinct roles during epithelial cell colonization .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "bacteriology", "medicine", "and", "health", "sciences", "pathology", "and", "laboratory", "medicine", "intracellular", "pathogens", "hela", "cells", "pathogens", "biological", "cultures", "microbiology", "salmonellosis", "epithelial", "cells", "bacterial", "diseases", "secretion", "systems", "enterobacteriaceae", "cell", "cultures", "bacteria", "bacterial", "pathogens", "research", "and", "analysis", "methods", "specimen", "preparation", "and", "treatment", "infectious", "diseases", "staining", "microbial", "physiology", "animal", "cells", "medical", "microbiology", "microbial", "pathogens", "biological", "tissue", "cell", "lines", "salmonella", "bacterial", "physiology", "cell", "staining", "cell", "biology", "anatomy", "virulence", "factors", "epithelium", "biology", "and", "life", "sciences", "cellular", "types", "cultured", "tumor", "cells", "cytosol", "organisms" ]	2017	A second wave of Salmonella T3SS1 activity prolongs the lifespan of infected epithelial cells
Secondary metabolites , including toxins and melanins , have been implicated as virulence attributes in invasive aspergillosis . Although not definitively proved , this supposition is supported by the decreased virulence of an Aspergillus fumigatus strain , ΔlaeA , that is crippled in the production of numerous secondary metabolites . However , loss of a single LaeA-regulated toxin , gliotoxin , did not recapitulate the hypovirulent ΔlaeA pathotype , thus implicating other toxins whose production is governed by LaeA . Toward this end , a whole-genome comparison of the transcriptional profile of wild-type , ΔlaeA , and complemented control strains showed that genes in 13 of 22 secondary metabolite gene clusters , including several A . fumigatus–specific mycotoxin clusters , were expressed at significantly lower levels in the ΔlaeA mutant . LaeA influences the expression of at least 9 . 5% of the genome ( 943 of 9 , 626 genes in A . fumigatus ) but positively controls expression of 20% to 40% of major classes of secondary metabolite biosynthesis genes such as nonribosomal peptide synthetases ( NRPSs ) , polyketide synthases , and P450 monooxygenases . Tight regulation of NRPS-encoding genes was highlighted by quantitative real-time reverse-transcription PCR analysis . In addition , expression of a putative siderophore biosynthesis NRPS ( NRPS2/sidE ) was greatly reduced in the ΔlaeA mutant in comparison to controls under inducing iron-deficient conditions . Comparative genomic analysis showed that A . fumigatus secondary metabolite gene clusters constitute evolutionarily diverse regions that may be important for niche adaptation and virulence attributes . Our findings suggest that LaeA is a novel target for comprehensive modification of chemical diversity and pathogenicity . Aspergillus fumigatus is a saprophytic filamentous fungus with no known sexual stage . Prolific production of asexual spores ( conidia ) and nearly ubiquitous distribution in the environment ensures constant host exposure to its spores , at a density of 1 to 100 conidia/m−3 [1] . The innate immune system enables spores to be eliminated from lung epithelial tissue with ease in immunocompetent vertebrates . However , immunocompromised individuals are at risk for pulmonary disease as a consequence of A . fumigatus infection . Of particular concern is invasive aspergillosis , which occurs when hyphal growth proliferates throughout pulmonary or other tissues . Invasive aspergillosis has an associated mortality rate ranging from 50% to 90% depending on the patient population [2] . As the number of immunocompromised patients has increased in recent decades due to immunosuppressive chemotherapy treatments , HIV/AIDS , and solid organ and bone marrow transplantation , the incidence of invasive aspergillosis has increased more than 4-fold in developed nations [2] . Several A . fumigatus secondary metabolites or natural products ( e . g . , conidial melanins and mycotoxins ) have been implicated as affecting virulence [3–7] . However , the exact mechanisms by which many of these compounds might affect disease outcome are unknown , nor is it clear in most cases whether these factors play direct or indirect roles in pathogenicity . In contrast to most genes involved in primary metabolism , genes encoding secondary metabolite biosynthetic enzymes exist in contiguous clusters within the genome [8 , 9] . LaeA was originally identified as a transcriptional regulator of secondary metabolite gene clusters in Aspergillus nidulans and A . fumigatus [10 , 11] , including gliotoxin in the latter . Gliotoxin has long been suggested to be a major virulence attribute in invasive aspergillosis [12–14] . However , whereas a ΔlaeA mutant shows reduced virulence in a mouse model of invasive aspergillosis [11] , inactivation of gliotoxin biosynthesis alone does not [15–17] . Therefore , we reasoned that because LaeA is a transcriptional regulator , perhaps acting at a chromatin remodeling level [9 , 18] , a microarray experiment comparing the transcriptomes of ΔlaeA , wild-type , and complemented ΔlaeA control strains would yield further insight into LaeA-mediated A . fumigatus virulence attributes . We uncovered an unprecedented view of LaeA global regulation of mycotoxin islands , nearly all found in nonsyntenic regions of the Aspergillus genome . Because secondary metabolite gene cluster regions are evolutionarily diverse and may affect virulence attributes , LaeA is a novel target for comprehensive modification of chemical diversity . Transcriptional profiles of the wild-type , ΔlaeA , and ΔlaeA complemented strain were determined by comparisons of relative transcript levels between ( 1 ) ΔlaeA versus wild-type and ( 2 ) wild-type versus complemented control strain . All strains were grown under identical conditions ( 25 °C , liquid shaking culture , glucose minimal media , 60 h ) for three biological replicates . The condition and time point were chosen on the basis of optimal production of secondary metabolites [10 , 11] . The comparison of ΔlaeA versus wild-type was used to determine gene expression patterns specific to the ΔlaeA mutant , while the wild-type versus complemented strain comparison was conducted as a control , because the difference between these two strains is the presence of an ectopic copy of a selectable marker for hygromycin resistance . The processed signal intensity ratios for the three ΔlaeA versus wild-type replicates were analyzed using the significance analysis of microarrays ( SAM ) method [19] , as described in Materials and Methods . In total , 943 genes were significantly differentially expressed . Figure S1 shows a heat map of a subset of these loci , depicting normalized expression ratios for the three ΔlaeA versus wild-type experiments and the three wild-type versus complemented control experiments . The high quality of the data is indicated by the consistency of color between the replications and the relative lack of color in the control lanes . Of the 943 genes showing significant differences in expression between ΔlaeA and wild-type by SAM analysis , 415 showed increased expression in ΔlaeA and 528 showed decreased expression . Table 1 and Figure S2 indicate functional categories for these genes ( defined as described by the Gene Ontology Consortium , http://www . geneontology . org ) . The most remarkable discovery was the near-global suppression of secondary metabolite gene expression in the ΔlaeA mutant . Nearly all ( 97% ) of the secondary metabolite gene cluster loci showed decreased expression in ΔlaeA , with a mere three genes in this category showing increased expression in ΔlaeA . This was in contrast to all other functional categories , which showed substantial proportions of both increased and decreased expression in the mutant , possibly reflecting indirect effects due to loss of production of multiple metabolites . In addition to genes with unknown function ( 39% ) and genes involved in secondary metabolism ( 11% ) , other major categories included genes encoding proteins involved in transmembrane transport ( 8% ) and those involved in information processing ( 4% ) , and cell wall biogenesis ( 4% ) . Statistical analysis of the overrepresentation of different Gene Ontology categories and Pfam protein domains within the set of 943 differentially regulated genes is shown in Tables S1 and S2 , respectively . Interestingly , LaeA appeared to influence expression of a subset of species- and lineage-specific genes not strongly conserved with other fungal species . Only 18% and 44% of all genes significantly differentially expressed in the mutant have putative orthologs in Saccharomyces cerevisiae and Neurospora crassa , respectively , compared to an average of 33% and 58% of all A . fumigatus genes . Many , but not all , of these genes were classified as secondary metabolism genes . Moreover , there are about 120 differentially expressed genes; again , most , but not all , are present in secondary metabolism clusters ( Table 2 ) , which have no detectable orthologs in Aspergillus oryzae and A . nidulans . Considering this overwhelming tight and directed transcriptional control of secondary metabolite loci by LaeA , below we focus on such genes as possible members of the LaeA-regulated A . fumigatus pathogenicity arsenal . Although initial genome analysis suggested the presence of 26 secondary metabolite gene clusters [20] , subsequent analysis ( G . Turner , N . D . Fedorova , V . Joardar , J . R . Wortman , and W . C . Nierman , unpublished data ) has provided support for only 22 clusters . Of the 13 secondary metabolite gene clusters whose expression was influenced by LaeA in the condition used for microarray analysis , ten are particularly strongly affected , with a majority of genes within these clusters being significantly down-regulated in ΔlaeA as indicated by SAM . Three additional clusters have at least one gene encoding a critical enzyme such as a nonribosomal peptide synthetase ( NRPS ) or a polyketide synthase showing decreased expression in ΔlaeA . Additionally , 38% ( 23 of 71 ) of all P450 monooxygenases show differential expression in ΔlaeA , also associated with secondary metabolite biosynthesis and/or detoxification . Fifteen of these genes encoding P450 monooxygenases are found in secondary metabolite gene clusters . Table S3 gives normalized expression ratio values for all 22 gene clusters in A . fumigatus . Table 2 summarizes the current state of knowledge regarding function of LaeA-regulated secondary metabolite gene clusters . These include clusters dedicated to production of conidial melanins , fumitremorigens , gliotoxin , and ergot alkaloids such as festuclavine , elymoclavine , and fumigaclavines A , B , and C ( Table 2 ) [4 , 17 , 21–27] . Figure 1 depicts the chromosomal landscape of those regions most strongly regulated by LaeA . To confirm these microarray results , quantitative real-time reverse-transcription ( RT ) -PCR ( QRT-PCR ) was performed on one major class of secondary metabolite genes , those encoding NRPSs [28] . As indicated in Table 3 , relative expression levels for NRPSs that showed differential expression between mutant and wild-type in the microarray study also were dramatically reduced upon QRT-PCR analysis . In all cases , complementation of the laeA defect restored NRPS gene expression to wild-type levels ( Table 3 ) . Notably , because the microarray analysis determines only relative expression and not absolute levels of transcript , we could not conclude whether secondary metabolite clusters not showing differential expression in mutant versus wild-type were not affected by LaeA or were simply not induced under the growth condition used . To further examine these possibilities , we assessed the expression of a subset of NRPSs thought to encode siderophore-biosynthesizing enzymes . Although siderophores do not fit neatly into a definition of secondary metabolites , which are dispensable in laboratory growth conditions [29] , these molecules are produced from clustered genes and are critical for pathogen growth in blood serum [30] . Because iron was included in the media used for the microarray study , we investigated whether the Δlae mutant was deficient in expression of siderophore gene cluster NRPSs under iron-limiting conditions . As previously reported [29] , low iron conditions induced transcriptional upregulation of several NRPS genes known or predicted to be involved in siderophore biosynthesis ( Table 4 ) . Normalized expression levels of the siderophore NRPSs in the low iron condition relative to high iron conditions were NRPS2/sidC , 2 . 245 ± 0 . 449; NRPS3/sidE , 68 . 595 ± 13 . 725; and NRPS4/nps6/sidD , 28 . 509 ± 4 . 704 . NRPS7 transcripts were not detectable in these experiments . In contrast to Reiber et al . [29] , NRPS3/sidE showed the highest induction to the low iron conditions in our experiments . This discrepancy might be explained by differential sensitivity of the semiquantitative RT-PCR method used by Reiber et al . compared to our QRT-PCR methodology or subtle differences in culture conditions . We also noted that the complemented control strain with an ectopic copy of laeA showed increased expression of NRPS3/sidE in both low and high iron conditions . Comparison of the ΔlaeA mutant and controls by QRT-PCR analysis indicated differential expression of NRPS3/sidE in low ( inducing ) iron conditions . In high iron conditions , NRPS4/nps6/sidD , NRPS3/sidE , and possibly NRPS2/sidC showed decreased expression in the ΔlaeA mutant ( Table 4 ) . Interestingly , NRPS7 was not detectable in these experiments . In the low iron condition , expression of actin did decrease in the ΔlaeA mutant . However , the dramatic decrease in expression of NRPS3/sidE ( 1 , 000-fold less ) seen in the ΔlaeA background strongly suggests that LaeA regulates the expression of at least this NRPS . Little is known about the function of SidE , although it has been speculated to be involved in siderophore biosynthesis on the basis of homology to SidC [29] . It remains to be determined whether NRPS3/sidE is involved in siderophore production , a process known to be critical to virulence [31 , 32] , or whether it synthesizes an iron-responsive compound with a distinct function . Regardless of the function of SidE , these experiments show that LaeA is also involved in controlling expression of other secondary metabolite clusters not induced by the environmental conditions used in the microarray experiments . Cluster 18 ( Figure 1 ) on Chromosome 6 , strongly differentially expressed in ΔlaeA , encodes the genes required for gliotoxin biosynthesis . Gliotoxin is arguably the most well-studied mycotoxin produced by A . fumigatus . First identified in 1936 , this compound has immunosuppressive properties in vitro [12] and in vivo [13 , 14] , although its direct contribution to pathogenicity is only beginning to be understood [15–17] . Like all other compounds in the epipolythiodioxopiperazine class , gliotoxin is a cyclic dipeptide with an internal disulfide bridge that can undergo redox cycling ( for a recent review , see [33] ) . Immunosuppressive activity of gliotoxin is due at least in part to negative regulation of the transcription factor nuclear factor–κB , which occurs by inhibition of proteasome-mediated degradation of the nuclear factor–κB inhibitor IκBα [34 , 35] . Gliotoxin is also known to be cytotoxic and can evoke both apoptotic [36–39] and necrotic [40 , 41] cell death . Recently , gliotoxin was shown to trigger the release of apoptogenic factors by the host mitochondrial protein Bak [42] . The secondary metabolism gene cluster responsible for gliotoxin production was recently identified by bioinformatic analysis [43] and has been experimentally confirmed [15 , 17] . Despite the known immunosuppressive activities of the molecule and its detection in blood serum of patients with invasive aspergillosis [44] , three recent studies using genetic mutants of the gliotoxin gene cluster demonstrated that gliotoxin is not a virulence factor in murine models of invasive aspergillosis [15–17] . However , these same studies presented evidence that gliotoxin could adversely affect T cells , neutrophils , and mast cells and , we offer , likely acts synergistically with other LaeA-regulated toxins . The ΔlaeA mutant is impaired in gliotoxin production during growth in culture as well as growth in vivo in murine models of invasive aspergillosis [10 , 15] , and the microarray results presented here confirm that LaeA strongly influences expression of genes in this cluster under the condition investigated . Secondary metabolite cluster 1 on Chromosome 1 , which is differentially expressed in ΔlaeA , contains an atypical NRPS called Afpes1 that is required for virulence in an insect model of invasive aspergillosis [4] . Afpes1 shows greatest homology to NRPSs that produce siderophores or destruxins , including one paralog required for virulence of the plant pathogen Alternaria brassicae [45] . However , the Afpes1 cluster is thought to be unlikely to produce either of these compounds , because destruxin toxin has not been detected in A . fumigatus [4] and expression of Afpes1 was not responsive to iron levels [4 , 21] . Deletion of Afpes1 alters conidial morphology and hydrophobicity as well as melanin synthesis and results in increased susceptibility to reactive oxygen species , implying altered conidial melanin and/or rodlet composition [4] . Most of these characteristics are common to the ΔlaeA phenotype [11] , possibly implicating a role of the Afpes1 metabolite in the attenuated virulence of ΔlaeA . A . fumigatus synthesizes several clavine ergot alkaloids , compounds that can be partial agonists or antagonists of serotonin , dopamine , and α-adrenalin receptors , thus affecting nervous , circulatory , reproductive , and immune system function [46] . The role of these compounds in invasive aspergillosis has not been determined . In addition to having the receptor-modulating activities mentioned , the festuclavine ergot alkaloid produced by A . fumigatus is cytostatic and is directly mutagenic in the Ames assay [47 , 48] . Recently , Coyle and Panaccione [25] showed that deletion of an A . fumigatus dimethylalleletryptophan synthase ( DMAT synthase ) homologous to dmaW of the ergot-producing species Claviceps purpurea eliminated all known ergot alkaloids , confirming its predicted function in the first committed step of ergot alkaloid production ( i . e . , addition of dimethylallyl diphosphate to l-tryptophan to result in 4-methylallyl-tryptophan ) . The biochemical activity of the A . fumgiatus DmaW enzyme was also confirmed by Unsöld and Li [22] , who subsequently characterized a reverse prenyltransferase in the same gene cluster that converts fumigaclavine A to fumigaclavine C [23] . These genes are located in secondary metabolite gene cluster 4 on Chromosome 2 , which is strongly differentially expressed in ΔlaeA . Melanins found in conidia are one of the few described virulence factors in A . fumigatus [5 , 6 , 24] . Lack of melanins leads to increased susceptibility to reactive oxygen species produced by the host innate immune response during infection as well as altered ( smooth ) conidial morphology [5 , 7] . However , the scarcity of nonpigmented A . fumigatus spores in nature has drawn into question the clinical relevance of melanins as virulence factors [1] . Conidia of ΔlaeA are pigmented , but altered expression of alb1 in the mutant has been reported previously and at least one unidentified spore metabolite is missing in ΔlaeA [11] . There is significant differential expression of the 1 , 8-dihydroxynapthalene–melanin gene cluster in ΔlaeA under the condition investigated in this study . Expression of this gene cluster is also regulated by cAMP/protein kinase A signaling [49] as is LaeA itself [10] , perhaps a suggestion that in this case LaeA control of this cluster may be both directly and indirectly mediated by protein kinase A . Additionally , a LaeA-regulated supercluster on Chromosome 8 is likely to produce multiple compounds . Recently , two genes in this cluster have been reported to encode biosynthetic enzymes for the tremorgenic mycotoxin fumitremorgin B and related compounds [26 , 27] . The cyclo-l-Trp-l-Pro derivative fumitremorgin B is cytotoxic , inhibiting cell cycle progression at G2/M , and thus has been of interest as a potential anticancer agent . The pathway involves generation of the cyclic dipeptide brevianamide F by the NRPS brevianamide synthetase [27] , prenylation of brevianamide F by the prenyltransferase FtmPT1 to tryptrostatin B [26] , and subsequent conversion in several steps to fumitremorgen B . Thus , LaeA-mediated influence on expression of ftmPT1 and ftmPT2 would govern the production of this entire class of diketopiperozine compounds . Once again , however , the specific effects of these compounds on pathogenicity during invasive aspergillosis are unknown . The fact that LaeA promotes expression not only of these secondary metabolite gene clusters but an additional eight others confirms its role as a master controller of secondary metabolism . The importance of several of these compounds in toxicity studies also underscores relevance of LaeA during infection [11] . We suggest the possibility that virulence attributes are not influenced as much by individual metabolites as by the blend of LaeA-regulated toxins , which , in combination , may confer an advantage to the pathogen . Comparative genomic analysis between A . fumigatus and related species indicates overlap between A . fumigatus–specific genes and genes differentially expressed in ΔlaeA ( N . D . Fedorova and W . C . Nierman , unpublished data ) . In total , 68% of A . fumigatus secondary metabolite genes do not have orthologs in the closely related species A . clavatus ( N . D . Fedorova and W . C . Nierman , unpublished data ) . Additional secondary metabolite genes do not have orthologs in more distantly related Aspergilli such as A . oryzae and A . nidulans [20] . The variability of secondary metabolite clusters may be explained by the fact that many of them are located in highly divergent telomere-proximal regions characterized by frequent chromosomal rearrangements [20 , 50] . For example , 54% of the clusters showing differential expression in ΔlaeA in the conditions described here were found within 300 kb of telomeres . It should be noted that , in addition to the secondary metabolite clusters , other genes with significantly lower expression in ΔlaeA also show some positional specificity within the genome but to a much lesser extent ( unpublished data ) . Further analysis also showed that A . fumigatus telomere-proximal clusters tend to have larger numbers of genes than clusters located closer to the centromeres , suggesting that the former may accumulate additional genes more easily ( N . D . Fedorova , J . R . Wortman , and W . C . Nierman , unpublished data ) . Initial comparative genome analyses indicate that the telomere-proximal regions ( and to a lesser extent , synteny breakpoints and intrasyntenic regions ) appear to be a hotbed of diversity , not only between Aspergillus species but even between different strains of the same species [51 , 52] . The genomes of two A . fumigatus strains have been sequenced: the clinical isolate Af293 ( by The Institute for Genomic Research , Rockville , Maryland , United States ) and isolate CEA10 ( under contract from Elitra Pharmaceutical by Celera Genomics and made available by Merck; B . Jiang and W . C . Nierman , personal communication ) . These strains show an overall divergence of 2% , and the majority of this variation is in telomere-proximal and synteny breakpoint regions . Similarly , microarray experiments also supported high divergence in these regions when Af293 was compared to the unsequenced A . fumigatus strains Af294 and Af71 [20] . Many secondary metabolite clusters appear to be associated with transposons ( known to be active in A . fumigatus [51] ) and transposase-like sequences ( Table S3 ) . The finding that these transposable elements often flank or are embedded in many of the clusters may represent one mechanism for generating the diversity of secondary metabolites in aspergilli . Whether or not there is a connection between LaeA function and transposon activity has yet to be established . In total , these analyses suggest that secondary metabolite clusters are located in the regions that undergo extensive rearrangements , which may result in subsequent alterations in secondary metabolite production and , therefore , have major impacts on niche adaptation between different species of fungi or between strains of the same species . Other examples include a non–aflatoxin-producing clade of A . flavus , better known as the food-fermenting A . oryzae used in the production of traditional Asian products such as miso and soy sauce , which may have arisen as a result of telomere-proximal rearrangements [53] . Similarly , genotypic variability between strains of Fusarium compactum also proved to be a major determinant of metabolite production and geographic distribution [54] . In Fusarium graminearum , the major cause of wheat and barley head blight , intraspecific polymorphic variations in a trichothecene mycotoxin gene cluster were correlated with chemotype differences , host range , and fitness [55] . In light of such examples , it is interesting to speculate about the role of LaeA in chemotype evolution and niche adaptation . It is possible that variation at any particular secondary metabolite gene cluster could result in less efficient control by LaeA . This potential has been demonstrated in A . nidulans [18] . Conversely , LaeA itself is a major target for comprehensive changes in the entire complement of secondary metabolites . The clustering of secondary metabolite biosynthetic genes has been suggested to reflect their evolutionary history [8 , 9 , 20 , 51 , 56 , 57] . Several models have been proposed to explain the establishment and maintenance of secondary metabolic gene clusters in filamentous fungi . The “selfish cluster” hypothesis proposes that selection occurs at the level of the cluster and promotes maintenance of the cluster as a unit , possibly through horizontal transfer events [56] . However , there is only limited evidence for widespread horizontal transfer of secondary metabolism gene clusters , with penicillin being a notable exception [58] . Alternative models suggest that clusters are maintained due to coregulation mechanisms , likely at the level of chromatin regulation [8 , 9] . LaeA may provide a mechanistic means of secondary metabolism gene cluster coregulation and maintenance . Certainly LaeA demonstrates a positional bias for local gene regulation , as transfer of genes into or out of a secondary metabolite cluster leads to respective gain or loss of transcriptional regulation by LaeA [18] . This has been speculated to occur through regulation of nucleosome positioning and heterochromatin formation [9] . Our results confirm that LaeA plays a central role in regulation of chemical diversity in A . fumigatus . Furthermore , genomic regions that are transcriptionally controlled by LaeA are species and even strain specific , suggesting that they may serve as niche adaptation factors . The loss of laeA results in a great decrease in repertoire of secondary metabolites , which appears to impact the infection process . Therefore , LaeA constitutes a novel target for the production of an array of factors critical to success during pathogenesis . Furthermore , LaeA is a tool to identify metabolite gene clusters that may impact virulence , allowing the correlation of specific secondary metabolite clusters with virulence even in absence of knowledge about the mycotoxin itself . Three prototrophic A . fumigatus fungal strains were used in this study . Af293 ( the wild-type clinical isolate used in the A . fumigatus genome sequencing project [20] ) , TJW54 . 2 ( ΔlaeA ) [11] , and a complemented control strain TJW68 . 6 ( ΔlaeA + laeA ) [11] were grown in triplicate at 25 °C in liquid minimal media [59] with shaking ( 280 rpm ) for 60 h . Profiles of secondary metabolites extracted from the media with chloroform were compared by thin-layer chromatography , and the results confirmed that the ΔlaeA strain showed reduced levels of multiple secondary metabolites under this condition ( [10] and unpublished data ) . Total RNA was isolated from fungal mats , labeled , and hybridized with a DNA whole-genome amplicon microarray [20 , 60] in three independent biological replicates . To analyze siderophore NRPS gene expression under low- or high-iron conditions , 50-ml liquid cultures were grown as described [29] , with low-iron media containing 25 g/L glucose , 3 . 5 g/L ( NH4 ) 2SO4 , 2 . 0 g/L KH2PO4 , 0 . 5 g/L MgSO4 ( heptahydrate ) , and 8 mg/L ZnSO4 ( heptahydrate ) ( pH 6 . 3 ) . High-iron media was identical except for the addition of Fe ( III ) Cl3 to a final concentration of 300 μM . Cultures were grown at 37 °C , 280 rpm , and samples were collected at 24 h postinoculation . All glassware was subjected to sequential treatment with 1 mM and 5% HCl as described [29] . Total RNA was extracted from Aspergillus strains by use of TriZOL reagent ( Invitrogen , http://www . invitrogen . com ) according to the manufacturer's instructions . RNA was further purified by two extractions with phenol:chloroform:isoamyl alcohol ( 25:24:1 ) and then labeled with Cy-3 or Cy-5 dye and hybridized as previously described [20] . The generation of the whole genome array has been described [20] . QRT-PCR was used to ( 1 ) confirm the expression level trends observed in the microarray experiment and ( 2 ) investigate NRPS gene expression under iron-limiting conditions . Expression of select NRPSs putatively regulated by LaeA was examined . Total RNA from two or three biological replicates was pooled in equal amounts ( 2 μg per sample ) for each Aspergillus strain , wild-type AF293 , TW54 . 2 , and TW68 . 6 , and treated with Ambion Turbo DNA-free DNase I ( Ambion , http://www . ambion . com ) to remove contaminating genomic DNA . A total of 500 ng of DNase I–treated total RNA from each sample was reverse transcribed with Superscript III reverse transcriptase ( Invitrogen ) . Real-time RT-PCR was conducted with 20-μl reaction volumes with the iQ SYBR green supermix ( Bio-Rad , http://www . bio-rad . com ) , 2 μl of a 1:6 dilution of first-strand cDNA , and 0 . 4 μl of each 10 μM primer stock . Primer sequences were previously reported [28] . No reverse transcriptase controls ( NRT ) were used to confirm elimination of contaminating genomic DNA . Real-time RT-PCR was performed using an iQ Cycler Real-Time PCR detection system ( Bio-Rad ) . PCRs for each NRPS were done in triplicate and melt curve analysis was performed immediately following the PCR to confirm the absence of nonspecific amplification products and primer dimers . The relative expression levels of NRPS genes between A . fumigatus wild-type strain AF293 , the ΔlaeA mutant , and the complemented control strain were calculated using 2−ΔΔCt method with iQ cycler system software . All values were normalized to expression of the A . fumigatus actin gene and relative to the wild-type strain for each condition analyzed . Gene expression ratios were determined for triplicate comparisons of ( 1 ) wild-type and ΔlaeA and ( 2 ) ΔlaeA and the complemented control strain . Prior to statistical analysis , the LOWESS normalization method was used to remove any systematic bias from the raw expression ratios [61] . Loci showing significantly different expression were identified using the SAM method for one-class designs that has been previously described in detail [19] , implemented in the TM4 suite's MultiExperiment Viewer ( http://www . tm4 . org ) [62 , 63] . This allowed identification of genes whose mean expression across experiments is significantly different from a user-specified mean ( log2 = 0 , corresponding to identical mRNA levels in the mutant and wild-type strains ) . Genes with scores above the significance threshold and exceeding the cutoff value of zero for the false discovery rate ( the most conservative setting ) were designated as significantly differentially expressed between mutant and wild-type . The delta value cutoff in SAM was chosen to capture the maximum number of significant genes while maintaining the reported estimated false discovery rate at zero . Genes down-regulated in ΔlaeA were further analyzed by the Expression Analysis Systematic Explorer ( EASE ) [64] within TM4 to identify overrepresented Gene Ontology terms and Pfam domains . Fisher's exact test probabilities and step-down Bonferroni corrected probabilities are reported from the EASE analysis to indicate which terms are overrepresented in the down-regulated gene set .	Patients with suppressed immune systems due to cancer treatments , HIV/AIDS , or organ transplantation are at high risk of infection from microbes . Some of the most deadly infections for such patients arise from a fungal pathogen , Aspergillus fumigatus . This species , like several of its close relatives , can produce an array of small chemical compounds that influences both the infection process and its environmental niche outside of the host . The genes dedicated to production of each compound are clustered adjacent to each other in the genome . One protein named LaeA is a master regulator of such clustered small molecule genes , and removal of the gene encoding LaeA cripples the organism's ability to infect . We conducted a genome-wide microarray experiment to identify small molecule gene clusters controlled by the presence of LaeA in A . fumigatus . In doing so , we identified actively expressed gene clusters critical for small molecule production and potentially involved in disease progression . These results also provide insight into evolutionary events shaping the organism's collection of chemical compounds .	[ "Abstract", "Introduction", "Results/Discussion", "Materials", "and", "Methods" ]	[ "secondary", "metabolism", "aspergillus", "fumigatus", "mycotoxin", "fungal", "immunology", "microbiology", "computational", "biology", "pathogenicity", "gene", "expression", "regulation", "genetics", "and", "genomics" ]	2007	Transcriptional Regulation of Chemical Diversity in Aspergillus fumigatus by LaeA
Populations of genetically identical eukaryotic cells show significant cell-to-cell variability in gene expression . However , we lack a good understanding of the origins of this variation . We have found marked cell-to-cell variability in average cellular rates of transcription . We also found marked cell-to-cell variability in the amount of cellular mitochondrial mass . We undertook fusion studies that suggested that variability in transcription rate depends on small diffusible factors . Following this , in vitro studies showed that transcription rate has a sensitive dependence on [ATP] but not on the concentration of other nucleotide triphosphates ( NTPs ) . Further experiments that perturbed populations by changing nutrient levels and available [ATP] suggested this connection holds in vivo . We found evidence that cells with higher mitochondrial mass , or higher total membrane potential , have a faster rate of transcription per unit volume of nuclear material . We also found evidence that transcription rate variability is substantially modulated by the presence of anti- or prooxidants . Daughter studies showed that a cause of variability in mitochondrial content is apparently stochastic segregation of mitochondria at division . We conclude by noting that daughters that stochastically inherit a lower mitochondrial mass than their sisters have relatively longer cell cycles . Our findings reveal a link between variability in energy metabolism and variability in transcription rate . Genetically identical populations of cells can exhibit cell-to-cell variations in the amount of individual gene products; this can result in phenotypic diversity [1] , [2] . The study of cellular variability was pioneered by Delbrück in the mid-forties , who measured differences in the number of phages produced by individual Escherichia coli [3] . Fluctuations in the small numbers of molecules involved in gene expression have been indicated as a source of this variation , and current experimental and theoretical approaches seek to anatomize the potential sources of variability , or “noise” . Variation between cells could be due to global factors such as cell cycle position or differences in numbers of transcription factors . Such changes can affect all genes and so constitute “extrinsic” sources of variability . In contrast , “intrinsic” noise is identified as molecular variation that occurs at the level of single genes and their products [4] . Cell-to-cell variability could be mainly the combined effect of large amounts of intrinsic variation or might be attributable to more system-wide extrinsic variation . In the following we investigate how global factors can influence transcription rate across the eukaryotic cell . Experiments investigating gene expression noise suggest that gene expression variability has a mix of intrinsic and extrinsic sources [5] , [6] . Intrinsic noise has been modelled extensively , and we have a relatively refined idea of its origin in the molecular machinery of transcription , translation , and degradation [1] , [2] , [7] , [8] . The magnitude of extrinsic noise is largest at intermediate levels of gene expression and dominates when gene expression is high [6] , [7] , [9] . However , the sources of extrinsic noise are not as well characterised as those of intrinsic noise [7] , [10] . Studies carried out in yeast have , for example , suggested cell size , cell shape , cell cycle stage , and fluctuations in an as yet unidentified upstream regulator as potential sources of extrinsic noise [9] , [11]–[13] . While there has been discussion of variability in the process of transcription both in polymerase binding and in transcription elongation , e . g . , [14]–[16] , this is often with the principal aim of understanding intrinsic noise: in the following we will investigate how extrinsic factors might modulate transcription rate . To investigate the origins of global variability in eukaryotic gene expression we undertook a study of global transcription rate . We define global transcription rate as the average rate of production of transcripts within the nucleus of a single cell . Our results , obtained using direct measurement approaches , demonstrate that there is marked cell-to-cell variability in global transcription rate . The elongation rate of RNA polymerase II ( RNA pol II ) is a likely determinant of transcriptional rate , and we demonstrate that RNA pol II elongation is very sensitive to ATP concentration . We find that differences in [ATP] between cells relate to the transcription rate variability observed . We further find that the amount of mitochondrial mass and total membrane potential ( indicated by total cellular luminescence of CMXRos dye ) both correlate with transcription rate . Finally , we find that there is pronounced variability in mitochondrial mass in cellular populations and that a source of this variability is asymmetric segregation of mitochondria during mitosis . In combination these findings suggest that variability in mitochondrial content represents a likely source of global variability in transcription rate in eukaryotic cells . We directly measured transcription rate by recording levels of bromouridine ( BrU ) incorporated into nascent RNA [17] , [18] . The intensity of the BrU signal in RNA containing BrU ( Br-RNA ) rises with time , reaching a plateau after 1 h of incubation ( Figure S1 ) , when equilibrium between synthesis and degradation is reached . In these experiments BrU levels were analysed on confocal sections , providing a measure of the transcription rate per unit of nuclear volume . After a short pulse of BrU ( 30 min ) the amount of Br-RNA produced by different cells ( Hela ) varied dramatically across the cell population ( Figures 1A and S1A ) . This variation in BrU incorporation per unit of nuclear volume was not limited to Hela cells , but was observed in other established mammalian cell lines ( murine erythroleukemia cells and Chinese hamster ovary cells ) , immortalised cultures ( EBV-transformed lymphoblasts and mouse embryonic stem cells ) , and , importantly , in primary cells ( lymphocytes and primary human fibroblasts ) with coefficients of variation ( CVs ) ranging from 0 . 3 to 0 . 6 ( data not shown; CV is the standard deviation of the data points divided by their mean ) . Many factors could possibly contribute to variability in BrU incorporation . One source of variation could be differences in staging between cells in a population [9] , [12] . We observed that the variability in total nuclear BrU incorporation remains substantial throughout the cell cycle , from a CV of 0 . 36 in G1 to 0 . 35 in G2 ( Figure S1D and S1E ) , and thus , in agreement with previous studies performed in yeast [11] , the cell cycle was ruled out as a principal source of variability in BrU incorporation . Another source of variability in BrU incorporation could be differences in the number of active molecules of RNA pol II between individual cells . Therefore , we estimated the number of active RNA pol II molecules in different cell types , using run-on experiments [18]–[20] ( see Figure S2A and S2B ) . The results suggest that the amount of active RNA pol II molecules was approximately constant per unit of nuclear volume in a given population and across the different cell types analysed . This suggests that the variation observed in BrU incorporation is not due to differences in the number of active RNA pol II molecules between individual cells . We next asked whether variability in transcription rate by RNA pol II could account for the differences in BrU incorporation observed . The transcription cycle by RNA pol II can be understood as follows: free RNA pol II molecules interact with DNA , making a complex that can either be abortive ( binding to DNA and not transcribing , or transcribing a very short transcript ) or that can proceed into elongation mode after being modified . Once RNA pol II elongating molecules finish the transcription cycle , they become free and diffuse throughout the nucleoplasm . This simple model thus involves steps with different kinetic properties , which we exploited to gain insight into the rate of transcription of RNA pol II in single cells . We generated a cell line ( C23 ) in which a GFP-tagged version of wild-type RNA pol II was introduced into Chinese hamster ovary cells containing a temperature-sensitive mutation in the largest catalytic subunit of RNA pol II ( tsTM4 ) . At the restrictive temperature , only the wild-type GFP–RNA pol II was functional [21] , complementing the endogenous RNA pol II mutant ( tsTM4 ) and thereby enabling the mutant Chinese hamster ovary cells to grow normally [22] . We performed fluorescence loss in photobleaching ( FLIP ) analysis of the wild-type GFP–RNA pol II ( Figure 1B–1D ) and obtained Koff values for RNA pol II consistent with the presence of at least two populations of RNA pol II molecules ( Figures 1D and S3 ) , as has been previously suggested: one freely diffusible ( short half-life ) , and another associated with the DNA ( long half-life ) [21] , [23] . When we analysed the Koff data from individual cells of the DNA-associated population ( long half-life ) , we found a huge variation between cells ( Figure 1E ) , with a half-life ( t1/2 ) ranging from 2 . 5 to 30 min with a mean of 10 . 1 min and a standard deviation of 4 . 9 min , suggesting the existence of significant variation in rates of transcription elongation . The FLIP analysis exhibited a CV of 0 . 49 , comparable with the variation in BrU incorporation we observed in this cell type ( CV = 0 . 46 ) . To assess whether the differences in Br-RNA between cells correspond to variation in transcription rate , we performed FLIP analysis in a group of C23 cells , followed by BrU incorporation . This experiment showed a strong relationship between DNA-associated RNA pol II t1/2 and Br-RNA production ( Figure 1F ) . The faster the RNA pol II was dissociating from the DNA , the more Br-RNA was produced , supporting the suggestion that variability in the rate of DNA dissociation was coupled to variability in the rate of transcript elongation . Transcription through intact chromatin involves the removal of histone H2B in order to destabilise the nucleosome [24] , and consequently the dynamic properties of histone H2B reflect transcription elongation rate [25] . We therefore analysed the rate of exchange of fluorescently tagged histone H2B as a complementary approach to assess RNA pol II elongation rates in individual cells . Half of the nuclei of Hela cells expressing histone H2B–GFP were photobleached , and the decay of the signal in the unbleached halves was analysed . H2B–GFP showed a bi-exponential decay with a short t1/2 population that exchanges in a transcription-dependent manner ( Figure S4 ) ( ∼7% of the histone H2B–GFP ) . The t1/2 of the fast-turnover histone H2B–GFP showed a wide range of values , consistent with different cells transcribing at different speeds ( Figure 1G and 1H ) . This was again corroborated by experiments where cells were incubated with BrU after photobleaching , showing a good relationship between H2B ( t1/2 ) and Br-RNA production ( Figure 1I ) . As in the case of RNA pol II , the more dynamic the exchange of H2B , the more Br-RNA was produced , and vice versa . Taken together , these results provide good evidence that transcription elongation varies significantly between different individual cells within an otherwise homogenous population . Next , we asked whether all the elongating RNA pol II molecules in a given cell were transcribing at a similar speed . In order to analyse only the nascent transcripts we limited the BrU pulse to 15 min and immediately “froze” cells with sarkosyl [18] . We measured the intensity of multiple individual Br-RNA foci within each nucleus ( Figure 1J and 1K ) . We plotted the CV of the intensity of these nascent transcripts ( Br-RNA foci ) versus the mean intensity of these foci in the same cell , and carried out the analysis in cells exhibiting different amounts of transcription ( Figure 1L ) . The data show scant change in the CV , consistent with all the polymerases that share the same nucleus transcribing at similar speed . There , thus , appears to be a global factor coupling the transcription rates of all foci across the nucleus ( the variability in the rate of expression between different foci in the same nucleus is independent of the average rate of expression in the nucleus ) . To summarise , we found a marked variability in the levels of steady state incorporation of BrU in genetically identical populations ( Figures 1A and S1A ) ( and this appears to be independent of cell cycle position; Figure S1E ) . We then investigated the connection with transcription rate . The RNA pol II experiments suggested that there was marked cell-to-cell variability in the rate of dissociation of RNA pol II from the DNA ( even though run-on data suggested that the amount of associated RNA pol II is relatively constant between cells; Figure S2A and S2B ) . The H2B–GFP experiments ( Figure 1H ) suggested this was related to cell-to-cell variation in transcription elongation rate . Both of the bleaching experiments suggested a correlation between DNA dissociation rates , RNA elongation rates , and the levels of BrU incorporation ( Figure 1 ) . This leaves the factor responsible for this cell-to-cell variation in global transcription rate unexplored , but , as the experiments in Figure 1L show , the factor appears to be affecting all transcription foci equally in the nucleus . Next , we investigated whether the global factor responsible for the variation in transcription rate was soluble . In a first approach we incubated cells with BrU for 30 min and analysed the intensity in the nucleus and mitochondria in individual cells . This experiment showed a strong correlation between BrU incorporation in these two compartments ( Figure 2A and 2B ) , suggesting the factor is not nucleus specific . In a second experiment , we fused Hela cells with polyethylene glycol and after 2 . 5 h we carried out BrU incorporation for 30 min . This experiment showed that nuclei sharing the same cytoplasm have almost identical levels of BrU incorporation per unit of nuclear volume , showing an average CV of 0 . 04 ( in contrast , the average CV for randomly selected pairs of nuclei from different cells was 0 . 50 ) . The same was observed when the dynamic properties of RNA pol II–GFP or histone H2B–GFP were analysed in fused cells ( Figure 2C and 2D ) ( note that in Figures 2D and S12D the CV is the average of the CVs calculated for pairs of nuclei ) . Both sets of experiments suggested the existence of a small soluble factor responsible for the variation . An obvious candidate is differences in substrate content ( nucleotides ) available to RNA pol II in each cell . This was supported by the observation that “in vitro” transcription using a fixed concentration of bromouridine triphosphate ( BrUTP ) as a tracer showed a much lower degree of variability than BrU incorporation “in vivo” ( CV<0 . 10 ) ( Figure S5 ) . BrUTP incorporation “in vitro” was performed in permeabilized cells , guaranteeing an even concentration of precursors to all cells . Based on these results , we sought to analyse the relationship between nucleotide precursors and BrU incorporation . However , even though we have a good knowledge of NTP concentrations in cell populations [26] , there are no methods available to measure the nucleotide content in individual cells that are appropriate in this case . Instead , we studied the behaviour of RNA pol II with respect to [NTPs] . We used a nonradioactive method to measure the kinetic properties of RNA pol II , attached to the appropriate template , in the natural environment of the cell nucleus . Our method is based on measurements of the amount of incorporated BrUTP in nascent RNA , detected by immunofluorescence . We checked the dependence of the speed of transcription on different substrate concentrations . Cells were incubated with a cocktail containing different concentrations of all the NTPs except for ATP , which was fixed at a cell physiological level of 1 mM ( henceforth NTP will refer to UTP , CTP , and GTP only ) . Plotting transcription rate , V , versus [NTP] yields a hyperbolic curve ( Figure 2E ) , consistent with Michaelis-Menten kinetics with a Km of 80±10 µM ( R2 = 0 . 996 ) ( Figure 2E and 2F ) . This suggests that RNA pol II activity depends on the nucleotide content of the cell . However , the concentration of NTPs inside the cell is believed to be in the millimolar range [26] . From Figure 2E , this means that RNA pol II is effectively working at full speed with respect to NTPs ( even if NTP concentration falls from 1 mM to 250 µM ) . Therefore , [NTP] is unlikely to be the factor responsible for the observed variation . Some models for transcription in the literature have explicit and implicit energy dependences ( see [27] for an example ) . Given this energy dependence , we also studied the RNA pol II activity with respect to [ATP] ( this time fixing NTP concentration at 100 µM ) . The plot of V versus [ATP] showed a sigmoidal curve ( Figure 2G ) . A plot of [ATP]/V versus [ATP] ( Figure 2H ) [28] emphasises this . It is thus possible that RNA pol II behaves as an allosteric enzyme ( Hill coefficient of 1 . 5±0 . 34; R2 = 0 . 99; Km 870±450 µM ) with respect to ATP . An allosteric behaviour of RNA pol II has not to our knowledge been previously reported , possibly because all other studies have been performed either in vitro with purified enzymes or without the near-physiological conditions necessary to minimise the perturbation of essential macromolecular complexes . Our transcription system uses physiological salt concentrations and macromolecular crowding agents that keep the molecular complexes as close as possible to “in vivo” conditions . The apparent allosteric behaviour of RNA pol II is consistent with evidence that active RNA pol II forms structures containing several molecules [18] , [20] , [29] . There are also reports of more simple viral RNA polymerase molecules that oligomerize and show cooperativity [30] . Another explanation for this allosteric behaviour could be an effect of ATP on other proteins that influence the catalytic activity of RNA pol II . Obvious candidates are remodelling factors and/or DNA helicases that are generating template for RNA pol II in an ATP-dependent manner . In this category we can find the ATPase CHD1 ( chromo-ATPase/helicase–DNA-binding domain ) , which remodels nucleosomes in vitro and appears to function in both elongation and termination [31] . Another example is the remodelling complex SWI/SNF , which is also ATP dependent and associates with the RNA pol II holoenzyme [32] . Therefore , the activity of all these factors should affect the apparent activity of RNA pol II . To study if this was the case we decided to uncouple transcription from remodelling . We reasoned that by decondensing chromatin , remodelling factors would not limit the availability of DNA , and therefore these factors would contribute very little , if at all , to the kinetics of RNA production . We explored such a possibility by repeating the study of the relation between RNA pol II kinetics and [ATP] in swollen cells . Incubation of cells in hypotonic buffer for 10 min induced chromatin decondensation ( Figure S6 ) , and in these swollen nuclei the kinetic behaviour of RNA pol II with respect to [ATP] was hyperbolic ( Figure 2I ) , in contrast to the sigmoidal kinetics observed in unswollen native cells . This hyperbolic behaviour with respect to [ATP] has also been reported for remodelling factor ( s ) [33]; the sigmoidal kinetics of RNA pol II with respect to [ATP] may be the result of two consecutive sub-processes ( elongation and remodelling ) with hyperbolic kinetics . Chromatin remodelling effects have been suggested as a cause of intrinsic noise [2] , so it is interesting to note their possible role in global variability . Whatever its origin , sigmoidicity seems to be dependent on the native status of these molecules on the natural template , which means that it probably reflects an in vivo scenario . As the intracellular [ATP] is believed to be ∼1 mM [26] ( close to the RNA pol II Km of ∼870 µM , found in our conditions ) , small fluctuations in [ATP] are likely to affect transcription elongation in vivo . ( This paper is concerned with the connection between transcription rate and mitochondrial function , but we also investigated the connection between mitochondrial mass , ATP , and protein synthesis; more details can be found in Figure S13 . ) We presented evidence that the global factor modulating transcription rate does so for both nuclear and mitochondrial genes ( and so is not nuclear specific; Figure 2B ) . Fusion studies suggested this factor is small and rapidly diffusing ( Figure 2D ) . In vitro studies indicate a sensitive dependence of transcription rate on [ATP] ( at around cellular concentrations ) , while this is not the case for other NTPs ( Figure 2E and 2G ) . Decondensing the chromatin eliminates this sensitivity ( Figure 2H ) . The experiments described above suggest that the differences seen in BrU incorporation could be a reflection of cellular heterogeneity in ATP content . Indeed , in population studies where we perturbed intracellular [ATP] , we observed a direct relationship between BrU incorporation and [ATP] ( Figure 2J ) . A similar effect was observed in the rate of dissociation ( t1/2 ) of RNA pol II ( Figure S7 ) . By sorting cells according to their mitochondrial content ( using MitoTracker Green FM dye ) , we found that cells with a higher transcription rate per unit volume of nuclear material also have more mitochondrial mass ( Figures 3A , 3B , and S8 ) . Using similar sorting experiments , we found evidence that a crude measure of cellular [ATP] covaried with mitochondrial content ( Figure S9A and S9B ) . We explored this correlation further using another indicator of [ATP] . ATP is a product of mitochondrial function so we assessed the mitochondrial membrane potential ( Δψ ) , which is the driving force for ATP production [34] . Cells were sorted according to tetramethyl rhodamine methyl ester ( TMRM ) levels ( an indicator of Δψ ) , and we found evidence for a correlation with an approximate measure of cellular [ATP] ( Figure S9C ) . Single-cell studies showed that both total membrane potential and also transcription rate are slowly varying ( Figure S10; Video S1 ) . For further discussion , see Text S1 . To study the relationship between Δψ and the rate of BrU incorporation per unit of nuclear volume we used MitoTracker Red ( CMXRos ) , a fixable probe ( TMRM is not fixable ) sequestered inside the mitochondria that depends on Δψ . After incubation with both reagents we quantified the signals in individual cells . The two parameters showed a strong relationship , suggesting that total membrane potential relates to the BrU incorporation ( Figure 3C and 3D ) . To give further support we used phosphorylated ribosomal protein S6 ( P-S6 ) as a reporter of the energy state of the cell . P-S6 is located downstream in the mTOR pathway . mTOR is a homeostatic [ATP] sensor , and phosphorylation of its targets is dependent on ATP concentration [35] . One target of mTOR is the ribosomal S6 kinase ( S6K1 ) that phosphorylates the ribosomal protein S6 [36] . The use of P-S6 as a reporter for energy status was validated by induction of energy stress after deprivation of glucose and incubation with deoxyglucose ( DG ) for 12 h , which resulted in depletion of cellular ATP ( [36] and this study , data not shown ) . As predicted , P-S6 decreased in response to energy depletion ( ) , working as a surrogate reporter of the energy status of the cell . Next , we incubated cells for 30 min with BrU , and after immunolabelling with BrU and P-S6 antibodies , we observed a correlation between both signals ( Figure 3E ) , and both decreased in a manner proportional to the concentration of DG ( Figure 3F ) . We also increased the intracellular concentration of ATP by incubation with succinate at 5 and 10 mM , which increased [ATP] to 135% of normal levels , resulting in an increase in transcription rate and reduction in transcription rate variability ( assessed by measuring total nuclear BrU incorporation and H2B t1/2 exchange ( Figure 3G and 3H ) . If mitochondrial activity is coupled to variability in transcription rate , then changing mitochondrial function by altering the presence of anti- or prooxidants might affect this rate and its variability . We undertook studies using the antioxidants dithiothreitol ( DTT ) and MnTMPyP and prooxidants diamide and N-ethylmaleimide ( NEM ) ( Figure S12 ) . These studies suggested that the presence of antioxidants increases transcription rate and reduces rate variability , with the opposite holding for prooxidants ( Figure S12B and S12C ) . For further discussion , see Text S2 . In summary , we find evidence suggesting that transcription rate per unit volume of nuclear material covaries with the mitochondrial mass of cells ( Figures 3A , 3B , and S8 ) . We also found that a measure of membrane potential ( integrated over the cell ) correlated with transcription rate per unit volume ( Figure 3D ) . By modulating intercellular nutrients we modulated intracellular [ATP] and found that this also correlated with the degree of BrU incorporation ( Figure 2J ) . Further indirect studies gave support to this connection between ATP levels , mitochondrial mass , and transcription rate ( Figures 3F–3H , S9 , and S12 ) . In order to understand the origin of cell-to-cell differences in transcription rate , and given the observed connection between transcription rate and mitochondrial mass , we measured the mitochondrial content in Hela cells using MitoTracker Green FM dye . This staining demonstrates that Hela cells are heterogeneous in terms of mitochondrial content ( Figure 4A and 4B ) . We investigated the asymmetric segregation of mitochondria between daughter cells as a possible source of this heterogeneity . We used for this analysis a stable cell line containing mitochondria tagged with yellow fluorescent protein ( YFP ) . A plasmid encoding subunit VIII of cytochrome c oxidase fused with YFP was transfected into an epithelial-like human cell line derived from a bladder carcinoma ( ECV304 ) . The tagged subunit is incorporated into mitochondria , diffusing rapidly throughout the interior of the mitochondrion [37] . This makes this chimeric protein an ideal reporter to study the behaviour of mitochondrial mass at mitosis . We focused on cells in telophase or late mitosis ( Figure 4C ) , where we measured the mitochondrial mass for each daughter cell as the integrated intensity of the mitochondrial signal [38] . This analysis showed that cells generically segregate mitochondria in an uneven manner ( Figure 4C and 4D ) . Given the observation that mitochondrial mass segregates asymmetrically , one can ask whether this is relevant to cell physiology . Cell tracking experiments showed that mitochondrial content at mitosis correlates with cell cycle length . The daughter cells with more mitochondria progressed through the cell cycle proportionately faster than their sisters ( Figure 4E ) . To rule out the trivial explanation that asymmetry in mitochondrial content was an effect of asymmetry in the volume of daughter cells , we used ECV304 cells expressing DsRed , which is a soluble protein and distributes evenly throughout the cell . The analysis of the ratio of DsRed between daughters ( ratio of cell volumes ) versus the ratio of time to complete a cell cycle did not show a clear relationship ( Figure 4F ) , making it unlikely that asymmetries in the volume of daughter cells are the principle cause . We also found that daughters that inherit more mass than their sisters also have a higher rate of translation of some proteins ( Figure S13A ) ; this further suggests that the uneven inheritance of mitochondria has an effect through the cell cycle . For further discussion , see Text S3 . Since mitochondrial segregation is correlated with variation in cell cycle length , one might want to understand how mitochondrial partition at birth is controlled . We analysed whether the process of mitochondrial segregation had a memory . We measured the mitochondrial content ratio between daughters at birth ( F1 ) , followed each daughter for one cell cycle , and measured the mitochondrial content ratio of daughters in generation F2 . When we compared F1 versus F2 in terms of asymmetry , there was no clear relationship ( Figure 4G ) . The time to division of generation F1 cells was largely independent of the interdivision times of respective F2 cells ( Figure 4H ) . This gives us a more refined view of the stochastic character of mitochondrial segregation . We have found evidence for variability in mitochondrial mass within a population ( Figure 4A ) . A possible cause is asymmetric segregation of mitochondrial mass at division ( Figure 4C and 4D ) . Daughter cells that inherit more mitochondrial mass progress through their cell cycles faster and can show a faster rate of protein synthesis ( Figures 4 and S13A–S13D ) . We found no strong evidence for a dependence between one cell cycle duration and the next ( Figure 4G and 4H ) . This paper investigated the connection between two forms of cellular variability: variation in mitochondrial mass and variation in global transcription rate . We found marked heterogeneity in the amount of mitochondrial mass present in cells ( Figure 4B ) and evidence for an origin of this variability in the stochastic partition of mass at point of division ( Figure 4D ) . We further found that this variation has a cell physiological correlate: daughters that inherit relatively smaller amounts of mitochondrial mass than their sisters have longer cell cycles ( Figure 4E ) . We also presented evidence for global ( Figure 1L ) variability in transcription rate ( Figures 1E , 1H , and S1A ) . While our experiments suggest that the numbers of bound RNA polymerases are constant and transcription rate variability is independent of cell cycle stage ( Figures S1E , S2 , and S5 ) , they also suggest that a small diffusing factor may be responsible for this global transcription rate modulation ( Figure 1D ) . Given the above , we hypothesized , first , that there was aconnection between global transcription rate variability and variability in cellular mitochondrial content and , second , that this was mediated by variation in the fast-diffusing factor ATP . Studies in permeabilized cells ( Figure 2G ) found a sensitive dependence of transcription rate on [ATP] . In vivo perturbation studies also found a correlation between cellular [ATP] and transcription rate ( Figure 2J ) . We found that cells with more mitochondrial mass transcribe faster per unit volume of nuclear material ( Figure 3B ) . We further found that those with higher total membrane potential ( as indicated by CMXRos ) transcribed faster ( Figure 3D ) . We also found evidence correlating levels of ATP with mitochondrial mass and total membrane potential ( Figure S9B and S9C ) . Finally , we found that perturbing mitochondrial function with anti- or prooxidants perturbed transcription rate variability ( Figure S12 ) . Studies thus far have left our understanding of the origins of global variability in gene expression in higher eukaryotes unclear [2] . This paper suggests that cell-to-cell variability in mitochondria is coupled to cell-to-cell variability in global transcription rate . For in vivo transcription , cells were incubated in the presence of different concentrations of BrU ( Sigma ) for different times ( stated in figure legends ) . Incubation for 1 h with 100 µM DRB or for 1 h with 1 µg/ml actinomycin D prior to BrU incubation abolished BrU incorporation completely ( data not shown ) . For individual transcript analysis , Hela cells were grown on coverslips at low density then incubated for 15 min with 5 mM BrU , washed with PBS , and treated with 0 . 375% sarkosyl , 25 U/ml ribonuclease inhibitor , 10 mM EDTA , and 100 mM Tris-HCl ( pH 7 . 4 ) for 10 min at 20°C . Next , coverslips were tilted to allow the cell content to run out for 5 min . Samples were air-dried and fixed with 4% paraformaldehyde for 10 min and processed for Br-RNA detection . For transcription in vitro we used the conditions described in [18] plus 5% Ficoll 400 . For detection of primary transcripts , we used mouse anti-IdU/BrdU ( 5 mg/ml; Caltag Laboratories ) . Secondary antibodies were donkey anti-mouse IgG tagged with Cy3 ( 1/200 dilution; Jackson ImmunoResearch ) . The immunodetection procedure was performed as described in [18] , [19] . DNA was stained with 200 nM TO-PRO-3 ( Molecular Probes ) for 5 min , then slides were mounted in Vectashield ( Vector Laboratories ) , and images were collected using a Radiance 2000 confocal microscope ( BioRad Laboratories ) . Intensities in the nucleoplasm were measured using EasiVision software ( Soft Imaging Systems ) and data exported to Excel ( Microsoft ) for analysis . For cell fusion experiments , Hela cells were grown on coverslips to 80% confluence . Cells were fused using polyethylene glycol as described by Schmidt-Zachmann et al . [39] . After 2 . 5 h cells were incubated with 2 . 5 mM BrU for 30 min and then immunolabelled as described above . A clone stably expressing GFP–RNA pol II ( C23 ) [22] was cultured at 39°C , and images were collected with the microscope stage heated to 39°C . Fluorescence images were collected using a confocal microscope ( Zeiss LSM 510 META ) , with an EC PlnN 40×/1 . 3 oil objective , with the pinhole completely open . We selected a rectangle at the bottom half of each nuclei where we applied 100% laser power , in order to bleach all the fluorescent molecules in the rectangle . This operation was repeated every 5 s for a period of 1 , 200 s , and we analysed the decay of the fluorescence in the unbleached top half . Fluorescence intensity was analysed in MetaMorph 6 . 1 ( Universal Imaging ) . Curves were analysed using Sigma Plot 8 . 0 for Windows . For the analysis we assumed that there were two populations , freely diffusible , bound to DNA and fully engaged in transcription . For the fitting we allowed the two components to optimise with no restriction . Data were fitted to two populations with exponential decay ( always R2>0 . 99 ) . Fixing the slow population to an average speed rendered unacceptable fittings with the second population . We were concerned with the possible artefacts induced by FLIP . Therefore , transcription “run on” experiments were performed on photobleached cells , which demonstrated no alteration in the transcription pattern or intensity in the bleached area ( data not shown ) . Hela cells expressing histone H2B–GFP [25] were used to study the dynamics of histone H2B . FLIP was performed as for C23 cells , but the time was reduced to 10 min of photobleaching and the temperature was set at 37°C . The decay curves can be fitted to a bi-exponential decay . The two initial points were discarded because they correspond with the free population of histone H2B . One possible problem with the use of the exchange of histone H2B–GFP as a transcription reporter is the impact of its overexpression . However , in the cell line used , histone H2B–GFP represents 10% of all cellular histone H2B [25] . The production of natural histone H2B is reduced in preserving the normal amount of histones , which means that no overexpression occurs in this cell line [25] . The fraction of free histone in the cell line used is around 1% of the total H2B–GFP , which corresponds with the fraction bleached in the first two cycles of bleaching . This population was not considered for the analysis . Even if any hyperexpression occurs in this cells , only the fraction of molecules bound to DNA and not the t1/2 will be affected , which is the parameter studied . In agreement with this interpretation we did not observed a correlation between the initial fluorescence before bleaching of histone H2B–GFP and t1/2 ( Figure S4C ) . Tripsinized Hela cells were stained with MitoTracker Green FM dye ( Molecular Probes ) for 15 min in DMEM or TMRM ( Molecular Probes ) for 30 min , following manufacturer guidelines . Then , cells were sorted on a fluorescence-activated cell sorter ( MoFlo; DakoCytomation ) to purify populations of cells with different mitochondrial content or membrane potential . MitoTracker Red ( CMXRos ) was used following the manufacturer guidelines ( Molecular Probes ) . Cells were stained for 10 min in vivo after being grown in BrU for 30 min . Br-RNA was detected as previously described . For superoxide detection cells were incubated with 20 nM for 12 h with MitoSox ( Molecular Probes ) and then grown in BrU for 30 min . Cells were analysed using wide confocal cytometry [38] . ATP depletion experiments were carried out by incubation of cells for 12 h with different concentrations of DG ( Sigma ) . In another set of experiments ATP was depleted by incubation with 10 mM sodium azide ( Sigma ) and 6 mM DG in HBSS for 1 h ( BioWhittaker ) . ATP concentration was determined using the kit ATP Bioluminescence Assay Kit HS II ( Roche ) following manufacturer instructions . For antioxidant treatments , cells were incubated for 18 h with MnTMPyP ( CalBiochem ) or DTT ( Sigma ) . MnTMPyP was used at 50 , 25 , and 12 . 5 µM . DTT was used at 1 , 000 , 500 , 250 , and 125 µM . GSH was depleted by incubation with 200 , 100 , or 50 µM diamide ( Sigma ) for 2 h . CE–mitoRFP-W vector was generated from the pHR-SIN-CSGW vector [40] by exchanging the SFFV promoter for a human EF1a promoter and the GFP reporter for mitochondrial DsRed2 isolated from pDsRed2-Mito ( Clontech ) . Lenti lox vector expressing GFP , Emerald , or Cherry was generated as described in [41] . Lentiviruses were pseudotyped with the vesicular stomatitis virus G ( VSVG ) protein by transient transfection of 293T cells [41] . Viral stocks were prepared by ultracentrifugation , and viral particles were used for Hela H2B–GFP infection; 2 wk after infection a clone was selected . For in vivo analysis , cells were plated in a 48-well plate at low cell density , and left for 12 h to attach . Then the plates were transferred to the Cell IQ platform ( Chip-Man Technologies ) . Images were recorded every 30 min , for at least 6 d . Images were analysed using MetaMorph 6 . 1 . After completion of mitosis , the ratio of the integrated intensity of the fluorescent signal between daughter cells was measured as described in [38] .	Though pairs of cells may have identical genes , they still show behavioural differences . These cell-to-cell differences may arise from variations in how genes are transcribed and translated by the cellular machinery . Identifying the origins of this variation is important as it helps us understand why genetically identical cells can show a range of responses to the environment . In this work , we measured the rate at which the genes yield transcripts in cultured human cells . We found marked cell-to-cell variability in average rates of transcription . This variability is related to mitochondrial content . Cells with a higher mitochondrial mass have a faster rate of transcription , and we show that part of this variability is due to the unequal distribution of mitochondria to daughter cells when cells divide . Additionally , we find that cells that inherit more mitochondria divide earlier . These findings make a connection between variability in transcript production and variability in cellular mitochondrial content .	[ "Abstract", "Introduction", "Results/Discussion", "Materials", "and", "Methods" ]	[ "cell", "biology/nuclear", "structure", "and", "function", "cell", "biology/cell", "growth", "and", "division", "cell", "biology", "cell", "biology/plant", "genetics", "and", "gene", "expression", "cell", "biology/chemical", "biology", "of", "the", "cell", "cell", "biology/gene", "expression" ]	2010	Connecting Variability in Global Transcription Rate to Mitochondrial Variability
The interferon-induced dynamin-like MxA GTPase restricts the replication of influenza A viruses . We identified adaptive mutations in the nucleoprotein ( NP ) of pandemic strains A/Brevig Mission/1/1918 ( 1918 ) and A/Hamburg/4/2009 ( pH1N1 ) that confer MxA resistance . These resistance-associated amino acids in NP differ between the two strains but form a similar discrete surface-exposed cluster in the body domain of NP , indicating that MxA resistance evolved independently . The 1918 cluster was conserved in all descendent strains of seasonal influenza viruses . Introduction of this cluster into the NP of the MxA-sensitive influenza virus A/Thailand/1 ( KAN-1 ) /04 ( H5N1 ) resulted in a gain of MxA resistance coupled with a decrease in viral replication fitness . Conversely , introduction of MxA-sensitive amino acids into pH1N1 NP enhanced viral growth in Mx-negative cells . We conclude that human MxA represents a barrier against zoonotic introduction of avian influenza viruses and that adaptive mutations in the viral NP should be carefully monitored . Avian influenza A viruses sporadically transmit from waterfowl , their natural reservoir , into the human population [1]–[5] . These zoonotic viruses usually cannot propagate in the new human host , nor do they readily transmit between humans [6]–[8] . In rare cases , however , influenza A viruses of avian origin break the species barrier and establish new virus lineages in humans . In the last 100 years , the introduction of an influenza A virus with a novel nucleoprotein ( NP ) gene segment occurred only on two occasions , both of which led to pandemics: in 1918 ( “Spanish” H1N1 ) an avian virus and in 2009 ( pH1N1 ) a reassortant virus ( comprising gene segments of two swine influenza viruses ) established a stable lineage in humans [9] , [10] . In contrast , the 1957 ( “Asian” H2N2 ) and the 1968 ( “Hong-Kong” H3N2 ) pandemics were caused by genetic reassortment events , whereby the circulating human strains acquired some gene segments from avian sources but kept , among others , their 1918-derived NP [11] . To overcome the species barrier , multiple adaptations to the new host are required [6] . Theoretically , two categories of adaptive mechanisms can be envisaged . One comprises adaptations to cellular factors which promote viral infection yet differ between hosts . These include , for example , changes in the viral hemagglutinin during the adaptation of avian influenza A viruses to humans [12] , or altered binding of viral proteins to different cellular importins [11] , [13] . The second category comprises adaptations to counteract cellular restriction factors that inhibit virus replication . These factors are part of the intrinsic and innate host defense mechanisms and may exert a strong selective pressure against newly invading viruses . Surprisingly little is known about adaptive mutations that overcome such host restriction factors and facilitate trans-species transmission of influenza viruses . The human interferon ( IFN ) system represents a major innate defense against zoonotic viruses . Among the many antiviral factors induced by IFNs , the MxA protein is one of the most potent characterized to date [14] . It is a key effector molecule inhibiting influenza A virus as well as several other human RNA viruses [15] , [16] . MxA is a dynamin-like large GTPase which consists of an N-terminal globular GTPase domain , a bundle signaling element , and a C-terminal helical stalk . The recent atomic resolution of the MxA structure revealed that it forms stable tetramers and oligomers which assemble in a criss-cross manner via the stalk [17] , [18] . A current model proposes that , upon viral infection , MxA recognizes the incoming vRNPs and starts to self-assemble into rings , resulting in a higher-order oligomeric complex that blocks vRNP function [18] , [19] . In accordance with this model , recent findings suggest that NP determines the relative sensitivity of influenza A viruses toward the antiviral action of MxA . Avian influenza viruses were found to be generally more sensitive to MxA than human strains [20] , which was subsequently shown through reassortant viruses to be dependent on the origin of NP [21] . These findings suggest that human strains acquire adaptive mutations in NP to evade MxA restriction . Here , we identified the amino acids critical for MxA resistance in the two NP proteins introduced into the human population in 1918 ( by the “Spanish” H1N1 influenza A virus ) and in 2009 ( by the pH1N1 strain ) . These residues clustered into two distinct but overlapping “patches” in the body domain of the protein . Introduction of these amino acids into an MxA-sensitive H5N1 NP was sufficient to render the avian polymerase resistant to MxA . Surprisingly , the resistance-associated substitutions resulted in impaired viral growth in both mammalian and avian cells when introduced into recombinant H5N1 virus A/Thailand/1 ( KAN-1 ) /04 . The amino acid clusters identified here are highly conserved in circulating human isolates and virtually absent in NPs of avian influenza A viruses . Several of the amino acids that confer increased resistance to human MxA are also conserved in influenza A viruses of the classical swine lineage , correlating with resistance of these viruses to swine Mx1 . These findings suggest that multiple adaptive amino acid changes would be required for H5N1 viruses to both escape from MxA restriction and maintain viral fitness . Partial adaptation in an intermediate host , such as the pig , might facilitate this demanding process . We have previously shown that the polymerase activity of A/Thailand/1 ( KAN-1 ) /04 ( H5N1 ) is highly sensitive to inhibition by murine Mx1 , a close homolog of human MxA , in a polymerase reconstitution assay , and that this sensitivity is determined by the NP gene [21] . The H5N1 NP is of typical avian origin and resembles the avian H5N1 amino acid consensus sequence [21] . We therefore used this assay to identify the amino acids in NP of either the 1918 ( A/Brevig Mission/1/1918 ) or the 2009 ( A/Hamburg/4/2009 ) pandemic H1N1 strain ( Figure 1A ) critical for Mx1 resistance . In this assay , the polymerase activity was measured in the presence of overexpressed Mx1 . In addition , we determined the polymerase activity in the presence of the inactive mutant Mx1-K49A [22] . Mx1 resistance was defined as the relative activity of the viral polymerase in the presence of Mx1 divided by the activity obtained with Mx1-K49A . Substitution of the H5N1 NP with the NP of the pandemic 1918 strain [23] rendered the H5N1 polymerase largely Mx1-resistant ( Figure 1B ) . An alignment of the amino acid sequences of the 1918 NP with the NP of the Mx1-sensitive H5N1 strain revealed differences at 14 positions , including 4 positions in the C-terminal domain , namely amino acids 373 , 377 , 473 and 482 ( Figure 1A ) . An artificial chimera ( 1918-NP ) , consisting of the N-terminal 365 amino acids of the 1918 NP and the C-terminal domain ( amino acids 366 to 498 ) of the H5N1 NP , behaved like the full-length 1918 NP , indicating that the 4 C-terminal differences in the 1918 protein do not contribute to the Mx1 resistance phenotype ( Figure 1B ) . To investigate which of the 10 remaining 1918-specific amino acids in the chimeric protein contributed to Mx1 resistance , 1918-NP mutants harboring single H5N1-derived substitutions were tested . A major decrease in Mx1 resistance was observed for the mutations P283L and Y313F , while a less pronounced phenotype was observed for I100R and several other mutants ( Figure 1C ) . Various combinations of these putative adaptive mutations revealed that the triple mutant I100R , P283L , and Y313F led to a similar degree of Mx1 sensitivity as observed using the H5N1 NP ( Figure 1D ) , whereas combinations of the remaining seven mutations failed to reduce Mx1 resistance ( Figure 1D ) . Consistently , the 1918-NP , carrying the mutations I100R , P283L and Y313F reduced Mx1 resistance by 50% ( as compared to 1918-NP ) also in the context of the 1918 polymerase ( Figure S1A–B ) . To test whether the amino acids apparently responsible for Mx1 resistance of 1918 NP could also confer Mx1 resistance to an Mx1-sensitive NP , we introduced the mutations R100I , L283P , and F313Y into H5N1 NP . We also tested the exchange R100V in NP , since screening of the NCBI influenza database revealed that valine rather than isoleucine is commonly found at this position in seasonal strains [24] . Single amino acid exchanges slightly increased Mx1 resistance , while the combination of all three substitutions resulted in resistance comparable to the 1918 NP , irrespective of I or V at position 100 ( Figure 1E ) . Importantly , this enhanced Mx1 resistance was not simply achieved by a higher polymerase activity , as it did not strictly correlate with increased activity in the presence of the antivirally inactive mutant Mx1-K49A . Nevertheless , some Mx1 resistance-enhancing amino acids appeared to improve the polymerase activity for unknown reasons in the absence of Mx1 or presence of Mx1-K49A protein ( Figure S1C ) . Next , we evaluated the capacity of NP from the 2009 pandemic H1N1 influenza A virus ( pH1N1 ) to confer Mx1 resistance in the context of the H5N1 polymerase . Figure 2A shows that pH1N1 NP rendered the H5N1 polymerase activity relatively resistant to Mx1 inhibition , as previously reported [21] . Sequence comparisons between the NP of pH1N1 and H5N1 origin revealed that pH1N1 NP carried only one ( V100 ) out of the three Mx1 resistance determinants identified in 1918 NP ( Figure 1A ) . We therefore assumed that different amino acids contribute to Mx1 resistance in pH1N1 NP than in 1918 NP . Since the pH1N1 NP differs by 32 amino acids from the H5N1 sequence ( Figure 1A ) , we did not assay all individual amino acid positions , but rather focused on discordant and surface exposed amino acids in close proximity ( 27 Å ) to the resistance cluster identified in the 1918 NP , utilizing the published NP crystal structures [25] , [26] . Five of the resulting 10 pH1N1-specific amino acids , which were closest to the 1918 resistance cluster , were analyzed in the H5N1 polymerase reconstitution assay using pH1N1 NP mutants harboring single H5N1-derived substitutions ( Figure 2B ) . A significant decrease in Mx1 resistance was observed for the mutations D53E , H289Y and V313F , while a less pronounced phenotype was observed for V100R and K305R ( Figure 2B ) . Next , pH1N1-specific amino acids were introduced into H5N1 NP and tested for their contributions to Mx1 resistance in the H5N1 polymerase reconstitution assay ( Figure 2C ) . While no individual amino acid substitution had a major effect , the combination of mutations at 4 positions ( E53D , R100V , Y289H , and F313V ) enhanced Mx1 resistance to a similar extent as 1918 NP . The additional mutations R305K , I316M , T350K , R351K , V353I , and Q357K together further increased Mx1 resistance to the degree of pH1N1 NP ( Figure 2C ) . Although we observed variations in NP expression levels ( Figure 2C ) , these differences did not correlate with Mx sensitivity . Together , these results demonstrate that the cluster of amino acids conferring Mx1 resistance differs between the NP of the 2009 and 1918 pandemic strains , although the crucial residues in both cases are located in the surface-exposed body domain . Next , we investigated whether the identified amino acid clusters in NP of the 1918 and the pH1N1 strains also confer resistance to human MxA . Consistent with the findings observed with Mx1 , both 1918 and pH1N1 NP increased resistance in the H5N1 polymerase reconstitution assay ( Figure 3A ) , however , the 1918 NP confers greater resistance to human MxA than to murine Mx1 , whereas the opposite is true for the pH1N1 NP ( Figure 2C and 3A ) . Importantly , the mutant H5N1 NP containing the 1918-derived Mx1 resistance determinants R100V , L283P , and F313Y exhibited an MxA resistance approximately 85% of that by 1918 NP itself ( Figure 3A ) . To identify additional amino acid residues that contribute to the increased MxA resistance of the 1918 NP , we changed single amino acid positions in the 1918*-NP to avian residues . This revealed that D16 , in addition to V100 , P283 , and Y313 , contributed to MxA resistance ( Figure S2 ) . To confirm the relevance of this finding , we tested H5N1 NP harboring all four mutations ( G16D , R100V , L283P and F313Y ) . This mutant NP displayed an MxA resistance comparable to 1918 NP ( Figure 3A ) . Next , the pH1N1-specific adaptive mutations were tested in the context of H5N1 NP . In particular E53D , R100V and F313V increased resistance to MxA , while introduction of the additional mutations Y289H , R305K , I316M , T350K , R351K V353I and Q357K was required to achieve a resistance comparable to pH1N1-NP ( Figure 3A ) . These results indicate that the adaptive mutations in 1918 or pH1N1 NP lead to increased resistance for both murine Mx1 and human MxA . Again , the observed resistance towards MxA did not strictly correlate with polymerase activity ( Figure S3 ) . The atomic crystal structure of the H5N1 NP [26] revealed that the 1918-specific amino acids 100 , 283 and 313 form a surface exposed cluster in the body domain of the viral NP ( Figure 3B ) . Amino acid 16 is located in the N-terminal region of NP that is predicted to form a flexible loop adjacent to the 1918 cluster ( Figure S4 ) . The amino acids forming the pH1N1 cluster are located in the same area of the NP body domain as the 1918 cluster ( Figure 3C ) . V100 , P283 , and V/Y313 mainly responsible for Mx1 or MxA resistance ( Figure 3B and C ) , we analyzed the NP sequences of various isolates deposited in the NCBI Influenza Virus Sequence Database [24] . In avian isolates , each resistance-conferring amino acid could be identified in only ≤1% of the sequences ( n = 5350 ) investigated ( Table 1 ) . In contrast , in classical seasonal human isolates representing H1N1 , H1N2 , H2N2 , and H3N2 subtypes ( n = 4969 ) , the resistance-associated amino acids D16 , I/V100 , P283 , and Y313 were each found at very high frequencies ( >98% ) . Similarly , analyses of the human-derived pH1N1 NP sequences ( n = 4104 ) revealed a high conservation of amino acids D53 , I/V100 and V313 ( >99% ) . Intriguingly , a chronological sequence comparison revealed that additional Mx resistance-enhancing mutations occurred in the NP of strains which are classified as descendents of the 1918 virus , namely R305K and R351K ( Figure 4 , Figure S5 ) . These mutations emerged in early seasonal H1N1 viruses and were maintained in subsequent H2N2 and H3N2 strains ( Figure 4 ) . Taken together , these findings suggest a continuous selection pressure for increased Mx resistance in seasonal influenza viruses . Since pH1N1 NP is derived from an influenza A virus of the classical swine lineage [10] , [27] , we analyzed the NP sequences of a number ( n = 393 ) of corresponding swine isolates obtained between 1930 and 2012 . Amino acids I/V100 were highly conserved ( >99% ) , but D53 and V313 were not present in any of the NP sequences , which instead harbored the avian consensus amino acids at these positions ( Table 1 ) . We therefore anticipated that NPs of the classical swine influenza strains would confer less MxA resistance than pH1N1 NP . Indeed , NP of one of the first swine isolates such as A/swine/Iowa/1976/1931 displayed comparatively poor MxA resistance in the H5N1 polymerase reconstitution assay ( Figure 5A ) . Importantly , NP of the classical swine influenza A virus lineage acquired the additional mutations 305K , 351K , 353I and 357K over time ( Figure 4 ) , resulting in a gradual increase in MxA resistance ( Figure 5A , pH1N1-NP-D53E , V313F , M316I , third column from the left and Figure S6A ) . These changes , however , were not sufficient to confer MxA resistance comparable to that of pH1N1 NP . Interestingly , NP of the recent triple reassortant swine isolate A/swine/Ohio/02026/2008 ( H1N1 ) and a hypothetical NP precursor of pH1N1 ( pH1N1-NP-D53E , V313F , M316I ) share the adaptive mutations 100I/V , 289H , 305K , 350K , 351K , 353I and 357K . The hypothetical NP precursor of pH1N1 was created by altering the human pH1N1 specific positions D53 , V313 and M316 to the consensus found in classical swine H1N1 strains ( D53E , V313F , M316I ) . These mutations are found at the branching point of classical swine influenza viruses and pH1N1 viruses ( Figure 4 ) . The hypothetical NP precursor conferred only partial MxA resistance in the H5N1 ( Figure 5A and Figure S6A ) as well as in the pH1N1 background ( Figure S6C ) . These findings suggested that further adaptive mutations are needed which may affect MxA recognition or otherwise improve NP functions such as binding to viral ( e . g PB2 [28] or cellular components ( importins [13] , helicases [29] , [30] ) . Indeed , acquisition of E53D , F313V , and I316M was required to gain the full resistance of pH1N1 NP ( Figure 5A and Figure S6C ) . The antiviral potency of Mx proteins of pigs is still insufficiently characterized [31]–[33] . In the present reconstitution assay using HEK293T cells , porcine Mx1 ( Sus scrofa domestica ) decreased the H5N1 polymerase activity to 50% ( Figure 5B ) , whereas human MxA reduced the activity to approximately 10% ( Figure 5A ) , despite similar expression levels of both Mx proteins ( Figure 5C ) . Likewise , the H5N1 polymerase activity in the presence of different NPs from several distinct swine isolates was not significantly affected by porcine Mx1 ( Figure 5B , Figure S6B ) . To further test the antiviral strength of porcine Mx , we used porcine cells for the polymerase reconstitution assay . We found essentially the same extent of inhibition by the porcine Mx1 as in human 293T cells . As shown in Figure S7A , porcine Mx1 reduced the H5N1 polymerase activity to ca . 50% in swine NPTr or NSK cells [34] . In contrast , human MxA reduced the activity to 10% in porcine cells ( Figure S7A ) . We conclude that the antiviral effect of porcine Mx1 is weak both in porcine and human cells . Together , these data suggest that while there is a clear selection pressure for swine influenza A viruses to acquire Mx resistance , the selection pressure in the porcine host is comparatively weak . Clearly , additional adaptive mutations are required to escape MxA restriction in humans . Re-transmission of pH1N1 [35] from human to swine resulted in 17% of the documented cases in a substitution of aspartic acid at position 53 to the avian consensus glutamic acid ( D53E ) ( swine pH1N1 in Table 1 and Figure 4 ) , a mutation that confers loss of resistance to human MxA ( Figure 5A ) , but not to porcine Mx1 ( Figure 5B ) . This might suggest that MxA resistance-enhancing mutations are not necessarily favorable for NP function and might therefore cause impaired viral fitness . To compare the replication fitness of viruses containing MxA-sensitive or MxA-resistant NPs , we infected MDCKII cells ( which do not express antivirally active Mx proteins [36] ) with pH1N1 or mutant viruses with enhanced MxA sensitivity . A recombinant pH1N1 virus with the single D53E reversion ( pH1N1-NP-D53E ) grew equally well as the parental pH1N1 virus ( Figure 5D ) . In contrast , the pH1N1 precursor virus lacking three MxA resistance-enhancing mutations ( pH1N1-NP-D53E , V313F , M316I ) grew to approximately one log10 higher infectious titers throughout the course of infection ( Figure 5D ) . These results demonstrate that reversions to the original amino acids of the putative swine precursor virus ( Figure 4 ) provided a strong growth advantage in the absence of an antivirally active Mx . Thus , the acquisition of MxA resistance appears to cause some growth disadvantage . To confirm this hypothesis , we tested the human H5N1 strain KAN-1 containing MxA resistance-enhancing mutations in MDCKII cells . Consistent with previous observations [36] , the polymerase activity of H5N1 was not affected in the presence of canine Mx1 or Mx2 ( Figure S7B ) . The triple mutant H5N1-NP-R100V , L283P , F313Y achieved reduced viral titers in the order of 1–2 log10 ( Figure 5E ) . The recombinant H5N1 virus with the single MxA resistance mutation L283P ( H5N1-NP-L283P ) grew slightly less well than the parental H5N1 strain , while the double mutant virus H5N1-NP-R100I , F313Y showed comparable growth ( Figure 5E ) . Intriguingly , the latter virus showed severely impaired replication efficiency in avian LMH cells which lack antiviral Mx proteins [37]–[39] ( Figure S8 ) . We conclude that MxA resistance is linked to impaired viral growth and may be easily lost in the absence of selective pressure . We argued that the acquisition of Mx resistance should also increase the pathogenicity of H5N1 in Mx1-positive mice . To test this hypothesis we selected the double mutant virus H5N1-NP-R100I , F313Y which had almost wild-type growth characteristics in tissue culture ( Figure 5E ) and exhibited comparable polymerase activity ( Figure S8A ) , in spite of its Mx resistance-enhancing mutations ( Figure 1E ) . We could not include the triple mutant virus H5N1-NP-R100V , L283P , F313Y in our studies due to the emergence of escape mutants ( data not shown ) and its strong attenuation . First , we compared the growth of wild-type and mutant H5N1 viruses in Mx1-negative BALB/c mice . Infection with 10 PFU of wild-type H5N1 virus lead to pronounced weight loss and death of all animals , as expected [40] . In contrast , infection with the same challenge dose ( 10 PFU ) of mutant H5N1-NP-R100I , F313Y virus resulted in survival of all BALB/c mice without significant weight loss ( Figure 6A and B ) , indicating that the two amino acid substitutions associated with Mx resistance caused impaired viral growth . Indeed , challenge of BALB/c mice with 1000 PFU of the mutant virus resulted in viral lung titers that were 15-fold reduced as compared to wild-type virus at 48 h after infection ( Figure 6C ) . Next , the growth properties of the two viruses were studied in congenic Mx1-positive mice . In these animals , infection with wild-type H5N1 virus produced no pronounced pathological effects , even at high doses of 106 PFU ( Figure 6D and E ) . In contrast , Mx1-positive mice showed significant weight loss and mortality when challenged with 106 PFU of mutant H5N1-NP-R100I , F313Y virus ( Figure 6D and E ) . To assess viral growth in Mx1-positive mice , viral lung titers were determined at various time points after intranasal infection with 104 PFU ( Figure 6F ) . Two days after infection , the titers in mice infected with the mutant virus were approximately 28-fold lower than those in mice infected with the wild-type virus , demonstrating the attenuating effect of the Mx resistance-enhancing mutations in NP . Four days after infection , a 5-fold difference was observed , and 6 days after infection the mutant virus was still present in 5 out of 9 Mx1-positive mice with titers up to 2×105 PFU . In contrast , only low titers of wild-type virus were detected at the same time in 2 out of 9 animals . We conclude from these experiments that H5N1 viruses harboring MxA resistance-enhancing mutations partially overcome the antiviral effect mediated by Mx1 . Influenza A viruses sporadically transmit from the avian reservoir into the human population . Here we describe specific mutations found in the NP of the 1918 and 2009 pandemic viruses that confer resistance to the IFN-induced human MxA GTPase , a major restriction factor for influenza and other orthomyxoviruses . As MxA strongly inhibits transcription and replication of the viral genome early in infection , its antiviral activity can be readily analyzed in polymerase reconstitution ( minireplicon ) assays [20] , [21] . Using this assay , we identified a cluster of surface-exposed amino acids in the body domain of NP crucial for Mx resistance . Interestingly , different amino acid positions were identified in 1918 and pH1N1 NP , yet all were located in the same domain . All resistance-associated amino acids are conserved in previous and current human influenza A viruses ( Table 1 ) , and the continuing acquisition of resistance-enhancing mutations ( Figure 7 ) suggests strong positive selection pressure by MxA . Of note , mutations conferring MxA resistance are absent in avian influenza A viruses , although we did observe the emergence of adaptive NP mutations in avian-derived viruses circulating in swine ( Figure 4 and Figure 7 ) . These substitutions in NP not only increased resistance to swine Mx1 but also to human MxA , supporting the theory that swine are an excellent intermediate host for the generation of viruses with pandemic potential . Influenza A viruses carrying a novel NP gene were introduced into the human population in 1918 and 2009 [10] , [27] , but the distinct MxA resistance clusters in the NP genes of these pandemic viruses suggest independent evolution ( Figure 7 ) . In case of the 2009 pH1N1 virus , the NP gene originated from the classical swine lineage , which itself is of avian origin [41] . Our data suggest that the swine precursor virus of pH1N1 acquired additional amino acid changes , which together increased the ability to counteract human MxA . The evolution of the 1918 NP is comparatively less clear . Recent data suggest that both the 1918 and the classical swine virus lineage share a common avian ancestor [9] . It is unresolved whether the avian precursor virus was first transmitted to humans and then onto swine or vice versa [9] . In the latter case , the precursor virus may have first adapted to swine Mx1 through the mutation R100I in NP which is found in early swine isolates [24] . In the first case , mutations at position 283 and 313 might have been lost after transmission from humans to swine , as observed with the pH1N1 virus . However , analyses of host specificity markers which discriminate human from avian influenza viruses indicate that four adaptive mutations in NP ( 16D , 283P , 313Y and 357K ) were likely required for transmission of the 1918 precursor virus to humans [42] . Remarkably , all of these mutations contribute to MxA resistance ( Figure 3 ) and may have evolved in a pre-pandemic phase in the human population . Since circulating human influenza A virus strains maintain these adaptive mutations ( Table 1 ) , it is conceivable that viruses are under constant selection pressure mediated by MxA . We observed that the acquisition of Mx resistance had a negative effect on viral growth in the absence of MxA . When MxA resistance-enhancing mutations were introduced into highly pathogenic avian H5N1 viruses , the recombinant viruses grew less well than the wild-type H5N1 virus in MxA-negative MDCKII cells ( Figure 5E ) , in BALB/c mice ( Figure 6C ) and even in avian cells ( Figure S8 ) . These mutations have no major effect in the viral polymerase reconstitution assay ( Figure S1C and S3 ) , and it is unclear which step of the viral replication cycle is attenuated in infected cells . If MxA resistance-associated amino acids are also counter-selected in circulating avian influenza strains , then the emergence of MxA resistance in the avian reservoir is expected to be an extremely rare event ( Table 1 ) . Of note , human H5N1 isolates have developed few if any of the identified MxA resistance-enhancing mutations ( Table 1 ) , most likely due to the associated strong attenuation [43] ( Figure 5E , 6A–C ) . Perhaps for this reason , the 1957 and 1968 pandemic viruses retained the well-adapted 1918-origin NP , despite acquiring other avian genome segments by reassortment . Thus , more than 90 years passed before a new NP lineage was established 2009 in humans , a process aided by gradual adaptation of the new NP in swine . Correspondingly , we observed an increase in viral growth of pH1N1 lacking the MxA resistance-enhancing mutations D53 , V313 and M316 in MDCKII cells ( Figure 5D ) which do not express antivirally active Mx proteins [36] . We therefore propose that passage of viruses in hosts with weak or inactive Mx proteins ( such as swine or laboratory mouse strains , respectively ) would result in a loss of MxA resistance . In fact , experimental adaptation of the pandemic 2009 virus to Mx1-negative mice led to the acquisition of mutations in NP which strongly diminished Mx resistance , such as D101G [44] ( Figure S9 ) . Similarly , mutations conferring MxA resistance were lost ( D53E , D101G , H289L ) following re-transmission of pH1N1 from human to swine ( Figure 4 ) . The mechanism by which MxA exerts its antiviral function during infection or in the polymerase reconstitution assays is presently not known . MxA may block the viral life cycle at several early steps by interfering with the processes of vRNP entry and intracellular transport [45] , as well as primary [45] and secondary transcription [46] . We proposed a model in which MxA recognizes vRNPs and begins to self-assemble into rings , thereby sterically inhibiting vRNP function [15] , [18] , [19] . This model has recently been suggested also for mouse Mx1 , but in a modified version involving in addition also the polymerase subunit PB2 [28] , in agreement with previous functional work implicating PB2 as putative target of mouse Mx1 [47] , [48] . Modeling the present MxA resistance clusters into the available vRNP structure [49] revealed that the sites on NP are most likely solvent-exposed , and thus accessible to cellular factors . Initial contact of MxA to single binding sites on NP might be weak but reinforced by oligomerization , involving multiple repetitive contacts exposed on the many NP molecules that form the vRNP . Weak but extensive contacts to repetitive viral target motifs have been demonstrated for other intracellular restriction factors . For example , TRIM5α specifically binds to several surface-exposed amino acids of the capsid protein of HIV-1 , thereby forming an array or lattice on top of the viral capsid [50] , [51] . To date , physical interaction between influenza A virus NP and MxA could be demonstrated after covalent protein crosslinking [52] and thus MxA might also bind to free NP in the cytoplasm thereby blocking polymerase activity indirectly . However , it is also likely that further cellular proteins modulate MxA activity and its interaction with viral proteins . Previous work identified a number of potential MxA co-factors , but their contribution is still unclear [53]–[56] . One promising candidate is the helicase UAP56 , a DEAD box RNA helicase which was shown to prevent double-strand RNA formation and subsequent innate immune activation in influenza virus-infected cells . UAP56 binds both MxA and NP [29] , [57] , in the latter case via the N-terminal domain of NP which contains the MxA resistance-enhancing mutation G16D [58] . Nonetheless , the significance of this observation and the role of UAP56 for antiviral activity remain to be demonstrated . In summary , we have found functional and evolutionary evidence that the human MxA GTPase provides an efficient barrier against zoonotic introduction of influenza A viruses into the human population . Thus , the human MxA is a significant driving force in influenza A virus nucleoprotein evolution . We therefore propose that amino acids known to contribute to MxA resistance should be monitored as a strong indicator for the pandemic potential of newly emerging influenza A viruses . All animal experiments were performed in compliance with the German animal protection law ( TierSchG ) . The mice were housed and handled in accordance with good animal practice as defined by FELASA ( www . felasa . eu/guidelines . php ) and the national animal welfare body GV-SOLAS ( www . gv-solas . de/index . html ) . The animal welfare committees of the university of Freiburg , as well as the local authorities ( Regierungspräsidium Freiburg ) approved all animal experiments . Canine MDCKII , porcine NSK and NPTr cells [34] , and human HEK 293T cells were maintained in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal calf serum , 2 mM L-glutamine and 1% penicillin-streptomycin . Chicken hepatocellular epithelial cell line ( LMH ) [59] was grown in DMEM supplemented with 8% fetal calf serum 2% chicken serum , 2 mM L-glutamine and 1% penicillin/streptomycin . The pHW2000 rescue plasmids [60] and pCAGGS expression plasmids [61] coding for NP were used for site directed mutagenesis . The coding region of MxA was cloned into pCAGGS , whereas murine Mx1 was expressed using pcDNA 3 . 1 [21] . The cDNA of porcine Mx1 ( poMx1 ) corresponding to the full length 1992 nt long open reading frame as described in [62] , encoding a Flag tag at its 5′-end was cloned into pCAGGS using KpnI and XhoI . poMx1 cDNA was generated from mRNA isolated from IFNα-2a-treated cell cultures from domestic pig ( Sus scrofa domestica ) . The recombinant viruses A/Hamburg/4/09 ( pH1N1 ) and A/Thailand/1 ( KAN-1 ) /04 ( H5N1 ) , and the NP-mutant viruses were generated by the eight-plasmid reverse-genetics system as described previously [21] . All recombinant viruses were plaque purified on MDCKII cells . Virus stocks were prepared on MDCKII cells and titers were determined by plaque assay . HEK 293T cells seeded in 12-well plates were transfected using the Nanofectin transfection reagent ( PAA Laboratories ) according to the manufacturer's protocol . 10 ng of pCAGGS plasmids encoding PB2 , PB1 , and PA and 100 ng of NP-encoding plasmid were cotransfected with 100 ng of the firefly luciferase- encoding viral minigenome construct pPolI-FFLuc-RT , which is flanked by the noncoding regions of segment 8 of influenza A virus . The transfection mixture also contained 30 ng of pRL-SV40 , a plasmid constitutively expressing Renilla luciferase under the control of the simian virus 40 promoter to normalize variations in transfection efficiency . To evaluate the antiviral potential of Mx1 and MxA , we cotransfected Mx1- or MxA- encoding plasmid . A simultaneous experiment with cotransfection of the antivirally inactive mutants Mx1-K49A or MxA-T103A , respectively , was used as a control . To achieve equal amounts of transfected DNA , an empty vector plasmid was added . Twenty-four hours post transfection , cells were lysed and firefly and Renilla luciferase activities were measured using the dual luciferase reporter assay ( Promega ) according to the manufacturer's protocol . Reconstitution of the viral polymerase complex in avian and porcine cells was performed as above with the exception that the minigenome RNA was expressed under the control of a chicken [63] or porcine Pol I promoter ( pSPOM2 ) [34] . MDCKII cells seeded in 6-well plates were incubated with virus at a multiplicity of infection ( MOI ) of 0 . 001 in PBS+/+ containing 0 . 2% BSA for 1 h at 37°C . The inoculum was removed and 3 ml infection medium ( DMEM supplemented with 0 . 2% BSA ) , additionally containing 1 µg/ml TPCK-treated trypsin for pH1N1 viruses , was added . Virus titers in cell culture supernatants were determined at the indicated time points by plaque assay and are expressed as PFU per ml . For determination of viral transcript levels in virus-infected MDCKII cells , cells were seeded in 6-well plates and infection was carried out with infection media . After the indicated time point post infection , cells were harvested in Trizol™ and RNA was purified according to the manufacturer's protocol ( Invitrogen ) . Primer extension analysis was performed as described [63] using specific primers for the NA segment ( mRNA , cRNA and vRNA ) and cellular 5sRNA . BALB/c mice were obtained from Janvier ( Straßburg ) and congenic BALB . A2G-Mx1 mice ( designated BALB-Mx1 ) carrying the functional Mx1 allele [64] were bred locally . Six- to eight-week-old mice were anesthetized with a mixture of ketamine ( 100 µg per gram body weight ) and xylazine ( 5 µg per gram ) administered intraperitoneally ( i . p . ) and inoculated intranasally ( i . n . ) with the indicated doses of viruses in 50 µl phosphate-buffered saline ( PBS ) containing 0 . 2% bovine serum albumin ( BSA ) . Mice were monitored daily for weight loss until 14 days postinfection ( p . i . ) . Animals with severe symptoms or more than 25% weight loss were euthanized . Lung homogenates were prepared using the FastPrep24 system ( MP Biomedicals ) . Briefly , after addition of 800 µl of PBS containing 0 . 2% BSA , lungs were subjected to two rounds of mechanical treatment for 10 s each at 6 . 5 m/s . Tissue debris was removed by low-speed centrifugation . The LD50 values were calculated based on the infectious dose ( PFU ) . All animal work was conducted under BSL 3 conditions in accordance with the guidelines of the local animal care committee . The program PyMOL ( www . pymol . org ) was used to assign the indicated positions in the structural model of the NP of A/HK/483/97 ( H5N1 ) ( PDB code:2Q06 ) . The program I-TASSER ( zhanglab . ccmb . med . umich . edu/I-TASSER ) was used to generate a full length NP model of A/Thailand/1 ( KAN-1 ) /04 ( H5N1 ) , including amino acids 1–20 . Alignments and phylogenetic analyses were conducted with MEGA5 [65] . For maximum likelihood ( ML ) tree inference , the GTR substitution model assuming gamma distribution ( four gamma categories ) and invariant sites was selected , and the initial tree was made automatically . Bootstrap analysis was performed with 1 , 000 replications . The optimal substitution model was selected on the basis of the Bayesian information criterion ( BIC ) and the corrected Akaike information criterion ( AICc ) using a model test implemented in MEGA5 .	Influenza A viruses of avian or swine origin sporadically enter into the human population but do not transmit between individuals . In rare cases , however , they establish a new virus lineage in humans . The mechanisms by which invading viruses overcome the species barrier are not well understood , but multiple adaptations to the new host are required . Surprisingly little is known about adaptive mutations that overcome restriction factors of the intrinsic and innate host defense system . In this study , we have identified adaptive mutations in pandemic strains A/Brevig Mission/1/1918 and A/Hamburg/4/2009 that confer resistance to the interferon-induced antiviral factor MxA which is a dynamin-like large GTPase that recognizes the incoming viral nucleocapsids and blocks their function . The resistance-enhancing mutations changed several amino acids in the viral nucleoprotein which is the main nucleocapsid component . These mutations were sufficient to increase the pathogenicity of an avian influenza virus strain in a Mx-positive mouse model . Interestingly , the resistance-associated amino acids are counter-selected in circulating avian influenza strains , because they compromise general viral replication fitness . The present data indicate that the innate immunity factor MxA provides a barrier against zoonotic introduction of influenza A viruses and that adaptive mutations in the nucleoprotein must be carefully monitored .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "medicine", "infectious", "diseases", "immunity", "virology", "innate", "immunity", "biology", "microbiology", "viral", "diseases", "viral", "evolution" ]	2013	Pandemic Influenza A Viruses Escape from Restriction by Human MxA through Adaptive Mutations in the Nucleoprotein
Lymphatic filariasis is a chronic , disabling and often disfiguring condition that principally impacts the world’s poorest people . In addition to the well-recognised physical disability associated with lymphedema and hydrocele , affected people often experience rejection , stigma and discrimination . The resulting emotional consequences are known to impact on the quality of life and the functioning of the affected individuals . However , the management of this condition has focused on prevention and treatment through mass drug administration , with scant attention paid to the emotional impact of the condition on affected individuals . This study aimed to determine the prevalence and severity of depression among individuals with physical disfigurement from lymphatic filariasis in Plateau State , Nigeria . A cross-sectional 2-stage convenience study was conducted at 5 designated treatment centers across Plateau State , Nigeria . All available and consenting clients with clearly visible physical disfigurement were recruited . A semi-structured socio-demographic questionnaire , Rosenberg Self-esteem and a 9-item Patient Health Questionnaire ( PHQ-9 ) were administered at the first stage . Those who screened positive ( with a PHQ-9 score of five and above ) were further interviewed using the Depression module of the Composite International Diagnostic Interview ( CIDI ) . Ninety-eight individuals met the criteria and provided consent . Twenty percent of the respondents met criteria for depression , with the following proportions based on severity: Mild ( 42 . 1% ) , Moderate ( 31 . 6% ) and Severe ( 26 . 3% ) . History of mental illness ( OR 40 . 83 , p = 0 . 008 ) ; Median duration of the illness was 17 years ( IQR 7 . 0–30 years ) and being unemployed ( OR 12 . 71 , p = 0 . 003 ) were predictive of depression . High self-esteem was negatively correlated ( OR 0 . 09 , p<0 . 004 ) . Prevalence of depression is high among individuals with lymphatic filariasis and depression in sufferers is associated with low self-esteem and low levels of life satisfaction . Neglected Tropical Diseases ( NTDs ) are a group of disabling conditions that are among the most common infections affecting the world’s poorest people [1] . Many NTDs lead to chronic and often disfiguring conditions that result in significant disability and affect more than 1 billion people across the world [2] . Lymphatic filariasis ( LF ) is a mosquito-borne disease caused by filarial parasitic worms like Wuchereria bancrofti , Brugia malayi and Brugia timori [3] . Global estimates suggest that 120 million are affected in 80 countries throughout the tropics and subtropics with people at risk exceeding 1 . 3 billion [4 , 5] . LF often manifests as enlargement of the entire leg or arm , the genitals ( scrotal hydrocele in men ) , vulva and breasts leading to significant physical disfigurement [3] . Significant social stigma is associated with this stage of the disease [6] , and the psychosocial problems linked with the condition are believed to be more severe than the physical ones [7] . In the 2010 Global Burden of Disease study , the Disability Adjusted Life Years ( DALYs ) associated with depressive illness in lymphatic filariasis was found to be twice that of the physical consequences of the disease ( 5 . 09 million global DALYs vs 2 . 78 million respectively ) [8] . Despite this report , very few studies have been carried out to explore the association between depressive illness and lymphatic filariasis . This study , therefore , aimed to determine the prevalence and severity of depression and the social and clinical factors associated with depression in individuals with lymphatic filariasis in Plateau State , North Central Nigeria . Ethical clearance for the study was obtained from the Institutional Research Ethical Committee of Jos University Teaching Hospital , Plateau State . Signed or thumb-printed written consent was individually obtained from each participant after due explanation of the purpose of the study and the voluntary nature of their participation . Respondents below the age of 18 years , required the consent of their parent or guardian in addition to their assent . The medical data as well as respondent scores on the instruments were anonymized in order to protect confidentiality . Data analysis was carried out using the Statistical Package for Social Sciences ( SPSS ) , version 21 , software using descriptive statistics to yield frequencies , percentages and proportions . Level of significance was kept at 5% . Logistic regression was used to identify predictors of depression among the respondents in this study using a few of the co-variates . A total of ninety-eight participants , who met the inclusion criteria and gave consent for the study , were interviewed . Ninety-four ( 95 . 9% ) had full documentation and were analysed . The majority , 58 respondents ( 61 . 7% ) , were female . Other socio-demographic details are provided in Table 1 . The median duration of illness was 17 years ( IQR 7 . 0–30 . 0 years ) . Twenty-three ( 24 . 5% ) had the illness for more than 30 years . 21 respondents ( 22 . 3% ) rated their level of functioning as poor . Other details of functioning , as well as perceived adequacy of support are presented in Table 2 . Nineteen respondents ( 20% ) met criteria for depression , using CIDI , with the severity of the depression being Mild [8 ( 42 . 1% ) ] , Moderate [6 ( 31 . 6% ) ] and Severe [5 ( 26 . 3% ) ] . See Fig 1 . Fourteen of the depressed respondents ( 73 . 7% ) were female . Furthermore , 69 respondents ( 73 . 4% ) reported low self-esteem . See Table 3 . Logistic regression analysis revealed that history of mental illness ( OR 40 . 83 , p = 0 . 008 ) ; duration of the illness between 11–20 years ( OR 5 . 02 , p = 0 . 079 ) , being unemployed ( OR 12 . 71 , p = 0 . 003 ) and Self Esteem ( OR 0 . 09 , p = 0 . 004 ) were predictive of depression in the cohort . High self-esteem was negatively correlated to depression . See Table 4 below . Beyond the physical burden of living with lymphatic filariasis , people with the condition also have significant psychiatric complications ( particularly depression ) . Appropriate treatment for depression has been found to improve outcomes in other chronic conditions [22] . Therefore , given the high prevalence of depression , providing access to mental health screening and interventions should be integral to NTD programmes . Thus , in addition to attending to physical needs , emotional needs should also be routinely assessed and catered for . It is recommended that staff should be sensitised to the high risk of depression and be trained to recognize basic signs and symptoms of depression . Simple screening instruments for depression , such as the PHQ-9 can also be utilized in routine clinics , and those found to be depressed can be treated or referred appropriately . Health promotion strategies geared towards assessing and addressing factors associated with depression ( for example self-esteem ) should be incorporated in routine community engagement , including health talks in the clinic . These are simple steps that can easily be incorporated into specific NTD services , and general health care for endemic populations , providing the possibility of improving the quality of life of affected persons , and reducing the negative impact of depression on an already marginalised population .	Lymphatic filariasis is a chronic illness that is disabling and often results in disfigurement . Affected people experience rejection , and stigma and discrimination , which can result in significant emotional consequences . Overall functioning and the quality of life of such individuals can be further affected by this exclusion and psychosocial impacts . Little or no attention is presently paid to the emotional impact of this disease in the overall management of people affected . The study , therefore , aimed to determine the prevalence and severity of depression , as well as associated socio-demographic factors , in individuals with physical disfigurement from lymphatic filariasis in Plateau State , Nigeria . Ninety-four consecutive consenting individuals with physically disfiguring lymphatic filariasis at 5 established treatment centers across Plateau State , Nigeria , were recruited and had semi-structured sociodemographic , Patient Health ( PHQ-9 ) , the depression module of Composite International Diagnostic Interview ( CIDI ) and Rosenberg Self-esteem questionnaires administered using a 2-stage design . Twenty percent of the sample were found to be depressed , while history of mental illness , duration of the illness , being unemployed , and religion were predictive of depression . High self-esteem was negatively correlated . The study underscores the need to go beyond just the physical needs of individuals with lymphatic filariasis . Management must be holistic and attention must be focused on the emotional sequelae of lymphatic filariasis .	[ "Abstract", "Introduction", "Methods", "Results", "Conclusion" ]	[ "medicine", "and", "health", "sciences", "sociology", "tropical", "diseases", "geographical", "locations", "social", "sciences", "parasitic", "diseases", "health", "care", "filariasis", "neglected", "tropical", "diseases", "lymphatic", "filariasis", "africa", "mood", "disorders", "social", "discrimination", "health", "risk", "analysis", "mental", "health", "and", "psychiatry", "economics", "nigeria", "people", "and", "places", "economic", "history", "helminth", "infections", "quality", "of", "life", "depression" ]	2017	Prevalence of depression and associated clinical and socio-demographic factors in people living with lymphatic filariasis in Plateau State, Nigeria
Using a chromatin immunoprecipitation-paired end diTag cloning and sequencing strategy , we mapped estrogen receptor α ( ERα ) binding sites in MCF-7 breast cancer cells . We identified 1 , 234 high confidence binding clusters of which 94% are projected to be bona fide ERα binding regions . Only 5% of the mapped estrogen receptor binding sites are located within 5 kb upstream of the transcriptional start sites of adjacent genes , regions containing the proximal promoters , whereas vast majority of the sites are mapped to intronic or distal locations ( >5 kb from 5′ and 3′ ends of adjacent transcript ) , suggesting transcriptional regulatory mechanisms over significant physical distances . Of all the identified sites , 71% harbored putative full estrogen response elements ( EREs ) , 25% bore ERE half sites , and only 4% had no recognizable ERE sequences . Genes in the vicinity of ERα binding sites were enriched for regulation by estradiol in MCF-7 cells , and their expression profiles in patient samples segregate ERα-positive from ERα-negative breast tumors . The expression dynamics of the genes adjacent to ERα binding sites suggest a direct induction of gene expression through binding to ERE-like sequences , whereas transcriptional repression by ERα appears to be through indirect mechanisms . Our analysis also indicates a number of candidate transcription factor binding sites adjacent to occupied EREs at frequencies much greater than by chance , including the previously reported FOXA1 sites , and demonstrate the potential involvement of one such putative adjacent factor , Sp1 , in the global regulation of ERα target genes . Unexpectedly , we found that only 22%–24% of the bona fide human ERα binding sites were overlapping conserved regions in whole genome vertebrate alignments , which suggest limited conservation of functional binding sites . Taken together , this genome-scale analysis suggests complex but definable rules governing ERα binding and gene regulation . Cellular transcriptomes are dictated by complex interactions between signal transduction pathways , general and specific transcription factors , chromatin remodeling proteins , and the RNA polymerase complexes . Precise transcriptional responses are achieved in part by targeting transcription factor complexes to the cis-regulatory regions of target genes via specific binding site sequences . The importance of these binding site sequences is reflected in the conservation of sequence motifs in coregulated genes and through evolution . A number of studies have examined transcriptomic changes to breast cancer cells following estrogen treatment [1–7] . Estrogen receptors ( ERs ) ( specifically ERα and ERβ ) are ligand-dependent transcription factors that mediate cellular responses to hormone exposure in vertebrate development , physiological processes , and endocrine-related diseases . ERα , in particular , has been implicated in the etiology of breast cancer and is a major prognostic marker and therapeutic target in disease management [8] . At the molecular level , ERs interact either directly with genomic targets encoded by estrogen response elements ( EREs ) ( 5′-GGTCAnnnTGACC-3′ ) or indirectly by tethering to nuclear proteins , such as AP-1 and Sp1 transcription factors [9–11] . The mechanisms of ER binding site specificity , however , are not clear since these binding site sequence motifs are ubiquitous in the genome , and there is no discernable difference between functional and nonfunctional sites by computational modeling approaches . This ambiguity is likely due to a lack of systemic information on binding site usage and architecture and mechanistic complexity involving additional transcription factors and epigenetic modifications [12 , 13] . Chromatin immunoprecipitation ( ChIP ) assays have facilitated characterizations of in vivo protein-DNA complexes such as histone modifications and recruitment of transcription factor complexes to specific binding sites [14] . Coupled with microarray technology , the ChIP-on-chip experiments have resulted in more global binding site maps for a number of human transcription factors in proximal promoters and in specific genomic regions . For example , Carroll and colleagues recently mapped ER binding sites first in human Chromosomes 21 and 22 and more recently across the entire human genome [15 , 16] . In spite of the success of ChIP-on-chip studies , there remain caveats regarding probe specificity and performance , including constraints on probe design in certain genomic regions and potential biases introduced by amplification protocols . Thus , the analysis of ERα binding sites using alternative genome-wide technologies is warranted . Previously we developed a high throughput cloning and sequencing approach for mapping full-length transcripts . By employing specialized cloning techniques and vectors , paired-end diTags ( PETs ) from ends of transcripts and this gene identification signature ( GIS ) can be sequenced and mapped precisely to the genome [17] . The GIS-PET technique increases sequencing efficiency by 30-fold as compared to sequencing the entire transcript insert . We subsequently showed that binding site fragments from ChIP experiments can also be subjected to PET analysis to generate an unbiased whole-genome map of p53 tumor suppressor protein binding sites and demonstrated an association of binding sites and adjacent target genes with p53 functions in patient tumor samples [18] . The PET technology has also been utilized to study OCT4 and Nanog binding sites in stem cell biology [19] . To obtain a global map of ERα binding sites in breast cancer cells , we applied ChIP-PET to generate a library of ERα binding sites in MCF-7 cells following estrogen treatment . We then combined the binding site data with hormone-responsive and breast tumor sample microarray gene expression studies to discern correlations between ER binding , transcriptional activity , and disease status and outcome in patients . We also compared our findings with data from ChIP-on-chip studies to evaluate the performance of each respective technology . Herein , we describe the comprehensive cartographic results and outline the insights they provide into binding site usage and molecular mechanisms of ER transcriptional regulatory functions . Hormone-deprived MCF-7 cells were treated with 10 nM estradiol for 45 min , and then DNA-bound receptor complexes were isolated through ChIP using anti-ERα antibodies ( HC-20 , Santa Cruz Biotechnology , http://www . scbt . com ) . Prior to generating the PET library for sequencing , we qualified the ChIP products by measuring DNA fragment size and enrichment of known ER binding site in the pS2/TFF1 gene promoter after immunoprecipitation to ensure sample quality . The ChIP DNA fragments ranged from 300 bp to 1 kb , and there was an 80-fold enrichment of ER binding at the known pS2/TFF1 ERE as compared to the irrelevant antibody control . Once ChIP DNA quality had been confirmed , the PET library was constructed and sequenced as described previously [18] . A total of 635 , 371 PETs were sequenced , of which 361 , 241 ( ~56 . 86% ) were unambiguously mapped to unique loci in the human genome ( hg17/NCBI build 35 ) and localized to 136 , 158 distinct genomic coordinates . One of the first questions we asked was whether the sequencing of these 635 , 371 clone “equivalents” in the form of pair-end tags provided sufficient representation of the chip library . To assess the degree of saturation of the PET library sequenced ( the total number of distinct ChIP DNA fragments that can be captured from the library ) , we fitted a Hill Function [20] using extrapolated and historical sequencing data ( see Figure S1 ) . The degree of saturation of the ER ChIP PET library is estimated at 73 . 24% ( 136 , 158 actual/185 , 915 expected ) , suggesting that ~73% of the extrapolated hypothetical limit of coverage by our library was sequenced . Sequencing results are summarized in Figure 1A . Overlapping uniquely mapped PETs that form PET clusters ( see Figure 1B ) have been shown to be highly enriched for “true” binding sites [18] . We previously set the selection parameter ( i . e . , number of PETs within a given cluster ) for high probability binding regions by using the goodness-of-fit analysis employing a mixture of two standard Pareto distributions to model the signal and noise within the dataset [18] . MCF-7 cells , however , pose a special analytical challenge in that regions of gene amplification in the cell line also appeared to amplify cluster PET numbers from low quality binding sites ( Figure S2 ) . Therefore , we devised an alternative strategy that normalizes regions of gene amplification so that all chromosomal regions can be directly compared . Using this “adaptive maximum overlap PET ( moPET ) threshold” approach ( unpublished data ) and setting the false-positive rate of <0 . 01 , 1 , 234 moPET3+ ER binding site clusters were then defined as high quality binding regions and were used for all subsequent analysis . Among the high quality binding regions , 45% are moPET3 clusters , another 20% are moPET4s , and the remaining 35% have five or more PETs in overlap regions within the clusters ( see Figure 1C ) . An indication that the ChIP-PET experiment and the analytical methods were properly executed is that many known ER binding sites , including the pS2/TFF1 , GREB1 , ADORA1 , and CYP1B1 EREs are present in the defined set of regions [6 , 21 , 22] . The complete list of 1 , 234 high quality binding regions and their chromosomal location can be found in Table S1 . ERα binding regions defined by ChIP-PET are located in every chromosome in the human genome , with the exception of the Y chromosome , which is not present in MCF-7 cells derived from a female breast cancer patient ( see Figure S3A and S3B ) . When regions of gene amplification are accounted for , the frequency of binding clusters per chromosome generally corresponds to the size and gene density of the chromosome , and ER does not appear to localize to specific chromosomal regions within the genome . Figure S3C shows ER binding clusters distribution relative to the nearest University of California Santa Cruz ( UCSC ) Known Genes ( KG ) ( see Materials and Methods ) . Binding regions were mapped to the precise positions relative to the 5′ and 3′ ends of transcripts in the UCSC KG database . Only 5% of the regions map to the proximal promoter regions , defined as 0–5 kb upstream of the transcriptional start site ( TSS ) , where the vast majority of current known EREs have been identified and characterized thus far . The largest fraction ( 38% ) of binding regions map to intragenic regions of transcripts and are localized to introns , whereas 23% are within 100 kb from the 5′ start sites , and 19% are within 100 kb of 3′ polyadenylation sites . Only 20% of the ER binding regions are located in gene deserts where the nearest KG is >100 kb away . These findings initially suggest that DNA-bound ER can interact with the transcriptional machinery through both proximal- and distal-acting mechanisms , and these interactions are not likely to be limited by binding site orientation ( 5′ or 3′ ) relative to the TSSs . Intriguingly , functional ER binding sites were rarely in exons and when in exons were in probable untranslated regions . We did not detect any binding regions that mapped to a protein-coding domain of a transcriptional unit . These observations further suggest a dynamic selection of ER binding sites that excludes exonic regions and raise the possibility that transcription factor binding sites ( TFBSs ) in exons may undergo negative selection during evolution . As a preliminary assessment on the fidelity of the 1 , 234 high quality binding regions , we considered presence of putative ERE motif as a proxy for a real binding event . A total of 13 base pair sites that were at most two Hamming distance away ( i . e . , two base deviation ) from the consensus ERE ( GGTCA-nnn-TGACC ) were called putative EREs . Upon scanning the 1 , 234 binding regions , we discovered that 884 ( ~71% ) binding regions contained at least one ERE-like sequence , 25% encoded putative half-ERE sites , and the remaining 4% bore no ERE sequence motifs whatsoever ( Figure 2 ) . To further confirm the validity of the discovered binding regions , we selected 107 out of the 1 , 234 high quality clusters for further ChIP-qPCR validation ( Figure 3A ) . All clusters containing a full putative ERE ( 47 sites ) showed significant ( >2-fold over control ) enrichment , and based on this 100% success rate the 884 genomic loci of ER binding containing consensus ERE motifs are highly likely to encompass true ER binding sites . We also tested 37 sites with half EREs and validated ER binding in 84% ( 31 of 37 ) of selected sites . A similar success rate was found for the non-ERE ChIP-PET clusters , as 19 out of 23 tested sites ( 83% ) showed binding . The median fold enrichment of the validated sites containing a full ERE was 81-fold , which was considerably higher than the median fold enrichment observed for half- and non-ERE-containing PET clusters ( 36- and 51-fold , respectively ) . These results support the idea that EREs tend to encode high affinity ER binding sites whereas the half- and non-ERE binding likely support moderate affinity binding , perhaps through indirect tethering mechanisms . This is further supported by the enrichment of EREs as the number of PETs in the moPET clusters increase , corresponding to higher ChIP efficiency and potentially reflecting the higher affinity binding . The positive gradient of the curve supports the notion that higher moPET clusters are more likely to contain full ERE-like sequences ( p = 3 . 204e−8 ) ( Figure 3B ) . Thus , the canonical ERE sequences appears to be a hallmark of ER control on a genome-wide scale . Out of the ten clusters that failed validation , eight of the loci that were misclassified as true binders belonged to the moPET 3 category , which would be considered in the borderline confident range; one was a moPET 4 cluster , whereas a false positive in a moPET7 was located in an amplified region of the MCF-7 genome on Chromosome 20 ( which we have shown overestimates the binding efficiency of a DNA fragment to ER ) . When adjusted for the frequency of full , half , and no ERE motifs in the PET-defined binding loci , our validation rate for binding calls is 94% . These validation results are in line with our previous whole-genome ChIP-PET analysis of p53 , Oct4 , and Nanog binding sites [18 , 19] and compares favorably with other genome-wide technologies such as ChIP-on-chip . To examine whether ER ChIP-PET binding sites containing putative full EREs harbor transcriptional enhancer activities , we PCR amplified 11 binding sites from MCF-7 genomic DNA and cloned them upstream of the pGL4-TATA luciferase reporter . For negative and positive controls we used pGL4-TATA and pGL4-2ERE-TATA , respectively . These constructs were transiently transfected into hormone-depleted MCF-7 cells , treated with either ethanol or E2 for 18–24 hours , and then assayed for luciferase activity . A total of eight out of 11 ER ChIP-PET binding site reporter constructs tested were E2 responsive ( Figure 4A ) . To show that the transcriptional enhancer activities of the ER ChIP-PET binding sites are mediated via EREs , we mutated the putative ERE motifs in the eight E2 responsive constructs and transfected them into MCF-7 cells . Destroying the putative EREs resulted in either significant reduction or complete loss of estrogen response , thus demonstrating the EREs of the ER ChIP-PET binding sites are responsible for their enhancer properties ( Figure 4B ) . Other technologies have been used to map ER binding sites on a genomic scale . Carroll et al . using ChIP and human genome tiling arrays have mapped ER binding sites in Chromosomes 21 and 22 and across the human genome in MCF-7 cells [15 , 16] . We first compared our mapped ER binding sites to the previously published results in Chromosomes 21 and 22 [15] . In the 1 , 234 binding sites mapped by ChIP-PET , we detected 36 ER binding sites in these chromosomes and found that 20 binding sites were identified by both techniques ( Figure 5A ) and 16 sites were unique to ChIP-PET . We then tested ER binding to technology-specific and overlapping sites by conventional ChIP to determine the validity of the mapped sites . We selected 25 sites detected by one technique or by both for further validation by ChIP and quantitative PCR ( Figure 5B ) . The six sites identified by both technologies were validated as bona fide ER binding sites and had the greatest fold enrichment ( median ≈ 300× ) . All nine of the sites discovered only by ChIP-PET were validated compared to nine of the ten selected sites discovered by ChIP-on-chip alone . The median fold enrichment of the sites identified solely by the ChIP-PET approach was higher than that of the ChIP-on-chip ( medians ~45× versus ~22× ) . A total of four of the ten sites discovered only by ChIP-on-chip were also associated with ChIP-PET clusters but not deemed high probability sites since three had two PETs in their cluster , and one site was a one PET singleton ( unpublished data ) . Moreover , these four ChIP-on-chip sites overlapping with lower probability PET clusters had higher fold enrichment for ER binding than the remaining sites with only ChIP-on-chip supporting evidence ( ~37× versus ~12× median ) , suggesting that conjoint assignment of sites by the two technologies even at suboptimal thresholds may identify higher quality ER binding sites . To assess the two technology platforms across the entire genome , we also compared the 1 , 234 ChIP-PET ER binding sites identified in this study to the recently published whole-genome ChIP-on-chip ER binding site map of 3 , 665 binding sites [16] and found 624 ( 50 . 6% ) ChIP-PET sites in common with the ChIP-on-chip data and 610 ( 49 . 4% ) sites unique to the PET technology ( see Figure 5C ) . These results are consistent with the data of Chromosomes 21 and 22 where 44 . 4% ( 16/36 ) of the ChIP-PET sites are unique ( see Figure 5A ) . It is likely that the difference between the two platforms are due to lower affinity ER binding sites being more susceptible to constraints and limitations inherent in the detection technologies and to possible differences in biological handling of cell lines in each study . Moreover , there appears to be content differences between the discovery capacity of the two technologies . An inherent disadvantage of the ChIP-on-chip approach is that arrays mask sites that contain repetitive sequences , whereas the output of the ChIP-PET technology is completely unbiased in regard to the presence or absence of repetitive sequences . To this point , we found that ~27 . 9% of the base pairs in bona fide binding sites discovered by the pair-end tag approach were associated with repetitive sequences , whereas 5 . 3% of those in Carroll et al . bore repeats . Taken together , these results further suggest that the ChIP-PET and the ChIP-on-chip are complementary methods for the identification of TFBSs in a whole genome manner . The binding site map denotes ER transcriptional regulatory potential for a large number of genes . To determine the specific transcriptional responses corresponding to estrogen treatment and ER recruitment to cis-regulatory sites in MCF-7 cells , we performed gene expression profiling experiments using Affymetrix U133 microarrays on a time course following estradiol exposure . Differentially expressed genes were selected based on a q-value cut-off of less than 2% using a stringent significance analysis of microarrays analysis algorithm . We identified 802 probe sets , representing 544 unique named genes , whose expression levels were up-regulated in response to 10 nM E2 treatment for 12 , 24 , or 48 h , and 1 , 168 probe sets corresponding to 704 unique named genes , were down-regulated following hormone treatment . When combined with the ER binding site mapping data ( within 100 kb of and closest to high quality binding regions identified by ChIP-PET ) , 171 up-regulated genes and 116 down-regulated genes were associated with high confidence ER binding sites . Table S2 contains the complete listing of estrogen responsive genes with adjacent ER binding sites identified in this study . We next examined whether there was a preference for positioning of the ER binding sites in up-regulated versus down-regulated genes . Our analysis revealed that there was a statistically significant association between the presence of an ER ChIP-PET cluster near an up-regulated gene and an under-representation of ER ChIP-PET clusters associated with down-regulated genes ( p = 8 . 379e−08 ) ( see Table 1 ) . The over-representation of ER ChIP-PET clusters that can be associated with E2-up-regulated genes is particularly noticeable at the early 12-h time point . That more binding site clusters are associated with E2 up-regulated genes ( 60% ) compared with down-regulated genes ( 40% ) suggests that the ER protein more frequently is directly involved in the transcriptional regulation of E2 up-regulated genes as compared to down-regulated genes . This finding is in concordance with a previous study we conducted in T-47D human mammary carcinoma cells , where we found that out of 89 genes identified as direct target genes ( i . e . , E2-responsive and cycloheximide insensitive ) , 59 ( 66 . 3% ) were up-regulated and only 30 ( 33 . 7% ) were down-regulated by E2 treatment [6] . To further confirm the observed association of binding sites with the transcriptional response , we examined the association of the 107 sites validated by conventional ChIP qPCR to E2 regulated gene expression data . Of the 107 sites tested , 22 sites were found to be associated with an E2-regulated probe from the Affymetrix dataset . Out of these 22 sites , 16 were up-regulated by E2 whereas six were down-regulated by E2-treatment . The 16 sites associated with E2 up-regulated probes all showed high levels of enrichment ( ~25–473× ) when analyzed by ChIP qPCR , whereas the six sites associated with down-regulated genes showed lower levels of enrichment ( ~1–25× ) ( Figure 6A ) . These data suggested that some potential mechanistic association exist between high affinity ER binding sites and induction of gene expression by ER . Exploring the association between ER binding sites and the directionality of the transcriptional response further , we mapped the locations of the binding sites relative to the start and termination sites of E2 up- and down-regulated genes as assessed by time course studies using Affymetrix expression arrays . As shown in Figure 6B , binding clusters associated with E2-induced genes are found , significantly above background , 50 kb upstream of the TSS , 50 kb downstream of the TSS within the early introns , and within 25 kb downstream of the termination site of their associated genes . Intriguingly , approximately 35 ER binding sites were found within 5 kb of the transcriptional initiation sites of up-regulated genes ( Figure 6B ) . No such enrichment was detected with down-regulated genes or with genes randomly selected from the UCSC KG database . Taken together , our results suggest that ER binds to many sites in the genome , most bearing a discernable ERE-like sequences , and that genes induced by estrogen are significantly more likely to have an ER binding site within 50 kb of the TSS . Manual analysis of the intronic binding sites showed no evidence for internal alternative TSSs . Genes down-regulated by estrogen show no such positional enrichment and appear to be associated with lower ChIP efficiency ER binding sites . Moreover , genes repressed by estrogen are usually down-regulated later than those induced ( 48 h versus 12 h; as also observed elsewhere [3 , 16] ) . This suggests that genes repressed by ER may require further synthesis and recruitment of other factors to the ER binding sites and that the mechanism of gene repression is topographically distinct from that of gene induction . Supporting this is our observation that binding sites for up-regulated genes have higher moPET counts than binding sites for down-regulated genes ( p = 0 . 0005 ) . When sampled for validation by quantitative PCR , ER binding sites associated with induced genes had much higher fold enrichment for ER occupancy than repressed genes ( see Figure 6A ) . We posited that the genes adjacent to the ER binding sites identified by ChIP-PET are putative ER target genes and should reflect ER function in vivo . To assess this possibility , we examined whether the behavior of the collection of adjacent genes could determine the ER status of human breast cancers . All genes within 100 kb of an ER binding site or with an ER binding cluster within an intron were used to cluster 251 breast tumors from a cohort from Uppsala , Sweden previously analyzed for gene expression using Affymetrix U133 A and B arrays [23] . Our results showed that this proximate gene list could easily segregate ER-positive and ER-negative breast tumors ( see Figure 7A ) , whereas a random gene list from the U133 A and B gene set could not ( unpublished data ) . Statistical analysis using the Fisher's exact test showed a highly significant separation based on ER-status with p = 3 . 914e−12 . Moreover , patients with ER-positive like expression profiles , based on the ER-associated genes , have better disease-specific survival over ten years of follow-up as compared to those with the ER-negative like profiles ( p = 0 . 0057 ) ( Figure 7B ) . This is consistent with all current knowledge of the impact of ER-status in breast cancer prognosis . These results provide strong evidence that the ERα binding sites identified using ChIP-PET enrich for ER responsive genes that are associated with the biology of human breast cancers . We have previously shown that ERE sequences in promoter regions of putative ER target genes are not highly conserved , even though both conserved and nonconserved sites appear to be involved in ER binding [6] . Carroll et al . , however , indicated conservation of ER binding sites based on their analysis of binding sites discovered in Chromosomes 21 and 22 [15] . To resolve this apparent difference in our observations , we performed comparative analysis of binding site regions and ERE motifs across species using this genome-wide dataset . The overall conservation of a binding region is measured using the base-by-base conservation score and the presence of conserved elements ( PhastCons score and PhastCons Conserved Elements [24] ) , ( see Materials and Methods ) . Using this approach , although a clear conservation signal is visible , as compared to a randomly generated set of regions , the actual proportion of binding regions that are conserved is hidden . To analyze this conservation further , we examined the presence of PhastCons Elements in the ER binding sites . Surprisingly , only 273 ( 22% ) of the initial 1 , 234 binding regions overlap with such conserved elements , but partitioning the binding sites using this criterion showed that these 22% carry most of the conservation signal ( Figure 8 ) . Using size-matched random samples of genomic location we estimate the enrichment of conserved sites to be only approximately 13% ( unpublished data ) . Since TFBS motifs are short , typically 10–20 bp , they may not necessarily be located in a conserved region as detected by standard algorithms . We sought , therefore , to identify ER binding sites with ERE-like sequences and determine whether the specific ERE-like motifs are conserved in homologous regions in chimpanzee , mouse , and dog regardless of conservation of surrounding sequences . We extracted the sequences associated with the 1 , 234 binding regions in human ( hg17 ) and identified the corresponding homologous regions in chimpanzee ( panTro1 ) , mouse ( mm5 ) , and dog ( canFam2 ) using the tool liftOver ( UCSC Genome Browser utility tool , http://genome . ucsc . edu/cgi-bin/hgLiftOver ) . We then scanned for the presence of consensus EREs with up to two mutations . Using optimized 500-bp windows in human and chimpanzee and 1-kb windows in mouse and dog ( see Table 2 ) , 754 of the 1 , 234 human binding regions contained a full ERE . As expected , the vast majority ( 698 or 93% ) of the homologous binding windows in chimpanzee also contain ERE-like sequences . The ERE motif is also found in 283 ( 38% ) of the mouse homologous regions and in 357 ( 47% ) of the dog windows . Because the ERE-like sequences are common in any genome , we considered a site bearing a conserved motif only if the ERE is retained in chimp , mouse , and dog . Using this more stringent criterion , we found 179 ( 24% ) of the sites bearing motif conservation in all four species with a background of ~7% . This suggests that ~17% of the sites are under positive selection . Taken together , the sequence and motif conservation results indicate that the majority of binding sites identified in this study are poorly conserved between primates and other mammalian species , and the conservation of binding sites reported previously [15 , 16] likely resulted from a minority ( 22%–24% ) of highly conserved sites when assessed by multiple conservation metrics . We should note that the actual conservation of binding sites may be higher than observed due to alignment errors [25] . Even with adjustment for this potential error , however , there will likely be a large number of nonconserved ER binding sites . Though the ERE appears to be the dominant recognition sequence for ER on DNA , other transcription factors and their binding sites are also involved in directing ER to their specific target sites . Indeed , Carroll and colleagues discovered an enrichment of forkhead binding site motifs within ER binding regions of human Chromosomes 21 and 22 and demonstrated a role for the FOXA1 transcription factor in facilitating ER's ability to bind EREs and regulate target gene expression . In other nongenomic based studies , ER is also known to bind DNA indirectly through interactions with Sp1 and AP-1 transcription factors [9 , 10] . To determine the presence of additional TFBS motifs in the 1 , 234 ChIP-PET binding sites across the genome , we analyzed the 1 , 234 cluster sequences for putative TFBS based on TRANSFAC ( professional version 9 . 1 ) using the accompanying MATCH program [26] with the “minimize False Positive” setup . To compute the statistical significance , we generated a background sequence set , matching the total length of 1 , 234 clusters , using a third order Markov Chain sequence model trained on the whole human genome ( hg17 ) and scanned them similarly for putative TFBS . This was done 1 , 000 times . For each TFBS matrix , the average number of sites found per nucleotide represents the background probability of finding its putative sites . The p-values were computed under the binomial distribution and were adjusted for multiple hypotheses testing using the conservative Bonferroni correction . Table 3 lists the top matrices ( see Supporting Information for additional details of the analysis ) . As expected , the predominant sequence motif enriched in ER binding sites is the ERE , Interestingly , however , a large number of transcription-factor binding site motifs are statistically enriched in these ER binding regions even when corrected for multiple sampling suggesting that ER may interact extensively with other transcription factors at the DNA sites . Previous investigations have shown that FOXA1 bound to forkhead binding site motifs adjacent to EREs and interacted with ER , as do AP-1 and Sp1 . We also found FOXA1 , AP-1 , and Sp1 binding motifs significantly associated with the GIS-PET clusters . To further assess the functional significance of detected binding sites , we performed RNA interference experiments using small interfering RNA ( siRNA ) to knock-down Sp1 and then examined the expression of ten estrogen-responsive genes , which from our data were found to have adjacent ChIP-PET ER binding sites and predicted Sp1 binding sequences ( Figure 9 ) . Transfections with Sp1 siRNA constructs reduced Sp1 protein levels by 85% ( see immunoblot , Figure 9A ) as compared to the luciferase siRNA controls . Sp1 knock-down reduced basal expression levels of all the genes examined and had significant impact on estrogen-responsive induction of MN1 , JMJD2B , TBX2 , SL9A3R1 , CDC6 , and KIBRA and more moderate effects on GREB1 , IGFBP4 , RARA , and MYC as compared to the luciferase and vehicle-treated controls ( Figure 9A ) . In ChIP experiments where we examined four estrogen responsive genes ( Figure 9B and 9C ) , we observed that recruitment of ER to the ChIP-PET site was greatly increased by the presence of estrogen ( E2 ) and that this ER recruitment was reduced after knock-down of Sp1 in three of the four genes ( Figure 9B ) . By contrast to ER , Sp1 was present at a lower level at the ChIP-PET site ( note the lower fold recruitment ) , and the recruitment level of Sp1 was not affected by treatment with E2 ( Figure 9C ) . We also investigated these parameters in three estrogen-responsive genes in which their ChIP-PET region contained an ERE but was not enriched for Sp1 sites . For the three genes assessed ( FOS , BCL2 , and PGR ) , we found that E2 treatment increased their gene expression ( mRNA level ) and that Sp1 knock-down also reduced their gene expression after E2 . In these genes , we saw a very low level of Sp1 at the ChIP-PET region that was not altered by E2 treatment ( unpublished data ) . We believe that our observations are in keeping with the fact that these genes all contain Sp1 binding sites close to the promoter , shown previously to be important in their gene regulations [27–30] , so that some Sp1 presence and impact of Sp1 knock-down would be expected . Looping of the distal enhancer to a proximal region that binds Sp1 would result in the presence of both ER and Sp1 in our ChIP assays . This might be similar to the direct interaction of a distal signal-specific enhancer binding factor ( NF-κB ) region with the proximal transcription factor Sp1 binding region , as reported recently for the tumor necrosis factor α-inducible regulation of the monocyte chemoattractant protein-1 gene , MCP-1 [31] . Our findings suggest that the predicted binding sites found in ER ChIP-PET clusters and their associated transcription factors are likely involved in ER-mediated transcriptional regulation , although the extent and impact of their involvement may differ due to gene-specific cis-regulatory architecture and transcription factor complex recruitment , and , in the case of Sp1 , interactions between ER-associated effects and promoter proximal regulatory functions . Using these specific validated associations as reference points to assess the relative importance of other associated binding site motifs , we found that 46 other transcription factors are as significantly associated with the 1 , 234 GIS-PET binding sites as these three validated cis-partners of ER . This suggests that a wider range of transcription factors may partner with ER at the site of DNA binding than previously thought . Given these findings , we next asked whether there was discernable structure within the ER binding sites relative to the transcription factor response elements that are significantly over-represented . To this end , we assessed whether some motifs were nonrandomly distributed within a 500-bp window encompassing the center of the PET defined binding site or centered on the main ERE or half ERE ( Tables S3 , S4 , and S5 ) . Our results show that factors such as CDX , PAX , AP-1 , SF1 , and MAF are distributed within these sites in a nonrandom fashion . To dissect the anatomy of these adjacent binding motifs , we examined the specific position of these motifs vis-à-vis the central ERE ( Figure 10 ) . We plotted the position of the second transcription factor binding motif against the frequency of such an occurrence and found that SF1 , PAX2 , PAX3 , MAF , and AP-1 co-exist with the central ERE in an ordered manner . Surprisingly , all these factors appear to have significant overlap specifically at the ERE site itself , the most striking being SF1 and PAX3 . It did not escape our attention that the observed overlap could be arising from the inherent similarity of the other comotif with ERE or half-site ERE . However , if we use a validated cofactor AP-1 as a model , upon alignment of all the sequences of the AP-1 binding sites and their associated ERE ( see Table 4 ) , we discovered an inordinate number of AP-1 sites positioned in the place of a cognate ERE half site . These associated AP-1 sites could represent either truly functional AP1 sites , or degenerate ERE recognition sequences but with sufficient similarity to potentially be recognized by ER , or both . This half-site mimicry by the ERE of other transcription factor binding motifs was seen with MAF/BACH , and most intense with SF1 and PAX3 where a large proportion of those binding sites replace an ERE half site at ER binding regions . This was not seen in other adjacency candidates such as AML-1 where no internal ordering was noted . For CDX where nonrandom order was detected , the structure appeared to be an under-representation within 50 bp to around the EREs . Examination of the consensus binding motifs of these transcription factors reveal that SF-1 , BACH/MAF , and PAX3 contain sequences usually just one base different from the ERE half site and could by chance generate an acceptable ERE half site ( Figure 11 ) . Moreover , as in the case with AP-1 , the 5′ flanking sequences of these sites all contain the AP-1 consensus dinucleotide TG , which renders the ERE half site into a good AP-1 consensus . That these 8- to 13-mer recognition consensus sequences can be so frequently found as part of an ERE suggest that these factors may interact with ER in binding cis-regulatory sequences of target genes . It has been determined that ER can interact with AP-1 and Sp1 factors to regulate gene expression through a tethering mechanism where the DNA binding moiety is AP-1/Sp1 . In our genome-wide analysis , we asked whether our ER binding sites without discernable EREs had a predominant transcription factor binding motif . Our results show that predominantly forkhead transcription factors , followed by SRY recognition sequences are significantly enriched in these regions ( Table 5 ) . AP-1 sites , though not on the enriched list is however very similar to the MAF recognition sequences , which appear as borderline significant after SRY . Since AP-1 can bind to MAF sites , AP-1 involvement in these purely tethered sites is projected . Thus , surprisingly , ER binding sites without EREs appear highly enriched for recognition motifs for the forkhead family of transcription factors and above that of the known AP-1 interacting factors . Whole genome analysis of transcription factors provides an unbiased view of their regulatory dynamics . Here we present a genome-wide analysis of the DNA binding sites of ERα as present in the MCF-7 breast cancer cell line and map these sites to transcripts regulated by estrogen . We used a cloning and sequencing based technology and identified 1 , 234 high probability binding sites using an algorithm that minimizes false positives from amplified regions of the genome . That 94% of a sample of these sites could be validated by standard ChIP suggests that the majority of the 1 , 234 sites identified by ChIP-PET represent bona fide binding regions for ERα . Of note is that 96% of the validated binding sites harbored either full ERE-like ( 71% ) or solely half-ERE motifs ( 25% ) . Only 4% had no ERE-like sequences detectable using a two-position degeneracy cut-off , and therefore a pure tethered mechanism of ER transcriptional regulation must occur infrequently . This dispersed nature of these 1 , 234 sites vis-à-vis the TSSs makes the direct molecular assessment of whether these adjacent genes can be regulated by ER highly impractical . We sought to resolve this problem by examining the clinical behavior of these genes adjacent to ER binding sites . We posited that if these adjacent genes were under ER regulation , then their expression in breast cancers should readily determine ER status of primary breast cancers . Our results using a cohort of 251 breast cancers showed that these putative ER regulated genes can significantly separate ER status in breast tumors and therefore represent a transcriptional regulatory cassette that appears to affect ER response . We further examined this question by studying the behavior of these genes in MCF-7 cells as assessed using expression arrays . Though only 23% of the genes proximal to an adjacent ER binding site are responsive following estrogen treatment , this represents a significant enrichment of bona fide ER binding sites adjacent to estrogen responsive versus unresponsive genes ( p = 3 . 018433e − 51 ) ( unpublished data ) . Therefore , our in vitro and in vivo probabilistic analysis all point to the biological significance of the ER binding sites identified by our ChIP-PET analysis in the regulation of gene expression by ER . It is important to note the ability of ChIP-PET to identify , in an unbiased manner , bona fide ER binding sites among nearby EREs predicted only by computational methods . For example , for the carbonic anhydrase XII ( CA12 ) gene , matrix-based computational approaches used to identify potential cis-regulatory elements directing ER-regulation of CA12 indicate that five putative EREs reside in the proximal 5′ 5 kb with an additional ERE found in the first intron of the gene . However , we have identified a moPET 5′ binding site approximately 6 kb 5′ to the gene , which was found to be the major regulatory site directing ER-mediated transactivation as a distal enhancer ( D . H . Barnett and B . S . Katzenellenbogen , unpublished data ) . CA12 mRNA is up-regulated by estradiol in MCF-7 breast cancer cells [3 , 32] and in other ER-positive cells [33] and is positively associated with ERα status in primary breast tumors [34] . Hence , our findings highlight the ability of ChIP-PET to identify previously undiscovered enhancers of biologically relevant target genes . Much of the research of ER transcriptional regulation has focused on a few EREs located within the proximal promoter . We have shown with our global binding site data that , in fact , the vast majority of sites are located in distal or intragenic regions relative to the nearest regulated transcripts . Our genome wide analysis confirmed the more limited observations previously seen in Chromosomes 21 and 22 that only a small portion ( 5% ) of the binding sites are within 5 kb of the TSS and consistent with our previous predictions [6 , 35] . Intriguingly , however , detailed analyses revealed that the statistical preponderance of genes responsive in MCF-7 cells to E2 adjacent to ChIP-PET identified ER binding sites were up-regulated rather than down-regulated . Moreover , the location of these sites next to E2 induced genes showed an obvious enrichment around the TSS both in prestart locations and in 5′ introns and within 50 kb from the TSS ( Figure 6B ) . The number of these sites is small when the entirety of genes regulated by ER is considered and therefore would have been missed by a less specific analysis . This distribution of the ER binding sites relative to the induced transcripts indicates diversity in both proximal and distal mechanisms in regulating RNA polymerase activity and suggests that the proposed looping mechanisms [15] may play a more prominent role in ER-mediated transcriptional regulation than previously thought . We have further mapped the entire transcriptome of the MCF-7 cell line using a full-length cDNA library sequencing approach [17] . In sequencing pair-end tags of over 500 , 000 full-length cDNA equivalents , we found that 13% of the 22 , 115 individual transcripts identified were novel . When novel transcripts from MCF-7 are accounted for , 90% of the 1 , 234 high quality ER binding sites are within 100 kb of transcript boundaries ( G . Bourque , C . L . Wei , and E . T . Liu , unpublished data ) . This apparent distance restriction may reflect structural and spatial constraints on the distal effects of the bound ER on promoters . Equally intriguing is the possibility that ER-mediated gene repression may use mechanisms very different than gene induction , and that genomic topography ( i . e . , binding site location and affinity ) may have a significant role . Consistent with this is our quantification study using ChIP-qPCR on ChIP-PET identified ER binding sites where genes repressed by ER uniformly had ER binding sites that had the lowest fold induction after E2 exposure ( ~1–25 ) , as compared to those binding sites adjacent to induced genes ( ~25–473 ) , and were less likely to harbor a full ERE-like motif . When all sites are taken into account and measured by the number of overlapping PETs ( moPETs ) up-regulated genes have significantly higher moPET counts than down-regulated genes ( p = 4 . 575e − 4 , unpublished data ) . Moreover , in reporter assays performed with the 11 candidate ER binding sites , the only three that did not induce transcription off a TATA promoter were sites associated with repressed genes . These observations are consistent with previous findings that deviations from the full ERE motif reduces ER binding affinity and that the binding site dynamics may differ in genes that are induced by ER than those repressed by ER [11] . The large number of bona fide and nonproximal ER binding sites reported here represents ideal candidates for further characterization of these distinct mechanisms . It is known that ER can regulate gene expression not by direct DNA binding but through association with an intermediary transcription factor such as AP-1 . Theoretically , this mechanism of ER transcriptional regulation does not require ER binding to an ERE . Our motif searches in these non-ERE sites revealed that the predominant motifs in the pure tethered bin are those for the forkhead transcription factors , SRY , with MAF reaching borderline statistical significance ( p = 0 . 056 ) . MAF recognizes sequences related to the AP-1 target site and are considered as part of the larger AP-1 family of transcription factors and , therefore , our results suggest that AP-1 and MAF can bind to these sites [36] . The interesting observation is that in the absence of a minimum of an ERE half site , the fold enrichment of ER binding in these sites is lower ( a median of 51-fold enrichment of binding as compared to 81-fold for ERE; Student's t-test p = 0 . 027 ) . Moreover , our analysis of AP-1 sites within EREs show perfect orientation with one half site with similarities to AP-1 recognition sequences , and the second ( cognate ) half site primarily an ERE recognition sequence . These “hybrid” sites show higher levels of ER binding . This suggests that AP-1–associated tethering may favor sites with ERE half-site “anchors . ” Indeed , previous analysis of the ERE half site associated with the AP-1 site found in the progesterone receptor promoter showed that the integrity of the ERE half site is required for ER and AP-1 binding and estrogen responsive promoter activity [37] . These genome wide approaches to nuclear hormone receptor binding sites are revealing in that the large number of validated binding sites provide statistical power in assessing underlying motif structure in the binding sites . The results of our motif search analysis also point to the potential involvement of a number of other transcription factors participating in ER transcriptional regulatory activity . Included in the list of putative transcriptional coregulators is FOXA1 , which has been previously shown to be required for ER functions [15] . However , the fact that 46 other factors are enriched in the ER binding sites with the same probability as the proven interactors of FOXA1 , AP-1 , and Sp1 suggests the potential for highly complex interactions . Of course not all cis-partner transcription factors will be expressed in every cell type . But though it would be highly improbably that each co-occurrence will predict binding by both factors , our analysis of Sp1 action on ten estrogen-responsive genes with adjacent ChIP-PET ER binding sites and predicted Sp1 binding sequences showed down-regulation of all ten . Moreover , we have validated the effect of adjacent GATA3 and BACH interactions in ER binding to EREs ( J . Thomsen and E . T . Liu , unpublished data ) . This suggests that our algorithms to predict adjacent transcription-factor binding are potentially highly accurate . Perhaps even more interesting is the systematic order of these potential partner transcription factors relative to the position of the central ERE in bona fide ER binding sites . Consistent with the model where AP-1 binding appears “anchored” by an adjacent ERE is that AP-1 is distributed in a nonrandom manner within a 500-bp window of an ER binding site . In this distribution , the sequence of a number of full EREs are actually composite binding elements with an AP-1 site posing as an ERE half site . These composite EREs are seen with SF-1 , MAF/BACH , AP-1 , and PAX2 and PAX3 . All these factors have recognition sequences that overlap with ( but are distinct from ) the ERE half site . Unexpectedly , highly skewed positioning was found with the SF-1 and PAX3 recognition sequences ( Figure 10 ) , where a large proportion of these response elements are positioned as the second ERE half site within bona fide ER binding sites . Although such overlap may be cues for inherent similarity of the computational model between ER binding sites and other factor binding sites , in the case with SF-1 , it has been previously observed that SF-1 response elements can also bind ERα , but not ERβ [38] . Interestingly , SF-1 knock-out mice exhibited ovarian abnormalities and sterility resembling tissues from ER and aromatase knock-out animals , further suggesting an interaction between SF-1 and the ER-estrogen axis [39] . Thus , such composite sites are potential points of exchange for transcription factors possibly switching to and from homodimer and heterodimer states of occupancy and represent a potential mechanism to augment heterogeneous response to estrogen exposure . We have previously reported very little conservation of ERE motifs within promoter regions of human and mouse genes even though conserved and nonconserved sites both bind ER [6] . In the promoter regions of putative direct target genes , approximately 6% of predicted EREs were conserved in the mouse . In contrast , Carroll and colleagues reported conservation in sequences flanking ER binding sites they experimentally mapped to human Chromosomes 21 and 22 and in their whole-genome study [15 , 16] . To reconcile these apparent differences , we examined the 1 , 234 ChIP-PET ER binding sites and determined conservation in both flanking sequences and detected ERE motifs . Using similar analytical approaches as those used by Carroll et al . , we also find evidence of conservation within the 500-bp windows around the discovered binding sites . However , a more in-depth analysis showed that the conservation signal observed was driven by only 22% of all sites tested . There was limited conservation regardless of whether local sequence similarity or presence of an ERE motifs were used as the metric for conservation ( Table 2 ) . Thus , the conservation also observed by Carroll et al . is likely due to a small number of highly conserved sequences and does not represent global conservation of binding sites [15 , 16] . We have noted that the conservation may be underestimated due to alignment errors in the comparative analysis of whole-genome sequence data [25] , but these errors will not fully explain the large number of nonconserved sites . The list of genes with conserved ER binding sites does not appear to have functional coherence ( unpublished data ) . Genes classically thought of as prototypes of ER responsiveness , such as pS2/TFF1 , and the progesterone receptor have bona fide ER binding sites in the human MCF-7 cell line that are not conserved by sequence or motif presence across mammalian species . Moreover , both conserved and nonconserved sites are associated with ER-regulated genes . A total of 287 of all 1 , 234 binding sites ( 23 . 3% ) are associated with ER-regulated genes , while 63 of the 273 conserved binding sites ( 23 . 1% ) are associated with regulated genes ( not significantly different ) . The limited conservation of ER binding sites does not imply that the genes that are important in ER function are not regulated by ER , but that the precise DNA targets may differ . Given the distance of 100 kb , in which an occupied ERE can potentially regulate its associated promoter , there is much flexibility in the placement of ER regulatory elements . Nevertheless , these observations indicate that there are likely species-specific differences in the components and the dynamics of estrogen action and that results from animal studies need to be interpreted with this caveat in mind . In summary , our work provides a new cartography of ER binding on a genome-wide scale . The collective configuration of these binding sites has revealed fundamental rules that describe the characteristics of a bona fide ER recognition motif . The dominance of the ERE , the distributed nature of the binding sites distant to their associated genes , the separate nature of up- versus down-regulated genes , the importance of adjacent binding motifs of other transcription factors , and the frequency of composite ER response elements are all findings that would have been difficult to assess on a gene-by-gene basis . Data from this work will provide the experimental targets that will further dissect the intricacies of ER transcriptional regulation . MCF-7 cells were grown to 80% confluence in D-MEM/F-12 ( Invitrogen/Gibco , http://www . invitrogen . com ) supplemented with 10% FBS ( Hyclone , http://www . hyclone . com ) . Cells were washed with PBS and incubated in phenol red-free D-MEM/F-12 medium ( Invitrogen/Gibco ) supplemented with 0 . 5% charcoal-dextran stripped FBS ( Hyclone ) for 24 h in preparation for 17β-estradiol ( E2; Sigma , http://www . sigmaaldrich . com ) treatment . Estrogen-deprived MCF-7 cells were treated with 10 nM E2 for 45 min prior to the ChIP procedures . ChIP was carried out as described previously [6] using the HC-20 anti-ERα antibody ( Santa Cruz Biotechnology ) . Following ChIP , DNA fragments were either pooled for PET library generation or analyzed for ER binding at specific sites by real-time PCR . Proper DNA fragment length and ER binding to the known pS2/TFF1 ERE were confirmed by gel electrophoresis and real-time PCR , respectively . ChIP assays using antibodies to Sp1 were performed as previously described [40] . The antibodies used were from Santa Cruz Biotechnology ( Sp1 PEP-2 , rabbit IgG ) and Upstate Biotechnology ( Sp1 ) ( http://www . upstate . com ) . DNA obtained from ChIP was analyzed by quantitative real-time PCR using specific primers for the ER binding sites closest to selected ER-regulated genes . PCR quantification was performed on the ABI7500 Real-time PCR System ( Applied Biosystems , http://www . appliedbiosystems . com ) with 20 μl reaction volume consisting of 20 ng of ChIP samples or 20 ng of input DNA as templates , 0 . 2 μM primer pairs , and 10 μl of 2× SYBR Green PCR Master Mix ( Applied Biosystems ) . For each PCR run , the samples underwent 40 amplification cycles . Fluorescence was acquired at the conclusion of each cycle at 60 °C during the amplification step . Around 140 ng of ChIP DNA were used for construction of the ChIP-PET library for mapping ER binding sites in the human genome , following a procedure described previously [18] . Briefly , End-It DNA End-Repair Kit ( Epicentre , http://www . epibio . com ) was used to repair the ends of the ChIP DNA . DNA fragments larger than 500 bps were selected by using cDNA size fractionation columns ( Invitrogen ) and cloned into pGIS-3a vector [18] , which contains the Mme I cassettes flanking the cloning site ( XhoI ) . The ligation mixture was transformed into the One Shot Top10 Electrocomp Cells ( Invitrogen ) . A total of 2 . 3 million clones were obtained . Around 90% of the clones contained inserts . We plated out 1 . 2 million clones on LB-agar ( ampicillin 50 ng/ml ) and scraped off the cells for plasmid DNA isolation . Around 10 μg of purified plasmid DNA mixture was digested with MmeI and end-polished with T4 DNA polymerase to remove the 3′-dinucleotide overhangs . The resulting plasmids containing a signature tag from each terminal of the original ChIP DNA insert were self-ligated to form single-PET plasmids . These were then transformed into One Shot Top10 Electrocomp Cells ( Invitrogen ) to form a “single-PET library . ” We plated out 1 . 2 million clones from this library on LB-agar ( ampicillin 50 ng/ml ) and extracted plasmid DNA from the cells . Around 250 μg of plasmid DNA were digested with BamHI to release the 50 bp PETs . About 600 ng of single-PETs were PAGE-purified , then concatenated and separated on 4%–20% gradient TBE-PAGE . Appropriate size fraction ( 600–1 , 100 bps ) of the concatenated DNA was excised , extracted , and cloned into EcoRV-cut pZErO-1 ( Invitrogen ) to form the final ChIP PET library . The clones were grown on LB-Agar ( Zeocin 25 μg/ml ) . The plasmids were prepared and sequenced using ABI3730 DNA analyzer . All microarray experiments were carried out on Affymetrix U133 A and B GeneChips . MCF-7 cells were treated with 10 nM E2 for 12 , 24 , and 48 h and RNA extraction , labeling , and hybridizations were performed according to manufacturer protocols . Affymetrix analysis software was used to perform the preliminary probe-level quantitation of the microarray data . These data were further normalized using the RMA [41] normalization method . The default option of RMA ( with background correction , quantile normalization , and log transformation ) was used to generate the normalized intensity for each probeset . Differentially expressed genes were identified at each time point separately using the three untreated at the time point as controls against the three treated samples . The SAM [42] statistical method was used to select differentially expressed genes . Genes were selected based on the q-value less than 2% . Experiments using patient samples were performed as described in a previous publication [23] , and the data used in this study were obtained from the Uppsala cohort from the previous study . ER ChIP-PET binding sites were amplified from MCF-7 genomic DNA by PCR and cloned into the pGL4-TATA vector ( a minimal TATA box upstream of pGL4-Basic ) by homologous recombination using the In-Fusion CF Dry-Down PCR Cloning kit ( Clontech , http://www . clontech . com ) . Putative EREs were mutated using the QuickChange Site Directed Mutagenesis kit according to the manufacturers instructions ( Stratagene , http://www . stratagene . com ) . MCF-7 cells , grown in hormone-depleted medium for at least 3 d , were cotransfected with the ChIP-PET constructs and HSV-TK renilla with Fugene ( Roche , http://www . roche . com ) . After the cells were treated with 10 nM estradiol or ethanol for 18–24 h , cell lysates were harvested and assessed for firefly and renilla luciferase activity using the Dual Luciferase Reporter Assay system ( Promega , http://www . promega . com ) . Estrogen-deprived MCF-7 cells were transfected with Sp1 SMARTpool or GL3 luciferase control siRNA ( Dharmacon , http://www . dharmacon . com ) , according to the manufacturer's instructions . After 72 h , cells were treated with 1nM E2 for 4 h . Total RNA was harvested and prepared using Trizol reagent ( Invitrogen ) . Quantitative real-time PCR was performed as previously described [33] . The fold change in expression was calculated using the ribosomal protein 36B4 as an internal control as previously described [3 , 35] . Primer sequences are available upon request . Proteins were extracted from MCF-7 cells using RIPA buffer , separated on SDS-PAGE , transferred to nitrocellulose membrane , and immunoblotted using anti-Sp1 antibodies ( Upstate Biotechnology ) . PET sequence extraction and mapping were done as described previously [18 , 19] , using the PET-Tool [43] . The mapped PET sequences were further processed , annotated , and visualized using the T2G genome browser , our in-house genome browser developed based on the UCSC genome browser . To assess the saturation of the library , we fitted a Hill Function: where x is the number of PETs sequenced and f ( x ) is the number of distinct PETs mapped into the genome among x PETs sequenced . The parameters were chosen to ascertain the completeness of the library and to gain insight on the sequencing effort required for attaining higher saturation level . Using the nonlinear least-square Marquardt-Levenberg algorithm [44] and the historical sequencing data , we obtained a fit with a = 185 , 915 ( ±4 . 362 ) , b = 1 . 04144 ( ±2 . 704e−5 ) , and c = 239 , 414 ( ±12 . 07 ) . The underlying aberrant genome of MCF-7 presented an additional challenge in determining which of the ChIP-PET clusters were truly bound by ER . Presence of amplified regions [45] , with high and varying copy numbers , increased the probability of those regions being sampled during the ChIP assays , which translated into unusual overall ChIP-PET enrichments in multiple genomic pockets . Relying solely on the raw count of overlapping PETs would introduce undesirable false positives . To address this issue , we have developed a binding region identification algorithm ( unpublished data ) that produces lower false positives when predicting binding clusters in amplified regions , compared to using raw counts . When assessing the likelihood of a given ChIP-PET cluster being bound by ER , the two-stage algorithm first estimates the amount of noisy PETs surrounding the cluster of interest within its 25-kb flanking regions . Based on the estimated noise level and the neighborhood size ( i . e . , 50 kb ) , a moPET cut-off value can be calculated , such that the false positive probability is less than 1e−2 . If the given ChIP-PET cluster has a stronger overlapping region ( i . e . , higher moPET value ) than the calculated cut-off , we consider the cluster to be truly bound by ER . The rich presence of putative EREs points to the canonical and dominant theme of direct ER-DNA interaction . Nevertheless , ER interplays with other transcription factors have previously been reported and are expected , for it to exert wider and more diverse regulatory roles . These high quality binding regions present an unprecedented opportunity for the study of regulatory partners of ER . We employed a three-pronged approach to mine the binding regions for potential enrichment of binding motifs of other transcription factors , where the first assessed the enrichment of certain motifs in a given set of sequences , the second tested whether putative motifs of other transcription factors exhibited certain spatial correlation with respect to the main ERE or half ERE motif , and lastly the Genomatix suite was used for a low-throughput high quality semi-automatic assessment and visualization of potential comotifs . For the first and second sets of analysis , putative binding sites were identified based on weight matrices available in TRANSFAC ( professional version 9 . 1 ) and using the accompanying MATCH program [26] with the “minimize False Positive” configuration . To compute the statistical significance of motif enrichment in a given set of sequences , a background sequence set , with its total length matching that of the sequence set , was generated using a third-order Markov Chain sequence model ( trained on the whole human genome [hg17] ) and was similarly scanned for putative TFBS . This was done 1 , 000 times , and for each TFBS matrix , the average number of sites found per nucleotide represents the background probability of finding its putative sites . The p-values for motif enrichment were computed under the Binomial distribution and were adjusted for multiple hypotheses testing using the conservative Bonferroni Correction procedure . Evaluation of spatial correlation between main ERE or ERE half sites was carried out using Kolmogorov-Smirnov test . A 500-bp sequence window was defined for each binding region , centering on its main ERE or ERE half site . The putative binding sites of each transcription factor were tested whether they were uniformly distributed within the sequence window . PhastCons scores are base-by-base values between 0 and 1 that give a measure of evolutionary conservation in eight vertebrate genomes ( human , chimp , mouse , rat , dog , chick , fugu , and zebrafish ) based on a phylogenetic hidden Markov model , phastCons [24] , and Multiz alignments [46] . PhastCons Conserved Elements identify regions of the genome with high conservation scores . These tracks were obtained through the UCSC Genome Browser [47] . A binding site is identified as sequence conserved if its overlapping region overlaps any PhastCons Conserved Elements . Motif conservation analysis was carried out as follows: ( 1 ) sequences centered on the middle of the overlapping region of the 1 , 234 binding regions in human ( hg17 ) were identified; ( 2 ) corresponding homologous regions in chimpanzee ( panTro1 ) , mouse ( mm5 ) , and dog ( canFam2 ) were identified using the tool liftOver ( UCSC Genome Browser utility tool ) ; ( 3 ) corresponding fasta sequences were extracted; and ( 4 ) all sequences were scanned for the consensus ERE motif allowing for two mismatches . Process was repeated for various window sizes in human varying from 250 bp to 5 kb ( unpublished data ) .	Estrogen receptors ( ERs ) play key roles in facilitating the transcriptional effects of hormone functions in target tissues . To obtain a genome-wide view of ERα binding sites , we applied chromatin immunoprecipitation coupled with a cloning and sequencing strategy using chromatin immunoprecipitation pair end-tagging technology to map ERα binding sites in MCF-7 human breast cancer cells . We identified 1 , 234 high quality ERα binding sites in the human genome and demonstrated that the binding sites are frequently adjacent to genes significantly associated with breast cancer disease status and outcome . The mapping results also revealed that ERα can influence gene expression across distances of up to 100 kilobases or more , that genes that are induced or repressed utilize sites in different regions relative to the transcript ( suggesting different mechanisms of action ) , and that ERα binding sites are only modestly conserved in evolution . Using computational approaches , we identified potential interactions with other transcription factor binding sites adjacent to the ERα binding elements . Taken together , these findings suggest complex but definable rules governing ERα binding and gene regulation and provide a valuable dataset for mapping the precise control nodes for one of the most important nuclear hormone receptors in breast cancer biology .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "oncology", "mammals", "computational", "biology", "homo", "(human)", "genetics", "and", "genomics" ]	2007	Whole-Genome Cartography of Estrogen Receptor α Binding Sites
Biological and social networks are composed of heterogeneous nodes that contribute differentially to network structure and function . A number of algorithms have been developed to measure this variation . These algorithms have proven useful for applications that require assigning scores to individual nodes–from ranking websites to determining critical species in ecosystems–yet the mechanistic basis for why they produce good rankings remains poorly understood . We show that a unifying property of these algorithms is that they quantify consensus in the network about a node's state or capacity to perform a function . The algorithms capture consensus by either taking into account the number of a target node's direct connections , and , when the edges are weighted , the uniformity of its weighted in-degree distribution ( breadth ) , or by measuring net flow into a target node ( depth ) . Using data from communication , social , and biological networks we find that that how an algorithm measures consensus–through breadth or depth– impacts its ability to correctly score nodes . We also observe variation in sensitivity to source biases in interaction/adjacency matrices: errors arising from systematic error at the node level or direct manipulation of network connectivity by nodes . Our results indicate that the breadth algorithms , which are derived from information theory , correctly score nodes ( assessed using independent data ) and are robust to errors . However , in cases where nodes “form opinions” about other nodes using indirect information , like reputation , depth algorithms , like Eigenvector Centrality , are required . One caveat is that Eigenvector Centrality is not robust to error unless the network is transitive or assortative . In these cases the network structure allows the depth algorithms to effectively capture breadth as well as depth . Finally , we discuss the algorithms' cognitive and computational demands . This is an important consideration in systems in which individuals use the collective opinions of others to make decisions . A goal of many of network studies ( e . g . [1]–[4] ) is to predict the effects of perturbations , such as extinction and predation events , on network structure . Making these predictions requires information about network connectivity ( e . g . is the network scale-free , exponential , etc . ) . When the connectivity is non-uniform , it is also important to quantify variation at the node level in order to identify nodes that , if removed , are likely to impact negatively network stability . This is well recognized and many useful methods have been developed for measuring this variation [1]–[2] , [5]–[14] in a range of networks , including the world-wide web [5] , food webs describing trophic interactions [1] , [2] , networks of interactions between genes and proteins [6]–[10] , and social networks , in both animal and human societies [11]–[16] . Patterns of connectivity can also influence node function in the larger system of which the network is a part . For example , in previous work on the behavioral causes of multi-scale social structure in primate societies [14] , [17]–[22] it was found that group consensus about an individual's ability to win fights – its social power ( see Sec . Primate communication network ) –is population coded in a status signaling network . In this system , individuals use subordination signals to communicate to adversaries that they perceive themselves to be the weaker opponent . The signals are often repeated and are always unidirectional ( emitted by one individual in a pair but not the other ) . A single signal indicates that the sender perceives the receiver capable of using force successfully against him . The frequency of signals emitted ( over some defined time period ) indicates the strength of the sender's perception that the receiver can successfully use force against him . In the work cited above it was demonstrated that consensus in the group about individual ability to successfully win its fights can be calculated by quantifying uniformity in the weighted in-degree distribution of signals sent to by its senders and weighting this score by the total number of signals received ( this calculation is described in Sec . Shannon consensus ) . The resulting score for may not be the preferred score for of any specific group member , but can be said to reflect the group's collective view about how good is at winning fights . Correspondingly , the rank order associated with the distribution of scores in the population might not match the preferred rank order of any single individual , but as the outcome of integrating over all of the individual opinions , it can be said to be the consensus social power rank order . The data indicate that individuals can estimate their own social power and also know something about how others in the group are collectively perceived [17] , [20] , [23] , [24] . Consequently , social power is informative about the likely cost of interaction when interactions are not strictly pair-wise ( a common feature of these systems and the reason why a consensus-based definition is important ) [17]–[19] . Under heavy-tailed power distributions , in which a few individuals are disproportionately powerful , conflict management mechanisms like third-party policing ( a critical social function ) can emerge and are performed by nodes in the tail of the power distribution [14] , [20] . Policing is an important social function because by controlling conflict it facilitates edge building by nodes in the signaling as well as other social networks [18] , [20] . These results suggest that ( 1 ) network structure can encode node function and that ( 2 ) measures that quantify agreement in node connectivity patterns can be used to decode this population coding of node function . In Table 1 we give several examples of other networks in which node function might also be population coded and consensus estimation could be useful for identifying important nodes . In principle , consensus about node state or function can be quantified by measuring the uniformity of a node's weighted in-degree distribution [14] , as in the above example , by measuring the “flow” into and out of a node ( depth ) , or using simple counts . To capture these competing notions of consensus , we introduce a variety of alternative information theoretic , diffusion , and count algorithms that capture breadth and depth to different degrees , and so serve as hypotheses about how functional variation in nodes is encoded in interaction networks via consensus . The algorithms take an interaction network as input and produce a vector of scores for the nodes in the network as output . We interpret the score of node as the collective opinion , or consensus , about state or its capacity to perform a given behavior . We note that the algorithms only quantify agreement in the connectivity patterns; what the consensus is about– state– depends on the type of interactions in the interaction network . For example , in the work on power in primate societies mentioned above , the interaction matrix contained directed subordination signals . These signals have special properties that allow them to reliably encode information about the ability to win fights , which is the basis of power [19] . We discuss the importance of the interaction matrix for the interpretation of consensus in greater detail in Sec . Background and motivation . After introducing the algorithms , we compare their mathematical properties , and in a few cases , establish approximate equivalence . We introduce three data sets that we use to empirically evaluate how well the output of the algorithms predicts node function out of sample . We investigate the properties of these algorithms that make them predictive measures of consensus . The data sets include a status communication network in a primate society , a network of collaborating condensed matter physicists from a prominent journal , and a functional linkage network of yeast genes that influence viability and growth . Finally , we assess the sensitivity of the algorithms to systematic error at the node level and strategic manipulation of the network by nodes or small sub-sets of the network . Here we consider one additional algorithm , the Borda count , for computing consensus on networks . The Borda count is an algorithm that is traditionally used to determine the outcome of an election . Each member of a voting population ranks the candidates of the election . This is analogous to each individual in a primate group emitting signals to others in accordance with whom they perceive as more or less likely to use force successfully . The Borda count aggregates these preferences into one ranking over the candidates . Supposing that there are candidates , each voter gives votes to his highest preference , to his next highest choice , on down to one vote to his least favorite candidate . A voter can rank candidates equally and the candidates' votes in this case are the average of the numbers of votes they would have received were they not tied . A candidate's score is the sum of his votes from each voter . In the signaling case , the receiver of the most signals from a given individual will receive n “votes” and the receiver of the fewest signals from that individual will receive one “vote” . In unweighted networks , each individual “voter” divides the group into nodes with whom he does or does not interact , giving the same number of votes to the individuals in each class . Mathematically , we define a matrix such that is the rank given node by node , where gives rank to its highest preference and rank to its lowest preference , and define the vector is the Borda consensus scores for node . The Borda count is more coarse-grained than the total frequency of interactions received because information about the number of interactions received is lost and only the ordinal ranking of nodes by the number of interactions received is used . It does , however , convey information about agreement among interaction partners . If we find that a node has a high score under the Borda count , this indicates that many other nodes rank the receiver highly and agree about its relative value to them . Hence like Shannon Consensus it should be intrinsically sensitive to certain kinds of bias in the interaction matrix ( see Sec . Empirical Comparison for further discussion ) . All of the algorithms we compare provide some measure of consensus in a network about the state of a given node , such that we expect they are positively correlated ( these data are presented in Sec . Basics of data set ) . In fact , we can describe these correlations by deriving mathematical relationships between some of the algorithms . The mathematical relationships between the breadth algorithms are easiest to see . , and are related by the definition of and a simple theorem about :Consider the definitions of and :These definitions make the mathematical relationships and on the one hand and obvious . If the network is unweighted , then we can write the following algorithms as a function of in-degree , :where is constant across nodes and depends on the total number of edges in the network . In this case , the rankings generated by these algorithms will be the same , although the actual values will be different . Eigenvector Centrality can be related to the redistribution probabilities and in-degree . Recall that is the stochastic transition matrix where denotes the probability of walking from node to node and Eigenvector Centrality is defined by the equation Since is stochastic , for all and we can choose such that . Since , this gives If we let , then and we can show thatwhere . In the case that for all , these bounds can be combined to giveThis bound gives an indication of how is related to the number of interactions received and the redistribution weights used in the calculation of . As we increase the redistribution weight , the minimum possible Eigenvector Centrality scores increases . In general , nodes that engage in more interactions and that interact with nodes with few other interaction partners will have higher Eigenvector Centrality scores . Much of the research on consensus aims to determine how a group comes to a single decision , such as which direction to move , who should be president , etc . [30]–[33] . In this study our aim is somewhat different . Our goal is to quantify how much consensus there is in the group ( e . g . network ) about the state of a node ( is it on or off , is it capable of performing a target function , etc . ) . Hence the interpretation of consensus turns on the meaning of the edges in the network , represented by the data in the interaction matrix , as much as on the algorithm applied to the matrix to compute the consensus scores for the nodes . It is therefore critical that the interaction data used to construct the matrix be chosen carefully . Below we provide basic details about the three test systems –a primate status communication network , a collaboration network , and a functional gene linkage network . We provide the biological interpretation of the edges in the networks and of node state , and we introduce the functional data used to empirically evaluate the algorithm's performance . We note that the mechanistic basis for consensus as an important network measurement is best understood for the primate communication network , and this fact is reflected in that section's length . In Table 1 we provide interpretations of consensus scores for several different kinds of networks in addition to those we describe below . We are using a primate communication network in a large captive social group of pigtailed macaques ( Macaca nemestrina ) to measure social power , operationalized as group consensus about individual ability to win fights . We are using a collaboration network to measure reputation , defined operationally as group consensus about whether to work with a given scientist . We are using a network of functional linkages between genes to measure gene importance , defined operationally as group consensus about whether to be functionally linked with a given gene ( this “decision” could be made in either developmental or evolutionary time ) . Each algorithm produces as output a vector of scores for nodes in the network . In Table S1 in Text S1 we present the correlations between these outputs for each network . The distribution of consensus scores for each of the networks , according to each of the algorithms considered in this paper , is presented in Figures S2 , S3 and S4 . Within each data set , most algorithms suggest roughly similar distributions . In the case of the signaling network , these distributions look heavy-tailed , which is consistent with the distributions of functional data . Additionally , for the signaling network , two of the algorithms– Shannon Consensus and Weighted Simple Consensus– produce distributions that are not significantly different than normal after log transform , indicating they are consistent with the log-normal distribution . Our predictor variables are the social power indices produced by the consensus algorithms . Dependent variables include: support solicited – requests for support received by a third-party to a fight from fight participants ( should be positively correlated with power ) ; intervention cost – operationalized as the intensity of aggression received by an intervener in response to its interventions into fights among group members ( should be negatively correlated with ) ; and intensity of aggression used by an intervener during its intervention ( should be negatively correlated ) ( these variables are defined and the data collection methods are described in Section Methods and in [14] ) . These dependent variables are corrected for underlying variation in tendency to fight ( see Section Methods ) . All algorithms are significantly correlated with the dependent variables ( , Table 4 ) . The best predictor of the dependent variables is Weighted Simple Consensus , followed closely by Shannon Consensus , and Eigenvector Centrality . The worst predictors are David's Score , Borda Count , and Simple Consensus . The most highly predictive algorithms have very similar values , so it is hard to differentiate between them based on their predictive power alone . However , as we discuss below , these algorithms vary in their sensitivity to source biases and in their computational and cognitive complexity . In this social system , there are a few individuals in the tail of the power distribution who are disproportionately powerful [14] , [17] . This is borne out in our data , as the correlation between the algorithm scores and the dependent variables is substantially higher for the top quartiles than the bottom quartiles ( Figure 1 ) . Our predictor variables are the reputation indices produced by the consensus algorithms . The dependent variable is total amount of grant money awarded to a PI or CoPI by the National Science Foundation ( see Section Methods ) . Of all the algorithms we consider , only Eigenvector Centrality is significantly correlated with this external variable ( , Table 4 ) . Two reasons , one mathematical and one sociological , appear to account for this result . First , Eigenvector Centrality can distinguish between nodes that have identical local neighborhoods . In-degree can only take integer values and there is presumably an upper bound on the number of possible collaborators given time and other constraints . In this network , the highest in-degree observed is so that there are possible values , , a node's in-degree can take . As Eigenvector Centrality can take any value between and , it can give different scores to nodes with the same in-degree . In other words , Eigenvector Centrality uses global information to differentiate between nodes that are locally identical . This effect is not as noticeable in the subordination signaling network because there are only individuals in the signaling network and therefore less degeneracy in the in-degree distribution . Second , it is perhaps not surprising that for this kind of network Eigenvector Centrality is more predictive of the dependent variable than the breadth algorithms – although physicists involved in the process of awarding grants to others are expected to recuse themselves when confronted with an application from one of their own collaborators , they may be more likely to award grants to collaborators of their collaborators . Therefore , having many collaborators may not be that helpful in receiving grant money , but scientists whose collaborators have many collaborators may have an advantage . Our predictor variables are the importance indices produced by the consensus algorithms . The dependent variables are the viability and competitive fitness of organisms with mutated versions of the gene . For each of our algorithms , the importance scores for essential genes are significantly higher than the importance scores for non-essential genes ( , Table 4 ) . Similarly , for each algorithm , the importance scores are significantly negatively correlated with the competitive fitness variable ( , Table 4 ) . The most predictive algorithms are , in order , Eigenvector Centrality , Simple Consensus , the Borda count , and Shannon Consensus . In differentiating between essential and non-essential genes , Eigenvector Centrality is marginally better than the other algorithms . In predicting competitive fitness , the four most predictive algorithms perform equally well . With both external variables , the test statistics are noticeably smaller for the Graph Laplacian than for the other algorithms . As we showed above , on unweighted networks , On both the collaboration and linkage networks , nodes with high in-degree tend to interact with many other highly connected nodes . For both networks , we find high correlations between in-degree , , and the sum of the in-degrees of a node's neighbors , ( , , for the collaboration network and , , for the linkage network ) . Nodes that have many interactions with other highly connected nodes receive low Graph Laplacian scores , a counterintuitive result that suggests the Graph Laplacian is not a robust measure of consensus . We summarize the predictive performance of the algorithms on the three data sets in Table 5 . An important question in evaluating the performance of a consensus algorithm is how sensitive the algorithm is to deficiencies in the data in the interaction matrix . Aspects of this question have been addressed in previous work . Ghoshal et al . [57] showed that in scale-free networks of sufficient size , if all edges in the network are shuffled but the in-degrees maintained , the ranking of the nodes according to eigenvector centrality is not severely perturbed . This type of shuffle allows the researcher to simulate the effects of missing or noisy data in the interaction matrix on an algorithm's output . We are particularly interested in the effects on the algorithm's output of nodes systematically making errors in their assessments of the states of other nodes or nodes attempting to manipulate social structure by “loading the deck” or inflating the consensus scores of nodes by , for example , manipulating the weighted degree distribution . ( One way to manipulate the weighted degree distribution is to inflate a node's weighted in-degree by sending many signals . ) Capturing this kind of “deficiency , ” which we call source bias requires a different kind of shuffle . First , we measure in our interaction matrices the correlation between a node's Shannon entropy ( as defined in Sec . Shannon consensus ) and the total frequency of interactions it receives ( weighted in-degree or in-degree ) ( see Table S1 in Text S1 ) . If entropy and in-degree were poorly correlated , we could independently evaluate the effects of receiving many interactions from receiving interactions from many individuals . However , this is not the case on the data sets we consider . We break the correlation by systemically shuffling the data in the matrices such that we create matrices with strong source biases but conserve the total number of interactions ( e . g . signals ) received . We now have two matrices –the original , unshuffled matrix , and the shuffled matrix , . We then compute consensus scores for the nodes using the unshuffled and shuffled matrices and assess how much the rank order changes under the shuffle . More specifically , for a given pair of interaction partners in the network , say nodes and , we construct a matrix in which the target node , receives all of its interactions from partner node , . If the original network is directed , we hold constant the out-edges of in addition to holding constant weighted in-degree . If the original network is undirected , we maintain the symmetry . The subordination signaling network is small enough so that we can perform this shuffle for every pair of partners . However , the collaboration network and the functional linkage networks are too large to exhaust every pair of partners , so we choose of the nodes that are also represented in the functional data sets . Partner nodes are chosen at random from the target node's neighbors . An algorithm is said to be sensitive to source bias if the rank order of the shuffled matrix , differs from the rank order of the original matrix , . Large changes in the rank order indicate that the test algorithm tends to give higher scores to nodes that interact with many neighbors than to nodes that interact strongly with just one other node and is an indication that the algorithm is sensitive to source biases . We find that Shannon Consensus , and the Graph Laplacian , , tend to be quite sensitive to source biases ( Figure 2 ) . This is expected , as and depend on the entropy of the receiving distribution , which is by design in the shuffled matrices . By definition , in-degree , , is maximally insensitive to source bias as we hold it constant in our shuffle . Eigenvector Centrality , , is also fairly insensitive , but the explanation why is initially counter-intuitive . As can be seen in Figure 2 , Eigenvector Centrality appears to be particularly insensitive to the shuffle for the subordination signaling network , as the rank order for the shuffled and unshuffled matrices for that network is very similar . The reason for this is that in the subordination signaling network individuals who receive many signals receive some of these signals from partners who themselves receive many signals . In addition individuals who receive many signals send very few signals . Hence there is information about breadth encoded in the second and third order connections ( and so forth ) in the network . Even if we shuffle the matrix so that all of an individual's signals come from a single other node , as long as we hold constant the out-edges of , the in-edges to are likely to be from an individual who itself receives relatively many signals . Eigenvector Centrality , by emphasizing paths through the network , takes these second and third order connections into account . It is consequently likely to get the rank order right , even after we reduce the diversity or breadth in the first order or direct connections , as long as the second and third order connections in the shuffled matrices encode information about the first order connections in the unshuffled matrices . See Text S1 , Section Sensitivity of eigenvector centrality on transitive networks for more discussion of the relationship between transitivity and sensitivity to source bias . This suggests that measures of consensus that emphasize depth– paths through the networks –also implicitly measure consensus breadth when there is some degree of either assortativity or transitivity in the network , and work well because of these features . In the absence of transitivity , or when transitivity is very low , depth measures like Eigenvector Centrality , should not perform well as measures of consensus , unless , as in the case of the Graph Laplacian , they explicitly incorporate Shannon information . In the Text S1 we provide details on an additional analysis we performed to evaluate algorithm sensitivity to source bias . The results reported in this paper and elsewhere ( [17] , see also [60]–[62] ) suggest that at least in social networks nodes may be making strategic decisions about social interactions using knowledge of how they are perceived by the group . For example , the individuals in the primate study group appear to estimates of their relative power to make decisions about whether to intervene in conflicts [17] . This requires that they have some knowledge of moments or properties of the distribution of power ( e . g . approximate variance ) . An important question is how individuals extract this information [22] , [63] . More generally , what do animals know about social structure and collective dynamics , how precise are their estimates , and what heuristics might they use to make calculations [64] ? It would be useful , for example , to be able to quantify the algorithmic complexity of each algorithm so that we could rank calculations by some measure of computational difficulty ( see also [65] ) . Ideally , we would also like to know how sensitive each algorithm is to the input data . ( e . g . is the exact number of signals received by individual critical , or will a rough estimate do ? ) for the output distribution of power to be a useful predictor of out of sample data . Addressing this robustness question would help to determine how much room there is to relax the mathematical requirements of a given algorithm , and find a heuristic simple enough for this study species . Ranking the algorithms by their algorithmic complexity is a long way off , if achievable at all . As is illustrated in Figure S1 , we can only crudely rank the algorithms given what we know about the minimum number of steps each requires in order to estimate critical quantities from an empirical perspective – the absolute power of individual , the relative power of ( e . g . where it falls in a power distribution of a given type ) , and the moments of the power distribution . In most circumstances it seems unlikely that we , or the animals , would be interested in an isolated individual's score . This is because it is not her power value that is important , but rather where an estimated value falls in a distribution of power scores . Yet calculation of absolute and relative power require different computational approaches and a preliminary assessment suggests that the difficulty of these steps varies across algorithms . We discuss these issues in greater detail in Text S1 , Section Computational complexity . In addition to approaching the problem of complexity mathematically , we can approach it empirically by asking how sensitive the algorithms are to imperfect information in the input matrices . For example , perhaps the individuals in our system cannot discriminate based on identity and can only remember classes of individual ( e . g . male or female , or matriline x or y , etc . ) , signals or signalers , or an interaction history of length . By coarse-graining the input data , it is in principle possible to test how sensitive the algorithms are to this kind of imperfect information resulting from various cognitive or spatial constraints . Aspects of this question have been addressed in previous work , as discussed in Section Sensitivity of the algorithms to source biases . However , many questions remain open for future work . If node function in many different systems is collectively encoded in interaction networks and this information is decodable by quantifying the agreement in network connectivity patterns , this would suggest that consensus formation is at the core of sociality . Consider the primate society used as a model system in this paper . Power in our primate study group is a critical social variable . However power is not a simple variable . The distribution of power does not map directly onto a distribution of body sizes or even a distribution of fighting abilities . Rather it consolidates as multiple interacting individuals learn about fighting abilities and signal about this to reduce social uncertainty [14] , [19] , [21] , [22] , [63] . When the statistics used to operationalize an aggregate social property , like power structure , are more than simple counts over strategies , and when the inputs are not simply individual traits but network data , we need to worry explicitly about the mappings between behavioral strategies and decision-making at the microscopic level and social organization [65]–whether we are working with the social organization of primates or of cells forming a tissue . A central question becomes , How do strategies get collectively combined by multiple components to produce macroscopic social properties ? How much degeneracy characterizes this mapping ? Once we can describe the developmental dynamics giving rise to an aggregate social property , we will be in a position to study how the social processes producing power and other kinds of social structure have evolved in a wide range of systems . The data set , collected by J . C . Flack , is from a large , captive , breeding group of pigtailed macaques that was housed at the Yerkes National Primate Research Center in Lawrenceville , Georgia . The physicist collaboration network was collected by Mark Newman , as described in [38] , and is available at http://www-personal . umich . edu/~mejn/netdata/ . The data were initially collected from the Los Alamos e-Print Archive , now the arXiv at http://arxiv . org . Since initial publication in 2001 , the network has been updated with collaborations from the arXiv through 2005 . scientists are represented in the network and the collaborations occurred between January , 1995 and March , 2005 . The National Science Foundation makes the data about awarded grants publicly available at http://www . nsf . gov/awards/about . jsp . For each scientist in the collaboration network , we searched this database for any grant concerning condensed matter physics on which the scientist was one of the investigators . If the scientist was awarded more than one grant , we summed the total amount of grants awarded him or her . Grant data was available for of the scientists in the collaboration network . The grants were awarded between September , 2008 and September , 2012 , with one grant starting in September , 2004 . The functional linkage network was constructed by Lee et al . , as described in [56] and [51] , and is available at http://www . yeastnet . org/ . Functional linkages between genes are associations that “represent functional constraints satisfied by the cell during the course of the experiments” [56] . Evidence of a functional linkage between two genes was provided by mRNA coexpression levels , the results of protein interaction experiments , phylogenetic profiles , and the co-occurrence of the two genes in a scientific paper [51] , [56] . Lee et al . combined these data to calculate the log-likelihood that two genes are involved in a similar function . In our analyses , we say an edge is present if its log-likelihood score is greater than and is absent otherwise . The resulting network has nodes . The Saccharomyces Genome Database maintains information about the phenotypic effects of genes in the yeast genome at www . yeastgenome . org . Two phenotypic effects reflect a gene's overall importance . One measure is the viability of organisms with a mutant version of the gene and a second measure is the competitive fitness of organisms with a mutant version of the gene . The viability measure is binary: a mutation to a gene can lead to either a viable or an inviable organism . An inviable organism is one that is unable to grow under standard growth conditions for S . cerevisiae , defined as glucose-containing rich medium ( YPD ) at C . A gene's competitive fitness is given by the relative growth rate of an organism with a mutated version of the gene compared to one with the normal genotype . Greater competitive fitness is indicated by a relative growth rate of greater than . These experiments can be performed in various media: we only used those performed in minimal medium to standardize our comparisons . More information is available at http://www . yeastgenome . org/help/function-help/phenotypes . These phenotype data are available for of the genes in the linkage network .	Decision making in complex societies requires that individuals be aware of the group's collective opinions about themselves and their peers . In previous work , social power , defined as the consensus about an individual's ability to win fights , was shown to affect decisions about conflict intervention . We develop methods for measuring the consensus in a group about individuals' states , and extend our analyses to genetic and cultural networks . Our results indicate that breadth algorithms , which measure consensus by taking into account the number and uniformity of an individual's direct connections , correctly predict an individual's function even when some of the group members have erred in their assessments . However , in cases where nodes “form opinions” about other nodes using indirect information algorithms that measure the depth of consensus , like Eigenvector Centrality , are required . One caveat is that Eigenvector Centrality is not robust to error unless the network is transitive or assortative . We also discuss the algorithms' cognitive and computational demands . These are important considerations in systems in which individuals use the collective opinions of others to make decisions . Finally , we discuss the implications for the emergence of social structure .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "complex", "systems", "mathematics", "ecology", "applied", "mathematics", "biology", "computational", "biology", "behavioral", "ecology" ]	2013	A Family of Algorithms for Computing Consensus about Node State from Network Data
Maintenance of the correct redox status of iron is functionally important for critical biological processes . Multicopper ferroxidases play an important role in oxidizing ferrous iron , released from the cells , into ferric iron , which is subsequently distributed by transferrin . Two well-characterized ferroxidases , ceruloplasmin ( CP ) and hephaestin ( HEPH ) facilitate this reaction in different tissues . Recently , a novel ferroxidase , Hephaestin like 1 ( HEPHL1 ) , also known as zyklopen , was identified . Here we report a child with compound heterozygous mutations in HEPHL1 ( NM_001098672 ) who presented with abnormal hair ( pili torti and trichorrhexis nodosa ) and cognitive dysfunction . The maternal missense mutation affected mRNA splicing , leading to skipping of exon 5 and causing an in-frame deletion of 85 amino acids ( c . 809_1063del; p . Leu271_ala355del ) . The paternal mutation ( c . 3176T>C; p . Met1059Thr ) changed a highly conserved methionine that is part of a typical type I copper binding site in HEPHL1 . We demonstrated that HEPHL1 has ferroxidase activity and that the patient’s two mutations exhibited loss of this ferroxidase activity . Consistent with these findings , the patient’s fibroblasts accumulated intracellular iron and exhibited reduced activity of the copper-dependent enzyme , lysyl oxidase . These results suggest that the patient’s biallelic variants are loss-of-function mutations . Hence , we generated a Hephl1 knockout mouse model that was viable and had curly whiskers , consistent with the hair phenotype in our patient . These results enhance our understanding of the function of HEPHL1 and implicate altered ferroxidase activity in hair growth and hair disorders . Iron is an essential trace element and constituent of important cellular proteins that include hemoglobin , myoglobin , flavoproteins , cytochromes and various non-heme enzymes . Transfer of electrons via oxidation–reduction ( redox ) reactions results in the conversion of iron between its ferrous [Fe ( II ) ] and ferric [Fe ( III ) ] forms . This property allows iron to participate in vital biological processes including oxygen transport , DNA biosynthesis and oxidative phosphorylation [1] . In mammals , iron homeostasis is precisely regulated to ensure proper iron acquisition , transfer , and storage while also preventing the donation of electrons to molecular oxygen that would otherwise lead to the generation of toxic free radicals [2] . Dietary iron is absorbed predominantly in the duodenum and traverses both the apical and basolateral membranes of absorptive epithelial cells to reach the plasma . For non-heme iron , one or more ferrireductases ( e . g . , DCTYB ) on the apical surface of duodenal cells first converts Fe ( III ) to Fe ( II ) [3] . Fe ( II ) is then imported by an apical divalent metal-ion transporter ( DMT1 ) [4–6] into the cytosol where it can be stored in the iron-storage molecule ferritin or exported into plasma by the basolateral iron exporter ferroportin [7 , 8] . The transmembrane protein ferroportin transports Fe ( II ) into the plasma , but Fe ( II ) must be oxidized to Fe ( III ) for incorporation into transferrin . This function is carried out by hephaestin ( HEPH ) , which is a member of the multicopper oxidase family that facilitates the conversion of Fe ( II ) to Fe ( III ) in an enzymatic reaction that uses oxygen as an electron acceptor [9] . HEPH is the predominant multicopper ferroxidase expressed in the basolateral membrane of absorptive intestinal cells [10] . Mice harboring loss-of-function mutations in Heph ( Sex-linked anemia or Sla mice ) show marked accumulation of iron in the intestinal mucosa and systemic iron deficiency , owing to a deficit in iron export [11] . Ablation of Heph either specifically in the intestine or in the whole body also leads to iron accumulation in duodenal enterocytes and reduction in intestinal iron absorption [12] . These findings show the specific role of Heph in intestinal enterocytes in maintaining whole body iron homeostasis . Ceruloplasmin ( CP ) , a paralog of HEPH and the principal ferroxidase in plasma , oxidizes Fe ( II ) to Fe ( III ) and is involved in the release of Fe ( III ) from multiple cell types , allowing iron to bind transferrin in blood and extracellular fluid [9 , 13] . Genetic deficiency of CP ( aceruloplasminemia ) in humans leads to iron overload in the liver , brain and pancreas , and results in progressive neurological disease and diabetes [14] . Similarly , mutations in the Cp gene in mice lead to decreased iron export and increased iron retention in the liver and spleen [13] . HEPHL1 or zyklopen , another member of the multicopper oxidase family , was identified and studied in BeWo cells , a placental cell line . Molecular modeling of HEPHL1 revealed a typical six-domain multi-copper ferroxidase structure with type I , type II and binuclear type III copper binding sites , predicted to coordinate six copper atoms[15] . Like HEPH , HEPHL1 is predicted to anchor to the plasma membrane due to the presence of a putative transmembrane region at the C-terminus . A role for HEPHL1 in mediating iron efflux in placental trophoblasts during iron transport from mother to developing fetus has been proposed [15–17] , but not yet confirmed; the role of HEPHL1 in iron transport remains uncertain . In this study , we demonstrate a potential physiological role for HEPHL1 based upon two different pathogenic HEPHL1 mutations found in a patient who clinically presented with abnormal hair ( pili torti and trichorrhexis nodosa ) , combined-type attention deficit hyperactivity disorder ( ADHD ) , speech articulation disorder , increased joint mobility , severe heat intolerance , and chronic leg pain . We show that each mutation adversely affects the ferroxidase activity and post-translational modification of HEPHL1 . Remarkably , complete ablation of Hephl1 in mouse leads to a curly whisker ( vibrissae ) hair phenotype , supporting an important role for the ferroxidase activity of HEPHL1 in hair development . We evaluated a 5-year old Caucasian male of non-consanguineous Native American and Mexican descent with abnormal hair growth , and early cognitive delays ( see S1 Text for a full clinical phenotype description ) . At birth , his hair was thick and black and distributed evenly on his scalp . He had no eyebrows but did have full eyelashes . Anterior hair loss gradually progressed to total alopecia by six months of age . His hair then regrew in a patchy distribution , sparsely in the temporal areas and more along the crown . Physical examination at age 5 years revealed elfin facies , absent lateral third of his eyelashes , sparse eyebrows and coarse hair texture ( Fig 1A , left ) , reminiscent of X-linked recessive Menkes disease ( MIM 309400 ) . Light microscopy of both short and long hair demonstrated pili torti and trichorrhexis nodosa ( Fig 1A , right ) . Given the possible diagnosis of Menkes disease , we performed relevant biochemical testing . The patient had normal levels of serum copper ( 115 μg/dL; normal range 85–150 ) and ceruloplasmin ( 34 . 7 mg/dL; normal range 24–46 ) , and normal plasma catecholamines ( norepinephrine 95 pg/mL; normal range 80–498 , epinephrine estimation <23 pg/mL with an interfering peak present , and dopamine 17 pg/mL; normal range 3–46 ) . Molecular analysis of the Menkes disease-associated gene , ATP7A , was performed using multiplex PCR and direct DNA sequencing as previously described [18] . The patient had a previously reported single nucleotide polymorphism , H1178Y , in exon 18 of ATP7A; this polymorphism , however , occurs in the normal population at an expected frequency of 1% [19] . No other sequence alterations in the coding regions or intronic splice junctions of the ATP7A gene were found . Together , the biochemical and genetic analyses argued against the diagnosis of Menkes disease . A possible diagnosis of Ectrodactyly-ED-Clefting ( EEC ) , Rapp Hodgkin ( ( MIM 129400 ) or a related syndrome was also ruled out by our failure to find any disease-associated mutation in PCR amplified genomic DNA for analysis of TP63 ( p63; TP73L ) exons 5–8 and 13–14 , and their flanking splice sites . Subsequent whole exome sequencing ( WES ) and follow-up Sanger sequencing of genomic DNA from the proband identified compound heterozygous mutations in the HEPHL1 gene; both parents were heterozygous carriers ( Fig 1B ) . The maternal variant ( NM_001098672: c . 1063G>A; p . Ala355Thr ) is predicted to cause a canonical splice site interruption while the paternal variant ( NM_001098672: c . 3176T>C; p . Met1059Thr ) is a missense mutation leading to a Met1059Thr change in a copper oxidase domain of HEPHL1 . Both variants are present in the Exome Aggregation Consortium ( ExAC ) Browser and Genome Aggregation Database ( gnomAD ) ( see URLs ) in the European population at extremely low frequencies ( 0 . 0001074 and 0 . 000101 for the maternal and paternal variants , respectively ) . Neither variant is present in the homozygous state in the ExAC/gnomAD databases . Search for other patients with biallelic variants in HEPHL1 using GeneMatcher and through other collaborators did not yield any matches . The combined annotation dependent depletion ( CADD ) Phred scores , which rank the deleteriousness of single nucleotide variants within the human genome , were 26 . 1 and 27 . 3 for the maternal and paternal variants , respectively . Both variants are predicted to be disease causing by Mutation Taster ( see URLs ) and "probably damaging" by PolyPhen-2 ( see URLs ) . Webserver wInterVar ( see URLs ) , which classifies genetic variants according to the ACMG/AMP 2015 guidelines [20] , ranked both variants as pathogenic with evidence codes PS3 , PM4 , PP3 and PS3 , PM1 , PP3 for the p . Ala355Thr and p . Met1059Thr variants , respectively . These in silico analysis tools support the contention that both variants are likely to affect the expression and function of HEPHL1 . The maternal variant of HEPHL1 , a conserved missense mutation in the last nucleotide of exon 5 ( c . 1063 G>A ) , is predicted to cause a disruption in splice site function . We found that HEPHL1 mRNA is expressed in wild type ( WT ) iPS cells at levels several times greater than those of skin fibroblasts ( S1 Fig ) . Therefore , to explore the effect of the maternal variant on splicing , we performed RT-PCR analysis on control and patient-derived iPSc mRNA samples , using a forward primer that binds to the 5’ UTR and a reverse primer that binds to exon 6 . The correctly spliced transcript was present in both control and patient cells ( Fig 2A gel , upper band , lane 2 and 3 ) , but the patient’s cells also contained a transcript that spliced directly to exon 6 , bypassing exon 5 ( Fig 2A , lower band , lane 3 ) . This led to an in-frame deletion of 255 nucleotides ( 85 amino acids ) from the maternal transcript , hence from hereon refereed to as c . 809_1063del; p . Leu271_Ala355del . Sequencing of PCR products confirmed these findings ( Fig 2A ) . The 85 amino acids encoded by exon 5 include three critical amino acid residues ( H304 , C347 and H352 ) that coordinate with a distal methionine ( M357 ) to constitute a typical type I copper binding site in domain 2; so any resulting translation product would be expected to have altered ferroxidase activity . To examine the expression of the paternal variant ( c . 3176T>C ) , direct sequencing of the PCR product obtained from the RT-PCR reaction was performed using a forward primer in exon 17 and reverse primer in exon 20 ( Fig 2B ) . The mutation affects methionine at a highly conserved position ( Fig 2C ) . Homology modeling using ceruloplasmin as a template ( 4enz . pdb ) shows that an “integral" copper ion is bound to Met 1059 , His 1003 , His 1054 , and Cys 1049 of domain 6; so the mutation to threonine could impair the assembly of this copper-binding site . Ion binding in the nearby sites , referred to as the labile and holding sites [21 , 22] , may also be influenced by the mutation ( Fig 2D ) . Next , we compared total HEPHL1 mRNA levels in control and patient iPSc using quantitative real time PCR analysis and a TaqMan reagent that detected the boundary between exons 11 and 12 . We found no significant difference between levels of mRNA expression in control and patient iPSc ( Fig 2E ) , suggesting that the exon skipping in the maternal transcript did not lead to nonsense-mediated decay and that overall expression of the full length paternal and the truncated maternal transcript in the patient was similar to that of wild-type cells . We were unable to analyze HEPHL1 protein status in iPSc or fibroblasts due to the lack of a suitable antibody that can recognize endogenous HEPHL1 protein . HEPHL1 is predicted to be a copper-dependent ferroxidase due to its similarity in sequence and putative structure to the known ferroxidases , CP and HEPH [15] . Since both patient HEPHL1 mutations are predicted to disrupt type I copper binding sites , we hypothesized that ferroxidase activity would be severely reduced . Our attempts to measure endogenous ferroxidase activity of HEPHL1 in iPSc or fibroblasts cell lines failed , likely because baseline ferroxidase activity is below the limit of detection in these cells . To circumvent this problem , we performed overexpression studies in HEK293 cells using a WT-HEPHL1 construct and constructs for the paternal ( HEPHL1 M1059T ) and maternal ( HEPHL1 Δexon 5 ) alleles . These were created by site-directed mutagenesis using a commercially available HEPHL1 expression vector with a C-terminal myc-DDK tag ( Origene , Rockville , MD ) as template . Transfection of these constructs into HEK293 cells resulted in robust expression , as illustrated by western blot analysis using anti-DDK antibody ( Fig 3A , lanes 2–4 ) . To measure HEPHL1 ferroxidase activity , we prepared protein extracts under native conditions from the overexpressing HEK293 cells . Equivalent amounts of protein extracts were then separated by nondenaturing gel electrophoresis followed by an in-gel ferroxidase assay . After incubation of the gel in saturated ferrous ammonium sulfate solution for 2 h , the gel was incubated with a ferrozine solution . Ferrozine turns red-purple when Fe ( II ) is bound . Oxidation of Fe ( II ) to Fe ( III ) indicates ferroxidase activity and can be seen by the formation of a discrete clear area ( band ) in the gel . As shown in Fig 3B , a discrete band was observed in WT-HEPHL1 lane , suggestive of active ferroxidase; the level was slightly higher than that attributed to 3 μg of purified ceruloplasmin , a known ferroxidase ( lane 5 ) . In contrast , we did not observe any ferroxidase activity when HEPHL1 M1059T or HEPHL1 Δexon 5 were expressed ( Fig 3B , lanes 3 and 4 ) . It remained possible , however , that HEPHL1 M1059T or HEPHL1 Δexon 5 retained some low level of ferroxidase activity . Therefore , we developed a simple and quantitative ferrozine-based colorimetric assay to quantitatively assess ferroxidase activity by the conversion of Fe ( II ) to Fe ( III ) ( S2 Fig ) . We used this assay to measure ferroxidase activity in lysates of HEK293 cells expressing HEPHL1 constructs . Expression of WT-HEPHL1 led to a significant reduction in the absorbance of the Fe ( II ) –ferrozine complex as compared to the mock-transfected cell lysate ( Fig 3C ) , reflecting significant ferroxidase activity . Consistent with the results of the in-gel assay , expression of either HEPHL1 M1059T or HEPHL1 Δexon 5 resulted in non-detectable conversion of Fe ( II ) to Fe ( III ) . In fact , the Fe ( II ) –ferrozine absorbance was identical to that of mock-transfected cells , indicating that both mutants were catalytically inactive in the assay . To confirm that the inability of HEPHL1 M1059T and HEPHL1 Δexon 5 to oxidize Fe ( II ) was not simply due to lack of expression , we carried-out an anti-DDK western blot on lysates . As shown in lower panels of Fig 3C , WT-HEPHL1 , HEPHL1 M1059T and HEPHL1 Δexon 5 were robustly expressed at comparable levels . Taken together , these results provide strong evidence that the paternal and maternal mutations in HEPHL1 each completely abolish ferroxidase activity . Immunoblot analysis of lysates from HEK293 cells transfected with WT-HEPHL1 consistently identified an additional , higher molecular weight species , presumably a post-translationally modified form of HEPHL1 ( Fig 3A , lane 2 and Fig 3C , lane 2 ) . This form was lost in HEPHL1 M1059T and HEPHL1 Δexon 5 ( Fig 3A , lanes 3 and 4 , Fig 3C , lanes 3 and 4 ) , suggesting that these mutations interfere with the post-translational modification of HEPHL1 that , as shown previously for HEPH , likely involves glycosylation [23] . To explore this , we immunoprecipitated WT-HEPHL1 and HEPHL1 M1059T using anti-DDK beads and treated with glycosidases that cleaved both N-linked and O-linked glycans . As shown in Fig 3D , the higher molecular weight form of WT-HEPHL1 completely disappeared after the treatment , indicating that this higher molecular weight form is glycosylated HEPHL1 ( Fig 3D , lanes 3–5 ) . Treatment with glycosidases also slightly increased the electrophoretic migration of HEPHL1 M1059T ( Fig 3D , lanes 6–8 ) suggesting that HEPHL1 M1059T may also be glycosylated , albeit to a lesser extent . As further evidence for HEPHL1 glycosylation , mass-spectrometry analysis revealed that WT-HEPHL1 is N-linked glycosylated at three conserved asparagines ( N161YT , N407AS , N772RT ) , while HEPHL1 M1059T is glycosylated at only two of these sites ( N161YT , N772RT ) . Interestingly , the Δexon 5 mutant failed to show glycosylation at any of the three sites ( S3 Fig ) . Additionally , we detected two O-linked glycosylation sites in WT-HEPHL1 ( S1076 and T1077 ) , but glycosylation of neither site was affected in HEPHL1 M1059T and HEPHL1 Δexon 5 . To determine whether copper might be required for HEPHL1 N-linked glycosylation , we transfected HEK293 cells with the WT-HEPHL1 construct and treated the cells over a 2-day period with either ammonium tetrathiomolybdate ( ATTM ) or bathocuproinedisulfonic acid ( BCS ) to deplete copper . Using three different amounts of untreated , ATTM-treated , and BCS-treated cell lysates , we ran an anti-DDK western blot . As shown in Fig 3E , untreated WT-HEPHL1 appeared in both the unmodified and the high molecular weight glycosylated forms ( lanes 1–3 ) , but treatment of cells with ATTM ( lanes 4–6 ) or BCS ( lanes 7–9 ) markedly reduced the higher molecular weight , glycosylated form . The lower , un-modified form of HEPHL1 remained unchanged . These results suggest that copper is required for maintaining HEPHL1 in the mature glycosylated form , whether by fostering glycosylation or by inhibiting deglycosylation . Several studies have shown the importance of copper for the enzymatic activity and/or structural integrity of HEPH and CP [23 , 24] . Since both paternal and maternal mutations affect the residues involved in type I copper binding sites in HEPHL1 , we investigated whether decreased copper availability leads to impaired activity of HEPHL1 . We therefore transfected HEK293 cells with WT-HEPHL1 , HEPHL1 M1059T or HEPHL1 Δexon 5 constructs and treated the cells with BCS . WT-HEPHL1 transfected cells that were treated with BCS had the same ferroxidase activity as for mock transfected HEK293 cells ( Fig 3C , BCS treatment ) , and significantly lower than its activity in the absence of BCS , suggesting that copper depletion eliminated the ferroxidase activity . As expected , no further change in ferroxidase activity was found in either HEPHL1 M1059T or HEPHL1 Δexon 5 transfected cells upon treatment of cells with BCS . Taken together , these results indicate that the ferroxidase activity seen in HEPHL1 is copper-dependent . Given that the patient mutations abrogated the ferroxidase activity of HEPHL1 , we assessed the basal intracellular iron content in dermal fibroblasts . The mutant fibroblasts showed a statistically significant increase in basal iron content compared to that of control fibroblasts ( Fig 4A ) . Western blot analysis also showed an increased expression of ferritin in patient cells ( Fig 4B , lanes 1 and 2 ) . Treatment of fibroblasts with FeCl3 led to a further increase in ferritin expression with more pronounced accumulation in patient cells ( Fig 4B , lanes 3 and 4 ) . Treatment of cells with the iron chelator deferoxamine ( DFO ) reduced the ferritin expression in both control and patient fibroblasts ( Fig 4B , lanes 5 and 6 ) , consistent with the modulation of ferritin expression by intracellular iron content . We were curious if impaired ferroxidase activity of HEPHL1 also influences the activity of copper dependent enzymes . We therefore measured the activity of the lysyl oxidase ( LOX ) enzyme , which requires copper [Cu ( II ) ] for enzymatic activity . To our surprise , lysyl oxidase activity in the patient's fibroblasts was significantly reduced compared to unaffected controls ( Fig 4C ) . To further explore this phenomenon , we analyzed the two different forms of the LOX protein ( i . e . , the immature prolysyl oxidase and the mature , active enzyme ) in whole cell extracts and the medium collected from confluent cultures of the patient’s and control fibroblasts . Immunoblot analysis using an antibody that detects all the forms of LOX showed no difference in the levels of the 52 kDa prolysyl oxidase in whole cell extracts ( lanes 1–3 ) . However , levels of the 29 kDa , catalytically functional LOX were significantly decreased ( by 37% ) in the patient’s fibroblast medium compared to that of two unaffected controls ( lanes 4–6 ) . This finding suggests that reduction in the levels of lysyl oxidase and the corresponding enzymatic activity in the patient’s fibroblasts were due to altered post-translational processing of this enzyme in the absence of functional HEPHL1 . To explore the physiological consequences of loss of HEPHL1 activity , we generated Hephl1 knockout ( Zp-/- ) mice . Exon 2 of the Hephl1 gene was chosen for gene targeting because it is located near the start of the protein-coding region and encodes residues that make up the Type II copper binding site required for ferroxidase activity . Removal of exon 2 ( 245 bp ) also leads to a frameshift that introduces an early stop codon . Quantitative PCR analysis of Hephl1 mRNA extracted from Zp+/+ and Zp-/- placental tissues confirmed the loss of exon 2 . In addition , only residual expression of downstream exons was detected ( Fig 5B ) , indicating instability of the Hephl1 transcript lacking exon 2 . Phenotypically , Zp-/- mice were viable and bred successfully , and the genotype ratios in the progeny were consistent with normal perinatal viability . All mice with ablation of Hephl1 exhibited short , curled whiskers ( vibrissae ) throughout life ( Fig 5C and 5D ) . Heterozygous mice ( Zp+/- ) did not exhibit the curly whisker phenotype , indicating that this phenotype is recessive . Although a full characterization of these mice was beyond the scope of the present study , we investigated whether the lysyl oxidase abnormalities we observed in the patient’s fibroblasts are recapitulated in our mouse model . We established mouse embryonic fibroblasts ( MEFs ) cultures from Zp+/+ and Zp-/- embryos and measured lysyl oxidase activity . Consistent with the results in patient’s fibroblasts , we also found a significant reduction in the activity of lysyl oxidase in Zp-/- MEFs ( Fig 4E ) . Taken together , our results strongly suggest that lysyl oxidase activity is regulated by HEPHL1 . The identification of biallelic mutations ( p . Leu271_ala355del and p . Met1059Thr ) in the HEPHL1 gene of a 5-year-old boy prompted us to investigate the function of normal and mutant HEPHL1 proteins . First , we examined the copper-binding sites . A unique characteristic of multi-copper oxidases is the presence of at least one of each of three types of copper binding sites: type I , type II and binuclear type III . A type I copper site accepts electrons from the substrate while the trinuclear cluster , comprising a type II and a binuclear type III site , operates as a center where dioxygen is reduced to two water molecules [9] . CP , a six-domain multi-copper oxidase , contains three type I copper sites ( in domains 2 , 4 and 6 ) and a trinuclear cluster at the interface of domains 1 and 6 [25] . Based on the known crystal structure of CP , homology modeling indicates that all type I , II , and III copper sites for the 6 copper ions in CP are also present in HEPHL1 [15] . In the patient , both the paternal and maternal mutations affect residues involved in the binding of copper at type I sites , which leads to loss of ferroxidase activity . A highly conserved methionine involved in the patient’s paternal mutation ( M1059 ) is part of the most essential type I copper binding site that is couple to the trinuclear cluster in domain 6 . The maternal deletion of exon 5 removes 85 residues in domain 2 of HEPHL1 , including three residues ( H304 , C347 , H352 ) that coordinate with a distal methionine ( M357 ) to constitute a typical type I site . For CP , while the type I site in domain 2 is atypical [25] , a mutation at this site was shown to prevent copper incorporation , suggesting that this domain 2 site is critical for the ferroxidase function of CP [24] . The deletion of much of domain 2 in HEPHL1 resulting from the maternal mutation presumably disrupts the overall protein structure , eliminating copper binding in domain 2 and quite possibly elsewhere . We consistently found that wild-type-HEPHL1 appeared as two separate bands , with the higher molecular weight band presumably representing a post-translationally modified form of HEPHL1 . For the structurally similar ferroxidase HEPH , the higher molecular weight form reflects glycosylation of the protein . Transfection experiments performed in polarized differentiated T84 cells and subsequent pulse-chase analysis revealed that , although HEPH is synthesized as a single chain polypeptide of 144 kDa , it runs as a double band because the newly synthesized HEPH is modified to a mature 161 kDa species by N-linked glycosylation [23] . Similarly , in this paper , we show that the higher molecular weight form of HEPHL1 ( Fig 3D ) is glycosylated . Whereas previous studies have revealed that the percentage of glycosylated HEPH was higher than the non-glycosylated form [23] , we observed a smaller percentage of the glycosylated form of WT-HEPHL1 compared to the non-glycosylated form . This may indicate that transiently expressed proteins have not reached full maturation after 48 h . Interestingly , we found that the glycosylation and ferroxidase activity of HEPHL1 is copper-dependent . Using HEK293 cells overexpressing WT-HEPHL1 , we showed that copper is necessary for the subsequent glycosylation and formation of mature catalytically active holoenzyme . The chelation of copper with BCS abrogated both glycosylation and ferroxidase activity , likely due to rapid degradation of the unstable apo-HEPHL1 moiety . Further studies will be needed to determine the precise function of glycosylated and non-glycosylated forms of HEPHL1 . These HEPHL1 mutations have functional consequences . First , both the paternal ( M1059T ) and maternal ( Δexon 5 ) mutants , which involve copper binding sites , showed complete lack of ferroxidase activity . This was associated with a modest but significant increase in intracellular iron content and ferritin expression in the patient’s cultured fibroblasts ( Fig 4A and 4B ) , supporting a role for HEPHL1-mediated ferroxidase activity in modulating intracellular iron content . Although HEPHL1 does not appear to be the principal ferroxidase in blood , and the patient’s serum iron was 107 μg/dL ( normal , 30–110 ) , HEPHL1-mediated ferroxidase activity could play a more specific role in certain tissues or cells , where loss of its activity cannot be fully compensated by other ferroxidases . Analysis of mRNA expression suggests that HEPHL1 is robustly expressed in rodent placenta and embryo and at a lower level in heart and kidney , while no expression was found in liver or enterocytes . At the protein level , HEPHL1 was detected in rat placenta , mouse mammary tissue and whole embryo , as well as BeWo , MCF7 and T47D cell lines . Interestingly , no protein expression of HEPHL1 was detected in mouse serum or enterocytes , where CP and HEPH , respectively , are prominently expressed [15] . This distribution suggests a tissue-specific requirement for ferroxidases . Another function of HEPHL1 is likely manifest in the hair follicle , which depends upon transition metals for its structure [26] . A recent genome-wide transcriptome analysis by deep RNA-sequencing identified Hephl1 as a signature gene of the mouse hair follicle’s transient amplifying cells ( HF-TACs ) from postnatal day P ( 5 ) back skin [27] . Interestingly , expression of Hephl1 in TACs is significantly higher than both Heph and Cp ( www . hair-gel . net ) , suggesting a unique role for Hephl1 in these specialized progenitor cells in regulating hair follicle morphogenesis . Indeed , our patient with biallelic HEPHL1 mutations displayed sparse , twisted , brittle and easily broken scalp hairs , with similar findings in the eyebrows and eyelashes . The boy’s hair also exhibited two different structural abnormalities , i . e . , trichorrhexis nodosa ( broken , nodular shafts ) and pili torti . Pili torti ( hair twisted ) is a rare , congenital or acquired condition characterized by twisted and flattened hair shafts at irregular intervals , caused by changes in cell shape and thickness of both the outer and inner root sheaths [28] . The Online Mendelian Inheritance in Man ( OMIM ) database identified 24 entries for pili torti illustrating its association with a wide spectrum of neurological disorders including Björnstad syndrome [29] ( MIM 262000 ) , ectodermal dysplasias such as Rapp-Hodgkin syndrome [29] ( MIM 129400 ) as well as Menkes disease [29] ( MIM 309400 ) , also known as kinky hair syndrome [30 , 31] . We also note that Belted Galloway cattle with homozygous loss-of-function mutations in HEPHL1 have hypotrichosis [32] , and our Hephl1 knockout mouse exhibits curly whiskers resembling pili torti . A third function of HEPHL1 could be to serve as a cuprous oxidase , catalyzing the conversion of Cu ( I ) to Cu ( II ) . Other ferroxidases such as bacterial CueO , yeast Fet3p and human CP also exhibit cuprous oxidase activity , using Cu ( I ) as substrate [33–36] . Kinetic analysis of Fet3p showed that both ferroxidase and cuprous oxidase reactions are catalyzed at the same type I copper site [35] . Along with the data presented here showing that the patient’s mutations completely abrogated ferroxidase activity , this raises the possibility that HEPHL1 mutations might also interfere with the cuprous oxidase activity of the enzyme . The end result would be reduced availability of Cu ( II ) for lysyl oxidase . Indeed , we found that the levels of the catalytically active form of lysyl oxidase ( Cu ( II ) bound holoenzyme ) and its corresponding enzymatic activity were significantly diminished in fibroblasts from our patient . Consistent with this analysis , we also observed a significant reduction in lysyl oxidase activity in MEFs derived from Hephl1 knockout embryos . Taken together , these results provide an insight into a possible physiological role of HEPHL1 in regulating the activity of lysyl oxidase and likely other cuproenzymes . In summary , we have established HEPHL1 as a physiologically important ferroxidase whose deficiency in a human , as in animals , results in structural hair abnormalities . The patient's mutations lead to abolished ferroxidase activity , and the patient's cultured fibroblasts exhibited increased iron accumulation and reduced amounts of mature , secreted lysyl oxidase and corresponding enzymatic activity . Of interest , we observed increased joint mobility in our patient , which could be a consequence of decreased lysyl oxidase activity causing lax connective tissue . We do not know if our patient’s mild cognitive delays , heat intolerance , and chronic leg pain are related to the HEPHL1 mutations . Identification of additional patients with HEPHL1 loss-of-function mutations should shed further light on the clinical phenotype associated with this condition , and the Hephl1 knockout mouse can be used to further investigate HEPHL1 function . The patient and family members were enrolled in the National Institutes of Health Undiagnosed Diseases Program ( UDP ) [37–39] and in protocol 76-HG-0238 , “Diagnosis and Treatment of Patients with Inborn Errors of Metabolism and Other Genetic Disorders” approved by the Institutional Review Board ( IRB ) of the National Human Genome Research Institute ( NHGRI ) . Written informed consent was obtained , and the patient was evaluated at the NIH Clinical Center . All work performed on mice was in accordance with the National Institutes of Health ( NIH ) guidelines , as described in the Guide for the Care and Use of Laboratory Animals of the NIH , and with approval from the Office of Laboratory Animal Care at the University of California , Berkeley , and the QIMR Berghofer Medical Research Institute Animal Ethics Committee . Genomic DNA was extracted from blood leukocytes following standard protocols . Whole exome sequencing ( WES ) was performed using an Agilent 38MB Sure Capture System . Image analysis and base calling were performed using Illumina Genome Analyzer Pipeline software ( versions 1 . 13 . 48 . 0 ) with default parameters . Reads were aligned to a human reference sequence ( UCSC assembly hg19 , NCBI build 37 ) using the Efficient Large-scale Alignment of Nucleotide Databases ( Illumina , San Diego , CA , USA ) package . Genotypes were called at all positions where there were high-quality sequence bases using the Most Probable Genotype Bayesian algorithm [40] , and variants were filtered using the graphical software tool VarSifter v1 . 5 [41] . Validation of HEPHL1 sequence variants and segregation with disease were confirmed by Sanger sequencing . Genomic DNA flanking the site of paternal or maternal variants was amplified with HotStar Taq DNA polymerase ( Qiagen , Valencia , CA ) using the following primer sets: set 1 paternal , CTTTCCTGGGACATTCCAAA and TCCTGTTTTGGGGGTCTACA; set 2 maternal , CCAGCCACCTTCCTTACAAC and TGAGCACTAGTGACTGTGTGGTT . PCR conditions were: initial denaturation at 95°C for 5 min , followed by 35 cycles of denaturation at 95°C for 30 s , annealing at 55°C for 30 s , extension at 72°C for 30 s , followed by a final extension step at 72°C for 5 min . PCR products were purified using ExoSAP-IT ( Affymetrix ) and sent to the Macrogen service center ( Macrogen USA , Rockville , MD ) for sequencing . Sequencing files were evaluated using Sequencher v5 . 0 software ( Gene Codes Corporation , Ann Arbor , MI ) . Primary dermal fibroblasts from the patient were derived from a forearm skin biopsy and cultured in DMEM supplemented with 10% fetal bovine serum ( Thermo Fisher Scientific , Waltman , MA ) , MEM nonessential amino acid solution ( Sigma-Aldrich , St . Louis , MO ) , penicillin–streptomycin and 2 mM L -glutamine ( Sigma-Aldrich ) . Control fibroblasts were obtained from the ATCC ( PCS-201-012 , Manassas , VA , USA ) and Coriell Institute of Medical Research ( GM08398 , GM00942 Camden , New Jersey ) . Induced pluripotent stem ( iPS ) cell derivation and characterization using early passage dermal fibroblasts from the patient were done at ALSTEM ( ALSTEM cell Advancements , Richmond , CA ) by ectopic expression of OCT4 , SOX2 , KLF4 , and c-MYC genes using episomal plasmids . Routine culture of iPS cells was performed using mTesR1 with 5X Supplement ( STEMCELL Technologies , CA ) as described in detail under ALSTEM protocol iPS11 ( www . alstembio . com/protocols ) . All cell lines were grown in a 37o C incubator with 5 . 0% CO2 . Total RNA was extracted from control and patient cells using a RNeasy Mini Kit ( Qiagen ) . Genomic DNA contamination was removed using the Turbo DNA free kit ( Thermo Fisher Scientific ) following the manufacturer’s instructions . Single-stranded cDNA was synthesized using the High Capacity RNA-to-cDNA kit ( Applied Biosystems ) . To analyze HEPHL1 mutations at the cDNA level , PCR amplifications of the control and patient cDNA were carried out using forward primer CTGCTTGAGTTACATCCACAA and reverse primer TGACCCTTCATCTTGGGGTA for the maternal variant c . G1063A , and forward primer CTTTCCTGGGACATTCCAAA and reverse primer GACCCAGATTCTTGCCAAAG for the paternal variant c . T3176C . PCR products were separated on a 1% agarose gel and extracted using a gel extraction kit ( Qiagen ) . Direct sequencing of the PCR products was carried out by Sanger sequencing as described above . For relative quantitation of the HEPHL1 transcript levels , real time PCR was performed with an ABI7300 Genetic Analyzer ( Applied Biosystems ) using the FAM-labeled TaqMan Probe and primers mix ( Thermo Fisher Scientific , HS01376171_m1 ) ; the primers anneal to the junction between exon11/exon 12 in HEPHL1 . Gene expression values were normalized to the expression of the reference gene glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) ( Thermo Fisher Scientific Hs03929097_g1 ) . A three-dimensional model of human HEPHL1 was built based on its homology to ceruloplasmin , using Prime software tools ( Schrodinger , LLC ) . Of the 1043 HEPHL1 residues modeled , ranging from Thr 26 to Asn 1068 , 541 ( 52% ) are identical to the nearest residue in the ceruloplasmin template structure 4enz . pdb [22] . The model was rendered using the programs MolScript [42] and Raster3D [43] . The pCMV6 plasmid containing the entire sequence of the human HEPHL1 ORF ( NM_001098672 ) with a C-terminal MYC-DDK tag [myc-DDK-HEPHL1 , hereafter WT ( wild type ) -HEPHL1] was obtained from Origene ( RC214648 , Origene Technologies , Rockville , MD , USA ) . The Q5 Site-Directed Mutagenesis Kit ( New England Biolabs ) was used to generate both the maternal expression construct lacking exon 5 ( myc-DDK-HEPHL1 Δexon 5 , hereafter HEPHL1 Δexon 5 ) and the paternal expression construct ( myc-DDK-HEPHL1 M1059T , hereafter HEPHL1 M1059T ) using the manufacturer’s protocol . pCMV6-Entry , an empty vector with a C-terminal Myc-DDK Tag ( PS100001 , Origene Technologies ) was used as a control/mock vector in the transfection assays . All constructs were verified by DNA sequencing . HEK293 cells in the exponential growth phase in 10 cm culture dishes were transiently transfected with 10–12 μg of WT-HEPHL1 , HEPHL1 Δexon 5 , or HEPHL1 M1059T constructs using Lipofectamine 2000 reagent ( Thermo Fisher Scientific ) according to the manufacturer’s instructions . Cells were incubated for 48 h with a change of medium the day after transfection . For copper-chelation , 200 μM of either ammonium tetrathiomolybdate ( ATTM ) ( Sigma-Aldrich ) or 200 μM bathocuproinedisulfonic acid ( BCS ) ( Sigma-Aldrich ) was added to the medium immediately after transfection , and the medium ( with drug ) was changed the day after transfection . Cells were collected , washed twice with PBS and lysed using native lysis buffer ( Abcam , Cambridge , MA ) with Protease ( Roche ) and Phosphatase ( Roche ) Inhibitors . Lysates were kept on ice for 30 min followed by mechanical shredding by passing through a 25 5/8 G needle 5–7 times . After centrifuging to pellet insoluble material , supernatants were transferred to a new tube and protein concentrations were measured using the DC protein assay ( Bio-Rad , Hercules , CA ) . To each sample , equal amounts of 2X Laemmli Sample Buffer ( Bio-Rad ) with beta-mercaptoethanol were added before boiling at 95°C . Samples were then fractionated on 4–15% Mini-Protean TGX Stain Free Gel ( Bio-Rad ) and run at 150V , followed by transfer onto a low fluorescence PVDF or nitrocellulose membrane using the Trans-Blot Turbo system ( Bio-Rad ) . Membranes were blocked in Odyssey blocking buffer ( LI-COR ) for 2 h at room temperature and probed with an anti-DDK mouse monoclonal antibody ( Origene TA50011-100 , Clone OTI4C5 ) overnight at 4°C . Membranes were then washed with TBST ( Tris-buffered saline with 0 . 05% Tween 20 ) followed by incubation with IRDye 800 CW or IRDye 680RD goat anti-mouse secondary antibody ( LI-COR Biosciences ) for 1 h at room temperature . Images were captured using the LI-COR Odyssey CLx imaging system . Mouse monoclonal anti-Vinculin ( V9131 , Sigma-Aldrich ) was used to confirm equivalent sample loading . Lysates were prepared from transiently transfected HEK293 cells with WT-HEPHL1 , HEPHL1 Δexon 5 , or HEPHL1 M1059T constructs as described above for immunoblot analysis . To visualize in-vitro ferroxidase activity , equal amounts of 2X native PAGE sample buffer ( Bio-Rad ) were added to lysates and samples were fractionated on 4–15% Mini-Protean TGX Stain Free Gels ( Bio-Rad ) at a constant 100 V for 21 min , then at 33 V for 7 h , using native PAGE running buffer ( Bio-Rad ) . Gels were then incubated at 37°C for 2 h in 0 . 00784% ferrous ammonium sulfate [Fe ( NH4 ) 2 ( SO4 ) 2 6H2O] solution prepared in 0 . 1 M sodium acetate , pH 5 . 0 . After incubation , gels were placed in 15 mM ferrozine Solution ( Sigma-Aldrich ) and monitored for clear area ( band ) formation . Images were captured using the Bio-Rad ChemiDoc imaging system . For the quantitative measurement of in vitro ferroxidase activity , equivalent amounts of lysates were incubated with 30 μL of anti-DDK Beads ( Origene ) for 4 h at 4o C with rotation . After incubation , beads were washed three times with PBS containing 0 . 5% Triton X100 and re-suspended in 0 . 5 mg/mL ferrous ammonium sulfate solution prepared in 0 . 1 M sodium acetate , pH 5 . 0 , and samples were then incubated at 37o C for 1 . 5 h ( 50 μL total reaction volume ) . Following incubation , 250 μL of 0 . 1 M sodium acetate and 30 μL non-reducing iron detection reagent ( 6 . 5 mM Ferrozine , 2 . 5 M ammonium acetate ) were added to the beads , and beads were further incubated at RT for 30 min . After centrifugation of samples for 1 minute at 10 , 000 rpm , 280 μL of each sample was carefully transferred to a 96 well plate , and the iron concentration was measured by absorbance at 550 nm using a SpectraMax plate reader . Purified human ceruloplasmin ( 25 μL; Athens Research and Technology ) , used as positive control , was diluted in 0 . 1M sodium acetate pH 5 . 0 and incubated with 25 μL of 0 . 5 mg/mL ferrous ammonium sulfate and processed identically in parallel . Iron standards were prepared by adding 0 , 5 , 10 , and 25 μL of 1 mM iron standard ( Bio Vision ) to a final volume of 50 μL in 0 . 1 M sodium acetate , pH 5 . 0 , and processed identically as described above , except for the use of a reducing iron detection reagent [6 . 5 mM ferrozine , 2 . 5 M ammonium acetate , 1 M ascorbic acid] in place of the non-reducing detection reagent . Ceruloplasmin was used to show that the in vitro ferroxidase assay can detect the conversion of Fe ( II ) to Fe ( III ) in an enzyme concentration-dependent manner . HEK293 cells were transiently transfected with WT-HEPHL1 , HEPHL1 M1059T , or HEPHL1 Δexon 5 constructs . After transfection , cells were collected , washed with PBS and suspended in 62 . 5 mM Tris-HCl ( pH 6 . 8 ) , and then an equal amount of 2X SDS buffer ( 62 . 5 mM Tris-HCl [pH 6 . 8] , 6% SDS , 10% glycerol , 20 mM iodoacetamide , protease and phosphatase inhibitors ) was added . Cell lysates were sonicated and centrifuged and diluted ten times with Triton X-100 buffer ( 50 mM Tris-HCl [pH 7 . 5] , 150 mM NaCl , 0 . 5% Triton X-100 , 1 mM EDTA , 10 mM iodoacetamide , protease and phosphatase inhibitors ) . Equivalent amounts of lysates were incubated with anti-DDK agarose beads ( Origene; clone OT14C5 ) for 4 h at 4°C with rotation . After incubation , immunoprecipitates were washed three times with Triton X-100 buffer . For the glycosylation analysis , enzymatic release of glycans from the immunoprecipitated HEPHL1 was performed directly on beads using NEB deglycosylation mix II to remove potential N-linked and/or O-linked glycans . The deglycosylated or non-deglycosylated protein bands were excised from the gel and digested using chymotrypsin and trypsin followed by the Liquid Chromatography-tandem Mass Spectrometry ( LC-MS/MS ) analysis ( see S2 Text for a full description of the method and instrument used for LC-MS/MS ) . Quantitation of intracellular iron concentration in dermal fibroblasts was performed as described by Reimer et al . [44] . Briefly , cells were lysed using 50 mM NaOH for 2 h , neutralized by addition of equal volume of 10 mM HCl and sonicated . Protein concentration was determined using the DC protein assay , and lysates were then treated with freshly prepared acidic KMnO4 solution ( 1 . 4M HCl and 4 . 5% ( w/v ) KMnO4 in H2O ) at 60°C for 2 h with shaking . The total iron content of the lysates was determined by the addition of an iron detection reagent ( 6 . 5 mM ferrozine , 1 M ascorbic acid , 2 . 5 M ammonium acetate ) followed by measurement of absorbance at 550 nm . To measure the expression of ferritin in fibroblasts , a rabbit polyclonal anti-ferritin heavy chain antibody ( Santa-Cruz Biotechnology , sc-25617 ) was used . Lysyl oxidase activity was measured using an Amplite Fluorimetric Lysyl Oxidase Assay Kit ( AAT Bioquest Inc . , Sunnyvale , CA ) following the manufacturer’s instructions . Briefly , 50 μL of cell culture medium from fibroblasts growing in the presence or absence of the lysyl oxidase inhibitor 2-aminopropionitrile ( BAPN , Sigma-Aldrich ) were collected and mixed with an equal volume of assay mixture containing horseradish peroxidase ( HRP ) and Amplite HRP substrate . After incubation at 37°C , the change in fluorescence was measured using a fluorescence plate reader ( Molecular Probes ) at Ex/Em 540/590 nm . The specific fluorescence attributed to lysyl oxidase activity in each sample was determined after subtracting the fluorescence obtained in the presence of BAPN . Total protein contents were determined after lysis of cells in SDS buffer and used for the normalization of lysyl oxidase activity . To visualize different forms of lysyl oxidase , unaffected controls and patient’s fibroblasts were grown to confluence in T-75 flasks and then incubated with phenol red and serum-free medium for another two days . Equal amounts of conditioned cell media were collected and concentrated using an ultra-15 centrifugal filter unit ( EMD Millipore ) . Cell lysates were prepared by lysis of cells in Triton X-100 buffer ( 150 mM NaCl , 1% Triton X-100 and 50 mM Tris ( pH 8 . 0 ) , with protease and phosphatase inhibitors ) . Western blot analysis was performed using a LOX specific antibody ( Abcam , Ab174316 ) . Heterozygous mice with conditional ( floxed ) Hephl1 allele were commercially created for us by the UC Davis Mouse Biology Program . The targeting construct used to generate the mice contained LoxP sites that flanked exon 2 , a 11 . 5 kb 5’ arm of homology ( containing exon 1 ) , a 11 . 2 kb 3’ arm of homology ( containing exons 3–6 and the 5’ portion of exon 7 ) , a Diphtheria Toxin A ( DTA ) cassette , and a Neomycin ( neo ) cassette flanked by FRT sites . JM8 . F6 C57BL/6N embryonic stem ( ES ) cells were microinjected with the targeting construct and one colony that successfully incorporated the targeting construct and had a high percentage of cells ( 88% ) with the expected karyotype was selected . The cells were injected into BALB/c blastocysts and implanted into pseudo-pregnant females . Three high percentage chimeric males were generated ( estimated 75–85% C57BL/6 cells based on coat color ) and bred with wild-type C57BL/6 females . Black pups were genotyped and females heterozygous for the Hephl1 floxed allele ( still bearing the FRT-flanked neo cassette ( Zpfl ( neo ) /+ ) were used for further breeding . In order to remove the neo cassette , the Zpfl ( neo ) /+ females were crossed with a C57BL/6J mouse bearing the FlpE transgene [ ( “FlpE” mouse , B6 . Cg- Tg ( ACTFLPe ) 9205Dym/J , The Jackson Laboratory ) kindly provided by Dr . Ian Tonks at QIMR Berghofer] . When pups inherit a copy of the FlpE transgene , the FlpE recombinase protein is expressed and excises regions of DNA flanked by FRT sites , in this case the neo cassette . To make the Hephl1 floxed strain ( Hephl1tm1 . 1Vul , denoted as Zpfl/fl hereafter ) , the Zpfl/+ FlpE mice lacking the neo cassette were first bred with wild-type C57BL/6J mice , and then those containing a Hephl1 floxed allele , but lacking the FlpE transgene ( Zpfl/+ ) , were backcrossed onto the C57BL/6J strain for multiple generations until at least 97% C57BL/6J . These Zpfl/+ mice were then bred together to generate Zpfl/fl mice . To make the whole body Hephl1 knockout strain ( Hephl1tm1 . 2Vul , hereafter referred to as Zp-/- ) , Zpfl/fl mice were bred with C57BL/6J mice bearing the Cre recombinase transgene driven by the eIIa promoter , which is ubiquitously active ( “eIIa-cre” mice , B6 . FVB-Tg ( EIIa-cre ) C5379Lmgd , The Jackson Laboratory ) . Expression of Cre recombinase leads to excision of the region in the DNA between the LoxP sites ( exon 2 of Hephl1 ) . Progeny containing both the eIIa-cre transgene and the Hephl1 floxed allele ( Zpfl/+ EIIa ) were then bred together and pups that likely contained germline deletion of Hephl1 were selected by genotyping . These mice were then backcrossed onto the C57BL/6J line to remove both the FlpE and eIIa-cre recombinase transgenes and to generate mice at least 97% C57BL/6J and heterozygous for the Hephl1 knockout allele ( Zp+/- ) . To generate Zp-/- for this study , Zp+/- mice were then bred together . Genotyping was done by PCR using forward primer CCTTATGACTACAGTGAACAGGGTTCTG and reverse primer CTACTCTCTGGCCCTTGCTTTTGC to amplify the wild-type allele and forward primer CGACGGCCAGTGAATTGTAATA and reverse primer GTGATAGAGCTGAGATGGCGCAA to amplify the knockout allele . Quantitative analysis of Hephl1 mRNA expression was carried out using total RNA extracted from Zp+/+ and Zp-/- frozen placental tissues by TRIzol reagent ( Thermo Fisher Scientific ) . cDNA was synthesized using M-MLV Reverse Transcriptase ( Thermo Fisher Scientific ) and an oligo ( dT ) primer as per manufacturer’s instructions . Real-time quantitative PCR was performed with a CFX 384 detection system ( Bio-Rad ) using iTaq Universal SYBR Green supermix . Each sample was analyzed in triplicate and gene expression was calculated from the Cq value using the standard curve method . Gene expression levels were normalized to the expression of the housekeeping gene hypoxanthine guanine phosphoribosyltransferase ( Hprt ) . Primer validation and analyses were in accordance with the MIQE guidelines[45] . The primer pairs used were: Hephl1 Exon 2 forward: GGTGGGATCTACAAGAAGGCG; Hephl1 Exon 2 reverse: GTCTCCCACTTCTGCCCTCA; Hephl1 Exon 18–19 forward: GTTTGCTGATCACCCAGGAACA; Hephl1 Exon 18–19 reverse: TCCAGAAGGCGTCTTGGTAGAAT; Hprt forward: GGACTGATTATGGACAGGA; Hprt reverse: GAGGGCCACAATGTGATG . Mouse embryonic fibroblasts ( MEFs ) were derived from embryonic day ( E ) 12 . 5 embryos according to established procedures [46 , 47] and cultured using high glucose DMEM ( Thermo Fisher Scientific ) supplemented with 10% FBS , 1% GlutaMAX ( Thermo Fisher Scientific ) and 1% penicillin–streptomycin . For the measurement of lysyl oxidase activity , MEFs of defined genotype at passage 3–4 were seeded into 12-well plates at a density of 2 . 5 x105 cells/well . When the cells reached confluence , the medium was carefully aspirated and replaced with fresh phenol red-free complete culture medium . Twenty-four hours later , 500 μM of BAPN was added to the wells . After another 24 hours , the cell surface was flushed gently several times with the existing culture medium , which was then collected for lysyl oxidase ( LOX ) activity measurement . LOX activity was measured as described for human fibroblasts . Cells were then washed once with PBS , and 200 μL of lysis buffer ( 25 mM Tris-HCl , pH 7 . 2 , 25 mM NaCl with 2% SDS ) was added into each well to lyse the cells . Total protein contents from the clarified supernatants of the cell lysates were determined by BCA protein assay ( Thermo Scientific ) and used for normalization of the fluorescence readings . The percentage of lysyl oxidase-specific fluorescence for each MEF cell line was calculated by dividing the difference in fluorescence between untreated and BAPN-treated wells by the fluorescence obtained from the untreated wells .	Multi-copper ferroxidases play a critical role in maintaining iron homeostasis in humans . Two well-characterized ferroxidases , ceruloplasmin and hephaestin , facilitate iron transport in different tissues by oxidizing ferrous iron to the ferric form , which is subsequently carried by transferrin . Hephaestin like 1 ( HEPHL1 ) is a new member of the multicopper oxidase family , with a typical six-domain multi-copper ferroxidase structure containing type I , type II and binuclear type III copper binding sites , predicted to coordinate six copper atoms . The physiological role of HEPHL1 in iron homeostasis is unknown . In this study , we provide initial insights into a possible physiological role for HEPHL1 by functionally characterizing two different mutations found in a patient who presented with abnormal hair ( pili torti and trichorrhexis nodosa ) , combined-type ADHD , speech articulation disorder , increased joint mobility , severe heat intolerance , and chronic leg pain . Whole exome sequencing revealed biallelic loss of function mutations in HEPHL1 . We show that both mutations adversely affect the ferroxidase activity of HEPHL1 . Remarkably , complete ablation of Hephl1 in the mouse leads to a curly whisker ( vibrissae ) hair phenotype , supporting an important role for the ferroxidase activity of HEPHL1 in hair growth and hair disorders .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "medicine", "and", "health", "sciences", "integumentary", "system", "molecular", "probe", "techniques", "immunoblotting", "fibroblasts", "animal", "models", "model", "organisms", "animal", "anatomy", "connective", "tissue", "cells", "experimental", "organism", "systems", "glycosylation", "molecular", "biology", "techniques", "zoology", "research", "and", "analysis", "methods", "hair", "artificial", "gene", "amplification", "and", "extension", "animal", "cells", "proteins", "animal", "studies", "gene", "expression", "connective", "tissue", "biological", "tissue", "mouse", "models", "molecular", "biology", "animal", "physiology", "biochemistry", "cell", "biology", "anatomy", "post-translational", "modification", "vibrissae", "genetics", "biology", "and", "life", "sciences", "cellular", "types", "glycobiology", "polymerase", "chain", "reaction" ]	2019	Biallelic HEPHL1 variants impair ferroxidase activity and cause an abnormal hair phenotype
The secretion of extracellular vesicles ( EVs ) in helminth parasites is a constitutive mechanism that promotes survival by improving their colonization and adaptation in the host tissue . In the present study , we analyzed the production of EVs from supernatants of cultures of Echinococcus granulosus protoscoleces and metacestodes and their interaction with dendritic cells , which have the ability to efficiently uptake and process microbial antigens , activating T lymphocytes . To experimentally increase the release of EVs , we used loperamide , a calcium channel blocker that increases the cytosolic calcium level in protoscoleces and EV secretion . An exosome-like enriched EV fraction isolated from the parasite culture medium was characterized by dynamic light scattering , transmission electron microscopy , proteomic analysis and immunoblot . This allowed identifying many proteins including: small EV markers such as TSG101 , SDCBP , ALIX , tetraspanins and 14-3-3 proteins; proteins involved in vesicle-related transport; orthologs of mammalian proteins involved in the immune response , such as basigin , Bp29 and maspardin; and parasite antigens such as antigen 5 , P29 and endophilin-1 , which are of special interest due to their role in the parasite-host relationship . Finally , studies on the EVs-host cell interaction demonstrated that E . granulosus exosome-like vesicles were internalized by murine dendritic cells , inducing their maturation with increase of CD86 and with a slight down-regulation in the expression of MHCII molecules . These data suggest that E . granulosus EVs could interfere with the antigen presentation pathway of murine dendritic cells inducing immunoregulation in the host . Further studies are needed to better understand the role of these vesicles in parasite survival and as diagnostic markers and new vaccines . Human echinococcosis is a zoonotic cestode disease caused by the larval stages of Echinococcus ( family Taeniidae ) . It is considered as a re-emerging and neglected disease that causes serious chronic lung and liver diseases . The two Echinococcus species of greater public health importance and economic concern worldwide are Echinococcus multilocularis , responsible for alveolar echinococcosis ( which is restricted to the Northern hemisphere ) , and Echinococcus granulosus responsible for cystic echinococcosis ( which is globally distributed ) [1 , 2] . The larval stage of these parasites develops as metacestodes ( fluid-filled cysts ) in the viscera ( mainly in the liver ) of mammalian intermediate hosts . Metacestodes are formed by a thin cellular layer ( germinal layer ) from which protoscoleces ( larval form that can develop either in an adult worm in the final host or in a secondary hydatid cyst in an intermediate host ) bud . The cells of the germinal layer secrete the laminated layer , an acellular and carbohydrate-rich layer that surrounds the metacestode [2 , 3] . The laminated layer , only present in the genus Echinococcus , appears to be a key component of the host-parasite interface , being involved in the maintenance of the cyst physical integrity and in the interaction with the host immune system [2 , 4] . This structure is composed of mucins containing defined galactose-rich carbohydrates , and , in E . granulosus , is also accompanied by calcium inositol hexakisphosphate deposits ( InsP6 ) [3] . These helminth parasites lack digestive and excretory systems but have developed active endocytic and exocytic cellular processes to regulate metabolite uptake and excretion [5] . In previous studies , we have determined an increased exocytosis rate in the E . granulosus larval stage , which could be controlled by calcium concentration and in which proteins such as calcineurin and calpains are involved [6] . These proteins have been reported to be involved in unconventional vesicle-mediated protein secretion and in inflammatory responses [7] . Recently , it has been suggested that the endo/exosomal vesicular trafficking pathways share common features with autophagy [8 , 9] , which is an active process in E . granulosus both in basal conditions and after pharmacological treatment [10 , 11] . Helminth parasites release several molecules , such as proteases , glycolytic enzymes and protease inhibitors into the mammalian hosts [12] . These products are known as excretory/secretory products which are exposed to the host immune system and could be involved in its modulation and the consequent parasite survival [13] . In this context , extracellular vesicles ( EVs ) are considered interesting target structures due to their potential role in parasite-parasite and host-parasite communication [14 , 15] . Depending on their intracellular site of origin , composition and size , EVs are classified into exosomes , ectosomes or microvesicles , and apoptotic and autophagic vesicles [16–18] . Although EVs exhibit a varied range of sizes , exosomes are considered small vesicles ( sEVs ) of typically 30–150 nm which originate from the inward budding of late endosomes that form multivesicular bodies ( MVBs ) with intraluminal vesicles ( ILVs ) [19] . When MVBs fuse with the plasma membrane , ILVs are released as exosomes from the cell surface . Consequently , the biochemical composition of exosomes is associated with their biogenesis , including proteins from the endosomal-sorting complexes required for transport ( ESCRT ) pathway [20] . Although , the EV composition is presumably context-dependent , no universal and specific EV markers are yet available [21] . Nevertheless , Kowall et al . [22] have recently proposed the proteins Syntenin-1 ( Syndecan Binding Protein -SDCBP- ) and Tumor Susceptibility Gene 101 ( TSG101 ) as markers of bona fide exosomes in mammalian systems . Microvesicles comprise larger structures than exosomes ( usually 100–1000 nm ) and are directly produced by budding from the plasma membrane , generally as a consequence of an external stimulus that causes an intracellular Ca+2 increase [23] . Several studies have reported that helminth parasites secrete EVs that could play important roles during infection [24–28] . It is known that cestodes such as Taenia crassiceps , Mesocestoides corti and E . multilocularis secrete EVs with protein and miRNA cargo that can modulate the host immune system [29] . Recently , dos Santos et al . [30] confirmed the presence of EVs in hydatid fluid from fertile and infertile metacestodes of E . granulosus , whereas simultaneously , Siles-Lucas et al . [31] demonstrated that E . granulosus cysts secrete exosome-like vesicles into the hydatid fluid and that these vesicles contain proteins involved in cyst survival . Since information on the function of EVs in this cestode is still limited , the aims of this study were to characterize the EVs produced by the larval stage of E . granulosus and to investigate the interaction of EVs with host cells , to find out whether this interaction plays a role in host immunomodulation . In addition , given that the release of EVs can depend on different stimuli like calcium increase and therapeutic treatment , we analyzed the occurrence of exosome-like vesicles and the protein composition of EVs released from control parasites and parasites treated with loperamide , a calcium channel agonist with anti-echinococcal effect [32] . The animal study was carried out in agreement with National Health Service and Food Quality ( SENASA ) guidelines , Argentina and with the 2011 revised form of The Guide for the Care and Use of Laboratory Animals published by the U . S . National Institutes of Health . The Animal Experimental Committee at the Faculty of Exact and Natural Sciences , Mar del Plata University approved the experimental protocols ( permit number: 2555-08-16 ) . Specific-pathogen free female CF-1 mice ( 28–35 g ) were provided by the SENASA . A minimum number of animals were used in each experiment . The animals ( five mice per cage ) were kept under controlled laboratory conditions ( temperature ±20°C , 12 hour light/12 hour dark with lights off at 8 . 00 p . m . ) . They were maintained with water and food ad libitum , monitored daily and placed in a clean cage with fresh sawdust every 3 days . E . granulosus metacestodes were obtained from the peritoneal cavity of mice injected with 1500 protoscoleces in suspension . For each experiment , the infected mice were anesthetized with ketamine-xylazine ( 50 mg/kg/mouse-5 mg/kg/mouse ) and sacrificed by cervical dislocation at 6–8 months post infection . All efforts were made to minimize suffering . Echinococcus granulosus protoscoleces were obtained from lung and liver of infected cattle presented for routine slaughter at the abattoir in the province of Buenos Aires , Argentina . The viscera were transported to the laboratory where the hydatid cysts were aseptically opened to remove the laminar and germinal membranes along with the hydatid fluid and the protoscoleces . Protoscoleces were exhaustively washed in Phosphate Buffered Saline ( PBS ) and maintained in sterile conditions until in vitro culture . Protoscolex in vitro culture ( n = 3 , 000/9 . 5cm2 ) , and viability assays were carried out as previously described [33] . Briefly , they were cultured in medium 199 ( Gibco ) supplemented with antibiotics ( penicillin , streptomycin , and gentamicin 100 μg/ml ) and glucose ( 4 mg/ml ) in Leighton tubes at 37°C without changing the medium . Vitality was determined by methylene blue exclusion test . Otherwise , E . granulosus metacestodes ( 10–20 cysts for each drug treatment , with diameters ranging between 5 and 15 mm and free from the adventitial layer ) were aseptically obtained from the peritoneal cavities of CF-1 mice 6–8 months after intraperitoneal infection with protoscoleces ( n = 1500 ) [34] . They were cultured in the same conditions than protoscoleces and viability was assessed based on the collapse of the germinal layers . Since it is known that intracellular calcium increase plays a role in exosome release [35–38] , and the loperamide can rise the cytosolic free Ca+2 concentration ( [Ca+2]i ) [39] , the addition of this drug at the parasite cultures , could ensure the high EVs production from E . granulosus . In vitro protoscolex- and metacestode-sub-lethal treatments were assayed with loperamide dissolved in dimethyl sulfoxide ( DMSO ) at 20 and 50 μM as final concentrations . Parasites incubated in culture medium containing 0 . 1% DMSO were used as controls . Changes in [Ca+2]i using Fluo-3 acetoxymethyl ester ( Fluo-AM ) probe were fluorometrically monitored [33] . Experiments were performed with 5 x 103 protoscoleces and incubated with 50 μM of loperamide for 4 h . Pretreatments with 1 mM EGTA plus 100 μM BAPTA-AM calcium chelators were performed for 15 min . Then , fluorescence was recorded with a spectrofluorimeter ( model F-4500; Hitachi ) . The excitation and emission were set at 488 nm and 505–530 nm , respectively . Parasite-autofluorescence was individually corrected and untreated controls were included in each replication . Experiments were done in quintupled . Statistical analysis was done with the nonparametric Mann–Whitney test , a p-value of less than 0 . 05 was considered significant . Extracellular vesicles were enriched by differential centrifugation [40] . Briefly , 9000 protoscoleces or 45 cysts were maintained in serum-free media for 5 days and incubated in control conditions or with 20 μM of loperamide for 16 more hours . Following , the parasite culture medium was collected and centrifuged at 300 xg , 10 min; then at 2000 xg , 10 min and finally at 10000 xg for 30 min to remove large dead cells and large cell debris . The supernatant was ultracentrifuged at 100 , 000 xg for 1 h to pellet the vesicles in an Optima LE-80k ultracentrifuge ( Beckman ) using a 90 Ti rotor . To remove contaminating proteins , the pellet was washed with 3 ml of PBS and finally centrifuged at the same high speed . EVs were resuspended in 30 μl PBS and protein concentration was determined by absorbance at 280 nm with a Nanodrop spectrophotometer and by Bradford method for supernatants from the ultracentrifugation step . Statistical analysis was done with the Kruskal-Wallis test with Dunn's multiple comparisons post-test; p-values of < 0 . 05 were considered to be significant . Finally , EVs were stored at -80°C until experimental use . The 100 , 000 xg supernatant was collected , lyophilized , resuspended in 100 μl nuclease-free water and stored at -20°C for protein content comparison . Size distribution profile and number of the vesicles isolated from protoscoleces of E . granulosus were performed with DLS using a Zetasizer Nano ( Nano ZSizer-ZEN3600 , Malvern , U . K . ) at the Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas ( NIFTA- Argentina ) . Briefly , samples were captured at 25°C ± 1°C and diluted 1:50 in pre-filtered PBS and measure at a scattering angle of 173° with a He-Ne 633 nm laser . A total of 6 scans , each with duration of 60 s were recovered for each sample . Extracellular vesicles were fixed in 2% paraformaldehyde in PBS and send refrigerated to a Transmission Electron Microscopy ( TEM ) external service for analysis ( Centro Regional de Investigaciones Básicas y Aplicadas de Bahía Blanca -CRIBABB- Argentina ) . Procedures were carried out as described in [40] . EV-preparations were placed on a 300-mesh Formvar coated copper grids , negative stained with 1% ( w/v ) uranyl acetate for 1 min and examined at 100kV in a JEOL JSM 100CX II transmission electron microscope . Fiji software was used to evaluate the diameter of the vesicles observed in MET pictures . Data were statistically compared using the Kruskal-Wallis test with Dunn's multiple comparisons post-test; p values of < 0 . 05 were considered to be significant . Additionally , control and loperamide-treated protoscoleces were also fixed and observed by TEM as previously described [11] . Briefly , parasites were fixed with 3% glutaraldehyde in sodium cacodylate buffer for 24 h at 4°C . Then , they were send refrigerated to a TEM external service for analysis ( Servicio Central de Microscopía Electrónica de la Facultad de Ciencias Veterinarias , Universidad Nacional de La Plata ) where they were post-fixed in 2% OsO4 in cacodylate buffer , dehydrated in a graded acetone series and subsequently embedded in resin epoxy and examined with a JEM 1200 EX II ( JEOL Ltd . , Tokio , Japan ) transmission electron microscope at 80 kV . The purified EVs ( 100 μg ) and the supernatants were run 1 cm into the resolving gel of a 10% SDS-PAGE . Then , the gel was stained with colloidal Coomassie Blue G-250 and the samples cut from the gel and sent to the CEQUIBIEM proteomic service ( Buenos Aires , Argentina ) for mass spectrometry analysis and protein identification . Briefly , the samples were incubated with 20 mM dithiothreitol for 45 min at 56°C for reduction , and with 20 mM iodoacetamide for 45 min at room temperature in darkness for alkylation . Then , the samples were digested using trypsin and were processed by nano-HPLC ( EASY-Spray Accucore , Thermo Scientific , West Palm Beach , FL , USA ) coupled to a mass spectrometer with Orbitrap technology ( Q-Exactive , Thermo Scientific , West Palm Beach , FL , USA ) allowing peptide separation and identification . The sample ionization was made by electrospray ( EASY-SPRAY , Thermo Scientific , West Palm Beach , FL , USA ) and the data analysis was carried out by the Proteome Discoverer software version 1 . 4 , Thermo Scientific . Only proteins with at least two peptides in two replicates were selected for further analyses . The in silico analyses to establish the subcellular location and Gene Ontology ( GO ) classification of the identified proteins were performed using the UniProt database and software ( http://www . uniprot . org/ ) . Additionally , these proteins were also classified using the Reactome pathway database ( https://reactome . org/ ) and manually based on data from the available literature . Moreover , the identified proteins were compared with those cataloged in the ExoCarta database ( http://www . exocarta . org/ ) . The proteins identified as “uncharacterized , hypothetical , conserved or expressed protein” were analyzed and classified based on the presence of conserved domains using ProDom ( http://prodom . prabi . fr/prodom/current/html/form . php ) , CDART ( https://www . ncbi . nlm . nih . gov/Structure/lexington/lexington . cgi ) and CDD ( https://www . ncbi . nlm . nih . gov/cdd/ ) . Finally , to determine whether the uncharacterized proteins were secreted by classical or nonclassical secretory pathways , we used SignalP server ( http://www . cbs . dtu . dk/services/SignalP/ ) and SecretomeP server ( http://www . cbs . dtu . dk/services/SecretomeP/ ) [41] , respectively . These softwares were developed for bacterial and mammalian systems but they are also used in helminths [42 , 43] . Isolated EVs were lysed in CytoBuster protein extraction reagent ( Novagen ) , supplemented with protease and phosphatase inhibitors ( Thermo Fisher Scientific ) . Protein quantification was performed using the BCA Protein Assay ( Pierce ) . A volume of 10 μl of EV-ultracentrifugated pellet ( containing 30 μg of proteins in control samples and 69 μg of proteins in loperamide-treated samples ) were loaded for all samples and analyzed simultaneously on 10% SDS-PAGE under non-reducing conditions . Polypeptides were electroblotted onto a nitrocellulose membrane ( HyBond C; Amersham , Argentina ) at 43 mA for 60 min . Following , the membranes were incubated in blocking solution ( TBST: 20 mM Tris-HCl , 150 mM NaCl , 1% Tween-20 , pH 7 . 6 containing 2% bovine serum albumin for 4 h at 20°C ) and were probed with a 1:1000 dilution of mouse monoclonal antibodies raised against human CD9 ( BD Pharmingen , clone M-L13 ) and CD63 ( MEM-259 , ImmunoTools clone MEM-259 ) to detect cross-reactivity with E . granulosus tetraspanins of approximately 25 and 50 kDa respectively . The corresponding antibodies datasheets do not allow the identification of the “antigen portion” used for their generation . Therefore , we analyzed the identity along the entire antigenic protein and the proteins of interest through sequence alignment . Also , we performed an antibody recognition ability analysis based on the identification of similar linear and conformational epitopes between human immunogen and parasite tetraspanins using BepiPred 2 . 0 ( http://www . cbs . dtu . dk/services/BepiPred/ ) and CEP-Conformational Epitope Prediction Server- ( http://196 . 1 . 114 . 49/cgi-bin/cep . pl ) . Finally , the blots were incubated with anti-rabbit immunoglobulin peroxidase-linked species-specific whole antibody ( GE Healthcare , cat no . NA934V ) and ECL reagents ( GE Healthcare , cat no . RPN2106V1 ) to detect the chemiluminescent signal on film . A protein extract from human peripheral blood mononuclear cells ( PBMCs ) was used as a positive control . Isolated EVs were labeled with PKH26 fluorescent dye ( Sigma-Aldrich ) according to the manufacturer’s instructions . In brief , 10 μl purified EVs were resuspended in 10 μl of Diluent C and mixed gently with PKH26 ( added to a final concentration of 2 μM ) in 150 μl of final volume for 35 min at 37° C in darkness . To stop the staining , they were incubated with BSA 1% for 10 min at room temperature . Then , the samples were washed with PBS followed by ultracentrifugation at 100 , 000 xg for 1 h to remove the excess dye and finally were resuspended in 150 μl of PBS . Negative controls consisted of the resuspended pellet after ultracentrifugation step labeled with the fluorescent dye alone , without purified EVs Bone Marrow-derived Dendritic Cells ( BMDCs ) were produced by flushing bone marrow of femurs and tibias of CF-1 mice ( 6–8 weeks old ) as previously described with minor modifications [44] . Briefly , cells suspensions were depleted of erythrocytes with RBC lysing buffer ( BD Bioscience , San Jose , CA ) . Cells were plated at 1 x 106/ml in 6-well culture plates with 3 ml of supplemented RPMI 1640 ( 5% heat-inactivated fetal calf serum -Gibco; Invitrogen- , 100 U/ml penicillin/streptomycin , 10 μg/ml gentamicin and 2 mM L-glutamine , -all from Life Technologies , Grand Island , NY- ) . To induce DC-differentiation , cells were cultured in presence of 100 ng/ml Flt3-L ( R&DSystems ) at 37°C in 5% CO2 for 6 days . Finally , DC-population was characterized by flow cytometry using fluorescence-conjugated monoclonal antibodies ( mAbs ) directed against CD11b ( M1/70 ) , CD11c ( HL3 ) , CD3 ( 145-2C11 ) , CD45R/B220 ( RA3-6B2 ) , SiglecH ( eBio440c ) , CD172a ( P84 ) and CD24 ( M1/69 ) ( eBiosciences , San Diego , CA ) . Approximately 70–90% of the cells were CD11c+ . Endocytosis of EVs by BMDCs and maturation assays was performed by flow cytometry and confocal microscopy . BMDCs ( 1 x 106 cells/ml ) were cultured with or without 150 μl PKH26 labeled-extracellular vesicles purified ( 10 μl pellets recovered by ultracentrifugation coming from culture supernatants of 3000 protoscoleces ) from control or 20 μM loperamide-treated samples for 30 min at 37°C . Cells were then washed gently , pelleted and maintained for 18 h in culture before harvested . Incubation of BMDCs with EVs at 4°C was used as negative control of endocytosis . In addition , to determine DC maturation , the cells were stimulated for 18 h with 100 ng/mL of lipopolysaccharide ( LPS , Sigma-Aldrich Co , positive control ) . Fluorescein isothiocyanate ( FITC ) or phycoerythrin-conjugated mAbs directed to CD11c ( HL3 ) , CD40 ( HM40-3 ) , CD80 ( 16-10A1 ) , CD86 ( GL1 ) , MHC class I ( AF6-88 . 5 . 5 . 3 ) and MHC class II ( M5 / 114 . 15 . 2 ) were from eBioscience ( San Diego , CA , USA ) . In all cases , isotype-matched control antibodies were used , and a gate ( R1 ) was defined in the analysis to exclude all nonviable cells and debris , based on size and propidium iodide staining . The analysis was performed using a PartecCyflow Space ( Sysmex , UK ) flow cytometer , and the FlowJo software ( Treestar ) . The results are expressed as the mean fluorescence intensity or as the percentage of positive cells . Data were statistically compared using the Kruskal-Wallis test with Dunn's multiple comparisons post-test; p values of < 0 . 05 were considered to be significant . Immediately after incubation of the BMDCs with the EVs , cells were then harvested and plated on alcian blue-treated coverslips ( 12 mm ) during 20 min at room temperature . Then , the cells were washed with PBS-BSA 2% in a wet chamber and fixed in 4% PFA and permeabilized with 0 . 05% saponin . Afterward , they were incubated with mAb MHC class II-FITC antibody ( eBioscience , San Diego , CA , USA ) for 1 h at 37°C , washed and incubated with 50 ng/ml DAPI ( Sigma-Aldrich , USA ) to counterstained nuclei . Coverslips were mounted on glass slides using Fluoromount G . Immunofluorescence and images were acquired with an inverted confocal laser scanning microscope ( Nikon , Confocal Microscope C1 ) using a 60 x oil immersion objective with an excitation/emission wavelength 485/538 nm for FITC , 358⁄461 nm for DAPI and 551/567 nm for PKH26 . Fluorescent intensity and co-localization analysis were performed using Histogram and Coloc 2 plugins in Fiji software . Briefly , to quantify MHCII modulation in cell surface , a total of ten cells in absence or presence of EVs was analyzed . Image files were loaded as separate image stacks . Then , surrounding background was subtracted before different region of interest ( ROI ) were analyzed to obtain the mean intensity values . For co-localization of MHCII molecules with EVs labeled with PKH26 the Pearson’s coefficient ( r ) was used to analyze the correlation of the intensity values of green and red pixels in dual-channel images . Statistical analysis was done with the nonparametric Mann–Whitney test , a p-value of less than 0 . 05 was considered significant . Since exocytosis could be regulated by [Ca+2]i , loperamide was used as an exocytic stimulus applying a concentration and time of incubation that guaranteed parasite viability [32] . Loperamide exposure ( 50 μM ) increased free [Ca+2]i 6-fold over a 4 h incubation period in comparison with the control ( Fig 1A ) . The fluorescence signal diminished around control values after pretreatment with a mixture of EGTA ( an extracellular chelator ) plus BAPTA-AM ( a membrane-permeable calcium chelator ) in the medium . In order to determine whether E . granulosus larval stage produces EVs , we followed a series of centrifugal steps of increasing speed as was described by Théry et al . [40] from which small EVs ( sEVs or nanovesicles ) were mainly isolated [22] . The identification and characterization of EVs focused on DLS for size distribution determination ( Fig 1B ) and TEM for morphology assessment ( Fig 1C and 1D ) . The diameter of the majority of EVs was within the expected size range for exosome-like vesicles . The 96 , 2% of EVs from control protoscoleces was among 30–90 nm ( 45 . 04 ± 10 . 06 nm ) and the 100% of EVs from loperamide-treated parasites showed a size range among 35–110 nm ( 68 . 18 ± 25 . 35 nm ) . However , a minor population of EVs from control protoscoleces ( 3 . 8% ) shows a size of 135 . 3 ± 44 . 25 nm ( Fig 1B ) . TEM analysis confirmed the presence of sEVs with the typical cup-shaped structures of 25–150 nm ( Fig 1C and 1D ) in accordance with DLS outcome . EVs from protoscoleces and hydatid fluid exhibit similar sizes while those from metacestodes display higher diameters even though they are not statistically significant ( Fig 1D ) . Additionally , EVs were more abundant in protoscolex-cultures than in metacestode-cultures and their hydatid fluids indicated by protein concentration ( 6 ± 1 μg/μl , 1 . 1 ± 0 . 5 μg/μl and 0 . 7 ± 0 . 5 μg/μl , respectively , Fig 1C and 1E ) . TEM pictures from entire protoscoleces also show structures compatible with EVs ( exosome-like vesicles , microvesicles ) and MVBs with ILVs associated with the tegument and surface of protoscoleces ( S1 Fig ) . Besides , sub-lethal treatment of protoscoleces with the drug induced ultrastructural changes in the tegument , such as disorganization of the distal cytoplasm and lack of glycocalyx and microtriches , while it was unaltered in control condition ( S1 Fig ) . To characterize the protein-cargo and analyze the possible differences between the EVs obtained from untreated- and loperamide-treated parasites we carried out a proteomic analysis of the sEV-enriched fraction of parasites incubated in both conditions . In samples from metacestodes , we identified a very low number of proteins due to a low recovery of total proteins . Nevertheless , we could successfully identify 5 and 13 proteins in control and loperamide condition , respectively . Among these proteins , gelsolin , heat shock protein 70 , tetraspanins and 14-3-3 protein which are usually present in exosomes , were identified . Otherwise , a total of 298 proteins were identified in samples of loperamide-treated protoscoleces . Of these proteins , 112 were common to both control and loperamide-treated samples whereas the remaining 186 proteins were exclusively found in the treated sample ( Fig 2A and S1 and S2 Tables ) . Proteins were classified into 10 categories based on the analysis with Reactome database and information from the literature . In spite of the fact that loperamide increased the identified-protein number and the release of EVs including exosome-like vesicles , the abundance of proteins involved in calcium homeostasis , immune system/host interaction and antigens was greater in control conditions . On the other hand , proteins involved in metabolism , transport of molecules , signal transduction , vesicle-mediated transport/membrane trafficking and developmental biology/cellular migration were expressed in higher proportion in loperamide samples respect to the control ( Fig 2B ) . A detailed examination at peptide level revealed that most of the common proteins showed a peptide abundance of 2-fold and 3-fold enriched in drug-treated samples compared with untreated-samples , except for some proteins such as multidrug resistance proteins , Ca-ATPase and Antigen 5 , where the expression was increased between 5-fold and 9-fold in the sEVs released from loperamide-treated parasites ( S1 Table and S1 and S2 Files ) . In addition , we observed a high prevalence of uncharacterized proteins which are of special interest for their putative role in the parasite-host interaction . They were analyzed and classified based on the presence of conserved domains using ProDom , CDART and CDD which allowed the identification of 7 putative antigenic proteins , 2 tetraspanins , 2 thioredoxin-like proteins , 1 galectin/galactose-binding lectin , among others ( S3 Table ) . Besides , we performed a sequence analysis using SignalP and SecretomeP which revealed that 17% and 23% of uncharacterized proteins were secreted by classical and nonclassical secretory pathways , respectively ( S3 Table ) . Moreover , we compared our data with those of ExoCarta database . We observed that 56 out of 298 proteins share homology with the proteins listed in ExoCarta comprising some exosomal markers such as syntenin 1 and TSG101; proteins present in specific sEVs such as tetraspanins , ALIX , annexin A6 and EH-domain containing protein; proteins present in multiple EVs such as heat shock 70 kDa proteins , annexins , beta-actin and tubulin alpha-1C; and proteins present in large EVs such as eukaryotic elongation factor 2 and actinin ( Fig 2C ) . We also identified other proteins that account for the presence of ectosomes such as Rho family of small GTPases ( including RhoA and Cdc42 ) and proteins implicated in MVB trafficking such Rab family of small GTPases and SNARE proteins ( S1 and S2 Tables ) . On the other hand , certain proteins were exclusively detected in EVs under loperamide-treatment such as TSG101 ( exosomal marker tumour susceptibility ) , EPS8-like protein ( as part of epidermal growth factor receptor ( EGFR ) signalling ) , prominin-1 ( CD133 , a pentaspan protein ) and the transforming growth factor-beta-induced protein ig-h3 ( TGFBI , also known as keratoepithelin ) . Finally , tetraspanins , which are very prevalent in exosomes and are involved in their biosynthesis were assayed by immunoblot analysis . The anti-CD9 and anti-CD63 antibodies used are directed against proteins which showed 26–30% or 21–23% amino acid identity with the four Eg-tetraspanins identify in our proteomic analysis ( EUB61600 and EUB60810 in control and loperamide-samples and EUB63772 and EUB54099 only from loperamide-samples ) . Additionally , the antigenic regions of both human CD9 and the Eg-tetraspanins are coincidental , exposed , and with coil structure , suggesting that the probability of recognition by anti-CD9 include the four parasite proteins . The transmembrane tetraspanins homolog to human CD9 ( with an expected size of approximately 25 kDa ) were enriched in the EV preparations from loperamide-treated samples ( Fig 2D ) . The immunodetection of tetraspanins in drug-treated samples was surely due to higher protein loading of this sample in relation with the control . The sub-optimal protein concentration from control samples could result in the observed undetected signal . Dendritic cells ( DCs ) are the unique professional antigen-presenting cell able to link the initiation of an antigen-specific response to the microbial mediators forming a decisive interface between innate and adaptive immunity . When they capture foreign antigens in the periphery undergo phenotypic , functional and migratory changes that allow them to present processed antigenic peptides to naïve T cells . In this way , DCs handle the adaptive immune system in a proinflammatory or tolerogenic profile [45] . In this context , we performed a functional analysis to disclose potential roles of E . granulosus EVs in host-parasite interactions . To monitor the internalization of EVs , we stained them with the red-fluorescent lipophilic compound PKH26 . Coincubation of 1 x 106 BMDCs for 30 min at 37°C with labeled and extensively washed EVs revealed that these cells internalized the vesicles as can be seen by confocal microscopy analysis in Fig 3A . As expected , when EVs and BMDCs were incubated at 4°C , no ingestion of vesicles was observed due to the loss of fluidity and endocytic capacity of the plasma membrane . When we compare control with LPS-treated BMDCs we observed similar amounts , but a different cellular distribution of the vesicles internalized . When EVs were exposed to control BMDCs , the dye PKH26 show a homogeneous fluorescent pattern with some punctate dot structures ( Fig 3A , S2 Fig ) . Remarkably , the internalized vesicles by LPS-treated BMDCs show a dotted pattern at the perinuclear region . In addition , higher levels of the PKH26 dye have been observed in BMDCs incubated with purified EVs coming from loperamide-treated protoscoleces . It is also important to note the colocalization of the EVs and MHCII indicated by the high Pearson’s R value ( 0 . 95–0 . 96 ) specially in EVs from treated parasites ( Fig 3A ) . This suggests that they are located at the same cellular site , probably related to endosomal-lysosomal compartments where preformed MHCII molecules are stored . Addiotionally , we performed a control to evaluate the labeling specificity using the dye purified without EVs which revealed only a 3–4% of cells with diffuse signal and nonspecific extracellular fluorescence in comparison with the 40% positive labeled cells observed in presence of EVs ( S2 Fig ) . Due to vesicles secreted by E . granulosus also transport antigenic proteins , we , therefore , use the MHCII and the costimulatory molecule CD86 as markers of the maturation of DCs . A slight down-regulation in the expression of MHCII was evidenced by confocal microscopy when BMDCs were in contact with purified EVs from control protoscoleces ( Fig 3A and 3B ) . Conversely , FACS analysis showed that the E . granulosus EVs induce the up-regulation of CD86 whereas similar expression was observed for CD40 , CD80 , MHCI and MHCII in a unified CD11c+ DCs population ( Fig 3C and S3 Fig ) . In our assays , we have detected 3 different subpopulations of dendritic cells . Plasmacytoid dendritic cells ( pDC , CD11c+ B220+ SinglecH+ ) and conventional dendritic cell ( cDC ) : CD11b+ like DCs ( CD11c+ , CD172a+ ) and CD8+like DCs ( CD11c+ CD24+ and CD172a- ) . However , the proportion of pDCs ( with mainly antiviral and antitumor activity ) was always less than 10% , and the 3 populations analyzed separately did not show any changes in the response against the stimulation of the exosomes-like from E . granulosus . Extracellular vesicles have been widely related to the host-parasite relationship and cell comunication and are considered relevant for the infection and persistence of the parasite in the host [15 , 46] . In this work , we isolated and characterized for the first time the EVs produced in in vitro cultures of E . granulosus protoscoleces and metacestodes obtained from infected mice , and we analyzed their biological function after contact with host DCs . In the course of the production of this manuscript , two studies evidenced EVs in hydatid fluid [30 , 31] and another one evidenced EV secretion by E . multilocularis [29] . As mentioned in the Results section , we isolated sEVs from the larval stage of E . granulosus which were similar in size , morphology ( Fig 1 ) and protein content ( Fig 2 , S1 and S2 Tables ) to the exosomes characterized in reports refered to Echinococcus sp . , other parasites ( Echinostoma caproni , Heligosomoides polygyrus , Fasciola hepatica , Schistosoma japonicum , Taenia crassiceps , Mesocestoides corti , among ohers ) and mammals [24–26 , 28–31 , 47] . Dynamic light scattering assay and TEM images show exosome-like structures between 25 and 150 nm in diameter , which displayed a usual cup-shaped morphology ( Fig 1B and 1C ) , thus allowing us to confirm the presence of exosome-like vesicles . Besides , mass spectrometry allowed us to identify a number of well-known exosomal and sEV markers , including TSG101 , SDCBP and ALIX ( Fig 2C , S1 and S2 Tables ) . Also , in accordance with previous reports , our results of immunoblot showed that these EVs were enriched in CD9-like tetraspanin , which could have a role in stabilizing membrane microdomains and increasing exosome production ( Fig 2D ) [22 , 48] . Although some authors have previously identified CD63-like tetraspanins from E . granulosus [49 , 50] , in the present study we were not able to detect them under our western blot conditions probably due to a low antibody cross-reactivity against parasite tetraspanins . These proteins have been suggested as promising targets for vaccination or anti-parasitic therapy against E . multilocularis and Opisthorchis viverrini , respectively [51 , 52] . In the present study , proteomic analysis allowed us to identify 112 and 298 proteins from EVs of control and loperamide-treated protoscoleces respectively of which 38 and 56 were specifically enriched in orthologs of mammalian proteins present in exosomes , respectively ( Fig 2A and 2C ) . Additionally , they were compared with those previously reported for Echinococcus and other flatworm parasites ( S1 and S2 Tables ) revealing that the majority of components found in EVs ( such as exosomal markers , proteins involved in MVB biogenesis , vesicle trafficking , enzymes , chaperones and cytoskeleton proteins ) and several antigens were common among helminth parasites [26 , 28 , 29 , 31] . Coincidentally with Siles Lucas et al . [31] we found in the ten most common proteins SDCBP , ezrin/moesin/radixin and antigen 5 ( S1 and S2 Tables ) . The proteins coming from the membrane and cytosol included: heat shock proteins , signal or scaffolding proteins , endosome-membrane proteins ( Annexins and Rabs ) , Heteromeric G proteins and Rab effectors ( otoferlin and synaptotagmin-like protein ) , which are associated with exosome biogenesis ( Fig 2C ) [20] . Given that these proteins are representative of the ESCRT dependent pathway , we propose that this could be the main route involved in the exosome biogenesis in the Echinococcus larval stage , as previously described for adult F . hepatica [28] . In this line of evidence , the TEM images of E . granulosus protoscoleces displayed in S1 Fig demonstrated the potential occurrence of structures similar to MVBs in the distal cytoplasm of the larvae and the presence of EVs . All these findings support the assumption that the isolated nanovesicles in our preparations are exosome-like vesicles . However , although we had a population enriched in nanovesicles ( with absence of nuclear and mitochondrial proteins and presence of known and uncharacterized proteins belonging to non-classical secretory pathways , Fig 1B and S1–S3 Tables ) , our purification protocol lacked a gradient separation step . Thus , we cannot discard the presence of contaminating soluble proteins and other types of EVs secreted by the parasites . Indeed , we identified proteins present in microvesicles ( indicated with asterisks in Fig 2C ) and soluble proteins previously identified as excretory/secretory products in this cestode [53–55] . Probably , the observed increases in intracellular calcium trigger a signal cascade that stimulates the release of microvesicles from the tegumental surface , also increasing the ratio of these vesicles in loperamide-treated samples respect to the controls [23] . Supporting this idea , in loperamide-treated samples , we observed differential expression of proteins involved in microvesicle biogenesis such as ADP-ribosylation factor 6 , as reported in F . hepatica ( S2 Table ) [28] . To boost EV release and to analyze their protein cargo in relation with untreated conditions , we added loperamide into the parasite cultures . This drug increased [Ca+2]i in both E . granulosus larval forms and in other cell systems ( Fig 1A ) [39] , enhanced the EV production in the cestode ( Fig 1 ) , as previously reported in C2 C12 myoblast cells [56] , and affected the protein composition of the released exosome-like vesicles ( Fig 2A and 2C , and S2 Table ) . Although total protein content was higher in loperamide-treated samples than in controls , the proportion of parasite antigens and calcium homeostasis proteins ( otoferlin , PDCD6 , dysferlin , among others ) was lower than in controls ( Fig 2A and 2B , S1 and S2 Tables ) . Therefore , the molecular characteristics of the EVs released from protoscoleces under physiological conditions suggest that these vesicles could mediate biological aspects of the parasitic life cycle involved in parasite-parasite and/or parasite-host interactions . On the other hand , proteins exclusively detected in sEVs under loperamide-treatment included TSG101 , which has been previously reported to increase after cancer chemotherapy [57] , and several other proteins such as EPS8-like protein , prominin-1 , TGFBI and CDC42-interacting protein , which could induce migration and cell proliferation between the host and the parasite [58–60] . Besides , the EVs purified from drug-treated protoscoleces were enriched in multidrug resistance proteins and glutathione S-transferases , which could represent a potential mechanism of the parasite to reduce the chemotherapeutic effectiveness of the drug , as previously reported in cancer cells [57 , 61 , 62] . The analysis of sEVs obtained under loperamide treatment was helpful to corroborate that these sEVs were similar in size and quality to control sEVs even in presence of initial ultrastructural alterations ( S1 Fig ) encouraging further analysis of sEVs characterization using other antiechinococcal drugs to determine the occurrence of antigenic and/or immunoregulatory proteins . Similarly to that observed by TEM in protoscolex cultures , the supernatants of E granulosus metacestode cultures were enriched in exosome-like vesicles ( Fig 1Ce and 1D ) . In support of this , nanovesicles crossing the laminated layer of Echinococcus metacestodes have been described through ultrastructural analysis of the tegument [63 , 64] . In the present study , the metacestodes used to obtain EVs were collected from the inner part of the cystic masses that are surrounded by the adventitial layer . This strategy allowed the isolation of cysts free of the collagen capsule and host cells and consequently of EVs only of parasite origin . Recently , it has been described that the metacestode stages of T . crassiceps and M . corti secrete EVs in close contact with the host , but that in E . multilocularis EVs are retained by the laminated layer [29] . The laminated layer of E . multilocularis is much thinner than that of E . granulosus; therefore , in the latter , the EVs would be mostly retained . Interestingly , the laminated layer of E . granulosus possesses crystalline granules containing deposits of the calcium salt of InsP6 ( Ca5H2L_16H2O , where L represents fully deprotonated InsP6 ) , which are absent in E . multilocularis [3] . This molecule has been reported to bind to numerous proteins that regulate intracellular vesicle traffic [65 , 66] . It is known that a true exchange of macromolecules across of the laminated layer occurs between the host and the parasite , with constant vesicular trafficking through the tegument [3 , 67] . This movement may depend on signature organellar targeting motifs within the proteins and on their interactions with certain components of the laminated layer . High-affinity InsP6-binding proteins include components of E . granulosus exosome-like vesicles such as synaptogmins , ATP-dependent RNA helicases ( DExD/H-box protein family ) , pleckstrin , ezrin/radixin/moiesin , gelsolin and galectin ( S1–S3 Tables ) [68] . Thus , it could be considered that extracellular InsP6 binds to proteins on E . granulosus exosome-like vesicles , acting as a “dynamic anchorage” that promotes the passage across the laminated layer and thus accounting for the EVs observed in the metacestode culture medium ( Fig 1Ce ) . It is widely known that exosome-like vesicles , particularly those derived from helminth parasites , mediate the immune modulation through their protein- , lipid- and RNAs-cargo [24 , 69 , 70] . Proteins contained in helminth exosome-like vesicles can modify host responses to favor parasite survival , proliferation and dissemination [24 , 28] . Thus , in the present study , we first analyzed the interaction of exosome-like vesicles with BMDCs since the latter can adsorb exosomes and thus modify T cell responses or internalize these vesicles by endocytosis , present antigens and modify their own function conditioning T cell responses [71 , 72] . Bone Marrow Dendritic Cells internalized exosome-like vesicles from E . granulosus thus promoting their maturation ( Fig 3 ) . An interesting finding was that the exposure of the DCs to EVs induced an unconventional activation profile , with increase in CD86 , but with a slight decrease in MHC class II molecules ( Fig 3 ) . The MHCII increase was only statistically significant by confocal microscopy analysis probably due to its ability to detect both intracellular and cell surface molecules in contrast to FACS analysis where only surface molecules can be detected ( S3 Fig ) . These maturation pattern on the course of differentiation of DCs have been previously reported during the exposure to hydatid fluid or purified antigen B [73] . This activation profile could be related to the modulation of the parasite to avoid antigenic presentation favoring the scape to both immunosurveillance and an effective immune response . Based on mass spectrometry analysis and in silico functional categorization of E . granulosus EVs , in the present study and increasing the data previously reported in Echinococcus , we identified several proteins with immunomodulatory functions such as a B-Cell Receptor Associated Protein 29 ( Bp29 ) , which negatively modulates the membrane expression of MHCI in HeLa cells [74] . Another protein well represented in the EVs analyzed was basigin ( also known as Extracellular Matrix Metalloproteinase Inducer–EMMPRIN- or cluster of differentiation 147 -CD147- ) , a member of the immunoglobulin superfamily which acts as the principal receptor that mediates chemotaxis by cyclophilins and regulates the responsiveness of lymphocytes by inhibition of T cell proliferation [75 , 76] . An orthologous protein to CD147 , associated with the negative regulation of T-reg cells , has also been previously identified in F . hepatica [77] . Besides , here we identified a maspardin ortholog ( known as MAST or ACP33 protein ) , which is highly conserved in metazoans and is able to bind to CD4 , acting as a negative regulator in T cell activation [78] . In the same line of evidence , we detected numerous antioxidant proteins that could neutralize the oxidative free radicals generated by host phagocytic cells such as thioredoxin peroxidase and glutathione S-transferase which have been previously described in excretory/secretory products from E . granulosus protoscoleces and hydatid fluid [53] and comprise a major detoxification system in parasites [79 , 80] . Also , we identified annexins , which act as glucocorticoid-regulated proteins that promote the resolution of inflammation by limiting neutrophil recruitment and production of proinflammatory mediators and inducing macrophage reprogramming toward an alternative phenotype [81–83] . These effects could explain the type 2 immune response observed in patients with cystic echinococcosis [73] . Other proteins found in our purified EVs were peptidyl-prolyl cis-trans isomerases which show potential immunomodulatory activity , as the alteration in DC function by cytokine production , which leads to expansion of CD4+ Treg cells in Schistosoma mansoni [84] . Also , we found proteins that showed identity with parasite tegumental proteins which could inhibit polymorphonuclear cell chemotaxis and induce IL-4-T lymphocytes and non-complement fixing antibodies ( IgG4 ) in patients with cystic echinococcosis [85] . In this way , due to their ability to suppress the innate and adaptive immune response , EVs could also be valuable tools to improve inflammation-associated disease [24] . On the other hand , and opposed to the immunoregulatory response , certain antigenic proteins described in protoscoleces and metacestodes were detected in E . granulosus EVs . EVs transfer antigens to DCs and T lymphocytes more efficiently than soluble peptides [70] . In our proteomics data , we found 14-3-3 proteins which have been reported to provide 97% protection against E . multilocularis challenge infection in rodent and the production of a specific humoral response in rhesus macaque models , respectively [86 , 87] , the endophilin B1 , which is highly expressed in the tegument of Taeniidae metacestodes and responsible for a strong immune recognition in sera from patients with cystic and alveolar echinococcosis and chronic neurocysticercosis [88] , antigen 5 , an immuno-dominant serine-protease with heparan sulphate proteoglycan- and calcium-binding sites , and antigen P-29 immunologically related to antigen 5 used as a marker for post-surgery surveillance of cystic echinococcosis patients [89 , 90] . Antigen 5 is highly immunogenic in human infections although its highly cross-reactive glycan moieties may involve a parasite evasion mechanism [91–95] . Particularly , this antigen showed greater abundance in the EVs obtained from loperamide-treated parasites than in the controls . This increase correlates with higher positivity rates to antigen 5 test detected in albendazole-treated patients in comparison with untreated patients [96] , probably indicating that the antigen 5 is released not only as a soluble protein but also associated with EVs . In summary , our study showed for the first time the secretion of EVs in E . granulosus protoscoleces and metacestodes obtained in infected mice . Our study also demonstrated conserved size , shape and content of particular excretory/secretory proteins of EVs , suggesting an important role of EVs in the maturation process of DCs which are essential for the coordination of specific immune responses . Nevertheless , additional work is necessary to shed light on the functionality of these DCs stimulated with EVs from E . granulosus in antigen presentation , cytokine release , and activation of T cell population . It will also be interesting to elucidate the components of EVs and establish whether they can be used as diagnostic markers for parasitic diseases , as new vaccines , and/or as treatment tools [97] .	Human cystic echinococcosis , caused by chronic infection with the larval stage of Echinococcus granulosus , affects over 1 million people worldwide . This helminth parasite secretes numerous excretory/secretory products that are in contact with host tissues where it establishes hydatid cysts . In this study , we comprehensively characterized extracellular vesicles ( EVs ) from E . granulosus protoscoleces and metacestodes , and demonstrated for the first time that the exosome-like vesicles from helminths can interact with host dendritic cells and carry several immunoregulatory proteins . This study provides valuable data on cestode-host immune communication . Nevertheless , further research on EVs is needed to fully understand their role in the parasite-host interface and obtain new data concerning their function as therapeutic markers and diagnostic tools .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "invertebrates", "medicine", "and", "health", "sciences", "vesicles", "immune", "cells", "cestodes", "helminths", "antigen-presenting", "cells", "immunology", "parasitic", "diseases", "animals", "dendritic", "cells", "cellular", "structures", "and", "organelles", "animal", "cells", "proteins", "flatworms", "biochemistry", "cytoskeletal", "proteins", "eukaryota", "cell", "biology", "exosomes", "protein", "domains", "biology", "and", "life", "sciences", "cellular", "types", "organisms" ]	2019	Extracellular vesicles from Echinococcus granulosus larval stage: Isolation, characterization and uptake by dendritic cells
Pressure ulcers are costly and life-threatening complications for people with spinal cord injury ( SCI ) . People with SCI also exhibit differential blood flow properties in non-ulcerated skin . We hypothesized that a computer simulation of the pressure ulcer formation process , informed by data regarding skin blood flow and reactive hyperemia in response to pressure , could provide insights into the pathogenesis and effective treatment of post-SCI pressure ulcers . Agent-Based Models ( ABM ) are useful in settings such as pressure ulcers , in which spatial realism is important . Ordinary Differential Equation-based ( ODE ) models are useful when modeling physiological phenomena such as reactive hyperemia . Accordingly , we constructed a hybrid model that combines ODEs related to blood flow along with an ABM of skin injury , inflammation , and ulcer formation . The relationship between pressure and the course of ulcer formation , as well as several other important characteristic patterns of pressure ulcer formation , was demonstrated in this model . The ODE portion of this model was calibrated to data related to blood flow following experimental pressure responses in non-injured human subjects or to data from people with SCI . This model predicted a higher propensity to form ulcers in response to pressure in people with SCI vs . non-injured control subjects , and thus may serve as novel diagnostic platform for post-SCI ulcer formation . In the United States , it is estimated that approximately 250 , 000 people live with spinal cord injury ( SCI ) . Approximately 12 , 000 new cases occur each year [1] , with total direct costs for treating all cases of SCI exceeding $7 billion annually [2] , [3] . Pressure ulcers are common , costly and life-threatening complications for people with SCI . The prevalence of pressure ulcers in people with SCI is estimated to range from 8% to as high as 33% [4] . Post-SCI pressure ulcers are caused by a combination of impaired sensation , reduced mobility , muscle atrophy , as well as reduced vascularity and perfusion [5] . The current consensus is that pressure alone or pressure in combination with shear force cause localized injury to the skin and/or underlying tissue , usually over a bony prominence [6] . Several pathways have been identified for pressure/shear-induced ulceration , the major one being tissue ischemia . Prolonged tissue ischemia may cause inflammation , necrosis , and the eventual formation of a pressure ulcer [7] , [8] . Tissue inflammation is the common physiological reaction caused by tissue ischemia before necrosis occurs . We have focused our attention on this complex biological process . Inflammation is a central , modulating process in many complex diseases ( e . g . sepsis , infectious disease , trauma , and wound healing ) , and is a central driver of the physiology of people with SCI [9]–[12] . However , inflammation is not an inherently detrimental process: properly regulated inflammation is required for successful immune response and wound healing [9] , [13] , [14] . Inflammation is a prototypical complex , nonlinear biological process that has defied reductionist , linear approaches [15]–[18] . Dynamic computational simulations , including ordinary differential equation ( ODE ) - and agent-based models ( ABM ) , have been employed to gain insights into inflammation . These simulations have been useful in integrating mechanistic information and predicting qualitative and quantitative aspects of the inflammatory/wound healing response [19]–[22] . The purpose of the present study was to integrate blood flow data and the process of skin injury , inflammation , and healing using a hybrid model that combines ABM and ODE into a single computational model . Agent-based modeling is an object-oriented , rule-based , discrete-event method of constructing computational models , and this technique can be used to model complex biological systems in which the behavior of individual components/agents , as well as pattern formation and spatial considerations are important [23] . Systems of ODE are well-suited for describing processes ( or physiological responses ) that can be approximated as well-mixed systems [22]–[26] . Modeling with differential equations ( ordinary or partial ) is the most widely used method of mathematical modeling . The main advantage of this approach is that there is a well-developed mathematical theory of differential equations which helps to analyze such equations and in some cases completely solve them [23] , [25] , [26] . To model a complex biological system as an ABM , the system is divided into small computational units ( “agents” ) , with each agent obeying a set of rules that define the behavior of this agent . These simple rules , performed stochastically by agents in the model , lead to a complex , often emergent behavior of the system as a whole . In many cases , agents need only local information on the state of the system , rather than being affected by the global system state . As such , ABM's are particularly well suited to representing the transition between mechanisms at one scale of organization to behavior observed at another . The object-rule emphasis of an ABM greatly simplifies the process of model construction without loss of important features in the system , and also allows for modeling biological processes that are known to have both local and global features [23] . Our primary goal in this study was to gain translationally-useful insights into post-SCI pressure ulcer formation using dynamic , mechanistic computational modeling . However , several issues exist with the use of either ABM's or ODE's alone in modeling the pressure ulcer formation . It is difficult to analyze the output of ABM's in order to derive insights into qualitative regimes or primary drivers of outcome . In addition , simulating ABM's is more computationally intensive than simulating ODE-based models . On the other hand , real-life systems are often too complex to be modeled using only ODE , and the corresponding equation-based models may become too complicated to carry out practically useful results . Hybrid modeling is an emerging technique that involves combining diverse types of computational models into a single simulation [27]–[29] . In this approach , ODE can be used to define certain agent rules ( low-level details ) , and ABM to describe the behavior of the high-level components of our system . In the present study , we utilized ODE to model properties tissue ischemia , and an ABM to model the stochastic , pressure-driven ulcer formation behavior in people with and without SCI . Using this approach , we find that a model calibrated with blood flow data predicted a higher propensity to form ulcers in response to pressure in SCI patients vs . non-injured control subjects . The skin blood flow data used for computing the parameters of the differential equation model were collected from 12 adults ( six with SCI and six without ) . This study was approved by the University of Pittsburgh Institutional Review Board ( IRB# PRO08060015 ) , and was carried out after obtaining informed consent from the participants . The age range of the subjects recruited for this study was 20–50 years old . The actual age in each group was: subjects with spinal cord injury ( 26 , 27 , 35 , 35 , 43 , 48 years old ) ; subjects without any neurological deficits ( 21 , 25 , 29 , 35 , 36 , 44 years old ) . There was no statistically significant difference in age between the two cohorts of subjects ( data not shown ) . For people with SCI , only those with ASIA [30] , a scale for classification of spinal injury , grade A and B , one-year post-injury and non-ambulatory are recruited . The reactive hyperemic response was induced with 60 mmHg of pressure for 20 min on the sacral skin , with the participants lying on their stomach on a mat table . A laser Doppler probe was located at the center of the indenter to collect the skin blood flow . Instrumentation details are published previously [31] . A sample blood flow data collected in the experiment is demonstrated in Figure S1 . The raw blood flow data of all tested subjects are provided in Dataset S1 and the plots of these data are shown in Dataset S2 . The hybrid model utilized in our study is comprised of an ABM of skin/muscle injury , inflammation , and ulcer formation along with an ODE model of blood flow and reactive hyperemia . The ABM portion of the model comprises interactions among oxygen , pro-inflammatory elements , anti-inflammatory elements , and skin damage , with realistic predictions of the pattern , size , and progression of pressure ulcers . All rules of this ABM were generated based on literature reviews and previously-described ABM's of diabetic foot ulcer formation [21] and simplified pressure ulcer formation [32] . The ODE portion of the model simulates the ischemia-induced reactive hyperemic response , and is derived from a previous circuit model [33] . Figure 1 shows the model representation of the pressure ulcer formation . Figures 2A&B depict the model components and their interactions within the hybrid model , with the solid rectangles , ellipses and arrows representing the components of the ABM portion and the dashed ellipse and arrows representing the components of the ODE portion of the model . The ABM portion of the model is based on our previously-developed models [21] , [32] . This ABM is a simplified model that simulates inflammation and reactive hyperemic response ( as the result of applied pressure ) in a small segment of tissue ( epithelial cells in the model ) . We implemented this ABM in SPARK ( Simple Platform for Agent-based Representation of Knowledge; freely downloadable at http://www . pitt . edu/~cirm/spark ) [34] , following an extensive process of literature search and creation of graphical diagrams that incorporate known biological influences [20] , [35] , [36] . From such diagrams and based on our prior work on modeling of the formation of diabetic foot ulcers [21] , we constructed rules by which individual agents ( e . g . cells or cytokines ) interact with each other and bring about biological effects . The ABM portion of the model consists of key cells and diffusible inflammatory signals assumed to be involved in the process of formation of a pressure ulcer . A similarly parsimonious approach was used to construct the rules and relationship among agents , with the goal of generating a high-level view of the process of pressure ulcer formation . The components and inter-relationships among the agents and variables of the pressure ulcer ABM are presented in Figure 2 . Importantly , our model adheres to our prior work on the importance of the positive feedback loop of tissue damage/dysfunction→inflammation→tissue damage/dysfunction [22] , [25] . The main components of the ABM portion of the model are: structural/functional skin cell ( nominally epithelial cells ) ; inflammatory cells ( nominally macrophages ) ; blood vessels; an aggregate pro-inflammatory cytokine agent ( nominally TNF-α ) ; an aggregate anti-inflammatory/pro-healing cytokine ( nominally TGF-β1 ) ; and oxygen . These agents interact according to the following rules . Epithelial cells are damaged by applied pressure . A damaged epithelial cell produces TNF-α . Epithelial cells also are damaged by excessive amount of TNF-α . A severely damaged epithelial cell dies . An epithelial cell can be healed by TGF-β1 , and the healing rate is proportional to the amount of oxygen at the position of the epithelial cell . Macrophages are attracted by TNF-α , and they also produce TNF-α and TGF-β1 . Each macrophage has a fixed lifespan ( measured in simulation steps ) and a macrophage dies after several simulation steps . Blood vessels create new macrophages and release oxygen . The rate of macrophage production and oxygen release depends on the amount of blood flowing through a blood vessel . The ODE portion of the model ( see below ) is incorporated into blood vessel rules , which specify how the oxygen is produced . Blood flow depends on the pressure applied on a blood vessel . A blood vessel dies if the surrounding epithelial cells die . There are also global model rules which specify how oxygen , TNF-α , and TGF-β1 diffuse and evaporate . Physical pressure in ABM portion of the model is applied periodically . More specifically , the pressure is applied for a fixed period of time . The pressure is then released for the same amount of time , and the process repeats . A specific model parameter ( called Pressure Interval ) specifies the pressure time interval . A detailed description of ABM rules and parameters is given in Text S1 . Ischemia-induced hyperemia ( the reactive hyperemic response ) is a sudden increase in skin blood flow following tissue ischemia [37] . Hyperemia is a normal physiological response that can be easily induced with non-damaging ischemic events , and it has been used in numerous fields to examine endothelial function [38] and vascular activity [39] . We incorporated an ODE model of reactive hyperemia into the pre-existing ABM of ulceration in order to link measurable parameters of reactive hyperemia to the process of ulceration induced by repeated cycles of pressure and ischemia/reperfusion . To do so , we adopted the ODE-based circuit model of de Mul et al [40] . These authors suggested that the reactive hyperemic response could be modeled as the circuit shown at Figure 3 , with R ( resistance ) representing vascular resistance , C ( capacitance ) representing vessel compliance , V ( t ) representing the input blood flow pressure , and I ( current ) representing blood flow . I2 ( t ) represents the skin blood flow ( specifically , reactive hyperemia ) as measured using a laser Doppler flowmetry system . The ODE system derived from the circuit model has the following formNote that here we have only two differential equations for I1 ( t ) and I2 ( t ) . I3 ( t ) , I4 ( t ) , I5 ( t ) , and I6 ( t ) can be algebraically eliminated . We are interested in modeling a situation when an occlusion occurs in the input blood flow due to application of an external pressure . De Mul et al [40] model such a situation by considering the following stepwise input blood flow functionHere V0 is the aortic pressure . Based on this expression of V ( t ) , an explicit solution for I2 ( t ) can be derived with initial conditions I1 ( 0 ) = I2 ( 0 ) = 0 . This solution has the following formHere I2 , rest , a , b , p1 , and p2 are constants expressed in terms of R1 , R2 , R3 , R4 , C1 , C2 , V0 . We used this explicit solution for I2 ( t ) for finding parameter values of the circuit model ( the ODE portion of the model ) based on available blood flow experimental data . In our agent-based simulations , the input blood pressure was a periodic function . In order to obtain the blood flow in these simulations , we used the ODE explicitly in our ABM . The main components of SPARK models are Space , Data Layers , Agents , and the Observer [34] . Space is analogous to the physical space , and provides a context within which the model evolves . Data Layers provide a convenient way of tracking variables in space . Data layers update in time simultaneously at all positions . This is a computationally efficient way of handling processes such as diffusion and evaporation without employing an agent at each position to carry out the calculation . Agents can move , perform functions , interact with each other , and also interact with the space they occupy . Each agent has a set of behaviors and rules of action . The Observer contains information about space and all agents in the model . We extended SPARK with a feature for simple incorporation of ODE into an ABM . Epithelial cells , blood vessels , and macrophages were implemented as agents in SPARK . Oxygen , TNF-α , and TGF-β1 were implemented as data layers in SPARK . Pressure was implemented as a global model variable that periodically changes during the model simulation process . The ODE portion of the model is integrated into the code of blood vessel agents . The following example shows how ODE's were added into SPARK-PL code: equations [ I4 = ( V - R1 * I1 ) /R4 I3 = ( V - R1 * I1 - R2 * I2 ) /R3 I5 = I1 - I2 - I4 I6 = I2 - I3 Dt I1 = ( dV - I5/C1 ) /R1 Dt I2 = ( I5/C1 - I6/C2 ) /R2 ] All variables in the example above are local variables of a blood vessel agent . Equations describe the evaluation of these variables in time . Each time step , the equation is integrated on the interval [t1 , t1+dt] , where t1 is the current simulation time and dt is the global parameter which specifies the time step . The output values of the equations are used in other rules defined for a blood vessel agent . V represents the input blood pressure which is a periodic function in our simulations which depends on three parameters:Here , Vmax and Vmin represent maximal and minimal blood pressures respectively; Tp is the pressure interval parameter of the model; k = 0 , 1 , 2 , etc; t is the number of simulation ticks . In other words , we set V = Vmin when the external pressure is applied and V = Vmax when the external pressure is released . The SPARK source codes of this hybrid model are provided in Dataset S3 . The ODE-based portion of the model was fit to data on blood flow for two different groups of subjects: a control group ( CTRL ) and an SCI group , as follows . We initially fixed parameters of the agent-based portion of the model . We chose these parameters based on a literature search . Only the approximate scale of parameters could be selected in this fashion , since our ABM is a simple , lumped-parameter model . With this set of parameter values , the ABM produces qualitative behavior commensurate with normal inflammation and wound healing [21] . Raw blood flow data was filtered with low pass filters . The filtered data were averaged over all six subjects in each group . Figures 4A and 4B depict the averaged reactive hyperemia blood flow data in people with and without SCI , respectively . We note that Figure 4A tend to oscillate more than Figure 4B . Depending on the level , and severity of injury , the reactive hyperemic response as measured with skin blood flow varied in people with SCI as compared to people without any neurological deficits . One main difference was the rate of increase and decrease in the skin blood flow of the reactive hyperemic response [41] , in other words , one subject's peak blood flow may occur at 0 . 5 minute , and the other one may occur at 2 . 0 minute . With this variation , the blood flow oscillates more in Figures 4A as compared to Figures 4B . Another possible explanation is that , the skin blood flow as measured with the laser Doppler flowmetry system does oscillate naturally . When the skin blood flow signal was computed with Fourier transform , previous studies have identified that different frequency bands represent different physiological control mechanism of the blood flow [42] . Therefore the oscillation of skin blood flow is inevitable . We also note that the data in our simulation focused on the first 4 minutes . The interesting portion of the experimental data is the time when the peak blood flow occurs . We obtained approximately 10 minutes of raw data after releasing the pressure . The important information includes the time of the peak and the rate of decrease after the peak; both these values can be extracted from first 4 minutes after the pressure is released for all recorded data . We believe that it is simpler and more reliable to fit the ODE parameters based on the most important part of the experimental data ( i . e . the first 4 minutes ) , since the rest of the data do not contain any important information for model fitting . We then calibrated the ODE portion of the model based on the averaged data . Calibration was done using the following error function which measures the distance between actual ( averaged ) data and simulated results:Here i is the group index , i . e . , i is either CTRL or SCI . Ei ( p ) is the error for the i-th group; y ( p , k ) is the value of the model function evaluated at the point k with the parameter vector p . Mi ( k ) is the averaged i-th data at the point k . Calibration was performed using Matlab R2011 ( The Mathworks , Inc . , Natick , MA , USA ) . We used the explicit expression of I2 ( t ) for finding best-fit parameters . The values of Vmax were assumed to be 85 mmHg for the control group and 75 mmHg for the SCI group , the same pressure values as in the experiments . For all other parameters , we defined possible lower and upper bounds . For the control group , we set 200 as the upper bound of all parameters , and 0 . 01 as the lower bound for all parameters except R4 , for which we chose 190 as the lower bound since it is assumed that R4>>R1 , R2 , R3 [40] . Then we randomly selected 1000 points in the space of parameters and ran the standard Matlab minimization function fminsearch for all these initial points , and picked the best fit results . The search of best-fit parameters for the SCI group was carried out in a similar way . The only differences were that the value of Vmax = 75 , and in addition we changed the upper bounds of C1 and C2 and set them equal to the best-fit values of C1 and C2 for the control group . This change was made to reflect the fact that C1 , 2SCI<C1 , 2CTRL [43] . Figures 5A and 5B show the best-fit simulation results , which minimize the error function Ei ( p ) in data from people with and without SCI , respectively . Table 1 lists the values of the best-fit parameters for both group with the ratios calculated in the Figure 6 to show the significant change of parameters for people with and without SCI . The results show that vascular resistance ( R1 ) is significantly increased and that blood vessel compliance ( C1 , C2 ) is decreased in the SCI group by comparing with the control group . We next sought to determine the behavior of our simulation under a more clinically realistic setting , in which pressure to tissues alternates with periods of pressure relief . We also sought to determine if , once partially calibrated with blood flow data from control vs . SCI subjects , our model would predict differential propensity to ulcerate between these two groups of patients . We simulated the application of medium-scale pressure on the skin with different frequencies , first applying a pressure on the skin for a given period of time ( pressure interval ) , releasing the pressure for the same amount of time , and then repeating the process . Using the parameters obtained as described above , we ran the model simulations for both groups and compared the outcome . We ran the model for 2000 steps with various values of the pressure interval parameter . All other ABM parameters were fixed . We assumed Vmax = V0 ( i . e . , Vmax = 85 for the control group and Vmax = 75 for the SCI group ) and Vmin = 40 for both groups . We initially examined the minimal value of the pressure interval that would be predicted to result in substantial tissue damage ( death of some epithelial cell agents ) . Figures 7A and 7B show the SPARK simulation results for control and SCI subjects . Green squares represent healthy epithelial cells , red squares represent damaged epithelial cells , red circles represent blood vessels , and blue circles represent macrophage . For the control group , the minimal value of the pressure interval was 205–210 simulation ticks ( Figure 7A ) ; in contrast , for the SCI group , the minimal value was 105–110 simulation ticks ( Figure 7B ) . We also performed subject-specific fitting of the ODE parameters and measured the minimal value of the pressure interval resulting in substantial tissue damage for each subject . The results are given in Table 2 . The average subject-specific value of the minimal pressure interval was 207 for control subjects and 168 for SCI subjects . These results agree qualitatively with our findings for the averaged data presented above: the minimal pressure interval is larger for the control group . We next examined the predicted effect of turning frequency on control and SCI subjects . Figures 8A and 8B show how the predicted health of epithelial cells progresses over time for simulations of the control and SCI groups , respectively , over varying pressure on/off cycles . Increasing the frequency ( or applying pressure for a short period of time and then subsequently relieving this pressure ) , we obtained an outcome in which a pressure ulcer did not form: when the simulated pressure is applied , the tissue is damaged somewhat , but when the pressure is relieved tissue health is restored . Also , simulated damage/dysfunction was predicted to increase more rapidly in the SCI group vs . the control group when the pressure interval was increased . The components of the inflammatory response are time-driven , highly interconnected , and interact in a nonlinear fashion [15]–[18] , [44] . The systems biology community has integrated mathematical and simulation technologies to understand complex biological processes [45] . More recently , we have suggested translational systems biology as a framework in which computational simulations are designed to facilitate in silico clinical trials , simulations are appropriate for in vivo and specifically clinical validation , and mechanistic simulations of whole-organism responses could guide rational therapeutic approaches [25] . Agent-based models have emerged as a useful complement to ODE-based models for elucidating complex biological systems , including inflammation , wound healing , angiogenesis , and cancer [19] , [21] , [23] , [36] , [46]–[49] . In the present study , we utilized a hybrid modeling approach that combines the both features of ODE and agent-based models . Using this approach , we integrate data regarding blood flow properties in SCI patients and compare them to data from control subjects . Our analysis suggests that , based on an abstraction of these blood flow properties and a stochastic model of tissue inflammation and ulcer formation , and in agreement with the literature [50] , SCI patients are predicted to be more prone to ulceration . Our study , along with prior work [28] , [51] , [52] , suggests that such hybrid modeling methodology could have a wide application in modeling complex , multiscale biological systems . Despite the lack of sensation and motor function after SCI , several physiological changes at the chronic stage of SCI ( more than 12 months since injury ) increase a person's susceptibility to develop pressure ulcers , including changes in body composition ( increased proportion of fatty tissue ) and vascularity [5] . The linkage between changes in vascularity , epithelial function and pressure ulcer formation in people with SCI is not fully explored . Therefore , this pilot hybrid model was aimed at simulating pressure ulcer development by including a key vascular response ( reactive hyperemia ) observed in human subjects . The goal of our previous research was to find the optimal turning frequency for patients with SCI [32] . The goal of the present model is the improvement of our previous model by coupling an ODE model of the reactive hyperemic response observed experimentally to an ABM based on rules derived from the literature . This model was capable of simulating the intensity in epithelial cell damage as a function of changes of amount and duration of localized pressure on the skin of people with and without SCI . Results from the best-fit parameters of the circuit model set showed differences in vascular resistance ( R1 ) and blood vessel compliance ( C1 , C2 ) between the two groups . The arterial resistance was bigger while the capillary resistance was smaller , respectively , in subjects with SCI as compared to controls . Changes in vascularity in people with SCI may be caused by denervation of sympathetic nervous system [53] as well as physical inactivity [54] . Our finding of increased vascular resistance in the arterial system was consistent with previous studies . With the loss of supraspinal control of the vascular system after high level of injury , people with SCI were reported to have increased vascular resistance in order to maintain the vascular tone by compensating for the loss of supraspinal sympathetic control [55] . Additionally , the increased vascular resistance may result from preservation of α-adrenergic tone . The increased vascular resistance could also result from vascular adaptation to deconditioning with the loss of motor function [56] . One prior study found that there was an increased activation of the receptor of the endothelin-1 , which increases the vascular tone [56] . The results of decreased vascular resistance in the capillary system were not consistent with observations regarding vascular resistance in the arterial system . The capillary resistance was not investigated in previous studies; thus , our findings regarding vascular resistance in the arterial system may not be generalized to the capillary resistance , since the vascular resistance was measured with venous occlusion plethysmography in previous studies and the measurement was not directly on capillary blood flow . In addition , the measurement of reactive hyperemia in our study was at the lower back using an indenter , whereas the aforementioned previous studies measured this response at lower limbs with cuff . Future study on structural changes in capillary system and vascularity of the microcirculation might be beneficial in understanding the linkage to ulceration . Results from the analysis of the best-fit parameters of the circuit model set also showed that the vessel compliance is smaller in people with SCI as compared to the controls . De Groot et al . found that the femoral artery compliance is smaller in individuals with SCI [43] , and they suggested that this physiological change may be due to inactivity of the muscle since arterial compliance could be enhanced with functional electrical stimulation . Our model validation studies suggest that the minimal amount of repeated pressure required to cause endothelial cell damage would be smaller in subjects with SCI . People with SCI are susceptible to ulcer formation , and there are several physiological changes that may contribute to the susceptibility of pressure ulcer development in this population . For example , people with complete SCI had decreased cross-sectional area of muscle fibers [57] and increased fat mass in lower limbs [58] . A recent study from Linder-Ganz et al . directly pointed out the relationship between physiological changes after injury and the pressure ulcer formation by using finite element model . They found that with the use of the same seat cushion , people with SCI had greater deep muscle stress as compared to controls [59] . To date , there is no study that investigated the direct linkage between changes in vascularity and ulcer formation in people with SCI . We were not aware of the underlying mechanism of changes in vascularity and the ulcer formation . However , from the rules and results of our model , it is indicated that changes in vascularity may play a role in decreased tolerance of pressure and endothelial function that leads to more severe damage with the same amount and duration of pressure . There are several limitations of this study . This study only recruited limited numbers of subjects ( six CTRL and six SCI ) , and people with SCI and controls were not matched for comparison . If additional subjects were used for the model calibration , the conclusion could be reached at a higher level degree of confidence . Though the ages of the subjects in the cohorts were not identical , there was no statistically significant difference with regard to age between the two groups of patients . In addition , previous studies [60] , [61] found that the reactive hyperemic response was not different between healthy elderly population and healthy adults; these authors only found an impaired reactive hyperemic response among individuals in a hospitalized elderly population . Since there was no statistically significant difference in age between non-injured and SCI-injured subjects in our studies , and since all subjects recruited in our studies were healthy and not hospitalized during the time of the study , age is unlikely to be a significant factor in our data analysis . This is a pilot study developing this hybrid model of ulcer formation with different input of people with and without SCI . For a more realistic simulation , the ABM portion of the model could be expanded by incorporating additional physical and biological components , such as shear force and reperfusion injury , which may contribute to the formation of the pressure ulcer . Nevertheless , in this work , we present a first attempt to construct a biological model in a single computational platform where mathematical and agent-based models work in a seamless manner , and the result of the model reveals useful insight into the ulceration in people with and without SCI . In conclusion , we used a hybrid approach combining ordinary differential equations related to blood flow along with an agent-based model of skin injury and subsequent inflammation in a single modeling platform , in order to investigate pathogenesis difference between people with SCI and without SCI in the process of ulcer formation . Our current finding suggests that people with SCI have higher propensity to form ulcers in response to pressure than non-injured control subjects .	Pressure ulcers are costly and life-threatening complications for people with spinal cord injury ( SCI ) . To gain insight into the pathogenesis and effective treatment of post-SCI pressure ulcers , we constructed a computer simulation in a hybrid modeling platform which combines both equation- and agent-based models . The model was calibrated using skin blood flow data and reactive hyperemia in response to pressure and predicted a higher propensity to form ulcers in response to pressure in people with SCI vs . non-injured control subjects . The methodology we present in the paper may eventually be used as a novel platform to study post-SCI ulcer formation , as well as serving as a framework for other biological contexts in which agent-based models and mathematical equations can be integrated .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "systems", "biology", "biochemical", "simulations", "computer", "science", "computer", "modeling", "immunology", "biology", "computational", "biology", "computerized", "simulations", "immune", "response", "immune", "system" ]	2013	Hybrid Equation/Agent-Based Model of Ischemia-Induced Hyperemia and Pressure Ulcer Formation Predicts Greater Propensity to Ulcerate in Subjects with Spinal Cord Injury
Density-Enhanced Phosphatase-1 ( DEP-1 ) de-phosphorylates various growth factor receptors and adhesion proteins to regulate cell proliferation , adhesion and migration . Moreover , dep-1/scc1 mutations have been detected in various types of human cancers , indicating a broad tumor suppressor activity . During C . elegans development , DEP-1 mediates binary cell fate decisions by negatively regulating EGFR signaling . Using a substrate-trapping DEP-1 mutant in a proteomics approach , we have identified the C . elegans β-integrin subunit PAT-3 as a specific DEP-1 substrate . DEP-1 selectively de-phosphorylates tyrosine 792 in the membrane-proximal NPXY motif to promote integrin activation via talin recruitment . The non-phosphorylatable β-integrin mutant pat-3 ( Y792F ) partially suppresses the hyperactive EGFR signaling phenotype caused by loss of dep-1 function . Thus , DEP-1 attenuates EGFR signaling in part by de-phosphorylating Y792 in the β-integrin cytoplasmic tail , besides the direct de-phosphorylation of the EGFR . Furthermore , in vivo FRAP analysis indicates that the αβ-integrin/talin complex attenuates EGFR signaling by restricting receptor mobility on the basolateral plasma membrane . We propose that DEP-1 regulates EGFR signaling via two parallel mechanisms , by direct receptor de-phosphorylation and by restricting receptor mobility through αβ-integrin activation . Protein phosphorylation is one of the most common post-translational modifications used by eukaryotic cells to regulate various aspects of protein function . Signaling through the conserved Epidermal Growth Factor Receptor ( EGFR ) pathway involves receptor auto-phosphorylation as well as the phosphorylation of several downstream signal transduction molecules once an EGF ligand has bound to and activated the receptor tyrosine kinase [1] . On the other hand , protein phosphatases are key components of inhibitory networks that attenuate EGFR signaling before and after ligand binding [2] . The genomes of vertebrates and invertebrates encode a large number of predicted phosphatase genes , of which many are implicated in human diseases [3] . However , the physiological substrates of most protein phosphatases are not well defined , and it is often difficult to correlate a specific phosphatase activity with in vivo changes in protein phosphorylation [2] . The Density Enhanced Phosphatase DEP-1 , also known as PTPRJ , PTP-η or CD148 , belongs to the class III Receptor Protein Tyrosine Phosphatase ( R-PTP ) family [4 , 5] . Like the other R-PTPs of this family , DEP-1 contains a single intracellular catalytic tyrosine phosphatase domain , a transmembrane domain , and multiple extracellular fibronectin type III repeats ( Fig 1B ) . DEP-1 was originally isolated as a phosphatase whose expression is up-regulated in contact-inhibited , dense cell cultures [5] . The mouse dep-1 gene was independently identified as the colon cancer susceptibility locus Scc1 , and human dep-1 is frequently deleted or mutated in various cancer types , such as thyroid , colon , lung , pancreatic , and breast cancer [2 , 6] . DEP-1 inhibits cell motility in contact-inhibited cell cultures and exhibits tumor-suppressor activity when overexpressed in cancer cells [7–10] . A multitude of potential DEP-1 substrates have been identified , including various growth factor receptors such as EGFR , PDGFR , VEGFR , FLT-3 , MET , the ERK-2 kinase , the p85 subunit of PI3K as well as cell-cell junction proteins like p120ctn , β-catenin and γ-catenin , occludin and ZO-1 [4 , 11 , 12] . We have previously identified the C . elegans dep-1 ortholog in a forward genetic screen for negative regulators of the EGFR/RAS/MAPK signaling pathway [11] . DEP-1 controls cell fate decisions by inhibiting the activation of the EGF receptor ( termed LET-23 in C . elegans ) in different tissues . Most prominently , dep-1 is required for a binary cell fate switch during vulval development . Towards the end of the second larval stage , the uterine anchor cell ( AC ) secretes the EGF-like growth factor LIN-3 to induce the differentiation of the adjacent vulval precursor cells ( VPCs ) P3 . p to P8 . p [13] . All six VPCs express LET-23 EGFR on their basolateral membrane , but the VPC that is located closest to the AC ( P6 . p ) sequesters most of the LIN-3 EGF signal and responds by adopting the primary ( 1° ) vulval cell fate ( Fig 1A ) . P6 . p then produces a lateral DELTA signal that activates the NOTCH signaling pathway in the adjacent VPCs P5 . p and P7 . p . NOTCH signaling in these VPCs induces the expression of several inhibitors of the EGFR/RAS/MAPK pathway , such as the MAPK phosphatase LIP-1 , to inhibit the 1° and induce the alternate secondary ( 2° ) cell fate [14–16] . In this context , DEP-1 is part of a negative feedback loop , in which EGFR signaling down-regulates DEP-1 expression in the 1° VPC , while NOTCH signaling maintains high DEP-1 expression in the 2° VPCs ( Fig 1A ) . As a consequence of this signaling cross-talk , the EGFR remains fully active in the 1° VPC , but is repressed by DEP-1 in the 2° VPCs . Genetic epistasis experiments indicated that DEP-1 inhibits signaling at the level of the EGFR and biochemical studies indicated that the EGFR is a DEP-1 substrate [11] . To explore new functions of DEP-1 in C . elegans , we searched for additional DEP-1 substrates through a proteomics approach and thereby identified the β-integrin subunit PAT-3 . Integrins are type I transmembrane glycoproteins that function as cell adhesion molecules and form heterodimers consisting of one α and one β subunit . The C . elegans genome encodes a single β-subunit , PAT-3 , and two α-subunits , PAT-2 and INA-2 ( INtegrin Alpha ) . pat-2 or pat-3 loss-of-function mutations cause a “Paralyzed and Arrested at Two-fold” embryonic lethal ( PAT ) phenotype because PAT-2 and PAT-3 are essential to attach the body wall muscles to the overlaying epidermis during elongation of the embryo [17] . Mutations in ina-1 , on the other hand , cause larval lethality with pleiotropic defects , as the INA-1/PAT-3 complex is required for the correct migration of various cell types [18] . Integrins play important roles in transducing extra- and intra-cellular signals in either direction , in what is referred to as either “inside-out” or “outside-in” signaling [19–21] . In their active , open conformation , integrins bind to a variety of cell-surface , extracellular matrix ( ECM ) or soluble protein ligands . The phosphorylation status of the β-integrin cytoplasmic tail controls the recruitment of different cytoplasmic adaptor proteins to focal adhesion sites . In particular , the cytoplasmic FERM domain protein talin , called TLN-1 in C . elegans , binds preferentially to the un-phosphorylated membrane-proximal NPXY792 motif in activated β-integrins , thereby connecting the F-actin cytoskeleton to the αβ-integrin complex [22] . The membrane-distal NPXY motif on the other hand , binds to kindlin and regulates intracellular integrin trafficking [19 , 21 , 22] . The C . elegans integrins have so far not been implicated in the regulation of the EGFR/RAS/MAPK signaling pathway . Though , there are several reports of cross-talk between integrins and the EGFR signaling pathway in mammalian cells [23] . In order to identify additional substrates of C . elegans DEP-1 , we undertook a mass spectrometry-based approach , in which proteins enriched through pull-down experiments were analyzed by LC-MS/MS . For this purpose , the intracellular domain of DEP-1 carrying the substrate trapping mutation D1241A ( DEP-1intra-DA , Fig 1B ) [11] was expressed as a GST fusion protein in E . coli , affinity-purified on glutathion-sepharose and incubated with total protein extracts of mixed-stage C . elegans wild-type N2 animals ( Fig 1C and 1D and materials and methods ) . To distinguish between DEP-1 substrates that specifically bound to the substrate-trapping mutant and other DEP-1 binding proteins , we performed parallel pull-down experiments with the wild-type intracellular domain ( DEP-1intra-wt ) and with GST alone as negative control ( Fig 1B and 1D ) . Protein complexes were separated by SDS-PAGE and stained with colloidal Coomassie-blue . We identified three prominent protein bands besides the DEP-1intra-DA bait with molecular weights of around 90 kD , 130 kD , and 170 kD that were present exclusively in the DEP-1intra-DA pull-down ( Fig 1E ) . After fractionation of the SDS-gel ( dotted lines in Fig 1E ) , proteins were processed for in-gel tryptic digestion and C18 ZipTip purification . Liquid chromatography and tandem mass spectrometry ( LTQ Orbitrap LC-MS/MS ) in conjunction with a database search against a C . elegans proteome led to the identification of 585 proteins in three independent pull-down experiments . Ninety-seven of the identified proteins were not present in the GST negative control ( Fig 1F , see S1 Table for the full list ) . Among the 50 proteins ( represented by 404 peptides ) that bound preferentially to DEP-1intra-DA , 99 peptides were assigned to the α-integrin subunit PAT-2 ( MW = 135 . 9 kD ) and 54 peptides to the β-integrin subunit PAT-3 ( MW = 90 . 1 kD ) ( Fig 1F ) . Furthermore , 41 peptides were from NID-1 ( MW = 174 . 4 kD ) , an ECM component that interacts with the extracellular domain of integrins [24] . Since PAT-2 and PAT-3 peptides were found in each of the three replicate experiments using DEP-1intra-DA , the α-integrin and the β-integrin are the strong candidates for DEP-1 substrates . Although we have previously reported that DEP-1intra-DA specifically interacts with the EGFR homolog LET-23 [11] , LET-23 was not present among the 50 proteins that preferentially bound to DEP-1intra-DA , presumably because LET-23 protein abundance in total animal extracts is too low to be detected in this assay ( bottom 5% of all proteins ) [25] . Tyrosine phosphorylation of the intracellular domain of the β-integrin subunits is known to regulate integrin activity [20 , 21] , but tyrosine phosphorylation of the α subunits has so far not been reported . We thus hypothesized that DEP-1 de-phosphorylates PAT-3 in a complex with PAT-2 . To further characterize the binding of DEP-1intra-DA to PAT-3 and PAT-2 , we performed pull-down experiments using purified DEP-1intra proteins and total protein extracts from worms expressing functional PAT-3::GFP or PAT-2::GFP transgenes [26 , 27] . Western blotting with anti-GFP antibodies revealed a specific interaction of DEP-1intra-DA with PAT-3::GFP , whereas no binding of DEP-1intra-wt to PAT-3 was detected ( Fig 2A , top ) . Similar results were obtained with total protein extract from PAT-2::GFP expressing worms . PAT-2::GFP did interact with DEP-1intra-DA , though a weaker interaction was also observed with DEP-1intra-wt ( Fig 2A , bottom ) . To test whether the interaction between PAT-3 and DEP-1 depends on the catalytic phosphatase domain , we added the PTP inhibitor sodium orthovanadate ( Na3VO4 ) to the protein extracts before performing the binding experiment [28] . The interaction of DEP-1intra-DA with PAT-3::GFP was abolished in the presence of 5 mM Na3VO4 ( Fig 2B ) . Since only the substrate-trapping DA mutant of DEP-1 interacted with PAT-3 and this interaction was blocked by orthovanadate treatment , we conclude that PAT-3 is most likely a direct substrate of DEP-1 . The phosphatase domain of DEP-1 is located in the cytoplasmic region , indicating that DEP-1 de-phosphorylates a phospho-tyrosine residue in the cytoplasmic tail of PAT-3 . The PAT-3 intracellular domain contains three conserved tyrosine residues , Y772 , Y792 , and Y804 ( Fig 2C ) . Y792 and Y804 are part of two conserved NPXY motifs , whose tyrosine phosphorylation regulates integrin activation and the recruitment of cytoplasmic adaptor proteins to the plasma membrane [17 , 21 , 22] . However , the protein purification and LC-MS/MS conditions we used did not allow the detection of phospho-peptides . To determine , which of these tyrosine residues are recognized by DEP-1 , we generated mutant pat-3::gfp transgenes , in which the cytoplasmic tyrosines Y772 , Y792 , Y804 , as well as the conserved threonine triplet TTT796-798 located between Y792 and Y804 were replaced with alanines ( Y772A , Y792A , Y792A Y804A , Y804A , and TTT796-798AAA ) . The interaction of DEP-1intra-DA with PAT-3 containing either the Y792A mutation in the membrane-proximal NPXY motif alone or in combination with the Y804A mutation in the membrane-distal NPXY motif was significantly reduced ( Fig 2D and 2E ) . By contrast , the Y804A single mutation did not affect the binding of DEP-1intra-DA to PAT-3 . Furthermore , the TTT796-798AAA mutation reduced the interaction of PAT-3 with DEP-1intra-DA , possibly because this triple mutation in the vicinity of the NPXY792 motif prevents substrate recognition by DEP-1 . Taken together , the protein interaction experiments suggest that DEP-1 selectively de-phosphorylates tyrosine Y792 in the membrane-proximal NPXY motif in the cytoplasmic tail of PAT-3 . We have previously identified dep-1 as a negative regulator of vulval induction [11] . Epistasis analysis indicated that DEP-1 negatively regulates EGFR/RAS/MAPK signaling by inhibiting LET-23 EGFR in the vulval precursor cells ( VPCs ) . We thus investigated whether the de-phosphorylation of PAT-3 β-integrin by DEP-1 mediates the inhibitory activity DEP-1 exerts on EGFR signaling . For this purpose , we used CRISPR/Cas9-mediated genome editing to create a non-phosphorylatable pat-3 ( Y792F ) allele , in which the Y792 phosphorylation site was replaced with Phenylalanine ( F ) ( pat-3 ( zh105 ) , see materials and methods ) . To exclude potential off-target effects , we sequenced all predicted off-target sites with a score of at least 0 . 2 [29] , but found no additional mutations at these sites ( S2 Table ) . Furthermore , the pat-3 ( zh105 ) allele was backcrossed four times to the wild-type N2 strain to remove unlinked mutations before analysis . pat-3 ( Y792F ) single mutants are viable and exhibit a superficially wild-type morphology with normal levels of vulval induction ( Fig 3A and 3B ) . dep-1 ( lf ) single mutants also develop a wild-type vulva due to the redundancy between the various negative regulators of the EGFR/RAS/MAPK pathway ( Fig 4A ) [11] . However , combined with loss-of-function ( lf ) mutations in other negative regulators of EGFR/RAS/MAPK signaling , such as a lf mutation in the MAP kinase phosphatase lip-1 , dep-1 ( lf ) causes a frequent transformation of the 2° cell fate of the P5 . p and P7 . p descendants into a 1°-like fate , a so called Adjacent Primary Fate ( Apf ) phenotype ( Fig 3C and 3E ) [11] . The dep-1 ( lf ) ; lip-1 ( lf ) Apf phenotype is characterized by the detachment of the P5 . p and P7 . p descendants from the cuticle and an enlarged diameter of the vulval lumen ( Fig 3C and 3F ) . The pat-3 ( Y792F ) allele partially suppressed the Apf phenotype of dep-1 ( lf ) ; lip-1 ( lf ) double mutants ( Fig 3D ) . In particular , the frequency of animals , in which the P5 . p or P7 . p descendants were transformed into a 1°-like fate , was significantly decreased by the pat-3 ( Y792F ) allele ( Fig 3E ) . The suppression of the Apf phenotype was further reflected by a reduced vulval lumen diameter in dep-1 ( lf ) ; pat-3 ( Y792F ) ; lip-1 ( lf ) triple mutants compared to the dep-1 ( lf ) ; lip-1 ( lf ) double mutants ( Fig 3F ) . We next investigated the genetic interaction between pat-3 ( Y792F ) and mutations that either increase or decrease the activity of the EGFR/RAS/MAPK pathway . First , we analyzed a gain-of-function mutation in the let-60 gene , which encodes the sole C . elegans RAS family member [30] . let-60 ras ( gf ) single mutants contain on average 3 . 8 induced VPCs per animal ( i . e . the Vulval induction Index VI = 3 . 8 ) and exhibit a penetrant Multivulva ( Muv ) phenotype due to the hyper-activation of the downstream MAPK pathway ( Fig 3G ) . However , the RAS/MAPK pathway in the let-60 ras ( gf ) background remains sensitive to the upstream signal from the EGFR and to lateral inhibition by NOTCH [31] . Hence , let-60 ( gf ) animals only rarely exhibit an Apf phenotype ( Fig 3G ) . By contrast , dep-1 ( lf ) ; let-60 ( gf ) double mutants exhibit a slightly enhanced Muv ( VI = 4 . 1 ) and a penetrant Apf phenotype ( Fig 3G ) [11] . The VI as well as the frequency of the Apf and Muv phenotypes were decreased in dep-1 ( lf ) ; pat-3 ( Y792F ) ; let-60 ( gf ) triple mutants , pointing at an inhibitory effect of the pat-3 ( Y792F ) allele on EGFR/RAS/MAPK signaling . The lin-7 gene encodes a PDZ domain adaptor protein that is required for basolateral localization of LET-23 and efficient receptor activation by the inductive LIN-3 EGF signal [32] . lin-7 ( lf ) mutants exhibit a partial vulvaless ( Vul ) phenotype that is efficiently suppressed by dep-1 ( lf ) ( Fig 3H ) . Consistent with an inhibitory effect of PAT-3 on the EGFR signaling pathway , the introduction of the pat-3 ( Y792F ) mutation into the dep-1 ( lf ) lin-7 ( lf ) background caused a slightly decreased VI ( statistically not significant ) and an increased penetrance of the Vul phenotype ( Fig 3H ) . Thus , the pat-3 ( Y792F ) allele partially suppresses the effects of dep-1 ( lf ) on mutations that either increase or decrease the activity of the EGFR signaling pathway . These results indicate that DEP-1 attenuates EGFR signaling in part by de-phosphorylating the Y792 residue in the PAT-3 β-integrin subunit . Since the pat-3 ( Y792F ) mutation only partially reverted the dep-1 ( lf ) phenotype , DEP-1 must inhibit EGFR signaling through additional , PAT-3 independent mechanisms . One alternate mechanism of DEP-1 action is probably be the direct de-phosphorylation of the EGFR [11 , 12] . We next examined the effects of reduced αβ-integrin activity on EGFR/RAS/MAPK signaling . Since constitutive pat-2 ( lf ) or pat-3 ( lf ) mutations cause embryonic or early larval lethality , we first expressed a dominant-negative pat-3 mutant in the VPCs under the control of the bar-1 promoter ( Pbar-1::pat-3 ( dn ) ) [27 , 33] . In the sensitized let-60 ras ( gf ) background , the Pbar-1::pat-3 ( dn ) transgene caused an increase in vulval induction and a similar Apf phenotype as observed in dep-1 ( lf ) ; let-60 ( gf ) mutants ( Fig 4A–4D and 4I ) . In addition , we used the tissue-specific RNAi strain let-60 ( n1046gf ) ; rde-1 ( ne219lf ) ; zhEx418[lin-31::rde-1] to reduce pat-2 and pat-3 activity specifically in the Pn . p cells , which include the VPCs [34] . Pn . p cell-specific RNAi against both , pat-2 and pat-3 , increased the vulval induction in the let-60 ( gf ) background when compared to vector control animals ( Fig 4E–4G and 4J ) . ( Note that the overall reduced VI in the Pn . p cell-specific RNAi strain is due to the rde-1 ( lf ) background [34] ) . Moreover , the pat-2 and pat-3 RNAi treated L4 larvae showed a similar Apf phenotype as observed in dep-1 ( lf ) ; let-60 ( gf ) double mutants ( Fig 4C ) . By contrast , RNAi of the second α-integrin subunit ina-1 did not significantly change vulval induction in the let-60 ( gf ) background ( Fig 4J ) . The conserved NPXY motifs in the β-integrin C-terminal tail serve as docking sites for adaptor proteins that mediate various cellular effects of the integrins . During integrin activation , the NPXY motifs in the β-integrin cytoplasmic tail are de-phosphorylated , thereby permitting the recruitment of the talin and kindlin adaptor proteins from the cytoplasm to the plasma membrane [22] . Talin binds specifically to the membrane-proximal NPXY792 motif of β-integrins and links the F-actin cytoskeleton to focal adhesion sites , while kindlin binds to the membrane distal NPXY motif that controls integrin trafficking . We therefore examined whether the single C . elegans talin ortholog TLN-1 mediates the inhibitory effect of PAT-2 and PAT-3 on the EGFR signaling pathway . tln-1 RNAi in the let-60 ( gf ) background caused a strong increase in the VI and similar vulval morphology defects as observed after Pn . p cell-specific pat-3 or pat-2 RNAi ( Fig 4H and 4J ) . Thus , the PAT-2/PAT-3 αβ-integrin complex and its cytoplasmic adapter protein TLN-1 function as negative regulators of EGFR signaling during vulval development . According to our biochemical and genetic interaction studies , loss of dep-1 function is predicted to cause the constitutive phosphorylation of the membrane-proximal NPXY792 motif in the PAT-3 β-integrin cytoplasmic tail . This may alter the intracellular trafficking of the αβ-integrin complex or its interaction with the cytoplasmic adaptor protein TLN-1 talin [20 , 22] . We therefore examined the subcellular localization of the CRISPR/Cas9 engineered endogenous PAT-3::GFP ( zh115 ) and GFP::TLN-1 ( zh117 ) reporters in the vulval cells of wild-type and dep-1 ( lf ) ; lip-1 ( lf ) larvae ( see materials and methods ) . In addition , we engineered an endogenous Y792F mutant PAT-3 ( Y792F ) ::GFP reporter ( zh116 ) to test if mutation of the phosphorylation site in the membrane-proximal NPXY792 motif altered PAT-3 localization . Wild-type PAT-3::GFP was expressed on the lateral membranes of the vulval cells and along the basal laminae that separate the vulval cells from the uterus ( Fig 5A ) . Since PAT-3::GFP was also expressed in the adjacent uterine cells , both tissues are likely to contribute to the strong signal along the basal laminae . PAT-3::GFP localization on the lateral membranes of the vulval cells was most prominent at the onset of vulval invagination in mid to late L3 larvae ( Fig 5A ) , allowing us to specifically observe PAT-3::GFP membrane recruitment in the vulval cells without the contribution from the ventral uterine cells that do not express DEP-1 [11] . To quantify the membrane localization of PAT-3::GFP , we generated intensity profiles from mid-sagittal optical sections across the 2° cells ( Fig 5D ) and calculated from these plots the standard deviation of the mean as a measure of membrane recruitment , as described previously [35] ( Fig 5E ) . We then examined PAT-3::GFP localization in dep-1 ( lf ) ; lip-1 ( lf ) double mutants because the abnormal invagination of the 2° cells observed in this background ( see Fig 3C ) might be a result of reduced integrin activation . However , we did not observe an obvious change in PAT-3::GFP localization in the 2° vulval cells of dep-1 ( lf ) ; lip-1 ( lf ) mutants compared to wild-type larvae at the same stage ( Fig 5B , 5D and 5E ) . Similarly , the localization of the mutant PAT-3 ( Y792F ) ::GFP reporter was not significantly changed compared to the wild-type PAT-3::GFP reporter ( Fig 5C–5E ) . Thus , the de-phosphorylation of the membrane-proximal NPXY792 motif by DEP-1 did not appear to alter PAT-3 β-integrin localization . We next examined the membrane recruitment of the TLN-1 talin adaptor protein in the different genetic backgrounds . The GFP::TLN-1 reporter showed a very similar expression pattern and subcellular localization as the PAT-3::GFP reporter , with prominent expression in all differentiating vulval and uterine cells . In addition to the basal and lateral plasma membrane-associated signal , GFP::TLN-1 expression was also observed in the cytoplasm of the vulval cells of wild-type L3 larvae ( Fig 5F ) . In the dep-1 ( lf ) ; lip-1 ( lf ) background , GFP::TLN-1 exhibited a marked reduction of the signal on the lateral membranes and a more diffuse intracellular signal when compared to wild-type larvae at the same stage ( Fig 5G and 5I ) . Quantification of the signal indicated a more uniform distribution of GFP::TLN-1 in dep-1 ( lf ) ; lip-1 ( lf ) mutants ( Fig 5K ) . In addition , we examined GFP::TLN-1 localization in the pat-3 ( Y792F ) mutant background . The recruitment of GFP::TLN-1 to the lateral VulD membrane was enhanced in the pat-3 ( Y792F ) background compared to the wild-type background ( Fig 5H–5K ) , suggesting that TLN-1 preferentially binds to unphosphorylated PAT-3 , as reported for mammalian talin [22] . We conclude that DEP-1 promotes the membrane recruitment of TLN-1 in the 2° vulval cells by de-phosphorylating the NPXY792 motif in the PAT-3 β-integrin cytoplasmic tail , but DEP-1 has no measurable effect on PAT-3 localization . To examine how the PAT-2/PAT-3 αβ-integrins and TLN-1 talin affect EGFR signaling , we measured LET-23 EGFR dynamics in the vulval cells after RNAi knock-down of the different components of the integrin complex . For this purpose , we performed in vivo fluorescence recovery after photobleaching ( FRAP ) experiments using a functional LET-23::GFP reporter , as previously described [34 , 36] . A portion of the basal membrane in the vulval cells at the Pn . px stage was bleached in pat-2 , pat-3 or tln-1 RNAi treated animals , and the recovery of the LET-23::GFP signal was measured over a five minute period ( Fig 6A and 6B ) . When compared to empty vector treated control animals , RNAi of pat-2 , pat-3 or tln-1 resulted in an overall accelerated recovery of the LET-23::GFP signal ( Fig 6C ) . We used a curve fitting algorithm to calculate the mobile receptor fraction ( i . e . the fraction of the bleached LET-23::GFP signal that recovered after an infinite time period ) and the recovery time t1/3 , the time after which one third of the bleached signal had recovered , as described [34] ( S3 Table ) . This analysis indicated that the faster recovery of LET-23::GFP after pat-2 , pat-3 or tln-1 RNAi was due to an increase in the total mobile receptor fraction as well as to an increased receptor mobility reflected by the shorter t1/3 ( Fig 6D and 6E ) . The changes in LET-23 EGFR dynamics indicate that the PAT-2/PAT-3 αβ-integrins restrict the mobility of the receptor on the basal plasma membrane of the vulval cells together with TLN-1 talin . The increased mobility of LET-23 EGFR is consistent with the enhanced activation of the RAS/MAPK pathway after knock-down of integrin components ( shown in Fig 4 ) , as receptor trafficking is intimately coupled to downstream signaling [34] . DEP-1/PTPRJ was originally identified as a phosphatase that is up-regulated in confluent , contact-inhibited mammalian cell cultures [5] . Reduced DEP-1 activity results in enhanced phosphorylation and endocytosis of the EGFR , leading to increased cell proliferation [12] . Conversely , overexpression of DEP-1 causes rearrangements of the actin cytoskeleton , reduced focal adhesion kinase activity and defects in directed cell migration [7 , 37] . Although numerous DEP-1 substrates including the EGFR [11 , 12] and several other receptor tyrosine kinases have been identified , the relevant substrates that mediate the various effects DEP-1 exerts on cell proliferation and motility remain to be elucidated . By using the substrate-trapping DEP-1 ( D1241A ) mutant in a proteomic approach , we have identified the β-integrin subunit PAT-3 as a novel DEP-1 substrate in C . elegans . DEP-1 shows a remarkable specificity for the membrane-proximal NPXY792 motif in the PAT-3 cytoplasmic tail , whereas the membrane-distal NPXY804 does not appear to be recognized by DEP-1 . In order to test the physiological significance of this interaction , we engineered a non-phosphorylatable β-integrin mutant by replacing the codon encoding the Y792 residue in the membrane proximal NPXY motif with an F codon in the endogenous pat-3 β-integrin locus . Since the Y792F amino acid substitution in PAT-3 does not cause the embryonic lethal Pat phenotype observed in pat-3 ( lf ) mutants [17] , the Y792F mutant β-integrin likely forms active focal adhesion complexes together with PAT-2 α-integrin and retains its ability to interact with cytoplasmic adaptor proteins . However , the pat-3 ( Y792F ) allele caused the partial reversion of the hyperactive EGFR signaling phenotype observed in dep-1 ( lf ) mutants . This genetic interaction between pat-3 ( Y792F ) and dep-1 ( lf ) provides good evidence indicating that PAT-3 is indeed a physiologically relevant substrate of DEP-1 . Since the pat-3 ( Y792F ) mutation did not completely rescue the vulval phenotypes of dep-1 ( lf ) mutants , the un-phosphorylated form of PAT-3 β-integrin mediates some but not all of the inhibitory effects of DEP-1 on EGFR signaling . Most likely , the direct de-phosphorylation of the EGFR accounts for the integrin-independent activity of DEP-1 [11] . We thus propose that DEP-1 acts via two redundant mechanisms to inhibit EGFR signaling ( Fig 7 ) . Phosphorylation of the NPXY motifs in the β-integrin cytoplasmic tail by Src family kinases ( SFK ) is a critical signal inducing a switch of the integrins from the active , open to the inactive , closed conformation , which results in a decreased adhesion to the ECM and increased cell motility [21] . Conversely , de-phosphorylation of the NPXY motifs promotes integrin activation and strengthens cell adhesion . To our knowledge , DEP-1 is the first protein phosphatase known to promote integrin activation via specific de-phosphorylation of a conserved NPXY motif . Cytoplasmic talin proteins bind with their FERM domains preferentially to the un-phosphorylated , membrane-proximal NPXY792 motif in active integrin complexes and stabilize the open integrin conformation [21] . Talin binding strengthens the links between integrins and the cortical F-actin cytoskeleton , thereby stabilizing the focal adhesion complexes . The pat-3 ( Y792F ) mutant can therefore be viewed as a constitutively open , active form of β-integrin that interacts with talin irrespective of SFK activity . The membrane-distal NPXY804 motif , on the other hand , binds to the integrin co-activator kindlin and regulates integrin recycling via sorting nexins [22] . Since DEP-1 specifically de-phosphorylates the membrane-proximal NPXY792 motif , DEP-1 promotes the recruitment of talin to the cell cortex . By contrast , DEP-1 does not appear to affect integrin localization , which involves the membrane-distal NPXY804 motif . Moreover , DEP-1 shows a highly tissue-specific expression pattern during C . elegans development , with most prominent expression in the vulval cells during the larval stages , in the excretory duct cell precursors in the embryo , and in a pair of sensory neurons in the head [11] . PAT-3 and PAT-2 , on the other hand , are broadly expressed in many tissues during all stages of C . elegans development [17] . It thus seems likely that multiple protein phosphatases are required for integrin activation in different tissues . This probably explains why dep-1 ( lf ) mutations neither cause a general loss of integrin activity nor a lethal PAT phenotype . Components of the ECM were previously shown to bind to and stimulate the phosphatase activity of DEP-1 [38] . Thus , DEP-1 may be part of a focal adhesion complex that coordinates cell adhesion with receptor tyrosine kinase ( RTK ) signaling . Multiple modes of crosstalk between RTK pathways and focal adhesion complexes have been reported ( reviewed by [23] ) . First , signal transduction by active integrins depends on intracellular signaling molecules that are shared with a number of receptor tyrosine kinase ( RTK ) signaling pathways , such as Src family kinases , SHC adapter proteins , focal adhesion kinase ( FAK ) , Jun-dependent kinase , PKC and PI3K . For example , FAK and SHC are both downstream effectors of different RTK signaling pathways [39] . However , mutations in kin-32 , which encodes the C . elegans FAK ortholog [40] , or in shc-1 do not alter vulval induction in let-60 ( gf ) mutants , indicating that neither FAK nor SHC regulate EGFR signaling during vulval development [41] . Second , integrins form clusters with the EGFR on the plasma membrane to create a micro-environment , in which the EGFR can efficiently interact with EGF ligands and downstream signaling molecules [23 , 42] . This interaction may involve the induction of EGFR auto-phosphorylation by integrin binding , or signal amplification of EGF-induced receptor activation through integrin-ECM cross-linking [43] . In most of these cases , integrins function as positive regulators of RTK signaling . By contrast , our genetic data indicate that the C . elegans integrins attenuate EGFR signaling in the vulval cells . RNAi against the talin homolog tln-1 , pat-2 α-integrin or pat-3 β-integrin all cause an Apf phenotype characteristic of hyperactive EGFR signaling . Thus , TLN-1 talin is likely to be a critical intracellular mediator of the inhibitory effect , which the integrins exert on the EGFR pathway . The results of the FRAP analysis further indicate that reducing integrin or talin expression causes an increased intracellular mobility and an accelerated recovery of the EGFR on the basolateral plasma membrane . The control of the basolateral EGFR mobility is of particular importance in the vulval cells because the inductive EGF signal is secreted exclusively by the AC , which is located in the somatic gonad facing the basal compartment of the VPCs [32 , 44] . Hence , only the fraction of EGFR molecules localized on the basolateral plasma membrane can bind EGF and activate RAS/MAPK signaling . We thus propose that the activated αβ-integrin/talin complex alters the basolateral membrane micro-environment by recruiting F-actin bundles . Enhanced F-actin recruitment likely reduces the mobility of the EGFR on the basolateral membrane , which in turn inhibits receptor activation and endocytosis [12] . We have previously reported a similar effect on LET-23 EGFR trafficking and signaling in erm-1 ( lf ) mutants [34] . The ezrin homolog erm-1 encodes another FERM domain protein that links F-actin to the basolateral cortex . Taken together , our results point at a model , in which DEP-1 controls EGFR signaling via two distinct mechanisms ( Fig 7 ) . On the one hand , DEP-1 inhibits EGFR signaling by direct de-phosphorylation of the cytoplasmic receptor tail [11 , 12] . On the other hand , DEP-1 restricts EGFR mobility by regulating the activity of the αβ-integrin/talin complex . This dual inhibitory function of DEP-1 ensures a tight regulation of the EGFR signaling pathway . Our findings also provide a good rationale for the reported tumor suppressor activity of DEP-1 in human cancer [6] . Loss of DEP-1 function in cancer cells could simultaneously activate cell proliferation via hyperactive EGFR signaling and increase tumor cell mobility and invasion by reducing integrin-mediated cell adhesion . The strains used for the experiments were derivates of Bristol strain N2 of Caenorhabditis elegans . The animals were cultivated under standard conditions at 20°C as described previously [45] . Unless noted otherwise , the mutations used have been described previously and are listed below by their linkage group . Standard methods were used to construct double and triple mutants . Transgenic animals were generated by microinjection of purified plasmids into the syncytial gonads of young adult worms [46] . All constructs were injected at a concentration of 50 ng/μl ( except for the Pbar-1::pat-3 ( dn ) construct , which was injected at 150ng/μl ) with myo-2::mCherry as transformation marker at a concentration of 2 . 5 ng/μl or pUnc-119 at 10 ng/μl . The total concentration of DNA was adjusted to 150 ng/μl by adding the plasmid pBluescript-KS . The vulval induction index ( VI ) was scored at the L4 larval stage using Nomarski optics as described [14] . For Nomarski analysis , animals were mounted on 4% agarose pads in M9 solution containing 20 mM tetramisole hydrochloride . LGI: tln-1 ( zh117 ) [gfp::tln-1] ( this study ) . LGII: dep-1 ( zh34 ) [11] , unc-4 ( e120 ) [45] , rrf-3 ( pk1426 ) [47] , lin-7 ( e1413 ) . LGIII: unc-119 ( e2498 ) , unc-119 ( ed3 ) , unc-119 ( ed4 ) ( all [48] ) , pat-3 ( zh105 ) [pat-3Y792F] , pat-3 ( zh115 ) [pat-3::gfp] , pat-3 ( zh116 ) [pat-3Y792F::gfp] ( all this study ) . LGIV: lip-1 ( zh15 ) [14] , let-60 ( n1046 ) [49] . LGV: rde-1 ( ne219 ) [50] . Integrated arrays ( transgene , cotransformation marker ) : LGIV: zhIs038[let-23::GFP; unc-119 ( + ) ][34] . LG unknown: qyIs43[pat-3::gfp + ina-1 ( genomic ) , unc-119 ( + ) ][27] , qyIs15 [zmp-1>HA-βtail] [27] , jeIs2222[pat-2::gfp , rol-6 ( su1006 ) ] [26] . Extrachromosomal arrays ( transgene , cotransformation marker ) : zhEx418[lin-31::rde-1 , myo-2::mCherry] [34] , zhEx419[pat-3::gfp Y792A Y804A , myo-2::mCherry] , zhEx420[pat-3::gfp Y772A , myo-2::mCherry] , zhEx432[pat-3::gfp Y772A Y804A , myo-2::mCherry] , zhEx456[pat-3::gfp Y792A , myo-2::mCherry] , zhEx457[pat-3::gfp Y804A , myo-2::mCherry] , zhEx458[pat-3::gfp TTT796-798AAA , myo-2::mCherry] , zhEx524[pat-3::gfp , myo-2::mCherry] , zhEx528[pat-3::gfp Y804A , myo-2::mCherry] ( all this study ) . The intracellular domain of DEP-1 ( wild-type and D1241A ) was cloned into the BamHI site of the E . coli expression vector pGEX-2TK ( Pharmacia ) as described [11] . Recombinant proteins were affinity-purified on glutathione Sepharose according to the manufacturer’s protocol , except that protein expression was induced in BL21 bacteria at 18°C , and fusion proteins were washed in 20 mM NaP pH = 8 . 0 , 250 mM NaCl , 1% Triton-X . Approximately 50 μg of each DEP-1 fusion protein ( wild-type/D1241A ) and 100 μg of GST as a negative control were used in each binding reaction . To prepare N2 worm extracts , mixed-stage liquid cultures were cleaned by sucrose flotation , resuspended in lysis buffer ( 100 mM Tris pH = 8 . 0 , 150 mM NaCl , 1 mM EDTA , 1 mM DTT , 0 . 5% NP-40 , 1x protease inhibitor cocktail; Roche ) , shock frozen in liquid nitrogen , and homogenized in a swing mill ( MM300; Retsch ) . Thawed worm extract was then centrifuged for 10 min at 4°C and 10 , 000g to remove insoluble components . About 2 . 5 mg of total protein extract was used in each reaction . Binding was performed at 4°C overnight , followed by three washes with lysis buffer . Bound proteins were eluted by boiling the beads for 5 min in 30 μl Laemmli buffer and separated on a 4–15% linear gradient SDS-gel ( Biorad Nr . 161–1104 ) , followed by colloidal Coomassie blue staining according to the manufacturer’s protocol ( Roti-Blue; Roth ) . Differential protein bands were excised with a scalpel into small pieces and prepared for in-gel tryptic digestion . Thereby the gel pieces were washed and dehydrated three times in 50% Acetonitrile and dried in SpeedVac . 10 mM Dithiothreitol ( DTT ) in 25 mM Ammonium bicarbonate ( AMBIC ) , pH = 8 . 0 was added to cover gel pieces and incubated for 45 min at 56°C . After DTT was removed , 50 mM Iodoacetamide ( IAM ) in AMBIC 25 mM was added to cover gel pieces and incubated for 1 hour at room temperature in the dark . IAM was removed and gel pieces were washed twice in 50% Acetonitrile ( ACN ) and dried in the SpeedVac . 50 ng trypsin in AMBIC were added for enzymayic digestion and incubated over night at 37°C . To extract the peptides , gel pieces were incubated three times for 15 min with 50% ACN/5% trifluoroacetic acid ( TFA ) and once with 100% ACN . The peptides were dried in a SpeedVac and resolubilized in 5 μl 3% ACN/0 . 1% formic acid ( FA ) . Finally , samples were desalted with C18 ZipTip ( Millipore ) according to the manufacturer’s protocol . Samples were then analyzed on a LTQ-Orbitrap mass spectrometer ( Thermo Fischer Scientific , Bremen , Germany ) coupled to an Eksigent-Nano-HPLC system ( Eksigent Technologies , Dublin ( CA ) , USA ) . Solvent composition at the two channels was 0 . 2% formic acid , 1% acetonitrile for channel A and 0 . 2% formic acid , 80% acetonitrile for channel B . Peptides were resuspended in 3% ACN and 0 . 2% formic acid and loaded on a self-made tip column ( 75 μm × 70 mm ) packed with reverse phase C18 material ( AQ , 3 μm 200 Å , Bischoff GmbH , Leonberg , Germany ) and eluted with a flow rate of 200 nl per min by a gradient from 3 to 40% of B in 55 min , 48% B in 60 min , 97% B in 68 min . Full-scan MS spectra ( 300−2000 m/z ) were acquired with a resolution of 60 000 at 400 m/z after accumulation to a target value of 500 000 . Collision induced dissociation ( CID ) MS/MS spectra were recorded in data dependent manner in the ion trap from the five most intense signals above a threshold of 500 , using a normalized collision energy of 35% and an activation time of 30 ms . Charge state screening was enabled and singly charge states were rejected . Precursor masses selected twice for MS/MS were excluded for further selection for 120s . The exclusion window was set to 20 ppm , while the size of the exclusion list was set to a maximum of 500 entries . Samples were acquired using internal lock mass calibration set on m/z 429 . 088735 and 445 . 120025 . MS/MS spectra were exported to Mascot generic format ( mgf ) and searched against a C . elegans database ( wormbase , April 2010 ) containing reverse entries and common proteomics contaminants . Mascot search engine version 2 . 2 was used for protein identification ( Perkins et al . , 1999 ) , using the following parameters: peptide tolerance at 5 ppm , MS/MS tolerance at 0 . 8 Da , peptide charge of 2+ or 3+ , trypsin as enzyme allowing up to one missed cleavage , carbamidomethylation on cysteines as a fixed modification and oxidation on methionine and phosphorylation of tyrosine as a variable modification . Only peptides with a maximum of 2 ( 3 for semi-tryptic digest ) missed cleavage sites were allowed in database searches . Mascot results ( dat files ) were compared using Scaffold 3 . 0 ( Proteome Software ) filtering the data for proteins identified with at least 2 peptides and a 95% protein Probability . Approximately 10 μg of purified GST::DEP-1 ( wild-type and D1241A ) and 40 μg of GST as a negative control were incubated with ca . 800 μg total worm extract over night at 4°C in each binding reaction . Followed by washing with lysis buffer ( see above ) , bound proteins were eluted by boiling the beads for 5 min in Laemmli buffer . GFP tagged proteins were detected on Western blots of 10% acrylamide gels with monoclonal Anti-GFP antibody ( Roche , Cat . No . 11 814 460 001 ) . For phosphatase inhibitor experiments , 5 mM Na3VO4 was added to the lysis buffer . Genome editing was performed using the co-CRISPR strategy described by [51] . For a detailed description of the procedure and the construction of the plasmids used , see S1 Methods . RNAi was performed by feeding worms with dsRNA-producing E . coli as described [52] with the following modifications: The worms were synchronized with hypochloride solution , and L1 larvae ( P0 ) were placed on growth media plates containing 3 mM IPTG and allowed to grow at 20°C . The F2 generation was analyzed . The PAT-3::GFP and GFP::TLN-1 reporters were imaged with an X-light spinning disc microscope using a 70 μm pinhole , a 475 nm solid state light source ( Lumencor Spectra light engine ) and a CMOS ( Hamamatsu ORCA flash 4 . 0 ) or EMCCD ( iXon Ultra 888 ) camera . For each animal , around 50 stacks were taken at 0 . 3–0 . 5 μm z-spacing . Images were deconvolved using the Huygens software package ( Scientific Volume imaging ) and the three mid-sagittal sections around the AC were projected to generate intensity plots with the Fiji software [53] using a custom script described in [35] . For LET-23::GFP FRAP analysis , RNAi treated L3 larvae at the Pn . pxx stage were analyzed using a confocal scanning microscope ( Zeiss LSM710 equipped with 458/488/514 nm argon laser ) as described [36] . Curve fitting was done using the solver function in MS Excel as described previously [34] . The raw data and data analysis are found in S3 Table .	The phosphorylation of proteins by kinases is one of the most common post-translational modifications that regulates protein function in a variety of processes . Protein kinases are usually counteracted by specific phosphatases that remove the phosphate groups from proteins . We have undertaken a systematic biochemical ( “proteomics” ) approach to identify the substrates of the Density-Enhanced Phosphatase DEP-1 . The DEP-1 phosphatase has previously been identified as a growth inhibitor and tumor suppressor that de-phosphorylates different growth factor receptors to inhibit intracellular signal transduction and cell proliferation . Here , we have identified a ß-integrin subunit as a specific DEP-1 substrate . Integrins consist of one α and one β subunits that function as cell adhesion molecules mediating the attachment of cells to the extracellular matrix . The de-phosphorylation of integrins causes their activation and stabilizes the adhesion sites . By engineering a ß-integrin mutation that cannot be phosphorylated and hence does not depend on the DEP-1 phosphatase , we show that integrin activation not only permits extracellular matrix adhesion but also attenuates epidermal growth factor signaling . Our findings point at a dual role of the DEP-1 phosphatase in regulating EGFR signaling by simultaneously de-phosphorylating integrins and growth factor receptors .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "methods" ]	[ "phosphorylation", "invertebrates", "rna", "interference", "caenorhabditis", "enzymes", "egfr", "signaling", "enzymology", "animals", "phosphatases", "animal", "models", "membrane", "proteins", "caenorhabditis", "elegans", "integrins", "model", "organisms", "experimental", "organism", "systems", "epigenetics", "cellular", "structures", "and", "organelles", "research", "and", "analysis", "methods", "genetic", "interference", "cell", "adhesion", "proteins", "extracellular", "matrix", "gene", "expression", "cell", "membranes", "biochemistry", "signal", "transduction", "rna", "cell", "biology", "nucleic", "acids", "post-translational", "modification", "genetics", "nematoda", "biology", "and", "life", "sciences", "cell", "signaling", "organisms" ]	2017	β-Integrin de-phosphorylation by the Density-Enhanced Phosphatase DEP-1 attenuates EGFR signaling in C. elegans
The circadian clock underlies daily rhythms of diverse physiological processes , and alterations in clock function have been linked to numerous pathologies . To apply chemical biology methods to modulate and dissect the clock mechanism with new chemical probes , we performed a circadian screen of ∼120 , 000 uncharacterized compounds on human cells containing a circadian reporter . The analysis identified a small molecule that potently lengthens the circadian period in a dose-dependent manner . Subsequent analysis showed that the compound also lengthened the period in a variety of cells from different tissues including the mouse suprachiasmatic nucleus , the central clock controlling behavioral rhythms . Based on the prominent period lengthening effect , we named the compound longdaysin . Longdaysin was amenable for chemical modification to perform affinity chromatography coupled with mass spectrometry analysis to identify target proteins . Combined with siRNA-mediated gene knockdown , we identified the protein kinases CKIδ , CKIα , and ERK2 as targets of longdaysin responsible for the observed effect on circadian period . Although individual knockdown of CKIδ , CKIα , and ERK2 had small period effects , their combinatorial knockdown dramatically lengthened the period similar to longdaysin treatment . We characterized the role of CKIα in the clock mechanism and found that CKIα-mediated phosphorylation stimulated degradation of a clock protein PER1 , similar to the function of CKIδ . Longdaysin treatment inhibited PER1 degradation , providing insight into the mechanism of longdaysin-dependent period lengthening . Using larval zebrafish , we further demonstrated that longdaysin drastically lengthened circadian period in vivo . Taken together , the chemical biology approach not only revealed CKIα as a clock regulatory kinase but also identified a multiple kinase network conferring robustness to the clock . Longdaysin provides novel possibilities in manipulating clock function due to its ability to simultaneously inhibit several key components of this conserved network across species . A variety of physiological processes such as sleep/wake behavior , body temperature , hormone secretion , and metabolism show daily rhythms under the control of the circadian clock which is intrinsic to the organism . Perturbation of clock function has been implicated in numerous pathologies including circadian sleep disorders , cardiovascular disease , cancer , and metabolic disease [1]–[4] . The close association of the circadian clock with diverse physiological processes and diseases implies that identification of clock-modulating compounds could form the basis for therapeutic strategies directed towards circadian rhythm-related disorders , shift-work fatigue , and jet lag . The manifestation of circadian disorders at the level of the whole organism can be caused by dysfunction of the clock at the level of intracellular networks , as single cells exhibit circadian rhythms in a cell-autonomous manner [5]–[6] . In mammals , these cellular oscillators are organized in a hierarchy , in which the suprachiasmatic nucleus ( SCN ) of the hypothalamus constitutes the central circadian pacemaker controlling behavioral rhythms , while peripheral clocks in other tissues control local rhythmic outputs [1] , [3] , [7] . In the intracellular circadian network , the clock genes and their protein products form transcriptional feedback loops: CLOCK and BMAL1 transcription factors activate expression of Per and Cry genes , and PER and CRY proteins ( PER1 , PER2 , CRY1 , and CRY2 ) in turn inhibit their own transcription to generate rhythmic gene expression [3] , [8] . In addition to transcriptional regulation , post-translational modification of clock proteins provides another level of regulation , as most clock proteins undergo rhythmic phosphorylation [9] . Hamster tau mutants showing a short period behavioral rhythm have a missense mutation in the CKIε gene [10] , and human familial advanced sleep phase syndrome ( FASPS ) with early sleep times is attributed to missense mutations of PER2 and CKIδ genes [11]–[12] . CKIδ and CKIε phosphorylate PER to trigger proteasomal degradation , and tau and FASPS mutations lead to higher PER degradation than wild type , consistent with the short period phenotype [13]–[15] . Supporting the functional importance of CKIδ/ε , application of the known CKI inhibitors IC261 , CKI-7 , and D4476 causes period lengthening in cultured cells [14] , [16]–[17] . Generation of CKIε and CKIδ deficient mice [15] , [18] as well as the development of the CKIε-selective inhibitor PF-4800567 [19] revealed the minimal , if any , role of CKIε in period length regulation and also demonstrated a dominant role for CKIδ . In contrast , potential roles of CKI family members other than CKIδ/ε are less characterized: They show much less binding with PER1 than that of CKIε [20]–[22] , and knockdown of CKIα-like , a homolog of CKIα , has no period effect in cultured cells [23] . Together with CKIδ/ε , GSK-3β and CK2 are also implicated in period regulation . GSK-3β phosphorylates PER2 , CRY2 , REV-ERBα , CLOCK , and BMAL1 for functional regulation [24]–[28] , and pharmacological and RNAi-based inhibition of GSK-3β causes period shortening in cultured cells [29]–[30] . Conversely , inhibition of CK2 causes period lengthening [29] , [31]–[33] , and CK2-mediated phosphorylation regulates PER2 and BMAL1 functions [31]–[32] , [34] . Genetic and molecular biological studies over the past two decades have identified more than a dozen genes that form the core of the mammalian circadian network [3] , [8] , [35] . However , it is clear that more clock components and modulators remain to be discovered [36] . Considering the limitations of conventional biological approaches associated with lethality , pleiotropy , and functional redundancy of closely related proteins , introduction of new strategies will accelerate the identification of novel clock mechanism . Chemical biology approaches are attractive candidates , because they utilize small molecules as proof-of-concept probes for biological systems and can be effective in discovering novel biological mechanisms and evaluating their effects in vivo by complementing the limitations of conventional biological approaches [7] , [37] . Furthermore , the circadian clock network can be a good target for chemical biology approaches due to the quantitative readout of an oscillation . To discover new chemical probes for dissecting biological mechanisms , it is valuable to screen comprehensive , large-scale compound libraries containing hundreds of thousands of compounds , because a wide variation of chemical structures has the advantage of probing many classes of potential targets . Although it is technically challenging to identify proteins specifically affected by a novel compound [38] , this process might also be necessary for known compounds , given that even well-characterized kinase inhibitors have off-targets unrelated to the primary effect [39]–[40] . Combined with conventional biological approaches , the chemical biology approach is expected to provide an effective way to identify novel components of the circadian clock [41] . We previously developed a cell-based high-throughput circadian assay system to perform compound screening [29] . In this system , Bmal1-dLuc reporter cells derived from human U2OS osteosarcoma cells show robust luminescence rhythms on 384-well plates by expressing a rapidly degradable luciferase under the control of a mouse Bmal1 gene promoter . We initially tested a chemical library containing 1 , 280 well-characterized compounds ( LOPAC; Library of Pharmacologically Active Compounds ) and found 11 compounds that change period length of the luminescence rhythms in a dose-dependent manner . The kinase inhibitors among the hit compounds revealed novel roles of GSK-3β and CK2 in the mammalian clock mechanism as described above . Furthermore , many of the hits were previously known to alter the circadian period in other organisms and tissue preparations , demonstrating the predictive value of the high-throughput assay system [29] . Together , these observations indicate the effectiveness of small molecules as probes and/or modulators of the circadian clock mechanism . A similar LOPAC screen in NIH3T3 and U2OS cells identified CKIδ/ε-dependent phosphorylation as a chemically sensitive process of the clock [42] . It was found that the CKIδ/ε-targeting compounds cause much larger period lengthening than CKIδ gene knockout [18] , [29] , [42] , but the molecular mechanism underlying the strong effects of the compounds remains unknown . The present study aimed to apply chemical biology methods to probe the clock mechanism with novel small molecules through a circadian screen of a structurally diverse library of ∼120 , 000 uncharacterized compounds . We found a purine derivative , longdaysin , that dramatically lengthens the circadian period . Identification and characterization of longdaysin-target proteins revealed the roles for protein kinases CKIα and ERK2 in period regulation , as well as confirmed the importance of CKIδ . Simultaneous inhibition of these three kinases drastically lengthened the circadian period , illustrating a new facet of the clock mechanism whose robustness is conferred in part by a multiple kinase network . By applying a high-throughput circadian assay system using human U2OS cells with Bmal1-dLuc reporter [29] , we analyzed approximately 120 , 000 uncharacterized compounds corresponding to diverse chemical scaffolds [43]–[44] at a final concentration of 7 µM . We identified a number of compounds with different scaffolds that lengthened the circadian period of cellular luminescence rhythms . Among them , we selected one purine derivative compound 1 ( Figure 1A ) for follow-up studies , because it strongly lengthened the period in a dose-dependent manner and showed less effect on the amplitude of Bmal1-dLuc rhythms ( Figure S1 ) . A preliminary structure-activity relationship study helped to identify a derivative of compound 1 that is 3 times more potent and able to generate >10 h period change at a concentration of 10 µM ( Figure 1B and Table 1 ) . We termed this derivative “longdaysin” ( Figure 1A ) , based on its prominent period lengthening effect . We further investigated the effect of longdaysin ( Figure 1C–E ) by using primary cells and tissues isolated from mPer2Luc knockin mice harboring a mPer2Luc reporter [45]–[46] as an additional clock-controlled reporter different from Bmal1-dLuc used in the screen . Longdaysin consistently caused dose-dependent period lengthening in adult tail fibroblasts ( Figure 1C ) and lung explants ( Figure 1D ) , which represent peripheral clocks , and in SCN explants ( Figure 1E ) , which represent the central clock . The effect of longdaysin was reversible , as the period length returned to normal after washout of the compound ( Figure S2 ) . Taken together , these results demonstrate that longdaysin potently lengthens the circadian period in multiple mammalian cells including SCN neurons . In order to identify potential biological targets of longdaysin by affinity-based proteomic approaches [38] , we synthesized longdaysin analogs with an aminohexyl linker , based on the preliminary structure-activity relationship analysis . Among them , compound 2 with a linker at the C2 position ( Figure 2A ) retained the period lengthening effect in the cell-based circadian assay ( Figure 2B ) . We then prepared agarose-conjugated compound 3 ( Figure 2A ) and incubated it with U2OS cell lysate in the presence or absence of 100 µM longdaysin as a soluble competitor ( Figure 2C ) . Proteins that bound to the affinity resin , and could be competed off by free longdaysin , were separated by SDS-PAGE and analyzed by liquid chromatography-tandem mass spectrometry ( LC-MS/MS ) . This analysis yielded 10 proteins ( Figure 2D ) including the protein kinases ( highlighted in blue ) CKIδ ( CSNK1D ) , CKIα ( CSNK1A1 ) , ERK2 ( MAPK1 ) , CDK7 , and p38α ( MAPK14 ) . Independent affinity chromatography followed by Western blotting with specific antibodies confirmed both the binding of the protein kinases to the affinity resin as well as decreased binding in the presence of free longdaysin ( Figure S3 ) . Furthermore , in vitro kinase assays revealed that longdaysin inhibited CKIδ , CKIα , ERK2 , and CDK7 activities ( IC50 = 8 . 8 , 5 . 6 , 52 , and 29 µM , respectively; Figure 2E and Table 1 ) , while it had much less effect on p38α ( unpublished data ) . In contrast , compound 1 inhibited CKIδ , CKIα , and ERK2 with ∼3 times less potency than longdaysin and inhibited CDK7 similarly to longdaysin ( Figure 2E and Table 1 ) . The difference in potency between longdaysin and compound 1 against CKIδ , CKIα , and ERK2 was consistent with their cellular period effects ( Table 1 ) , suggesting an involvement of these three kinases in the period regulation . To identify the protein ( s ) mediating longdaysin effect on period length , we first tested the contribution of CKIδ , a well-characterized kinase in period regulation [12] , [18]–[19] , [47] , by using embryonic fibroblasts prepared from CKIδ deficient mice harboring the mPer2Luc knockin reporter [18] . In a 384-well plate format , the period of CKIδ deficient ( CKIδΔ2/Δ2 ) cells was 1 . 1 h longer than that of wild type ( CKIδ+/+ ) cells ( CKIδΔ2/Δ2 , 25 . 9±0 . 5 h; CKIδ+/+ , 24 . 8±0 . 9 h; n = 48 ) , consistent with a previous report [18] . We found that longdaysin lengthened the period in a dose-dependent manner in CKIδ deficient cells as well as in wild type cells ( Figure 3A ) . This result indicates the presence of additional longdaysin-target ( s ) that regulate period length besides CKIδ . To investigate the effects of RNAi-mediated inhibition of potential longdaysin-targeted kinases on the circadian period , we conducted knockdown experiments by applying four independent siRNAs against each gene . At least two siRNAs for CSNK1D ( encoding CKIδ ) , CSNK1A1 ( CKIα ) , and MAPK1 ( ERK2 ) caused period lengthening ( Figure 3B , red box ) , while those for CDK7 and MAPK14 ( p38α ) did not , thus proposing CKIδ , CKIα , and ERK2 as the potential clock-acting targets of longdaysin . We also tested siRNAs against the close homologs of these three kinases ( CSNK1E for CSNK1D , CSNK1A1L for CSNK1A1 , and MAPK3 for MAPK1 ) and found that the homologs had little or no effect on the period ( Figure 3B ) . The minor period effects for MAPK14 , CSNK1E , and CSNK1A1L are in line with previous reports [23] , [42] . We further looked at the primary screening data from our genome-wide RNAi study [33] in which we used four siRNAs different from this study by combining two siRNAs as a pool . Among the 10 longdaysin-interacting proteins identified by the affinity chromatography ( Figure 2D ) , only CSNK1D , CSNK1A1 , and MAPK1 showed period lengthening of the reporter with both siRNA pairs ( Figure S4 , red box ) , supporting important roles for CKIδ , CKIα , and ERK2 in period regulation as longdaysin targets . We then characterized the period lengthening effects of siRNAs against CSNK1D , CSNK1A1 , and MAPK1 by using an 8-point dilution series of the effective siRNAs . All siRNAs tested gave dose-dependent changes of the period ( Figure 3Ci ) and reduction of the target gene mRNA levels without affecting the levels of closely homologous genes ( Figure 3Cii ) . The only exception was CSNK1A1 si1 , which has sequence similarity against CSNK1A1L mRNA ( 96% identical on a 23 bp stretch ) and reduced its level at higher dose ( Figure 3Cii ) . The correlation between period effect and mRNA knockdown effect matched well for two siRNAs against CSNK1A1 or MAPK1 ( Figure S5 ) . The proportional changes of circadian function by dose-dependent knockdown of CSNK1D , CSNK1A1 , and MAPK1 ( Figure 3C ) are common characteristics among the core clock components and clock modifiers [23] , [33] . Taken together , these results illustrate the involvement of CKIα and ERK2 in the period regulation , as well as confirming the importance of CKIδ . We further tested knockdown of all three kinases CSNK1D , CSNK1A1 , and MAPK1 in combination , in order to determine if their concomitant reduction could explain the strong effect of longdaysin as an inhibitor of all three kinases . Combinatorial knockdown of the three genes caused strong and dose-dependent lengthening of the period to >10 h ( Figure 3Di ) . The multiple gene knockdown effect ( Figure 3Dii , red line ) matched well with the theoretical sum of the effect of single gene knockdown ( black line ) . These results suggest that the knockdown of these three kinases works in an additive manner to cause prominent period lengthening similar to that generated by longdaysin . In contrast to the well-characterized roles of CKIδ/ε-mediated phosphorylation of PER proteins [9]–[21] , the functions of CKIα and ERK2 in period regulation have yet to be characterized . To examine the interaction of CKIα and ERK2 with the core clock proteins , we co-expressed HA-tagged kinases with Flag-tagged clock proteins in HEK293T cells . Immunoprecipitation assay revealed interactions of both CKIα and ERK2 with PER1 and PER2 , and to a lesser extent , CRY1 and CRY2 ( Figures 4A , S6A , and S6B ) . Co-expression of PER1 with CKIα generated lower electrophoretic mobility forms of PER1 ( Figure 4B ) , which disappeared upon phosphatase treatment ( Figure S7A ) , suggesting CKIα-dependent phosphorylation of PER1 . This modification relied on the kinase activity of CKIα , because the kinase-dead K46R mutant of CKIα [CKIα ( KR ) ] [48] did not cause the mobility-shift of PER1 ( Figure 4B ) . In contrast , ERK2 showed no detectable effect on the PER1 mobility ( unpublished data ) . Treatment of the cells with longdaysin reduced the CKIα- and CKIδ-dependent mobility-shift of PER1 ( Figures 4C , S7B , and S7C ) , consistent with a potential mode of longdaysin action through CKIα and CKIδ . As phosphorylation of PER1 modulates its stability [16]–[17] , [21] , we further tested the effect of CKIα on the stability of PER1 by expressing luciferase-fused PER1 protein ( PER1-LUC ) in HEK293T cells . Luminescence changes were monitored following inhibition of de novo protein synthesis by cycloheximide treatment . Co-expression of CKIα but not CKIα ( KR ) accelerated PER1-LUC degradation relative to that of LUC ( Figure 4D ) . The CKIα- and CKIδ-dependent degradation of PER1 was inhibited by longdaysin treatment in a dose-dependent manner ( Figure 4E ) . Similar results were obtained by using PER1 without the LUC fusion ( Figure S8A ) . In contrast , we found that CKIα had no effect on the stability of a PER2-LUC fusion protein while CKIδ accelerated its degradation ( Figure S8B ) . These results demonstrated the selectivity of CKIα against PER1 degradation over PER2 . Longdaysin inhibited the CKIδ-mediated PER2-LUC degradation in a dose-dependent manner ( Figure S8C ) , suggesting its role in regulating both PER1 and PER2 stabilities through CKIδ and/or CKIα . We then investigated the effect of longdaysin on the protein level of endogenous PER1 in Bmal1-dLuc U2OS cells during circadian cycles . The cells were synchronized with medium change and collected every 6 h from 28 h to 58 h after the medium change ( Figure 4Fi ) . Consistent with the longdaysin-dependent shift of the second trough of Bmal1-dLuc luminescence rhythm ( 34 , 38 , and 46 h for 0 , 3 , and 9 µM longdaysin , respectively; Figure 4Fi ) , the second peak of PER1 protein rhythm shifted in parallel ( 40 , 40–46 , and 52 h for 0 , 3 , and 9 µM longdaysin , respectively; Figure 4Fii ) . Furthermore , 3 or 9 µM longdaysin treatment strongly up-regulated overall protein amount of PER1 compared with 0 µM control ( Figure 4Fii ) without affecting its mRNA level ( Figure 4Fiii ) , demonstrating post-transcriptional increase of endogenous PER1 by longdaysin . The progressive phosphorylation of PER1 was still observed in the presence of longdaysin ( Figure 4Fii ) , possibly because of the phosphorylation by kinase ( s ) that was not affected by longdaysin . Collectively , these results provide a possible mechanism of longdaysin action for period regulation through the CKIα- and CKIδ-mediated control of PER1 stability . Lastly , we investigated if longdaysin had any in vivo efficacy by using zebrafish , which provide a useful model system for studies on circadian rhythms at the level of the whole organism [49] , and have conserved CKI and ERK family genes [50]–[51] . By using transgenic zebrafish harboring a per3-luc reporter [52] , we first established an in vivo circadian assay to investigate the effects of compounds . Larval per3-luc fish were entrained in 12 h light/12 h dark cycles from day 3 to 6 postfertilization and then placed in an individual well of a 96-well plate to monitor luminescence rhythms under constant darkness . By using this assay , we found that longdaysin treatment caused >10 h period lengthening in a dose-dependent manner in per3-luc reporter fish ( Figure 5A and 5B ) , without affecting body size ( Figures 5C and S9 ) . The in vivo period changes were similar to those observed in mammalian tissues and cells ( Figure 1 ) , showing the prominent characteristics of longdaysin as a period lengthening compound . The present study highlighted the effectiveness of the chemical biology approach in dissecting circadian clock mechanisms . Our large-scale small molecule screening identified a novel compound longdaysin that exhibited a drastic effect on the circadian period of not only a variety of mammalian cells but also zebrafish in vivo . As a first attempt to determine the molecular mechanism underlying such a large period effect , we conducted affinity-based proteomics and siRNA-mediated knockdown analyses . Our results revealed CKIδ , CKIα , and ERK2 as targets of longdaysin for period regulation . Effective concentrations of longdaysin against CKIδ and CKIα in a cell-based PER1 degradation assay were similar to those in in vitro kinase assays ( Table 1 ) , suggesting efficient cell permeability of the compound . Treatment with 10 µM longdaysin consistently inhibited CKIδ and CKIα activities in vitro and their effects on PER1 phosphorylation and degradation , resulting in a 13 h period lengthening in U2OS cells . The increasing period effect at the range of 3 to 24 µM , in which longdaysin considerably inhibited ERK2 in vitro , supported the role of ERK2 at higher longdaysin concentration . In mammals , the CKI family of Ser/Thr kinases contains seven members ( α , β , γ1 , γ2 , γ3 , δ , and ε ) . While the roles of CKIδ/ε in the circadian clock mechanism have been extensively studied , the inhibition of CKIδ/ε alone is insufficient to explain the drastic effect of longdaysin . Knockout of these genes has only a modest effect on period length and the effect of longdaysin was also observed in CKIδ deficient cells . We found that CKIα , in addition to CKIδ and CKIε , binds to PER1 and regulates its stability . This observation is reminiscent of the CKIα- , CKIδ- , and CKIε-mediated regulation of β-catenin and Ci , key players in the Wnt and Hedgehog signaling pathways [53]–[54] . Similar to NFAT transcription factors that are the targets of CKIα and/or CKIε [55]–[56] , the CKI docking site of PER1/2 contains a FXXXF motif necessary for CKIε binding [56] . Although PER1/2 bind with CKIα/δ/ε , the affinity of CKIα is much lower than CKIδ/ε ( Figure S6C ) . A recent study demonstrated that two amino acid residues in the CKI kinase domain cause weaker affinity of CKIα for PER1 compared with CKIε [22] . The low affinity will be advantageous to release PER proteins from CKIα for subsequent regulations , such as phosphorylation by other kinases and degradation . Disruption of the circadian rhythm in CKIδ deficient fibroblasts by overexpression of dominant negative form of CKIε [57] may be mediated also by perturbation of CKIα-dependent regulation , because of the tight binding of dominant negative CKIε with PER proteins . On the other hand , ERK1 and ERK2 MAP kinases have been well characterized in the resetting mechanism of the clock [58] . Our results demonstrated a role for ERK2 in the regulation of circadian period as well . Attenuation of the circadian rhythms in SCN explants by treatment with the MEK ( ERK kinase ) inhibitor U0126 [59] could potentially be explained by strong inhibition of both ERK1 and ERK2 . Similar to CKIδ and CKIα , ERK2 bound to PER1/2 , suggesting PER protein as a key node in phosphorylation-dependent period regulation by multiple kinases . The effect of ERK2 on PER phosphorylation and function will be addressed in future studies . CKIδ , CKIα , and ERK2 are involved in diverse cellular processes such as cell proliferation and apoptosis [53] , [60] . Consistently , CKIδ and ERK2 are required for normal development as revealed by gene knockout studies [18] , [61]–[62] , while CKIα deficient mice are not reported yet . In addition to the regulation of PER by these kinases , it is possible that the regulation of other clock proteins and/or changes in cellular physiology may also affect the circadian period . Therefore , it is important to identify specific residues of PER responsible for the CKIδ- , CKIα- , and ERK2-mediated period regulation . In contrast to the CKIδ-dependent progressive phosphorylation of PER1 , CKIα caused a smaller mobility-shift ( Figure 4B , C ) , suggesting a key role for site-specific phosphorylation rather than a global change of phosphorylation level . In Neurospora , a quantitative mass spectrometry approach identified >75 in vivo phosphorylated residues of the clock protein FRQ [63] . Interestingly , phosphorylation of two distinct regions leads to opposing effects on FRQ stability and circadian period [63] . In mammals , phosphorylation site mapping via mass spectrometry identified 21 phosphorylated Ser/Thr residues in PER2 overexpressed in HEK293 cells [14] . Identification of PER1 phosphorylation sites and characterization of the role of each residue will lead to the understanding of CKIδ- , CKIα- , and ERK2-mediated regulation of PER1 function and the circadian period . Furthermore , the phenotypic differences between PER1 and PER2 observed in CKIα-dependent regulation of stability ( Figure S8B ) and CKIε-mediated control of nuclear translocation [64] could be explained by comparing phosphorylation sites and their functions . We found that combinatorial knockdown of CKIδ , CKIα , and ERK2 worked additively for prominent period lengthening ( Figure 3D ) , similar to that caused by longdaysin . In contrast , knockout of CKIδ ( Figure 3A ) , knockdown of single kinase ( Figure 3B , C ) , and CKI inhibitors D4476 and IC261 ( Figure S10 ) all showed smaller period effects . These observations indicate that the network of multiple kinases confers robustness to the clock mechanism . A single small molecule such as longdaysin inhibiting the multiple pathways simultaneously can significantly perturb the clock system and elicit unexpectedly long period . Previous screening of the LOPAC chemical library identified several kinase inhibitors that cause large period lengthening [29] , [42] . These compounds have the potential to inhibit CKIδ/ε [39]–[40] , [42] , although their primary target is CDK , p38 MAPK , JNK , CK2 , or VEGFR signaling pathway . Because of the high conservation of the kinase domain between CKIδ and CKIα , these compounds are also likely to inhibit CKIα . Considering our current finding , the inhibition of the primary target ( CDK , p38 MAPK , JNK , CK2 , or VEGFR signaling pathway ) in combination with CKIδ/ε/α may be essential for the large period effect of these compounds . Supporting this idea , CK2 acts cooperatively with CKIε to regulate PER2 stability [32] . Having multiple targets might be a common characteristic of therapeutically effective compounds , such as sunitinib and sorafenib for cancer treatment [65] . Our zebrafish experiments clearly showed an in vivo effect of longdaysin in a vertebrate , and further optimization of longdaysin in mammalian systems may provide a chemical starting point for the identification of small molecule therapeutics specifically designed for ameliorating circadian disorders . All animal studies were approved by the University of California , San Diego , Institutional Animal Care and Use Committee and performed in accordance with the guidelines . Synthesis of compound 1 , longdaysin , compound 2 , and compound 3 is described in Text S1 . The dilution series of the compounds was made on 384-well plates by using a robotic liquid handling system ( MiniTrak , Perkin-Elmer ) . The compound screen was done with the high-throughput circadian assay system as described previously [29] . In brief , Bmal1-dLuc U2OS cells were suspended in the culture medium ( DMEM supplemented with 10% fetal bovine serum , 0 . 29 mg/ml L-glutamine , 100 units/ml penicillin , and 100 µg/ml streptomycin ) and plated onto 384-well white solid-bottom plates at 20 µl ( 2 , 000 cells ) per well . After 2 d , 50 µl of the explant medium ( DMEM supplemented with 2% B27 , 10 mM HEPES , 0 . 38 mg/ml sodium bicarbonate , 0 . 29 mg/ml L-glutamine , 100 units/ml penicillin , 100 µg/ml streptomycin , 0 . 1 mg/ml gentamicin , and 1 mM luciferin , pH 7 . 2 ) was dispensed to each well , followed by the application of 500 nl of compounds ( dissolved in DMSO; final 0 . 7% DMSO ) . The plate was covered with an optically clear film and set to luminescence monitoring system equipped with a CCD imager ( ViewLux , Perkin Elmer ) . The luminescence was recorded every 2 h for 3–4 days . In follow-up studies , the luminescence was recorded every 100 min by using a microplate reader ( Infinite M200 , Tecan ) . The period parameter was obtained from the luminescence rhythm by curve fitting program CellulaRhythm [29] or MultiCycle ( Actimetrics ) , both of which gave similar results . Luminescence rhythms of adult tail fibroblasts [46] and embryonic fibroblasts [18] from mPer2Luc knockin mice were analyzed similarly to U2OS cells , except that 1 , 800 cells were plated per well . Because of the low luminescence intensity of the fibroblasts , the higher sensitivity ViewLux imager was used for rhythm recording . Explants of lung and SCN were dissected from mPer2Luc knockin mice [45] and cultured in explant medium as described previously [46] . The medium was changed every week with increasing concentration of longdaysin each time ( from 0 to 9 µM , final 0 . 7% DMSO ) . The luminescence was recorded every 10 min with LumiCycle luminometer ( Actimetrics ) , and the period parameter was obtained by using LumiCycle Analysis software ( Actimetrics ) . U2OS cells kept in confluence ( 2×108 cells ) were collected with ice-cold PBS and homogenized by using Dounce homogenizer in 5 ml of lysis buffer ( 25 mM MOPS , 15 mM EGTA , 15 mM MgCl2 , 1 mM DTT , 60 mM β-glycerophosphate , 15 mM p-nitrophenyl phosphate , 1 mM Na3VO4 , 1 mM NaF , 1 mM phenyl phosphate , 10 µg/ml leupeptin , 10 µg/ml aprotinin , 10 µg/ml soybean trypsin inhibitor , 100 µM benzamidine , pH7 . 2 ) . The homogenate was sonicated and centrifuged ( 16 , 000×g ) at 4°C for 20 min . The resulting supernatant was split into two , and each portion was incubated with or without 100 µM longdaysin ( final 0 . 1% DMSO ) at 4°C for 10 min ( Figure 2C ) . Then , 120 µl of compound 3 [50% slurry in bead buffer ( 50 mM Tris , 250 mM NaCl , 5 mM EDTA , 5 mM EGTA , 0 . 1% NP-40 , 5 mM NaF , 10 µg/ml leupeptin , 10 µg/ml aprotinin , 10 µg/ml soybean trypsin inhibitor , 100 µM benzamidine , pH 7 . 4 ) ] was added to the mixture and incubated at 4°C for 1 h with rotation . The agarose beads were washed 6 times with 2 ml of the bead buffer . The bound proteins were eluted with SDS sample buffer and separated by SDS-PAGE ( 4%–12% gradient gel , Invitrogen ) . The gel was CBB stained , and the gel lane for each condition was cut horizontally into 24 pieces . All gel bands were subjected to LC-MS/MS analysis as described previously [66] . Tandem MS data were analyzed using Sequest ( ThermoFinnigan , San Jose , CA; Version 3 . 0 ) . Sequest was set up to search a Homo sapiens subset of the EBI-IPI database ( Version 3 . 32 ) to which a reversed copy of the protein database was appended , assuming the digestion enzyme trypsin . Sequest was searched with a fragment ion mass tolerance of 0 Da and a parent ion tolerance of 3 . 0 Da . Iodoacetamide derivative of cysteine was specified in Sequest as a fixed modification . Oxidation of methionine was specified in Sequest as a variable modification . Scaffold ( version Scaffold_2_05_00 , Proteome Software Inc . , Portland , OR ) was used to validate MS/MS based peptide and protein identifications . Peptide identifications were accepted if they could be established at greater than 95 . 0% probability as specified by the Peptide Prophet algorithm . Protein identifications were accepted if they could be established at greater than 99 . 0% probability and contained at least three unique peptides . Protein probabilities were assigned by the Protein Prophet algorithm . Crude differential quantitation of proteins identified in both pulldown experiments was performed by comparing the number of assigned peptides . The CKIδ , CKIα , CDK7 , and ERK2 kinase assays were performed on 384-well plates ( 10 µl volume ) . The reaction mixture was as follows: for CKIδ , 2 ng/µl CKIδ ( Millipore , 14-520 ) , 50 µM peptide substrate RKKKAEpSVASLTSQCSYSS corresponding to human PER2 Lys659-Ser674 [47] , and CKI buffer ( 40 mM Tris , 10 mM MgCl2 , 0 . 5 mM DTT , 0 . 1 mg/ml BSA , pH 7 . 5 ) ; for CKIα , 1 ng/µl CKIα ( Invitrogen , PV3850 ) , 50 µM CKI peptide substrate ( Anaspec , 60547-1 ) , and CKI buffer; for CDK7 , 5 ng/µl CDK7 ( Millipore , 14-476 ) , 100 µM Cdk7/9 peptide substrate ( Millipore , 12-526 ) , and CKI buffer; for ERK2 , 1 . 5 ng/µl ERK2 ( Millipore , 14-550 ) , 0 . 8 µg/µl MBP ( Millipore , 13-104 ) , and ERK buffer ( 50 mM Tris , 10 mM MgCl2 , 0 . 5 mM DTT , 1 mM EGTA , pH 7 . 5 ) . Five hundred nl of compound was added to the mixture ( final 5% DMSO ) , and the reaction was started by adding ATP ( final 5 µM ) . After incubation at 30°C for 3h , 10 µl of Kinase-Glo Luminescent Kinase Assay reagent ( Promega ) was added , and the luminescence was detected to determine remaining ATP amount . All of the tested compounds did not inhibit luciferase activity directly . IC50 value was obtained by using Prism software ( GraphPad Software ) . siRNAs against protein kinase genes ( obtained from Human Protein Kinome Set , Integrated DNA Technologies ) were tested on 384-well plates in Figure 3B , and resynthesized siRNAs ( Table S1 , Integrated DNA Technologies ) were tested on 96-well plates in Figure 3C , D by using Bmal1-dLuc U2OS cells as described previously [29] , [33] . In brief , for 96-well plates , the siRNA was spotted onto white solid-bottom plates , and 60 µl of Opti-MEM ( Invitrogen ) containing 0 . 4 µl of Lipofectamine 2000 ( Invitrogen ) was dispensed onto each well . After incubation at room temperature for 20 min , 60 µl of the cells in DMEM supplemented with 20% fetal bovine serum was dispensed ( 6 , 000 cells/well ) . The cells were cultured overnight , and the medium was changed to 180 µl of the culture medium . After 2 d , the medium was changed to 180 µl of the explant medium , and the plate was covered with optically clear film . The luminescence was recorded every 36 min by using the Tecan luminometer . The period parameter was obtained from the luminescence rhythm by using MultiCycle software . Bmal1-dLuc U2OS cells were transfected with siRNAs as described above and harvested just before the change to the explant medium ( i . e . , the cells were unsynchronized at the time of harvest ) . Total RNA preparation and RT-qPCR were performed as described previously [29] , [33] . The primers for qPCR are listed in Table S2 . HEK293T cells ( 1 . 25×106 cells ) were reverse transfected on 6-well plates by Lipofectamine 2000 with 1 µg each of expression vectors for C-terminally 3XFlag-tagged clock protein ( in p3XFLAG-CMV-14 , Sigma ) and N-terminally HA-tagged kinase ( in p3XFLAG-CMV-14 ) . For ERK2 , 0 . 05 µg of expression vector with 0 . 95 µg of empty vector was used because of its efficient expression . After 24 h , the cells were collected with ice-cold PBS and suspended in 100 µl of incubation buffer [50 mM Tris , 50 mM NaCl , 2 mM EDTA , 10% glycerol , 1 mM DTT , Complete Protease Inhibitor Cocktail ( Roche ) , Phosphatase Inhibitor Cocktail 1 and 2 ( Sigma ) , pH 8 . 0] . The mixture was supplemented with NP-40 ( final 1% ) and incubated on ice for 15 min , followed by centrifugation ( 16 , 000×g ) at 4°C for 10 min . A part of the resulting supernatant ( 40 µl ) was incubated with 0 . 4 µg of anti-HA antibody ( Roche , 11867423001 ) cross-linked with Dynabeads Protein G ( Invitrogen ) at 4°C for 2 h with rotation . The beads were washed twice with the incubation buffer supplemented with 1% NP-40 . The bound proteins were eluted with SDS sample buffer , separated by SDS-PAGE ( 4%–12% gradient gel ) , and analyzed by Western blotting with anti-Flag antibody ( Sigma , F1804 ) or anti-HA antibody conjugated with HRP ( Roche , 12013819001 ) . For the analysis of PER1 electrophoretic mobility-shift , the cell extracts were separated by SDS-PAGE ( 3%–8% gradient gel , Invitrogen ) and analyzed by Western blotting with anti-Flag antibody or anti-α-tubulin antibody ( Santa Cruz Biotechnology , sc-32293 ) . Protein concentration of each sample was measured by the Lowry method using DC protein assay ( BioRad ) . HEK293T cells ( 6 . 0×104 cells ) were reverse transfected on 96-well white solid-bottom plates by Lipofectamine 2000 with 40 ng each of expression vectors for C-terminally luciferase-fused PER1 ( in p3XFLAG-CMV-14 ) and N-terminally HA-tagged kinase ( in p3XFLAG-CMV-14 ) . For luciferase ( in p3XFLAG-CMV-14 ) , 2 ng of expression vector with 38 ng of empty vector was used because of its efficient expression . After 48 h , the medium was supplemented with luciferin ( final 1 mM ) and HEPES-NaOH ( pH 7 . 2; final 10 mM ) . After 1 h , cycloheximide ( final 20 µg/ml ) was added to the medium , and the plate was covered with optically clear film . The luminescence was recorded every 10 min by using the Tecan luminometer . Half-life was obtained by using Prism software ( GraphPad Software ) . Bmal1-dLuc U2OS cells were plated onto 6-well-plates ( 2 . 0×105 cells/well ) . After 2 d , the medium was replaced with 2 ml explant medium containing 0 , 3 , or 9 µM longdaysin . The plate was covered with film and kept at 36°C . At indicated time points , the cells were collected with ice-cold PBS and stored at −80°C . Then the cell pellets were homogenized in SDS sample buffer and analyzed by Western blotting with anti-PER1 antibody ( Cosmo Bio , KAL-KI044 ) or anti-α-tubulin antibody . In parallel , luminescence rhythms of the cells plated on 35 mm dishes were recorded with LumiCycle luminometer at 36°C . The per3-luc transgenic line [52] was obtained from Zebrafish International Resource Center . Hemizygote larval fish were entrained in 12 h light/12 h dark cycles from day 3 to 6 postfertilization . They were then placed in an individual well of a 96-well white solid-bottom plate with 180 µl of E3 solution ( 5 mM NaCl , 0 . 17 mM KCl , 0 . 33 mM CaCl2 , and 0 . 33 mM MgSO4 , pH 7 . 0 ) containing 0 . 5 mM luciferin , 0 . 013% Amquel Plus Instant Water Detoxifier ( Kordon brand; Novalek , Hayward , California , United States ) , and various concentrations of longdaysin ( final 0 . 1% DMSO ) . The plate was covered with optically clear film , and the luminescence was recorded every 36 min by using the Tecan luminometer at 25°C . The period parameter was obtained from the luminescence rhythm by using MultiCycle software .	Most organisms show daily rhythms in physiology , behavior , and metabolism , which may be advantageous because they anticipate environmental changes thus optimize energy metabolism . These rhythms are controlled by the circadian clock , which produces cyclic expression of thousands of output genes . More than a dozen components of the circadian clock are called clock genes , and the proteins they encode form a transcription factor network that generates rhythmic gene expression . In this study , we set out to control the function of the circadian clock and to identify new clock proteins by means of chemical tools . We tested the effects on the clock in human cells of around 120 , 000 uncharacterized compounds . Here we describe identification of a novel compound “longdaysin” that markedly slows the circadian clock both in cultured mammalian cells and in living zebrafish . By using longdaysin as a chemical probe , we found new proteins that modulate clock function . Because defects of clock function have been linked to numerous diseases , longdaysin may form the basis for therapeutic strategies directed towards circadian rhythm-related disorders , shift-work fatigue , and jet lag .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "chemical", "biology", "biochemistry", "molecular", "biology", "cell", "biology" ]	2010	High-Throughput Chemical Screen Identifies a Novel Potent Modulator of Cellular Circadian Rhythms and Reveals CKIα as a Clock Regulatory Kinase
The worldwide burden of snakebite is high , especially in remote regions with lesser accessibility to professional healthcare . Therefore , adequate first aid for snakebite is of the utmost importance . A wide range of different first aid techniques have been described in literature , and are being used in practice . This systematic review aimed to summarize the best available evidence concerning effective and feasible first aid techniques for snakebite . A systematic literature screening , performed independently by two authors in the Cochrane Library , MEDLINE and Embase resulted in 14 studies , fulfilling our predefined selection criteria , concerning first aid techniques for snakebite management . Data was extracted and the body of evidence was appraised according to the GRADE approach . The pressure immobilization technique was identified as the only evidence-based first aid technique with effectiveness on venom spread . However , additional studies suggest that proper application of this technique is not feasible for laypeople . Evidence concerning other first aid measures , such as the application of a tourniquet , suggests avoiding the use of these techniques . The practical recommendation for the treatment of snakebite in a first aid setting is to immobilize the victim , while awaiting the emergency services . However , given the low to very low quality of the data collected , high quality randomized controlled trials concerning the efficacy and feasibility of different variations of the pressure immobilization technique are warranted . Venomous snakes occur worldwide , with the exception of a few remote islands , regions of high altitude and the arctic regions [1] . Not surprisingly , ophidiophobia , or fear of snakes , is commonly reported [2] . It has also been demonstrated that humans are able to detect snakes faster than other , less harmful stimuli , suggesting the presence of an internal , evolutionary conserved warning system [3 , 4] . Despite this , snakebites occur frequently , with a global estimate of 421 , 000 to 1 , 842 , 000 cases of snake envenomation and 20 , 000 to 94 , 000 deaths each year [5] . The prevalence is especially high in the tropical regions of South and Southeast Asia , Latin America and sub-Saharan Africa , with estimates of 13 . 33 , 12 . 59 and 11 . 11 cases of snakebite/100 , 000 inhabitants , respectively . However , the accuracy of these numbers has been questioned [6 , 7] . Furthermore , studies in which data was collected through household surveys instead of official records suggested that the actual incidence of snakebite might be even higher , as many snake bitten subjects fail to present themselves to healthcare centers due to remoteness or a preference for traditional healers [7] . Snakebite victims that survive their encounter with a snake often also suffer from permanent disability . Several snake venoms , such as those from vipers and some cobra species , induce local necrosis , which can lead to amputations [8] , further increasing the estimated burden of snakebite [9] . Different studies have shown that people living in rural areas are at higher risk of encountering snakebite than people living in urban areas [7 , 10 , 11] . This might be due to a higher presence of snakes in rural areas , but also due to occupational hazards , as many people living in rural areas are occupied in agriculture , which has been shown to be a risk factor for snakebite [10–13] . Furthermore , snakebite victims are often adult males in the professionally active age range [10–12] . Therefore , snakebite is considered to be a condition with a high economic impact in an economically vulnerable population [14] . The high burden of snakebite and the fact that snakebite mostly occurs in rural areas , with less accessibility to professional health care and therefore rapid antivenom therapy , illustrate that adequate first aid treatments are of the utmost importance for achieving a positive outcome on both mortality and morbidity after a snakebite . In literature , many different techniques , and a combination thereof , are claimed to be effective for the treatment of snakebite [15 , 16] . These include techniques suggested to deactivate the venom , such as the application of electroshocks , cryotherapy or the use of traditional medicine and concoctions , a collection of practices where mixtures of herbs , oils and other products are being ingested or applied to the bite wound . Furthermore , techniques which are supposed to remove venom from the bite wound include suction of the wound , by mouth or specialized suction devices , incision/excision of the bite wound , irrigation of the bite wound , or the use of “snake stones” , which are believed to absorb the poison out of the wound . Methods proposed to limit the spread of the venom in the body include application of a tourniquet , which completely blocks the blood flow to the bitten limb , and the pressure immobilization technique . The latter technique involves application of a pressure bandage at sufficiently high pressures to block lymphatic flow , but without actually applying a tourniquet , together with immobilization of the bitten limb [17] . This systematic review is the first in its kind to synthesize the available evidence concerning suggested first aid measures for snakebite , thus facilitating evidence-based decision making during the development of snakebite first aid guidelines for laypeople . For this , the following PICO question was formulated: In people with snakebites ( P ) , is a certain first aid intervention ( I ) , compared to another first aid intervention or no intervention ( C ) , effective and feasible for laypeople as a first aid treatment to increase survival , tissue healing , functional recovery , pain , complications , time to resumption of usual activity , restoration to the pre-exposure condition , time to resolution of the symptoms or other health outcome measures ( including adverse effects ) ( O ) ? The following databases were searched for relevant studies from their date of inception to March 10 , 2016: The Cochrane Library for clinical trials and systematic reviews , MEDLINE ( using the PubMed interface ) for systematic reviews , experimental and observational studies and Embase ( via the Embase . com interface ) for systematic reviews , experimental and observational studies , using the search strategies described in S1 File . Titles and abstracts of retrieved articles were scanned , and for relevant articles the full-texts were obtained and studied . Studies that did not meet the predefined selection criteria , as described below , were excluded . The reference lists of included studies and also the first 20 similar articles in PubMed were screened for other relevant publications . The searches and study selection procedures were performed independently by two reviewers ( BA and VB ) . Any discrepancy between the reviewers was resolved by consensus or by consulting a third reviewer ( EDB ) . For the population ( P ) , studies concerning people with snakebites or healthy volunteers with “mock” snakebites were included . The interventions ( I ) that were included in this systematic review were interventions for the first aid management of snakebites that can be applied by laypeople without medical background . We excluded interventions for the management of snakebites that are not feasible to be performed in a first aid setting where laypeople are the first aid providers . We selected studies that compared ( C ) the interventions to any other first aid intervention or no intervention . Concerning the outcomes ( O ) , we included ( 1 ) survival , functional recovery , pain , complications , time to resumption of usual activity , restoration of the pre-exposure condition , time to resolution of symptoms or other health outcome measures ( including adverse effects ) for studies involving snakebite victims , ( 2 ) spread of mock venom for studies investigating the efficacy of pressure immobilization and ( 3 ) quality of the bandage applied and tension generated for studies investigating the feasibility of pressure immobilization . The following experimental study designs were included: ( quasi or non- ) randomized controlled trials , controlled before and after studies or controlled interrupted time series , if the data were available . For studies concerning the feasibility of first aid interventions , non-controlled before and after studies were also included , since this is typically measured with that type of study design . Observational studies of the following types were also included: cohort and case-control study , controlled before and after study or controlled interrupted time series , if the data were available . We excluded observational studies if the intervention was already studied in experimental studies , letters , comments , narrative reviews , case reports , cross-sectional studies , animal studies , ex vivo or in vitro studies , conference abstracts unless no other relevant data was available , studies reporting no quantitative data , studies reporting only means , but no standard deviations ( SDs ) , effect sizes , p-values . Only studies reported in English were selected . Data concerning study design , study population , outcome measures ( expressed as mean difference , odds ratio or risk ratio ) and study quality were independently extracted from the included studies by two reviewers ( BA and VB ) using an in advance prepared form . Any discrepancy between the reviewers was resolved by consensus . Data and p-values were extracted directly from the publications , unless it is stated that these were calculated from raw data available using the Review Manager software [20] . Outcomes from the selected studies without raw data or statement of significance were not extracted . Data are represented as mean±standard deviation ( SD ) or relative risk ( RR ) with 95% CI ( confidence interval ) , unless otherwise stated . The overall quality of “the body of evidence” was determined using the GRADE approach [21] . Evidence from experimental studies started with an initial “high” quality level , and evidence from observational studies with an initial “low” quality level . The evidence was then assessed for limitations in 5 domains , for which the quality of evidence could be downgraded , namely limitations in study design , indirectness , imprecision , inconsistency and reporting bias . Limitations in study design were assessed at the level of the individual study using the items listed by GRADE . The overall quality was assessed separately for ( 1 ) experimental studies concerning efficacy of pressure immobilization , ( 2 ) experimental studies concerning feasibility of the application of pressure immobilization to be performed by laypeople and ( 3 ) observational studies concerning other first aid techniques ( tourniquet application , suction , traditional medicine , snake stones , incision of the bite wound ) . A search in The Cochrane Library , MEDLINE and Embase resulted in a total of 3 , 893 retrieved references ( Fig 1 ) . After removing 956 ( BA ) and 1 , 132 ( VB ) duplicates , the titles and abstracts of 2 , 928 ( BA ) and 2 , 761 ( VB ) records were screened on relevance regarding the PICO question . For 81 ( BA ) and 101 ( VB ) publications , a full-text was obtained and eligibility was assessed , resulting in 12 articles that matched the predefined selection criteria . The majority of publications excluded had an inappropriate study design . A search in the references and similar articles lists of these publications resulted in 2 additional publications matching the selection criteria , leading to a total of 14 included articles . An overview of the in- and excluded studies can be found in Table 1 and S2 Table , respectively . Of the 14 included articles , 7 were experimental [23–29] and 7 were observational studies [13 , 16 , 22 , 30–33] . 4 experimental studies evaluated the efficacy of a first aid treatment , i . e . variants of the pressure immobilization technique , on simulated snake bites [23–26] , while 3 others examined the feasibility of pressure immobilization to be performed by laypeople [27–29] . The observational studies all examined the outcomes of different applied first aid procedures in snakebite patients [13 , 16 , 22 , 30–33] . An overview of the study characteristics of the included studies is shown in Table 1 . Two controlled trials performed by Anker et al . examined the effectiveness of two different pressure immobilization techniques on the spread of a subcutaneously injected radioactive “mock venom” , consisting of either saline , containing 0 . 2 or 0 . 3 μCi/kg Na131I [23] , or saline , containing 0 . 2 μCi/kg 125I-labeled porcine insulin [24] , in the leg of healthy volunteers . Both studies compared the effectiveness of a crepe or elastic compression bandage ( 55±5 mmHg ) and a wooden splint on the injected limb or a cloth pad bound firmly over the site of injection ( >70 mmHg ) with 2 broad bandages to no first aid treatment . A study by Tun Pe et al . used a variant of the pressure cloth studied by Anker et al . , namely a rubber pad instead of a cloth pad bound firmly at the site of injection , together with immobilization of the injected limb by splinting [26] . The effectiveness of this first aid technique was assessed on the spread of a mock venom , consisting of 12 or 20 μCi Na131I , injected subcutaneously in the leg of healthy volunteers and compared to no first aid treatment . Howarth et al . used a within subjects design to study the effect of pressure immobilization , a crepe bandage of 50–70 mmHg with a splint applied to one upper and one lower limb , on the spread of a mock venom , consisting of 0 . 1 ml 10 MBq 99mtechnetium antimony sulfur colloid , subcutaneously injected in both upper and lower limbs of healthy volunteers [25] . The spread of the mock venom was compared between the treated limbs and the corresponding untreated limbs , both in rest and during walking . Norris et al . compared the correct application of a pressure bandage on the own or the investigator’s upper and lower limbs between a group of health care workers and a group of laypeople after receiving only written instructions [28] . In contrast , Canale et al . studied whether intense training influenced the correct application of an elastic bandage on a simulated snakebite victim’s lower limb by a test group consisting of both healthcare workers and laypeople , compared to before training , when no instructions were given [27] . Simpson et al . performed a randomized controlled trial comparing the correct application of an elastic bandage on the upper or lower limb of a simulated snakebite victim by two groups of volunteers receiving either intense training or only written instructions [29] . Furthermore , the retention of the acquired skills was measured in the group receiving intense training immediately versus three days after receiving the training . The observational studies included are cohort studies describing outcomes following the application of different first aid techniques in snakebite victims presenting at health care facilities after being bitten by South American rattle snakes [22] , vipers [30] , lance-headed vipers [31] , Russell’s vipers [32] , Chinese cobras [33] or unspecified snakes [13 , 16] . The use of a tourniquet was studied in all 7 observational studies , while 2 studies also looked at the effects of incision [16 , 30] , 2 studies examined the use of “snake stones” and traditional medicine and concoctions [13 , 16] and one study examined the use of suction [16] . A structured synthesis of the findings from the included studies can be found in S3 Table , a narrative overview is given below . An overview of the limitations in study design for the included studies individually , according to the GRADE approach is shown in Table 2 . An overall assessment of the body of evidence is further elaborated below . This study aimed to summarize the best available evidence concerning the effectiveness and feasibility of first aid treatments for snakebites . A broad search strategy was used , to identify a wide range of first aid measures that are claimed to be effective . However , for several suggested first aid treatments , such as electroshock therapy or cryotherapy , no studies were found that met the predefined selection criteria . A total of 7 experimental and 7 observational studies that addressed the PICO question were identified . The experimental studies all concern the pressure immobilization technique , based on the use of a crepe or elastic bandage . This technique received a lot of attention in Australia , and is being recommended in official Australian first aid guidelines [34 , 35] . However , the effectiveness of this technique has only been demonstrated in animal models [17 , 36] , with evidence from human studies being limited to case reports [37 , 38] . Three of the studies on pressure immobilization efficacy meeting the selection criteria of this systematic review , favor a modified version of this technique , involving a localized cloth or rubber pad , firmly pressed on the site of the bite wound , with or without splinting , instead of a crepe or elastic bandage compressing the whole limb [23 , 24 , 26] . One study suggests that keeping a person still delays the spread of the venom [25] . However , the feasibility of correctly applying pressure immobilization using an elastic bandage is questionable , especially regarding the tension generated [27–29] . Intense training is warranted , but even then , retention is low [27 , 29] . Studies on the feasibility of applying a firmly strapped cloth or rubber pad are unavailable . Furthermore , it needs to be taken into account that the pressure immobilization technique might not be appropriate for any type of snake venom . The technique has a theoretical basis for limiting the spread of neurotoxic venoms , such as those produced by elapids , but less for necrotic venoms , such as those produced by vipers [39] . However , no controlled studies have been performed in real-life snakebite patients yet , leaving this controversy unresolved . Studies concerning other first aid techniques were all observational . For tourniquets , most outcomes that were studied show no benefit of using a tourniquet in snakebite victims [13 , 16 , 22 , 30–33] . Moreover , the few outcomes that do show significant differences between tourniquet treated snakebite victims and victims receiving no tourniquet or no first aid show harmful effects of tourniquets on local symptoms [30 , 31 , 33] . For two outcomes , inconclusive evidence was found [13 , 16 , 22] . The bulk of evidence thus indicates that tourniquet use is not indicated for the treatment of snakebite . The evidence available for other first aid measures is scarce , with evidence for the use of incisions , snake stones , traditional medicine , concoctions and suction being extracted from only 3 studies [13 , 16 , 30] . Concerning the use of incisions , 3 outcomes were not significantly different between snakebite victims treated with incisions and those treated without incisions or receiving no first aid [16 , 30] . One outcome , duration of hospital stay , differed in favor of incision , while another , local swelling , differed in favor of no first aid treatment . For snake stones , 3 outcomes did not differ between subjects treated with snake stones or those treated without snake stones or who received no first aid [13 , 16] . One other outcome , the amount of antivenom required , had inconclusive results . The use of traditional medicine or concoctions had no statistically significant effect for 2 outcomes , an inconclusive effect for 1 outcome , the amount of antivenom required , and a harmful effect for another outcome , the incidence of death or disability [13 , 16] . Finally , suction was shown to be ineffective for the treatment of snakebite on 3 outcomes [16] . In addition , there is a potential threat for the caregiver who could be exposed to the poison when performing oral suction . In conclusion , these alternative methods for the treatment of snakebite are most likely not beneficial and perhaps even harmful . Most of these management strategies are applied by traditional healers , who might be preferred over healthcare professionals in first instance . The use of this type of ineffective pre-hospital care might cause a delay in the presentation of the snakebite victim to healthcare facilities , further increasing the detrimental impact of the snakebite on morbidity and mortality . Habib et al . previously showed a 1% increase in odds of dying from snakebite for every 1 h delay in healthcare facility presentation in a case-control study of snakebite victims in north-eastern Nigeria [40] . Evidence concerning the time to application for specific first aid measures and their influence on the timing of presentation at a healthcare facility is currently unavailable . This systematic review has some limitations . The fact that the best available evidence was collected for different first aid techniques led to the inclusion of studies with differing study designs , which implies substantial heterogeneity between studies . The study populations , interventions and outcomes assessed differed between studies , thus complicating the comparison between different first aid techniques . Therefore , it was both unfeasible and unwarranted to perform meta-analyses . Secondly , the sample sizes studied were small , limiting the generalizability of the conclusions made . Thirdly , substantial bias was present in the included studies , as discussed in the quality of evidence paragraph of the results section . Fourthly , the indirectness of the experimental studies on the efficacy of pressure immobilization , all performed using mock venoms , further limits our confidence in the reported results . Thus , the overall quality of the available evidence was low to very low , according to the GRADE approach [21] . Studies on the efficacy and feasibility of pressure immobilization in real-life snakebite victims are crucial to draw trustworthy conclusions concerning this technique . The evidence collected in this systematic review has been used for the development of a first aid guideline for sub-Saharan Africa [41] , according to the principles of Evidence-Based Practice [19] , which is being updated in 2016 . No new evidence , concerning first aid treatments for snake bites , could be identified in the 2016 update . This summary of best available evidence has been presented to a panel of first aid experts , who have made a recommendation , based on the available evidence , taking into account the needs and preferences of the target group , i . e . African laypeople encountering a case of snakebite . The only first aid measure that is supported with evidence is pressure immobilization , but it appeared difficult to apply this technique correctly . However , keeping a mock-bitten victim still had a beneficial effect on the spread of the venom [25] . In addition , the inhibition of venom spread by pressure immobilization might not be virtuous for venoms which induce localized necrosis and edema , such as those from e . g . vipers , which occur frequently in sub-Saharan Africa [1] . Unfortunately , no studies on the use of the pressure immobilization technique have been performed in sub-Saharan Africa yet . Therefore , the final recommendation is as follows: “Make sure no additional bites are encountered . Try to identify the snake , but do not try to catch it . Reassure the victim , tell him/her to lie down and move as little as possible . Contact the emergency services immediately . Remove any jewelry , watch or tight clothing . Immobilize the affected limb and control the victim’s vital parameters until the emergency services arrive . ” This systematic review on first aid measures for the treatment of snakebite by lay first aid providers , has revealed that none of the in the literature suggested measures is proven to be both effective and feasible for the treatment of snakebite . The pressure immobilization technique was found to be effective but not feasible for laypeople . Therefore , evidence supporting a first aid guideline used in daily practice is limited to supportive therapy until professional help arrives . However , given the low quality of the evidence found , high quality studies concerning the efficacy and feasibility of different forms of pressure immobilization are warranted .	The Belgian Red Cross-Flanders develops first aid guidelines that specifically target laypeople . In the context of updating the first aid guidelines for sub-Saharan Africa , we aimed to summarize the best available evidence for the treatment of snakebite , feasible for laypeople . Of the numerous first aid measures supported in literature and used in practice , we could only find evidence concerning effectiveness for the pressure immobilization technique on the spread of snake venom , which involves application of a firm pressure bandage on the bitten limb , together with immobilization of the limb . However , studies concerning its feasibility suggest this technique is difficult for laypeople to apply correctly . Keeping the limb immobilized on the other hand had a beneficial effect on the spread of the venom . However , given the low to very low quality of the evidence , high quality trials concerning the effectiveness and feasibility of different variations of the pressure immobilization technique are needed .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion", "Conclusion" ]	[ "traditional", "medicine", "medicine", "and", "health", "sciences", "radiochemistry", "toxins", "pathology", "and", "laboratory", "medicine", "tropical", "diseases", "vertebrates", "animals", "toxicology", "toxic", "agents", "observational", "studies", "research", "design", "radioactivity", "reptiles", "complementary", "and", "alternative", "medicine", "neglected", "tropical", "diseases", "snakebite", "research", "and", "analysis", "methods", "venoms", "vipers", "chemistry", "research", "assessment", "snakes", "physics", "nuclear", "physics", "squamates", "systematic", "reviews", "biology", "and", "life", "sciences", "physical", "sciences", "amniotes", "organisms" ]	2016	The Treatment of Snake Bites in a First Aid Setting: A Systematic Review
The possibility of using computer simulation and mathematical modeling to gain insight into biological and other complex systems is receiving increased attention . However , it is as yet unclear to what extent these techniques will provide useful biological insights or even what the best approach is . Epstein–Barr virus ( EBV ) provides a good candidate to address these issues . It persistently infects most humans and is associated with several important diseases . In addition , a detailed biological model has been developed that provides an intricate understanding of EBV infection in the naturally infected human host and accounts for most of the virus' diverse and peculiar properties . We have developed an agent-based computer model/simulation ( PathSim , Pathogen Simulation ) of this biological model . The simulation is performed on a virtual grid that represents the anatomy of the tonsils of the nasopharyngeal cavity ( Waldeyer ring ) and the peripheral circulation—the sites of EBV infection and persistence . The simulation is presented via a user friendly visual interface and reproduces quantitative and qualitative aspects of acute and persistent EBV infection . The simulation also had predictive power in validation experiments involving certain aspects of viral infection dynamics . Moreover , it allows us to identify switch points in the infection process that direct the disease course towards the end points of persistence , clearance , or death . Lastly , we were able to identify parameter sets that reproduced aspects of EBV-associated diseases . These investigations indicate that such simulations , combined with laboratory and clinical studies and animal models , will provide a powerful approach to investigating and controlling EBV infection , including the design of targeted anti-viral therapies . Computer simulation and mathematical modeling are receiving increased attention as alternative approaches for providing insight into biological and other complex systems [1] . An important potential area of application is microbial pathogenesis , particularly in cases of human diseases for which applicable animal models are lacking . To date , most simulations of viral pathogenesis have tended to focus on HIV [2–7] , and employ mathematical models based on differential equations . None have addressed the issue of acute infection by the pathogenic human herpes virus Epstein–Barr virus ( EBV ) and its resolution into lifetime persistence . With the ever-increasing power of computers to simulate larger and more complex systems , the possibility arises of creating an in silico virtual environment in which to study infection . We have used EBV to investigate the utility of this approach . EBV is a human pathogen , associated with neoplastic disease , that is a paradigm for understanding persistent infection in vivo and for which a readily applicable animal model is lacking ( reviewed in [8 , 9] ) . Equally important is that EBV infection occurs in the lymphoid system , which makes it relatively tractable for experimental analysis and has allowed the construction of a biological model of viral persistence that accounts for most of the unique and peculiar properties of the virus [10 , 11] . We are therefore in a position to map this biological model onto a computer simulation and then ask how accurately it represents EBV infection ( i . e . , use our knowledge of EBV to test the validity of the simulation ) and whether the matching of biological observation and simulation output provides novel insights into the mechanism of EBV infection . Specifically , we can ask if it is possible to identify critical switch points in the course of the disease where small changes in behavior have dramatic effects on the outcome . Examples of this would be the switch from clinically silent to clinically apparent infection and from benign persistence to fatal infection ( as occurs in fatal acute EBV infection and the disease X-linked lymphoproliferative [12] , for example ) , or to clearance of the virus . Indeed , is clearance ever possible , or do all infections lead inevitably to either persistence or death ? Such an analysis would be invaluable . Not only would it provide insight into the host–virus balance that allows persistent infection , but it would also reveal the feasibility and best approaches for developing therapeutic interventions to diminish the clinical symptoms of acute infection , prevent fatal infection , and/or clear the virus . A diagrammatic version of the biological model is presented in Figure 1 . EBV enters through the mucosal surface of the Waldeyer ring , which consists of the nasopharyngeal tonsil ( adenoids ) , the paired tubal tonsils , the paired palatine tonsils , and the lingual tonsil arranged in a circular orientation around the walls of the throat . Here EBV infects and is amplified in epithelium . It then infects naïve B cells in the underlying lymphoid tissue . The components of the ring are all equally infected by the virus [13] . EBV uses a series of distinct latent gene transcription programs , which mimic a normal B cell response to antigen , to drive the differentiation of the newly infected B cells . During this stage , the infected cells are vulnerable to attack by cytotoxic T cells ( CTLs ) [14] . Eventually , the latently infected B cells enter the peripheral circulation , the site of viral persistence , as resting memory cells that express no viral proteins [15] and so are invisible to the immune response . The latently infected memory cells circulate between the periphery and the lymphoid tissue [13] . When they return to the Waldeyer ring they are occasionally triggered to terminally differentiate into plasma cells . This is the signal for the virus to begin replication [16] , making the cells vulnerable to CTL attack again [14] . Newly released virions may infect new B cells or be shed into saliva to infect new hosts , but are also the target of neutralizing antibody . Primary EBV infection in adults and adolescents is usually symptomatic and referred to as infectious mononucleosis ( AIM ) . It is associated with an initial acute phase in which a large fraction ( up to 50% ) of circulating memory B cells may be latently infected [17] . This induces the broad T lymphocyte immune response characteristic of acute EBV infection . Curiously , primary infection early in life is usually asymptomatic . In immunocompetent hosts , infection resolves over a period of months into a lifelong persistent phase in which ∼1 in 105 B cells carry the virus [18] . Exactly how persistent infection is sustained is unclear . For example , once persistence is established , it is unknown if the pool of latently infected memory B cells is self-perpetuating or if a low level of new infection is necessary to maintain it . Indeed , we do not know for sure that the pool of latently infected B cells in the peripheral memory compartment is essential for lifetime persistence . It is even unclear whether the virus actually establishes a steady state during persistence or continues to decay , albeit at an ever slower rate [17] . In the current study we describe the creation and testing of a computer simulation ( PathSim ) that recapitulates essential features of EBV infection . The simulation has predictive power and has utility for experiment design and understanding EBV infection . One practical limitation of available simulation and modeling approaches has been their inaccessibility to the working biologist . This is often due to the use of relatively unfamiliar computer interfaces and output formats . To address these issues , we have presented the simulation via a user-friendly visual interface on a standard computer monitor . This allows the simulation to be launched and output to be accessed and analyzed in a visual way that is simple and easily comprehensible to the non-specialist . The computer model ( PathSim ) is a representation of the biological model described in the Introduction . A schematic version of both is shown in Figure 1 . To simulate EBV infection , we created a virtual environment consisting of a grid that describes a biologically meaningful topography , in this case the Waldeyer ring ( five tonsils and adenoids ) and the peripheral circulation , which are the main sites of EBV infection and persistence . The tonsils and adenoids were composed of solid hexagonal base units representing surface epithelium , lymphoid tissue , and a single germinal center/follicle ( Figure 2A–2C; Video S1 ) . Each hexagonal unit had one high endothelial venule ( HEV ) entry point from the peripheral blood and one exit point into the lymphatic system ( Figure 2A ) . Discrete agents ( cells or viruses ) reside at the nodes ( red boxes ) of the 3-D grid ( white lines ) . There they can interact with other agents and move to neighboring nodes . Agents are assessed at regular , specified time intervals as they move and interact upon the grid . Virtual cells were allowed to leave the Waldeyer ring via draining lymphatics and return via the peripheral blood and HEVs ( Figure 2A and 2B; Video S1 ) as in normal mucosal lymphoid tissue [19] . A brief summary list of the agents employed in our simulation , and their properties and interactions , is given in Table 1 . In this report we refer to actual B cells as , for example , “B cells” , “latently infected B cells” , or “lytically infected B cells” , and their virtual representations as “virtual B cells” , “BLats” , or “BLyts” . Similarly , we refer to actual virus as virions and their virtual counterparts as virtual virus or Vir . A full description of the simulation , including a complete list of agents , rules , the default parameters that produce the output described below , and a preliminary survey of the extended parameter space is presented in M . Shapiro , K . Duca , E . Delgado-Eckert , V . Hadinoto , A . Jarrah , et al . ( 2007 ) A virtual look at Epstein-Barr virus infection: simulation mechanism ( unpublished data ) . Here , we will first present a description of how the virtual environment was visualized and then focus on a comparison of simulation output with the known biological behavior of the virus . Simulation runs were accessed through an information-rich virtual environment ( IRVE ) ( Figures 2 and 3; Videos S1 and S2 ) , which was invoked through a Web interface . This provided a visually familiar , straightforward context for immediate comprehension of the spatial behavior of the system [20] . It also allowed specification of parameters , run management , and ready access to data output and analysis . Figure 3 demonstrates how the time course of infection may be visualized . Usually the simulation was initialized by a uniform distribution of Vir over the entire surface of the Waldeyer ring , thereby seeding infection uniformly . However , in the simulation shown in Figure 3A , virtual EBV was uniformly deposited only on the lingual tonsil . Figure 3B–3D shows the gradual spread of virtual infection ( intensity of red color indicating the level of free Vir ) to the adjacent tonsils . It can be seen in this case that the infection spreads uniformly to all the tonsils at once , implying that it was spreading via BLats returning from the blood compartment and reactivating to become BLyts , rather than spreading within the ring . Examples of infectious spread between and within the tonsils can also be seen in Video S2 . In this paper we present a comprehensive model of EBV infection that effectively simulates the overall dynamics of acute and persistent infection . The fact that this simulation can be tuned to produce the course of EBV infection suggests that it models the basic processes of this disease . To achieve this , we have created a readily accessible , virtual environment that appears to capture most of the salient features of the lymphoid system necessary to model EBV infection . Achieving infection dynamics that reflect an acute infection followed by recovery to long-term low-level persistent infection seems to require access of the virus to a blood compartment where it is shielded from immunosurveillance . Because we cannot perform a comprehensive parameter search ( due to the very large parameter space involved ) , we cannot unequivocally state that the blood compartment is essential . What is clear though , is that persistence is a very robust feature in the presence of a blood compartment , and that we could not achieve an infection process that even remotely resembles typical persistent EBV infection in its absence . The areas in which the simulation most closely follows known biology are summarized in Table 2 and include the peak time of infection , 33–38 d , compared to the incubation time for AIM of 35–50 d [21] . This predicts that patients become sick and enter the clinic at or shortly after peak infection in the peripheral blood , a prediction confirmed by our patient studies , where the numbers of infected B cells in the periphery always decline after the first visit [17] . An important feature of a simulation is its predictive power . Our analysis predicted that access to the peripheral memory compartment is essential for long-term persistence . This is consistent with recent studies on patients with hyper-IgM syndrome [31] . Although these individuals lack classical memory cells , they can be infected by EBV; however , they cannot sustain persistent infection and the virus becomes undetectable . Unfortunately , those studies did not include a sufficiently detailed time course to see if time to virus loss coincided with the simulation prediction of 1–2 mo . Another area where the simulation demonstrated its predictive power was in the dynamics of viral replication . In the simulation it was unexpectedly observed that the level of Vir production plateaued long before BLats , predicting that the levels of virus shedding , unlike latently infected cells , will have leveled off by the time AIM patients arrive in the clinic . This prediction , which contradicted the common wisdom that virus shedding should be high and decline rapidly in AIM patients , was subsequently confirmed experimentally ( V . Hadinoto , M . Shapiro , T . Greenough , J . Sullivan , K . Luzuriaga , D . Thorley-Lawson ( 2007 ) On the dynamics of acute EBV infection and the pathogenesis of infectious mononucleosis ( unpublished data ) and see also [22 , 23] ) . The simulation also quite accurately reproduces the relatively large variation in virus production over time , compared to the stability of B latent . This difference is likely a consequence of stochasticity ( random variation ) having a relatively larger impact on virus production . This is because the number of B cells replicating the virus at any given time is very small , both in reality and the simulation , compared to the number of infected B cells , but the number of virions they release when they do burst is very large . This difference may reflect on the biological requirements for persistence of the virus since a transient loss in virus production due to stochasticity can readily be overcome through recruitment from the pool of B latents . However , a transient loss of B latents would mean clearance of the virus . Hence , close regulation of B latent but not virion levels is necessary to ensure persistent infection . Although there is now a growing consensus that EBV infects normal epithelial cells in vivo [27–29] , the biological significance of this infection remains unclear . The available evidence suggests that epithelial cell infection may not be required for long-term persistence [25 , 26] , and this is also seen in the simulation . The alternate proposal is that epithelial infection might play an important role in amplifying the virus , during ingress and/or egress , as an intermediary step between B cells and saliva . This is based on the observation that the virus can replicate aggressively in primary epithelial cells in vivo [30] . In the simulation , epithelial amplification had no significant effect on the ability of Vir to establish persistence . This predicts that epithelial amplification does not play a critical role in entry of the virus , but leaves open the possibility that it may be important for increasing the infectious dose present in saliva for more efficient infection of new hosts . The simulation is less accurate in the precise quantitation of the dynamics . Virtual acute infection resolves significantly more slowly and persistence is at a higher level than in a real infection . In addition , virtual persistent infection demonstrates clear evidence of oscillations in the levels of infected cells that have not been detected in a real infection . The most likely explanation for these discrepancies is that we have not yet implemented T cell memory . Thus , as the levels of virtual infected cells drops , the immune response weakens , allowing Vir to rebound while a new supply of virtual CTLs is generated . Immunological memory would allow a more sustained T cell response that would produce a more rapid decline of infected cells , lower levels of sustained persistence , and tend to flatten out oscillatory behavior , thus making the simulation more quantitatively accurate . This is one of the features that will be incorporated into the next version of our simulation . It remains to be determined what additional features need to be implemented to sharpen the model and also whether and to what extent the level of representation we have chosen is necessary for faithful representation of EBV infection . Our simulation of the Waldeyer ring and the peripheral circulation was constructed with the intent of modeling EBV infection . Conversely , our analysis can be thought of as the use of EBV to validate the accuracy of our Waldeyer ring/peripheral circulation simulation and to evaluate whether it can be applied to other pathogens . Of particular interest is the mouse gamma herpesvirus MHV68 [32 , 33] . The applicability of MHV68 as a model for EBV is controversial . Although it also persists in memory B cells [34] , it appears to lack the sophisticated and complicated latency states that EBV uses to access this compartment . However , one of the simplifications in our simulation is that the details of these different latency states and their transitions are all encompassed within a single concept , the BLat . We have also assumed a time line whereby a newly infected BLat becomes activated and CTL sensitive , migrates to the follicle , and exits into the circulation , where it is no longer seen by our virtual CTLs . In essence , we have generalized the process by which the virus proceeds from free virion to the site of persistence in such a way that it may be applicable to both EBV and MHV68 . Thus , we might expect that the overall dynamics of infection may be similar even though detailed biology may vary . As a first step to test if this concept had value , we performed an analysis based on studies with MHV68 where it was observed that the levels of infected B cells at persistence were unaffected by the absolute amount of input virus at the time of infection [35] . When this parameter was varied in the simulation , we saw the same outcome . This preliminary attempt raises the possibility that the mouse virus may be useful for examining quantitative aspects of EBV infection dynamics . The last area we wished to investigate was whether we could identify biologically meaningful “switch” points , i . e . , places in time and space where relatively small changes in critical parameters dramatically affect outcome , for example , switching from persistence to clearance to death . We have observed one such switch point—reactivation of BLats upon return to the Waldeyer ring—that rapidly switches the infection process from persistence to death . How this might relate to fatal EBV infection , X-linked lymphoproliferative disease , is uncertain . However , viral production is a function both of how many B cells initiate reactivation and how efficiently they complete the process . We believe that most such cells are killed by the immune response before they release virus [16] , so defects in the immune response could allow more cells to complete the viral replication process and give the same fatal outcome . The ability to find such conditions for switch points could be very useful in the long term for identifying places in the infection process where the virus might be optimally vulnerable to drug intervention . The easiest place to target EBV is during viral replication; however , it is currently unclear whether viral replication and infection are required for persistence . It may be that simply turning off viral replication after persistence is established fails to eliminate the virus because the absence of new cells entering the pool through infection is counterbalanced by the failure of infected cells to disappear through reactivation of the virus . If , however , a drug allowed abortive reactivation , then cells would die without producing infectious virus and new infection would be prevented . This models the situation that would arise with a highly effective drug or viral mutant that blocked a critical stage in virion production ( e . g . , viral DNA synthesis or packaging ) , so that reactivation caused cell death without release of infectious virus . A similar effect could be expected with a drug or vaccine that effectively blocked all new infection . This is another case in which studies with the mouse virus , where non-replicative mutants can be produced and tested , may be informative as to whether and to what extent infection is required to sustain the pool of latently infected B cells and persistence . The simulation could then be used to predict how effective an anti-viral that blocked replication , or a vaccine that induced neutralizing antibodies , would need to be at reducing new infection in order to cause EBV to be lost from the memory pool ( for a more detailed discussion of this issue see M . Shapiro , K . Duca , E . Delgado-Eckert , V . Hadinoto , A . Jarrah , et al . ( 2007 ) A virtual look at Epstein-Barr virus infection: simulation mechanism ( unpublished data ) ) . Most modeling of virus infection to date has tended to focus on HIV and use differential equations [2–7] . One such study involved EBV infection [36] , but to our knowledge none outside of our group has addressed the issue studied here of acute EBV infection and how it resolves into lifetime persistence . In preliminary studies of our own , modeling EBV infection with differential equations that incorporate features common to the HIV models , with parameters physiologically reasonable for EBV did not produce credible dynamics of infection ( K . Duca , unpublished observations ) . Although we do not exclude the possibility that such models may be useful for simulating EBV , we took an agent-based approach because it is intuitively more attractive to biologists . Such models are increasingly being recognized as an effective alternative way to simulate biological processes [37–39] and have several advantages . The main advantage is that the “agent” paradigm complies by definition with the discrete and finite character of biological structures and entities such as organs , cells , and pathogens . This makes it more accurate , from the point of view of scientific modeling . It is also less abstract since the simulated objects , processes , and interactions usually have a straightforward biological interpretation and the spatial structure of the anatomy can be modeled meticulously . The stochasticity inherent to chemical and biological processes can be incorporated in a natural way . Lastly , it is generally much easier to incorporate qualitative or semi-quantiative information into rule sets for discrete models than it is for such data to be converted to accurate rate equations . The major drawback to agent-based models is that there is currently no mathematical theory that allows for rigorous analysis of their dynamics . Currently , one simply runs such simulations many times and performs statistical analyses to assess their likely behaviors . Developing such a mathematical theory remains an important goal in the field . In summary , we have described a new computer simulation of EBV infection that captures many of the salient features of acute and persistent infection . We believe that this approach , combined with mouse modeling ( MHV68 ) and EBV studies in patients and healthy carriers , will allow us to develop a more profound understanding of the mechanism of viral persistence and how such infections might be treated and ultimately cleared . Details of the AIM patient populations tested have been published previously [17] . Adolescents ( ages 17–24 ) presenting to the clinic at the University of Massachusetts at Amherst Student Health Service ( Amherst , Massachusetts , United States ) with clinical symptoms consistent with acute infectious mononucleosis were recruited for this study . Following informed consent , blood and saliva samples were collected at presentation and periodically thereafter . Diagnosis at the time of presentation to the clinic required a positive monospot test and the presence of atypical lymphocytes [21] . Confirmation of primary Epstein–Barr infection required the detection of IgM antibodies to the EBV viral capsid antigen in patient sera [40] . These studies were approved by the Human Studies Committee at the University of Massachusetts Medical School ( Worcester , Massachusetts , United States ) and by the Tufts New England Medical Center and Tufts University Health Sciences Institutional Review Board . All blood samples were diluted 1:1 in 1x PBS . The technique for estimating the absolute number of latently infected B cells in the peripheral blood of patients and healthy carriers of the virus is a real-time PCR–based variation of our previously published technique [17] , the details of which will be published elsewhere ( V . Hadinoto , M . Shapiro , T . Greenough , J . Sullivan , K . Luzuriaga , et al . ( 2007 ) On the dynamics of acute EBV infection and the origins of infectious mononucleosis ( unpublished data ) ) . To measure the absolute levels of virus shedding in saliva , individuals were asked to rinse and gargle for a few minutes with 5 ml of water and the resultant wash processed for EBV-specific DNA PCR using the same real-time–based PCR technique . We have performed extensive studies to standardize this procedure that will be detailed elsewhere ( V . Hadinoto , M . Shapiro , T . Greenough , J . Sullivan , K . Luzuriaga , et al . ( 2007 ) On the dynamics of acute EBV infection and the origins of infectious mononucleosis ( unpublished data ) ) . In the simulation , B cells are either uninfected ( BNaïve ) , latently infected ( BLat ) , or replicating virtual virus ( BLyt ) ; we do not distinguish blast and memory B cells . In the biological model , newly infected B cells in the lymphoepithelium of the Waldeyer ring pass through different latency states , which are vulnerable to attack by cytotoxic T cells ( CTL latent ) . Subsequently , they become memory B cells that enter the peripheral circulation and become invisible to the immune response by turning off viral protein expression . In the simulation , all these latency states are captured in the form of a single entity , the BLat . In addition , the blood circulation and lymphatic system are both represented as abstract entities that only allow for transport of BNaïves and BLats around the body . Virtual T cells are restricted to the Waldeyer ring . This simplification is based on the assumption that , in the biological model , EBV-infected cells entering the peripheral circulation are normal and invisible to CTLs , because the virus is inactive , and therefore the peripheral circulation simply acts as an independent pool of and a conduit for B latent . Operationally , therefore , BLats escape TLats in the simulation simply by entering the peripheral circulation . Consequently , unlike the biological model , BLats are vulnerable to TLats whenever they reenter the lymph node . Each agent ( e . g . , Vir or a BNaïve ) has a defined life span , instructions for movement , and functions that depend on which other agents they encounter ( for example , if a Vir encounters a BNaïve , it infects it with some defined probability ) . The agents , rules , and parameters used are based on known biology wherever possible with simplifications ( see above ) where deemed appropriate . A brief description and discussion of the agents and their rules is given in Table 1 . A detailed listing is provided in M . Shapiro , K . Duca , E . Delgado-Eckert , V . Hadinoto , A . Jarrah , et al . ( 2007 ) A virtual look at Epstein-Barr virus infection: simulation mechanism ( unpublished data ) . At each time point ( 6 min of real time ) , every agent is evaluated and appropriate actions are initiated . The simulation is invoked through a Web interface ( IRVE; see movies linked to Figure 2 , and [20] ) that allows a straightforward visual , familiar , and scalable context for access to parameter specification , run management , data output , and analysis . This has the additional advantage that it readily allows comprehension of the spatial behavior of the system ( e . g . , “how does the infection spread ? ” ) . The simulation may also be invoked from the command line . Through the Web , users can process simulation data for output and analysis by a number of common applications such as Microsoft's Excel , University of Maryland's TimeSearcher [41] , and MatLab . We have developed display components that encapsulate multiple-view capabilities and improved multi-scale interface mappings . The IRVE is realized in the international standard VRML97 language . The simulation can be rerun and reanalyzed using a normal VCR-type control tool , which allows the operator , for example , to fast forward , pause , rewind , or drag to a different time point , and to play back runs or analyze simulation output dynamically . In the IRVE , any spatial object ( including the global system ) can be annotated with absolute population numbers ( as a time plot and/or numeric table ) or proportional population numbers ( as a bar graph ) for any or all of the agents . Spatial objects themselves can be animated by heat-map color scales . The intensity of the color associated with each agent is a measure of the absolute level of the agent; so , for example , as the level of free Vir increases , so will the level of intensity of the associated color ( in this case red ) both within the single units and in the entire organ . In our simulation we manage multiple views of the dynamic population values through a higher order annotation called a PopView ( population view ) . A PopView is an interactive annotation that provides three complementary representations of the agent population . The representations can be switched through in series by simple selection . The default view is a color-coded bar graph where users can get a quick , qualitative understanding of the agent populations in a certain location at that time step . The second is a field-value pair text panel , which provides numeric readouts of population levels at that time step . The third is a line graph where the population values for that region are plotted over time . Because of the large amount of time points and the large number of grid locations , the IRVE manages an integrated information environment across two orders of magnitude: “Macro” and “Micro” scales . Through the standard VRML application the user has a number of options including free-navigational modes such as: fly , pan , turn , and examine . This allows users to explore the system , zooming in and out of anatomical structures as desired . In addition , the resulting visualization space is navigable by predefined viewpoints , which can be visited sequentially or randomly through menu activation . This guarantees that all content can be accessible and users can recover from any disorientation . The Visualizer manages Macro and Micro scale result visualizations using proximity-based filtering and scripting of scene logic . As users approach a given anatomical structure , the micro-scale meshes and results are loaded and synchronized to the time on the users' VCR controller .	The possibility of using computer simulation and mathematical modeling to gain insight into biological systems is receiving increased attention . However , it is as yet unclear to what extent these techniques will provide useful biological insights or even what the best approach is . Epstein–Barr virus ( EBV ) provides a good candidate to address these issues . It persistently infects most humans and is associated with several important diseases , including cancer . We have developed an agent-based computer model/simulation ( PathSim , Pathogen Simulation ) of EBV infection . The simulation is performed on a virtual grid that represents the anatomy where EBV infects and persists . The simulation is presented on a computer screen in a form that resembles a computer game . This makes it readily accessible to investigators who are not well versed in computer technology . The simulation allows us to identify switch points in the infection process that direct the disease course towards the end points of persistence , clearance , or death , and identify conditions that reproduce aspects of EBV-associated diseases . Such simulations , combined with laboratory and clinical studies and animal models , provide a powerful approach to investigating and controlling EBV infection , including the design of targeted anti-viral therapies .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "agent", "based", "model", "infectious", "diseases", "epstein-barr", "virus", "computer", "simulation", "pathology", "virology", "immunology", "dynamics", "of", "infection", "computational", "biology" ]	2007	A Virtual Look at Epstein–Barr Virus Infection: Biological Interpretations
Mobile elements comprise close to one half of the mass of the human genome . Only LINE-1 ( L1 ) , an autonomous non-Long Terminal Repeat ( LTR ) retrotransposon , and its non-autonomous partners—such as the retropseudogenes , SVA , and the SINE , Alu—are currently active human retroelements . Experimental evidence shows that Alu retrotransposition depends on L1 ORF2 protein , which has led to the presumption that LINEs and SINEs share the same basic insertional mechanism . Our data demonstrate clear differences in the time required to generate insertions between marked Alu and L1 elements . In our tissue culture system , the process of L1 insertion requires close to 48 hours . In contrast to the RNA pol II-driven L1 , we find that pol III transcribed elements ( Alu , the rodent SINE B2 , and the 7SL , U6 and hY sequences ) can generate inserts within 24 hours or less . Our analyses demonstrate that the observed retrotransposition timing does not dictate insertion rate and is independent of the type of reporter cassette utilized . The additional time requirement by L1 cannot be directly attributed to differences in transcription , transcript length , splicing processes , ORF2 protein production , or the ability of functional ORF2p to reach the nucleus . However , the insertion rate of a marked Alu transcript drastically drops when driven by an RNA pol II promoter ( CMV ) and the retrotransposition timing parallels that of L1 . Furthermore , the “pol II Alu transcript” behaves like the processed pseudogenes in our retrotransposition assay , requiring supplementation with L1 ORF1p in addition to ORF2p . We postulate that the observed differences in retrotransposition kinetics of these elements are dictated by the type of RNA polymerase generating the transcript . We present a model that highlights the critical differences of LINE and SINE transcripts that likely define their retrotransposition timing . Mobile elements have constantly assaulted genomes , shaping and molding their structure and organization . In particular , mobile elements have flourished in mammals generating between 40–50% of their genomic sequence [1]–[3] . About one third of the human genome can be attributed directly or indirectly to the activity of the non-LTR retroelements also referred to as LINEs ( Long INterspersed Elements ) . LINE-1 ( L1 ) and its non-autonomous partners Alu , SVA , and retropseudogenes continue to amplify in the human genome . L1 and the SINE ( Short INterspersed Element ) , Alu , are by far the most numerous , adding up to 1 . 5 million copies [1] . Although Alu mobilization depends on L1 proteins [4] , they outnumber L1 inserts by 2 to 1 . Similarly , the sum of the total copies of all rodent SINEs outnumber L1 copies about 2 to 1 [2] , [3] . Alu and the rodent SINE inserts have been more successful than other non-autonomous retroelements , such as the retropseudogenes [5] . Size and sequence composition differences between SINEs and LINEs may allow the mammalian genome to better tolerate SINE insertions , reviewed in [6] . Negative selection has clearly played a role in reducing L1 copy number through ectopic recombination and elimination of many full length and nearly full length L1 inserts [7] . However , processes other than negative selection must influence the observed differences . The updated reports of diseases caused by de novo inserts ( where little , or no , selection has occurred ) show that Alu inserts outnumber those of L1 by about 2 to 1 [6] , [8] . Tissue culture assay systems indicate that L1 retrotransposition rates are consistently higher than those observed for SINEs [4] , [9] . This is possibly a reflection of the strong cis-preference contained by L1 [10] , [11] , while Alu must compete for L1 proteins in trans . How is it that Alu with a lower retrotransposition rate than L1 , contributes more de novo disease cases ? It is likely that multiple factors are involved , such as the ability to bind SRP9/14 [12] , [13] . Retroelements are mobile elements that amplify through an RNA intermediate in a process known as retrotransposition [14] . There are limited data on the details of the mechanism of LINE retrotransposition , and even less for SINE retrotransposition . The process begins with the generation of RNA ( Figure 1A ) . Active L1 elements express two proteins from a bicistronic mRNA: ORF1p[15] and ORF2p ( Figure 1B and C ) . Both L1 proteins are needed for L1 retrotransposition [16] . In contrast to L1 , ORF2p expression is sufficient for SINE retrotransposition [4] , [9] , [17] , while ORF1p may enhance the process [17] . ORF1p possesses nucleic acid chaperone activity [18] , [19] , an essential property for L1 retrotransposition [19] , [20] . ORF2p is a multifunctional protein with endonuclease and reverse transcriptase activities [21] , [22] . Both proteins are proposed to interact in cis [10] , [11] with the L1 RNA to form a cytoplasmic RNP complex interacting with polyribosomes [20] , [23] . SINE RNA is predominantly found in the cytoplasm as an RNP complex [12] , [24] , [25] ( Figure 1C ) and uses L1 protein ( s ) in trans for its mobilization . The endonuclease of the L1 ORF2p generates the first nick within the L1 endonuclease recognition sequence generating single stranded DNA that primes the reverse transcription [22] , [26] . Both L1 and Alu are proposed to undergo integration through a target-primed reverse transcription ( TPRT ) reaction [27] . To generate a new insertion , L1 and SINE elements must return to the nucleus either together or independently ( Figure 1D ) . Reported data suggest that retrotransposition-competent L1 RNPs may transit through the nucleolus [28] . The 3′ poly-A stretch or “A-tail” of LINEs , SINEs and processed pseudogenes is required for the priming of reverse transcription ( Figure 1E ) [4] , [29] . Unlike the post-transcriptionally generated A-tail of pol II RNAs ( mRNA ) , SINE A-tails are included within their sequence and play an important role in SINE retrotransposition [30] , [31] . The details of the final integration and ligation of the L1 or Alu inserts into the host DNA remain unclear . Recent reports indicate that cellular factors , such as DNA repair enzymes , may aid in the L1 retrotransposition process [32] , [33] . The final inserted sequence is typically flanked by direct repeats ( Figure 1G ) . Non-autonomous retrotransposed inserts , such as Alu , SVA , hYs and retropseudogenes share these hallmarks with L1 inserts , strongly suggesting that these elements use the L1 ORF2p endonuclease generated nick for their integration [34]–[36] . To date , all known SINEs are ancestrally derived from RNA pol III transcribed RNA genes , reviewed in [37] . The vast majority are derived from different tRNA genes and only two ( Alu and the rodent B1 ) originated from the 7SL RNA gene , a component of the signal recognition particle ( SRP ) [38] . Other examples of pol III transcribed repeats include the four hY genes ( hY1 , hY3 , hY4 and hY5 ) that likely contributed directly or indirectly to the generation of almost 1000 copies in the human genome by retrotransposition [36] , [39] . In contrast to SINEs , an internal RNA pol II promoter drives LINE transcription with the unusual ability to start transcription upstream of its location . Like other pol II RNAs , L1 transcription is regulated by different mechanisms , including promoter methylation [40] , transcriptional attenuation due to A-richness [41] , premature polyadenylation [42] , and the generation of different splice variants [6] . Additional studies suggest that at least some portion of the L1 mRNAs are capped [43] and that the capping enhances L1 translation [44] . Previously , an L1 element tagged with a green fluorescent protein ( EGFP ) retrotransposition cassette was used to detect L1 retrotransposition “near real time” [45] . The earliest detection of an L1 retrotransposition event was 48 h post-transfection . In this manuscript , we evaluate the timing of retrotransposition ( defined as the time required for a retroelement from the initial transcription step to complete an insertion ) of tagged Alu and L1 . We demonstrate that Alu elements only require about half of the amount of time as L1 to generate an insert . Our data demonstrate that the type of RNA polymerase dictates the retrotransposition timing , but does not determine the retrotransposition rate ( defined as the number of inserts a given element can generate , i . e . the “efficiency” of an element ) . After evaluating several potential time limiting steps , we show that the RNA polymerase type is an important early factor contributing to the divergent retrotransposition kinetics between LINEs and SINEs . Reverse transcriptase ( RT ) domains of multiple sources can be grouped into a family of shared sequence homology [NCBI cdd pfam00078 . 12] [46] , including the RT of the human immunodeficiency virus and L1 ORF2 protein . Endogenous RT activity is inhibited by two antiretroviral agents nevirapine and efavirenz [47] . L1 retrotransposition in a culture assay system can be suppressed by the addition of a variety of HIV RT inhibitors [48] , [49] . This system utilizes a tagged vector designed to allow expression of the reporter gene only when the retroelement goes through its reverse transcriptase-dependent amplification process ( Figure 2A ) . Therefore , only the newly inserted element will express the reporter gene ( e . g . neo ) . Using the established L1 and Alu retrotransposition tissue culture assays [4] , [16] , we evaluated the dose of , 2′ , 3′-didehydro-3′-deoxy-thymidine ( d4t ) required to abolish retrotransposition of L1 and L1 ORF2p driven Alu without adversely affecting cell growth and viability . Treatment of transiently transfected HeLa cells showed that both L1 and Alu activities presented a d4t activity inhibitory concentration 50 ( IC50 ) of about 2 µM ( Figure S1 ) . For our subsequent experiments we utilized d4t treatments at 50 µM ( 25 fold the IC50 ) to inhibit SINE and LINE retrotransposition in tissue culture . We selected this dose for its efficient inhibition of retrotransposition and lack of observed negative effects , determined by colony formation of an unrelated plasmid that expresses a functional neomycin resistance gene and integrates into genomic DNA by random integration rather than by an L1-dependent mechanism ( data not shown ) . We took advantage of the d4t inhibition to determine L1 and Alu retrotransposition kinetics in cultured cells . By treating cells with d4t at different time points after the transient transfection with the vectors expressing the tagged L1 or Alu plus ORF2p , we specifically inhibited the retrotransposition process at designated time periods ( shown in Figure 2B ) . Any detected L1 or Alu inserts are presumed to have completed the insertion process prior to the addition of the d4t , as inhibition of ORF2p RT activity would prevent the generation of the cDNA . Using this approach , we show that L1 inserts are not detected in cultured cells during the first 24 h post-transfection ( Figure 2C ) . Similar results were previously observed using a green fluorescent protein ( EGFP ) -tagged L1 element [45] , [50] . The earliest detection of L1 inserts occurred at 32 h post-transfection ( Figure S2 ) . In contrast , we can easily detect Alu inserts 24 h and sometimes as early as 18 h post-transfection ( Figure 2C ) . Generation of an RNA transcript is an essential first step of the retrotransposition cycle ( Figure 1A ) . Besides serving as a template for protein translation , L1 RNA acts as the insertion template during retrotransposition . Thus , transcriptional limitations or variations can directly impact retrotransposition of L1 elements as well as other retroelements . Previous studies demonstrate that L1 elements generate low amounts of full-length transcripts due to premature polyadenylation [42] , transcriptional inefficiency due to A-richness [41] , and multiple splicing events [6] . In all these studies , a decrease in the amount of L1 mRNA contributed to reduced retrotransposition and , conversely , the rate increased with higher amounts of full-length L1 RNA [42] , [51] , [52] . To determine whether L1 RNA transcription and processing contributes to the observed timing difference between L1 and Alu inserts , we performed a time course to evaluate the generation of the spliced RNA product in cells transiently transfected with L1mneo , AluYa5neoTET , and L1neoTET ( Figure 3 ) . Because the Alu construct is driven by RNA polymerase III , its tag ( neoTET ) contains a self splicing intron disrupting the neomycin gene . Therefore , we included an additional L1 construct that contains the exact same self splicing ( neoTET ) tag present in the Alu vector to control for any potential variations introduced by splicing dynamics . Full-length spliced and unspliced transcripts from Alu and both L1 constructs could be detected as early as 3 hours post-transfection ( northern blots shown in Figure S3 ) . The mneo and neoTET tagged L1 constructs exhibited similar kinetics for the spliced transcript ( only RNA that will generate G418R colonies when retrotransposed ) , peaking by 24 h and declining by 72 hours ( Figure 3 ) . Splicing efficiency of the RNA produced by different expression vectors was evaluated ( Table S1 ) . Equivalent splicing efficiency was observed for the L1 and Alu transcripts sharing the same neomycin cassette ( neoTET or mneo ) . Alu-tag transcripts were only detected in the cytoplasmic fraction at any of the time points evaluated ( data not shown ) , consistent with what has been previously reported for the authentic Alu “untagged” RNA [53] . Despite early L1 mRNA availability , no L1 inserts were observed at the 24 h time point . Spliced Alu transcripts peak around 48 h , declining by 72 h , much like L1 mRNA ( Figure 3 ) . However , in contrast to L1 , numerous Alu inserts are readily detectable by 24 h . These results demonstrate that the full-length properly spliced L1 RNA is generated in the same time period as the Alu RNA . Thus , it is unlikely that RNA transcription or variation in the type of splicing within the neo cassette account for the observed time difference between the generation of Alu and L1 inserts . Another difference between the Alu and L1 elements involves the length of the transcript , which could alter the time required by the reverse transcriptase to generate a full-length cDNA . In this assay system full length inserts are not required to generate a G418R colony . In both Alu and L1 assays , inserts are detected with the retrotransposition of the minimal unit of a functional neomycin gene , which is identical in length in both transcripts once the intron is removed . Therefore , the timing differences observed between these two elements should be independent of the transcript length . We next assessed whether the delay reflects the time required for translation of the L1 proteins and the ability to reach the nucleus ( Figure 1B–E ) . ORF2 protein has been notoriously difficult to observe by conventional techniques , such as western blot analysis [28] . As an alternative , the ORF2p activities can be evaluated . Because Alu elements require ORF2p for retrotransposition , evaluation of Alu retrotransposition serves as an alternate method to detect ORF2p activity . Therefore , we exploited the trans-complementation assay to monitor the ability of L1 to trans-mobilize Alu , using AluYa5neoTET as a reporter construct . We determined the Alu insertion kinetics in cells cotransfected with the AluYa5neoTET plus the L1 no tag vector . Multiple Alu inserts were detected as early as 24 h post-transfection ( Figure 4 ) , corroborating the availability of the ORF2p expressed from the L1 vector in the nucleus by 24 h . Equivalent results were observed when using a blasticidin tagged L1 to drive Alu retrotransposition ( data not shown ) . Under our experimental conditions , endogenous L1 present in HeLa cells does not significantly contribute to the generation of the G418R colonies as the Alu vector was unable to generate any inserts without L1 supplementation at 24 , 32 and 48 h post-transfection ( vector control , Figure 4 ) . A few solitary colonies ( 2 and 1 ) were observed at the 42 and 72 h time points . This observation clearly demonstrates that a full-length L1 vector generates enough ORF2p to reach the nucleus within 24 h and to mobilize a tagged Alu element in our assay system . Our observations are in agreement with previously published data demonstrating that cells transiently transfected with L1 exhibit extensive double strand breaks at 24 h post-transfection [33] . The observed DNA breaks are dependent on the endonuclease activity of the L1 ORF2p . Our data strongly suggest that translation and nuclear localization of ORF2p is unlikely to be the main limiting step for the observed differences between the L1 and Alu time requirements . In addition , pre-transfection of high amounts of ORF2p or any of the L1 factors ( proteins and/or RNPs ) in trans did not alter L1 retrotransposition timing ( Figure S4 ) . This is not surprising considering that L1 RNA exhibits a strong cis-preference for its own translated proteins for retrotransposition [10] , [11] . Pre-transfection with ORF2p showed a few more Alu inserts at early time points ( data not shown ) . However , this slight increase was not statistically significant ( Student's paired t-test , p = 0 . 297 ) . Transcripts generated from RNA polymerase II and III promoters differ in their capping , 3′ end processing , folding structures , post-transcriptional processing , interaction with translation factors and degradation pathways , reviewed in [54]–[56] . In addition , these two transcriptional complexes can be observed in different spatial locations in the nucleus indicating discrete transcriptional sites [57] , [58] . To evaluate the timing of retrotransposition of other pol III-driven genes we generated “tagged” versions of 6 human genes ( 7SL , U6 , hY1 , hY3 , hY4 and hY5 ) by cloning the genes with at least 300 bp of their upstream enhancer sequence 5′ of the neoTET cassette ( details in materials and methods ) . Although the “functional” genes are not SINEs per se , we selected these as examples of pol III-driven genes . The human genome contains multiple examples of retrotransposed copies with sequence homology to these genes [36] , [59] . Thus , these serve as our best examples of other human pol III-driven constructs . We also included in our analysis the pol III-driven B2 element as a known active rodent SINE [9] , [60] . In our d4t-assay system , all tagged pol III-driven elements generated inserts by 24 h post-transfection when supplemented with just L1 ORF2p ( Figure 5 ) . To better understand the RNA polymerase influence on retrotransposition , we also evaluated the time requirement of two pol II-driven ( CMV ) constructs: ORF1mneo and pol II Alu ( Figure 6A ) . We selected ORF1mneo because it generates a transcript of L1 ORF1 , which has previously been used to reflect retropseudogene activity [10] . The ORF1mneo vector can retrotranspose when a source of ORF2p is supplied in trans [10] . The pol II Alu ( pCMVYa5mneo ) contains an Alu tagged with the “mneo” cassette from the L1-tagged construct [61] , which contains pol III terminators ( 4 Ts ) that would generate truncated transcripts if the internal pol III A and B boxes in the Alu sequence are used for transcription . The “normal A-tail” at the end of the Alu sequence and 5′ of the neo cassette ( Figure 6A ) was not included in order to prevent potential internal priming for TPRT in the cDNA extension step ( Figure 1E ) , which would circumvent inclusion of the neo reporter gene in the retrotransposed copy . Thus , only the Alu body sequence was utilized in the construct . Just like the L1 construct , the A-tail used in the TPRT step is generated from the transcript polyadenylation by the RNA polymerase II from the SV40pA signal at the 3′ end of the neo cassette ( Figure 6A ) . Spliced and unspliced transcripts were detected from both constructs by 24 h ( Figure 6B ) . The tagged ORF1p transcript driven by an ORF2p generated one single insert at 24 hours ( Figure 6C ) , while the total number of colonies generated were 136 and 226 for 48 h and 72 h respectively . It is possible that the endogenous L1 expression in HeLa cells [6] affected the timing . However , our data on Alu retrotransposition indicates that effects from endogenous L1 expression under our experimental conditions are negligible ( Figure 4 ) . Most likely , the single G418R colony observed at 24 hours is due to a rare event that escaped d4t inhibition . A quantitative time course evaluation of the spliced RNA product in cells transiently transfected with ORF1mneo and AluYa5mneo further indicates that the availability of spliced product is not limiting retrotransposition timing ( Figure 6E ) . No pol II-generated Alu inserts were ever observed when supplemented with ORF2p under any conditions tested , representing a rate of less than 1×106 cells/µg of plasmid . However , retrotransposition of the pol II-driven Alu transcript occurred when it was supplemented with both ORF2p and ORF1p expression plasmid ( Figure 6D ) . Under these conditions , G418R colonies were observed at 48 h post-transfection , much like L1 and retropseudogene behavior . No colonies were ever observed at the 24 h time point in 5 independent experiments using triplicates for each time point . Swapping the RNA pol III for an RNA pol II promoter changed the retrotransposition requirements of the tagged Alu to reflect those observed for pseudogenes and LINEs . Recent data demonstrate that one amino acid substitution in the mouse L1 ORF1 protein dramatically affects retrotransposition rate and the ability to detect new inserts earlier [50] . We evaluated the insertion timing of the most efficient L1 available at the time , the synthetic mouse L1 ( L1m syn ) previously reported to increase retrotransposition efficiency by more than 200 fold relative to the wildtype L1spa element [52] . Despite having a much higher retrotransposition rate , L1m syn required 48 h to generate inserts even when spliced RNA could be detected as early as 3 hours post-transfection ( Figure 7 ) . There were a few ( 1 to 2 ) colonies at 24 hours or earlier but these are likely outlier observations as they only represent 0 . 001 of the total observed G418R colonies . Our data are consistent with the observation that all of the evaluated pol II-driven constructs require 48 h , while all of the pol III-driven constructs generate inserts by 24 h despite their very low retrotransposition rates relative to L1 ( Table 1 ) . Because of the large variation in retrotransposition rates , we opted to show the relative number of inserts in the figures for each construct by designating the 48 or 72 hour time point as 100% . While both U6 and Alu tagged transcripts , for example , can generate inserts by 24 hours , their retrotransposition rates ( i . e . , the actual number of observed inserts ) differ dramatically . The same is true for the tagged L1 and ORF1 RNAs . Throughout mammalian evolution different mobile elements have flourished within genomes . Retroelements such as LINEs and SINEs have been particularly successful , generating more than one third of human sequence mass . Interestingly , the parasitic non-autonomous SINE elements outnumber their autonomous LINE partners in the primate and rodent genomes . The success of SINEs is especially evident when compared to the copy numbers of other non-autonomous elements such as the retropseudogenes . Our data reveal differences between retropseudogenes , Alu , and L1 retrotransposition . When evaluating Alu and L1 retrotransposition kinetics , the tagged Alu transcript required less time to generate an insert . This timing difference can not be attributed to differences in the time required to generate functional transcripts or availability of L1 proteins . It is clear that full-length functional L1 transcripts can be detected as early as 3 hours post-transfection and are abundant by 24 h post-transfection . In addition , the difference observed between Alu and L1 kinetics could not be attributed to the type of detection cassette system ( self splicing or not ) or to the differences in the retrotransposition rates . L1 colonies were rarely observed ( Figure 7 ) at time points earlier than 48 h . These few observed G418R colonies possibly represent the rare event that circumvented inhibition by d4t ( in one experiment a colony was observed even at the zero time point ) . In our assay , production of L1 ORF2p is not limiting . Our data demonstrate that enough ORF2p is generated from an L1 construct to drive Alu insertions within 24 hours post-transfection , which indicates that ORF2p is made and readily available for Alu transcript mobilization . However , at this time we do not know if the ORF2p reaches the nucleus as a “free” protein or as part of an RNP with the L1 RNA or Alu RNA . As expected , due to the L1 cis-preference [10] , pre-transfections with ORF1p , ORF2p or other L1 components , such as full-length transcripts or RNPs , did not affect the L1 time requirement . Although unexpected , it is not totally surprising that Alu and L1 present different retrotransposition time requirements . Previous data show that , although Alu and L1 share the same insertion hallmarks , the two elements can exhibit differences in their behavior . For example , of two HeLa “cell lines , ” only one supports Alu retrotransposition while both support L1 retrotransposition [62] . In addition , Alu and L1 are selectively inhibited by different APOBEC3 proteins [62] . This corroborates our observations that cellular components differentiate between Alu and L1 retrotransposition . Our data suggest that the observed time differences are dependent on the type of RNA polymerase generating the transcript . Multiple features that distinguish these two transcript types may collectively or individually contribute to the observed differences in the retrotransposition timing between L1 and Alu elements . RNA capping , association with the translational machinery and ORF1 requirement are plausible factors that could influence SINE and LINE retrotransposition kinetics . As a pol II product , L1 mRNA is likely capped . Experimental evidence indicates that at least part of the L1 mRNA is capped [43] and that capping enhances L1 translation in vitro [44] . In contrast , pol III genes lack the 7-methylguanosine cap and are subjected to different processing in a spatially separate location of the nucleus [57] , [58] . L1 mRNA likely interacts with most , if not all , of the pol II protein complexes that assemble with the transcription of generic mRNAs , as evidenced by the premature polyadenylation and splicing of L1 transcripts [6] , [42] . Even though both pol II and pol III produced RNAs form complexes with various cellular proteins , the structure and composition of these RNPs varies dramatically . As a rule , pol III transcripts do not code for proteins and therefore interact with the translational machinery in a different manner than mRNA . Most known pol III transcripts fold to form a structured RNA and associate with a variety of proteins to form RNPs . Specifically , Alu interaction with SRP9 and SRP14 [12] is thought to transiently provide proximity to the ribosomal complexes and translating L1 RNA , allowing the Alu transcript to efficiently compete for the L1 factors required for retrotransposition [26] . It is also likely that the ability of the dimeric Alu to bind these proteins contributes to the dramatic difference in retrotransposition rates observed between Alu and other SINEs [9] , [13] . In contrast , the polyribosomes and translation machinery assemble with the L1 mRNA in a more stable complex to undergo translation . The cis-preference displayed by L1 [20] suggests that the L1 RNA must dissociate from the cellular translation machinery to form L1 RNPs as an intermediate step in the retrotranspositional process . These L1 complexes are composed of L1 RNA , ORF1p [20] , and likely ORF2 protein [11] . All three components are shown to co-purify in the polyribosomal fraction of the cytoplasm [11] , [23] . It is plausible that ORF1p directly competes with the cellular translation machinery for access to L1 mRNAs , transitioning the L1 transcript away from the polyribosomal fraction and into the retrotranspositionally competent RNPs . Because of their nature and subcellular localization , SINEs completely avoid these two potentially time consuming steps in their mobilization . Therefore , SINE transcripts may enter their retrotransposition cycle as soon as L1 ORF2p becomes available . The pol II-driven Alu transcripts that are most likely to associate with the cellular translational machinery , at least transiently , require L1 ORF1 protein in addition to ORF2 protein for retrotransposition in a manner reminiscent of retropseudogenes [29] . The retrotransposition time of the pol II-driven Alu parallels that of L1 . At this stage it is unclear what the role of ORF1p is in the trans-mobilization of retropseudogenes or the pol II Alu transcript . However , it is consistent with the above-discussed hypothesis implicating ORF1 protein in removing pol II RNAs from their expected cycle of translation and degradation . Thus , the pol II L1 and the pol III Alu transcript interactions with different cellular components may dictate the timing difference between L1 and Alu RNAs to form their respective retrotranspositionally competent complexes . The inefficient retrotransposition rate of the pol II-driven Alu construct suggests that the presence of an Alu sequence within an mRNA would not facilitate its retrotransposition by L1 factors . Although there is no available data on the SVA promoter , it is unlikely that the pol III polymerase drives SVA transcription due to the presence of numerous pol III terminators within its sequence . Thus , it is questionable whether the truncated antisense Alu-like sequences present in the SVA element contribute to the L1 trans-complementation of this retroposon as previously suggested [35] . In addition to assisting its own retrotransposition , the cis-preference exhibited by L1 may decrease cell damage by limiting random retrotransposition of cellular mRNA . A previous study demonstrated the co-localization of ORF1p and cellular proteins to stress granules[63] . The authors suggest that the sequestering of ORF1 protein in stress granules for degradation may prevent promiscuous binding of ORF1p to non-L1 mRNAs . Thus , as a side effect of L1 self-preference , retropseudogene formation is less likely [5] . In addition , this “cis-preference” could help the L1 transcript “escape” the ribosomal complex and degradation pathways . Once translation is completed , most transcripts decay by several known mRNA degradation pathways , reviewed in [56] . In contrast , pol III transcripts are meant to perform their function as RNA molecules in the cytoplasm or nucleus before degradation by the exosome [64] . Essentially , the functional molecule of pol III genes is the RNA , while for pol II genes the mRNA is an intermediary prior to the generation of the functional protein . In the case of L1 , the ORF1p may play an additional role by protecting the L1 RNA from degradation , increasing the chance of returning to the nucleus where the involvement of ORF1p in the L1 integration process has been previously suggested [18] , [23] . Thus , the requirement for both ORF1 and ORF2 proteins could contribute to the longer time needed for L1 transcripts to generate inserts . In addition , it is plausible that interactions with different cellular components during insertion , mediated by ORF1p , may contribute to the timing differences observed . We postulate that the differences observed in retrotransposition kinetics are dictated by the type of RNA polymerase generating the transcript . We propose an initial model where the cytoplasmic interactions of pol II ( L1 and mRNA ) and pol III transcripts and pathways influence the amplification kinetics of LINEs and SINEs ( Figure 8 ) . Overall , it is evident that the type of RNA polymerase generating the transcript alters the timing of mobile element insertion and remains a critical parameter in the classification of different types of retroelements . The basic transient L1 [16] or Alu [4] retrotransposition assay was performed as previously described with some minor modifications . Briefly , HeLa cells ( ATCC CCL2 ) were seeded in T-75 flasks at a density of 5×105 cells/flask or in 6 well plates at a density of 2 . 5–5×104/well . Transient transfections were performed the next day with Lipofectamine Plus following the manufacturer's protocol ( Invitrogen ) , with 3 µg of SINE-neoTET vector plus 1 µg pBud-ORF2opt or 1 µg of L1 no tag . For L1 assays 1 µg of JM101/L1 . 3 was used . Inhibitory effects on cellular growth or colony formation capabilities by the d4t treatment was evaluated by transfecting cells in parallel with 0 . 3 µg of a plasmid expressing neomycin resistance ( pIRES2-EGFP; BD Biosciences Clontech ) as a “toxicity” control . Following removal of transfection cocktail , the cells were treated with the appropriate media containing 400 µg/ml Geneticin/G418 ( Fisher Scientific ) alone or in combination with 50 µM d4t for selection and/or reverse transcriptase inhibition . After 14 days , cells were fixed and stained for 30 minutes with crystal violet ( 0 . 2% crystal violet in 5% acetic acid and 2 . 5% isopropanol ) . The inhibitor d4t- ( 2′ , 3′-Didehydro-3′-deoxy-thymidine; Sigma-Aldrich ) was freshly added to the selection media at the indicated time period . During the inhibitor treatment period all cells in the experiment were refreshed daily for the first week with the appropriate media . The rate of insertion efficiency ( retrotransposition rate ) was determined as the number of visible G418R-resistant colonies obtained at 72 h after transient transfection of 1×106 seeded HeLa cells with 1 µg of the neo tagged construct . RNA extraction and poly ( A ) selection was performed as previously described [42] . Total RNA was extracted using the recommended protocol for TRIzol Reagent ( Invitrogen ) from two 75 cm2 cell culture flasks at 3 , 6 , 24 , 48 , and 72 hours post-transfection . The PolyATract mRNA isolation system III ( Promega ) was used to select polyadenylated RNA species following the manufacturer's protocol . After separation in a 1% ( L1 ) or a 2% ( pol III constructs ) agarose-formaldehyde gel , the RNA was transferred to a Hybond-N nylon membrane ( Amersham Biosciences ) . The RNA was cross-linked to the membrane using a UV-light ( GS Gene linker , BioRad ) and pre-hybridized in 30% formamide , 1× Denhardt's solution , 1% SDS , 1 M NaCl , 100 µg/ml salmon sperm DNA , 100 µg/ml-1 yeast t-RNA at 60°C for at least 6 h . The 3′ region of the neomycin gene was amplified by PCR using the following primers T7neo ( − ) : 5′-TAATACGACTCACTATAAGGACGAGGCAGCG-3′ and Neo northern ( + ) : 5″- GAAGAACTCGTCAAGAAGG-3′ . The isolated PCR product was used as a DNA template to generate a 32P-CTP ( Amersham Biosciences ) labeled single strand-specific RNA probe using the MAXIscript T7 kit ( Ambion ) following the manufacturer's recommended protocol . We utilized material included in the kit to generate the riboprobe for the β-actin . The radiolabeled probe was purified by filtration through a NucAway Spin column ( Ambion ) . Hybridization with the probe ( final concentration of 4–12×106 cpm/ml ) was carried out overnight in the pre-hybridization solution at 60°C . Two ten-minute washes were performed at high stringency ( 0 . 1×SSC , 0 . 1%SDS ) at 60°C . The results of the northern blot assays were evaluated using a Typhoon Phosphorimager ( Amersham Biosciences ) and the ImageQuant software .	SINE retroelement amplification has been extremely successful in the human genome . Although these non-autonomous elements parasitize factors from LINEs , both the human Alu and the cumulative rodent SINEs have generated over one million copies in their respective hosts . Alu-induced mutagenesis is responsible for the majority of the documented instances of human retroelement insertion-induced disease . Our data indicate that SINEs require a shorter period of time to complete insertion than L1s , possibly contributing to the ability of Alu elements to effectively parasitize L1 components . We demonstrate that RNA polymerase changes the timing Alu requires to complete retrotransposition and creates the need for the L1 ORF1protein in addition to ORF2p . We postulate that the way cells manage pol III and pol II ( mRNA ) transcripts affects the timing of a transcript going through the retrotransposition pathway . We propose a model that highlights some of the critical differences of LINE and SINE transcripts that likely play a crucial role in their retrotransposition process .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "cell", "biology", "molecular", "biology" ]	2009	The RNA Polymerase Dictates ORF1 Requirement and Timing of LINE and SINE Retrotransposition
As an advanced approach to identify suitable targeting molecules required for various diagnostic and therapeutic interventions , we developed a procedure to devise peptides with customizable features by an iterative computer-assisted optimization strategy . An evolutionary algorithm was utilized to breed peptides in silico and the “fitness” of peptides was determined in an appropriate laboratory in vitro assay . The influence of different evolutional parameters and mechanisms such as mutation rate , crossover probability , gaussian variation and fitness value scaling on the course of this artificial evolutional process was investigated . As a proof of concept peptidic ligands for a model target molecule , the cell surface glycolipid ganglioside GM1 , were identified . Consensus sequences describing local fitness optima were reached from diverse sets of L- and proteolytically stable D lead peptides . Ten rounds of evolutional optimization encompassing a total of just 4400 peptides lead to an increase in affinity of the peptides towards fluorescently labeled ganglioside GM1 by a factor of 100 for L- and 400 for D-peptides . In the field of bioactive substances , peptides are drawing increasing attention as they close the gap between small molecules and proteins , combining the compact size and synthetic accessibility of the former with the high specificity in molecular recognition processes of the latter . Of particular interest in this context are tasks where targeting of an active compound to a defined cellular or molecular structure is desired , e . g . the site-specific delivery of drugs , vaccines , or contrast agents for molecular imaging applications [1] , [2] . To date , mainly antibodies are used in such situations [2] , [3] , yet the large size of an antibody ligand severely hampers tissue penetration and optical resolution , and its antigenicity and degradability limit its use in vivo . Hence , various approaches to artificially reduce ligand size while maintaining specificity are being pursued to establish the next generation of targeting molecules [4] . Small peptides built up of 10 to 20 amino acid residues which permit highly specific interactions with biological targets carry this concept to its final consequence [5] , [6] . Although the use of peptides in therapy and diagnostics may be hampered by their proteolytic lability or limited cell penetration , too , these obstacles can be overcome by building up proteolytically stable peptide isomers from D-amino acid residues or by coupling the peptides to membrane shuttles [7] , [8] . Far more challenging is the identification of peptide sequences that exhibit the necessary sensitivity and specificity of a targeting ligand . To date , high throughput screening of large peptide libraries is a common approach for the identification of peptide ligands , but with increasing ligand length the procedure rapidly reaches its limits . Beyond a length of 9–10 amino acids such libraries are no longer representative due to the exponentially growing peptide sequence space ( e . g . 1021 sequences for 16mer L-peptides ) . In order to overcome this limitation , computational structure based design methods suitable for reduction of the sequence space allocatable have been established . If the 3D molecular structure of the target is available it can be used in docking approaches for the design of peptide ligands for these targets using mere in silico procedures [9] , [10] . Another way to optimize peptide sequences for desired applications is the use of structural scaffolds [10] in molecular dynamics simulations . Both approaches work best with rigid proteinacious target molecules . Structure-independent design of peptides can be accomplished by e . g . sequence motif scanning [10] utilizing learning algorithms such as artificial neural networks . This technique , however , is limited to sequence data already present in training sets and often fails to create novelty . In protein design , directed evolution strategies which aim to improve candidates by iterative rounds of mutations and functional screenings constitute another way to optimize biomolecules [11] . These methods , which include gene-shuffling , site-directed mutagenesis and chimeragenesis , work on the DNA-level and hence are restricted to gene encoded optimization candidates . Therefore the incorporation of non-natural building blocks or the optimization of all D-peptides cannot be achieved with these techniques . Yet , the inclusion of a function-screening step in such directed evolution strategies represents a definite strength . In light of the above , it appears most reasonable to employ not a structure , but a function-driven strategy for the identification of peptides suitable for the desired applications [12] , [13] , [14] . We have devised such a strategy based on a molecular optimization process that mimics Darwinian evolution . The evolutionary process is initiated with a peptide library of random sequences or with lead peptides either of known rudimentary suitability or designed by structural considerations . The functional prowess of each peptide is assessed in an appropriate biological assay , in result of which all individuals are assigned “fitness values” . The resulting peptide population is operator-inspected and top candidates are selected to act as parent peptides for the follow up generation . In a computational step , an evolutionary algorithm ( EA ) is used by which the selected peptides are propagated in silico via crossing and mutating them , with the “fittest” candidates having the highest probability of passing on their “genetic information” , i . e . their peptide sequence , to produce a filial generation . We have applied this cooperative in silico and in vitro optimization methodology to identify peptidic ligands for the cell membrane glycolipid ganglioside GM1 , a potential target e . g . for diagnostic imaging applications at the mucosal wall or for mucosal vaccine delivery systems [15] . Evolutionary optimization of peptidic ligands is a complex process where numerous parameters and different evolutionary mechanisms may depend on and influence each other . In order to keep those variables at a manageable level a general framework was defined in the beginning: i ) the length of the ligand to be evolved was set to 16 amino acids , which was deemed a good compromise between synthetic accessibility and sequence space; ii ) a single most relevant criterion – optimal binding to the desired target – was selected as evolutionary goal and iii ) appropriate parameter settings and combinations of evolutionary mechanisms were selected on the basis of empirical in silico simulation studies . The latter was done by shaping 16mer peptides towards a defined characteristic ( molecular mass ) as “pseudo-fitness” . The fitness values were optimized in distance metric simulations , and the evolutionary optimization data were evaluated in order to identify settings which lead the algorithm to converge in a minimal generation count . As evolutionary goal we decided to optimize a peptide ligand for binding to ganglioside GM1 . This particular target molecule was chosen for several reasons . Firstly , carbohydrate molecules , e . g . on cell surface receptors , are a highly relevant class of biological targets , but they are demanding candidates in computational design due to their dynamic solution structure and microheterogeneity [16] . We reasoned that our structure-independent , function-driven approach should be particularly suited to identify ligands for such targets . Secondly , GM1 is a small target , its molecular mass of 1 . 6 kDa lies within the same range as that of a putative peptide ligand . Therefore its binding sites for different leads should largely overlap which heightens the probability for cross-bred offspring to also bind in this region . Proteins and other large target molecules , on the other hand , may offer multiple , independent binding sites for which leads can be identified . Such leads will not produce meaningful progeny upon crossing and thus would unreasonably discredit our approach . And lastly , GM1 is an attractive candidate from the biomedical point of view because it has already been singled out as potential target molecule [15] . Moreover , we already had identified a battery of structurally diverse alleged GM1 binders which could serve as lead sequences for the optimization process [17] and set up an in vitro assay which allowed the simultaneous investigation of a large number of ligand-target interactions . For this assay , all peptides are synthesized in arrays on cellulose membranes . These peptide libraries are probed in a dot-blot type biochemical binding procedure by incubating them with a fluorophore-labeled ganglioside GM1 derivative ( lysoGM1/DY650 ) [17] , and the fluorescence intensities of the individual peptide spots after excitation are quantitated . As initial population for the evolution process the previously identified 64 lead peptide sequences were used [17] . To enable the parallel identification of proteolytically stable D-peptide ligands corresponding retro-inverso D-peptides were submitted to the same process in parallel . The L- and D-peptides were analyzed for their capacity to bind to the GM1 probe , and ranked according to the fluorescence intensity of their respective spots ( Table S1 ) . Peptides yielding fluorescence signals above a statistically defined background [18] were manually selected as lead sequences for the subsequent evolutionary optimization processes ( Table 1 ) . In these lead peptides , the potential influence of individual amino acids on GM1-probe binding was estimated by an “alanine walk” experiment; arginine , phenylalanine , tryptophan and histidine were revealed to be critical for binding . To test the probe binding capacity of the peptide candidates in each generation of the evolutionary process , the peptide sequences proposed by the evolutionary algorithm were synthesized in duplicate onto cellulose membranes and probed with lysoGM1/DY650 . In each array , the “fittest” peptides of the corresponding parent generation were included as a reference to monitor the progress of the optimization process ( Figure 1 ) . The evolutionary algorithm ( EA ) that was used for the mating of the peptide sequences was a generic population-based heuristic optimization algorithm designed for the java runtime environment ( Figure 2 and Supplementary Figure S1 ) . Our evolutionary process differs from the standard genetic algorithm ( GA ) and evolutionary strategy ( ES ) onset [19] as well as from state of the art directed evolution of custom tailored characteristics [20] , [21] , [22] in several points . The work flow in the software is divided into two modules , one module is in charge of the general , ES-like cycle 2 and a second one for the internal , GA-like cycle 1 . A third , external cycle handles the determination of fitness values for all individuals of a population by the biochemical assay . At the beginning of cycle 1 , the fitness values assigned to the peptides are scaled by a fitness scaling function , and fitness proportional selection by stochastic universal sampling ( SUS ) of parent peptides is performed to create a mating pool of peptide sequences . Sequences from that pool enter cycle 2 and are first -times recombined , and the resulting peptides are then mutated to establish a finished filial generation of λ peptides . The peptide sequences created this way are synthesized in parallel for cycle 3 , where their fitness is determined in the biochemical assay . The results are manually inspected to select μ* candidates to act as parent peptides for the next generation . This process is repeated until optimized peptides that meet predefined criteria or display a consensus motif ( see below ) are obtained ( see Supplementary Figure S1 for a more detailed description of the algorithm ) . The fitness values that quantify the suitability of a peptide sequence were deduced from the fluorescence intensities of the respective peptide spot in the in vitro GM1-binding assay . The probability of each candidate sequence to participate in recombination events was determined by subjecting the fitness values to a fitness scaling function . Populations of filial peptides were created by applying the “evolutional parameters” - crossover rate , number of fracture sites and mutation rate - to the lead peptides . In contrast to the situation in biological systems , in our artificial setup these essential EA-parameter settings which strongly influence the performance of the optimization progress can be configured freely [23] . As no theoretical model for the global optimization of parameter settings in such algorithms is available to date [24] , we had to work out appropriate settings for our problem . As a starting point , the crossover rate was adjusted to 100% , i . e . all sequences underwent recombination , the number of fracture sites was set to 1 and the mutation rate to 7% [23] , i . e . an average of one sequence position in each 16mer peptide was mutated . The probability distribution onset for recombination and mutation were equal for all positions in the peptide sequences [23] . These initial parameter settings had been determined in empirical simulation experiments using distance metric calculations . Concerning the population size which also is an important parameter in the optimization procedure we had to consider that the number of peptides synthesized in each generation ought to be large enough to provide sufficient sequence variability for a successful evolution progress . We chose a population size of 200 peptides in each generation as this could be easily managed in form of a SPOT-synthesized peptide array with replicas . The influence of changes in the parameter settings on the results of the optimization process was investigated in the first generations of L-peptide evolution . To begin with , the scaling function which is applied to the fitness values of the peptides in order to increase the differences between these values before starting the mating process was varied . From the lead peptides , two populations of filial peptides were generated , one resulting from a square , the other from an exponential scaling function . By using the square fitness scaling function , candidate sequences with a high fitness value are more often selected for reproduction than it would be the case without or with e . g . a linear scaling function . The exponential fitness scaling function magnifies this effect and puts even more emphasis on only the very fittest candidates , strongly reducing the sequence space available in the filial generation . In our set up , exponential fitness scaling led to a premature decrease in sequence diversity and was therefore deemed inappropriate for a successful optimization and not pursued further . Manual population inspection as the driving force of a directed selection process ( cycle 3 ) was embedded into the flow of the evolutionary algorithm ( Figure 2 ) thereby introducing research experience into each optimization cycle . This knowledge was utilized to choose the number of succeeding peptides out of a population that shall - according to their fitness - serve as “parents” for the next generation . The influence of larger and smaller populations of such parent peptide sequences on the optimization process was investigated using sequence data from the second generation ( gen2 ) . We compared a choice of the 20 fittest ( “very fit” ) versus the 32 fittest peptides ( “very fit+fit” ) from gen2 as parent sequences to generate two populations of the third generation ( gen3 ) . While we observed that the increase in average fitness of the 22 top candidates of the filial generation was higher with only 20 “very fit” parents ( 1 . 8 fold ) than with 32 “very fit+fit” parents ( 1 . 5 fold ) , the total number of peptides with high fitness values was lower in the filial population derived from the smaller parent population than in the filial population derived from the larger parent population . We therefore chose a compromise between high increase in average fitness and high number of very fit filial peptides and decided on using the 25 fittest peptides of each generation as parents for the next evolution round from generation 4 onwards . A careful balance between the decrease of diversity in the peptide population and the increase in fitness of the candidate peptides had to be preserved in order to prevent inbreeding . The evenness of amino acid distribution in the peptide populations was therefore continually monitored [25] ( Figure 3 A ) and the decrease in diversity in the populations was counteracted by an elevation of the mutation rate to 12% in order to prevent premature termination of the optimization process . Even so , in 10 rounds of evolution consensus sequences had emerged ( Figure 3 B–E ) which could not be broken by a mutational rate of 12% and appeared to be a local fitness optimum reachable from the population sets of lead peptide sequences . Although the fraction of peptides which carried the consensus motif still increased at this point of the evolution - and hence the mean fitness of the population – no new sequence motifs were obtainable because of the lack of diversity in the peptide population and consequently no further optimization could be achieved . While we could observe basically no homology between the consensus motif and the respective sequence positions in the lead peptides , it is noteworthy that the amino acids identified by an alanine-scan in the lead peptides as “beneficial” for binding of the GM1-probe were - independent of their position - mostly aromatic ( tryptophan , phenylalanine ) and , to a lesser extent , apolar ( leucine , isoleucine ) and positively charged ( arginine , lysine ) amino acids . This tendency is reflected in the consensus motifs formed in generation 10 where the tryptophan content in the L-peptides rose to 18% and where the D-peptides even had a tryptophan at 50% of the N-terminal positions 1–8 . Our data are consistent with other studies [26] where arginine , phenylalanine and tryptophan were found to be important for binding in phage-mutant experiments with non-labeled GM1 . In the course of the 10-round molecular optimization process performed the fitness of the candidate L- and D-peptide sequences progressed steadily as shown in Figure 4 . The increase in fitness of the peptides could be determined by normalization of the fitness values of all peptides in one generation on the values of their respective parent peptides which were always synthesized and tested anew along with each filial generation ( Figure 1 ) . Since the parent peptides of a filial generation were identical to the fittest candidates of the previous generation a normalization chain over the entire evolutionary process was possible and eliminated potential synthesis-to-synthesis variations . This rendered data from different peptide array experiments comparable and also allowed readjustment of the laser intensity settings used for readout in the imager . The adaption of laser intensity used for the readout of the peptide arrays from different generations was necessary in order to stay in the maximal dynamic measuring range of the instrument . Over the total evolutionary process encompassing ten generations , the fitness of the 25 fittest candidates of each generation of L- and D-peptides increased steadily but the improvement did not follow a simple exponential growth curve . Logarithmic transformation revealed two exponential growth phases , a fast one over the first 5 generations and a second , slower one for the last five generations ( Figure S2 ) . Whether this deceleration is due to the increased mutation rate that was necessary to prevent inbreeding beyond generation 5 or whether it already represents the logistic growth typical for natural growth processes remains disputable . Over all ten generations an average fitness improvement factor per generation of 1 . 6 for the L- and 1 . 7 for the D-peptides was observed leading to a cumulative affinity improvement of about 100-fold for the L- and 400-fold for the D-peptide candidates ( Figure 4 ) . This gain in affinity was attained by synthesizing a total of just 2400 16mer L- and 2000 D-peptides out of a peptide space of 1021 permutations imaginable . These results demonstrate that the evolutionary optimization of a peptidic 16mer GM1 ligand is possible , that the improvement progresses rapidly , that in just 10 generations of 200 peptides each a 400fold improvement in affinity is achievable and that both L- and D-peptides can be optimized this way . Such a dramatic and rapid improvement is not self-evident . Yokobayashi and colleagues [20] for instance achieved only a 3fold improvement over 6 generations when optimizing a 6mer peptidic trypsin inhibitor . The reasons for this discrepancy may reside in the length of the peptide since the peptide space of a 6mer peptide solely composed of proteinogenic amino acids is 1014-times smaller than that of a respective 16mer; the lead structure may already be close to perfection such that there is only little room for improvement; and there may be colliding features in the optimization parameters that per se preclude the existence of an optimal candidate . Such limitations must always be kept in mind when setting up an EA-based molecular evolution procedure . An EA optimizes a molecule within a preset selection of constraints , nothing less but also nothing more . Neither does it correct shortcomings in assay design , ill-chosen amino acid pools or peptide length nor does it introduce properties one has not asked for . For example , a GM1-ligand evolved in a solid phase assay where solubility was not an issue may not be optimally suited for soluble GM1-targeting systems but could perform well on a particulate vaccine delivery system or contrast agent . Likewise , an all-D peptidic GM1-ligand can be perfect for gastrointestinal applications where all-L-peptides are broken down quickly [27] but the same all-D-peptide may be toxic upon systemic application because of its high stability . Another difficulty in the application of the procedure presented here might arise when peptidic ligands for large protein targets are to be developed . Here , the pool of lead sequences must be chosen carefully in order to prevent crossing of leads directed against topologically different binding sites on the target surface . Yet , even though such limitations exist , the EA-based molecular optimization procedure presented in this work is superior to classical high-throughput screening approaches as the latter suffer from similar shortcomings without the time- and material-saving advantages of the evolution process . In light of these considerations we are confident that the work presented in this study will be a valuable stimulus to fields where site specific delivery is crucial , such as drug targeting , molecular imaging and personalized medicine . Peptide arrays were synthesized onto cellulose membranes following a modified version of the procedure described by Frank [28] . The cellulose membranes were derivatized with epibromohydrine in dioxane ( 10% ( v/v ) containing 1% of 60% perchloric acid , 0 . 02 ml/cm2 membrane ) , for 3 h at room temperature ( RT ) according to Ast et al . [29] . The membranes were washed 3× with dioxane ( 0 . 13 ml/cm2 membrane , 5 min , RT ) and subsequently exposed to a solution of 4 , 7 , 10-trioxa-undecane-1 , 13-diamine ( 20% ( v/v ) in DMF , 0 . 21 ml/cm2 membrane ) for 3 h at RT . This solution was removed and the membranes were incubated in a solution of NaOMe in Methanol ( 5 M , 0 . 21 ml/cm2 membrane ) for 30 min , washed 7× with aqueous methanol ( 0 . 21 ml/cm2 membrane , 5 min , RT ) and air dried . For definition of the synthesis areas , 0 . 1 µl of 0 . 2 M Fmoc-β-alanine-pentafluorophenyl ( Pfp ) ester ( 0 . 2 M Fmoc-β-alanine-Pfp-ester in NMP ) were applied by the pipetting robot at defined positions on the membrane , then excess hydroxyl groups were blocked ( “capping” ) with a solution of acetic anhydride ( Ac2O ) and diisopropylethylamine in DMF ( 8% Ac2O , 15% diisopropylethylamine ( v/v ) , 0 . 13 ml/cm2 membrane , 1 h , RT , rocking ) and the membranes were washed 3× with DMF ( 0 . 13 ml/cm2 membrane , 3 min , RT ) . The Fmoc protecting groups were cleaved using piperidine ( 20% ( v/v ) in DMF , 1 ml/cm2 membrane , 20 min , RT ) . To confirm the presence of free amino functions in the spot areas , the membranes were washed 5× with DMF ( 0 . 13 ml/cm2 membrane , 3 min , RT ) , stained with bromophenol blue in DMF ( 0 . 01% ( v/v ) , 0 . 13 ml/cm2 membrane , 10 min , RT , rocking ) , washed 3× with 100% ethanol ( 0 . 13 ml/cm2 membrane , 3 min , RT ) and air dried . The above capping , washing , Fmoc-cleavage and staining steps were repeated between all synthesis cycles from the definition of the synthesis areas ( spots ) onwards , but after the third synthesis cycle capping was performed for 20 min only using 2% Ac2O ( v/v ) in DMF . For all peptide synthesis cycles , the amino acid building blocks were converted into their corresponding HOBt esters immediately before use by adding 1 . 25 moles diisopropylcarbodiimide per mole amino acid to a solution containing 0 . 4 M N-α-Fmoc-protected amino acid and 0 . 7 M HOBt in NMP ( final concentration: 0 . 2 M amino acid , 0 . 35 M HOBt , 0 . 25 M diisopropylcarbodiimide ) and allowing the mixture to react for 30 min at RT . Precipitates were removed by a short centrifugation step and 0 . 2 µl of these N-α-Fmoc-protected amino acid active esters were applied to the respective synthesis areas . Coupling of each amino acid was repeated 3 times and a minimum of 40 min reaction time was allowed in each synthesis cycle . After the last cycle the peptides were N-terminally acetylated with 2% ( v/v ) Ac2O in DMF ( 15 ml ( 0 . 13 ml/cm2 membrane ) , 20 min , RT , rocking ) . Side chain protecting groups ( except for Acm ) were removed by immersing the membranes twice in a cleavage cocktail ( 50% ( v/v ) trifluoroacetic acid , 3% ( v/v ) triisobutylsilane , 2% ( v/v ) water in dichloromethane , 0 . 09 ml/cm2 membrane , 1 h each , RT , rocking ) . Subsequently , membranes were washed 4× with dichloromethane , 3× with DMF , 4× with 1 M acetic acid and finally 3× with 100% ethanol ( each 0 . 13 ml/cm2 membrane , 3 min , RT ) . Membranes were air-dried , desiccated overnight in vacuo and stored in the presence of desiccant at −20 °C . To obtain a fluorescent GM1 without modification of its carbohydrate part which is essential for ligand binding , we substituted the generic fatty acid of the ganglioside's ceramide moiety with a dark red fluorescent dye . Lysoganglioside GM1 ( lysoGM1 ) was chosen as starting material as it already lacks the fatty acid residue and contains a free amino function instead . For preparation of the lysoGM1/DY650 conjugate , 200 µl of 2 mg/ml lysoGM1 in dry DMF were mixed with 225 µl of 2 mg/ml DY650-NHS in dry DMF , 7 µl of diisopropylethylamine were added , and the mixture was incubated for 36 h at 30°C under an argon atmosphere . The solvent was removed in vacuo , the residue was dissolved in 500 µl of acetonitrile/water ( 20% ( v/v ) ) and purified by HPLC on RP-18 silica gel ( A = water , B = acetonitrile , 20% B to 70% B , 1 ml/min , 60 ml , retention volume 40 ml ) . For screening of the peptide libraries , the cellulose membranes carrying the peptide arrays were blocked with casein/PBS ( 1% casein in Dulbecco's PBS ( D-PBS ) ( 1 . 47 mM KH2PO4 , 8 . 10 mM Na2HPO4 , 137 mM NaCl , 2 . 68 mM KCl , pH 7 . 4 ) for 3 h at RT and incubated over night at RT with a solution of lysoGM1/Dy650 ( approx . 10 ng/ml ( approx . 5 nM ) ) in casein/PBS . Then they were washed at RT 6×10 min with D-PBS-Tween ( D-PBS containing 0 . 5% ( v/v ) Tween20 ) , 2×10 min with D-PBS and the bound lysoGM1/DY650 was quantitated on the wet membranes in an Odyssey Infrared Imager ( excitation wavelength 680 nm , emission wavelength 700 nm ) . All experiments were performed in duplicate using 2 identical libraries that had been synthesized in parallel . Peptide sequences for alanine scan experiments were synthesized on cellulose membranes as described above . All sequence-positions in the L- and D-lead peptides were replaced with L-alanine for the L-peptides and D-alanine for the D-peptides , and the “point-mutated” sequences were probed for their lysoGM1/DY650 binding capacity as described above . An influence on the GM1 binding capacity due to an amino acid that had been replaced by alanine was assumed , if the fluorescence signal of the mutated peptide spot was increased or decreased by 50% in comparison to the original sequence . If the exchange of an amino acid residue lead to a decrease in signal of the mutated peptide by 50% the replaced residue was judged to be beneficial in GM1 binding and highlighted by boldface in table 1 . If the exchange of a residue lead to an increase in signal of the mutated peptide by 50% the replaced residue was judged to be disadvantageous for binding in the original sequence and underlined in table 1 . The work flow in the software for the evolutionary optimization of peptides is summarized in Figure 1 and depicted in detail in form of a 10-step-flow chart in Supplementary Figure S1 . L-Peptide lead sequences for the evolution process were identified ( step 1 ) by manual inspection from a library of 64 alleged GM1 binding peptides ( Supplementary Table S1 ) which had been screened with fluorophore-labeled lysoGM1/Dy650 [17] . Eleven peptides ( μ = 11 ) displaying a GM1-probe binding affinity above background , as determined by applying a statistically defined cut-off value [18] , were character encoded using a 1-letter code for the respective amino acids ( step 2 ) , assigned fitness-values ( step 3 ) according to their fluorescence signals in the assay and entered as lead peptides into the evolution process . Different parameter settings in the EA - fitness scaling function , parent population size , crossover rate , number of fracture sites and mutation rate - were worked out in an empirical simulation study based on a “pseudo” fitness function and artificial peptide motifs and optimized experimentally in the first generations of the evolution process . The initial gaussian configuration interval of probabilities for recombination and mutation was equal for all positions in the peptide sequences [23] . During the evolutionary process , each setup of the recombination and mutation operators was generated within the pre-configured gaussian distribution interval [23] . As a starting point a crossover rate of 100% with one fracture site in each sequence and a mutation rate of 7% were chosen . In generation 1 two subpopulations of λ = 200 peptide sequences each were generated by applying different scaling functions to the fitness values ( x ) of the lead peptides ( step 4 ) ; the first subpopulation ( gen1sq ) was created by using a square fitness scaling function:whereas the second subpopulation ( gen1ex ) was generated by using an exponential scaling function:Both functions have been chosen to markedly scale the fitness values within the entire dataset range and therefore increase the selection pressure . Generation 2 ( gen2 , λ = 200 peptides ) was created by using the μ* = 16 sequences with the highest fitness values from each , gen1sq and gen1ex , as parents . From this point onward , only the square fitness scaling function was used . In generation 3 the influence of the size μ* of the parent population resulting from manual inspection of the peptides ( step 9 ) was investigated . For that purpose , a first subpopulation ( gen3a , λ = 200 peptides ) was created from μ* = 32 good parents out of gen2 while a second subpopulation ( gen3b , λ = 200 peptides ) was based on the μ* = 20 best sequences from gen2 . The μ* = 25 sequences with the highest fitness values from gen3a and gen3b were taken as parent peptides for the creation of generation 4 . For all the following generations ( gen5–gen10 , λ = 200 peptides each ) , always the μ* = 25 best motifs of the entire population were used as parents . From generation 6 onwards ( gen6–gen10 , λ = 200 peptides each ) the mutational rate was elevated to 12% but the crossover rate of 100% and the single fracture site were kept . For identification of D-peptide lead sequences , the 64 alleged GM1 binding peptides were synthesized in form of their retro-inverso peptide isomers and the resulting library of D-peptides was analyzed as described for the L-peptides . The D-peptide lead sequences were selected accordingly ( Supplementary Table S1 ) . For the optimization process of the D-peptides , the parameter settings were adopted from the procedure conducted with the L-peptides ( crossover rate 100%; mutation rate 7% in gen 1–5 , 12% in gen 6–10; square scaling function ) . From the μ = 13 lead peptides selected as a starting population , λ = 200 filial peptides were generated in gen1 . The μ* = 17 fittest peptides from gen1 were chosen as parent peptides for gen2 , for all the following generations ( gen3–gen10 , λ = 200 peptides each ) the best μ* = 25 peptides were selected as parent sequences . For the L- and the D-peptides the evolutional process was continued until a consensus motif was reached ( step 10 ) and the evenness in the populations had reached a low plateau ( Figure 3 ) . All peptides synthesized and analyzed in this evolutionary process are listed in Supplementary Table S2 . In all populations of D- and L-peptides , the evenness was calculated which is a measure of the population diversity . To do so , the frequency f of any individual amino acid ( i ) at each sequence position in a population of peptides was counted , and the species richness S ( number of different amino acids which occur in total at this position ) was determined . The individual amino acid frequency fi was divided by the population size ( N ) ( i . e . number of peptides in the respective population ) to compute Pi ( Pi = fi/N ) which is the relative frequency of an amino acid in a given sequence position . With the relative frequency P for each amino acid ( i ) which is present at this sequence position , the Shannon Index ( H ) can be determined:On the basis of the Shannon index H the Pielou evenness index ( E ) ( E = H/Hmax ) [25] can be calculated with Hmax being the value for H in case that all amino acids available are present and equally distributed at this sequence position ( Hmax = lnS ) . The evenness index E can assume values between 1 ( all amino acids present in equal numbers at the respective sequence position , i . e . “even distribution” of all amino acid species ) and 0 ( only one amino acid present at the respective sequence position , i . e . “maximally uneven distribution” of the amino acid species ) . For the 16 mer peptides the evennesses per position were summed up to calculate the population evenness .	A clever identification procedure is crucial when peptidic ligands for diagnostic and therapeutic techniques such as in vivo imaging or drug targeting are to be developed . Here , we present a propitious and versatile approach for the discovery of peptide sequences with custom features that is based on an iterative computer-assisted optimization process . The methodology smartly combines in silico evolution with in vitro testing to quickly obtain promising peptide ligand candidates with desired properties . To validate our method in a proof of concept we tried to identify peptide sequences that can bind to a glycosidic cell membrane component . We applied the evolution process by starting out with a small population of peptide lead sequences and achieved a constant increase in affinity between the peptide candidates and their target molecule with each generation . After 10 rounds and a total number of only 4400 peptides synthesized and tested , a more than 100fold improvement in target recognition could be achieved . Since all kinds of building blocks useable in chemical solid phase peptide synthesis can in principle be employed in this evolutionary optimization process , our method should prove a most versatile approach for the optimization of peptides , peptoids and peptomers towards a preset functionality .	[ "Abstract", "Introduction", "Results/Discussion", "Materials", "and", "Methods" ]	[ "biomacromolecule-ligand", "interactions", "algorithms", "biochemistry", "computer", "science", "synthetic", "peptide", "biology", "proteomics" ]	2012	Molecular Evolution of Peptide Ligands with Custom-Tailored Characteristics for Targeting of Glycostructures
DNA mismatch repair greatly increases genome fidelity by recognizing and removing replication errors . In order to understand how this fidelity is maintained , it is important to uncover the relative specificities of the different components of mismatch repair . There are two major mispair recognition complexes in eukaryotes that are homologues of bacterial MutS proteins , MutSα and MutSβ , with MutSα recognizing base-base mismatches and small loop mispairs and MutSβ recognizing larger loop mispairs . Upon recognition of a mispair , the MutS complexes then interact with homologues of the bacterial MutL protein . Loops formed on the primer strand during replication lead to insertion mutations , whereas loops on the template strand lead to deletions . We show here in yeast , using oligonucleotide transformation , that MutSα has a strong bias toward repair of insertion loops , while MutSβ has an even stronger bias toward repair of deletion loops . Our results suggest that this bias in repair is due to the different interactions of the MutS complexes with the MutL complexes . Two mutants of MutLα , pms1-G882E and pms1-H888R , repair deletion mispairs but not insertion mispairs . Moreover , we find that a different MutL complex , MutLγ , is extremely important , but not sufficient , for deletion repair in the presence of either MutLα mutation . MutSβ is present in many eukaryotic organisms , but not in prokaryotes . We suggest that the biased repair of deletion mispairs may reflect a critical eukaryotic function of MutSβ in mismatch repair . DNA mismatch repair ( MMR ) is a major repair system in organisms ranging from bacteria to humans . The discovery that MMR defects cause the most common form of inherited colon cancer underscored the importance of this repair pathway to human health [1]–[6] . In eukaryotes , MMR involves recognition of mismatches created during replication by protein complexes that are homologues of bacterial MutS , followed by downstream processing events involving homologues of bacterial MutL [7]–[9] . There are two main recognition complexes , MutSα , a heterodimer consisting of Msh2 and Msh6 that recognizes base-base mismatches and small loops , and MutSβ , a heterodimer consisting of Msh2 and Msh3 that recognizes mainly loops [7]–[10] . The exact role that MutSβ plays in MMR is not clear . Loss of MutSβ causes only a weak mutator effect unless the assay is specific for insertion or deletion ( in/del ) mutations [11] , [12] . In general , there seems to be much less MutSβ protein than MutSα protein in yeast and human cells [13]–[15]; however , a recent report suggests that the relative amounts of MutSα and MutSβ vary in mouse tissues , with some tissues containing more MutSβ than MutSα [16] . A number of organisms such as Drosophila melanogaster and Caenorhabditis elegans apparently have no MutSβ although they have MutSα [17] . Most analysis of MutSβ MMR function has tended to center on its repair of loops compared to the repair of base-base mismatches by MutSα . However , two early studies of MutSβ and microsatellite instability in yeast found a surprising difference in loop repair and loss of MutSβ compared to loss of MutSα [18] , [19] . For example , using an assay for dinucleotide repeat slippage , Sia et al . found more insertions than deletions in wild-type cells , whereas complete loss of MMR resulted in approximately equal numbers of insertions and deletions; strikingly , cells containing only MutSα had many more deletions than insertions whereas cells containing only MutSβ had many more insertions than deletions [18] . The authors concluded that loops on the primer strand were repaired differently from loops on the template strand . The role of MutL proteins in MMR is less well understood , although they act downstream of initial mismatch detection [7]–[9] . In both yeast and mammalian cells , there are three MutL complexes: MutLα , MutLβ , and MutLγ [7]–[9] . Downstream processing usually involves MutLα , in yeast a heterodimer of Mlh1 and Pms1 [7]–[9] . In yeast , it appears that both MutLβ ( consisting of Mlh1 and Mlh2 ) and MutLγ ( Mlh1 and Mlh3 ) play a role in correction of deletion mutations , although the effect is minor and depends on a sensitive assay [20] , [21] . Although MutL proteins are not thought to have any specific recognition of mismatches , two mutations in PMS1 , pms1-G882E and pms1-H888R , were shown to result in substantial increases in +1 insertions but had essentially no effect on repair of base-base mismatches or deletions [22] . Biochemical analysis has given no information about how MMR could differentiate between mismatches that would lead to insertion versus deletions . The experiments above that have indicated that MMR might repair insertion and deletion loops differently have been rather limited , and we wished to examine in/del mutagenesis in an environment in which sequence context , transcriptional strand , and replication strand could be controlled . We had previously found that we could generate insertion mutations of various sizes and compositions in vivo via single-strand oligonucleotide ( oligo ) transformation that was subject to MMR [23] . In that case , the oligos produced loops on the primer strand of replication that in the absence of repair led to insertion mutations; we had not tested whether oligos could induce loops on the template strand of replication that would lead to deletion mutations . Here we show that , in the absence of MMR , oligo transformation can be used to induce template-strand loops that lead to deletion mutations ( deletion loops ) with essentially the same efficiency as primer-strand loops that lead to insertion mutations ( insertion loops ) . Using this assay we find that , when only MutSα is present , insertion loops are repaired with a greater efficiency than deletion loops , whereas in the presence of only MutSβ , insertion loops are poorly repaired , but deletion loops are efficiently repaired . Deletion loops are repaired almost as efficiently in strains containing pms1-G882E or pms1-H888R as in wild type strains , whereas insertion loops are not repaired . Surprisingly , repair of deletion loops in pms1-G882E or pms1-H888R mutant strains has a major dependence on MutLγ . Our data indicate that the biased repair of insertion versus deletion loops is dependent on interactions with the MutL proteins . We suggest that these properties of MMR can best be understood in an evolutionary sense in which MutSα represents the functions of bacterial MutS to repair base-base mismatches and in/del mismatches with a bias toward insertions , whereas MutSβ represents a new function present in some eukaryotes that complements MutSα function with respect to repair of deletion mismatches , chiefly through a different interaction with MutL proteins . We had previously used oligo transformation to study insertion mutations using the lys2ΔA746 frameshift reversion assay that requires restoration of a −1 frameshift in a region of the LYS2 gene indifferent to amino acid sequence [23] , [24] . We wanted to study the effects of MMR on deletion mutations as well as insertion mutations , but , because of the different affinities in binding of loop sizes by MutSα and MutSβ , it was necessary to compare the effects of insertion and deletion mismatches of the same size , requiring the use of two complementary reversion assays . We therefore used both the −1 lys2ΔA746 frameshift allele and the +1 frameshift allele lys2ΔBgl in the same LYS2 region [23]–[25] . In order to have reversion windows with known orientations relative to a dependable origin of replication and to have the different frameshift alleles as similar as possible , we used the LYS2 genes inserted in both orientations ( “same” and “opposite” ) at the HIS4 locus previously described [26] and inserted the frameshift alleles as described in MATERIALS AND METHODS . The −1 lys2ΔA746 frameshift allele was used to study +1 and −2 loops , and the +1 frameshift allele lys2ΔBgl was used to study −1 and +2 loops . The overall scheme for the assay is illustrated in Figure 1 . The efficiency of recognition by MMR is known to be dependent not only on the mispaired bases , but also on the sequence context surrounding the mispaired bases [10] , [27] . Therefore we used a collection of oligos that created different mispairs , in two sequence contexts ( Figure 1B ) . Because we transform with single-stranded oligos , the oligos can have the sequence of the transcribed strand and create a TC or GA insertion loop or have the sequence of the nontranscribed strand and create a TC or GA insertion loop , all in otherwise the same sequence context . This was done in two different locations within the reversion windows of the lys2 mutant alleles . Deletion loops are created by transforming with oligos lacking certain bases contained in the template strand; therefore different deletion loop sequences cannot be created in the same sequence context using the same strains . As detailed in Materials and Methods , an oligo was transformed into a given strain in three independent experiments , and the average number of transformants over background reversion events was determined . All oligos were transformed into two strains with opposite orientations of the LYS2 gene relative to the nearby origin of replication so that the effect of loops on the leading versus lagging strand could be assessed . The results of transformation with a selected set of oligos in strains containing or lacking certain components of MMR are given in Figure 2 and the full set of results is given in Figure S1 . Several patterns can be observed in the results in strains lacking MMR ( msh2 ) . Each pair of oligos differing only in the insertion bases gave results that were generally not statistically different from one another . Comparing any set of oligos ( e . g . TrL1-Lag-s ) , the difference between insertions and deletions was generally less than 2-fold , with a mixture of insertions , deletions , or neither predominating . Finally , as we have observed previously [23] , [28] , in all cases the number of insertions or deletions was greater when targeted to the lagging strand than to the leading strand , by an average of approximately 6-fold in these experiments . Therefore we can conclude that , in the absence of MMR , insertion and deletion mutations can be created at approximately equal efficiencies by oligos . In contrast to oligo transformation in the absence of MMR , one can see quite different patterns of transformation in strains containing only MutSα ( msh3 strains ) or MutSβ ( msh6 strains ) in Figure 2 . To compare the effect of MMR on transformation , we divided the average number of revertants obtained in the absence of MMR by the average number of revertants obtained in a given MMR background to give a Repair Ratio ( Table 1 ) . The larger the Repair Ratio , the more effectively the loop created by the oligo was removed . The results in strains containing only MutSβ ( msh6 strains ) are very consistent , as can be seen in Figures 2 and S1 and Table 1 . In every case , deletion mispairs were corrected much more efficiently than insertion mispairs; in Table 1 , the Repair Ratios for insertions range from 1 to 13 and for deletions from 59 to 310 . The results in strains containing only MutSα ( msh3 strains ) were more varied . Uniformly , deletion mispairs are poorly repaired , with a range of Repair Ratios of 2 to 9 ( Table 1 ) . Insertion mispairs are repaired with a wide range of efficiencies of 2 to 130 ( Table 1 ) . The one consistent difference is that within the same sequence context , a GA sequence in the loop is always repaired more efficiently than a TC . However , when only MutSα is present , insertion loops are repaired overall with much greater efficiency than deletion loops . Additionally , in the presence of only MutSα , insertion loops are repaired with somewhat greater efficiency when the loop is on the lagging strand compared to the leading strand , with an average ratio of 1 . 6 , whereas , when only MutSβ is present , deletion loops on the leading strand are repaired 1 . 6-fold more efficiently than on the lagging strand , a difference we previously found under other circumstances [23] . The difference between these two ratios is statistically significant as determined by a Mann-Whitney rank sum test ( P = 0 . 038 ) . A median measure of the insertion and deletion loop Repair Ratios is given in Table 2 , which illustrates the differing biases of MutSα and MutSβ . There is inherently more error associated with measurement of revertants in cells that are wild type for MMR , as the number of revertants can be decreased by over two orders of magnitude to quite low numbers . However , a consistent pattern emerges as observed both in Figures 2 and S1 and in Tables 1 and 2: 2-nt deletion mispairs are corrected more efficiently than insertion mispairs in strains wild type for MMR . Deletion mispairs are corrected with an efficiency somewhat greater than that of cells containing MutSβ alone ( usually less than two-fold ) , presumably reflecting the ability of MutSα to recognize deletion mispairs , albeit at a much lower efficiency than does MutSβ . Insertion mispairs are generally corrected with an efficiency greater than that observed in cells with MutSα alone , although in 5 cases , insertion mispairs were corrected less efficiently than in MutSα cells , and in one other case about the same ( Table 1 ) . One explanation for those situations could be a dilution in MutSα molecules due to Msh3 pairing with some of the Msh2 [13] , [14] . A dinucleotide repeat stability assay previously showed that 2-nt deletions were repaired with a greater efficiency than insertions in strains wild-type for MMR [18] , [19] . A screen for mutations in PMS1 found two mutants that resulted in large increases in +1 insertions but had no effect on deletions [22] . We tested those mutations in our assay system to see if they would have a similar effect on 2-nt in/del mispairs . The results are shown in Figures 3 and S2 . As described in Materials and Methods , the pms1 ( 761-904 ) Δ mutant was a precursor in construction of the two PMS1 point mutations; terminal deletions of that length have previously been shown to be nonfunctional [29] . Pms1 is needed for most repair , as the pms1 ( 761-904 ) Δ strains behave similarly to the msh2 strains . However , the msh2 strains generally had more transformants , averaging 1 . 7-fold more insertions and 2 . 5-fold more deletions ( Table S1 ) , suggesting that some in/del repair might be mediated by complexes lacking Pms1 . Strains containing either of the two PMS1 point mutations show an extreme difference in repair of insertion versus deletion mispairs that is evident in Figures 3 and S2 and given quantitatively in Tables 1 and S1 . Both mutant strains repair deletion mispairs but have little effect on insertion mispairs . The median effect of each mutation is presented in Table 2 . The effect of the two mutations , pms1-G882E and pms1-H888R are similar , but the pms1-H888R mutants appear to have a more distinctive effect , with almost no repair of insertion mispairs but more repair of deletion mispairs than the pms1-G882E mutants . Because there is very little repair of deletion mispairs in the absence of Pms1 ( Table S1 ) , the pms1 point mutants must be functional in deletion repair . Previously , the evidence for the differential effect of MutSα and MutSβ on in/del mutations came from a dinucleotide repeat assay , although an assay using one particular mononucleotide repeat indicated that the loss of either MutSα or MutSβ led to an increase mainly of deletions [18] . The pms1-G882E and pms1-H888R mutations had only been examined with mononucleotide repeats [22] . Therefore we wanted to examine whether the effects we observed on 2-nt in/del mispairs would be observed in similar 1-nt in/del mismatches . For that survey , we used only oligos in one location , and the results are presented in Figures 4 and S3; quantitative comparisons are given in Table 3 . There are similarities to the results with 2-nt in/del mismatches in terms of the opposing biases for insertions versus deletions , but the quantitative results differ , presumably due to the relatively greater affinity of MutSα recognition for 1-nt loops over 2-nt loops , and the correspondingly lower recognition of MutSβ for 1-nt loops compared to 2-nt loops . MutSα has an overall much greater effect on suppression of 1-nt in/del mismatches than does MutSβ , and MutSα has substantial activity on 1-nt deletion loops in contrast to its activity on 2-nt deletion loops ( Figures 4 and S3 ) . Even so , MutSα has a consistently greater activity toward 1-nt insertion mismatches , whereas the MutSβ activity is the reverse . In contrast , the pms1-G882E and pms1-H888R mutants have about the same lack of insertion repair as exhibited on 2-nt in/del mispairs ( Tables 3 , S2 ) . However , deletion repair in the pms1-G882E and pms1-H888R mutants is much more efficient than that in strains containing only MutSβ , indicating the involvement of MutSα in 1-nt deletion loop repair . The median Repair Ratios are given in Table 4 and illustrate that in contrast to the situation with 2-nt loops , there is relatively more repair of deletions with MutSα only and relatively less repair of deletions with MutSβ only , and in wild-type cells insertions and deletion mismatches are corrected with indistinguishable efficiency . How can the specificity of the pms1-G882E and pms1-H888R mutations best be understood ? The pms1-G882E and pms1-H888R mutant strains appeared to be similar to msh6 strains lacking MutSα for 2-nt deletion repair; we therefore examined strains containing both msh6 deletions and pms1 mutations to determine if they appeared to be in the same pathway . Because the pms1 mutations fail to repair insertion loops , we could only examine the effect on deletion loop mispairs . The results are given in Figures 5 and S4 and Tables 1 , 2 , and S1 for 2-nt deletions . Results in the double mutants , pms1-H888R msh6 and pms1-G882E msh6 , are not distinguishable from the single pms1 mutant results ( Table 2 ) . For 1-nt deletion loop repair , MutSα is much more important than in 2-nt loop repair and as noted above , the repair in the pms1 mutants is more efficient than in the presence of only MutSβ . Repair of 1-nt deletion loops in the double pms1-G882E msh6 and pms1-H888R msh6 mutants is much lower than in the single pms1 mutants ( Figure 4; Tables 3 , S2 ) , indicating that much of the deletion loop repair in the pms1 mutants must be due to the action of MutSα ( Table 4 ) . In order to determine if MutLγ ( composed of Mlh1 and Mlh3 subunits [21] ) might be involved in some of the observed repair , we examined strains with an MLH3 deletion . The results for 2-nt in/del mispairs are given in Figures 5 and S4 and Tables 2 and 5 . It is evident that MutLγ is not involved in repair of insertion mispairs , as Repair Ratios actually increased in the absence of MutLγ ( P = 0 . 019 ) ( perhaps due to a somewhat increased amount of MutLα ) . An mlh3 deletion resulted in an approximately 4-fold decrease in repair of deletion mispairs ( Table 2 ) ( P = <0 . 001 ) . Those results were expected given the limited effect previously found for mlh3 deletions [10] , [20] , [21] . The pms1-H888R mutation has less than a 2-fold effect on deletion repair , so one would have anticipated that the double mutant would be similar to the mlh3 mutant . Such was not the case as seen in Tables 2 and 5 . The double mutant had an almost 13-fold reduction in deletion repair compared to wild type . The difference between repair in mlh3 and pms1-H888R mlh3 strains is significant , with P = <0 . 001 . The same pattern was found in 1-nt in/del mismatch repair . A single mlh3 deletion has a relatively small effect on in/del repair , slightly raising the efficiency of insertion repair compared to wild type and slightly decreasing the efficiency of deletion repair , although the difference in both cases is marginally significant ( P = 0 . 05 ) ( Tables 3 and 4 ) . The pms1-H888R mutant has robust deletion repair , but the double mutant pms1-H888R mlh3 was reduced by 20 fold in deletion repair ( the difference is significant , with P = 0 . 029 ) ; deleting msh6 had no further effect ( Tables 3 and 4 ) . This result was particularly surprising , as MutSα is responsible for much of the 1-nt deletion repair and yet MutLγ has been thought to work only with MutSβ [21] . The biases we find here for repair of in/del mispairs had been previously observed in two different systems: a dinucleotide repeat assay for MutSα and MutSβ [18] , [19] and frameshift reversion assays for the pms1 mutants [22] . Given the limited scope of each of those experiments , it was not clear whether the results reflected a general property of the proteins involved , or were influenced by the DNA sequences involved in the particular assays used . Our results with a completely different assay system and with a variety of different sequences and gene strands and orientations lend confidence that our observations reflect an inherent difference in repair of insertion versus deletion loops by MMR . For 2-nt in/del mismatches , strains containing only MutSβ provide the clearest picture of a bias . As shown in Table 1 , the repair of all insertion loops tested is poor , ranging from 1 to 13-fold , and the repair of all tested deletion loops is robust , ranging from 60 to 300-fold . Although MutSβ has a measurable effect on repair of most insertions , it is only deletions for which it has a substantial effect . The effect in strains containing only MutSα is a bit more complex . The repair of deletion loops is uniformly low , ranging from 2 to 9-fold ( Table 1 ) . The repair of insertion loops is much more variable , with most ( 13/16 ) being repaired by a factor of 15 to 200-fold . The greater variability of repair initiated by MutSα compared to MutSβ is presumably a function of a stronger effect of sequence specificity [27] . The median Repair Ratios calculated in Table 2 indicate the remarkable difference between in/del repair mediated by MutSα and MutSβ although the individual data in Table 1 serve as a useful reminder that the particular repair of a given sequence can be quite variable . Our analysis of repair of 1-nt in/del mispairs was not as extensive as that for 2-nt in/del mispairs , but the results we obtained reveal a similar pattern ( Tables 3 and 4 ) . Presumably due to the greater affinity for MutSα for 1-nt in/del mismatches compared to 2-nt in/del mismatches ( and the converse for MutSβ ) the overall effect of MutSα on repair of both insertion and deletion mismatches is much greater than for 2-nt in/del mismatches , and the effect of MutSβ for 1-nt deletion mismatches is much less than for 2-nt mismatches ( Table 4 compared to Table 2 ) . However , one can see that for all tested combinations , insertion loops are repaired more efficiently than deletion loops when only MutSα is present , and the reverse when only MutSβ is present . In this context , it is interesting to observe the overall effect on in/del mismatches as observed in strains wild type for MMR . For 2-nt in/del mismatches , one sees that on average deletion mispairs are repaired significantly better than insertion mispairs ( Table 2 ) , whereas for 1-nt mispairs , there is no consistent bias in repair ( Table 4 ) , reflecting the relatively greater effect of MutSα on repair . How can the difference in repair of insertion versus deletion loops be explained ? For that question , the existing evidence from biochemistry is not very helpful , as no biochemical experiments have been done in which the strands of duplex DNA that have been used could be identified as primer or template strands in a replication complex . Recent structural studies reveal that MutSβ binding to in/del mismatches takes place in a very different manner from MutSα or MutS binding to mismatches [30] , [31] . Biochemical experiments have measured binding affinities , but what we measure here are overall repair efficiencies , for which the binding of MutSα or MutSβ is just the first step . After MutSα or MutSβ binding , the next step in MMR is an association with a MutL protein complex , which is usually MutLα . If the repair efficiencies we measure were purely the result of MutSα or MutSβ binding efficiencies , then we would expect that any mutant of MutLα would have equivalent effects on in/del mismatch repair . However , the PMS1 mutations we characterize , pms1-G882E and pms1-H888R , show extreme bias in in/del mismatch repair . As observed in Table 1 , the effects of both mutations are approximately the same , with the pms1-H888R mutation showing a bit stronger effect . Strains with the pms1-H888R mutation show almost no repair of any insertion , but very robust repair of deletion mispairs , ranging from 60 to 460-fold . A similar effect is seen for 1-nt in/del mispairs ( Table 3 ) . The initial characterization of those two mutations showed that they had only a modest effect on overall mutation rate and that their mutator effect was primarily on frameshift deletions [22] . Our results with 2-nt in/del mispairs suggested that the pms1-H888R mutant behaved very much like strains containing only MutSβ , and the double-mutant strains did not seem to be significantly different from either of the single mutants ( Table 2 ) . However , as seen in Table 4 , the situation is quite different with 1-nt in/del mispairs , as the double mutant strains are much worse at repair of deletions than either pms1 mutant strain , indicating that much of the deletion repair in the pms1-H888R or pms1-G882E mutant is due to MutSα activity . It is not clear how the two pms1 mutants affect MMR . The two pms1 mutations map into a region described as an Mlh1-interaction region [29] , but the interaction of the mutant proteins with Mlh1 was not found to be defective as judged by a two-hybrid assay [22] . A recent structure of the S . cerevisiae MutLα C-terminal domain permits a much better understanding of the location of the mutations within the MutLα protein [32] . At the time of the Erdeniz et al . paper , the initiating ATG codon was thought to be in a location such that the length of the Pms1 protein would be 904 aa . However , a genomic analysis found a different ATG codon to be the correct initiation site for translation , leading to a predicted protein length of 873 aa [33] . With that numbering , the two Pms1 mutations would be G851E and H857R . The crystal structure shows that the H857 residue is centered in the β8 β-sheet that is part of one of the most important regions of the heterodimerization interface , Patch 1 [32] . The G851 residue sits just outside the β8 β-sheet and so it is reasonable to suppose that a mutation in that residue could affect Pms1-Mlh1 interaction . One of the zinc atoms in the endonuclease site is stabilized by C848 and H850 [32] . That would put the G851 residue close to the endonuclease site , making it possible that the G851E substitution might interfere with the binding of the zinc atom and thus affect endonuclease activity . However , there is no indication in the structure that the H857 residue would influence endonuclease activity , and as we found above , the H857R mutation has a more distinctive mutator effect than the G851E mutation . Both of the mutations were found to have essentially wild-type base-base MMR activity [22] , and as Pms1 endonuclease activity is crucial for MMR function [34] , we consider it highly unlikely that the effects of the two mutations is on the endonuclease activity of Pms1 . In accordance with previous results , we find that the absence of Mlh3 leads to somewhat less effective repair of deletion mispairs ( Tables 2 and 4 ) [20] , [21] . The repair of insertion mispairs in an mlh3 background is more robust than in a wild-type background , suggesting that the loss of Mlh3 might lead to a somewhat greater amount of MutLα in the cell , with correspondingly greater repair of insertions . That view is consistent with the previous observation that overexpression of Mlh3 appears to result in lower levels of MutLα [35] . The surprise was the deletion repair observed in pms-H888R mlh3 mutants . Given the small effect of each individual mutation on deletion repair , one would have expected deletion repair to be robust in that mutant background . Instead , repair of both 1-nt and 2-nt deletion mispairs was synergistically compromised ( Tables 2 and 4 ) . Based on the prior results with the pms1-H888R mutants , it appeared that only insertion repair was compromised [22] . Our results suggest a different possibility: although the pms1-H888R mutant is functional for base mismatch repair , it functions relatively poorly in in/del mismatch repair . One possible explanation for this hypothesis involves the finding that MutS complexes recognizing mismatches are responsible for loading multiple copies of MutLα onto DNA [36] . MutSα recognizing a base-base mispair can interact with the pms1-H888R mutant to create a functional complex . However because of the orientation of the proteins mediated by their binding to PCNA , neither MutSα nor MutSβ when recognizing an insertion mispair can interact properly with the pms1-H888R mutant complex and there is very little insertion repair . When MutSβ recognizes a deletion mispair , the complex is positioned so that it is able to interact with the pms1-H888R mutant MutLα , although relatively poorly , giving Repair Ratios of 16–20 ( Tables 2 and 4 ) . This interaction is facilitated by MutLγ interacting with MutSβ , which then helps recruit multiple molecules of the pms1-H888R mutant complex . Repair of 2-nt deletion loops by MutSα is poor ( Repair Ratio of 5 . 5 , Table 2 ) ; however repair of 1-nt deletion loops by MutSα is much more robust ( Repair Ratio of 73 , Table 4 ) , although still less than insertion loop repair . Repair of 1-nt deletion loops in the pms1-H888R mutant is much greater than repair with only MutSβ present ( Repair Ratio of 170 compared to 22 , Table 4 ) , suggesting that much of the repair in the pms1-H888R mutant must be by MutSα . The fact that repair of 1-nt deletion loops in the pms1-H888R mlh3 background drops to the level of repair when only MutSβ is present suggests that MutSα-directed repair in the presence of the pms1-H888R mutation involves MutLγ . The very modest effect of the mlh3 mutation by itself shows that normal MutSα-directed repair of 1-nt deletion loops does not use MutLγ; confirmation of this suggestion would require additional experiments . One issue that has not been clear from previous experiments because of the modest effect of MutLγ on repair is whether there were certain mismatches that required MutLγ function , perhaps instead of MutLα , or whether the action of MutLγ always required MutLα and any mismatch was potentially susceptible to MutLγ function . Because each of our assays examines only one particular mismatch and because we see a strong effect in the mlh3 pms1-H888R background , we can draw several conclusions . 1 ) MutLγ functions only in repair of deletion loops and not insertion loops . 2 ) Any deletion loop is susceptible to being aided in repair by MutLγ . 3 ) MutLγ-mediated repair also requires MutLα . These conclusions do not mean that the effect of MutLγ deletion would be the same for all deletion loops: for both 1-nt and 2-nt deletion loops there is a range of about 4-fold in Repair Ratios , suggesting that certain mismatches could be more dependent for MutLγ on their repair . The above model , while compatible with our results , makes several predictions that may however prove difficult to study . The first is that the bias in repair of insertions compared to deletions is ultimately a function of the MutL complexes and not the recognition by MutS complexes . A role for MutLγ in the repair of some deletion mispairs had previously been detected [20] , [21] , so the idea that MutL complexes could be biased in in/del repair is not without precedent . Secondly , the bias observed in in/del repair mediated by MutSα and MutSβ indicate that they contact MutLα differently such that a deletion mispair recognized by MutSβ is more likely to be repaired than if the same mispair were recognized by MutSα , and vice versa for insertion mispairs . A major question then is how the MutS and particularly the MutL components could be oriented such that an insertion mispair was recognized differently from a deletion mispair . An important part of the explanation likely involves interactions of MMR proteins with the proliferating cell nuclear antigen , PCNA . PCNA is one of a family of DNA sliding clamps that encircles DNA , is essential for replication , and has binding sites for many proteins , including the replicative polymerases [37] and there is evidence that it can act as a scaffold to coordinate MMR through consecutive protein-protein interactions [38] . PCNA is required for MMR at a step preceding DNA resynthesis [39] , [40] , and MMR interactions with PCNA could be responsible for strand discrimination [41] , [42] . A variety of experiments demonstrated direct interactions of PCNA with Mlh1 , Msh3 , and Msh6 , and those interactions were important for proper MMR [40] , [43]–[46] . It is clear that interaction with PCNA is not sufficient to drive MMR , as there are other processes occurring . For example , engineering a mutation that blocked MutSα conformational change upon mismatch binding demonstrated that such change was necessary for MutLα binding [47] . PCNA is asymmetrical with respect to the replication fork , and this asymmetry can result in specific MutLα loading and subsequent endonucleolytic activation and thus proper strand discrimination as has been observed in human MMR [42] . Importantly for this work , experiments with various PCNA mutants suggested that the interactions of PCNA are different for Msh3 compared to Msh6 [48] . In addition , it has been recently shown in humans that in contrast with MutSα , PCNA and MutLα have the same binding site on MutSβ , suggesting that the interaction of MutSβ with PCNA and MutLα would be sequential [49] . These considerations suggest a mechanism by which the recognition of , for example , an insertion loop could be different for MutSα compared to MutSβ because of their different orientation to the duplex bulge due to their different PCNA interaction . It is not clear how subsequent interactions with MutL complexes are handled . In vitro studies suggest that MutSα is bound to PCNA on homoduplex DNA , and , when a mispair is encountered , the interaction with PCNA is either lost or changed [50] . The next step of interaction with MutL complexes could be sequential for both MutS complexes , with a loss of the MutS interactions [38] , but given the different nature of the MutS complex interactions with PCNA [49] , the nature of the interactions of MutSα and MutSβ with MutLα is likely to be very different . It is surprising to find that insertion and deletion mispairs are repaired with differing biases and that MutSα and MutSβ exhibit opposite biases for such repair . What might account for the development of an MMR system that would function in such a manner ? A recent analysis was done of multiple strains of over 40 bacterial and archaeal species . It was found that in species with no MMR system , expansions and contractions of simple sequence repeats were equally likely , whereas in species containing MMR systems , there was a bias toward contraction of simple sequence repeats [51] . Thus , it appears that bacterial and archaeal MMR systems , like yeast strains containing only MutSα , repair insertions better than deletions . It is possible that such a bias could have an evolutionary advantage , tending to reduce the length of simple sequence repeats . Although most eukaryotic species seem to have an MMR system , not all have a MutSβ; in fact two favorite model organisms , D . melanogaster and C . elegans , lack MutSβ , although they both have MutSα [17] . Structural evidence also shows that MutSα binds mismatches in a manner similar to MutS , whereas MutSβ binds mismatches quite differently [30] . This analysis would suggest that MutSα represents the bacterial MutS activity , whereas MutSβ represents a new activity in which the bias toward repair of deletion mispairs may have been equally or more important than the recognition of larger loops . Many eukaryotic organisms have abundant simple sequence repeats , including those in exons , and the addition of a more robust activity repairing potential deletion mispairs would help preserve those repeats in the genome . This new MutSβ activity , due to the MSH3 gene , not only had a recognition specificity different from that of MutSα , but interacted in a somewhat different manner with PCNA and MutLα and the new MutLγ complex that apparently does not usually interact with MutSα [21] . Domain swap experiments have shown that the mismatch recognition domain of Msh3 is not necessary for interaction with MutLγ , but rather another part of the Msh3 protein present in MutSβ [52] . Given the high degree of conservation , in both sequence and function , between MMR systems in yeast and mammalian cells , our results likely apply also to mammalian cells , although the experiments to test that are much more difficult to carry out . Repeat stability is a concern for mammalian cells , both in terms of various trinucleotide repeat diseases and in cancer [53] , [54] . In various trinucleotide repeat diseases , there is a strong involvement with MMR , but the effects are complicated [53] . In a mouse model of Friedreich ataxia which has GAA repeats , repeat instability was increased in the absence of MMR and there were enhanced deletions in the absence of MutSβ and an enhancement of both deletions and insertions in the absence of MutSα , with a relatively greater increase in insertions [55] . Those results are consistent with the activities we report here . However , repeat instability of other types of trinucleotide repeats shows a different effect , with MMR appearing to be required for expansion , for example [53] . Although there is not a complete understanding of such effects , many of them involve MutSβ and interactions with larger loops . For example , there are certain types of loops that are repairable by MutSβ and others such as CAG loops in which the loop appears to maintain MutSβ binding , thus preventing repair [56] . However , in an in vitro assay , 1 or 2 repeats of CTG/CAG were repaired in a process requiring MutSβ , but not larger loops , or substrates that contained multiple loops on both strands [57] . Some of the first analyses of MMR genes in humans demonstrated that defects in MMR led to Lynch syndrome or hereditary nonpolyposis colorectal cancer and that such cells manifested a greatly enhanced microsatellite instability [1] , [2] . Although the overall mutator effect of deficiencies in MMR is likely important in tumor formation and progression , genes containing exonic microsatellite sequences are a particularly susceptible target as any alteration in such sequences will likely lead to a strong phenotype [54] , [58] , [59] . Additionally there is some evidence that microsatellite repeats within introns and in 5′ and 3′ untranslated regions could also contribute to carcinogenesis [54] . Not only is the distribution of different tumor types generally different in MMR-defective mice compared to humans , but there is a marked difference depending on the particular defect in MMR [54] , [60] . Our results provide additional information on possible reasons for those differences . Part of the difference between the distribution of tumor types in mouse and human is likely due to the difference in the existence and sequence of regions in cancer target genes susceptible to in/del formation . Although we are able to induce approximately equal frequencies of insertion or deletion mispairs in the absence of MMR , spontaneous formation of primer or template loops could be at least partially a function of sequence , sequence context , and replication on the leading versus lagging strand , thus also implicating the relation of the gene to replication origin . Because there is plasticity in use of replication origins , the same gene could be replicated differently depending on tissue type [61] . Not only could the formation of a loop be influenced by its sequence and location near an origin , but as we have demonstrated previously [23] and also find here , there is a bias in repair by MutSα and MutSβ depending on the replication strand . There is some variability with MutSβ with different oligos , but there is even more pronounced variability with MutSα , with almost a 100-fold difference in repair between the best- and worst-repaired oligo ( Table 1 ) . In both yeast and human cells , there seems to be generally more MutSα than MutSβ in cells , so the likelihood of repair of a given in/del will depend on how well it is recognized by MutSα or MutSβ , which could depend on a variety of factors including sequence and perhaps location , whether it is an insertion or deletion loop , and on which replication strand it appears on . If there turns out to be significant variability in the relative amounts of MutSα and MutSβ in various tissues , as has been found in mouse [16] , the likelihood of repair could depend on tissue type . We demonstrate here the surprising finding that although the recognition of in/del mispairs is due to the MutS complex , it is the interaction with the MutL complex that biases the efficiency of repair of an insertion versus deletion mispair . Thus mutations in the genes encoding MutLα could influence not only the efficiency of repair but its bias in repair of in/del mispairs . The genotypes of strains used in these experiments can be found in Table S3 . All strains were derivatives of SJR2259 and SJR22609 [26] with LYS2 moved into HIS4 location . Mutant lys2 alleles either with [+1] ( lys2SΔBgl and lys2OΔBgl ) or [−1] ( lys2SΔA746 and lys2OΔA746 ) frameshifts were then introduced by two-step allele replacement [62] using plasmids pSR125 [63] or pSR786 [64] respectively . ‘S’ and ‘O’ refer to the orientation of the LYS2 gene - the same or opposite orientation relatively to original HIS4 orientation ( Figure 1A ) . Gene deletions were made using a PCR fragment generated from the collection of yeast gene deletions [65] . The pms1 point mutations were made using the delitto perfetto method [66] . The pCORE cassette was inserted into the PMS1 gene using primers GCP735 and GCP736 ( Table S4 ) creating the pms1 ( 761-904 ) Δ mutant . The pCORE cassette was then replaced by transformation with a PCR product from strain NEY398 or NEY402 [22] using primers GCP737 and GCP738 ( Table S4 ) . Oligos for transformation were gel purified ( Eurofins MWG Operon ) and are listed in Table S4 . Transformation by electroporation was performed essentially as described previously [28] , [67] . An overnight culture of yeast cells ( 0 . 5 ml ) was inoculated into 25 ml of YPAD [68] , incubated with shaking at 30° to an A600 of 1 . 3–1 . 5 , washed twice with cold H2O , and once with cold 1 M sorbitol . After the final centrifugation , all solution was removed from the cells and 150 µL of cold 1 M sorbitol added to resuspend the cells . After addition of 200 pmol oligo and 50 ng of pRS314 [69] plasmid DNA , the solution was mixed and transferred into a 2-mm gap electroporation cuvette and electroporated at 1 . 55 kV , 200 Ω , and 25 uF ( BTX Harvard Apparatus ECM 630 ) . Immediately after electroporation , the cell suspension was added into 5 ml YPAD to recover for 2 h with shaking at 30° . Then cells were centrifuged , washed with H2O , and plated on synthetic dextrose ( SD ) medium lacking lysine [68] . The number of Trp+ transformants resulting from the pRS314 plasmid served as a useful marker of successful transformations , but was not consistent enough to be used as an internal standard for transformation efficiency . In order to determine background reversion , the same strains were electroporated as described but without adding oligos . For each oligo and strain combination , three independent experiments were performed , and the mean and standard deviation of the number of total transformants calculated .	DNA mismatch repair is a major pathway that prevents both base substitution and insertion or deletion errors during replication . Most eukaryotes have two recognition complexes , MutSα and MutSβ , homologues of prokaryotic MutS and differing in their affinity for mismatches , with MutSα recognizing base-base mismatches and small insertion/deletion loops and MutSβ recognizing larger loops . We show that repair mediated by these complexes has opposite biases for insertion versus deletion mispairs with MutSα-directed repair favoring insertion loops and MutSβ-directed repair favoring deletion loops . This bias is mediated by differing interactions with downstream MutL complexes . We suggest that MutSα represents a prokaryotic MutS biased for repair of insertion loops and that MutSβ represents a new eukaryotic activity biased for repair of deletion loops .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[]	2013	Different Roles of Eukaryotic MutS and MutL Complexes in Repair of Small Insertion and Deletion Loops in Yeast
Topoisomerases are enzymes with crucial functions in DNA metabolism . They are ubiquitously present in prokaryotes and eukaryotes and modify the steady-state level of DNA supercoiling . Biochemical analyses indicate that Topoisomerase 3α ( TOP3α ) functions together with a RecQ DNA helicase and a third partner , RMI1/BLAP75 , in the resolution step of homologous recombination in a process called Holliday Junction dissolution in eukaryotes . Apart from that , little is known about the role of TOP3α in higher eukaryotes , as knockout mutants show early lethality or strong developmental defects . Using a hypomorphic insertion mutant of Arabidopsis thaliana ( top3α-2 ) , which is viable but completely sterile , we were able to define three different functions of the protein in mitosis and meiosis . The top3α-2 line exhibits fragmented chromosomes during mitosis and sensitivity to camptothecin , suggesting an important role in chromosome segregation partly overlapping with that of type IB topoisomerases . Furthermore , AtTOP3α , together with AtRECQ4A and AtRMI1 , is involved in the suppression of crossover recombination in somatic cells as well as DNA repair in both mammals and A . thaliana . Surprisingly , AtTOP3α is also essential for meiosis . The phenotype of chromosome fragmentation , bridges , and telophase I arrest can be suppressed by AtSPO11 and AtRAD51 mutations , indicating that the protein is required for the resolution of recombination intermediates . As Atrmi1 mutants have a similar meiotic phenotype to Attop3α mutants , both proteins seem to be involved in a mechanism safeguarding the entangling of homologous chromosomes during meiosis . The requirement of AtTOP3α and AtRMI1 in a late step of meiotic recombination strongly hints at the possibility that the dissolution of double Holliday Junctions via a hemicatenane intermediate is indeed an indispensable step of meiotic recombination . Topoisomerases are enzymes with crucial functions in DNA metabolism . They are ubiquitously present in prokaryotes and eukaryotes and modify the steady-state level of DNA supercoiling [1] . Topoisomerases are required for cellular processes , such as DNA replication , transcription , recombination and chromatin remodeling to release topological forces by cleaving the DNA backbone in a reversible manner . Topoisomerases are sorted into two basic classes that differ in their ability to create either single strand ( class I ) or double strand breaks ( class II ) . They can be subdivided into two classes each , which have been defined either by their chemical properties ( IA and IB ) or by structural differences between the enzymes ( IIA and IIB ) [1] , [2] . There are three topoisomerases in yeast: TOP1 , 2 and 3 . In contrast to TOP1 and 2 , which are well characterized and involved in DNA replication ( TOP1 , type IB ) or decatenation of linked chromosomes ( TOP2 , type IIA ) , the role of TOP3 ( type IA ) is poorly understood . Mutants of TOP3 in Saccharomyces cerevisiae show hyper-recombination , chromosome instability and do not sporulate , whereas top3 mutants in Schizosaccharomyces pombe are lethal [3] , [4] . In higher eukaryotes , two homologues of TOP3 were found and annotated as TOP3α and ß , respectively [5]–[7] . A knockout of TOP3α leads to early embryogenic lethality in mice and Drosophila melanogaster and to pleiotropic effects , such as germ cell proliferation abnormalities in Caenorhabditis elegans [8]–[10] . In contrast , mice mutated for TOP3ß do not exhibit lethality or growth abnormalities , but do possess a shortened lifespan [11] . We recently demonstrated that a T-DNA insertion line of Arabidopsis TOP3α ( top3α-1 ) also exhibits severe developmental defects resulting in lethality of the plantlets [12] . In animals and plants , TOP3α proteins show an intimate physical and genetic interaction with one or more RecQ helicases . A tripartite protein complex indeed exists and consists of i ) a RecQ helicase; ii ) a TOP3α homolog and iii ) a recently characterized protein named BLAP75 ( for Blooms associated protein 75kd , in mammals ) or Rmi1 ( for RecQ mediated instability 1 , in yeast ) that mediates the complex formation of all three proteins [13]–[16] . This complex has also been referred to as the RTR or BTB complex ( for RECQ/TOP3α/RMI or Blooms/TOP3α/BLAP75 , respectively ) , and is able to dissolve recombination intermediates such as double Holliday Junctions ( dHJs ) or disrupt D-loop structures that otherwise lead to cell cycle arrest or cell death in different organisms [17]–[19] . Although ample evidence has accumulated on the biochemical properties of RTR in vitro and some of its functions in somatic cells , the role of the complex in meiosis has remained obscure . Using the model plant Arabidopsis , we demonstrate that the respective RTR homologues ( RECQ4A , TOP3α and RMI1 ) have similar functions in somatic plant cells but that TOP3α as well as RMI1 are also required for the resolution of meiotic recombination intermediates . To elucidate the biological role of TOP3α we characterized two different T-DNA insertion lines named top3α-1 ( SALK_139357 ) and top3α-2 ( GABI_476A12 ) of the respective gene At5g63920 [20] , [21] . The sequences of the insertion site of top3α-1 have been previously described [12] . The T-DNA of top3α-2 is inserted into the 11th intron ( of a total of 23 ) accompanied by a genomic deletion of 37 bp and two small filler sequences ( Figure 1A ) . The nucleotide sequence of the mRNA of AtTOP3α was determined by RT-PCR and RACE [22] and deposited into GenBank ( acc . no . EU295446 ) . The T-DNA insertions of both lines are shown in Figure 1A and are located in the mid-region of the gene . No expression of the gene spanning the respective T-DNA insertion sites could be detected ( Figure S1A; Table S1 ) . Attop3α-1 shows severe developmental defects and barely germinates; the mutant has deformed cotyledons and is not able to form roots at all ( Figure 2A , upper panel ) . As demonstrated previously , this early lethality can be converted to a less severe phenotype in an Atrecq4A-4 background [12] . The phenotype of the second line , Attop3α-2 , is less severe but nevertheless , visibly exhibits growth deformations , such as dwarfing , curling and fasciated organs as well as sterility ( Figure 2A , lower panel ) . In their respective heterozygous mutant states , neither TOP3α insertion lines are visibly affected , and they can be propagated for at least four generations in our hands . In analyzing more than 40 homozygous top3α-2 plants , a single intact seed was never observed . Interestingly , this phenotype is very similar to the one obtained for the double mutant recq4A-4/top3α-1 . The weaker phenotype of the top3α-2 mutant indicates that this mutation is hypomorphic and allows somatic growth . Nevertheless , mitosis is severely impaired in this mutant . While analyzing mitotic divisions by DAPI staining , we found mitotic aberrations in 14 . 2% of the dividing cells , which is more than ten times higher than in the wild type Col-0 control ( Table 1 ) . Interestingly , we were able to detect a significantly similar level of mitotic errors in the top3α-1/recq4A-4 or top3α-2/recq4A-4 double mutants ( 15 . 6% and 10 . 5% , respectively , Table 1 ) . The recq4A-4 mutant behaved like the wild type , indicating that the protein is not required for proper mitosis . Using the full-length genomic sequence , we were able to complement the strikingly different phenotypes of both TOP3α insertion lines . A genomic fragment containing the complete gene including the promoter and terminator region was amplified , cloned into a binary vector and transformed into both heterozygous mutant lines via Agrobacterium tumefaciens using the floral dip method [23] . We tested plants homozygous for the T-DNA insertion in the TOP3α gene containing the complementation construct . For top3α-1 , two out of two and for top3α-2 , five out of five tested independent transformants showed a virtually complete rescue of all phenotypic traits: complemented top3α-1 as well as top3α-2 plants grew properly without fasciated organs and were as fertile as the wild-type ( Figure 2B ) . This analysis demonstrates that the phenotypes are indeed caused by the two different insertions in the TOP3α gene . To further characterize the role of TOP3α in Arabidopsis , we analyzed the DNA repair capacity of the top3α-2 line . As TOP3α and BLM act together in the RTR complex in mammals , we included an insertion mutant of the RECQ4A gene in the analysis , which in many respects can be regarded as a functional BLM homologue in Arabidopsis [12] , [24] . We applied a liquid medium assay and determined the weight of 4-week-old plantlets challenged with different concentrations of DNA damaging agents . None of the analyzed insertion lines showed an altered sensitivity towards bleomycin ( data not shown ) . Compared to Col-0 wild types , recq4A-4 and top3α-2 were more impaired in growth when challenged with either the genotoxic agent methylmethane sulfonate ( MMS ) or the crosslinking agent cisplatin ( Figure 3A and B ) . Furthermore , top3α-2 was more sensitive to the topoisomerase I inhibitor camptothecin ( CPT ) while recq4A-4 was not ( Figure 3C ) . Because the RTR complex plays an important role in the suppression of crossover ( CO ) recombination in somatic cells , we tested whether Attop3α-2 could be characterized by an enhanced frequency of somatic homologous recombination as was previously demonstrated for several AtRECQ4A insertion mutants [12] , [25] . For this purpose , we used the recombination substrate line IC9C . This line harbours a transgene with non-functional overlapping parts of the ß-glucuronidase gene . Restoration of the marker is possible only by using the sister chromatid or the homologous chromosome as a partner [26] . Each recombination event is represented as a blue stained sector . We determined the HR frequencies of the different mutant lines with and without the bleomycin challenge as a means of inducing double stranded breaks ( DSBs ) . The HR frequency of both mutant lines ( top3α-2 and recq4A-4 ) in the IC9C background was enhanced to a similar extent ( five- to seven-fold , respectively; Figure 4 ) . After induction of DSBs , however , the enhancement could no longer be detected in either mutant lines compared to wild types . This finding reflects the behaviour of the recq4A mutant . One possible explanation would be that RECQ4A and TOP3α are involved in replication-associated crossover-suppression , but are not required for DSB-induced intermolecular HR . According to the SDSA model , which is appropriate for the description of homologues DSB repair in somatic plant cells , no crossovers should occur [27] . Alternatively , RECQ4A and TOP3α might prevent replication failures that would initiate recombination . The most striking phenotype of the top3α-2 and the recq4A-4/top3α-1 double mutant is their absolute sterility . To define the essential role of AtTOP3α , we analyzed the meiotic division of top3α-2 in detail using DAPI staining of pollen mother cells ( PMC ) . Representative meiotic stages of PMCs are shown in Figure 5 . In the early stages of meiosis , the top3α-2 mutant appears normal , and the first defects become visible during the late prophase when more than five bivalent structures appear , indicating the presence of breaks in the chromosomes ( Figure 5 B2 ) . In metaphase I , further chromosome condensation is disturbed and fragmentation becomes more visible ( Figure 5 B3 ) . The most impressive effect is observed in anaphase I: some chromosomal DNA is pulled to the poles but appears to stick together with other fragments that are , by themselves , not heading for the poles while some fragments stay in the former location of the metaphase plate ( Figure 5 B4 ) . After decondensation of the chromosomal fragments in telophase I , we were unable to observe the usual progress of meiosis towards the second division . This suggests that top3α-2 is arrested at this stage ( Figure 5 B5 ) . We also analyzed recq4A-4 , which did not show such defects and behaved like Col-0 during meiotic divisions in PMCs ( Figure S2 ) . To further define the nature of DNA structures that are processed by TOP3α during meiosis , it was necessary to define whether the enzyme acts on replication or recombination intermediates , and in the latter case , what types of intermediates might be targeted . To test this , crosses of top3α-2 with two lines defective in either initiation ( Atspo11-2-3 ) or further progression ( Atrad51 ) of meiotic recombination were analyzed [28] , [29] . The Atspo11-2-3 mutant produces no visible meiotic DSBs and exhibits unpaired univalents which are randomly distributed during the meiotic division . If TOP3α is involved in the processing of recombination intermediates , we would expect that the top3α-2 phenotype would be suppressed by the spo11-2-3 mutation . This is indeed the case , as the combination top3α-2/spo11-2-3 shows a clear spo11-2-3 phenotype and no traces of chromosome fragmentation ( Figure 5 C1 to C7 ) . Furthermore , the arrest of top3α-2 observed after the first meiotic division can be overcome in the spo11-2 background ( Figure 5 C6 and C7 ) . The fact that TOP3α is required for a stage of meiotic recombination after strand invasion is revealed by the fact that the double mutant top3α-2/rad51 has the same phenotype as the rad51 mutant , which shows abnormal pachytene chromosomes ( Figure 5 D1 to D7 ) [29] . Most notably in the double mutant , the second meiotic division takes place and results in polyads instead of the typical tetrads observed in the wild type ( Figure 5 D6 and D7 ) . Thus , TOP3α acts on DNA structures that are formed after DSB induction and heteroduplex invasion . The enzyme , therefore , appears to be involved in the resolution of a certain class of heteroduplex intermediates . The RMI1/BLAP75 protein has been identified recently as the third partner of the RTR complex in mammals and yeast [17] . We therefore analyzed the Arabidopsis genome for a putative homologue of RMI1 . We identified two candidate genes , At5g63540 ( that we refer to below as AtRMI1 ) and At5g19950 . As insertion mutants of At5g19950 did not reveal sensitivity to DNA damaging agents or reduced fertility ( data not shown ) , we assumed that the putative ORF , in contrast to At5g63540 , was not a functional homologue of RMI1 . For At5g63540 , we performed a cDNA analysis obtaining the same sequence that was deposited in the GenBank database and predicted as a hypothetical protein of 644 amino acids ( acc . no . AY735746 ) . Interestingly , the genes of AtTOP3α and AtRMI1 ( At5g63540 ) are located in close vicinity on chromosome V , spaced by only 130 kilobases . We characterized two independent mutant lines of AtRMI1 . The first line , rmi1-1 ( SALK_93589 ) , shows a deletion of approximately 2 kb starting in the first exon . It therefore virtually resembles a true knockout because more than 60% of the entire gene is lost ( Figure 1B ) . Only a truncated version of the RMI1 mRNA is produced in the rmi1-1 line ( Figure S1B ) . The second line , rmi1-2 ( SALK_094387 ) , harbours a duplicated T-DNA that has two left borders pointing outwards in the 5th exon and is accompanied by a small deletion of 12 nt ( Figure 1B ) . We did not detect any expression in the rmi1-2 line spanning the insertion site ( Figure S1B ) . Both lines show a similar phenotype , but not to the same extent . Both rmi1-1 and top3α-2 are completely sterile and severely impaired in male and female meiotic divisions , while rmi1-2 shows a reduced fertility , producing about half the number of seeds as wild type plants . By analyzing meiotic anaphases , we found a defect in 51 out of 230 figures analyzed ( 20 . 8% ) . In the wild type , however , we detected only about 6% abnormalities ( 11/193 ) . Thus , it seems that only rmi1-1 can be regarded as a null mutant , at least with respect to meiosis . To characterize the role of AtRMI1 in somatic cells , we performed a sensitivity assay towards genotoxic agents and changes in HR to detect defects in mitosis . Both rmi1-1 and 2 failed to exhibit mitotic defects ( Table 1 ) . Our tests with the rmi1-2 mutant revealed that the plants are more sensitive to DNA damaging agents MMS and cisplatin , to a similar extent as the recq4A-4 mutant ( Figure 3A and B ) . No sensitivity to camptothecin was detectable ( Figure 3C ) . Moreover , tests performed with the rmi1-1 mutant in the IC9C background revealed the same kind of HR behaviour observed in the top3α and the recq4A mutants . The loss of RMI1 results in hyper-recombination under standard growth conditions , however , no significant difference was observed after the induction of DSBs with bleomycin in wild-types ( Figure 4 ) . Thus , similar to TOP3α and RECQ4A , RMI1 is involved in the suppression of spontaneous recombination but not in interchromosomal homologous DSB repair . It therefore seems that in plants , a fully functional RTR complex is present and required for repair of certain kinds of DNA damage and crossover-suppression in somatic cells . Cytological analyses of AtRMI1 revealed similar kinds of meiotic defects in both lines . As for top3α-2 , changes in meiotic progression were first detectable in the late prophase , appearing as fragmented DNA and more than five bivalent structures ( Figure 5 E2 ) . The most impressive meiotic defect is visible at anaphase I . In top3α-2 mutants , most of the chromosome mass reached the pole in fragmented form , however , the greatest amount of DNA in rmi1-1 remained at the position of the metaphase plate ( Figure 5 E4 ) . In both mutants as well as in the double top3α and recq4A mutants , an increased amount of entangled chromosomes that contained unresolved connections and were torn apart during anaphase I were observed . Furthermore , severe chromosome fragmentation occurred and long bridges between the chromosomes were visible ( Figure 5 B3 and D4 ) . Representative examples of anaphase I structures observed in the top3α-2 , both top3α mutants in recq4A-4 backgrounds as well as in rmi1-1 and 2 are illustrated in Figure 6 . Taking the number of visible fragments into account , it was clear that most chromosomes were broken more than once . The decondensation of chromosomes took place during telophase I , and like the top3α-2 mutant , the rmi1-1 mutant did not progress any further . The same effect was observed in ∼20% of the meiotic stages that were impaired in the hypomorphic rmi1-2 line . Our results clearly demonstrate that both TOP3α and RMI1 are essential for the proper resolution of recombination intermediates during the first meiotic division . Very little is known about the biological function of TOP3α in higher eukaryotes . This is because that deletion or disruption of TOP3α in mammals , worms , and insects leads to severe phenotypes comparable to the Arabidopsis T-DNA insertion line top3α-1 [8]–[10] , [12] . We were able to restore the viability , but not the fertility , of this mutant using a recq4A background , suggesting that TOP3α might also be essential for meiosis . The characterization of a second TOP3α mutant with a less severe somatic phenotype has enabled us to define specific functions of the protein during mitosis and meiosis . Our finding that the phenotype could be complemented by the wild-type gene to the same extent as for the top3α-1 mutant suggests that top3α-2 can be regarded as hypomorphic . Our experiments clearly demonstrate that AtTOP3α is involved in at least three different pathways: i ) replication-dependent DNA repair and suppression of CO-recombination in somatic cells , ii ) proper resolution of replication intermediates during mitosis , and iii ) resolution of homologous chromosomes during meiosis . Moreover , the results from single and several double mutants enable us to define the processes in which TOP3α is acting in concert with its partners of the predicted plant RTR complex , namely RECQ4A and RMI1 ( Figure 7 ) . Our results also show that a functional RTR complex is present in plants as previously demonstrated for other eukaryotes . First , homologues of all three genes are not only present in the genome of Arabidopsis but also distantly related plants , such as poplar , rice and moss ( according to our database searches ) . We demonstrated that the three single mutants of the respective genes in Arabidopsis exhibited a similar sensitivity to genotoxic agents that methylate or crosslink DNA . This clearly shows that the RTR-like complex is involved in somatic DNA-damage repair in A . thaliana . Moreover , this complex directs the repair of aberrant DNA-structures during replication into a pathway by which HR is suppressed [12] . The formation of DNA structures that could enhance HR is therefore either prevented , interrupted or even dissolved during later steps as it has been shown for the RTR complex in yeast and animals [13] , [15] , [18] , [30] . This CO-suppression function conserved in plants is demonstrated by the fact that insertion lines of all three RTR members exhibit an enhanced level of HR without induction of DSBs , indicating their involvement in a replication-associated repair pathway . Second , mutation of RECQ4A rescues the lethal phenotype of top3α-1 . This phenomenon is similar to that in S . cerevisiae . Here , mutation of the Slow Growth Suppressor 1 ( Sgs1 ) rescues the very slow growth of a top3 mutant [3] . The double mutant recq4A-4/top3α-1 or 2 , however , exhibits the same defects in mitotic division as the top3α-2 single mutant , indicating that a functional TOP3α is always required for mitosis . Our analyses revealed no mitotic defects for the recq4A-4 or rmi1-1 and rmi1-2 mutants , indicating that the RTR complex is not necessary for this function of TOP3α . Thus , TOP3α possesses a unique role during mitosis , which was sustained by the observation that top3α-2 was the only line that showed enhanced sensitivity in the camptothecin treatment assay . This chemical specifically blocks replication forks by trapping TOP1 covalently bound to DNA [31] , [32] , which may indicate that AtTOP3α can at least partially substitute for a class IB topoisomerase . Third , we showed that both TOP3α and RMI1 have an essential role during meiosis . Here , RECQ4A is either not involved or might be substituted by another helicase . This is supported by the findings that recq4A-4 does not show any visible meiotic defects [12] , [25] . Notably , besides the recq4A mutant , recq4B mutants and the double mutant recq4A/recq4B are also fully fertile , despite the fact that both genes are most closely related to the mammalian BLM homologue [12] . The data presented in our manuscript , however , do not exclude that RECQ4A is involved in the resolution of recombination intermediates during meiosis as it has been recently shown for the yeast homolog Sgs1 [33] , [34] . It seems that during meiosis , both TOP3α and RMI1 are essential for the proper resolution of homologous chromosomes , as mutations in both genes show very similar meiotic defects . The most intriguing feature is that TOP3α as well as RMI1 mutants do not progress through meiosis past telophase I . This is in contrast to all meiotic mutants described in plants to date , which all complete meiosis even if chromosome fragmentation occurs [35]–[40] . The telophase I arrest found here is most likely due to the fact that the homologous chromosomes ( HC ) are not separated properly , and a signal to proceed with meiosis , which may require resolution of the chiasmata between HC , is not given . An alternative explanation might be that the meiotic abrogation occurs due to the extensive damage that may mechanically hinder the progression of meiosis . Conversely , severe damage also accumulates in other meiotic mutants , which are still able to enter meiosis II . Our data unambiguously demonstrate that both TOP3α and RMI1 are required in the late steps of meiotic recombination . It is difficult , however , to define their particular role in detail . The lack of either TOP3α or RMI1 clearly leads to plain sterility due to meiotic arrest during telophase I , as the proteins appear to be involved in the resolution of recombination intermediates . A major question of debate has been what the nature of Holliday Junction resolvases responsible to resolve the CO-structures during meiosis . It has been shown that at least two different complexes are present in most eukaryotes . One protein complex consists of the MUS81/EME1 endonuclease , and the other is associated with RAD51C and XRCC3 [41]–[43] . As it is generally assumed that such an enzyme complex should be able to resolve dHJs into COs , we are reluctant to assume that TOP3α/RMI1 are participating in CO-formation because both proteins are involved in the suppression of COs in somatic cells . Furthermore , topoisomerases of class I are not classical HJ resolvases . Together with the help of a helicase , dHJs can be transformed into hemicatenanes , which then can be resolved into two unlinked double strands by topoisomerase-mediated strand passage [44] . The classical DSB repair ( DSBR ) model predicts a dHJ as the key intermediate , whose resolution leads to either CO or nCO [45] . This theoretical prediction has been sustained later on the molecular level in yeast by 2D gel analysis , demonstrating that the dHJ is a major heteroduplex intermediate in yeast meiosis [44] , [46] . A revised model of meiotic recombination was suggested in 2001 . It was postulated that a major number , if not all nCO recombinants , do not arise from dHJs but rather from synthesis-dependent strand annealing ( SDSA ) using an intermediate D-loop [47]–[49] . The revised DSBR model involves an intermediate called single end invasion ( SEI ) , which occurs at the transition state between D-loop and dHJ during yeast meiosis [48] . It was proposed that the CO-decision is made very early during leptotene/zygotene , probably by a differential loading of RAD51 or DMC1 [48] , [50] . Taking the revised model into account , it is tempting to speculate that TOP3α/RMI1 may function at the resolution step of dHJs ( Figure 7 ) . Our hypothesis is that TOP3α/RMI1 acts as a type of safeguard system to resolve dHJs to nCO that were not resolved by a bona fide resolvase , even though they had been channelled into the CO pathway . This mode of dHJ resolution by a topoisomerase type I has was proposed long ago [51] . Schwacha and Kleckner also discussed this potential resolution of mature dHJs via a topoisomerase function as either an alternative to the modified DSBR model or as a backup mechanism for the resolution of an HJ that aberrantly persists beyond pachytene [44] . We favour the hypothesis that the resolution step in the CO pathway of meiotic recombination is either surprisingly error prone or rate limiting . Whenever HJ resolution fails , only the alternative dissolution of an HJ into nCO guarantees the progression of meiosis [47] , [48] . Failure of the TOP3α/RMI1 function could lead to the persistence of HJs that would have been processed into hemicatenane structures . These structures , however , could only be cleaved by a topoisomerase and no longer by a resolvase . Thus , due to mechanical shearing of the irresolvable interlinked homologues , we observed the fragmentation of chromosomes and telophase I arrest . An alternative explanation would be that TOP3α is specifically required in meiotic chromosome condensation during or after recombination to remove torsional stress . If the enzyme is not functional , chromosome breakage may occur . We , however , disfavour this hypothesis as at least yeast TOP3 , in contrast to TOP1 and 2 , is insufficient to remove supercoils in vitro [52] . Moreover , it is not clear why an RMI1 homologue would be involved in such a function as , in contrast to TOP3α , it is not required for mitosis in Arabidopsis . Our results demonstrate that both genes , TOP3α and RMI1 , are essential for proper meiotic progression , and if disturbed , result in fragmented chromosomes and meiotic arrest at telophase I . In an independent study , the group of Mathilde Grelon ( Chelysheva et al . , submitted to PLoS ) identified RMI1-deficient plants in a screen for sterile Arabidopsis mutants . Their detailed analyses of the meiotic role of RMI1 are in complete accordance with our results presented here . Moreover , they showed that RMI1 is not required for synaptonemal complex formation . Analyzing crosses between rmi1 and dmc1 or rad51 , they were further able to demonstrate that the protein is required for meiotic DSBR using either the homologue or the sister chromatid as a template . Thus , two independent studies strongly indicate that higher eukaryotes possess a safeguard mechanism to guarantee the untangling of homologous chromosomes in the late meiotic prophase . This process seems to be conserved in yeast as it has been demonstrated that the mutation of the unique TOP3 of S . cerevisiae leads to both mitotic defects and sterility [3] , [53] . Although no detailed cytological analysis was performed , it could be demonstrated that the phenotype was the result of a defect in meiotic recombination . It is tempting to speculate that the role of TOP3α/RMI1-1 in meiosis is functionally related to their roles in suppression of sister chromatid exchanges ( SCE ) during replication in mitosis . Nevertheless , both processes seem to be different because in plants , RECQ4A is also involved in SCE suppression along with TOP3α and RMI1 . The fact that recq4a mutants are fertile does not necessarily imply that RECQ4A has no role in meiosis . It has been recently shown that SGS1 , the unique RecQ homologue of yeast , is indeed directly involved in meiosis by suppressing formation of joint molecules comprising three and four interconnected duplexes [54] . Important aims of future research will therefore be to elucidate the role of AtRECQ4A in meiois as well as to identify the putative helicase that , according to our hypothesis , might be involved in the dissolution of dHJs in concert with AtTOP3α and AtRMI1 in meiosis . The genotyping of the different mutants was perfomed as described [12] using primers flanking the T-DNA insertions ( Figure 1A and C and Table S1 ) . Interestingly , the 2 kb deletion found in rmi1-1 seems to have occurred during propagation of the seeds at the SALK institute . Indeed our seed sample of SALK_093589 contained a mixture of two different genotypes . Besides the allele with the deletion , we found also a T-DNA inserted as described by SALK . Since we were afraid that the T-DNA insertion might be unstable during propagation we decided to perform our experiments with the deletion mutant . This line was stable in its phenotype for at least 4 generations in our hands . The T-DNA integration sites and the deletion in rmi1-1 were determined via PCR using primer combinations specific for the left border of the respective T-DNA and genomic sequences within the respective gene ( Figure 1A; primer 2 , 2R , and LB; Figure 1C; primer 4 , 5R and LB1 ) . The Rapid Amplification of cDNA Ends ( RACE ) was carried out according to Matz et al . ( 1999 ) [22] . Mutagen and HR assays using the IC9C reporter line and the respective T-DNA insertion lines were perfomed as described [12] . Specific adaptions concerning these two general assays for sterile plants were as follows below Seeds were sterilized using 4% NaOCl solution . After two weeks of growth on solid germination medium ( GM ) , plantlets were transferred into six-well-plates containing 5 ml of liquid GM medium for each untreated control , or 5 ml GM including the mutagen samples , respectively . Five plants were used for each well . For the top3A-2 mutant only homozygous plants were picked with regard to their visible growth phenotype . The top3A-2 phenotype was confirmed via PCR screening . After 14 more days in the growth chamber , the fresh weight of each sample was determined using a fine scale balance . To avoid any aberrations resulting from excessive liquid , the plant material was dabbed off with paper towels before weighting . The mutagen assays were carried out at least four times for each mutagen . The assay was carried out under sterile conditions , transferring in each case 20 2-week old plantlets from solid into liquid GM medium ( 10 ml ) . Bleomycin was added dissolved in GM the next day in a final concentration of 5 µg/ml . After 6 days in the growth chamber the seedlings were transferred into staining solution . For the rmi1-1 and top3A-2 lines , homozygous plants were identified by means of PCR screening and only the data obtained from these seedlings were included in the analysis . Meiotic and mitotic figures were analysed as described [28] . To rescue the different phenotypes observed in the top3α-1 and 2 mutant lines , we generated a complementation construct of 8711 bp representing the entire gene of TOP3α . It included 1174 bp of the promoter region , 114 bp 5′-UTR , 6699 bp between start and stop codon , 203 bp of the 3′-UTR and 521 bp of the terminator region . The genomic region was amplified in two parts using linker-primers and a proofreading polymerase ( Phusion-Taq , New England Biolabs ) . The fragments cloned into the vector pPZP221 [55] were sequenced , subsequently combined via usage of an internal NdeI restriction site and transformed into the heterozygous top3α-1 and 2 plant lines . Successful transformed plants showing gentamycin resistance have been screened afterwards for their respective genotype concerning the TOPα gene . For this screening we used primers that were specific for the genomic locus only ( Table S2 ) .	The topoisomerases of the class IA are present in all three eukaryotic kingdoms—plants , fungi , and animals—and are involved in DNA replication and DNA repair . During the course of their action , they introduce transient single-strand nicks into DNA . In higher eukaryotes , two different classes of the enzymes are present: TOP3α and TOP3β . TOP3α is essential , as disruption of its function usually results in lethality of the affected organism . Using a mutant of TOP3α that retains some activity , we show that the protein has multiple , different functions in the model plant A . thaliana . Besides its action in somatic cells , where it is required for mitosis as well as DNA repair , we demonstrate that TOP3α together with its protein partner RMI1 is essential for meiosis . Here , both proteins are involved in DNA recombination—the exchange of information between parental chromosomes . Disruption of either TOP3α or RMI1 leads to grave defects and an early termination of meiosis , resulting in the sterility of the mutant plants . Our detailed analysis indicates that both proteins are involved in a late step of meiotic recombination , in a mechanism that prevents entanglement of the parental chromosomes . Thus , meiotic recombination seems to be more complex than previously anticipated .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "plant", "biology/plant", "genetics", "and", "gene", "expression", "molecular", "biology/dna", "replication", "molecular", "biology/chromatin", "structure", "biotechnology/plant", "biotechnology", "computational", "biology/comparative", "sequence", "analysis", "developmental", "biology/plant", "growth", "and", "development", "molecular", "biology/chromosome", "structure", "molecular", "biology/recombination", "plant", "biology/plant", "cell", "biology", "genetics", "and", "genomics/gene", "function", "plant", "biology/plant", "growth", "and", "development", "genetics", "and", "genomics/cancer", "genetics", "genetics", "and", "genomics/plant", "genetics", "and", "gene", "expression", "molecular", "biology/dna", "repair" ]	2008	Topoisomerase 3α and RMI1 Suppress Somatic Crossovers and Are Essential for Resolution of Meiotic Recombination Intermediates in Arabidopsis thaliana
Tumor-induced angiogenesis leads to the development of leaky tumor vessels devoid of structural and morphological integrity . Due to angiogenesis , elevated interstitial fluid pressure ( IFP ) and low blood perfusion emerge as common properties of the tumor microenvironment that act as barriers for drug delivery . In order to overcome these barriers , normalization of vasculature is considered to be a viable option . However , insight is needed into the phenomenon of normalization and in which conditions it can realize its promise . In order to explore the effect of microenvironmental conditions and drug scheduling on normalization benefit , we build a mathematical model that incorporates tumor growth , angiogenesis and IFP . We administer various theoretical combinations of antiangiogenic agents and cytotoxic nanoparticles through heterogeneous vasculature that displays a similar morphology to tumor vasculature . We observe differences in drug extravasation that depend on the scheduling of combined therapy; for concurrent therapy , total drug extravasation is increased but in adjuvant therapy , drugs can penetrate into deeper regions of tumor . The abnormal structure of tumor vasculature is one of the leading causes of insufficient and spatially heterogeneous drug delivery in solid tumors . Tortuous and highly permeable tumor vessels along with the lack of a functional lymphatic system cause interstitial fluid pressure ( IFP ) to increase within tumors . This elevated IFP results in the inefficient penetration of large drug particles into the tumor , whose primary transport mechanism is convection [1 , 2] . The abnormalities in tumor vasculature are caused by dysregulation of angiogenesis . Tumors initiate angiogenesis to form a vascular network that can provide oxygen and nutrients to sustain its rapid growth . The production of VEGF , a growth factor that promotes angiogenesis , is triggered by the chronic hypoxic conditions that are prevalent in tumors . Besides inducing angiogenesis , it leads to hyperpermeable blood vessels by enlarging pores and loosening the junctions between the endothelial cells that line the capillary wall [3 , 4] . Subsequently , excessive fluid extravasation from these vessels results in a uniformly elevated IFP in the central region of tumor nearly reaching the levels of microvascular pressure ( MVP ) while at the tumor periphery , IFP falls to normal tissue levels [1 , 5 , 6] . This common profile of IFP within tumors has been identified as a significant transport barrier to therapeutic agents and large molecules [1 , 7] . When IFP approaches MVP , pressure gradients along vessels are diminished and blood flow stasis occurs , diminishing the functionality of existing vessels [8–10] . Furthermore , uniformity of IFP in interior regions of tumors terminates the convection within tumor interstitium , hindering the transportation of large drugs [1] . While the lack of a transvascular pressure gradient inhibits convective extravasation of drugs , sharp IFP gradient at tumor periphery creates an outward fluid flow from tumors that sweeps drugs away into normal tissues [1] . Together these factors lead to the decreased drug exposure of tumor cells . It has been revealed that the application of antiangiogenic agents can decrease vessel wall permeability and vessel density , transiently restoring some of the normal function and structure of abnormal tumor vessels [4 , 11 , 12] . This process , which is called vascular normalization , is associated with a decrease in IFP and an increase in perfusion . Therefore , this state of vasculature enables increased delivery of both drug and oxygen/nutrients to the targeted tumor cells [11 , 13] . Normalization enhances convection of drug particles from vessels into tumor interstitium by restoring transvascular pressure gradients through IFP reduction [11 , 14 , 15] . It has shown some favorable results in preclinical and clinical trials regarding the enhancement of the delivery of large therapeutics such as nanoparticles [14 , 16 , 17] . Since nanoparticles benefit from the enhanced permeability and retention effect ( EPR ) , they are distributed in higher amounts to tumors relative to normal tissue . Accumulation of nanoparticles in normal tissues is relatively small compared to the standard small molecule chemotherapies , leading to decreased toxicity and side effects . However , the main transport mechanism for large drugs is convection in tumor microenvironment . Hence , when IFP is high , extravasation via convection is inhibited . Normalization due to its ability to decrease IFP seems promising in drug delivery for large drugs with its potential of restoring convective transportation . In both clinical and preclinical studies , it has been shown that antiangiogenic drugs demonstrate anti-tumor effects in various cancer types [18] . However , rather than using antiangiogenic agents alone , studies reveal that the combination of these agents with chemotherapy drugs yields favorable results with increased therapeutic activity . In some clinical studies [19–21] , bevacizumab combined with conventional chemotherapy has increased the survival and response rates among patients with gastrointestinal cancer compared to bevacizumab alone . This finding that antiangiogenic therapy in combination with chemotherapy can improve the efficacy of treatment has been observed for patients with various cancers including non-small cell lung cancer [22 , 23] , breast cancer [24–26] and ovarian cancer [27] . However , it is evident that there is a transient time window for vessel normalization [28 , 29] . In order to improve drug delivery , chemotherapy should coincide with this transient state of improved vessel integrity . Prolonged or excessive application of antiangiogenic agents can reduce microvascular density to the point that drug delivery is compromised [30] . Therefore , dosing and scheduling of combined therapy with antiangiogenic agents must be carefully tailored to augment the delivery and response to chemotherapy [12] . It is suggested that rather than uninterrupted application , intermittent cycles which can create re-normalization should be employed for antiangiogenic agent scheduling [31] . Due to the complex and interdisciplinary nature of the subject , there is a considerable amount of computational efforts on tumor vascularization and its consequences for the tumor microenvironment and drug delivery . Development of vasculature and intravascular flow dynamics are studied comprehensively [32–37] and in many studies chemotherapy is given through the discrete vessel system in order to calculate drug delivery to capillaries and tumor [33 , 34 , 37–39] . Mathematical models have included transvascular and interstitial delivery of drugs [37–39] . In addition to that , Wu et al . added tumor response to chemotherapy by applying nanoparticles and evaluating the decrease in tumor radius during chemotherapy for different microenvironmental conditions [39] . There are also some studies about the optimization of combination therapy in tumors [40] . In studies by the groups of Urszula Ledzewicz and Heinz Schäettler , changes in tumor volume after the administration of cytotoxic and antiangiogenic agents have been investigated by proposing a mathematical model and seeking optimal solutions for different treatment cases [41 , 42] . Compartment models have also been used to explore how antiangiogenic agents may provide assistance to chemotherapy agents in reducing the volume of drug-resistant tumors and by using a bifurcation diagram it is shown that the co-administration of antiangiogenic and chemotherapy drugs can reduce tumor size more effectively compared to chemotherapy alone [43] . Applications of chemotherapy drugs together with antiangiogenic agents have been studied by Panovska et al . to cut the supply of nutrients [44] . Stephanou et al . showed that random pruning of vessels by anti-angogenic agents improves drug delivery by using 2-D and 3-D vessel networks [45] . However , they did not associate this benefit with normalization of vasculature . Jain and colleagues laid out the general groundwork for relations between vessel normalization and IFP by relating vessel properties and interstitial hydraulic conductivity to changes in pressure profile due to normalization [15] . The subject is further investigated by Wu et al . by building a 3-D model of angiogenesis and adding intravascular flow to the computational framework [32] . They observed slow blood flow within the tumors due to almost constant MVP and elevated IFP profile . They show the coupling between intravascular and transvascular flux . Kohandel et al . showed that normalization enhances tumor response to chemotherapy and identified the most beneficial scheduling for combined therapy in terms of tumor response [46] . The size range of nanoparticles that could benefit from normalization has also been investigated [16] . In this study , following the continuous mathematical model developed by Kohandel et al . [46] which couples tumor growth and vasculature , we built a framework for tumor dynamics and its microenvironment including IFP . We use this system to evaluate the improvement in nanoparticle delivery resulting from vessel normalization . As the tumor grows , a homogeneous distribution of vessels is altered by the addition of new leaky vessels to the system , representing angiogenesis . As a consequence of angiogenesis and the absence of lymph vessels , IFP starts to build up inside the tumor inhibiting the fluid exchange between vessels and tumor and inhibiting nanoparticle delivery . Simulations give the distribution of the nanoparticles in the tumor in a time-dependent manner as they exit the vessels and are transported through interstitium . The activity of the drugs on tumor cells is determined according to the results of experimental trials by Sengupta et al . [47] . We apply drugs in small doses given in subsequent bolus injections . During drug therapy , both vessels and tumor respond dynamically . After injections of antiangiogenic agents , a decrease in vessel density accompanies the changes in vessel transport parameters , initiating the normalized state . Combining chemotherapy with applications of antiangiogenic agents , we are able to identify the benefits of a normalized state by observing the effects of different scheduling on IFP decrease , extravasation of drugs and tumor shrinkage . We found that in adjuvant combination of drugs , IFP and vessel density decrease together resulting in an increase in the average extravasation of nanoparticles per unit area in the interior region of tumor . In concurrent combination of drugs , IFP decrease is higher but vessel decrease is higher as well , creating a smaller enhancement in average extravasation per unit tumor area . However , even though average extravasation is smaller in this case , we observe an increase in homogeneity in drug distribution . Nanoparticles begin to extravasate even in the center of tumor through sparsely distributed vessels due to the sharp decrease in IFP . Therefore normalization enabled the drugs to reach deeper regions of the tumor . Following Kohandel et al . [46] , the Eqs ( 1 ) and ( 2 ) are used to model the spatio-temporal distribution of tumor cells and the heterogeneous tumor vasculature . In Eq ( 1 ) , the first term models the diffusion of tumor cells , where Dn is the diffusion coefficient , and the second term describes the tumor growth rate , where nlim is the carrying capacity and r is the growth rate . In the absence of the third and fourth terms , the Eq ( 1 ) has two fixed points: an unstable fixed point at n = 0 where there is no cell population and a stable fixed point at n = nlim where the population reaches its maximal density . The coupling terms αmn n ( x , t ) m ( x , t ) and dr n ( x , t ) d ( x , t ) indicate the interactions of tumor cells with vasculature and chemotherapy drug , respectively . Tumor cells proliferate at an increased rate αmn when they have vessels supplying them with nutrients and tumor cells are eliminated at rate dr if chemotherapy drug d ( x , t ) is present . ∂ n ( x , t ) ∂ t = D n ∇ 2 n ( x , t ) + r n ( 1 - n n l i m ) + α m n n ( x , t ) m ( x , t ) - d r n ( x , t ) d ( x , t ) . ( 1 ) The tumor vasculature network exhibits abnormal dynamics with tortuous and highly permeable vessels which are structurally and functionally different from normal vasculature . In order to create this heterogeneous structure , a coarse-grained model is used to produce islands of vessels . In Eq 2 , the average blood vessel distribution is represented with m ( x , t ) and the equation is formulated to produce islands of vascularized space with the term m ( x , t ) ( α + βm ( x , t ) + γm ( x , t ) 2 ) which has two stable points m = 1 and m = 0 corresponding to the presence and absence of vessels , respectively . Representation of tumor-induced angiogenesis is modified in this model by recruiting the terms αnm n ( 1 − n/nlim ) m and βnm∇ . ( m∇n ) . Here , the former attains positive values for tumor periphery due to the low cell density and in the central regions when cell density exceeds nlim , the term becomes negative creating a behavior which resembles to real tumors that has generally high vascularization in periphery and low vessel density in the center due to the growth-induced stresses [48] . The latter term leads the vessels that are produced in the periphery towards the tumor core . In this novel form , parameters relate to angiogenesis , βnm and αnm are changed as 0 . 5 and 0 . 25 , respectively . Remaining set of the parameters related to tumor and vessel growth can be found in Kohandel et al . [46] . Ar m ( x , t ) A ( x , t ) is the reaction of tumor vessels to antiangiogenic agent A ( x , t ) , which results in the elimination of vessels in the presence of antiangiogenic agent . ∂ m ( x , t ) ∂ t = D m ∇ 2 m ( x , t ) + m ( x , t ) ( α + β m ( x , t ) + γ m ( x , t ) 2 ) + β n m ∇ · ( m ∇ n ) + α n m n ( 1 - n n l i m ) m - A r m ( x , t ) A ( x , t ) . ( 2 ) For the initial configuration of tumor cells , a Gaussian distribution is assumed while the initial vascular distribution is obtained by starting from a random , positively distributed initial condition of tumor vessels . Darcy’s law is used to describe the interstitial fluid flow within the tissue: u = −K∇P , where K is the hydraulic conductivity of the interstitium ( mm2/s/mmHg ) and P is the interstitial fluid pressure ( IFP ) . For the steady state fluid flow , the continuity equation is: ∇ · u = Γ b - Γ ℓ , ( 3 ) where Γb ( 1/s ) represents the supply of the fluid from blood vessels into the interstitial space and Γℓ ( 1/s ) represents the fluid drainage from the interstitial space into the lymph vessels . Starling’s law is used to determine the source and the sink terms: Γ b = λ b m ( x , t ) [ P v - P ( x , t ) - σ v ( π c - π i ) ] , ( 4 ) Γ ℓ = λ ℓ P ( x , t ) . ( 5 ) The parameters in these equations are the hydraulic conductivities of blood vessels λb and the lymphatics λℓ , the vascular pressure Pv , interstitial fluid pressure P and the osmotic reflection coefficient σv . The capillary and the interstitial oncotic pressures are denoted by πc and πi , respectively . Hydraulic conductivities of blood and lymph vessels are related to the hydraulic conductivity of vessel wall ( Lp ) and the vessel surface density ( S V ) with the relation λ b , ℓ = L p S V . The osmotic pressure contribution for the lymph vessels is neglected due to the highly permeable lymphatics . Also , the pressure inside the lymphatics is taken to be 0 mm Hg [49] . By substituting Darcy’s law and Starling’s law into the continuity equation , we obtain the equation for IFP in a solid tumor: - K ∇ 2 P ( x , t ) = λ b m ( x , t ) [ P v - P ( x , t ) - σ v ( π c - π i ) ] - λ ℓ P ( x , t ) . ( 6 ) Pressure is initially taken to be the normal tissue value Pv and the initial pressure profile is set based on the solution of the above equation with the initial condition for tumor vasculature . The boundary condition ensures that pressure reduces to the normal value Pv in host tissue . For the transport of antiangiogenic agents A ( x , t ) , a diffusion equation is used: ∂ A ( x , t ) ∂ t = D A ∇ 2 A ( x , t ) + λ A m ( x , t ) ( A v - A ( x , t ) ) - Γ ℓ A ( x , t ) - k A A ( x , t ) , ( 7 ) where DA is the diffusion coefficient of antiangiogenic agents in tissue , λA is the transvascular diffusion coefficient of antiangiogenic agents , Av is the plasma antiangiogenic agent concentration and kA is the decay rate of antiangiogenic agents . The terms on the right hand side represent the diffusion of the antiangiogenic agents in the interstitium , diffusion through the vessels , the drainage of agents to the lymph vessels and the decay rate of the agents , respectively . We consider liposomal delivery vehicles for chemotherapy drug with their concentration denoted by d ( x , t ) . Since they are relatively large ( ∼ 100 nm ) , a convection-diffusion equation is used for the transport of these drug molecules: ∂ d ( x , t ) ∂ t = D d ∇ 2 d ( x , t ) + ∇ · ( k E d ( x , t ) K ∇ P ) + Γ b ( 1 - σ d ) d v - Γ ℓ d ( x , t ) - d r d ( x , t ) n ( x , t ) - k d d ( x , t ) , ( 8 ) where Dd is the diffusion coefficient of drugs in the tissue , kE is the retardation coefficient for interstitial convection , dv is the plasma drug concentration , σd is the solvent drag reflection coefficient , dr is the rate of drug elimination as a result of reaction with tumor cells and kd is the decay rate of the drugs . The terms on the right hand side represent the diffusion and the convection of the drugs in the interstitium , convection of the drugs through the vessels , the drainage of the drugs into the lymphatics , the consumption of drugs as a result of tumor cell interaction and the decay of the drug , respectively . Diffusion of the drug from the blood vessels is assumed to be negligible since transvascular transport of large drugs is convection-dominated . Since the time scale of the tumor growth is much larger than the time scale for the transport and distribution of the drug molecules , both antiangiogenic agent and chemotherapy drug equations are solved in steady state , i . e . ∂ d ( x , t ) ∂ t = ∂ A ( x , t ) ∂ t = 0 . Both drugs are administered to the plasma with bolus injection in each administration through an exponential decay function: A v ( t ) = A 0 e - t / t 1 / 2 A , ( 9 ) d v ( t ) = d 0 e - t / t 1 / 2 d , ( 10 ) In these equations , the terms A0 , d0 and t 1 / 2 A , t 1 / 2 d indicate the peak plasma concentration and the plasma half-lives of the antiangiogenic agent and chemotherapy drug , respectively . No-flux boundary conditions are used for the antiangiogenic agent and the chemotherapy drug . Parameters related to transport of interstitial fluid and transport of liposomes and antiangiogenic agents are listed in Tables 1 and 2 respectively . Some of the effective parameters in the equations above dynamically change to mimic the changes in tumor and its microenvironment . As the tumor grows , lymph vessels are diminished to ensure that there are no lymph vessels inside the tumor . Without the presence of tumor , vessel density can increase up to a specific value ( the dimensionless value of 1 ) . When vessel density is greater than 1 , it implies that they were produced by angiogenesis and leaky , thus their hydraulic conductivity is increased up to levels that is observed in tumors . During antiangiogenic treatment , vessel density is decreased and when it decreases below 1 , normalization occurs and the hydraulic conductivity returns to normal tissue levels . We started the simulations with a small tumor ( 0 . 2 mm radius ) and left it to grow for 30 days to an approximate radius of 13 . 5 mm . Vessels which were initially set as randomly distributed islands in the computational domain evolved into a heterogeneous state throughout the simulations due to the presence of tumor cells ( Fig 1 , vessel density ) . As the tumor grows , vessel islands become sparse in the interior region but their density increases by angiogenesis and they become leaky . By the end of the simulation , the leakiness of tumor vessels and the lack of lymphatic drainage inside the tumor causes elevated pressure in the interior region of tumor very similar to that suggested in literature [1 , 5] ( Fig 1 , IFP/Pe ) . We experimented with various drug regimens . To illustrate the improvement in drug delivery , we designed the cases given in Fig 2 . Dimensionless dose values are fixed in order to replicate the treatment response observed in [47] . Antiangiogenic treatment is adjusted such that at the end of administrations there is approximately a 50% decrease in MVD inside the tumor . A fixed chemotherapy drug dose is administered on days 23 , 25 and 27 while we change the day of antiangiogenic agent administration starting from the days 15 , 17 , 19 , 21 and 23 , continue to give them every other day in 4 or 5 pulses . We decrease the dose of antiangiogenic agents throughout the therapy because a better response in drug delivery is obtained with this way in our simulations . We present here four cases where only antiangiogenic agent administration starts on day 23 , only chemotherapy drug on day 23 , neoadjuvant therapy with antiangiogenic agents on day 19 and chemotherapy drug on day 23 and finally concurrent therapy with both of drugs starting on day 23 . The most beneficial results regarding the amounts of drugs extravasate in the interior parts of the tumor are yielded when the antiangiogenic treatment starts at day 19 ( case-3 in Fig 2 ) . As expected , antiangiogenic agents don’t have a profound effect on tumor cell density when they are applied alone ( Fig 3 , case-1 ) . In all cases , we observed greater drug extravasation near the tumor rim due to decreasing IFP in that region ( Fig 3 ) . It can be seen that fluid flow from vessels to the tumor is poor in the interior region for case-2 , but it starts to enhance in the same region in case-3 and case-4 . The main reason for this change is the introduction of a pressure gradient in the tumor center restoring drug convection . Therefore , in both case-3 and case-4 , tumor cell density is decreased in the interior region ( Figs 3 and 4b ) as a consequence of increased drug extravasation in the interior region of the tumor . We calculate the space average of cell density and IFP in each time step . Average cell density is calculated as ∫ ∫ A i n t n ( x , y , t ) d x d y ∫ ∫ A i n t d x d y ( 11 ) over area Aint whose boundary is set by the condition n ( x , y , t ) > 1 which represents the interior region of tumor ( corresponds to r < 6 mm for a tumor of radius 10mm ) . Average IFP is calculated as ∫ ∫ A P ( x , y , t ) d x d y ∫ ∫ A d x d y ( 12 ) over area A whose boundary is set by the condition n ( x , y , t ) > 0 . 1 which represents the value over whole tumor . When we evaluate average pressure over the entire area of the tumor , we observe a synergistic effect in reducing pressure arising from the combined application of antiangiogenic agent and chemotherapy which can be seen in Fig 4a , especially for case-4 . This synergistic effect also exhibits itself in tumor cell density in a less pronounced manner that can be observed from Fig 4b . This indicates improved combination treatment efficacy as an indirect result of decreasing IFP . According to our results , drug extravasation from vessels in the interior region of the tumor is nearly doubled for combination cases ( Fig 5a , case-3 and case-4 compared to case-2 ) . However , this improvement is not directly reflected on drug exposure due to reduced vessel density by antiangiogenic agents . Total drug exposure of unit area in tumor during treatment only improves approximately 20–25% . IFP during the applications of chemotherapy drug was the lowest for concurrent therapy ( case-4 ) . However , regarding tumor regression adjuvant therapy ( case-3 ) performed better , agreeing with the results of Kohandel et al . [46] . Even though decrease in vessel density and leakiness cuts off the supply of drugs , the decrease in IFP appearing for the same reasons seems to compensate in the interior region of tumor , resulting in better drug extravasation . When two drugs are given closer temporally , the resulting IFP decrease is maximized . This enables the convective extravasation of nanoparticles deep into tumors to places that are not exposed to drugs without combination therapy . In order to evaluate the effect of chemotherapy drugs that target tumor cell proliferation , we modified Eq 1 such that the chemotherapy drugs would directly act on tumor growth . The terms responsible for tumor growth ( 2nd and 3rd terms in the right-hand side of Eq 1 ) are multiplied by ( 1 − d ( x , t ) /dmax ) where dmax is maximum drug concentration that extravasated inside the tumor . In this scenario , small changes are seen in tumor cell densities between combination therapy and chemotherapy alone . However , we observe that in this form , extravasation of drugs is also increased in the central region as seen in Fig 6 , implying that normalization is also beneficial in this scenario . Using a mathematical model , we assess whether antiangiogenic therapy could increase liposome delivery due to normalization of tumor vessels . In order to do that , we first created a dynamic vessel structure that exhibits properties of tumor vessels created by angiogenesis as well as inherent vessels in the tissue . As the tumor grows , vessels in the central region begin to disappear due to increased tumor cell density in that region . Angiogenesis occurs in the tumor creating additional leaky vessels . The emergent vessel density is consistent with that observed in [59] , with decreasing density towards the tumor center along with randomly appearing clusters of vessels . IFP is found to be elevated throughout the tumor up to the levels of MVP and decreases sharply around the tumor rim as it is observed in various studies in the literature . [1 , 5 , 6] . We apply antiangiogenic agents in various regimens combined with chemotherapy and focus on large drugs ( liposomes ) whose delivery mainly depends on convection . As a result of the decrease in vessel density and leakiness due to the antiangiogenic activity , we expect a decrease in pressure which brings about a higher pressure difference between tumor and vessels . Transvascular convection depends on this pressure difference , hydraulic conductivity and density of vessels at the unit area . Since antiangiogenic agents decrease hydraulic conductivity ( i . e . leakiness ) and vessel density , by cutting the supply of drugs , the resulting increase in pressure difference should compensate for these effects , restoring extravasation in remaining vessels . In all simulations , liposome extravasation predominantly occurs in the tumor periphery due to low IFP levels , hence drugs preferentially accumulate in this area . Our result has been confirmed by experimental studies of drug distribution using large drugs such as micelles [60 , 61] , nanoprobes [62] and liposomes [59 , 63–66] in which peripheral accumulation is observed . As the application time between antiangiogenic agents and liposomes becomes shorter , the resulting decrease in IFP is maximized . This enables the convective extravasation of nanoparticles deep into tumors to places that could not previously be exposed to drugs before and liposome extravasation begins to appear in central region . However , that does not bring about maximum accumulation of liposomes consistently at all times . There is a trade-off between total drug accumulation and how deeply drug can penetrate inside the tumor . In our study , we find a balance between these two situations . It also shows us that IFP and drug accumulation are not always correlated , rather the maximum accumulation is achieved through the complex interplay between IFP , vessel density and leakiness . Current research by [63] also supports this view; in their mouse study , they point out that IFP is correlated with perfusion , perfusion is correlated with accumulation and the relationship between IFP and liposome accumulation is limited . In another significant study , tumor-bearing animals are subjected to combination therapy with liposomes and the antiangiogenic agent pazopanib in order to evaluate the effect of normalization via imaging drug distribution [65] . As a result of the decrease in MVP , they also observed a resulting decrease in IFP . Similar to our results , IFP is not the determinant of drug accumulation in their work . They have found that decreased leakiness of vessels inhibits delivery even though there is an IFP decrease as a result of antiangiogenic therapy . They have collected data for a single time point and observed a decrease in doxil penetration in combination therapy . They also point out that functional measures of normalization may not occur simultaneously which is also the case for our study . Throughout the combination therapy , we also observe periods where drug extravasation is limited and others where drug extravasation is improved . They have found the vessel permeability as a limiting factor in their study , however MVD [67] and tumor blood flow and blood volume [68] are also determinants of large drug accumulation . This shows that these measures of normalization are tumor type dependent and even within the same tumor they are dynamic which leads to variation in drug distribution . Among many different schedules , most of our trials did not show improvement in drug accumulation . We see that the dose of antiangiogenic agents should be carefully determined to ensure any delivery benefit . As stated by [30] , when we apply a large dose of antiangiogenic agents , significant IFP decrease is observed but the decrease in vessel permeability and the lack of vessel density lead to impaired liposome extravasation . At the other extreme , when we give small amounts of ant-angiogenic agents , it is seen that IFP decrease is not enough to make a significant improvement to liposome extravasation . In this model , intravascular flow is approximated as uniform to focus on the effects of transvascular delivery benefit of normalization . Due to abnormal vasculature , tumors are known to have impaired blood perfusion [69] due to simultaneous presence of functional and non-functional vessels . In this work , we simulate structural normalization of vessels without considering functional normalization which is associated with intravascular flow and results in increased perfusion [30] . Vessels within the tumor in this model have uniform functionality in terms of supplying blood flow . Hence , by decreasing vessel density in microenvironment due to antiangiogenic activity , we are decreasing blood perfusion . However , on the contrary , normalization is expected to enhance intravascular flow by decreasing pore size which restores intravascular pressure gradients and pruning non-functional vessels that interrupt circulation . Therefore , normalization brings about improved blood perfusion whereas here we decrease perfusion and improve the delivery only through improved convective extravasation by decreased IFP . In our simulations , the delivery benefit is underestimated since we decrease blood perfusion as a part of antiangiogenic activity . In [65] , they observed that MVD decrease did not change liposome accumulation because the eliminated vessels are the ones that are thought to be nonfunctional . In our previous study , we constructed a spherical tumor with uniform vessel density to investigate the benefit from normalization therapy and the results showed increased delivery in the interior regions of tumors of certain sizes [70] . In animal studies , it has been shown that the bulk accumulation of liposomes is not representative of efficacy since it is not informative about the drug accumulation within specific regions of tumors [67 , 71] and heterogeneous drug accumulation may result in tumor repopulation [72] . Therefore , it is important to understand the factors that yield heterogeneous accumulation and strive to avoid them to generate effective treatments . According to our results , it is plausible that administering targeted therapies using large drugs , normalization should be more useful since it can provide a simultaneous access to both tumor rim and center . The dose of chemotherapy should be increased in order to ensure similar drug exposure despite the sparser vessel density caused by antiangiogenic activity . This is the reason why targeted therapies are more suitable to seize the benefits from normalization , as they can be applied in greater doses without harming healthy tissue . When convective extravasation is restored in the central region , drugs can immediately reach to tumor center and increase the probability of treatment success and tumor eradication .	Tumor vessels being very different from their normal counterparts are leaky and lack organization that sustains blood circulation . As a result , insufficient blood supply and high fluid pressure begin to appear inside the tumor which leads to a reduced delivery of drugs within the tumor , especially in tumor center . A treatment strategy that utilizes anti-vascular drugs is observed to revert these alterations in tumor vessels , making them more normal . This approach is suggested to improve drug delivery by enhancing physical transport of drugs . In this paper , we build a mathematical model to simulate tumor and vessel growth as well as fluid pressure inside the tumor . This framework enables us to simulate drug treatment scenarios on tumors . We use this model to find whether the delivery of the chemotherapy drugs is enhanced by application of anti-vascular drugs by making vessels more normal . Our simulations show that anti-vascular drug not only enhances the amount of drugs that is released into tumor tissue , but also enhances drug distribution enabling drug release in the central regions of tumor .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "medicine", "and", "health", "sciences", "vesicles", "cardiovascular", "physiology", "engineering", "and", "technology", "cancer", "treatment", "clinical", "oncology", "drugs", "chemotherapeutic", "agents", "oncology", "angiogenesis", "developmental", "biology", "clinical", "medicine", "pharmaceutics", "nanoparticles", "nanotechnology", "oncology", "agents", "pharmacology", "cellular", "structures", "and", "organelles", "cancer", "chemotherapy", "liposomes", "drug", "delivery", "chemotherapy", "cell", "biology", "physiology", "biology", "and", "life", "sciences", "drug", "therapy", "combination", "chemotherapy" ]	2017	Quantifying the effects of antiangiogenic and chemotherapy drug combinations on drug delivery and treatment efficacy
In the absence of vaccines and limitations of currently available chemotherapy , development of safe and efficacious drugs is urgently needed for visceral leishmaniasis ( VL ) that is fatal , if left untreated . Earlier we reported in vitro apoptotic antileishmanial activity of n-hexane fractions of Artemisia annua leaves ( AAL ) and seeds ( AAS ) against Leishmania donovani . In the present study , we investigated the immunostimulatory and therapeutic efficacy of AAL and AAS . Ten-weeks post infection , BALB/c mice were orally administered AAL and AAS for ten consecutive days . Significant reduction in hepatic ( 86 . 67% and 89 . 12% ) and splenic ( 95 . 45% and 95 . 84% ) parasite burden with decrease in spleen weight was observed . AAL and AAS treated mice induced the strongest DTH response , as well as three-fold decrease in IgG1 and two-fold increase in IgG2a levels , as compared to infected controls . Cytometric bead array further affirmed the elicitation of Th1 immune response as indicated by increased levels of IFN-γ , and low levels of Th2 cytokines ( IL-4 and IL-10 ) in serum as well as in culture supernatant of lymphocytes from treated mice . Lymphoproliferative response , IFN-γ producing CD4+ and CD8+ T lymphocytes and nitrite levels were significantly enhanced upon antigen recall in vitro . The co-expression of CD80 and CD86 on macrophages was significantly augmented . CD8+ T cells exhibited CD62Llow and CD44hi phenotype , signifying induction of immunological memory in AAL and AAS treated groups . Serum enzyme markers were in the normal range indicating inertness against nephro- and hepato-toxicity . Our results establish the two-prong antileishmanial efficacy of AAL and AAS for cure against L . donovani that is dependent on both the direct leishmanicidal action as well as switching-on of Th1-biased protective cell-mediated immunity with generation of memory . AAL and AAS could represent adjunct therapies for the treatment of leishmaniasis , either alone or in combination with other antileishmanial agents . Protozoal infections are a worldwide health problem , particularly in the third world countries [1]–[2] , and account for approximately 14% of the world's population , who are at risk of infection . Leishmaniasis is considered by the WHO as one of the six major infectious diseases , with a high incidence and ability to produce deformities [3]–[4] . Therefore , finding a safe , effective and affordable treatment for such neglected tropical syndromes is a major concern and of high priority [3] . There are two main forms of leishmaniasis: cutaneous , characterized by skin sores; and visceral , which affects the internal organs ( e . g . the spleen , liver , and bone marrow ) . Visceral leishmaniasis ( VL ) is the more severe form , causing significant morbidity and mortality , if left untreated . In the current scenario , the disease is associated with the high cost of treatment and poor compliance . In addition , drug resistance , low effectiveness and poor safety have been responsible for retarding the treatment efficacy of current chemotherapy [5] . Concomitant infection with malaria or pneumonia increases the fatality of the illness if not diagnosed and treated in time . The problem of leishmaniasis has been worsened due to parallel infections in AIDS patients [6]–[7] . In the absence of a credible vaccine , there is an urgent need for effective drugs to replace or supplement those in current use . The pentavalent antimony compounds , which constitute the first line of drugs for treatment of leishmaniasis were developed before 1959 . The resistance to these drugs is now widespread in Bihar , India where 50–65% patients fail to be treated successfully with normal dose schedule of these first line drugs [8] . The new drugs that have become available in recent years for the treatment of VL are AmBisome , the excellent but highly expensive liposomal formulation of Amphotericin B ( AMB ) and the oral drug miltefosine , which has now been registered in India . The toxicity of these agents and the persistence of side effects even after modification of the dose level and duration of treatment are , however , severe drawbacks . Drug combinations like miltefosine/paromomycin and SbIII/paromomycin are also ineffectual , as Leishmania donovani is known to easily develop resistance [9] . In spite of rapid advances in synthetic chemistry that promises to offer new drugs , natural products continue to play an important role in therapy: Of the 1 , 184 new drugs registered between 1981 and 2006 , 28% were natural products or their derivatives . Another 24% of the new drugs had pharmacophores ( i . e . , functional groups with pharmacological activity ) derived from natural products [10] . Thus , a good starting point to find anti-parasitic natural products would be traditional medicinal plants that have been employed to treat infections , in Asia , Africa or America [11] . For both good scientific reasons and strong pragmatism , the WHO also advocates the use of traditional medicines for the treatment of these tropical diseases [12] . In the quest for new antileishmanial agents with negligible adverse effects , it was thus imperative to focus on alternative systems of medicine [7] including anti-parasitic plant extracts or secondary metabolites derived from them , as an alternative to synthetic drugs . It is well documented that a defective cell mediated immune response marks the progression of leishmaniasis and restoration of cellular immunity is critical to disease control [13] . The CD4+ as well as CD8+ T cells have been implicated in resolution of infection [14] . The Th1/Th2 dichotomy of CD4+ T cells is also evident in murine VL where the active diseased state is marked by a predominance of Th2 response whereas protection or cure is denoted by a strong Th1 response [15] . Further , recovery from the disease and resistance to reinfection is attributed to generation of long lasting immunological memory , which is dependent upon parasite specific memory T cells [16] . There is substantial evidence signifying that the immune system synergistically promotes the therapeutic efficacy of antiparasitic drugs [17] . Therefore , antileishmanial drugs that can quickly reverse the immune suppression of the infected host and polarize the response towards Th1 phenotype with generation of immunological memory , besides killing the parasites , are desirable . In the context of our study , many traditional medicinal plants have been shown to possess dual antileishmanial and immunopotentiating activities validating their use in folk medicine [18] , [19] . Artemisia annua ( Asteraceae ) , a well-known traditional medicinal plant , has been extensively used as antimalarial [20]–[21] and anticancer agent [22] . Recently the in vitro and in vivo efficacy of artemisinin ( one of the constituents of A . annua , A . indica and A . dracunculus ) against hepatocellular carcinoma [23] and experimental VL has been reported [24] . Flavonoids of A . annua have been linked to beneficial immunomodulatory activities in subjects affected from parasitic and chronic diseases [25] . The in vitro and ex vivo leishmanicidal activity of the A . annua leaves ( AAL ) and seed extracts ( AAS ) has been evaluated previously against L . donovani promastigotes and intracellular amastigotes by our group [26] . In the present study , we have explored the immunotherapeutic potential of AAL and AAS against VL in L . donovani infected BALB/c mice . Female BALB/c mice aged 6–8 weeks and weighing 20–25 g were used in the present study after prior approval from the Jamia Hamdard Animal Ethics Committee ( JHAEC ) for the study protocol ( Ethical approval judgment number is 459 ) . JHAEC is registered under the Committee for the purpose of supervision and control of experiments on animals ( CPCSEA ) . All animals were individually housed in the Central Animal House of Jamia Hamdard as per internationally accepted norms . The mice were kept in standard size polycarbonate cages under controlled conditions of temperature ( 23 ± 1°C ) , humidity ( 55 ± 10% ) , 12:12 h of light and dark cycle and fed with standard pellet diet ( Ashirwad Industries , Chandigarh , India ) and filtered water ( ad libitum ) . Fresh A . annua leaves and dried seeds with floral parts were collected from the Herbal Garden of Jamia Hamdard , washed , air-dried and ground separately and extracted with n-hexane as described previously [26] . The n-hexane extract of leaves ( AAL ) and seeds ( AAS ) were concentrated to dryness under reduced pressure at 35°C using a rotary evaporator and the semisolid paste further concentrated in a vacuum dessicator . Dosing solutions were prepared aseptically in dimethyl sulphoxide ( DMSO , cell culture grade ) , and diluted further in PBS ( 0 . 02 M phosphate buffered saline , pH 7 . 2 ) to achieve a final DMSO concentration not exceeding 0 . 2% , which is non-toxic . All reagents including AAL and AAS were free of lipopolysaccharide ( 0 . 2 ng/ml endotoxin ) as determined by the Limulus amoebocyte lysate assay . Leishmania donovani ( MHOM/IN/AG/83 ) promastigotes were grown in M199 medium , supplemented with 10% FBS , 2 mM glutamine , 100 units ml−1 penicillin , and 100 µg ml−1 streptomycin sulfate at 22°C . Late stationary phase promastigotes were obtained after incubation of the parasites for 4–5 days with starting inoculum of 1×106 parasites ml−1 [27] . Stationary phase L . donovani promastigotes were used to infect 6 to 8-weeks old BALB/c mice ( 2 ×107/animal ) through tail vein . Ten weeks post infection , parasite burden was confirmed in three arbitrarily selected animals; after which , mice were randomly assigned into seven groups of 10 mice each ( A–F ) . Group A – Control infected mice without any treatment ( INF ) ; Group B - Vehicle control mice that received normal saline orally ( VC ) . Test fractions and compounds were administered to three groups orally: Group C ( AAL ) ; D ( AAS ) ; and E artemisinin ( ART ) . These groups received three doses ( 50/100/200 mg/kg body weight {b . w . } ) daily for ten consecutive days . Group F - received Amphotericin B ( AMB , 5 mg/kg b . w . on alternate days over a 10 day period , intravenously ) and served as the positive control . Ten days post treatment , 5 mice per group were euthanized by carbondioxide asphyxiation , liver and spleen parasite burden determined from giemsa-stained multiple impression smears , and expressed as Leishman-Donovan Units ( LDU ) that was calculated as the number of parasites per 500 nucleated cells x organ weight in mg [28] . Percent reduction of parasite burden was calculated as: ( LDU of infected control - LDU of treated mice ) /LDU of infected control mice × 100 . Cure or protection correlated with a reduction in hepato-splenomegaly and elimination of parasites to negligible levels [28] . Fourteen days-post treatment; the remaining 5 mice per group were sacrificed for evaluating the immunological response . Freeze-thawed leishmanial antigen ( FT ) was prepared as reported previously [29] . Briefly , stationary-phase promastigotes , harvested after the third or fourth passage in liquid culture , were washed four times in cold 1× PBS and resuspended at a cell density of 2×108 cells ml−1 . The preparation was frozen and thawed at 80°C ( 30 min ) and 37°C water bath ( 15 min ) , alternately for 6 cycles , and stored at −70°C until use . Soluble leishmanial antigen ( SLA ) was prepared as reported previously [30] . In brief , the freezing-thawing cycles were repeated ten times , and the suspension finally centrifuged ( 5250 × g , 4 °C , 10 min ) . The supernatant containing soluble leishmanial antigen ( SLA ) was harvested and stored at −70°C until use . The protein content in FT and SLA was measured by the method of Lowry et . al . [31] . The delayed-type hypersensitivity ( DTH ) response in control infected and treated mice was determined as an index of cell-mediated immunity . The response was evaluated by measuring the difference in footpad swelling at 24 h , 48 h and 72 h following intradermal inoculation of the test footpad with 50 µl ( 800 µg ml−1 ) of FT compared to the PBS-injected contra-lateral footpad [14] . Proliferation of splenic and lymphatic lymphocytes as an index of cell mediated immune ( CMI ) response , was evaluated in spleen and lymph nodes ( axilliary , inguinal and popliteal ) by trypan blue dye exclusion as well as by carboxyfluorescein succinimidyl ester ( CFSE ) staining . Spleens from different groups of mice were homogenized and the erythrocytes lysed with lysis buffer ( 20 mM Tris , pH 7 . 4 containing 0 . 14 M NH4Cl ) at room temperature , 10 min . After centrifugation ( 1400 × g , 4°C , 10 min ) , the cells were washed with PBS and resuspended in complete RPMI-1640 medium ( supplemented with 25 mM HEPES ( pH 7 . 4 ) , 50 mM 2-mercaptoethanol , 100 U ml−1 penicillin , 100 µg ml−1 streptomycin and 10% FBS ) . Alternately , the homogenous suspension of lymph nodes was washed and resuspended in complete RPMI 1640 medium . The viability of both splenic and lymphatic lymphocytes as determined by trypan blue dye exclusion [24] exceeded 95% . For assessment of proliferation , the spleen ( 5 × 106 cells ml−1 ) or lymph node ( 2 × 106 cells ml−1 ) cells were cultured at 37°C for 48 h in a humidified atmosphere containing 5% CO2 in the presence of 10 µg ml−1 SLA or Con A ( 5µg ml−1 ) . Proliferation was ascertained by direct counting of viable cells after trypan blue dye exclusion [32] . Alternatively , for assessment of proliferation by CFSE dilution , the lymphocytes isolated from treated , infected and naïve mice were stimulated with SLA ( 10 µg ml−1 ) as described above . The lymphocytes ( 5 ×106 cells ml−1 ) were incubated with 1 µM CFSE . After 48 h , the cells were washed twice with PBS and finally resuspended in PBS . The cells were acquired in a BD LSR II flow cytometer following which the cell population was assessed and contour plots generated after appropriate gating [33] . Nitric oxide ( NO ) , a major microbicidal molecule killing intracellular Leishmania , is released during conversion of L-arginine into citrulline by nitric oxide synthase that is activated by the Th1 subset of CD4+ T cells . The nitrite , the primary , stable and non-volatile product of NO was quantified as an indirect correlate of NO production . The 48 h culture supernatants of peritoneal macrophages of differently treated , infected and naïve mice was analyzed for nitrite contents in the presence or absence of SLA ( 10 µg ml−1 ) and , in parallel re-stimulated with AAL and AAS ( 50 µg ml−1 ) as described previously [34] . Briefly , the Griess reagent ( 1% sulfanilamide and 0 . 3% N- ( 1-naphthyl ) ethylenediamine dihydrochloride in 5% H3PO4 ) was added to the culture supernatant at a 1:1 ratio and incubated for 15 min at room temperature . The optical density ( OD ) was determined at 550 nm using an ELISA reader . Sodium nitrite ( NaNO2 ) diluted in culture medium was used to generate a standard curve . Mice were bled at the time of treatment and at 10 days post treatment , and sera stored at −70°C until use . The specific serum IgG isotype antibody ( Ab ) response was measured by conventional enzyme-linked immunosorbent assay ( ELISA ) [14] . Briefly , wells of ELISA plates ( Nunc , Roskilde , Denmark ) were coated with FT ( 25 µg ml−1 ) and incubated overnight at 4°C . After washing three times with buffer ( 20 mM PBS , pH 7 . 2 containing 0 . 05% Tween 20 ) , the wells were blocked with 1% BSA for 2–3 h at room temperature . The plate was washed and mice sera at 1 , 000-fold dilution was added , followed by washing and incubation with isotype-specific goat anti-mouse IgG1 and IgG2a antibody ( Sigma Aldrich ) at 4°C overnight . The wells were then washed and incubated at 4°C overnight with peroxidase-conjugated rabbit anti-goat IgG ( Sigma Aldrich ) . The wells were washed and incubated with substrate solution ( o-phenylenediamine dihydrochloride , 0 . 8 mg ml−1 in 0 . 02 M phosphate-citrate buffer , pH 5 . 0 , containing 0 . 04% H2O2 ) for 30 min , and the absorbance read on an ELISA plate reader at 490 nm . The Th1 ( IFN-γ ) and Th2 ( IL-4 , IL-10 ) cytokine concentrations in the sera and culture supernatant of lymphocytes from different groups of mice were measured by a bead-based multiplex assay [35]-[36] . This assay used microspheres as the solid support and allowed simultaneous quantification of cytokines in a flow cytometer according to the manufacturer's instructions . Briefly , serum , culture supernatants from SLA-stimulated ( 10 µg ml−1 ) lymphocytes or the cytokine standards were mixed with equal volume of antibody-coated capture beads and subsequently incubated with biotin-conjugated secondary antibody mixture ( anti-mouse ) for 2 h at room temperature in the dark . Beads were then washed ( 400 × g , 4°C , 5 min ) and the supernatant was discarded carefully , leaving approximately 100 µl sample in each tube . This was repeated once , and the samples were further incubated with streptavidin–PE for 1 h at room temperature in the dark . After two further centrifugation steps as mentioned above , the beads were resuspended in assay buffer and read on a BD FACS Calibur ( BD Biosciences ) and analyzed with Cell Quest software . The data were processed using BD CBA software , with results based on a standard concentration curve . Lymphocyte phenotyping was performed as described previously [37] . The spleens ( 1/3 of the organ ) from differently treated and untreated BALB/c mice were placed in PBS and stored on ice prior to preparation of single cell suspension . The splenic erythrocytes were lysed as described above . After centrifugation ( 1400 × g , 4°C , 10 min ) , the cells were washed with FACS buffer ( PBS containing 1%FBS ) . The cell suspensions were refrigerated ( 4°C ) pending staining with antibodies . For each sample , 2 × 106 cells were stained with anti-CD4-FITC and anti-CD8-PE antibodies for 15 min on ice . The cells were then washed and resuspended in PBS for flow cytometric analysis which was performed on a LSR II flow cytometer equipped with DIVA software ( Becton Dickinson ) . Flow cytometry was performed for intracellular analysis of IFN-γ- producing CD4+ and CD8+ T lymphocytes at the single-cell level . Splenocytes from treated and untreated infected mice were stimulated with 10 µg ml-1 SLA for 24 h . Brefeldin A ( 10µg ml-1 ) was added to the culture and incubated for 1 hr . The cells were washed with FACS buffer and stained with APC and PE conjugated anti-CD4 and anti-CD8 antibody , respectively , washed and fixed with 100 µl of intracellular fixation buffer and permeabilized with permeable solution ( BD Pharmingen ) . The cells were subsequently stained with FITC-conjugated anti-IFN-γ or isotype-matched control monoclonal antibodies , and analyzed on a flow cytometer following acquisition . The CD4+ and CD8+ T cells were gated individually for determining the population of FITC positive IFN-γ- producing cells [14] . Splenic cells from differently treated and untreated BALB/c mice were suspended in RPMI-1640 medium after removing the red blood cells with lysis buffer as described above . Cells ( 1 × 107 cells ml−1 ) were washed thrice and incubated for 1 h at 37°C on petri plates . After removing the non-adherent T and B cells , the adherent macrophages were collected and washed with FACS buffer . To quantify the expression of co-stimulatory molecules ( CD80 and CD86 ) on CD11b+ and F4/80+ cells , 2 × 106 macrophages from each sample were stained with PE-labeled anti-CD80 , FITC-conjugated anti-CD86 and APC- labeled anti-CD11b or PE-Cy5- labeled anti-F4/80 monoclonal antibodies on ice for 15 min and washed with PBS . Cells were acquired on a BD LSR II flow cytometer equipped with DIVA software ( Becton Dickinson ) [38] . Spleen cells were isolated from differently treated and infected BALB/c mice , washed with FACS buffer and incubated for 30 min at 4°C with the following fluorochrome-conjugated anti-mouse antibodies: CD8-APC , CD62L-PE and CD44-FITC ( BD Pharmingen ) , and then fixed with 2% paraformaldehyde . Cell acquisition was performed with a BD LSR II flow cytometer [16] . Hepatic and renal functions of BALB/c mice were evaluated in treated and untreated mice as described previously [36] . Fourteen days post-treatment , mice were bled and sera were separated by centrifugation ( 5000 × g , 4°C , 2-3 min ) and stored at -70°C until use . The hepatic and renal functions was assessed by measuring the levels of serum glutamic oxaloacetic transaminase ( SGOT ) , serum glutamic pyruvic transaminase ( SGPT ) , alkaline phosphatase ( ALP ) , urea and creatinine using commercially available kits ( Span Diagnostics Ltd . ) . All the in vitro experiments were performed at least in triplicate . A minimum of five mice per group were used for in vivo experiments . The statistical significance of differences between groups was determined as described in the figure legends using ANOVA followed by Tukey's test by graph pad prism 5 software . P value of <0 . 05 was considered statistically significant . Error bars represent the standard error of the mean ( SEM ) . Results are from one of three representative experiments . AAL and AAS administered to 10 weeks infected BALB/c mice at 200 mg/kg b . w . for 10 consecutive days caused 95 . 45±2 . 05% and 95 . 84±1 . 95% ( P<0 . 001 ) reduction of parasite burden in spleen and 86 . 67±2 . 53% and 89 . 12±1 . 92% ( P<0 . 001 ) in liver , respectively ( Fig . 1A & B ) at 10 days post treatment . At 100 mg/kg b . w . , AAL and AAS induced 88 . 58±1 . 23% and 85 . 08±6 . 92% protection in spleen and 72 . 99±7 . 2% and 80 . 27±1 . 25% in liver , respectively . The lowest dose of AAL and AAS ( 50 mg/kg b . w . ) used in this study , caused more than 70% decrease in parasite load in spleen and approximately 50% in liver . ART was comparatively less effective since even the higher dose ( 200 mg/kg b . w . ) , could lower the parasite burden in liver as well as spleen by only 50% . With AMB ( 5 mg/kg b . w . ) , parasite elimination in liver and spleen was 94 . 02± 1 . 81% and 98 . 09±2 . 44% , respectively . AAL and AAS treatment ( 200 mg/kg b . w . ) also resulted in significant reduction ( 48 . 84% and 45 . 35% respectively ) in spleen weight compared to infected controls ( Fig . 1 C & D ) that was comparable with AMB ( 52 . 33% ) . Since cure of leishmaniasis is associated with an effective immune response , we investigated the possible immunological alterations induced by the treatment of AAL , AAS and ART in L . donovani-infected BALB/c mice at cure . Leishmanial antigen ( FT ) -specific IgG1 and IgG2a isotype levels were assessed in the sera of mice at 10 days post-treatment . Control-infected animals exhibited significantly higher IgG1 than IgG2a levels ( P ≤ 0 . 001 ) compared to treated groups ( Fig . 2 ) . The highest IgG2a/IgG1 ratio was found in AAL ( 2 . 09 ) and AAS ( 1 . 92 ) -treated mice at 200 mg/kg b . w . , followed by AMB treatment group ( 1 . 56 ) . In case of mice treated with ART , the IgG1 levels were significantly higher than IgG2a , resulting in decreased IgG2a/IgG1 ratio ( 0 . 75 ) even at the higher dose ( 200 mg/kg b . w . ) . Chemotherapeutic intervention and cure is generally associated with the acquisition of a DTH response and consequently “classical” cell-mediated immunity [39] . Hence , we investigated FT-induced DTH responses in infected BALB/c mice at 10 days post-treatment with AAL , AAS , ART and AMB . AAL and AAS treated mice showed the strongest DTH response at 200 mg/kg . b . w . ; a significant increase in footpad thickness was observed at 24 h ( 0 . 45±0 . 07 and 0 . 48±0 . 14 , respectively ) as compared with the INF ( 013±0 . 004 ) control group , followed by 100 and 50 mg/kg b . w . ( Fig . 3 ) . Whereas AMB ( 0 . 24±0 . 06 ) and ART ( 0 . 20±0 . 03 ) treated mice showed a marginal levels of DTH response . There was almost no change in DTH response at 48 versus 24 h; however , the DTH reactivity waned at 72 h in all the groups ( Fig . 3 ) . Active VL is characterized by marked T-cell anergy toward leishmanial antigens [40]-[42] . By direct enumeration under microscope , we observed a significant proliferative response of SLA-stimulated splenocytes and lymphocytes from mice at 10 days post-treatment with AAL and AAS . Maximum effect was found at 200 mg/kg b . w . followed by 100 and 50 mg/kg b . w . , whereas AMB and ART treatment showed marginal levels of SLA-specific lymphoproliferation in splenic and lymph node cells ( Fig . 4A & 4B ) . Alternately , lymphoproliferative capacity of lymphocytes after treatment with different groups was assessed by CFSE labeling . The percentage of normal cells that underwent division in spleen and lymph nodes was 17 . 8% and 16 . 6% , respectively . AAL and AAS treated ( 200 mg/kg bw ) groups exhibited the highest lymphoproliferative response in spleen ( 32 . 1% and 34 . 2% ) and lymph nodes ( 31 . 4% and 36 . 8% ) . AMB and ART treated groups induced low levels of lymphoproliferation ( 22 . 2% and 20 . 1% ) in spleen as well as lymph nodes ( 21 . 0% and 19 . 8% ) that was slightly higher than that observed in spleen ( 18 . 6% ) and lymph nodes ( 17 . 9% ) of infected control group ( Fig . 4C & 4D ) . The effect of AAL and AAS on macrophage function was assessed by measuring the amount of Nitric oxide ( NO ) produced by peritoneal macrophages of treated mice . Griess reagent was used to measure the nitrite levels , the stable end product of NO metabolism . The nitrite concentration ( µM ) was determined by extrapolation from a standard curve generated with sodium nitrite . In macrophages of AAL and AAS treated mice , a dose dependent NO production was observed upon in vitro re-stimulation with SLA followed by AAL and AAS stimulation and un-stimulation . Higher levels of nitrite were produced in AAL ( 11 . 18± 0 . 81µM ) and AAS ( 11 . 33±0 . 63µM ) treated mice ( 200 mg/kg/b . w . ) after re-stimulation with SLA as compared to INF ( 3 . 11±0 . 09µM ) control ( Fig . 5 ) . In contrast , AMB and ART treatment induced low nitrite levels ( 6 . 05±0 . 20 and 5 . 18±0 . 09µM , respectively ) in peritoneal macrophages . To evaluate the immune alterations , Th1 ( IFN-γ ) and Th2 ( IL-4 and IL-10 ) signature cytokines in serum and culture supernatants were estimated by bead-based multiplex assay . Mice treated with AAL and AAS ( 200 mg/kg b . w . ) induced significantly elevated levels of serum IFN-γ ( 2771±50 . 91 and 3033±396 . 69 pg ml−1 ) and reduced levels of IL-4 ( 4020±91 . 92 and 3961 . 5±208 . 24 pg ml−1 ) and IL-10 ( 4231 . 5±459 . 27 and 4077 . 5±35 . 0 pg ml−1 ) compared to untreated infected controls INF ( low IFN-γ; 1236±12 . 37 , high IL-4; 5696±79 . 9 and high IL-10; 5049±101 . 47 ) ( Fig . 6A ) . AMB and ART induced low levels of these cytokines compared to INF . Similar pattern of Th1 and Th2 cytokines was observed in the culture supernatant of lymphocytes from mice treated with AAL and AAS and that was significantly high compared with infected ( INF ) control ( Fig . 6B ) . It is well established that MHC class II-restricted CD4+ T cells are dominant during the development of immunity against Leishmania [43] . However , a few studies point to an essential role for CD8+ cells in immunity to primary infection with L . major [44] and also in the induction of long-term , vaccine-induced resistance against many intracellular pathogens [13] . A low population of CD4+ ( 8 . 2% ) and CD8+ ( 5 . 1% ) T cells were detected in the spleens of mice with established L . donovani infection ( Fig . 7 ) . The population of CD4+ and CD8+ T cells increased 10 days after treatment with 50 mg/kg b . w . of AAL ( 14 . 4% and 10 . 9% ) and AAS ( 16 . 8% and 12 . 7% ) , respectively . The increase was slightly more at a treatment dose of 100 mg/kg b . w . of AAL ( 18 . 4% and 13% ) and AAS ( 18 . 9% and 15 . 4% ) . The CD4+ and CD8+ T cell population was , however , highest at 200 mg/kg b . w . treatment with AAL ( 20 . 7% and 15 . 2% ) and AAS ( 21 . 8% and 16 . 12% ) ( Fig . 7 ) . These findings demonstrate a prominent inclination toward Th1 effector function and the involvement of both CD4+ and CD8+ T cells at cure with AAL and AAS treatment . These responses in case of ART and AMB treated groups were negligible . Both CD4 and CD8 T cells are source of IFN-γ and are essential for resolution of leishmaniasis [13] , [43] . In infected mice , a low frequency of CD4 ( 14 . 43±0 . 40% ) and CD8 ( 12 . 59±0 . 58% ) T cells secreting IFN-γ was detected which was elevated by AMB treatment ( CD4 20 . 13± 0 . 71% , CD8 18 . 02± 0 . 45% ) . However , the maximum induction of IFN-γ-producing CD4 and CD8 T cells was observed after AAL ( 32 . 05±0 . 55% , 27 . 16±0 . 42% ) and AAS ( 33 . 37±0 . 74% , 28 . 09±0 . 41% ) treatment at 200 mg/kg . b . w . In case of ART ( 200 mg/kg . b . w ) treatment no significant increase in the frequencies of IFN-γ producing CD4 ( 16 . 41±0 . 46% ) and CD8 ( 14 . 92±0 . 37% ) T cells were observed ( Fig . 8A & 8B ) . Ligands on antigen presenting cells ( APCs ) called co-stimulatory molecules interact with specific receptors on T cells providing the second signal which then leads to activation of the antigen stimulated T cells . CD80 and CD86 present on APCs are essential for the activation of lymphocytes and the secretion of cytokines . The expression of CD80 and CD86 co-stimulatory molecules was analyzed individually on CD11b ( monocytes ) and F4/80 ( macrophages ) . CD80 and CD86 expression and co-expression was particularly enhanced on F4/80 gated cells in comparison to CD11b positive cells . AAL and AAS treatment ( 200 mg/kg . b . w ) significantly up-regulated the expression of CD80 ( 20 . 3% and 21 . 13% ) and CD86 ( 14 . 6% and 16 . 0% ) as well as co-expression of CD80/86 ( 21 . 9% and 26 . 0% ) on F4/80 cells . AMB treatment was less effective in up-regulating CD80 ( 7 . 9% ) and CD86 ( 7 . 1% ) expression and CD80/86 ( 13 . 9% ) co-expression , which was followed by ART . ART induced the lowest level of CD80 ( 5 . 6% ) and CD86 ( 6 . 6% ) expression and CD80/86 ( 10 . 3% ) co-expression ( Fig . 9A ) Similar pattern of CD80 and CD86 expression and CD80/86 co-expression was modulated on CD11b positive cells . AAL and AAS treatment induced maximum up-regulation of CD80 ( 11 . 7% and 15 . 7% ) and CD86 ( 12 . 6% and 15 . 4% ) expression along with CD80/86 ( 13 . 1% and 17 . 0% ) co-expression in comparison with infection control where as in AMB and ART treated groups , the effect was less pronounced ( Fig . 9B ) . Resistant to Leishmania re-infection is attributed to generation of memory T cells in the host [16] . CD44 CD62L expression was low in infected control group ( 10 . 1% ) , which was up-regulated after treatment with AAL ( 13 . 8% ) and AAS ( 15 . 1% ) at 200 mg/kg b . w . demonstrating resolution of infection and generation of memory . AMB ( 9 . 7% ) and ART ( 9 . 4% ) treatment exhibited negligible effect on generation of memory T cells . Further the percentage of effector memory cytotoxic T cells ( CD44high CD62Llow ) was also increased following treatment with AAL ( 37 . 1% ) , AAS ( 38 . 8% ) and AMB ( 37 . 3% ) and no such up-regulation was evident in case of ART ( 27 . 9% ) treatment ( Fig . 10 ) . Estimation of ALP , SGOT and SGPT for liver dysfunction and urea and creatinine for renal dysfunction was done ten days post-administration of AAL , AAS and ART in normal BALB/c mice ( Table 1 ) as well as infected and treated mice ( Table 2 ) . AAL , AAS and ART ( up to 200 mg/kg b . w . ) treated group demonstrated normal levels of serum enzymes , indicating no in vivo toxicity . The fractions thus proved to be non-toxic in BALB/c mice used in antileishmanial screening . In the absence of effective vaccines , emerging resistance against current chemotherapeutic drugs or their combinations , and a stigma of being an AIDS-defining illness , improved therapy for leishmaniasis remains desirable . Plant extracts represent a natural library of potentially bioactive molecules that can activate intrinsic leishmanicidal mechanisms . In our earlier studies , we reported potent anti-leishmanial activity of AAL and AAS with selective elimination of the parasites without affecting host macrophages . The leishmanicidal effect was mediated by programmed cell death . α-amyrinyl acetate , β-amyrine and precursors of artemisinin were the major constituents in AAL and cetin , EINECS 211-126-2 and artemisinin precursors in AAS [26] . In an effort to realize the full therapeutic potential of AAL and AAS , in the present study , we have explored the efficacy of AAL and AAS against VL using L . donovani infected BALB/c mice . The major findings emerging from this study are that AAL and AAS ( 200 mg/kg b . w ) result in maximum clearance of parasites ( 85 to 90% ) from the liver and spleen of infected BALB/c mice as compared to untreated infected controls . Significant reduction ( 88%–96% ) in spleen weight was also observed with AAL and AAS . While , only marginal numbers of parasites were cleared from the liver and spleen upon treatment with artemisinin ( ART ) even at higher dose ( 200 mg/kg b . w . ) . Similar therapeutic effect has also been reported with the extracts of Tinospora sinensis [45] , Aloe vera [27] , Actinopyga lecanora [46] and with the essential oil of Chenopodium ambrosioides [47] and Bixa orellana [48] . VL is characterized by a variety of immunopathological consequences in man . The most remarkable of these are depression of CMI response and B cell activation [49] . As an index for CMI , DTH , a type IV hypersensitivity reaction was measured in treated mice . DTH develops when antigen activates sensitized TDTH cells resulting in secretion of IFN-γ and IL-2 [50] that promotes enhanced phagocytic activity of the recruited macrophages for effective killing of the parasites . DTH reaction is thus important in host defense system against Leishmania parasites . The importance of a positive DTH response in human leishmaniasis is illustrated by the fact that apparent clinical cure in the absence of a positive DTH response is often predictive of a relapsing infection [51] . Our results demonstrated that the DTH response was depressed in L . donovani infected BALB/c mice . However , treatment with AAL and AAS stimulated maximum DTH response at 24 h while negligible levels of DTH were induced with ART . Elicitation of DTH response has also been observed with Asparagus racemosus extracts [52] and Prunus cerasus treatment in infected BALB/c mice . Resistance against Leishmania infection remains largely associated with a polarized Th1 and an insufficient Th2 response . Cytokines such as IFN-γ and IL-4 direct immunoglobulin class switching in B cells to IgG2a and IgG1 , respectively [53] , as an indirect correlate of T helper subsets potentiated . Thus , IgG2a and IgG1 levels indirectly reflect the Th1/Th2 responses and hence their relative production is used as a surrogate marker for the induction of protective ( Th1 ) or deleterious ( Th2 ) type of immune responses . To assess the immunological status of the mice upon treatment , we evaluated serum levels of parasite-specific IgG1 and IgG2a . L . donovani infection in BALB/c mice resulted in increased IgG1 and decreased IgG2a levels . However , treatment with AAL and AAS showed three-fold decrease in IgG1 with approximately two-fold increase in IgG2a levels as compared to infected controls . ART treatment did not reveal any significant difference in isotype levels . Our data reflecting higher levels of IgG2a over IgG1 thus indicate that Th1-mediated protective immunity is generated by AAL and AAS treatment . Our results comply with the reports of Sachdeva et al . , [52] who showed decrease in IgG1 coupled with increase in IgG2a levels upon treatment of L . donovani infected BALB/c mice with A . racemosus in combination with cisplatin . Aqueous extract of A . racemosus has also been reported to result in significant increase in antibody titers [15] and upregulation of Th1 and Th2 cytokines [54] , suggesting Th1/Th2 adjuvant activity . Bhattacharjee et al . , [55] reported that the expression of Th1 signature cytokines ( IFN-γ and IL–2 ) is protective for VL whereas expression of Th2 cytokines viz . IL-4 and IL-10 increases during infection . Gomes et al . [56] showed that orally administered Kalanchoe pinnata selectively suppress IgG and IL-4 and up-regulates IFN-γ production in murine VL . Our studies are in agreement with these observations as AAL and AAS treatment generated a protective immunity through induction of IFN-γ and decline in IL-4 and IL-10 in serum as well as culture supernatants of spleen cells . The percentage of CD4 and CD8 T cells producing IFN-γ also increased after AAL and AAS treatment as depicted by intracellular staining . Further , AAL and AAS ( 200 mg/kg b . w . ) stimulated strong lymphoproliferative responses in lymph nodes as well as spleens , which was observed by CFSE dilution and trypan blue dye exclusion . The increased levels of IFN-γ correlated with the strong proliferative response and activation of Th1 subset of CD4+ T cells . Efficiency of chemotherapy in leishmaniasis is also impaired due to suppression of immune functions during the course of infection [57] . Disease outcome of VL is associated with various immunological dysfunctions . Successful chemotherapy requires strong cellular responses based on CD4+ and CD8+ T cells . Experimental mouse models of VL show that CD8+ T cells are important in control of L . donovani/L . infantum infection in the liver , through their ability to produce IFN-γ and/or their cytolytic activity [58] . Moreover , CD8+ T cells , together with CD4+ cells , are required to control and prevent reactivation of VL in mice [59] . Asparagus racemosus and Prunus cerasus have also been reported to enhance the percentage of CD4+ and CD8+ T cells in spleen of naive [60] and L . donovani infected BALB/c mice upon subsequent treatment [50] . The results of the present investigation revealed that the percentage of CD4+ and CD8+ T lymphocytes in spleens of L . donovani infected BALB/c mice were greatly augmented by AAL and AAS at 200mg/kg b . w . as compared to untreated infected controls as well as ART treated mice . The therapeutic implication of AAL and AAS in VL was further exploited by scoring the memory differentiation markers CD44 and CD62L . AAL and AAS induced generation of immunological memory as characterized by expression of CD62Llow and CD44high on CD8+ T lymphocytes . The stimulatory Th1/Th2 balance is dictated by the presence of other costimulatory stimuli simultaneously acting on T cells and antigen-presenting cells ( APCs ) that play crucial roles in eliminating intracellular pathogens . The optimal activation of naive T cells is achieved by occupancy of T-cell receptor ( TCR ) by the peptide-MHC complex displayed on the surface of APCs , delivery of co-stimulatory signals , and the presence of pro-inflammatory cytokines [61] . Ligation of CD28 with CD80 and CD86 is known to induce the secretion of IL-6 and IFN-γ by DCs for T and B cell activation , proliferation , and differentiation [62] . CD80 and CD86 expression has been reported to be down modulated in certain diseases [63]-[64] . The expression and co-expression of CD80 and CD86 was analyzed on CD11b+ and F4/80+ cells . Treatment with AAL and AAS significantly enhanced CD80 and CD86 expression and co-expression on both CD11b+ and F4/80+ cells however maximum expression was observed in case of F4/80+ population . Thus , our results suggest the potential of AAL and AAS in activating the APCs through co-stimulatory signals that eventually help in the generation of effective immune response by secreting various signaling molecules like IFN-γ for subsequent activation , proliferation , and differentiation of lymphocytes . ART treated mice did not show significant expression of CD80 and CD86 co-stimulatory molecules . Macrophages can be activated by different signals leading to their development into functionally distinct subsets with different disease outcomes . Thus , appropriate activation of macrophages is crucial for eliminating this intracellular pathogen . Macrophage stimulation is mediated by the products of Th1 and NK cells in particular , IFN-γ , which stimulates macrophages to produce inducible nitric oxide synthase ( iNOS , also known as NOS2 ) , an enzyme which catalyzes L-arginine to generate NO and citrulline [65] . NO is a toxic molecule that plays a major role in killing intracellular Leishmania parasites by the production of reactive oxygen species and generation of peroxinitrite . Such metabolites can cause protein , lipid and nucleic acid oxidation [66] . The function of NO in the leishmanicidal activity of activated macrophages has been demonstrated both in vitro and in vivo [67] . Our data demonstrate that AAL and AAS treatment in L . donovani infected BALB/c mice induces high levels of nitrite in SLA-stimulated macrophages ( Fig . 5A ) as compared to infection control as well as ART treated mice , suggesting that the inhibitory effect of the AAL and AAS on infection index is mediated by NO . In the absence of SLA , the NO production was muted but after re-stimulation with AAL and AAS , the NO production was upregulated . The impairment of kidney function and deterioration of liver function by chemotherapeutic agents is recognized as the main side effect and the most important dose limiting factor associated with their clinical use . There is a continuous search for agents , which provide nephro- and hepatic- protection against the renal and liver impairment induced by chemotherapeutic drugs for which allopathy offers no remedial measures . In the current study , AAL , AAS and ART were administered ( 50 , 100 and 200mg/kg b . w . ) in normal and L . donovani infected BALB/c mice . It was found that serum levels of SGOT , SGPT , ALP , urea and creatinine in treated mice were comparable to those in naive mice , indicating absence of nephro- and hepato-toxicity . Taken together , our findings indicate that treatment of infected mice with AAL and AAS significantly decreased the hepatic and splenic parasite load with reduction in spleen weight . AAL and AAS caused increased production of Th1 cytokines ( IFN-γ ) and concomitant decrease in Th2 signature cytokines ( IL-4 and IL-10 ) . The Th1 subset potentiation was also evident from class switching in B cells to produce higher levels of IgG2a over IgG1 and significant elicitation of DTH . AAL and AAS also resulted in higher CD4+ and CD8+ T cell numbers , lymphoproliferation , up-regulation of co-stimulatory molecules ( CD80 and CD86 ) on APCs and generation of NO . Cure as well as resistance against L . donovani infection was due to the parasite killing by AAL and AAS that was mediated by immunopotentiating effects shifting Th1/Th2 balance in favour of the host with induction of cell-mediated immunity as postulated in Fig . 11 . AAL and AAS may emerge as prospective antileishmanial therapy that may be administered alone or synergistically with current chemotherapeutic drugs , owing to their safety and ability to enhance disease healing Th1 immune responses .	Visceral leishmaniasis ( VL ) is a fatal , vector-borne tropical disease that affects the poorest sections of the society . The currently available drugs are toxic , expensive and have severe side effects . The problem is further compounded by emergence of VL-HIV co-infection and occurence of PKDL after apparent cure . Thus , alternate therapeutic interventions are needed in the absence of vaccines and mounting drug resistance . VL is also characterized by severe depression of cell-mediated immunity that complicates the efficiency of chemotherapeutic drugs . Restoration of the dampened immune system coupled with antileishmanial effect would be a rational approach in the quest for antileishmanial drugs . Plant derived secondary metabolites have been recommended for the containment of antiparasitic disease including leishmaniasis that synergistically aid in lifting the immune suppression . We previously reported in vitro antileishmanial activity of n-hexane fractions of Artemisia annua leaves ( AAL ) and seeds ( AAS ) that was mediated by apoptosis . In this study , we found significant reduction in liver and spleen parasite burden of Leishmania donovani infected BALB/c mice upon oral administration of AAL and AAS with concomitant immunostimulation and induction of immunological memory . The immunotherapeutic potentiation by AAL and AAS with no adverse toxic effects validates their use for treatment of this debilitating disease .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "medicine", "and", "health", "sciences", "clinical", "medicine", "clinical", "immunology", "complementary", "and", "alternative", "medicine", "herbal", "medicine", "biology", "and", "life", "sciences", "immunology", "microbiology", "parasitology" ]	2015	Th1-Biased Immunomodulation and Therapeutic Potential of Artemisia annua in Murine Visceral Leishmaniasis
Chlamydia spp . are intracellular obligate bacterial pathogens that infect a wide range of host cells . Here , we show that C . caviae enters , replicates , and performs a complete developmental cycle in Drosophila SL2 cells . Using this model system , we have performed a genome-wide RNA interference screen and identified 54 factors that , when depleted , inhibit C . caviae infection . By testing the effect of each candidate's knock down on L . monocytogenes infection , we have identified 31 candidates presumably specific of C . caviae infection . We found factors expected to have an effect on Chlamydia infection , such as heparansulfate glycosaminoglycans and actin and microtubule remodeling factors . We also identified factors that were not previously described as involved in Chlamydia infection . For instance , we identified members of the Tim-Tom complex , a multiprotein complex involved in the recognition and import of nuclear-encoded proteins to the mitochondria , as required for C . caviae infection of Drosophila cells . Finally , we confirmed that depletion of either Tom40 or Tom22 also reduced C . caviae infection in mammalian cells . However , C . trachomatis infection was not affected , suggesting that the mechanism involved is C . caviae specific . Chlamydia spp . are Gram-negative , obligate , intracellular bacterial pathogens that infect a wide range of hosts and cause various diseases . Three species infect humans . C . trachomatis is the leading cause of preventable blindness in developing countries [1] and the most common cause of bacterial sexually transmitted disease in developed countries [2] . Infection with C . pneumoniae leads to pneumonia , and in the past 10 years , C . pneumoniae has been implicated in atherosclerosis [3] and Alzheimer disease [4] , although the direct links between the bacteria and these diseases is still unclear . C . psittaci infects various animals and is responsible for pneumonia in humans [5] . Many Chlamydia species are recognized as animal pathogens [6] . C . muridarum infects mice and hamsters . C . suis , C . abortus , and C . felis infect swine , ruminants , and house cats , respectively . Finally , infection with C . caviae in guinea pig resembles ocular and genital infections caused by C . trachomatis in humans . Chlamydia are characterized by a biphasic developmental cycle that occurs exclusively in the host cell . The bacteria alternate between an infectious , metabolically inactive form called elementary body ( EB ) that is characterized by a condensed nucleoid , and an intracellular , metabolically active form named reticulate body ( RB ) . Once internalized , Chlamydia resides in a membrane-bound compartment , named the inclusion . Shortly after uptake , an uncharacterized switch occurs , leading to the differentiation of EBs into RBs . The RBs then start to replicate until the inclusion occupies a large part of the cytosol of the host cells . At the end of the cycle , which lasts 2 to 3 d depending upon the species , the RBs differentiate back into EBs . The host cell is lysed , leading to the release of EBs and the infection of neighboring cells [7 , 8] . Both bacterial and host factors contribute to the biogenesis of the inclusion , but little is known about the mechanisms involved . Chlamydia spp . possess a type III secretion system ( TTSS ) responsible for the secretion of effector proteins in the cytoplasm of the host cell . An example of such effectors is the family of highly hydrophobic Inc proteins . Some of them are present on the surface of the inclusion membrane and are thought , in combination with other bacterial effector proteins , to modify the host cell environment and allow bacterial replication [9–13] . During the cycle , Chlamydia targets various host cell functions in order to establish its replication niche and disseminate from cell to cell [14] . The bacteria acquire amino acids , nucleotides , and other precursors from the host cell . The mechanism of chlamydial entry is not well understood , but among others , heparan sulfate proteoglycans , tyrosine phosphorylation of the bacterial effector Tarp , and activation of small GTPases and signaling pathways leading to actin remodeling are involved in this process [15] . Once internalized , Chlamydia directs the trafficking of the nascent inclusion to a perinuclear localization via a mechanism involving microfilaments , microtubules , and the motor protein dynein [16] . The inclusion does not interact with the endocytic pathway [14 , 17] . However , it intercepts exocytic vesicles and lipids from the Golgi [18] . Some Rab GTPases are recruited to the inclusion membrane [19] , and a recent study suggests that Chlamydia targets host lipid droplets to enhance its intracellular survival and replication [20] . Finally , Chlamydia has the ability to modulate the programmed cell death pathway of infected cells [21 , 22] . During the early stage of infection , the infected cells are resistant to apoptosis signals but , by the end of Chlamydia developmental cycle , the programmed cell death pathway is induced , presumably to facilitate the release of the bacteria and the initiation of the next round of infection . In the past few years , Drosophila has been established as a useful model to dissect microbial pathogenesis [23] . Among others , Pseudomonas aeroginosa [24] , Mycobacterium marinum [25] , Salmonella [26] , and Listeria monocytogenes [27] successfully infect Drosophila adult flies . Host–pathogen interaction can also be analyzed in Drosophila S2 cells , which resemble embryonic hemocytes/macrophages . For example , the intracellular replication of L . monocytogenes [27 , 28] or Legionella pneumophila [29] in Drosophila cell lines is similar to the one observed in mammalian cells , and the first steps , but not the latest ( RB to EB differentiation ) , of C . trachomatis developmental cycle can be observed in Drosophila cells [30] . An important discovery was made by Clemens et al . , who reported that the simple addition of dsRNA to Drosophila cells in culture reduces or eliminates the expression of target genes by RNA interference ( RNAi ) , thus efficiently phenocopying loss-of-function mutations [31] . Combined with the sequence of the Drosophila genome , it has opened a new area of research , allowing scientists to test the involvement of any Drosophila gene in a given cellular process [32 , 33] . Several screens have already shed light on various cellular processes such as cell viability [33] , cytokinesis [34] , wnt signaling [35] , JAK/STAT signaling [36] , and mechanisms of host–pathogen interaction , including Listeria and Mycobacterium pathogenesis [37–39] , Candida albicans phagocytosis [40] , and L . pneumophila exploitation of the early secretory pathway [29] . We have investigated the possibility of using Drosophila Schneider's Line 2 ( SL2 ) cells [41] as a model system to dissect Chlamydia pathogenesis . We have shown that C . caviae enters , replicates , and performs a complete developmental cycle in Drosophila SL2 cells . We performed a genome-wide RNAi screen and identified 54 factors that , when depleted , inhibit C . caviae infection in Drosophila cells . We identified factors expected to have an effect on Chlamydia infection , but most importantly we also identified uncovered host factors , including components of the Tim-Tom complex . Clearly validating our approach , we showed that depletion of either Tom40 or Tom22 also reduced C . caviae infection in mammalian cells . We discuss how further investigation of the identified candidates may shed light on the molecular mechanisms involved in Chlamydia pathogenesis . Drosophila SL2 cells [41] were cultured at 25 °C in Schneider media ( Invitrogen ) supplemented with 10% heat inactivated FBS ( JRH ) . HeLa 229 cells were cultured at 37 °C with 5% CO2 in DMEM high glucose ( Invitrogen ) supplemented with 10% heat inactivated FBS ( Invitrogen ) . C . caviae , the guinea pig model of genital and ocular infection of C . trachomatis , were obtained from R . Rank ( University of Arkansas ) . C . trachomatis Lymphogranuloma venerum , Type II , were obtained from ATCC ( VR-902B ) . SL2 cell infection with GFP-expressing L . monocytogenes was conducted as previously described [37] . For propagation , HeLa 229 were incubated with C . caviae or C . trachomatis for 48 h in the presence of 2 μg/ml cycloheximide ( Sigma ) . The infected cells were centrifuged ( 10 min , 1 , 000 rpm ) and the cell pellet was resuspended in SPG buffer ( 218 mM sucrose , 3 . 76 mM KH2PO4 , 7 . 1 mM KH2PO4 , 4 , 9 mM glutamate [pH 7 . 4] ) . The cells were broken by passing them through a 261/2 gauge needle and the unbroken cells and nuclei were pelleted by centrifugation ( 10 min , 1 , 000 rpm ) . The supernatant was centrifuged ( 30 min , 12 , 000 rpm ) , and the bacterial pellets were resuspended in SPG buffer and stored at −70 °C . For Drosophila SL2 cell infection , C . caviae were diluted in Schneider media supplemented with 10% heat inactivated FBS and incubated with the cells at 30 °C for the indicated time . For HeLa 229 cell infection , C . caviae or C . trachomatis were diluted in DMEM high glucose supplemented with 10% heat inactivated FBS and incubated with the cells at 37 °C in the presence of 5% CO2 . One hour post infection , the bacteria were washed away and the cells were incubated with fresh media for the indicated length of time at 37 °C in the presence of 5% CO2 . The following primary antibodies were used: ( FITC ) -conjugated C5+C8 monoclonal antibodies directed against Chlamydia MOMP and LPS ( 1:300 , Argene ) , rabbit polyclonal anti IncA ( 1:200 , [42] ) , guinea pig polyclonal antibody directed against C . caviae EBs ( Kind gift of R . Rank , University of Arkansas ) , rabbit polyclonal antibody anti-hTom40 ( 1:500 , Kind gift of M . Ryan , La Trobe University , Australia [43] ) , mouse monoclonal anti-Tom22 ( 1:2000 , Sigma , clone 1C9–2 ) , and rabbit polyclonal anti-actin ( 1:10 , 000 , Sigma A2066 ) . The following secondary antibodies were used: goat anti-rabbit AlexaFluor 594 antibody ( 1:1 , 000 , Molecular Probes ) , fluorescein ( FITC ) -conjugated AffiniPure donkey anti-guinea pig IgG ( 1:500 , Jackson ImmunoResearch ) , peroxidase-conjugated goat anti-rabbit IgG ( 1:10 , 000 , Jackson ImmunoResearch ) , and peroxidase-conjugated goat anti-mouse IgG ( 1:10 , 000 , Jackson ImmunoResearch ) . At the indicated time , the cells were fixed for 30 min in PBS containing 4% paraformaldehyde . Immunostainings were performed at room temperature . Antibodies were diluted in PBS containing 0 . 16 μg/ml Hoechst ( Molecular Probes ) , 0 . 1% BSA , and 0 . 05% saponin . Samples were washed with PBS containing 0 . 05% saponin , and a final PBS wash was performed before examination under an epifluorescence microscope . Drosophila SL2 cells ( 108 ) were incubated at 30 °C with C . caviae ( MOI ∼ 5 ) , fixed 45 h post infection by addition of 0 . 125% glutaraldehyde / 2% paraformaldehyde in 0 . 1 M phosphate buffer ( pH 7 . 4 ) , postfixed with osmium tetroxide , dehydrated in ethanol , embedded in epoxy resin , sectioned , stained with 1% uranyl acetate , and examined by electron microscopy [44] . HeLa 229 cells cultured on coverslips were fixed in 2 . 5% glutaraldehyde in 0 . 1 M sodium cacodylate ( pH 7 . 4 ) for 1 h at room temperature , postfixed in 1% osmium tetroxide in the same buffer for 1 h at room temperature , stained in 2% uranyl acetate in 50 mM sodium maleate ( pH 5 . 2 ) for 1 h at room temperature , dehydrated in ethanol , and embedded in Embed 812 epoxy resin ( all reagents from Electron Microscopy Sciences ) . Ultra-thin sections ( 60 nm ) were obtained on a Reichert ultra microtome , transferred onto formvar- and carbon-coated hexagonal nickel grids , stained with 1% lead citrate and 2% uranyl acetate , and examined in a Tecnai 12 Biotwin electron microscope ( FEI Company ) . Random images of vacuoles were recorded at a magnification of 11 , 500 using a Morada CCD camera ( Olympus Soft Imaging Solutions ) . For quantitation of the percentage of vacuolar membrane or nuclear envelope covered by mitochondria , a grid with a distance of 560 nm between lines was superposed on top of the images , and the number of intersections of vertical and horizontal lines with membranes counted . The number of intersections of these lines with mitochondria was also counted , but mitochondria were counted as being associated with the vacuolar or nuclear membrane only if the distance between the point of intersection of the grid with the mitochondrial outer membrane and the closest vacuolar or nuclear membrane was 50 nm or less . The ratio of the number of intersection with mitochondria divided by the number of intersections with the vacuolar or nuclear membrane gives an estimate of the percentage of these membranes covered by mitochondria . Drosophila SL2 cells ( 108 ) were incubated at 30 °C with C . caviae . At the indicated time , the infected cells were processed as described above for Chlamydia propagation . The bacterial pellets were resuspended in 100 μl of SPG . To test for the presence of infectious C . caviae in the preparation , 300 μl of a 1:100 dilution were incubated with 6 . 104 HeLa cells seeded onto coverslips at 37 °C in the presence of 5% CO2 . After 1 h , the bacterial suspension was replaced by 500 μl of fresh medium . The cells were fixed 24 h post infection , stained , and the percentage of cells containing a large inclusion was determined by visual inspection using an epifluorescence microscope . The infection was performed in 384-well format such that 75% of the cells were infected . At the indicated time , the infected cells were collected and transferred to an eppendorf tube containing 100 μl of glass beads ( Sigma , G8772 ) and 300 μl of DMEM high glucose supplemented with 10% FBS . The cells were broken by vortexing for 1 min , and 40 μl of dilutions of the lysat were added to 4 . 103 HeLa 229 cells seeded in 384-well plate . After 1 h at 37 °C in the presence of 5% CO2 , the lysat was washed away and 40 μl of fresh media was added to each well . The cells were fixed and stained 24 h post infection and the percentage of infected cells was determined . Two sets of 42 384-well plates containing 0 . 25 μg of dsRNA per well were provided by the Drosophila RNAi Screening Center ( Harvard Medical School , Boston , Massachusetts , http://www . flyrnai . org ) . Drosophila SL2 cells ( 2 . 104 ) , resuspended in 20 μl of serum-free Schneider media , were seeded in each well and incubated 1 h at 25 °C before the addition of 20 μl of Schneider media containing serum . After 3 . 5 d , the cells were infected by addition of 10 μl of Schneider media containing C . caviae . The cells were centrifuged for 1 min at 1 , 000 rpm and incubated at 30 °C for 48 h . The cells were processed for immunofluorescence by using the DNA dye Hoeschst and FITC-conjugated C5+C8 monoclonal antibodies . An automated microscope was used to automatically track , focus , and capture fluorescent images of the cells within each well across an entire plate . One set of images was captured in the blue channel to detect the cells' nuclei and one set in the green channel to detect Chlamydia . The qualitative analysis of the image data was done by visual inspection . dsRNA used for validation and secondary assays were synthesized using a MEGAscript High Yield transcription kit ( Ambion ) according to the recommendation of the manufacturer . The protocol used for siRNA transfection was adapted from Dharmacon's HeLa cells transfection protocol . One volume of siRNA buffer containing 200 nM of siRNA was incubated with 1 volume of serum-free DMEM high glucose containing 5 μl/ml DharmaFECT-1 transfection reagent for 20 min at room temperature . Two volumes of DMEM high glucose supplemented with 20% FBS containing 5 . 104/ml HeLa 229 cells were added to each well and the cells were incubated at 37 °C with 5% CO2 for 3 d . The total volume was 40 μl in 384-well and 400 μl in 24-well . In 24-well format the transfection mix was replaced by 500 μl of fresh media 24 h post transfection . The knock down of Tom40 or Tom22 was performed as described above in 24-well plate . Three days post transfection , the cells were harvested in 100 μl of protein sample buffer and 20 μl of cell lysates were run on SDS-PAGE gels and analyzed by western blot using HPR-conjugated secondary antibodies and Amersham ECL western blotting detection reagents . Images were acquired using the Metamorph software ( Molecular Devices ) . The integrated morphometry analysis module was used to quantify the size of C . caviae inclusions . In an attempt to use Drosophila as a model system to study Chlamydia pathogenesis , we investigated C . caviae replication in Drosophila SL2 cells . For this purpose , 80% confluent Drosophila SL2 cells cultured in 96-well dish were incubated with C . caviae . At various times post infection , the cells were transferred to Concanavalin A–coated coverslips ( Sigma , 2 mg/ml ) in Schneider media for 2 h . The samples were then fixed and stained with the DNA dye Hoescht and a FITC-conjugated antibody directed against Chlamydia to stain the inclusion . As shown in Figure 1 , C . caviae is able to infect and replicate in Drosophila SL2 cells . Although most of the cells contained at least one bacterium 1 h post infection , only 20% to 30% of the cells had an inclusion 48 h post infection ( not shown ) , suggesting that some bacteria were actually cleared in the phagocytic SL2 cells . However , when the bacteria were successful in establishing their niche , the infected cells displayed a perinuclear inclusion whose size increased between 24 and 72 h post infection . At 96 h post infection , the size of the inclusions was more heterogeneous ( not shown ) and some cells displayed disrupted inclusions , suggesting that the developmental cycle was completed and that reinfection was occurring between 72 and 96 h post infection . We next determined whether C . caviae were undergoing a full developmental cycle in Drosophila SL2 cells . To this end , we determined whether the different developmental forms of C . caviae were present in the inclusion by electron microscopy . As shown in Figure 2A , 45 h post infection the bacteria were found in a membrane-bound compartment that occupies most of the cytosolic space . The inclusions mainly contained RBs and intermediate bodies ( IBs ) in the process of differentiating to EBs and are characterized by their DNA condensation stage , but they also contained some bacteria with an EB morphology ( Figure 2B ) , suggesting that in Drosophila SL2 cells , RBs start to differentiate back to EBs 45 h post infection . In order to demonstrate that infectious progeny was produced , C . caviae harvested from Drosophila SL2 cells at different times post infection were used to infect HeLa cells ( Materials and Methods; Figure 2C ) . When C . caviae were harvested 3 h post SL2 infection , 10% of the HeLa cells displayed an inclusion . This number decreased to less than 5% when the bacteria were isolated 24 or 48 h post infection , suggesting that a substantial amount of bacteria were either cleared or had differentiated into non-infectious RBs . In contrast , 12 . 5% and 19% of the HeLa cells contained a large inclusion when the bacteria were harvested 72 and 96 h post SL2 infection , respectively . After 96 h , the number of infected HeLa cells remained constant . These results indicate that infectious forms of C . caviae are produced in Drosophila SL2 cells . Moreover , they are in agreement with the immunofluorescence ( Figure 1 ) and electron microscopy ( Figure 2A and 2B ) data and confirm that 48 h post infection of Drosophila SL2 cells , the inclusion mainly contains RBs and IBs , whereas EBs are produced in the next 24 h . Taken together , these data show that C . caviae undergo a full developmental cycle in Drosophila SL2 cells and suggest that the cycle lasts 72 to 96 h . The TTSS of C . caviae was functional in Drosophila SL2 cells as shown by determining the presence of the Inc family protein , IncA , on the C . caviae inclusion membrane ( Figure 3 ) . Drosophila SL2 cells were fixed 48 h post infection with C . caviae , stained with the DNA dye Hoescht to visualize the nuclei ( N ) and the inclusions ( Inc ) , and antibodies directed against IncA . A ring-like signal ( IncA , red ) that surrounded the inclusion ( Inc , blue ) was observed , indicating that , in Drosophila SL2 cells , the TTSS of C . caviae is functional and that TTS substrates such as IncA , are delivered to the inclusion membrane . Sixteen thousand Drosophila genes were individually knocked down by RNAi and screened for their ability to reduce C . caviae infection of Drosophila SL2 cells . The assay was performed as follows ( Materials and Methods; Figure 4A ) . After 3 . 5 d of RNAi treatment , the Drosophila SL2 cells were incubated with C . caviae for 48 h . The infected cells were fixed and stained with a DNA dye and a Chlamydia-specific FITC-conjugated antibody . An automated microscope was used to capture fluorescence images that were subsequently analyzed by visual inspection . The primary screen was performed in duplicate . We identified 162 candidates that , when depleted , reduced C . caviae infection ( Table S1 ) . Figure 4B is representative of the phenotype observed: few cells displayed wild-type size inclusion ( middle top panel ) and the number of infected cells , as well as the size of the inclusion , was largely reduced ( middle bottom panel ) . The candidates were grouped into 14 functional categories ( Figure 5 ) : miscellaneous ( 32 ) , unknown ( 32 ) , metabolism ( 18 ) , transcription ( 14 ) , vesicular trafficking ( 12 ) , cytoskeleton ( 9 ) , mitochondria ( 8 ) , transporter ( 8 ) , kinase and phosphatase ( 7 ) , chromatin organization ( 5 ) , endosome and lysosome ( 5 ) , protein biosynthesis ( 5 ) , RNA processing ( 4 ) , and cell cycle ( 3 ) . The dsRNA targeting most of the candidates of the miscellaneous , metabolism , vesicular trafficking , cytoskeleton , mitochondria , transporter , kinase and phosphatase , and endosome and lysosome categories were resynthesized to confirm the phenotype observed in the primary screen ( Table S1 ) . Out of the 100 candidates retested , the phenotype was confirmed for 54 candidates in at least two out of three replicates ( Table 1 ) . The validation rate varied among the categories: miscellaneous ( 40% ) , metabolism ( 47% ) , vesicular trafficking ( 75% ) , cytoskeleton ( 75% ) , mitochondria ( 67% ) , transporter ( 37% ) , kinase and phosphatase ( 57% ) , and endosome and lysosome ( 100% ) . In an attempt to assay for Chlamydia specificity , the knock down of the candidates was tested for the inhibition of L . monocytogenes infection ( Table 1 ) . The knock down of most vesicular trafficking ( 9/9 ) , cytoskeleton ( 4/6 ) , and endosome and lysosome ( 5/5 ) candidates inhibited both C . caviae and L . monocytogenes infection , and an equal number of kinase and phosphatase candidates inhibited C . caviae or L . monocytogenes infection . The knock down of most miscellaneous ( 11/13 ) and metabolism ( 6/8 ) candidates and of all mitochondria ( 6/6 ) and transporter ( 3/3 ) candidates inhibited C . caviae infection only . These results suggest that the latter categories are likely to represent candidates specifically involved in C . caviae infection . The RNAi screen in Drosophila cells revealed that the silencing of six mitochondrial genes inhibited C . caviae , but not L . monocytogenes infection . Moreover , four out of the six candidates were members of the mitochondrial membrane translocase , a multiprotein complex involved in the recognition and import of nuclear-encoded mitochondrial proteins to the mitochondria [43 , 45] . Taken together , these results suggested a specific role of this machinery for optimal C . caviae infection in Drosophila cells . To address the relevance of these findings in Chlamydia pathogenesis , this observation was further investigated in mammalian cells . Tom40 or Tom22 expression was knocked down in HeLa 229 cells using a mix of four siRNA duplexes directed against their respective mRNA ( ThermoFisher ) . In addition , each siRNA was tested individually to rule out any potential off-target effects . The depletion of either Tom40 or Tom22 was assayed 3 d post transfection of the siRNAs by western blot analysis . As shown in Figure 6A , both Tom40 and Tom22 were efficiently depleted after incubation with the mix of four siRNAs or with individual siRNA duplexes . The effect of Tom40 or Tom22 depletion on C . caviae infection was analyzed . HeLa 229 cells were incubated for 3 d with either CDH1 siRNA control directed against E-Cadherin , or Tom40 or Tom22 siRNAs pooled ( mix ) , or individually ( 1 , 2 , 3 , 4 ) , infected with C . caviae for 24 h , and processed for immunofluorescence . The corresponding low and high magnification images are depicted in Figure 6B and 6C , respectively . The nuclei were labeled with the DNA dye Hoeschst ( Figure 6B and 6C: left panel , DNA , blue ) and the inclusions were stained with a guinea pig polyclonal antibody against C . caviae ( Figure 6B and 6C: middle panels , C . caviae , green ) . Although the number of infected cells was similar , the inclusions appeared smaller upon Tom40 or Tom22 depletion ( compare CDH1 middle panels to Tom40 or Tom22 middle panels ) . A computer-assisted analysis of the images was used to quantify the size of the inclusions ( Materials and Methods ) . In the control situation , we determined that each inclusion could be defined as a 10- to 150-μm2 object and 40% of the inclusions were larger than 30 μm2 . We defined 10- to 30-μm2 and 30- to 150-μm2 objects as small and large inclusions , respectively . The impact of Tom40 or Tom22 knock down on C . caviae ability to form large inclusions was analyzed ( Figure 6D ) . A 5- to 3-fold reduction in the percentage of large inclusions was observed upon depletion of either Tom40 or Tom22 , confirming the overall reduction in the size of the inclusions and suggesting that upon Tom40 or Tom22 depletion , C . caviae intracellular growth is impaired . Electron microscopy analysis of C . caviae inclusions in control or Tom40 depleted cells confirmed the immunofluorescence results . Although a mixed population of small and large inclusions was observed 24 h post infection , the overall size of Tom40 depleted cell inclusion was smaller ( Figure 7 ) . In addition , RBs had already started to differentiate back into EBs in control cells , and 85% of the inclusions contained more than 25% EBs . In contrast , although some IBs were present , very few EBs were visible in Tom40-depleted cells , and only 25% of the inclusions contained more than 25% EBs . This result suggested that , in addition to a reduction in intracellular growth , differentiation back into EBs is also lessened in Tom40-depleted cells . The electron microscopy results suggested that RB differentiation into EBs was reduced upon Tom40 or Tom22 depletion . We therefore investigated the production of infectious progeny by Tom40- or Tom22-depleted cells . The cells were incubated with the siRNA in pool or individually for 3 d before incubation with C . caviae for 48 h to allow completion of the developmental cycle . The infected cells were collected , lysed with glass beads , and dilutions of the lysate were used to infect fresh HeLa 229 cells ( see Materials and Methods ) . The cells were fixed 24 h post infection and the number of inclusion forming units ( IFUs ) was determined after assessment of the number of infected cells by immonulabeling ( Figure 8A ) . We observed a 2- to 3-fold reduction in the production of infectious progeny upon Tom40 or Tom22 depletion . On the contrary , a similar number of infectious C . trachomatis were recovered from control or Tom40- or Tom22-depleted cells ( Figure 8B ) . These results demonstrate that , as suggested by the electron microscopy analysis , the reduction in the size of C . caviae inclusions is accompanied with a decrease in the number of infectious progeny produced . Altogether , our results indicate that depletion of members of the Tom complex in mammalian cells have a detrimental effect on C . caviae intracellular replication , which impairs bacterial replication and differentiation . Since Tom40 or Tom22 depletion had no effect on C . trachomatis infection , our results also indicate that the mechanism involved is C . caviae–specific . We demonstrated that , similar to the situation in mammalian cells [8] , infectious forms ( EB ) of C . caviae enter Drosophila SL2 cells , differentiate into the replicative form ( RB ) , replicate within a membrane-bound compartment , and differentiate back from RBs to EBs . A previous report showed that different serovars of C . trachomatis , including C . trachomatis LGV serovar L2 , could initiate their developmental cycle in Drosophila S2 cells [30] . However , the later stages of the developmental cycle were not achieved . Similarly , we found that when Drosophila SL2 cells were incubated with C . trachomatis LGV serovar L2 most cells were also infected 1 h post infection . However , the pattern of staining did not change over a 72-h period post infection and the cells never displayed large perinuclear inclusions ( not shown ) , confirming that C . trachomatis developmental cycle is not complete in Drosophila SL2 cells . We noticed a difference in the morphology of C . caviae inclusion in Drosophila cells compared to mammalian cells . The inclusions appear multilobed in mammalian cells [46] , whereas they appeared as a single membrane-bound compartment in Drosophila SL2 cells . This morphology resembles that of C . trachomatis inclusions that are known to undergo homotypic fusion and are therefore monovacuolar . This observation suggests that C . caviae inclusions may also undergo homotypic fusion in Drosophila SL2 cells . IncA , a type III secretion ( TSS ) substrate known to be in involved in the homotypic fusion of the C . trachomatis inclusions [47] , was present on the surface of C . caviae inclusion in Drosophila SL2 cells . Since IncA from C . caviae can interact with itself [48] , it is possible that in Drosophila SL2 cells it participates to the biogenesis of a single large inclusion . If it is the case , some Drosophila factors probably interact with IncA and help promote the fusion . However , one cannot exclude that in Drosophila cells the homotypic fusion of C . caviae inclusions is IncA independent . Using the Drosophila cell / C . caviae model system , we have performed an RNAi screen and identified 54 host factors that , when depleted , reduced C . caviae infection . By testing the effect of the candidates' knock down on L . monocytogenes infection , we have identified candidates presumably specific of C . caviae infection . In the following section , we discuss their potential relevance in Chlamydia pathogenesis . The attachment of most Chlamydia species to the host cell is dependent on host cell heparan sulfate glycosaminoglycans ( GAGs ) [15] . C . caviae is no exception , because its adhesion is GAG dependent and can be blocked by heparin [49] . Drosophila contains two main glypicans: Dally ( Division abnormally delayed ) [50] and Dlp ( Dally-like protein ) [51 , 52] . They are composed of cell-surface heparan sulfate proteoglycans linked to the plasma membrane by a glycosyl phosphatidylinositol linker . Our screen showed that the knock down of Dlp reduced C . caviae infection , suggesting that Dlp may promote the attachment of C . caviae to the cell surface . Activation of Rho family of GTPases and actin remodeling has also been implicated in Chlamydia entry [15] . For example , Cdc42 and actin polymerization are involved in C . caviae entry of mammalian cells [53] , and we show here that their depletion also reduced infection of Drosophila cells . Rac1 , which is also involved in C . caviae entry in mammalian cells [53] , was not identified in our screen . The Drosophila genome contains two rac genes , and it is possible that the single knock down of one or the other was not sufficient to block C . caviae entry . In addition , we also identified Ssh , a phosphatase that controls actin reorganization through the dephosphorylation of cofilin [54] . Ssh was not previously reported to play a role in Chlamydia pathogenesis , but our data suggest that it may be implicated in regulating actin dynamics upon entry of C . caviae . After their internalization , C . trachomatis EBs direct the nascent inclusion to a peri nuclear area . This movement is dependent of microtubules and the motor dynein [16] . We have identified two candidates that are linked to motors and microtubules . The first candidate , pav , encodes a kinesin-like protein [55] . Although kinesin is not involved in the trafficking of C . trachomatis inclusion to the peri nuclear area [16] , the microinjection of antibodies against kinesin prevents the recruitment of mitochondria to C . psittaci inclusions and delays the developmental cycle [56] . A defect in mitochondria recruitment to the inclusion may therefore explain the phenotype observed upon pav knock down in Drosophila cells . The potential importance of mitochondria in Chlamydia infection will be further discussed in the following section . The second candidate related to microtubule is katanin-60 . In mammalian cells , katanin concentrates at the centrosome of the cell , where the p60 subunit exerts its microtubule severing activity and induces the release of microtubules from the centrosome [57] . C . trachomatis inclusions associate with centrosomes [58] . It is possible that Chlamydia interacts with the centrosome and induces a katanin-mediated local destabilization of the microtubule network , thus allowing the expansion of the inclusion . A recent study revealed a dynamic interaction between multi-vesicular body–derived constituents and C . trachomatis inclusion [59] . We have identified two candidates involved in lysosomal transport ( CG11814 ) and organization ( CG5691 ) , as well as two subunits of the v-ATPase ( VhaAC39 and VhaPPA1–1 ) . The identification of such factors suggests that , at some point during the developmental cycle , Chlamydia inclusions may interact with compartments of the endocytic pathway . Further analysis of theses candidates may shed light on the mechanism involved . C . trachomatis inclusion also intercepts vesicles and lipids from the Golgi [18] and targets lipid droplets [20] . We have identified several enzymes involved in fatty acid synthesis , desaturation , elongation , and oxidation . The identification of such enzymes reinforces the idea that the acquisition of lipids is an important aspect of Chlamydia intracellular replication , and further investigation may shed light on the host metabolism pathways targeted by Chlamydia . Our screen revealed that the knock down of members of the Tim-Tom complex , the multiprotein complex involved in the recognition and import of nuclear-encoded mitochondrial proteins to the mitochondria [45 , 60] , inhibited C . caviae infection in Drosophila cells . Importantly , we have shown that the knock down of two major components of the outer membrane complex of mitochondria , Tom40 and Tom22 , also inhibited C . caviae infection in mammalian cells . In the following section we discuss potential mechanisms that may explain the phenotype observed . The National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov/ ) accession numbers for the mammalian genes are CDH1 ( NM_004360 ) , Tom40 ( NM_006114 ) , and Tom22 ( NM_020243 ) .	Chlamydia spp . are intracellular bacterial pathogens that infect a wide range of hosts and cause various diseases , including preventable blindness in developing countries , sexually transmitted disease , and pneumonia . Chlamydia spp . are able to establish their replication niche inside the host cell , residing in a membrane-bound compartment that serves as a protector shield against immune surveillance and antimicrobial agents but also acts as a “filter” to exchange factors with the host cell . Despite the primary importance of Chlamydia for human health , little is known about the mechanisms underlying the infection process . The study of Chlamydia pathogenesis is challenging because Chlamydia spp . are not amenable to genetic manipulation and it is difficult to conduct extensive genetic approaches in the mammalian host . To circumvent these difficulties , we have used Drosophila cells to model Chlamydia infection . We conducted a genome-wide RNA interference screen and identified host factors that , when depleted , reduce Chlamydia infection . Validating our approach , we further showed that the identified factors were also required for infection in mammalian cells . This work will help us better understand the complex interaction between Chlamydia and its host and potentially identify novel targets for therapeutic treatment .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion", "Supporting", "Information" ]	[ "infectious", "diseases", "cell", "biology", "mammals", "homo", "(human)", "microbiology", "drosophila", "eubacteria" ]	2007	RNAi Screen in Drosophila Cells Reveals the Involvement of the Tom Complex in Chlamydia Infection
The World Health Organization aims for complete morbidity control of fishborne zoonotic trematodes ( FZT ) in endemic areas by 2020 . The main intervention tool for achieving this goal is regular use of preventive chemotherapy by offering praziquantel to those at risk in endemic areas . The purpose of this study was to investigate the effectiveness of preventive chemotherapy to control FZT in an endemic area in Northern Vietnam . We followed a cohort of 396 people who fulfilled the criteria for receiving preventive chemotherapy . Stool samples were examined by Kato-Katz technique for the presence of trematode eggs before , and two , 16 , 29 and 60 weeks after preventive chemotherapy . The prevalence of trematode eggs in stool was 40 . 2% before , 2 . 3% two weeks after and increased to a cumulative prevalence of 29 . 8% sixty weeks after preventive chemotherapy . The effectiveness of preventive chemotherapy as a main component in control of FZT is not well documented in most endemic areas . We found a high reinfection rate within the first year after preventive chemotherapy . Since these trematodes are zoonoses , preventive chemotherapy may not have sufficient impact alone on the transmission to have a lasting effect on the prevalence . Animal reservoirs and farm management practices must be targeted to achieve sustainable control of fishborne zoonotic trematode infections , hence control programs should consider a One Health approach . Foodborne trematodiasis is a group of Neglected Tropical Diseases with a considerable impact on human health , especially in Southeast and East Asia . An estimated 56 million people are infected with foodborne trematodes , of which approximately half is due to fishborne zoonotic trematodes ( FZT ) [1] . FZT infects animals and humans after consumption of raw or insufficiently cooked freshwater fish , in the case of humans often as a traditional dish [2] . Infections with FZT are believed to be emerging , and the rapid growth in aquaculture and consumption of fish raised in endemic areas has been proposed to contribute to infection [3] . FZT include the liver flukes Clonorchis sinensis and Opisthorchis viverrini as well as a large number of minute intestinal fluke ( MIF ) species . The prevalence of FZT often shows great regional differences within a particular country . There is also variation in predominance of species within specific geographical areas [4] , [5] . The morphological characteristics of eggs belonging to different species are difficult to differentiate by microscopy , and this may lead to unreliable data of their epidemiological distribution [6] . In Northern Vietnam , there is a mix of C . sinensis and MIF , especially Haplorchis spp . , in humans [7] , [8] . Studies from the same area have revealed high prevalence of different MIF-species , especially Haplorchis spp . , in dogs , cats and pigs , suggesting that domestic animals could play an important role as reservoir hosts and in the transmission of FZT [9] , [10] . The adult liver flukes enter the intrahepatic bile ducts , while the adult MIF remain in the small intestine . The most serious health hazard associated with liver fluke infection is the increased risk of cholangiocarcinoma . The MIF infections have been less studied , but the clinical manifestations may be less serious and MIF are not considered carcinogenic [5] . The morbidity of the liver trematodes , especially the predisposition for cholangiocarcinoma , is the main reason for establishing control measures [11] . WHO recommends preventive chemotherapy , that is large-scale distribution of anthelminthic drugs to populations at risk , as the main intervention strategy against liver flukes [12] , [13] . If the prevalence in a district is >20% , universal treatment of all individuals in the district is recommended once a year . If the prevalence is <20% , the recommendation is either a universal treatment every second year , or targeted treatment once a year of all those who report habitually eating raw fish [12] . A similar intervention strategy is used in the well-established programs to control other helminth infections like schistosomiasis , geo-helminth infections , onchocerciasis and filariasis [14] . The effectiveness of preventive chemotherapy on C . sinensis has been examined in Chinese studies [15] . Choi et al . published a large study with 14139 people , evaluating different regimes of universal treatment or selective treatment over a three-year period . The different regimes gave a reduction in prevalence of C . sinensis between 34 . 2–93 . 9% after one year following treatment and 72 . 7–95 . 6% after three years [16] . However , there are few studies from other endemic areas evaluating the effectiveness of preventive chemotherapy , and the information originates mostly from studies done more than 20 years ago of which several were conducted within the same study population [17]–[22] . A study from 1988 from a village in Thailand with a O . viverrini prevalence of around 90% showed that 88 . 4% of the people was egg negative two weeks after praziquantel treatment ( n = 704 ) . However , 87 . 7% of those negative after two weeks became reinfected within a year [23] . Nissen et al . found that half of the North Vietnamese farm dogs that had been given a single dose of praziquantel were reinfected with FZT within three months [24] . Hence they could not recommend anthelmintic treatment as the only intervention to control FZT in dogs . The use of preventive chemotherapy as an intervention to control FZT in Vietnam started in 2006 and has gradually increased since then . In 2011 such treatment was given to 128837 people , amounting to 532053 tablets of praziquantel ( NIMPE , annual report 2011 ) . Preventive chemotherapy is here given as selective chemotherapy to people confirming that they eat raw fish . Despite the large number of treatments , the effectiveness of preventive chemotherapy in Vietnam has not been thoroughly assessed . A pilot study from Nam Dinh province involving 21 FZT egg-positive people showed a prevalence of 29% and 71% six days and five weeks after having received praziquantel treatment , respectively [25] . The authors refer the infections as C . sinensis , but the species distribution in that province makes it more likely to be a mix of C . sinensis and MIF [8] . In the present study , the rate of reinfection with FZT was assessed after applying preventive chemotherapy for the first time in a cohort of people confirmed having eating raw fish in two communes in Northern Vietnam . Informed , written consent was obtained from participants aged 16 years or older . In accordance with Vietnamese regulations , guardians signed for minors under the age of 16 years . The study protocol was approved by the Ethics Committee at the National Institute of Malariology , Parasitology and Entomology ( NIMPE ) , Hanoi . The present study was part of the regular preventive chemotherapy program for FZT infections in the area . Hence , refusing to participate in the study did not prevent people from receiving preventive chemotherapy . Information about the disease and treatment was given initially according to standard praxis for the control program . Appropriate treatment was offered if other helminth eggs were found by microscopy . The practice of the present study implied intensified treatment compared with the normal setting for preventive therapy program , in which there is no testing after preventive chemotherapy and therefore no further treatment . The study was conducted between March 2011 and June 2012 in Nghia Hong commune , Nghia Hung district and Hai Hoa commune , Hai Hau district , both located in Nam Dinh province , in Northern Vietnam . The communes have approximately 7800 and 8700 inhabitants , respectively . Neither of the two communes had been part of the regional program for preventive chemotherapy prior to the study , and neither of the communes had been part of surveys in which people could have received anthelmintic treatment for the last three years prior to the study . Before the study , both communes were assessed by NIMPE , Hanoi , and found to be eligible for the regional program for preventive chemotherapy for FZT infections . Potential participants were approached following the standard procedure of this program . A village health worker visited the households and asked who had ever eaten raw fish . People confirming eating raw fish were invited to participate in preventive chemotherapy if they were not pregnant or suffering from liver or kidney disease , serious hypertension or acute stomach symptoms . Those selected received 50 mg/kg praziquantel , split in two doses on the same day . A randomly selected subgroup of those eligible for preventive chemotherapy were invited into the study and asked to deliver a stool sample three weeks prior to and two , 16 , 29 and 60 weeks after the preventive chemotherapy was given . Each stool sample was examined with Kato-Katz technique , using two slides for each sample [26] . The number of eggs per gram stool ( epg ) was recorded . A total of 539 people delivered a stool sample prior to preventive chemotherapy . From these , 75 people did not receive preventive chemotherapy and 68 failed to deliver any stool samples after preventive chemotherapy , hence , they were excluded from the follow up study . Overall , 396 people ( 73 . 5% ) had sufficient data to be included in the final analysis . Those found positive by microscopy after preventive chemotherapy received a treatment with praziquantel and were not examined further , but regarded as positive for the remaining of the study . Hence the reported figures for reinfection are cumulative . Missing stool samples after preventive chemotherapy were handled with imputation , i . e . a missing stool sample at time 16 weeks , 29 weeks and/or 60 weeks were regarded as negative if the previous test in the timeline was negative . The study design and the absolute figures of test results are presented in Figure 1 . Descriptive measures of prevalence and intensity of infection were performed with IBM SPSS Statistics ( Version 20 . 0 . Armonk , NY: IBM Corp ) . Intensity of infection is presented by median value , not mean , due to kurtosis of distribution . Confidence intervals around medians were created using bootstrapping method . The cumulative prevalence of infection is presented in a bar diagram using Microsoft Excel 2007 ( Redmond , Washington ) . The findings in this simple presentation were confirmed in an repeated measure analysis of variance ( SPSS ) . Confidence intervals around the prevalence were calculated in SPSS using Wilson's method without continuity correction . Multiple logistic regression ( SPSS ) were used for analysis of risk factors for reinfection . Reinfection among those negative and those positive before treatment is presented in a bar diagram using MicrosoftExcel 2007 . Study subjects included 242 men ( 61% ) with a median age of 50 years ( range 8–75 ) and 154 women ( 39% ) with a median age of 47 years ( range 15–75 ) . As seen in Figure 2 , the prevalence of small trematode eggs in stool before preventive chemotherapy for the two sites combined were 40 . 2% ( 95% confidence interval ( CI ) of 35 . 3–45 . 0 ) . The prevalence two , 16 , 29 and 60 weeks after preventive chemotherapy were 2 . 3% ( CI 0 . 8–3 . 7 ) , 10 . 9% ( CI 7 . 8–13 . 9 ) , 18 . 2% ( CI 14 . 4–22 . 0 ) and 29 . 8% ( CI 25 . 3–34 . 3 ) , respectively . The increase in prevalence after preventive chemotherapy is significant between all time points , as seen by the repeated measure ANOVA analysis ( table 1 ) . The FZT prevalence among all the 539 who delivered stool sample before preventive chemotherapy was 42 . 9% ( CI 38 . 7–47 . 0 ) , not different from the group of 396 constituting the cohort followed up . Results from the two study sites differed when analyzed separately ( Fig . 2 ) . Nghia Hong commune had a more rapid increase in FZT prevalence after chemotherapy . The prevalence after 60 weeks in Nghia Hong commune was the same as before preventive chemotherapy . In Hai Hoa commune , the FZT prevalence at the end of the study was less than half of that before preventive chemotherapy , and significantly lower than in Nghia Hong commune , as seen by the confidence intervals . The overall infection intensity , measured as median egg per gram stool , was low throughout the study ( Table 2 ) . However , the range between the highest and lowest epg was high , even as soon as 16 weeks after preventive chemotherapy . It is noteworthy that 76/118 ( 65% ) of those people egg-positive after 60 weeks were also positive before treatment , suggesting a relative high degree of reinfection in these people . When analyzing gender , age and previous infection as risk factors for reinfection at 60 weeks of follow up , we see in the single logistic regression model that significantly more men than women become reinfected . Increasing age is also associated with reinfection , as is previous infection . However , in the multiple logistic regression model only previous infection remains associated with reinfection , indicating confounding . Before treatment , a significantly higher proportion of the participating men than the participating women were infected ( 53% versus 20% , p<0 . 001 ) . A similar pattern was found 60 weeks after treatment , with 36% of the men and 20% of the women being infected ( p<0 . 001 ) . In the present study , we found a high reinfection rate of FZT within one year following treatment with praziquantel . The true reinfection rate might have been even higher as we applied a conservative analytical approach , categorizing a person as non-infected in case of non-participation during follow up . Also , as infected persons were treated during follow up , their contribution to FZT transmission was reduced . Our estimate of reinfection is thus conservative . We found a difference in degree of reinfection between the two communes . Whether this is due to a difference in the frequency of eating raw fish , intensity of FZT metacercariae in fish or due to other reason is not known . Figure 3 shows an increased rate of reinfection among those positive before treatment . This is confirmed in the multiple logistic regression model ( Table 3 ) . In this model we found that previous infection ( representing risk behavior ) was a significant risk factor for reinfection , whereas male gender and increasing age were not . This indicates that the seemingly close association between male gender , increasing age and reinfection we observed ( crude numbers ) , is possibly confounded by risk behavior of eating raw fish , a tradition mostly common among men [27] . FZT infections in Northern Vietnam are a mix of C . sinensis and MIF . C . sinensis seems to be much less common in fish than in humans . However , only very few studies have investigated the species distribution of FZT in humans . Phan et al . [28] examined 1543 fish from Nam Dinh province , Northern Vietnam , of which more than half were infected with FZT metacercariae . Only a single fish contained C . sinensis metacercariae . In humans , Trung et al . [8] identified the species of adult worms expulsed from 615 egg-positive people from Nam Dinh province . All the people had MIF , and 51 . 5% had in addition C . sinensis . We do not yet have an explanation for the seemingly different species distribution in fish and humans . One possible explanation is that it is a relatively rare event for a human to acquire an infection with C . sinensis compared to a MIF infection , but once acquired , C . sinensis infections are long lived , 26 years has been described [29] , [30] . Annual preventive chemotherapy may then decrease the proportion of C . sinensis infections in humans , as by far most of the reinfection will be due to MIF . Since C . sinensis probably have a higher morbidity and mortality than MIF , such a shift in species distribution may be an argument for preventive chemotherapy even if the reinfection rate is high . However , we do not know how preventive chemotherapy influences the species distribution , and such a shift has yet to be shown . Some aspects in the design of the Vietnamese control program may become a challenge . In Vietnam , the preventive chemotherapeutic dose of 50 mg/kg is split into two doses to minimize side-effects . This makes it impossible to have direct observed therapy . In the present study , the FZT prevalence dropped from 40 . 2% before preventive chemotherapy to 2 . 3% two weeks after . This drop is a combination of praziquantel efficacy and compliance . Hence , the compliance in our study was excellent , but compliance may become a problem especially if preventive chemotherapy is continued for several years and without direct observed therapy [31] . Vietnam has also found it cost-effective to perform preventive chemotherapy as selective chemotherapy by only treating those who confirm that they have ever eaten raw fish [32] , [33] . However , Phan et al . [27] found in a study involving 180 persons that 38 . 5% of those confirming eating raw fish were egg positive on microscopy , but also 25 . 8% of those who denied eating raw fish . A possible explanation for this may be that especially women often have the perception that they do not eat raw fish , but during food preparation they taste the raw fish dishes which is a significant risk [34] . Additionally , cross-contamination might occur because bowls with sauces are typically shared at meals , i . e . by people dipping pieces of raw fish and those dipping raw vegetables only . If this is to be confirmed , the current strategy on how to identify people to receive treatment may have to change . WHO promotes the use of preventive chemotherapy to control FZT . This is in accordance with the control strategy for several other helminths , like schistosomiasis , geo-helminth infections , onchocerciasis and filariasis . Even though preventive chemotherapy has shown to be efficient in reducing the prevalence and infection intensity of these infections , rapid reinfection after treatment is also well documented [35] . However , in FZT the the scarcity of studies provides little scientific support for this strategy . None of the mentioned helminth infections , with the exception of Schistosoma japonicum , are zoonoses , and hence have a quite different transmission than FZT . In the FZT lifecycle , both humans and domestic as well as wild animals pass eggs to the environment . The presence of infected snails and introduction of infected fish into the freshwater environment are also important in the transmission of FZT [36]–[38] . It is difficult to evaluate which factor makes the largest contribution to the transmission , and it is likely that this may vary according to FZT species , geographical area and type of aquatic environment . Further , it should be noted we do not know to what extend people infected with FZT obtained such infection from the consumption of raw dishes of fish originating from aquaculture or wild-caught . Nguyen et al . [9] , [10] found that dogs , cats and pigs are important sources of FZT eggs , in particular minute intestinal fluke eggs , to the environment in two endemic areas in Northern Vietnam . Cats and dogs deposit their stool freely , while pig manure is used as fertilizer for fish ponds and rice fields . As humans are only one of several reservoirs that contribute to the transmission , decreasing the egg output through preventive chemotherapy of humans may not have a lasting effect on reducing transmission and risk of FZT infection . We focused on reinfection in the first year after preventive chemotherapy . If repeated preventive chemotherapy decreases the prevalence of infections , it may have an effect on morbidity . However , if the human contribution to the lifecycle of FZT is only minor , transmission is not sufficiently affected , and it may be difficult to terminate the use of preventive chemotherapy after a limited number of years . Control of FZT , being a zoonosis , may in the long run benefit from an increased focus on a more integrated approach , as experienced in Thailand [4] . An integrated approach to reduce snail population and egg contamination has also been studied in Vietnamese fish nurseries by Clausen et al . [37] . Two intervention groups were studied; one farm management group with control of snail vectors and fecal pollution of pond , and one group with drug treatment of humans and animals . It was found that the intensity of metacercariae in fish was reduced with 91 . 7% in the farm management group . Even though veterinary public health has been mentioned by WHO as an intervention to overcome the Neglected Tropical Diseases , preventive chemotherapy has so far been the dominating pillar in the attempt to control FZT [12] , [14] . ‘One Health’ is a concept focusing on integration between human and veterinary medicine . This may be a useful concept if interventions such as discouraging the use of raw fish for animal feed , discouraging the use of untreated pig manure as fertilizer in ponds and regular anthelmintic treatment of roaming domestic animals are to be included in the control of FZT [10] . FZT has , apart from being a zoonosis , many similarities with other Neglected Tropical Diseases which coexist in the same population . A recent policy paper argues for multi-disease , multi-sectoral synergistic control interventions for helminth infections in the Western Pacific region [39] . In conclusion , we question if preventive chemotherapy alone is sufficient to control the FZT and suggest that a more integrated approach including One Health and improved aquaculture farm management should be explored .	Fishborne zoonotic trematodes ( FZT ) are small parasitic flatworms ( flukes ) living either in the intrahepatic bile ducts or in the intestines of humans and many species of animals and birds . Transmission occurs by eating infected raw freshwater fish . FZT are especially common in Southeast and East Asia . WHO suggests using preventive chemotherapy in the control of these parasites by providing an effective drug , praziquantel , once a year to either all inhabitants in the endemic areas or selectively to those who are at special risk because they eat raw fish . Preventive chemotherapy has been used for some years in Northern Vietnam , but the effectiveness has not been thoroughly examined . We followed a group of 396 people who had eaten raw fish and examined the stool for the presence of trematode eggs before preventive chemotherapy and two , 16 , 29 and 60 weeks after . The proportions of egg positive persons were 40 . 2% before , 2 . 3% two weeks after and increased to 29 . 8% sixty weeks after preventive chemotherapy . We found a high rate of reinfection . Contribution from animals to the transmission implies that preventive chemotherapy of humans alone will most likely be insufficient to control the FZT infections .	[ "Abstract", "Introduction", "Methods", "Results", "Discussion" ]	[ "medicine", "and", "health", "sciences", "infectious", "disease", "epidemiology", "tropical", "diseases", "parasitic", "diseases", "parasitology", "foodborne", "trematodiases", "gastroenterology", "and", "hepatology", "global", "health", "neglected", "tropical", "diseases", "infectious", "disease", "control", "veterinary", "science", "public", "and", "occupational", "health", "infectious", "diseases", "veterinary", "diseases", "zoonoses", "veterinary", "parasitology", "epidemiology", "clonorchiasis", "helminth", "infections", "diagnostic", "medicine", "biology", "and", "life", "sciences" ]	2014	High Reinfection Rate after Preventive Chemotherapy for Fishborne Zoonotic Trematodes in Vietnam
Fundamental properties of phasic firing neurons are usually characterized in a noise-free condition . In the absence of noise , phasic neurons exhibit Class 3 excitability , which is a lack of repetitive firing to steady current injections . For time-varying inputs , phasic neurons are band-pass filters or slope detectors , because they do not respond to inputs containing exclusively low frequencies or shallow slopes . However , we show that in noisy conditions , response properties of phasic neuron models are distinctly altered . Noise enables a phasic model to encode low-frequency inputs that are outside of the response range of the associated deterministic model . Interestingly , this seemingly stochastic-resonance ( SR ) like effect differs significantly from the classical SR behavior of spiking systems in both the signal-to-noise ratio and the temporal response pattern . Instead of being most sensitive to the peak of a subthreshold signal , as is typical in a classical SR system , phasic models are most sensitive to the signal's rising and falling phases where the slopes are steep . This finding is consistent with the fact that there is not an absolute input threshold in terms of amplitude; rather , a response threshold is more properly defined as a stimulus slope/frequency . We call the encoding of low-frequency signals with noise by phasic models a slope-based SR , because noise can lower or diminish the slope threshold for ramp stimuli . We demonstrate here similar behaviors in three mechanistic models with Class 3 excitability in the presence of slow-varying noise and we suggest that the slope-based SR is a fundamental behavior associated with general phasic properties rather than with a particular biological mechanism . Stochastic resonance ( SR ) has been extensively described in both bi-stable and excitable systems and is a classic example of noise enhanced processing [1]–[9] . Briefly , SR involves noise facilitating dynamic state transitions or threshold crossing , while permitting phase-locked response to a subthreshold signal . The interaction of signal , noise , and response nonlinearity maximizes signal encoding at a nonzero value of noise intensity . Here , we characterize the novel manner in which SR-like phenomena occur in phasic neuron models . Phasic neurons are characterized by the absence of repetitive firing to steady current injection and low-frequency input , yet show faithful responses to brief pulsatile and high-frequency signals [10]–[13] . In a classical SR system , often exemplified by non-phasic neurons , a signal can be detected without noise simply by making its amplitude adequately large . In contrast , deterministic phasic neurons will not respond to low-frequency input even if the signal amplitude is very large , making phasic neurons an ideal framework to study noise-gated coding [14] . We convey our insights by presenting detailed results for a phasic model [15] that has been widely used in modeling various auditory brainstem phasic neurons [16]–[18] that perform precise temporal processing and respond only to rapid transients and coincidences . We then examine other types of phasic models , showing that our findings are general . The Class 3 excitability , which is commonly used to define phasic responses [12] , can be created by different cellular mechanisms , such as recruiting a low-threshold potassium current ( IKLT ) [19]–[22] , inactivating the sodium current ( INa ) [12] , [23]–[24] , or steepening the activation of the high-threshold fast potassium current ( IKHT ) [25] . The phasic neuron model that is our primary focus here is a Hodgkin-Huxley ( HH ) type model with IKLT [15] . Combining the same phasic neuron model and whole-cell recordings in the medial superior olive ( MSO ) in gerbil , we have previously shown that adding noise enables phasic neurons to detect low-frequency inputs , which , alone , cause no spiking response [14] . Although this behavior seems to be consistent with SR , it is fundamentally different from SR for the reasons listed below . In a classical SR system , adding a small amount of noise to a subthreshold signal facilitates threshold crossing , such as a spike emission upon crossing a membrane voltage ( Vm ) threshold . When the intensity of the noise is properly chosen , the signal can be encoded by eliciting more spikes around the signal's peak and fewer spikes around its trough . The larger the amplitude of the subthreshold signal , the better the noise-gated encoding becomes; whereas , for supra-threshold signals noise will only degrade signal encoding . In this sense , we call the classical SR system an “amplitude-based stochastic resonance” . However , we discovered that phasic MSO neurons and a phasic neuron model [15] responded to the rising , falling , or trough phases , depending on the spectrum of the noise , but not to the signal's peak except for very large noise [14] . Here , we report that an essential feature of phasic neurons is that response “threshold” is better defined in terms of the slope rather than the amplitude of the input . We further show that the noise-gated signal encoding is sensitive to the slope of the signal , as opposed to its amplitude . For this reason , we label SR-like phenomena in phasic neurons as a “slope-based stochastic resonance” . In this study , we highlight the novelty of a phasic neuron's slope-based SR behavior by contrasting it with the qualitatively distinct amplitude-based SR and coding properties of tonic neurons . To this end , we first compare the dependence of the signal-to-noise ratio ( SNR ) on noise intensity obtained from a tonic model to that from a phasic model . In addition to analyzing SNRs , we pay attention to the temporal firing patterns , which are often overlooked when SR systems are concerned . Next , we show that the slope-based SR behavior of the phasic model can be reflected in a highly non-monotonic f-I ( firing rate vs . stimulus mean ) relation with the compelling feature that firing rate falls continuously to zero with increasing I . Such f-I curves have been reported for phasic neurons and models [24] , [26] , [27] . Finally , the slope threshold in response to a ramp stimulus , as is observed in noise-free conditions [28] , is reduced or diminished by the addition of noise . In total , we report that the influence of noise and any noise-assisted coding performed by phasic neurons is significantly distinct from that of tonic neurons . The occurrence of Class 3 excitability is often associated with outward currents , i . e . , IKLT or IKHT [25] , [29] . It is far less realized that strong inactivation of INa can also create Class 3 excitability [23] . We show that the slope-based SR behavior is observed with phasic models created by manipulating the voltage dependency of either IKHT or INa when the noise spectrum favors low frequencies . Our present study , complemented by our previous experimental results [14] , reveals that phasic neurons can have substantially different behaviors in noisy conditions compared to their behaviors in non-noisy conditions . The conventional views of phasic neurons being band-pass filters or slope detectors , which are all acquired in idealized conditions with no noise present , should be re-evaluated in noisy conditions . The response or bifurcation diagram of the tonic model shows repetitive firing over a range of steady current input , IDC ( Fig . 1A , left , green ) . An example time course ( IDC = 0 . 6 nA ) is plotted as an inset . In contrast , the phasic model shows typical Class 3 excitability ( Fig . 1A , right ) by having a unique stable steady state for all IDC . Note that the “phasicness” of the phasic model is relatively strong [14] so that no repetitive firing is observed even for large steady input current , unlike in some previous studies [29]–[31] . The phasic and tonic models also show different firing preferences for sinusoidal input with varying frequency and amplitude ( Fig . 1B and C; replotted from Fig . 1 in [14] ) . For the tonic model , the input threshold ( the lowest amplitude of a sinusoid that causes firing ) remains relatively constant for low frequencies . In contrast , the input threshold of the phasic model rises sharply on the low-frequency side . For this reason , phasic neurons are commonly viewed as band-pass filters , and consequently it is difficult to define a universal input threshold in terms of input amplitude . The threshold rise is not completely amplitude independent because 1 ) increasing the amplitude of a sinusoid steepens the zero-crossing slope , and 2 ) increasing the amplitude of a sinusoid is similar to decreasing the pre-ramp holding current of a ramp stimulus , which leads to decreased slope threshold . Nevertheless , for phasic neurons it is more natural to define the threshold in terms of an input slope/frequency . Another distinction between tonic and phasic firing is the spiking ratio ( the number of spikes per stimulus cycle ) for low-frequency sinusoidal inputs ( Fig . 1C ) . The tonic model fires more than one spike for low-frequency inputs ( left ) , whereas the phasic model fires only one spike in each cycle for most of the input conditions ( right ) . Representative time courses are plotted in Fig . 1B . Therefore , even if the phasic model responds to low-frequency inputs with extremely large input amplitude , the firing rate is low ( e . g . , 20 spikes/sec for a 20-Hz sinusoid with 4-nA peak amplitude ) . Based on this feature , later we show that the intensity of noise that optimizes signal encoding is different for the tonic and phasic models . Phasic neurons are often called slope detectors because they respond to fast-rising , but not to slow-rising , ramps [28] . Fig . 2 shows the Vm of the phasic model ( right ) in response to ramp current with different slopes ( left ) . The ramp elicits an action potential only when its slope ( dI/dt ) exceeds 0 . 55 nA/ms . In contrast , the tonic model fires action potentials to ramps with any slope , as long as the input amplitude reaches 0 . 3 nA ( not shown ) . Average firing rate and SNR are presented first to provide a general measure of the model behaviors , followed by detailed frequency and temporal responses . Fig . 3A shows that the average firing rate of both models increased monotonically with noise intensity ( σ ) . When the signal amplitude increased from 0 . 1 to 0 . 2 nA , the firing rate of the tonic model remained constant except at very low noise intensities . In contrast , when the amplitude of the signal increased from 1 to 2 nA , the firing rate of the phasic model decreased substantially . The relationship between firing rate and signal amplitude will be explored more thoroughly later . Fig . 3B shows the SNR obtained with the larger ( black ) and smaller ( gray ) signal amplitudes for both models . The SNR of the tonic model resembled the SNR of a classical SR system , showing an abrupt rise before a peak and a gradual decay after the peak [4] , [32] . In addition , the peaks of the SNRs for both signal amplitudes were obtained at the same noise intensity ( σ = 5 pA ) , consistent with an asymptotic theory of SR for weak signals ( see Equation 4 ) . The red dotted line is a fit to the SNR for the smaller signal amplitude ( black ) using Equation 4 . Although the SNR decayed faster than the fit , they had essentially the same shape . In contrast , the SNRs obtained with the phasic model did not show classical SR-like behavior ( Fig . 3B , right ) . For both signal amplitudes , the SNRs did not decrease significantly at high noise intensities . Presumably , the SNR will drop for very large noise intensity , however the membrane fluctuation caused by the largest noise intensity used here ( σ = 30 pA ) already reached 30 mV for the phasic model ( Fig . 3A , right , top horizontal axis ) . For the larger signal amplitude ( black ) , a distinct dip occurred in the SNR around σ = 17 pA ( marked with c ) , which yielded a SNR even lower than the SNR obtained with the smaller signal amplitude ( gray ) . Thus it is impossible to fit the SNR of the phasic model using Equation 4 . To understand why the SNRs of the phasic model had such unusual shapes , later we show more detailed frequency and temporal response patterns for the larger signal amplitude . Responses at several representative noise intensities ( marked in the SNR plots ) were chosen for demonstration . Here the SNR reflects at the system's output the signal power with respect to the noise power . Another frequently used metric in studies of SR is the spectral power amplification ( SPA ) [9]: the peak power at the signal's fundamental frequency normalized by the total signal power ( see Methods ) , a measure of gain of the subthreshold signal . For the tonic model , the SPA behaved similar to the SNR in that an optimal value of noise intensity can be identified to yield the highest signal gain ( Fig . 3C , left ) . Moreover , the amplification of the signal was approximately constant with respect to signal amplitude , as increasing the signal amplitude from 0 . 1 ( gray ) to 0 . 2 nA ( black ) did not noticeably change the gain . In contrast , the SPA for the phasic model kept increasing for a fixed signal amplitude up to the highest noise level tested ( Fig . 3C , right ) . In other words , with increasing noise intensity , the signal's gain was also increasing , leaving a relatively flat SNR ( Fig . 3B , right ) ; there was no optimal noise intensity to achieve the highest signal's gain . Another striking feature was that , as the signal amplitude increased from 1 ( gray ) to 2 nA ( black ) , the SPA as a measure of the signal's gain decreased significantly ( Fig . 3C , right ) . As will be described below , this was because for the larger signal amplitude , more output power was shifted from the signal's fundamental frequency to the first harmonic . Finally , it should be noted that because the phasic model did not favor low-frequency signals , the SPA as a measure of the signal's gain was considerably smaller than the SPA for the tonic model ( Fig . 3C ) . For the tonic model , the power-spectrum density ( PSD ) plots agreed with PSDs from classical SR systems with weak signals [4] in that a large peak occurred at the signal frequency ( 20 Hz ) with smaller peaks at the harmonics for low-intensity noise ( Fig . 4 , left , 1st column ) . The period histograms showed highest sensitivity to the signal's peak at all noise intensities with more uniformly distributed spikes occurring at high noise intensities ( Fig . 4 , left , 2nd column ) . The interspike-interval ( ISI ) histograms indicated that missing signal cycles only occurred on the rising phase of the SNR curve ( a ) ; once the SNR reached its peak ( b ) , there were always one or more spikes in each cycle ( b and d ) ( Fig . 4 , left , 3rd column ) . All of these behaviors were consistent with what are expected for neurons exhibiting classical SR behaviors [1] . In contrast , the phasic model was mostly sensitive to the signal's rising phase , indicated by the period histograms ( Fig . 4 , right , 2nd column ) . After SNR reached its peak ( b ) , the phasic model also responded to the signal's falling phase with a lower firing probability compared to the responses in the rising phase ( c and d ) . This preference for two distinct phases explained why the power at the first harmonic can be larger than the power at the fundamental frequency for certain noise intensities ( Fig . 4 , right , 1st column , c ) . This also explained why there was a peak at half signal cycle in the ISI histogram for certain noise intensities ( b and c ) . In addition , for very high noise intensities ( d ) , the phasic model still showed no response in the signal's trough , which is consistent with high SNR persisting at high noise intensities ( Fig . 3B , right ) . Note that these distinct features were observed with the phasic model for low-frequency signals . As the signal frequency increased , the SNR behaved more similar to the SNR of a classical SR system , and responses occurred around only one phase of each signal cycle ( e . g . , for 100 Hz , not shown ) . The above descriptions for the tonic model were generally independent of frequency . To make a more detailed comparison between the signal coding at the fundamental frequency and the first harmonic , the SNRs computed at the first harmonic are plotted in Fig . 5 . For the tonic model ( left ) , the SNRs at the first harmonic ( solid ) were always lower than the SNRs at the fundamental frequency ( dotted ) . This was also the case for the phasic model with the lower signal amplitude , except around the peak of the SNR ( Fig . 5 , right , gray ) . With the higher signal amplitude , there was a range of noise level ( ∼15 to 25 pA ) that yielded a higher response power value at the first harmonic than at the fundamental frequency ( Fig . 5 , right , black ) . In summary , the impact of noise on the encoding of low-frequency signals was different between the phasic and tonic models . In classical SR studies SNR is normally measured at the fundamental frequency of the signal and therefore does not capture how stimuli shape the temporal pattern of responses in other frequency bands . In particular , for large-amplitude stimuli there can be significant stimulus-response interactions at frequencies outside the stimulus spectrum , a hallmark of a nonlinear stimulus-response transfer function . The unusual SNR curves produced by the phasic model ( Fig . 3B , right ) are caused by significant firing at double the signal frequency for some noise intensities . Thus , the dip of the SNR computed from the fundamental frequency ( marked with c ) did not mean that the signal was badly encoded , but meant that it was encoded at a harmonic frequency of the fundamental . Although in some previous studies [9] , , nonlinear SR has been considered and quantified at higher harmonics , those studies did not associate such measurements with a clear temporal pattern , e . g . , firing at the rising and/or falling phases , as shown in the present study . We gain insight into the phasic model's unusual response properties by applying reverse correlation analysis and examining the spike-triggered averages ( STA ) of several dynamic quantities: the stimulus ( Fig . 6A ) , Vm ( Fig . 6B ) , the fast gating variable , w , of IKLT ( Fig . 6C ) , and the system's trajectory in the Vm-w phase plane ( Fig . 6D and E ) for condition c ( Fig . 4 ) . We select from a brief time window ( 4 ms ) centered on the rising or falling phases of the signal ( Fig . 6F ) . The stimulus STA indicates that , on average , a modest hyperpolarizing dip preceded the strong brief depolarizing component just prior to spike initiation ( Fig . 6A ) , consistent with previous findings [14] , [22] , [23] . As seen in the period histograms above ( Fig . 4 , right , 2nd column ) , the phasic model barely responded to the signal's peak for a wide range of noise intensity . This lack of response was due to the activation of IKLT , indicated by the high values of w ( Fig . 6C , black ) and the nearly flat voltage traces ( Fig . 6B , black ) during the positive half of the sinusoid . For spikes occurring on the signal's rising phase , the rises in Vm and w just before spike initiation were significantly slowed by the noise ( gray ) in comparison to the Vm and w responses to signal without noise ( green ) . For spikes occurring on the signal's falling phase , the hyperpolarizing noise dip led , on average , to a faster decrease in w before a spike ( black ) compared to the decrease of w caused solely by the signal ( green ) ( Fig . 6C ) . These observations can be rationalized by phase plane analysis , by comparing features and trajectories in the STA of Vm-w phase plane ( Fig . 6D and E , right ) , with those of the deterministic phase trajectory of the signal-induced ( noise-free ) response ( Fig . 6D and E , left ) . Due to the presence of IKLT , there is not a fixed voltage threshold for the phasic model [36] , [37] . Rather , the firing threshold is dynamic and involves Vm and w together , as affected by the input current . The full model ( Equation 1 ) is multi-dimensional; however , by considering a reduced two-dimensional model [38] , we reveal the dynamic threshold geometrically , as a separatrix curve in the Vm-w plane . For this reduction , we suppose that the sodium current ( INa ) activates instantaneously , i . e . we set m to . The nullclines and separatrixes are dynamic and move in this two-variable projection , depending on the stimulus and other dynamic variables . In order to demonstrate the dynamic aspects , we consider , first , the rising phase case and choose two points on the STA time course and trajectory ( gray in Fig . 6 A–C , D , right ) : one slightly before ( red circle ) and one slightly after ( purple circle ) the initiation of a spike . The corresponding phase points in the signal's trajectory are also marked ( Fig . 6D , left , triangles ) . In Fig . 6D , the nullclines and separatrixes were constructed with the variables h , n , p , z , and r set to their individual instantaneous values at the times chosen for the two “snapshots” ( the circles/squares ) . For the STA-driven case , these values were obtained from trial-averaging of the respective variables over the spike-generating trajectories . In the noise-free case , the threshold separatrix driven by signal alone moved upward as the signal increased ( Fig . 6D , left , red to purple ) . However , the phase point for the signal alone ( triangles ) also moved upward and ahead of the separatrix; no threshold-crossing occurred and the system remained subthreshold . In contrast , the mean spike-triggering noise , first hyperpolarizing , pushed the trajectory ( Fig . 6D , right , gray ) toward the w-nullcline ( blue solid ) . This push and proximity to the w-nullcline slowed the motion along the trajectory ( i . e . , dw/dt is small close to the nullcline ) , accounting for the slowed rise of the Vm and w time courses; while in the noise-free case ( Fig . 6D , left ) the trajectory was not slowed or close to the w nullcline . With this slowed growth of the IKLT the phasic model was hyperexcitable compared to that in the noise-free case for the same signal values . When the STA noise became depolarizing , the separatrix moved upwards rapidly ( Fig . 6D , right , red to purple ) , sweeping through the slowed phase point , thereby creating a threshold crossing . The geometrical analysis for the threshold and response dynamics during the signal's falling phase is analogous , showing how the STA-noise accelerated the trajectory to enable spike generation . Just before a spike , the hyperpolarizing noise pushed the STA phase point down and leftward to become farther away from the w nullcline ( Fig . 6E , right , squares ) than in the noise-free case ( Fig . 6E , left , triangles ) . This increased distance indicates that dw/dt was more negative in the STA-case , hence speeding up the motion and the decrease of IKLT . This accelerated decrease leads to a timely window for depolarizing fluctuations that , on average , swept the separatrix upwards through the phase point , creating a threshold crossing and spike . Movies that show the dynamic phase planes ( involving separatrixes , nullclines , and Vm-w phase points ) are included in the supplemental materials ( Video S1 ) . Although IKLT played a major role in creating the above behavior , the inactivation of INa , denoted by h , also made a small contribution in a way similar to w . For example , the hyperpolarizing noise slowed down the decrease of h in the rising phase and speeded up the increase of h in the falling phase . The phase-plane analysis in the Vm-h plane is also included in the supplemental materials ( Video S2 ) . The above simulations showed that noise can enable the phasic model to encode low-frequency signals , which alone cause no response , in a way essentially different from the classical SR . In fact , the slope-based SR behavior might be inferred from other properties of the phasic model , such as the f-I , f-A ( signal amplitude ) and f-slope curves obtained in the presence of noise . These properties are usually studied in noise-free conditions; however , in reality more or less noise is always present for a neuron . Below we will compare these properties for the tonic and the phasic models in noisy conditions , and explain why they are related to the behavior of SR . The phasic model used in the above simulations derives its Class 3 excitability through a negative feedback current , the IKLT , which activates below spike threshold . Here , we tested whether another two types of phasic models show similar slope-based SR behavior . First , we tested a phasic HH model with a steeper activation of the IK [25] compared to the original HH model [39] . The modification of IK is to achieve the Class 3 excitability , which was observed with squid giant axons but not with the original HH model . With a simulation temperature of 18 . 5°C , the phasic HH model showed a slope threshold around dI/dt = 3 . 5 ( mA/ms ) /cm2 for ramp inputs increasing from 0 to 50 mA/cm2 . When noise of different intensities was added , the slope threshold decreased and further disappeared ( for noise σ≥50 nA/cm2 ) in a way similar to the behavior of the phasic model ( Fig . 9 ) , except that multiple spikes occurred during the ramp ( not shown ) . The phasic HH model also showed a band-pass filtering property for noise-free sinusoidal input ( not shown ) similar to that of the phasic auditory brainstem model ( Fig . 1C , right ) , except that multiple spikes can occur in each signal cycle at medium-low frequencies . The phasic HH model did not fire to a 5-Hz sinusoidal signal up to 28 mA/cm2 and the voltage trace showed similar rectification as exhibited by the auditory brainstem model ( Fig . 1B , right ) . When white noise was added to a subthreshold signal ( As = 15 mA/cm2 ) , the negative feedback created by the IK was not strong or fast enough to prevent spikes at the signal's peak ( not shown ) . However , after the white noise was low-pass filtered with a cutoff frequency of <100 Hz , the phasic HH model also showed highest sensitivity to the signal's rising and falling phases and low response to the peak ( Fig . 10A ) , which resembled the temporal pattern obtained from the phasic auditory brainstem model ( Fig . 4 , right ) . Correspondingly , the f-I curve of the phasic HH model was monotonically increasing with white noise ( Fig . 10D , solid ) but started becoming non-monotonic when fcut was lowered to 100–150 Hz . Fig . 10D ( dotted ) shows an example non-monotonic f-I curve with the peak at 0 and minimal firing at ±15 mA/cm2 ( fcut = 63 Hz ) . The change of the f-I curve with noise spectrum confirmed that a non-monotonic f-I curve correlates with the slope-based SR for a certain noise profile . It should be pointed out that in previous physiological and computational studies showing the non-monotonic f-I curves [24] , [26] , [27] , Gaussian white noise was smoothed by exponential filters with τ = 1–3 ms . The spectrum of noise created this way is a decreasing function of frequency . Repeating the phasic HH model with smoothed white noise showed non-monotonic f-I curves for τ = 1–3 ms . A third cellular mechanism that can create Class 3 excitability is the inactivation of INa [23] , [24] . To test the role of INa inactivation alone in creating phasicness and the slope-based SR behavior , we shifted the sodium inactivation voltage sensitivity ( ) leftward by 15 mV [23] , while freezing the conductance of the IKLT to its resting value as we did for the tonic model . These manipulations created the Class 3 excitability and the slope-detecting ability ( a slope threshold around dI/dt = 0 . 22 nA/ms for ramp inputs increasing from 0 to 2 nA in a noise-free condition ) . When white noise was added to a 20-Hz subthreshold signal ( As = 2 nA ) , the model fired most at the signal's rising and falling phases , with less activity at the peak ( not shown ) . Lowering the cutoff frequency of the noise decreased the activity at the peak ( an example plotted in Fig . 10B for fcut = 1 kHz ) . Correspondingly , the f-I curve with white noise was increasing for most I values and decreased slightly for I>1 . 5 nA , while the f-I curve with low-pass filtered noise was highly non-monotonic ( not shown ) . In addition , a previous computational study [24] showed that by lowering the conductance of INa from 120 to 83 mS/cm2 , the standard HH model can also exhibit Class 3 excitability and non-monotonic f-I curves for exponentially smoothed noise ( τ = 1 ms ) . We simulated this model with low-pass filtered noise and found behaviors similar to what was described above for the phasic HH model with modified IK . Principal neurons in the MSO are shown to have both low-voltage inactivation of INa [23] and IKLT . With blockade of the IKLT , the inactivation of INa alone can cause MSO neurons to show phasic response for gerbils of postnatal day ( P ) 14 or 15 and older , but becomes more prominent for neurons >P17 [23] . In our previous study [14] , we showed that IKLT plays a major role in creating the slope-based SR response for MSO neurons of P14–16 . Here we repeated the experiments with three neurons from older animals ( P18–20 ) , for which INa is known to be highly inactivated . Fig . 10C ( left ) shows that in response to a 20-Hz signal ( As = 1 . 5 nA ) , the neuron ( P18 ) fired mostly to the signal's rising , falling phases and the trough . Low-pass filtered noise ( fcut = 1 kHz ) , instead of white noise , was used in the recordings because the electrode was not fast enough to generate white noise [14] . The neuron fired in the signal's trough because its membrane time constant ( 0 . 3 ms ) was so fast that it can integrate slow noise fluctuations even when the neuron was somehow hyperpolarized [14] . After the application of dendrotoxin-K ( DTX-K , a blocking agent selective for IKLT ) , the neuron started responding to lower noise intensities at the signal's rising phase ( Fig . 10C , right ) and clear firing preferences to the rising and falling phases can be seen for all noise intensities . Less firing in the trough was observed due to a steeper V-I relationship after DTX-K was applied [23] . For example , in the control condition at the signal's minimum the neuron was hyperpolarized by −5 mV from the resting potential , while the hyperpolarization increased to −15 mV by the same signal after DTX-K was applied , too far from spike threshold for noise to elicit spikes in the trough . Note that in the recordings the signal's negative part was scaled by a factor of 0 . 5 to prevent large hyperpolarizations . Similar results were obtained in the other two neurons recorded . Tonic neurons show classical SR behavior; that is , noise helps the detection of a subthreshold signal by generating spikes when the signal is near its peak , i . e . , when the Vm is closest to the firing threshold . This type of noise-controlled signal encoding is qualitatively similar to enlarging the amplitude of the signal . Indeed , the amplitude-frequency plot for sinusoidal input ( Fig . 1C , left ) shows a relatively constant input threshold ( ∼0 . 3 nA ) except at the high-frequency end . Thus enlarging the signal amplitude sufficiently can make the model respond to the signal even in the absence of noise . In contrast , such an input threshold in terms of input amplitude does not exist for the phasic model ( Fig . 1C , right ) ; for sufficiently low-frequency signals , enlarging the signal amplitude does not make the model fire . Consequently , classical SR theory does not capture the signal response of noisy phasic models . It is more appropriate to define the input threshold for a phasic model in terms of input slope or frequency . Adding noise to a signal with a frequency/slope below this threshold makes the phasic model fire , not because adding noise effectively enlarges the signal amplitude , but because noise transiently increases the slope/speed of the signal , or equivalently diminishes the slope/frequency threshold ( Fig . 9 ) . In this sense , it is not surprising to see that the phasic model is most sensitive to the signal's rising phase , where the slope of the signal is steep , rather than to the signal's peak , where the slope is zero ( Fig . 4 , right ) . Based on our findings , we call the noise-gated encoding of a low-frequency signal by phasic models a “slope-based SR” , in contrast to the general ( amplitude-based ) SR observed in tonic models . Classical SR theory for weak and slow signals fails to explain both the SNR curve and the temporal firing patterns for the phasic model . This failure is because the SNR , computed either at the signal's fundamental frequency or harmonics , does not capture the full firing properties of the phasic model . Previous studies [24] , [26] , [27] have shown that , in the presence of noise , tonic-firing neurons have monotonically increasing f-I curves , while phasic-firing neurons ( i . e . , Class 3 ) have highly non-monotonic f-I curves ( Fig . 7 ) . Because a monotonically increasing f-I curve predicts that the neuron will respond mostly to the peak of a low-frequency signal , we hypothesize that a non-monotonic f-I curve with peak firing rate at moderate I is suggestive of a slope-based SR . This hypothesis is supported by the phasic HH model and the phasic model with low-voltage inactivation of INa , for which by varying the cutoff frequency of the low-pass filtered noise , the monotonicity of the f-I curve is correlated to the slope-based SR behavior ( Fig . 10 ) . Some predictions for the phasic models can be derived from the f-I curves obtained with steady and time-varying I . First , in the quasi-static limit , the amount of firing in the falling phase should be equal to the firing in the rising phase . In contrast , with increasing signal frequency , firing in the rising phase should increase , while firing in the falling phase should decrease and eventually disappear . This trend was shown by the phasic model with the IKLT in response to the 20- vs . the 30-Hz signals ( Fig . 7 ) . Second , when the signal amplitude was small ( i . e . , a smaller range of I ) , the slope-based SR behavior will be less distinct compared to the response with large signal amplitude . Consistent with this prediction , we showed in our previous study [14] that the responses of the phasic auditory brainstem model to the signal's rising and falling phases were less distinct when the signal amplitude decreased from 2 to 1 nA . Previous studies describing non-monotonic f-I curves have not addressed the influence of the noise spectrum on firing rate [24] , [26] , [27] . In general , the fact that phasic models can remain sensitive to the slope of subthreshold ( low-frequency ) signals in the presence of noise is a continuity of phasicness as defined in a deterministic setting . Although all the phasic models tested here showed Class 3 excitability , we propose different degrees of “phasicness” characterized by a model's resistance of losing the non-monotonicity of the f-I curves when the noisy fluctuations become faster . For example , the phasic model with IKLT has the strongest phasicness among all phasic models studied because it maintained its phasicness ( detecting slopes and onsets ) even if the noise fluctuated as fast as a 25-kHz white noise . The other phasic models can maintain their phasicness only when the noise was relatively slow . However , low-pass filtered noise is a better model of neural fluctuations than white noise , because noisy input to neurons , in the form of random background synaptic events , is naturally spectrally limited due to synaptic filtering [40] . Therefore , the slope-based SR may be widely present with phasic neurons , and that a highly non-monotonic f-I relation can serve as an indicator . Note that here the degree of phasicness is different from the definitions used in other studies [29] , [31] . In those studies the term is used to describe the firing properties of a phasic neuron to noise-free step input when multiple spikes can occur at the onset , whereas in our study all the phasic models fired only one spike at the onset . Phasic neurons are labeled slope-detectors based on their sensitivity to the slope of a ramp current . That is , they do not respond to a current input that has a slope shallower than a threshold value even if the input amplitude is large [12] , [28] . Although this concept was developed for auditory brainstem neurons [28] , it is a characteristic of all phasic neurons , because if a steady current input causes no response , by continuity there will be no spiking for sufficiently slow ramp input . We showed that the slope threshold is lowered in the presence of weak noise , and further diminished when noise is strong enough to cause significant spiking in the absence of a signal ( Fig . 9 ) . When the slope threshold is lowered , a phasic neuron can respond to inputs that are below the threshold obtained without noise . In a classical amplitude-based SR system , noise brings the system above its amplitude threshold , effectively mimicking an upward shift in the response-area plot in Fig . 1C ( left ) . Correspondingly , in a slope-based SR system , noise brings the system above its slope threshold , which effectively moves in the rightward direction in the response-area plot in Fig . 1C ( right ) . The sensitivity of phasic models to input slopes can also be clearly observed in the f-I-dI/dt plots obtained with time-varying I ( Fig . 7D ) . The peak firing rate of the phasic models increased with dI/dt , while the firing rate of the tonic model was insensitive to dI/dt . The higher firing rate on the rising phase of the 30-Hz signal compared to the firing rate to the 20-Hz signal indicated that the 30-Hz signal was closer to the input threshold in terms of the frequency of a sinusoid ( i . e . , 32 Hz for As = 2 ) . Based on the non-monotonic noise-based f-I curves obtained with the phasic auditory brainstem model ( Fig . 7B ) , we predicted that the firing rate will decrease with increasing amplitude of a sinusoidal signal for moderate and strong noise ( Fig . 8 , right ) . In contrast , the tonic model showed increasing f-A curves at low noise intensities and relatively constant curves at high noise intensities ( Fig . 8 , left ) . These differences in the f-A curves indicate that noise plays a different role in affecting input-output relationships for tonic vs . phasic neurons . When firing rate encodes the amplitude of a periodic input ( A ) , the input dynamic range that evokes spikes is limited , since the firing rate remained zero when A varies below threshold ( Fig . 8 , black lines marked with a ) . Adding noise to the input linearizes the f-A curve , thereby achieving a larger input dynamic range [1] , [41]–[43] . For the tonic model , adding a small amount of noise ( e . g . , 4 pA ) can achieve this type of linearization , so that the firing rate increases with a relatively constant rate even in the subthreshold regime ( Fig . 8 , left , b ) . When the noise intensity further increases , the output dynamic range decreases as the firing rate becomes insensitive to A . For the phasic model , although a similar trend exists for weak noise ( e . g . , σ = 3 pA ) , the input dynamic range increased only around the input threshold ( 4 nA ) . The output dynamic range was also limited because the maximum firing rate was close to the signal frequency even with a small amount of noise added ( Fig . 8 , right , b ) . However , with strong noise added , the firing rate decreased relatively smoothly with A ( Fig . 8 , right , c ) and a large range of A can be encoded by the firing rate . In addition , because a large amount of noise can cause multiple spikes in each signal cycle , a large output dynamic range was also achieved ( Fig . 8 , right , c ) . It should be noted that although the firing rate decreases , the temporal precision increases with A . The two groups of spikes on the signal's rising and falling phases were less distinct for As = 1 nA compared to the 2-nA case [14] , although the SNRs were comparable ( Fig . 3B , right ) . The maximum slope of the input increases when signal frequency is kept constant and the signal amplitude is enlarged; however , the fraction of time spent around the maximum slope decreases , leading to a narrower time-window for the first crossing of the slope threshold . Phasic neurons are widely present in different sensory systems ( for review , see [29] ) . These neurons are thought to detect onset events , encode fast changes of its input , and maintain response temporal precision [28] , [29] . Our results suggest that in the presence of noise , phasic neurons can encode slow inputs and remain sensitive to changes of an input ( e . g . , the beginning and end of the positive cycle of a sinusoid ) , thereby extending their phasicness into the low-frequency region . The slope-based SR behavior reflects the tendency of phasic neurons to maintain its phasicness when moderate noisy fluctuations are applied; large intensity noise will , of course , reduce slope detection . In summary , slope-based SR is a continuity of phasic neurons' response properties over large input dynamics . Noise-gated or noise-assisted coding , of which SR is a classic example , is a popular topic in neural applications and other physical systems , where simple threshold models serve as a canonical model for SR [4] , [44] . Phasic systems offer a new avenue of research in noise-assisted coding , where the distinction of signal threshold is based on the slope of the signal , rather than the more traditional scenario of an amplitude threshold . We expect that new noise induced phenomena , qualitatively distinct from those previously described , will emerge as noise-driven phasic systems are studied further . Details of the phasic neuron model are described in [14] . Briefly , the auditory brainstem neuron model [15] contains a fast sodium current ( INa ) , a high-threshold ( IKHT ) and a low-threshold ( IKLT ) potassium currents , a hyperpolarization-activated cation current ( Ih ) , and a leak current ( Ilk ) . ( 1 ) Vm is the membrane voltage . is the current input . Membrane capacitance , = 12 pF; maximal channel conductances , = 1000 nS , = 150 nS , = 200 nS , = 20 nS , and = 2 nS; reversal potentials , = +55 mV , = −70 mV , = −43 mV , and = −65 mV . All the conductances and channel time constants are multiplied by a factor of 2 and 0 . 33 , respectively , to mimic the condition at 32°C , because our previous study [14] had slice recordings at 32°C . Although there are several currents with time-varying conductances , only the fast component of the IKLT , w , was playing a major role in the simulations with sinusoidal and noisy inputs [14] . The tonic model is created by fixing the gating variables , w and z , to the values obtained at resting potential [45] . A further modification of the Day's frozen model is to increase the from 1000 to 1500 nS , which enables a larger amplitude of the limit cycle and a broader input range for repetitive firing [14] . The tonic model created this way has the same membrane resting potential and input resistance as the phasic model does . The signal was a 20-Hz sinusoidal current ( unless otherwise specified ) , , with zero mean . The signal was kept subthreshold , and white noise ( 0–25 kHz ) was added to make the model spike . ( 2 ) Note that in our previous study [14] and the present physiological recordings , the negative part of the signal is multiplied by a factor of 0 . 5 to avoid excessive hyperpolarization of the neuron in whole-cell recordings . Here we used the unmodified sinusoid in the simulations because we were trying to make a direct comparison with classical SR systems where pure sinusoidal inputs are commonly used as signals . Two signal amplitudes were chosen ( As = 0 . 1 and 0 . 2 nA for the tonic model and As = 1 and 2 nA for the phasic model ) so that the detectability of signal from noise ( quantified by the signal-to-noise ratio ) was comparable between the two models . Our choice of the input amplitude for the phasic model is reasonable because MSO neurons , the phasic neurons that demonstrated similar properties compared to the phasic model [14] , have an average input threshold of 3–4 nA for step input [46] . The sampling frequency was 50 kHz . For each noise intensity , ∼5000 spikes were obtained unless stimulus duration reached 200 s . Fig . 11 shows an example of input stimulus ( top ) and the corresponding response of the tonic model ( middle ) . The detectability of the signal from the added noise was quantified by computing the signal-to-noise ratio ( SNR ) from the power spectrum of the spike train . Fig . 11 ( bottom ) shows an example of power-spectrum density ( PSD ) computed from the spike times of the tonic model . Spike times were re-sampled with a lower time resolution , 2 ms , producing a Nyquist of 250 Hz in the PSD plot ( higher frequency was unnecessary since the signal frequency was low ) . Two peaks are visible in the PSD plot , one at the signal frequency ( marked as Pf ) and another at the first harmonic ( marked as Ph ) . The SNR is computed as ( 3 ) where x is either the fundamental frequency ( x = f ) , or the first harmonic frequency ( x = h ) . Pbf is the baseline for the fundamental , computed as the average of a small range near Pf , and Pbh is the baseline for the first harmonic , computed as the average of a small range near Ph . In the following text , SNR will refer to the signal-to-noise ratio at the fundamental frequency unless otherwise specified . For simple spiking systems an adiabatic theory ( slow signal ) with weak signal amplitude approximates SNR as [4] ( 4 ) where ε is the signal amplitude , D is the noise intensity ( corresponding to the σ2/2 in the present study ) , and ΔU is the potential barrier separating the deterministic rest state and firing threshold . When Equation 4 was used to fit a SNR curve , D is chosen so that the peak of Equation 4 overlaps the peak of the SNR . Specifically , because Equation 4 reaches its peak when D = ΔU/2 , ΔU is chosen as twice the noise intensity where the peak of the SNR is obtained . Then the whole equation is scaled to match the peak SNR value . Note that Equation 4 predicts that the peak of the SNR is obtained with a fixed noise intensity invariant of the signal amplitude [32] . Another frequently used quantification of SR is the spectral power amplification ( SPA ) [9] . It is the output power at the signal frequency normalized by the input signal power , ( 5 ) Note that here the output power was the power of a discrete signal ( i . e . , spike times ) , while the input power was the power of a continuous signal ( i . e . , a sinusoidal current ) . In vitro data presented here is to confirm that strong sodium inactivation can replace IKLT to create phasic responses . Detailed experimental procedures are described in [14] . Briefly , gerbils ( Meriones unguiculatus ) aged P17–18 were used to obtain 150-µm brainstem slices . The internal patch solution contained ( in mM ) 127 . 5 potassium gluconate , 0 . 6 EGTA , 10 HEPES , 2 MgCl2 , 5 KCL , 2 ATP , 10 phosphocreatinine , and 0 . 3 GTP ( pH 7 . 2 ) . During recordings , sliceswere placed in a chamber with artificial cerebrospinal fluid ( ACSF ) containing ( in mM ) 125 NaCl , 4 KCl , 1 . 2 KH2PO4 , 1 . 3 MgSO4 , 26 NaHCO3 , 15 glucose , 2 . 4 CaCl2 , and 0 . 4 L-ascorbic acid ( pH 7 . 3 when bubbled with 95% O2 and 5% CO2 ) at 32±1°C . DTX-K ( 60 nM ) was added to the bath to block the IKLT . The perfusing rate of the oxygenated ACSF in the recording chamber was 2ml/min . An Axoclamp2A amplifier , in combination with Labview ( National Instruments ) , was used for stimulus generation , balance of series resistance , and data acquisition at 10 kHz .	Principal brain cells , called neurons , show a tremendous amount of diversity in their responses to driving stimuli . A widely present but understudied class of neurons prefers to respond to high-frequency inputs and neglect slow variations; these cells are called phasic neurons . Although phasic neurons do not normally respond to slow signals , we show that noise , a ubiquitous neural input , can enable them to respond to distinct features of slow signals . We emphasize the fact that , in the presence of noise , they are still sensitive to the change in stimulus , rather than to the constant part of the slow inputs , just as they are for fast inputs without noise . This feature distinguishes the response of phasic neurons from those of other neurons , which show more sensitivity to the amplitude of their inputs . We believe that our study has significantly broadened the understanding about the information-processing ability and functional roles of phasic neurons .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "neuroscience/sensory", "systems", "computational", "biology/computational", "neuroscience" ]	2010	Slope-Based Stochastic Resonance: How Noise Enables Phasic Neurons to Encode Slow Signals
Pest and pathogen losses jeopardise global food security and ever since the 19th century Irish famine , potato late blight has exemplified this threat . The causal oomycete pathogen , Phytophthora infestans , undergoes major population shifts in agricultural systems via the successive emergence and migration of asexual lineages . The phenotypic and genotypic bases of these selective sweeps are largely unknown but management strategies need to adapt to reflect the changing pathogen population . Here , we used molecular markers to document the emergence of a lineage , termed 13_A2 , in the European P . infestans population , and its rapid displacement of other lineages to exceed 75% of the pathogen population across Great Britain in less than three years . We show that isolates of the 13_A2 lineage are among the most aggressive on cultivated potatoes , outcompete other aggressive lineages in the field , and overcome previously effective forms of plant host resistance . Genome analyses of a 13_A2 isolate revealed extensive genetic and expression polymorphisms particularly in effector genes . Copy number variations , gene gains and losses , amino-acid replacements and changes in expression patterns of disease effector genes within the 13_A2 isolate likely contribute to enhanced virulence and aggressiveness to drive this population displacement . Importantly , 13_A2 isolates carry intact and in planta induced Avrblb1 , Avrblb2 and Avrvnt1 effector genes that trigger resistance in potato lines carrying the corresponding R immune receptor genes Rpi-blb1 , Rpi-blb2 , and Rpi-vnt1 . 1 . These findings point towards a strategy for deploying genetic resistance to mitigate the impact of the 13_A2 lineage and illustrate how pathogen population monitoring , combined with genome analysis , informs the management of devastating disease epidemics . As the cause of potato late blight , Phytophthora infestans is one of the most destructive plant pathogens within this genus of fungus-like oomycetes and widely known as the Irish potato famine pathogen [1] , [2] . P . infestans has migrated from Central or South America [3] , [4] , where it infects many native solanaceous hosts , and remains responsible for significant losses to key staple crops ( potato , tomato and other solanaceous plants ) worldwide [5] , [6] . Potato late blight management relies on regular applications of a range of anti-oomycete ‘fungicides’ . However , under optimal weather conditions the pathogen may complete several infection cycles a week on a susceptible host , with control failure leading to rapid epidemics and crop loss . Resistance breeding offers great potential but the durability of resistance conferred by R genes has been continually challenged by the evolution of new virulence traits within pathogen populations [7] . P . infestans is normally diploid with a heterothallic ( i . e . outbreeding ) mating system that requires co-infection of A1 and A2 mating types to form long-lived sexual oospores . A mixture of sexually compatible A1 and A2 mating types increases the opportunities for sexual reproduction , providing the pathogen with an evolutionary advantage via increased genetic diversity and oospores as a source of primary inoculum in the soil [8] , [9] . In the absence of oospores , in temperate regions the pathogen can only survive as asexual clones in potato tubers ( as seed , in discard piles or unharvested tubers ) . Mycelium from such infections generates sporangia that are carried by wind and rain-splash to a new host where they germinate directly or release multiple motile zoospores that infect , colonize and release new sporangia via host stomata . Many studies have demonstrated that , despite the theoretical advantages of sexual recombination [8] , a succession of clonal lineages of P . infestans have dominated the population in many potato production regions [7] , [10] . In Europe , the incursion of the A2 mating type occurred 135 years after the A1 type [11] . However until recently , the A2 type remained infrequent in most parts of Europe [10] , [12] , which limited the opportunities for sexual reproduction of the pathogen [10] , [13] , [14] . Conversely , in parts of Mexico and the Nordic regions of Europe , populations of P . infestans have more balanced A1:A2 mating type ratios and are genetically diverse , with sexually formed oospores that act as a source of primary inoculum [7] , [15] . Effective management of potato late blight is aided by an understanding of the characteristics of the contemporary pathogen population . For example , the aggressive and metalaxyl resistant A2 mating type US-8 lineage replaced the US-1 lineage which resulted in significant potato crop losses across the USA from 1985–1995 [16] . Pathogen genetic diversity has been monitored using a range of genetic markers [17] of which simple sequence repeats ( SSRs ) have recently proved effective for defining multilocus genotypes ( MLGs ) [18] . Key adaptive traits such as the ability of sporangia or zoospores to infect and colonise host tissue ( aggressiveness ) combined with efficient dissemination and , in temperate regions , survival from season to season ( fitness ) determine the success of particular P . infestans MLGs . Lesion growth rate and the period from inoculation to sporulation ( latent period ) are important components of aggressiveness [19] , [20] . Fitness , a measure of reproductive success [21] , is best studied in the field over several disease cycles . In a polycyclic disease such as potato late blight , even minor differences in aggressiveness or fitness can have a significant effect on the relative success of an MLG in the population . Traits such as ability to overcome specific host resistance , fungicide resistance or altered response to environmental conditions [22] are also important determinants of evolutionary success in the pathogen population . The sequenced genome of P . infestans strain T30-4 provides a ‘blueprint’ of the gene complement and genome architecture of this pathogen [23] . The assembly served as a reference sequence in this work . Recently , two additional isolates PIC99189 and 90128 were resequenced using 36 bp Illumina reads ( 10 . 4× and 17 . 1× coverage , respectively ) [24] . These projects revealed that P . infestans possesses a ‘two-speed genome’ with gene dense and gene-sparse repeat-rich regions . Gene-sparse regions ( GSRs ) are enriched in genes that are induced in planta and genes showing presence/absence polymorphism , copy number variation ( CNV ) or high nonsynonymous over synonymous substitution rates [24] . Effectors and other pathogenicity factors [23] that reside in these GSRs have the potential to evolve rapidly [24] , consistent with the pathogen's well-documented capacity to adapt to novel host resistance . These effectors include RXLRs , a class of host translocated proteins that carry an N-terminal signal peptide followed by an RXLR motif [23] , [25] . All known effector genes with Avr ( avirulence ) activity are in planta-induced genes of the RXLR type [26] . The study of the RXLR repertoire in emerging P . infestans lineages provides insights into the molecular basis of the infection phenotype on plants carrying the cognate R genes . In the present study , we investigated changes in the population of the late blight pathogen P . infestans in Great Britain ( GB ) and identified a major new lineage of P . infestans that first emerged in mainland Europe in 2004 . We investigated the factors driving this population change , demonstrating that 13_A2 MLG was amongst the most aggressive and fit MLGs in laboratory and field studies and able to overcome an important , previously durable source of host resistance . We sequenced the genome of an isolate of the 13_A2 MLG and compared it to the reference genome strain T30-4 . We identified genes unique to this MLG , signatures of positive selection and CNVs , in particular in the RXLR effector repertoire . We also studied patterns of gene expression during an infection time course and noted a remarkable extended biotrophic phase , with distinct sustained induction of genes including RXLR effectors in the 13_A2 MLG isolate compared to other reference isolates . Lastly , we evaluated the effectiveness of promising sources of R genes that recognise invariant Avr genes , demonstrating that they remain effective against a 13_A2 MLG isolate . Despite the differential expression of many RXLR effector genes , we present evidence of a common set of in planta-induced effectors which we consider ‘targets’ for durable late blight disease resistance breeding . We collected and determined the simple sequence repeat ( SSR ) -based [18] multilocus genotypes ( MLGs ) of 4 , 654 P . infestans isolates from 1 , 100 late blight disease outbreaks in Great Britain ( GB ) , sampled between 2003 and 2008 ( Table S1 in Text S2 , Figure S1 in Text S1 ) cross-referencing these to a sample of isolates ( n = 537 ) collected in previous GB surveys from 1982–1998 [13] , [27] , [28] . These SSR markers yielded between 2 and 25 alleles per locus and proved an effective tool to discriminate isolates within the GB pathogen population ( Figure S2 in Text S1 , Table S2 in Text S2 ) . The P . infestans population was dominated by clonal lineages with fewer than seven MLGs comprising >82% of the isolates each year ( Figure 1A , Table S3 in Text S2 ) . The A2 mating type frequency increased and genetic diversity reduced markedly over the years 2005 to 2008 ( Figure 1A and 1B , Figures S3 , S4 in Text 1 ) . A novel A2 mating type and metalaxyl resistant ( Table S2A in Text S2 ) MLG , termed 13_A2 , was first recorded in seven British potato crops from July 2005 and went on to rapidly displace other MLGs across the region ( Figure 1C ) . In 2006 MLG 13_A2 was prevalent in England from late May but not sampled in Scottish crops until late August ( Figure S5 in Text S1 ) which is consistent with a progressive crop-to-crop dispersal across the region in 2006 ( Figure 1C ) . Variation within the more variable SSR loci ( particularly G11 and D13 ) has allowed discrimination of minor variants amongst the 2 , 295 isolates of 13_A2 MLG in this study ( Figure 1B , Table S2B in Text S2 ) . P . infestans MLG 13_A2 was first detected in isolates collected from The Netherlands and Germany in 2004 , which is corroborated by other reports of A2 metalaxyl resistant isolates in continental Europe and suggests a north-westward migration to Great Britain ( GB ) ( Table S4 in Text S2 ) [29]–[31] . The ‘misc’ category of SSR genotypes is a composite of all the novel and rarely sampled MLGs representing diversity that is consistent with sexual recombination [15] . However , in contrast to some other regions of Europe where almost every isolate is genetically distinct [15] , this ‘misc’ category was recovered in GB disease outbreaks at a frequency of below 5% of the population from 2003 to 2008 ( Figure 1A ) indicating that the population remained largely clonal over this period ( Figure 1B and Figure S4 in Text S1 ) . We examined the selective forces behind the population displacement in extensive laboratory and field evaluations of the fitness of many isolates of P . infestans . Aggressiveness , ‘the quantity of disease induced by a pathogenic strain on a susceptible host’ [32] , is a key component of pathogen fitness and was estimated by measuring lesion size and latent period ( time elapsed from inoculation to spore production ) . Such adaptive traits contribute to the epidemiological success of this pathogen and closely correlate with spore production and infection frequency [19] . A detached leaflet laboratory screen of 26 P . infestans isolates on five contemporary potato cultivars varying in foliar late blight resistance ( Tables S5 and S6 in Text S2 ) was conducted at 13°C and 18°C . The isolates comprised representatives of the 9 MLGs in the 2006 British survey and reference isolates from other years and other European countries . MLG 13_A2 isolates consistently ranked among the most aggressive , showing among the shortest latent periods and the largest lesions of the MLGs tested , on all potato cultivars ( Figure 2 , Figures S6 , S7 , S8 in Text S1 ) . This effect was more pronounced at 13°C than at 18°C , suggesting that MLG 13_A2 is better adapted to cooler conditions . Consistent with its frequency in the population ( Figure 1C ) , MLG 6_A1 was also shown to be aggressive in this test . Measurements of the lesion size produced on two different potato cultivars by a 13_A2 MLG isolate ( 06_3928A ) and by the reference genome strain T30-4 [23] , showed that 06_3928A formed larger lesions , with a shorter latent period than T30-4 ( Figure S9 in Text S1 ) . Also , we observed marked differences in the pattern of induction of the Cdc14 gene in these two isolates during the biotrophic phase of infection on potato . This marker gene is associated with sporulation [33] , and was induced earlier and more strongly in the biotrophic phase of infection by 06_3928A than by T30-4 which is consistent with the shorter latent period in 06_3928A ( Figure S10 in Text S1 ) . The above experiments demonstrate that , in a single disease cycle , 13_A2 isolates tend to be more aggressive than other MLGs under laboratory conditions . We went further to examine the ability of MLG 13_A2 to compete directly with other MLGs over many disease cycles in a field epidemic via a ‘mark and recapture’ experiment . The central potato plant of each of 20 field plots ( five cultivars ) was inoculated with a mixture of five isolates: 13_A2 ( isolate 06_3928A ) and representatives of four other contemporary MLGs , including 6_A1 ( Table S5 in Text S2 ) . Infected leaves from the ensuing epidemic were sampled over 21 days and 716 blight lesions were fingerprinted using direct SSR analysis . 13_A2 was the most prevalent MLG recovered , being responsible for the disease in 93–100% of the lesions sampled ( Figure 3A ) . This high frequency was noted on all five cultivars which supported the field survey data showing a high recovery rate of 13_A2 MLG isolates from the ten most sampled cultivars ( Figure 3B and Figure S11 in Text S1 ) . In accordance with our results on the aggressiveness of 13_A2 at 13°C , the cool and wet conditions during the field trial ( Figure S12 in Text S1 ) may have favoured the spread of MLG 13_A2 . Combined , these results provide strong evidence that isolates of 13_A2 MLG are more fit and aggressive than other MLGs on many host cultivars and under field and laboratory conditions , and are consistent with data on other P . infestans population displacements [34] . In field trials since 2006 , significant levels of disease were observed on some cultivars known to be partially resistant to foliar blight since the 1990s , such as Stirling [35] and Lady Balfour , a cultivar used in organic production . This was supported in subsequent whole-plant resistance screens which indicated a collapse of Stirling's resistance ( Figure 4 ) . We examined the ability of many isolates of 13_A2 MLG to overcome foliar late blight resistance on eleven potato R differential plants that contain immune receptor genes derived from the Mexican species Solanum demissum . All isolates of 13_A2 were able to cause disease on all the differential plants , except R8 and R9 ( Table S5 in Text S2 and Figure S13 in Text S1 ) . This indicates that , in addition to being particularly aggressive on susceptible potato cultivars , isolates of 13_A2 caused more disease on a broader spectrum of late blight resistant potato cultivars than isolates belonging to other P . infestans MLGs . In late blight resistant potato plants , hypersensitive cell death and resistance are triggered by recognition of specific pathogen RXLR effectors by matching R proteins [26] . Effectors are pathogen proteins delivered inside plant cells to promote host colonization , for instance by suppressing plant immunity [36] . RXLR proteins , encoded by ∼563 genes in the P . infestans T30-4 genome [23] , are the main class of host translocated effectors . Some RXLR effectors are said to have an “avirulence” activity when acting as triggers of plant immunity . To determine the genetic features , in particular the effector gene repertoire , associated with the 13_A2 MLG phenotype , we generated ∼58-fold genome coverage Illumina paired-end reads of isolate 06_3928A ( see details in Text S3 ) . We processed the sequences first by aligning the reads to the reference genome of P . infestans strain T30-4 [23] , and then by performing de novo assembly of unaligned reads . In total , 95 . 6% of the 06_3928A reads aligned to the T30-4 sequence ( Table S7 in Text S2 ) . We detected 18 , 106 coding sequences with an average breadth of coverage of 99 . 2% ( Table S8 in Text S2 ) . We optimized bioinformatic parameters for calling single nucleotide polymorphisms ( SNPs ) to reach 99 . 9% accuracy and 85 . 8% sensitivity ( Figure S14 in Text S1 ) . Using these parameters , we identified 22 , 433 SNPs in 5 , 879 coding sequences of 06_3928A ( Tables S8 in Text S2 and Table S9 ) . This is similar to the 20 , 637 and 21 , 370 SNPs reported for P . infestans isolates PIC99189 and 90128 , respectively [24] ( Table S8 in Text S2 ) . Of the total SNPs discovered , 11 , 795 were unique to 06_3928A among the four examined strains , indicating a considerable degree of variation in the 13_A2 isolate ( Table S9 and Figure S15 in Text S1 ) . To detect signatures of positive selection in the 13_A2 lineage , we calculated rates of synonymous ( dS ) and nonsynonymous ( dN ) substitutions for every gene ( Table S10 ) . Of the 22 , 523 coding sequence SNPs , 11 , 421 are nonsynonymous ( 51% ) corresponding to an average dN/dS rate of 0 . 34 . Secreted protein genes , particularly RXLR effector genes , show higher dN rates compared to other categories ( Figure 5 ) . Of the 405 SNPs detected in RXLR genes , 278 are non-synonymous ( 69% ) corresponding to an average dN/dS rate of 0 . 53 ( Table 1 and S11 ) . RXLR effectors are modular proteins with N-termini involved in secretion and host-translocation while C-termini encode the effector biochemical activity [25] , [37] . The C-terminal domains of RXLR effector genes are highly enriched in nonsynonymous substitutions as previously noted in other oomycete species ( Figure 6 ) [38] . Several RXLR effector genes show high dN/dS ratios and multiple replacements in their C-terminal domain ( Figure S16A–C in Text S1 ) . In addition to RXLR effectors , other secreted proteins including a Kazal-like serine protease inhibitor show high dN/dS ratios ( Figure S16D in Text S1 ) . These amino acid polymorphisms could contribute to the enhanced aggressiveness and virulence phenotypes of this genotype . To estimate copy number variation ( CNV ) in the resequenced genome of P . infestans 13_A2 isolate 06_3928A relative to T30-4 , we used average read depth per gene and GC content correction ( see Text S3 ) . We detected 367 CNV events in 06_3928A genes , of which there are 320 duplications and 47 deletions ( Tables S12 , S13 ) . In agreement with other studies [23] , [24] genes showing deletions and duplications occur more frequently in the plastic gene sparse regions of the 06_3928A genome ( Figure S17 in Text S1 ) . RXLR effector genes show higher rates of CNV compared to other gene categories ( Figure S18 in Text S1 and Table S13 ) . We identified two RXLR effectors with ∼4–5 additional gene copies in the isolate 06_3928A and this was validated with a realtime PCR assay in 17 of 18 other isolates of 13_A2 MLG . Another 18 P . infestans MLGs had lower copy numbers suggesting the higher copy number duplications are unique to 13_A2 MLG isolates ( Figure S19 in Text S1 ) . Remarkably , 21% ( 10 out of 47 ) of the genes that are deleted in 06_3928A encode RXLR effectors ( Table S14 in Text S2 ) . 13_A2 MLG isolates are able to infect potatoes carrying the R1 gene ( Figure S13 in Text S1 ) which is consistent with our finding of an ∼18 Kb deletion encompassing the region surrounding the Avr1 RXLR effector gene in the 06_3928A isolate ( Figure S20 in Text S1 ) [26] , [39] . To identify sequences that are unique to 06_3928A , we performed de novo assembly of the unmapped Illumina reads and identified a total of 2 . 77 Mb contigs that did not align to P . infestans T30-4 sequences . Ab initio and homology based gene calling in these 06_3928A-specific contigs revealed 6 candidate RXLR effector genes absent in the T30-4 reference genome strain ( Table S14 in Text S2 ) . All 6 RXLR genes were subsequently confirmed by PCR on genomic DNA to be present in the 06_3928A isolate and absent in T30-4 ( see Text S3 , Table S15 in Text S2 ) . Among these , a highly divergent homolog of Avr2 evades recognition by the R2 resistance gene and explains the virulence of 06_3928A on R2 potatoes ( Tables S14 , S15 in Text S2 and Figure S13 in Text S1 ) [26] , [40] . Interestingly , the PCR testing also showed that the six novel RXLR genes in the 06_3928A isolate of 13_A2 MLG are present in various combinations in other multilocus genotypes ( MLGs ) sampled from Great Britain . This illustrates the heterogeneity of the RXLR effector repertoire that can occur within the wider P . infestans population . These findings point to a series of genetic polymorphisms that collectively contribute to the aggressiveness and virulence phenotype of the 13_A2 MLG . The phenotype of the 13_A2 MLG may not only result from changes in gene coding sequences as documented above , but also from changes in gene expression . An infection time course was performed by hybridizing NimbleGen microarrays with cDNA from potato leaves harvested at 2–4 days post inoculation ( dpi ) with P . infestans 06_3928A , the T30-4 reference genome strain , and a third strain , NL07434 , collected in 2007 in The Netherlands ( see Text S3 ) . We observed frequent expression polymorphisms between the three strains with 1 , 123 genes specifically induced in 06_3928A , compared with 110 in T30-4 and 891 in NL07434 ( Figure 7A , Table S16 ) . Remarkably , only 398 out of 4 , 934 genes were induced in all three strains indicating distinct isolate-specific sets of genes induced during potato infection ( Figure 7A ) . P . infestans effector genes are sharply induced during the biotrophic phase of infection , when the pathogen associates closely with living plant cells [23] , [26] . We identified 104 RXLR effector genes that are induced during biotrophy in 06_3928A compared to only 79 and 68 in T30-4 and NL07434 , respectively ( Figure 7A , Table S11 ) . Of these 104 RXLR genes , expression of 20 was specifically detected in the 06_3928A isolate but not in the other two ( Figure 7A , Figure S21 in Text S1 ) . In contrast , 18 RXLR effector genes are not induced in 06_3928A but are induced in at least one of the other two isolates ( Figure 7A ) . One of these genes , Avr4 is recognized by the R4 resistance gene [26] , [41] . The lack of induction of Avr4 in 06_3928A ( Figure S21 in Text S1 ) is consistent with the virulence of 13_A2 isolates on plants containing R4 ( Figure S13 in Text S1 ) . The updated repertoire of RXLR effectors and their expression profiles presented in this study provides additional data for systems biology approaches to understanding the role of effectors in plant-microbe interactions [42] . We noted a distinct temporal pattern of in planta gene induction in 06_3928A . Most up-regulated genes in this isolate showed sustained induction over 2 and 3 dpi in contrast to T30-4 and NL07434 , in which transcript abundance generally declines at 3 dpi ( Figure 7B–C , Table S16 ) coinciding with the onset of host tissue necrosis [23] . These findings prompted us to determine the extent to which gene induction patterns and disease progression correlate in 06_3928A and these other isolates . Microscopic observations of lesions caused by 06_3928A revealed significantly larger biotrophic zones during infection ( Figure 7D ) . The genes showing a sustained induction period in 06_3928A include putative virulence factors such as RXLR effectors , cell wall hydrolases , proteases and protease inhibitors ( Table S16 ) . The extended biotrophic phase of 06_3928A during host plant colonization , combined with expression of a range of effectors and other secreted virulence determinants , likely contribute to the enhanced aggressiveness ( Figure 2 ) and field fitness of MLG 13_A2 isolates . However , additional work is required to determine exactly which genes contribute to MLG 13_A2 aggressiveness and fitness . The genome analyses of MLG 13_A2 offers opportunities to identify useful forms of host resistance . The 45 “core” RXLR effectors showing in planta gene induction during biotrophy in all 3 examined strains include 5 known avirulence effector genes ( Figure 7A ) . Whilst homologs of Avr2 [40] and Avr3a [43] in the 06_3928A isolate contain sequence polymorphisms and are known to evade recognition in plants carrying the corresponding R2 and R3a genes ( Figure S13 in Text S1 ) , Avrblb1 [44] , Avrblb2 [45] and Avrvnt1 [46] occur as intact coding sequences that are induced during infection ( Figure 8A ) . These three Avr effectors are therefore predicted to be recognized by their cognate immunoreceptors . To determine whether 13_A2 MLG can infect plants carrying the Rpi-blb1 , Rpi-blb2 and Rpi-vnt1 . 1 resistance genes , we used isolate 06_3928A to inoculate stable transformant potato cv . Desiree lines expressing , independently , each of the three R genes . In each case , 06_3928A was unable to infect the R potatoes and triggered a typical hypersensitive response ( Figure 8B ) indicating that the three R genes are effective against this 13_A2 MLG isolate . Such sources of resistance will thus be an effective component of any integrated management system against late blight caused by genotype 13_A2 . We report the emergence of an aggressive clonal lineage of P . infestans , multilocus genotype ( MLG ) 13_A2 , and its rapid displacement of other genotypes within the Great Britain ( GB ) population . MLG 13_A2 has overcome previously durable disease resistances in potato , such as in cultivar Stirling and is resistant to phenylamide fungicides . Late blight caused by this lineage has thus proved challenging to manage and its migration to other potato growing regions of the world poses a threat to sustainable crop production . Therefore , there is a need , when developing a strategy for deploying disease resistance , to identify and respond rapidly to dramatic changes , and new epidemics caused by emerging genotypes within the pathogen population . Genome analyses of the 13_A2 isolate 06_3928A revealed a high rate of sequence variation and a remarkable pattern of extended biotrophic growth , which may explain 13_A2's aggressiveness and ability to cause disease on previously resistant potato cultivars . The genome analysis proved valuable in identifying RXLR effectors sensed by potentially durable potato resistance genes . This stresses the benefits of a crop disease management strategy incorporating knowledge of the geographical structure and evolutionary dynamics of pathogen lineages combined with data on their genome sequence diversity ( and in planta induced effector gene complement ) . Such data , when linked to the host R gene repertoire [47] , offers options for strategic deployment of host resistance with a positive impact on crop yield and food security . P . infestans isolates were obtained from more than 1 , 100 outbreaks of potato late blight across Great Britain ( GB ) from 2003 to 2008 . The locations of 672 outbreaks sampled in 2006 to 2008 and further details on sampling and pathogen characterisation are available ( Figure S1 in Text S1 and Text S3 ) . The mating type of each of 4 , 654 isolates collected in this study was tested by pairing with known tester isolates on Rye A agar plates . After an initial screen of the new A2 mating type lineages using the RG57 [48] RFLP probe ( Table S2A in Text S2 ) , all isolates were genotyped using 11 SSR markers [18] in 3 multiplexed PCR assays using fluorescently labelled primers on an ABI 3730 capillary sequencer ( Tables S2 and S3 in Text S2 and Text S3 ) . The SSR data were used to define MLGs , explore the relatedness amongst the multilocus genotypes ( MLGs ) and to describe the population change . Due to the presence of three alleles in some isolates , we calculated clonal distance [49] using the infinite alleles mutation model , to quantify genetic distance between MLGs . This distance essentially counts the number of alleles that differ between individuals . Isolates with null alleles were included , but any isolates that were not genotyped at one or more loci were excluded . Distance among multilocus genotypes was calculated in GenoDive ( Distributed by P . G . Meirmans at http://www . bentleydrummer . nl/software/software/Home . html ) . Minimum spanning networks were calculated by MINSPNET [50] and visualized using neato in the Graphviz package [51] . The numbers of isolates used to construct the trees were 748 , 795 , 1 , 072 , and 892 for 2003–2005 , 2006 , 2007 , and 2008 , respectively ( Figure 1B and Figure S4 in Text S1 ) . Representative isolates from the main MLGs from Great Britain plus a selection of reference isolates from other countries were used to examine two components of aggressiveness [19] ( lesion size and latent period ) on five contemporary potato cultivars ( Tables S5 and S6 in Text S2 ) as follows . For each cultivar , leaflets of a similar age and size were placed in clear plastic boxes ( 26 leaflets per box ) lined with moist tissue paper . After chilling to stimulate zoospore release , a droplet of 30 µl of inoculum ( approx 420 sporangia ) of each of the 26 isolates was applied to the centre of each leaflet . A total of 60 boxes of leaves were inoculated and 30 placed in a randomised block design with six replicate blocks in each of two adjoining illuminated walk-in growth rooms set at a constant 13°C or 18°C with 16/8 hours of light and dark . The 1 , 560 leaflets were scored daily for first symptoms ( i . e . infection period , IP ) , and sporulation ( i . e . latent period , LP ) and at six days post inoculation ( dpi ) , lesion size was measured in two orientations at right angles to each other using electronic calipers connected to a laptop computer . A randomised block field trial comprising four replicate 25 plant plots of the five potato cultivars used in the laboratory assay was established . In mid-July an equal mixture of sporangia of 5 isolates ( different MLGs ) were used to infect the lower leaves of the central plant in each plot . Once the disease had spread from the central plant , single lesions were sampled from the epidemic over the following three weeks and direct SSR fingerprinting of P . infestans from lesions pressed onto FTA cards ( Whatman , UK ) was used to determine the MLG . For additional details see Text S3 . Genome sequencing of P . infestans 13_A2 isolate 06_3928A was performed in 2G GAs ( Illumina Inc . ) and alignments were obtained with Burrows-Wheeler Transform Alignment ( BWA ) software package v0 . 5 . 7 with a seed length ( l ) of 38 and a maximum of mismatches ( M ) allowed of 3 as parameters [52] . Unmapped reads of P . infestans 13_A2 isolate 06_3928A were assembled using VELVET software package v1 . 0 . 18 [53] and mapped to the reference genome using NUCmer program from MUMmer software package v3 . 2 ( see details in Text S3 ) [54] . A False Discovery Rate ( FDR ) analysis was used to determine the performance of single nucleotide polymorphism ( SNP ) calling in the 06_3928A genome ( Figure S14 in Text S1 and Text S3 ) . Single nucleotide polymorphisms ( SNPs ) were called using a 90% consensus among reads calling a SNP with a minimum of 10× coverage ( Figure S16 in Text S1 ) . Rates of synonymous substitution ( dS ) , non-synonymous substitution ( dN ) and omega ( dN/dS ) were calculated using the yn00 program of PAML [55] by implementing the Yang and Nielson method [56] for every coding gene predicted in 06_3928A in comparison to the homologous gene in the reference genome strain T30-4 ( Figure 5 , Table S10 ) . Differences in frequencies of nonsynonymous minus synonymous SNPs were counted per 15 bp-long windows and sliding by 3 bp steps . Frequencies were calculated as the number of SNPs per bp per gene and averaged over 20 consecutive windows ( Figure 6A ) . The 20 windows adjacent to the RXLR motif were considered for each of the domains . Numbers of SNPs in RXLR gene domains were counted per 15 bp-long windows and sliding by 3 bp steps ( Figure 6B ) . A total of 118 RXLRs , 3 , 077 core orthologs and 2 , 442 gene-dense regions ( GDR ) genes that contain at least 1 SNP were analyzed ( Figure 6 ) . The NimbleGen microarray data are available in GEO under accession number GSE14480 for P . infestans T30-4 [23] and GSE33240 for P . infestans 06_3928A and NL07434 . Genes that are induced in planta were identified using a t-test ( p value<0 . 05 , >2 fold expression changes ) and False Discovery Rate ( FDR ) analysis ( q-value<0 . 05 ) [57] in samples from infected potato leaves relative to plate-grown in mycelia ( see more details in Text S3 ) .	We have documented a dramatic shift in the population of the potato late blight pathogen Phytophthora infestans in northwest Europe in which an invasive and aggressive lineage called 13_A2 has emerged and rapidly displaced other genotypes . The genome of a 13_A2 isolate revealed a high rate of sequence polymorphism and a remarkable level of variation in gene expression during infection , particularly of effector genes with putative roles in pathogenicity . Collectively , these polymorphisms , in combination with an extended biotrophic phase , may explain the aggressiveness of 13_A2 and its ability to cause disease on previously resistant potato cultivars . The genome analysis identified conserved effectors that are sensed by potato resistance genes . These findings provide options for the strategic deployment of host resistance with a positive impact on crop yield and food security . This work stresses the benefits of a crop disease management strategy incorporating knowledge of the geographical structure , evolutionary dynamics , genome sequence diversity and in planta-induced effector complement of pathogen lineages .	[ "Abstract", "Introduction", "Results/Discussion", "Materials", "and", "Methods" ]	[ "genome", "expression", "analysis", "plant", "biology", "population", "dynamics", "genome", "sequencing", "genome", "databases", "plant", "science", "plant", "pathology", "mutation", "databases", "population", "biology", "plant", "genetics", "comparative", "genomics", "biology", "disease", "dynamics", "plant", "pathogens", "genetics", "genomics", "genetics", "and", "genomics" ]	2012	Genome Analyses of an Aggressive and Invasive Lineage of the Irish Potato Famine Pathogen
The susceptibility of Anopheles mosquitoes to Plasmodium infections relies on complex interactions between the insect vector and the malaria parasite . A number of studies have shown that the mosquito innate immune responses play an important role in controlling the malaria infection and that the strength of parasite clearance is under genetic control , but little is known about the influence of environmental factors on the transmission success . We present here evidence that the composition of the vector gut microbiota is one of the major components that determine the outcome of mosquito infections . A . gambiae mosquitoes collected in natural breeding sites from Cameroon were experimentally challenged with a wild P . falciparum isolate , and their gut bacterial content was submitted for pyrosequencing analysis . The meta-taxogenomic approach revealed a broader richness of the midgut bacterial flora than previously described . Unexpectedly , the majority of bacterial species were found in only a small proportion of mosquitoes , and only 20 genera were shared by 80% of individuals . We show that observed differences in gut bacterial flora of adult mosquitoes is a result of breeding in distinct sites , suggesting that the native aquatic source where larvae were grown determines the composition of the midgut microbiota . Importantly , the abundance of Enterobacteriaceae in the mosquito midgut correlates significantly with the Plasmodium infection status . This striking relationship highlights the role of natural gut environment in parasite transmission . Deciphering microbe-pathogen interactions offers new perspectives to control disease transmission . Understanding how Plasmodium-Anopheles interactions contribute to the mosquito vector competence has received great attention lately , and the increasing knowledge promises to contribute to the development of new malaria control strategies [1]–[3] . Malaria still remains a serious health problem in developing African countries , causing more than 1 million deaths annually . Almost all these deaths are caused by the parasite Plasmodium falciparum whose major vector in Africa is Anopheles gambiae , which is widely distributed throughout the afro-tropical belt . A . gambiae s . s . is divided into two morphologically indistinguishable molecular forms , known as M and S , which are regarded as incipient species [4]–[6] . The M and S molecular forms exhibit ecological preferences [7] , [8] , but their respective epidemiological importance in malaria transmission has been poorly documented so far [9] , [10] . The susceptibility of Anopheles mosquitoes to Plasmodium infection is under genetic control [11]–[13] , but the large variability in oocyst number among closely related mosquitoes indicates that environmental factors also play a role . Multiple lines of evidence suggest that mosquito bacterial communities influence vector competence [14]–[17] . A protective role of Anopheles midgut bacteria against malaria infections was demonstrated by using antibiotic treatment to clear the gut microbiota , which resulted in enhanced Plasmodium infections [15] , [18] . Consistently , coinfections of bacteria with Plasmodium reduced the number of developing oocysts in the mosquito midgut , both in laboratory and field conditions [15] , [19] , [20]–[24] . Interestingly , Cirimotich et al . [24] recently described an Enterobacter bacterium isolated from wild mosquitoes in Zambia that confers refractoriness to P . falciparum infection . Mechanisms mediating this refractory phenotype remain elusive . Instead of eliciting immune responses leading to reduced levels of parasite burden , the experiments conducted by Cirimotich et al . [24] revealed that the inhibition of Plasmodium development by commensal microbiota occurs through production of reactive oxygen species by the Enterobacter bacteria that directly target Plasmodium parasites in the midgut [17] , [24] . Bacterial diversity in the Anopheles species is thought to be particularly low at the adult stage because of gut renewal during metamorphosis from pupae to adults . Nevertheless several bacterial species have been identified in the adult mosquito midgut using different conventional culture-mediated techniques [16] , [23] , [24] . These bacteria were acquired from the aquatic environment during immature stage development [25] , [26] , although vertical transmission ( from mother to offspring ) also has been documented [14] , [16] , [25] , [27] . Generally , knowledge on mosquito midgut bacterial communities remains largely unknown , mostly because of the limitations of isolating techniques based on culturing and to the low resolution of fingerprint analysis . However , the recent deployment of next generation DNA sequencing technologies has provided new opportunities to explore microbial diversity of complex environments [28]–[31] as well as to further investigate disease susceptibilities and host-bacteria-pathogen interactions [32]–[34] . In this study , we performed a meta-taxogenomic analysis of microbial communities in the midguts of adult mosquitoes originating from natural larval habitats in Cameroon . We further investigated correlations between midgut microbiota and the mosquito malaria infection status . Previous investigations of bacteria-Plasmodium interactions in the mosquito vector have considered laboratory-reared mosquitoes challenged with cultured bacteria and infected with a cultured P . falciparum line . Here , we challenged wild female mosquitoes with a natural isolate of P . falciparum , thereby offering an opportunity to examine natural bacteria/parasite associations . Our analysis revealed that the midgut bacterial diversity represents an important force shaping the mosquito vector competence , where bacteria of the Enterobacteriaceae genera benefit P . falciparum development . A total of 92 A . gambiae mosquitoes collected at the larval stage and reared to adults in the insectary were successfully fed through membrane feeders on gametocyte containing blood from a single individual . The origin of mosquitoes and their genetic characteristics ( molecular form and infection status ) are summarized in Table S1 . Mosquitoes of M molecular form were significantly more infected than those of S form ( 48 . 5% versus 27 . 1%; OR 0 . 39; 95% CI: 0 . 15–1 . 06; P = 0 . 044 ) . However , the comparison of infection prevalence ( number of infected mosquitoes ) between the different localities revealed a “sampling site effect” ( Fisher's exact test P<0 . 01 ) . Of those challenged with P . falciparum we then investigated the gut microbiota in mosquitoes originating from two different breeding sites . We used field mosquitoes from Mvan and Nkolondom and gut bacterial communities determined for 8 P . falciparum-PCR positive and 7 negative mosquitoes from each locality ( Table S2 ) . Pyrosequencing of 16S rDNA generated a total number of 663 , 798 sequence reads across the 3 hypervariable regions S1 , S2 , and S3 in the bacterial gene in 32 mosquitoes ( Table 1 ) . Few individuals failed to amplify the SSU regions ( 2 for S1 , 3 for S2 , and 5 for S3 ) , which was not linked to DNA quality as at least one region was successful for all samples , making it likely that technical problems in the PCR were responsible . After tag extraction and filtering of low-quality sequence tags , we obtained 575 , 284 reads for the analysis , representing 86 . 7% of the 454 reads . The average number of tags for all SSU regions combined per sample was 6 , 827 ( ±811 ) , read number per gut ranging from 3 , 305 to 10 , 169 . About 99 . 0% of sequence reads were successfully assigned , with unique tags representing 25 . 4% of the average tag number over the three SSU regions . We first compared the pyrosequencing data for the 23 gut samples that yielded sequence tags for all three 16S domains ( Table 1 ) . The comparison of the microbial communities between the 3 domains for seven midgut samples is given in Figure 1 . The three 16S domains overall provided a very similar picture of the bacterial populations , even if they differed for the exact percentages . When only the most abundant taxonomic categories were considered ( constituting >2% of the overall ) , the S1 domain reaches a lower percentage , indicating that this 16S library capable of identifying a greater number of minor clades ( Figure 1 ) . In addition , the S1 domain had better resolution , allowing more precise assigning of sequence tags ( see Figure 1 , mosquito NKD97 ) . We then performed further analyses on the S1 domain , for all 30 samples that were successful for pyrosequencing . The bacterial communities of the mosquito midgut belonged to 26 different phyla , among which , 5 represented more than 99% of the total microbiota: Proteobacteria , Bacteroidetes , Actinobacteria , Firmicutes , and Fusobacteria . We examined the relative abundance of the major classes , that is , detected in more than 30% of the samples and having an average abundance of >0 . 1% ( Figure 2 ) . The gut microbiota presented a large inter-individual variability and was dominated by few taxa . The first striking result came from laboratory-reared mosquitoes that exhibited a drastically different composition of midgut bacteria from field mosquitoes . More than 96% of tags corresponding to bacteria in the midguts of mosquitoes from the Ngousso colony were assigned to Flavobacteria , although this class accounted for only 0 . 38% ( ±0 . 24 ) of the sequence tags in field mosquitoes . Similarity searches against the SSU SILVA database ( release 108 ) identified Flavobacteria tags as belonging to Elizabethkingia sp . , which already has been isolated from A . gambiae midguts from different insectaries [15] , [35] , [36] . The guts of Ngousso mosquitoes also contained , to a lower extent , Gammaproteobacteria ( Pseudomonas sp . ) and Alphaproteobacteria ( Asaia sp . ) . In mosquitoes from natural habitats , midguts were mainly colonized by Proteobacteria ( 94% ) , and the most prominent classes were Gammaproteobacteria , Alphaproteobacteria and Betaproteobacteria ( Figure 2 ) . Figure 1 clearly shows the difference in bacterial composition between mosquitoes from the 2 breeding sites used in this study . In mosquitoes originating from Nkolondom , the intestinal bacterial flora is dominated by Alphaproteobacteria ( 68 . 65±7 . 38% ) , mainly of the genus Asaia sp . By contrast , the three major classes are almost equally represented in the midgut of mosquitoes from Mvan , although with large individual variability . The main taxa in this locality were Asaia , Sphingomonas , Burkholderia , Ralstonia , and Enterobacteriaceae . Enterobacteriaceae could not be assigned to more precise taxonomic ranks . Midgut bacteria were unevenly distributed among individual mosquitoes and between the different sampled localities . Several genera were found in all , or at least in a large majority , of the mosquitoes possibly representing the “mosquito midgut core microbiota” ( Table S3 ) . They included members of the genera Asaia , Burkholderia , Serratia , Ralstonia , Acinetobacter , Pseudomonas , Sphingomonas , Staphylococcus , Streptococcus , and Escherichia/Shigella . Of note , Asaia sp . was found in all samples , and its relative abundance showed great variation from one midgut to another , ranging from 1 . 49 to 98 . 95% in mosquitoes from Nkolondom and from 0 . 04 to 49 . 66% in those from Mvan . The group of unassigned Enterobacteriaceae also was identified in all field samples , with relative abundance ranging from 0 . 01 to 1 . 03% and from 0 . 04 to 71 . 51% in Nkolondom and Mvan , respectively . Other specific members of Enterobacteriaceae were frequent and represent a large proportion of the midgut bacterial communities: Serratia spp . accounted for 96 . 93% of the midgut bacteria in a mosquito ( NKD97 ) from Nkolondom , and Cedecea spp . encompassed 12% of the gut bacterial content in 2 mosquitoes from Mvan . Escherichia/Shigella was found in more than 85% of the mosquitoes , at low densities . The sequence of the Esp_Z Enterobacter ( JF690924 ) , despite its presence in our reference database , was absent from the analyzed reads . Of note , the midgut bacterial flora was mainly composed of Gram-negative communities . No Gram-positive bacterium was identified in the laboratory mosquitoes , whereas they represented 5% of the total microbiota of the field mosquitoes . Gram-positive bacteria belonged to the classes Bacilli and Actinobacteria . To determine whether all phylotypes present in the mosquito midgut microbiota were detected in this study , we performed a rarefaction analysis for each sample on tags from the S1 domain; rarefaction curves are shown in Figure S1 . The rarefaction curves decrease rapidly at approximately 2 , 000 sequences per sample and reach saturation at 3 , 000 , indicating that our sequencing effort was sufficient to catch the overall bacterial diversity in the mosquito midgut . The rarefaction curves show the large variability in bacterial complexity among samples , varying from 13 to 340 operational taxonomic units ( OTUs ) . In addition , they illustrate the paucity of clusters in the midgut of laboratory mosquitoes . They also revealed greater bacterial diversity in the samples from Mvan compared with those from Nkolondom ( 185±51 and 110±30 , respectively; t-test t = 2 . 385 , P = 0 . 025 ) . The Chao1 estimator , which gauges the number of unseen “species , ” predicted that we covered , on average , 81% of the species diversity across all samples . To confirm this result , we computed the ACE and Jackknife estimator indexes; both had higher values than the observed richness , indicating an underestimation of the gut microbial diversity ( see Table S2 ) . We characterized the species diversity in our set of mosquito midguts using the species richness , the Shannon diversity index ( H ) and the Simpson's diversity index ( D ) ; data are shown in Table S2 . No significant differences in the richness index were found comparing mosquito locality and/or P . falciparum prevalence . Significant differences of the diversity indexes were found when comparing mosquito locality ( Shannon , P = 0 . 0091 and Simpson , P = 0 . 0097 ) but none when comparing the infection prevalence of mosquitoes . Thus , at the genus level , the microbiota of mosquitoes from Mvan was more diverse than that from Nkolondom , but Plasmodium-infected and non-infected mosquitoes did not exhibit differences in their microbial diversity . The relationship between the class taxonomic rank of bacteria and the origin of the mosquitoes ( locality ) was evaluated using redundancy analyses ( RDA ) ( Figure 3 ) . The Eigen values of the first four axes were recorded at 0 . 304 , 0 . 214 , 0 . 331 , and 0 . 144 , respectively . The first two constrained axes explained around 50% of the total variance in the bacterial community and 100% of the species-environment relationship . The unrestricted Monte Carlo permutation test ( n = 499 ) indicated that all environmental variables were significant ( Nkolondom variable: F = 10 . 99 , P = 0 . 002; Mvan: F = 13 . 26 , P = 0 . 004; Mvan and laboratory variables fit collinearly ) . Thus the different classes of bacteria were not randomly distributed but linked to the breeding site where mosquitoes grew up . As seen in Figure 2 , Flavobacteria were related to laboratory mosquitoes , whereas Alphaproteobacteria were less diverse and related to the Nkolondom locality . The remaining classes cluster along the Mvan locality . These results confirm the higher diversity of bacterial taxa in mosquitoes collected in Mvan as compared with Nkolondom , and the paucity of the gut microbiota in laboratory-reared mosquitoes as compared with mosquitoes from the wild . We then investigated potential relationships between the gut microbial communities of field mosquitoes and the P . falciparum infection status . We performed the RDA by plotting the infection status and the origin of the field mosquitoes against the family taxonomic rank , allowing the analysis of more precise bacterial taxa ( Figure S2 ) . The first and second constrained axes corresponded to 35% and 7% of the total variance in the bacterial community , respectively , and explained all the cumulative percentage variance of the family-environment relationship . All environmental variables were significant ( Monte Carlo test , Nkolondom: F = 14 . 02 , P = 0 . 002; collinearity detected with Mvan variables; infection variable F = 3 . 00 , P = 0 . 042 ) . The first axis alone explained 84 . 1% of the variance of the family environment relationship and was related to the mosquito origin ( Mvan and Nkolondom ) . In concordance with the results already described for the Alphaproteobacteria class , the Acetobacteriaceae family was related to mosquitoes from Nkolondom , and most of the family is represented by Asaia spp . By contrast , the mosquitoes from Mvan exhibited a larger bacterial diversity . Interestingly , the RDA revealed a relationship between the Enterobacteriaceae family and the infection status along axis 2 ( Figure S2 ) . This result suggests that mosquitoes harboring Enterobacteriacae are more likely to be infected by P . falciparum . A correlation between the relative abundance of Enterobacteriaceae in the midgut and P . falciparum infection was further detected using the non-parametric Mann-Whitney test ( P = 0 . 004; Figure 4 ) , indicating that P . falciparum-positive mosquitoes were hosting more Enterobacteriaceae bacteria . We provide here an in-depth description of the microbial communities in the midgut of the malaria mosquito . Using pyrosequencing , we explored individual midgut samples from adult female mosquitoes collected at the larval stage in different natural environments , exposed to P . falciparum infection at day 5 after emergence and dissected 8 days after the infectious blood meal . We then examined the microbial diversity according to the origin of the mosquitoes and investigated putative correlations between the bacterial content and the malaria infection status by comparing midgut microbiota in P . falciparum-positive and P . falciparum-negative individuals . The adult mosquito midgut microbiota comprises five dominant phyla Proteobacteria , Bacteroidetes , Actinobacteria , Fusobacteria , and Firmicutes and presents some similarities with gut microbial communities from other invertebrate midguts , including mosquitoes [37]–[44] . Nonetheless , pyrosequencing revealed a higher diversity than more conventional molecular techniques , with an average of 147 . 64 OTUs ( ±88 . 49 , at a 0 . 04% occurrence threshold ) and an estimated richness of 72 . 27 ( ±31 . 70 ) taxa per field mosquito . Although bacterial richness is greater than previously described in mosquitoes , the vast majority of sequence tags ( >90% ) felt into few taxa and only 21 bacterial families , and 28 genera had an abundance of >1% in at least one mosquito midgut . Thus , the mosquito midgut is colonized by few dominant bacteria species , probably involved in metabolic functions . We used three different pairs of primers ( S1 , S2 , and S3 ) to amplify and analyze each midgut sample . The comparison of the bacterial diversity for the three libraries revealed similar patterns; however , the S1 library allowed more detailed identification of the mosquito microbiota . These data strengthen the previous observation that the SSU rRNA gene clone libraries are biased by the choice of the set of primers used for amplification and thereby distort the revealed biodiversity [45] . To our knowledge , this is the first 454 sequencing analysis where different couples of primers have been used to identify the bacterial diversity in biological samples . The analysis ensured that the primer sets used produced the most accurate view possible of the bacterial content of the mosquito midgut . Proteobacteria represented more than 90% of the bacterial gut content in the mosquitoes from the wild , whereas in laboratory-reared mosquitoes , more than 95% of sequence tags belonged to the Flavobacteria Elizabethkingia spp . The remaining tags from laboratory mosquitoes were assigned to the members of Gammaproteobacteria ( Acinetobacter , Pseudomonas ) , Firmicutes ( Staphylococcus , Streptococcus ) , and the Alphaproteobacterium Asaia sp . Bacterial richness and diversity seem to be particularly poor in the laboratory mosquitoes . We identified the Elizabethkingia spp . in 68% ( 19/28 ) of the field-collected mosquitoes , at low densities ( <0 . 5% of the total bacterial content ) , suggesting either that the bacterium has developed symbiotic associations with the mosquito midgut or that the bacterium is widespread in nature . The predominance of Elizabethkingia spp . in the midguts of the insectary-reared mosquitoes reflects that the bacterium has found a thriving niche in this environment where competition with other bacterial species is limited . Bacteroidetes are known as glucose degraders , and the large dominance of Elizabethkingia spp . in laboratory-reared mosquitoes is probably due to the mosquito food source [46] . Indeed , the midgut microbial diversity is directly associated with the individual diet [47]–[50] . In this study , all adult mosquitoes were maintained in our standard rearing conditions on a sterile glucose solution . The aquatic environment of the larval stages presented striking differences: larvae of the Ngousso strain were grown in clean spring water , whereas the immature stages of field mosquitoes were collected in natural breeding sites , water puddles , and flooded areas rich in biotic and abiotic components . Thus , our results indicate that the environmental conditions of the vectors are key determinants in shaping midgut microbiota . The drastic loss of microbial diversity from the wild to laboratory conditions may have important consequences on mosquito fitness and on the gut immune system . This undoubtedly explains the higher prevalence and intensity of P . falciparum infections in laboratory colonies of A . gambiae as compared with field-derived mosquitoes and pinpoints the limitations of using laboratory models to study host-pathogens interactions ( Morlais and Cohuet , unpublished ) . The great difference in the composition of gut bacteria between laboratory and field-collected mosquitoes as well as between mosquitoes originating from distinct breeding sites shows that most bacteria are commensally acquired from the environment . Field mosquitoes were sampled in their breeding sites at the larval stage and maintained in their aquatic habitats until adult emergence . We propose that the acquisition of endobacteria occurred from the aquatic environment , and possibly by vertical transmission routes . Indeed transstadial transmission has been demonstrated in Anopheles mosquitoes [27] , [51] , and despite of “gut sterilization” during mosquito metamorphosis from pupae to adult , which is believed to contribute to a reduction of the larval microbiota [52] , the bacterial clearance is not complete . Here , we show that the bacterial content of adult mosquitoes differed according to the breeding site where larvae were grown , and our results suggest that the composition of the midgut microbiota in adult mosquitoes relies on the bacterial richness of the native aquatic source . The 454 sequencing allowed the identification of both commensal and symbiotic bacteria , giving a broad description of the mosquito midgut microbial community . Bacterial taxa , such as Asaia or Burkholderia , are known insect symbionts , contributing to beneficial associations and possibly to an enhanced pathogen resistance [35] , [53]–[55] . We identified Asaia spp . as a predominant component of the gut microbiota in the mosquitoes from Nkolondom , representing more than 60% of sequence tags , but these bacteria also were found at lower abundance in all other mosquitoes even in those from the laboratory colony , which is indicative of a positive effect of this bacteria on mosquito fitness . Transmission of Asaia from adult to offspring occurs through an egg-mediated mechanism , but other modes of transmission , including contamination through the food source , have been described [35] , [51] , [56] . Several strains of Asaia colonize mosquito populations , including symbiotic and environmental isolates that follow distinct routes of transmission [35] . The difference of Asaia abundance in mosquitoes sampled in our two study sites , Nkolondom and Mvan , possibly underlies a genetic heterogeneity of the bacterium in the different environmental settings . By contrast to Asaia , Burkholderia spp . that were the dominant genera of the midgut microbiota in mosquitoes from Mvan , representing an average of 30% of sequence abundance , were not detected in the intestinal flora of the Ngousso colony . Thus , the infection by Burkholderia is not essential for growth and reproduction of the mosquito . Members of the genus Burkholderia are widespread in soil rhizospheres and plant surfaces , and some species are known to be associated with insects feeding on plants [41] , [54] , [57] , [58] . In the latter case , the Burkholderia symbiont is environmentally acquired by the nymphs [54] . Studies on the association between Burkholderia and the insect midgut revealed mutualistic relationships , where the symbiont presence increases the insect fitness or protects the insect from entomopathogenic fungi [54] , [55] . So far , as we know , the effect of the Burkholderia symbiont on malaria vectors is unknown . Further investigations on the microbiota dynamics through the mosquito life cycle , from egg to adult , are required to better define the nature of the microbe-insect associations and the most important microbial species critical for mosquito survival . Despite a larger diversity of the gut microbiota in wild mosquitoes , most bacteria species are sparsely distributed between individual mosquitoes . Only 20 genera were found in more than 80% of individuals and 60 in >50% . In insects , the gut microbiota differs according to the food source , and in blood-sucking insects , bacterial content is higher after a blood meal [15] , [23] , [59] , [60] . Here , because adult mosquitoes were fed the same diet , the high variability of taxa abundance results from individual variation , and the most abundant lineages represent the mosquito “core gut microbiota . ” The existence of a core gut microbiota , by which different bacteria species are sharing metabolic functions and maintain the gut homeostasis , is now emerging [34] , [61] , [62] . Because alteration of the microbiota composition has been related to the development of diseases or health disorders , the next challenge is to define members of the microbial community and/or the metabolic interdependencies essential to preserve optimal gut homeostasis [34] , [63] . The characterization of the mosquito core microbiota during the time course of Plasmodium infection will be a next step toward understanding the impact of gut bacteria on parasite development within the mosquito midgut . P . falciparum traverses the intestinal epithelium within 24 h after blood meal , at the peak of the digestion process; and whether parasites take advantage of intense competitive interactions for nutrient resources between bacteria to thwart the immune surveillance has to be investigated . Indeed , the gut microbiota is known to play an important role in protecting the host from potentially pathogenic microbes [64] , [65] . Protection occurs through different processes: stimulation of the mosquito immune response , competition for binding sites or nutrients and production of toxins [65]–[67] . However , despite the beneficial role of the microbiota , pathogens , such as helminthes and viruses have developed strategies for exploiting the gut microbiota to promote their transmission [68] , [69] , [70] . For the mosquito vector , our understanding is still at an early stage for how the natural resident microbial flora of the mosquito midgut contributes to its resistance to the Plasmodium [15] , [17] , [18] , [24] . In this study , we found that the abundance of Enterobacteriaceae is higher in P . falciparum-infected mosquitoes , suggesting that some microbe-parasite interactions may contribute to the successful development of the malaria parasite . However , whether Enterobacteriaceae have an effect on parasite survival or whether the increased level of Enterobacteriaceae is a consequence of Plasmodium development remains elusive . Alternatively , genetic factors , such as allelic polymorphism of immune genes , could regulate the variable levels of permissiveness of the mosquitoes as has been previously shown [71] . In contrast to our findings , previous studies reported the deleterious effect of bacterial infections on Plasmodium development in the mosquito [19] , [23]–[25] . Of interest in this context , several Enterobacteriaceae strains were able to inhibit the development of Plasmodium species in the mosquito midgut , among them Cedecea spp . , Serratia spp . , and Enterobacter spp . isolated from A . albimanus , A . stephensi , or A . arabiensis [19] , [22]–[24] . The Esp_Z Enterobacter strain isolated from A . arabiensis caught in Zambia [24] was not identified in any of the reads we analyzed . The possibility that this Enterobacter strain would have been absent from the PCR products because of competition with a different clade is unlikely as we used three different sets of primers . Therefore , we expect that in the gut of A . gambiae mosquitoes in Cameroon , the Esp_Z Enterobacter strain was below the 0 . 1% abundance threshold or absent . This Enterobacter strain was isolated in Zambia from wild-caught A . arabiensis mosquitoes , and differences in the mosquito species , as well as differences between the study areas , may explain why we did not find this bacterium in our material . Cirimotich et al . [24] recovered the Esp_Z on LB media , and culturing methods can lead to the artificial amplification of a bacterial strain present in minute amounts in an environmental sample . Therefore , it would be of interest to examine the presence/abundance of Esp_Z in wild-caught Zambian A . arabiensis using the methodologies described here . In our study , we analyzed the gut resident microbiota and revealed a positive correlation between commensal Enterobacteriaceae and Plasmodium infection , indicating that the P . falciparum infection phenotype under natural conditions results from more complex interactions than previously thought . Our data suggest a possible protective role of the Enterobacteriaceae on natural P . falciparum infection . Interestingly , it has been shown that commensal Enterobacteriaceae may promote intestinal homeostasis by enhancing immune receptors in the human colon [72] . For the mosquito , as described for the insect model Drosophila [73] , gut homeostasis could be maintained through the renewal of the intestinal epithelial layer that can be altered upon bacterial killing or through immune regulation . A major challenge now will be to correlate our data with quantitative phenotyping of the immune system of the gut . Previous reports on the susceptibility of the M and S molecular forms to P . falciparum infections relay contrasting findings [9] , [10] . In Cameroon , mosquitoes of the two molecular forms collected in a sympatric area exhibited similar susceptibility to P . falciparum infection [10] , whereas in Senegal , mosquitoes of the S form , derived from progenies of field-collected individuals , were more susceptible to P . falciparum than those of the M form [9] . In the present study with the mosquitoes collected in natural breeding sites and infected on the same blood donor , we found that the M form was more infected than the S form . However , a marked difference in the P . falciparum prevalence was observed according to the sampling site , and larger sample sizes of sympatric M and S populations of A . gambiae will be needed to further explore any difference of Plasmodium susceptibility between the two cryptic species . We propose that the composition of the gut microbiota may influence parasite transmission , which would explain the difference in infection levels between mosquito populations from diverse environments [9] , [74] . The mosquito susceptibility to Plasmodium infection is under host genetic control , and several candidate genes have a recognized role in the establishment of the pathogen in the mosquito midgut [11]–[13] . However , how the mosquito gut microbiota influences Plasmodium transmission has to be unraveled . The mosquito gut ecosystem remains poorly understood , and elucidating the precise role of the symbiotic and commensal flora on the regulation of the insect immune response and on the infection course of pathogens , such as Plasmodium parasites , will be of great interest . Pathogens and microbes likely depend on similar mechanisms for interacting with their hosts , and a better knowledge of the mosquito-microbiota interactions would open new avenues for vector disease control through manipulation of gut microbial communities . Furthermore , unraveling these strategies mounted by the parasites to cooperate with the resident microbiota will allow a better understanding of co-evolution of host-pathogen interactions . All procedures involving human subjects used in this study were approved by the Cameroonian national ethical committee ( statement 099/CNE/SE/09 ) . The gametocyte carrier used in this study was enrolled as a volunteer after his parents had signed a written informed consent . A . gambiae mosquitoes were sampled in aquatic habitats at the L4 and pupae stages in four localities in Cameroon using standard dipping technique [75] . In each locality , breeding sites were inspected visually for presence of larval stages . At each breeding site , 10 dips were taken with a standard dipper ( 300 ml ) and kept in a 5-liter container for transportation to the insectary at OCEAC . Anopheline larvae were identified morphologically; non-anopheline larvae and predators were removed . Larvae were kept in their original habitat water in a 3-liter plastic bucket and resulting pupae were collected daily for 2 days . Pupae were transferred to a plastic cup containing 20 ml of water from the breeding site , and the cup was placed in a 30×30 cm cage for emergence . The remaining larval collection was discarded after 2 days to avoid bias because of putative modifications of the biotic content of the aquatic habitats . Adult mosquitoes were maintained in standard insectary conditions ( 27±2°C , 85±5% RH , and 12 h light/dark ) and provided with 8% sterile sucrose solution . Female mosquitoes were fed on a single P . falciparum gametocyte carrier to avoid infection rate variability because of the blood donor . Infectious feeding was performed as previously described [71] , [76] . Females , 3 to 5 days old , were starved for 24 h and allowed to feed on the P . falciparum gametocyte containing blood for 35 minutes through membrane feeders . Unfed and partially fed mosquitoes were removed by aspiration and discarded . Fully engorged females were kept in the insectary until dissections 8 days after the infectious blood meal . Mosquitoes were surface sterilized in 70% ethanol for 5 minutes , then rinsed twice in sterile PBS solution , and midguts were dissected and stored individually at −20°C until processing . DNA was extracted using the DNeasy Blood &Tissue Kit from Qiagen ( Valencia , CA ) and quantified ( Nanodrop ND-1000 , NanoDrop Technologies , Montchanin , DE , USA ) . A 20-ng aliquot of DNAs was subjected to whole-genome amplification using the GenomiPhi V2 DNA Amplification Kit ( GE HealthCare , Uppsala , Sweden ) , and the GenomiPhi templates served to characterize molecular forms of A . gambiae and the P . falciparum infection status . Molecular forms were determined according to Fanello et al . [77] and the identification of mosquitoes that successfully developed malaria infection using a P . falciparum specific PCR amplifying a Cox gene fragment [78] . A total of 32 individual midguts were subjected to the 454-sequencing analysis . We included 2 samples of midguts dissected from mosquitoes of our local colony of A . gambiae , Ngousso . The Ngousso colony was established in January 2006 from larvae collected in breeding sites of an urbanized district of Yaounde , “Ngousso . ” Larval collections were conducted during a 2-month period; mosquitoes were blood fed for oviposition and then PCR screened for molecular form of A . gambiae . Ngousso mosquitoes belong to the M molecular and Forest chromosomal forms . Since then , the colony has been routinely maintained at the OCEAC insectary . The Ngousso samples served to provide an overview of the bacterial content of laboratory mosquitoes reared under standard insectary conditions and as an experimental control in this study . The 30 remaining samples were chosen among field mosquitoes fed on blood from the same gametocyte carrier . We selected both P . falciparum positive and P . falciparum negative midguts to assess putative differences of microbiota between non-infected and infected individuals in our P . falciparum-challenged mosquitoes . For each individual midgut DNA sample , we generated three PCR amplicon libraries . We targeted 3 different hypervariable regions of the 16S ribosomal RNA to allow accurate detection of the bacterial communities in each sample . Indeed , previous analyses showed that the set of primers used for amplification can have a strong impact on the biodiversity revealed; some abundant clades in a given sample could be foreseen , depending on the primers used [45] . The S1 library targeting the V4 hypervariable region was obtained using the forward primer 535F ( 5′-GTGCCAGCAGCCGCGGTAATA-3′ ) and the reverse primer 789R ( 5′-GCGTGGACTACCAGGGTATCT-3′ ) , the S2 library for the V5–6 region using the 326F ( 5′-CAAACAGGATTAGATACCCTG-3′ ) and the 1082R ( 5′-CGTTRCGGGACTTAACCCAACA-3′ ) primers , and the S3 library targeting the V5–6 region with the 1065F ( 5′-CAGGTGCTGCATGGCYGTCGT-3′ ) and the 1336R ( 5′-CGATTACTAGCGATTCC-3′ ) primers . Amplified DNA was purified and quantified using Picogreen fluorescent dye ( Molecular Probes , Eurogen , OR ) . Individual libraries were processed for 454 sequencing by ligating the 454 adapters coupled to MID tags , allowing the multiplexing of samples . The amplicon libraries were pooled in two separate batches and sequenced . The MIDs and 454 linkers were ligated after the PCR amplification and the pyrotags were sequenced unidirectionally . Pyrosequencing was performed at Genoscreen ( Lille , France ) using a Genome Sequencer FLX Titanium ( GS-FLX ) system ( Roche , Basel , Switzerland ) . In total , we recovered 663 , 651 sequence reads ( tags ) that were subjected to quality controls . All 454 sequences were deposited in Genebank ( SRS281724 . 1 and SRS281725 . 1 ) . Tags were extracted only if they contained the combination linker-MID-primer and the complement primer sequence at the 3′end . Tags were sorted in appropriate files according to their MID barcode and converted to the forward strand when necessary . A strict dereplication step was then applied that discarded tags with unidentified nucleotides ( Ns ) and those longer than 350 bp or shorter than 200 bp . Dereplicated tags were sorted by decreasing number of occurrences and clustered at k = 3 number of differences as described in Stoeck et al . [79]; this pipeline resulted in determining unique sequences . We next processed taxonomic assignments by implementing a new approach that clearly optimizes the successive assignments . We first extracted from the reference sequences of SILVA ( release 106 ) domains corresponding to the various possible couples of the primers . Extraction was first performed requiring a perfect match between each primer and a sequence and , when no match was found , 1 , 2 and 3 differences between each primer and a sequence were used successively . This pipeline then gave three reference databases , one per amplified 16S rDNA region , containing all reference amplicons putatively matching our tags . The S1 , S2 and S3 tagged databases contained 424 . 634 , 359 . 198 and 394 . 370 reference amplicons , respectively . In a second step , all unique tags were assigned a taxon using a global alignment method . Each amplicon of the reference database was considered if it had at least 70% similarity with a tag . The list of reference amplicons was sorted by decreasing percentage of similarity and rounded to an integer . For taxonomic assignments , the reference sequence with the highest percentage was used , and taxonomy to a given level was obtained by the consensus of these taxonomies when more than one result emerged . For example , a tag with 98% similarity to the class Gammaproteobacteria and Alphaproteobacteria was only assigned to the phylum Proteobacteria . When similarity was <80% , sequences were not assigned . Tags were clustered into OTUs according to their consensus taxonomy . For each mosquito sample and each amplified 16S rDNA region , OTU abundances represent relative abundances , the number of reads for the given OTU divided by the total of tags in the SSU region of that sample . Rarefaction curves were produced by plotting the number of unique sequence tags as a function of the number of randomly sampled tags . To generate rarefaction curves , we retained OTUs containing at least 2 sequence tags and encompassing the abundance threshold of 0 . 04% because rare sequences likely represent random sequencing errors and overestimate the overall diversity . Ecological indexes such as richness and diversity indexes ( Simpson , Shannon ) , were computed using the Vegan [80] and BiodiversityR [81] packages under the R software ( available at http://www . R-project . org ) [82] . Chao1 , ACE1 and Jackknife richness estimators were calculated using the SPADE software [83] . Indexes were calculated using values from the genus taxonomic rank , the lowest rank obtained with the 454 technology . Association between microbiota and environmental variables was assessed using a multivariate ordination test . We defined the different taxa present in the data set as “species variables” and the origin of the mosquitoes and the P . falciparum infection status as “environmental variables” for each individual . A detrended canonical correspondence analysis ( DCA ) was performed to determine the ordination method suitable for our data . The longest gradient we obtained was shorter than 3 . 0 , indicating that the constrained form of linear ordination method , the RDA , was the most appropriate test [84] . Redundancy analysis was performed using Canoco v . 4 . 5 Software [85] . The environmental variables were set as dummy variables ( 0 or 1 values ) . RDA and associated Monte Carlo permutation tests ( n = 499 ) were used to identify the measured environmental variable that contributed most significantly to the variation in the bacterial community data . The Monte Carlo test returns a p value associated with the effect of the environmental variable on the microbiota composition of the samples . Results were visualized on a biplot ordination diagram using CanoDraw extension .	During their development in the mosquito vector , Plasmodium parasites undergo complex developmental steps and incur severe bottlenecks . The largest parasite losses occur in the mosquito midgut where robust immune responses are activated . Variability in P . falciparum infection levels indicates that parasite transmission is the result of complex interactions between vectors and parasites , which rely on both genetic and environmental factors . However , in contrast to genetically encoded factors , the role of environmental factors in parasite transmission has received little attention . In this study , we characterized the midgut microbiota of mosquitoes derived from diverse breeding sites using pyrosequencing . We show that the composition of the midgut microbiota in adult mosquitoes exhibits great variability , which is likely determined by bacterial richness of the larval habitats . When field mosquitoes were collected at late immature stages in natural breeding sites and the emerging females challenged with Plasmodium falciparum in the laboratory , significant correlation was observed between P . falciparum infection and the presence of Enterobacteriaceae in the mosquito midgut . Greater understanding of these malaria-bacteria interactions may lead to novel malaria control strategies .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "ecology", "genetics", "immunology", "biology", "genomics", "evolutionary", "biology", "microbiology", "population", "biology", "genetics", "and", "genomics" ]	2012	Midgut Microbiota of the Malaria Mosquito Vector Anopheles gambiae and Interactions with Plasmodium falciparum Infection
The mosquito Aedes aegypti is the primary vector of dengue virus ( DENV ) infection in humans , and DENV is the most important arbovirus across most of the subtropics and tropics worldwide . The early time periods after infection with DENV define critical cellular processes that determine ultimate success or failure of the virus to establish infection in the mosquito . To identify genes involved in these processes , we performed genome-wide transcriptome profiling between susceptible and refractory A . aegypti strains at two critical early periods after challenging them with DENV . Genes that responded coordinately to DENV infection in the susceptible strain were largely clustered in one specific expression module , whereas in the refractory strain they were distributed in four distinct modules . The susceptible response module in the global transcriptional network showed significant biased representation with genes related to energy metabolism and DNA replication , whereas the refractory response modules showed biased representation across different metabolism pathway genes including cytochrome P450 and DDT [1 , 1 , 1-Trichloro-2 , 2-bis ( 4-chlorophenyl ) ethane] degradation genes , and genes associated with cell growth and death . A common core set of coordinately expressed genes was observed in both the susceptible and refractory mosquitoes and included genes related to the Wnt ( Wnt: wingless [wg] and integration 1 [int1] pathway ) , MAPK ( Mitogen-activated protein kinase ) , mTOR ( mammalian target of rapamycin ) and JAK-STAT ( Janus Kinase - Signal Transducer and Activator of Transcription ) pathways . Our data revealed extensive transcriptional networks of mosquito genes that are expressed in modular manners in response to DENV infection , and indicated that successfully defending against viral infection requires more elaborate gene networks than hosting the virus . These likely play important roles in the global-cross talk among the mosquito host factors during the critical early DENV infection periods that trigger the appropriate host action in susceptible vs . refractory mosquitoes . Dengue virus ( DENV ) represents a significant challenge for global public health where 2 . 5 billion people are estimated to be at risk of dengue related diseases [1]–[3] . The mosquito Aedes aegypti is the primary global vector of DENV . There are no effective vaccines or treatments available , with mosquito control remaining the only viable strategy for disease prevention . The spread of DENV is critically dependent upon successful completion of viral life cycles in the infected mosquito [4] . Understanding the basic mechanisms of how the mosquito successfully transmits DENV is a first requirement towards designing novel genetic control strategies . Upon mosquito blood feeding on a viremic human , DENV enters the mosquito mid-gut with the blood meal where it must establish an infection in mid-gut epithelial cells , the success of which is required for subsequent completion of the viral life cycle in the mosquito . The intrinsic ability of A . aegypti to host the virus is generally referred to as ‘vector competence’ . Several anatomical barriers including mid-gut infection barriers ( MIB ) or mid-gut escape barriers ( MEB ) contribute to reduced susceptibility of A . aegypti mosquitoes to DENV [5] . Though these infection barriers have been demonstrated to be influenced by genes within multiple quantitative trait loci ( QTL ) , the specific genes involved in conferring these infection barriers have not been identified . Natural populations of A . aegypti mosquitoes show extensive genetic variation that may account for varying degrees of susceptibility to DENV [5]–[9] . However , the mechanisms and genes that influence vector competence of A . aegypti to DENV are not well understood [10] . To better understand the global gene expression pattern of mosquito genes upon DENV infection , we performed genome-wide transcriptome analyses in A . aegypti susceptible and refractory strains at two early time points after challenge with DENV . Our data reveals that A . aegypti genes show transcriptional responses in a modular manner at early infection periods , wherein groups of genes are expressed in a tightly correlated manner . Our analysis further shows that specific biochemical pathways are enriched among these modularly expressed genes in the susceptible and refractory mosquitoes . This study was performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The animal use protocol was approved by the University of Notre Dame Institutional Animal Care and Use Committee ( Study # 11-036 ) . The A . aegypti strains Moyo-S ( MS ) and Moyo-R ( MR ) are sub-strains of the Moyo-in-Dry ( MD ) strain that was collected indoors from Shauri Moyo , near Mombasa , Kenya [11] . Sub-strains were originally selected for Plasmodium gallinaceum susceptibility and refractoriness , respectively [12] . Although not selected for DENV susceptibility , the sub-strains also show significant differences in mean DENV serotype 2 ( DENV-2 ) infection rates with ∼20% in one ( MR: DENV-2 refractory ) and ∼57% in the other ( MS: DENV-2 susceptible ) [13] , while the original MD strain shows natural low DENV infection rates of ∼13% ( unpublished data ) . The D2S3 strain was selected for high oral susceptibility to DENV [14] and shows ∼46% susceptibility to DENV-2 infection under our conditions [13] . Mosquitoes were reared and maintained in an environmental chamber following our standard conditions [13] . Cell culture and mosquito infections were performed as previously described [13] . Briefly , starved females were provided with an artificial infectious blood meal , freshly prepared using defibrinated sheep blood ( Colorado Serum Co . , CS1122 ) mixed with ( equal volume ) a dengue viral suspension . DENV-2 strain JAM1409 was cultured using Aedes albopictus C6/36 cells until 80–90% confluence in MEM-EBSS media ( HyClone SH3002401 ) supplemented with 25 mM Hepes buffer , 1 mM sodium pyruvate , 0 . 025 mg/ml Gentamycin , 1× ( 0 . 01 mM ) non-essential amino acids and 10% fetal bovine serum ( heat inactivated ) . A 0 . 1 multiplicity of infection ( MOI ) was used for infecting the mosquito cells . The MOI refers to the average number of viral particles that infect a single cell , which for our purpose is equal to plaque-forming units ( pfu ) per cell . The flasks were incubated at 28°C for 7 days after which the viral supernatant was obtained by centrifugation at 1500 rpm for 5 min at 4°C . All DENV infection work was performed in a BSL3 facility . A genome-wide transcriptome analysis was carried out using the NimbleGen oligonucleotide microarray format ( www . nimblegen . com ) . The custom-made high density array ( 385K format ) was designed with 60-mer oligos specific to 16 , 092 gene transcripts of gene build AaegL1 . 1 of A . aegypti ( www . vectorbase . org ) [15] . For each transcript , from 1 to 20 different unique probes were designed and used . However , for 99 . 4% of genes , 20 probes per each gene were used , with the lower number of probes-to-gene being associated with smaller transcripts . The NimbleGen design utilizes the uniqueness of probe sequences as one of the criteria for probe selection to avoid cross-hybridization with non-target genes . The details of the array design , sample description and expression data are available at Gene Expression Omnibus ( GEO ) under accession number GSE16563 . The layout of the probes in the array was made using row and column specifications with the standard 1∶4 format of NimbleGen ArrayScribe . Total RNA was purified from infected and control samples and forwarded to NimbleGen where labeling and hybridizations were performed following their standard procedures . RNA samples were obtained at 3 hr and 18 hr post blood feeding from DENV-2 infected and control females from each strain . Three independent feeding experiments were performed to obtain three biological replicates of samples for both the strains and the post-infection time points . Fully engorged females were isolated and maintained at 26°C with 84% humidity and in a 16-h light/8-h dark cycle . At each time point , RNA was extracted from 20 females per sample ( the blood meal was first removed from each female using a micro syringe needle ) using a Qiagen RNAeasy Kit as per manufacturer's instructions . The RNA was quantified by a Nanodrop spectrophotometer and quality of RNA was assessed by a Bioanalyzer . A total of 15 samples were used for array hybridizations . They included 12 test samples and three control samples . The test samples included three biological replicates for each of the four infected samples ( MS-3 hr , MR-3 hr , MS-18 hr and MR-18 hr ) . A control was prepared for each of the three replicates that consisted of RNA isolated from females fed with uninfected blood meals and pooled across both strains and time points . The two time points were chosen based on our unpublished observations that significant changes in host gene response to DENV infection are already evident within 24 hrs post-infection . The microarray data normalization was performed using the quantile normalization method [16] and the Robust Multichip Average ( RMA ) algorithm [17] . The Statistical Analysis of Microarray ( SAM ) software [18] was used to determine significantly ( δ = 1 . 61 , fdr = 0 . 52% ) differential expression between test and control samples . The expression modules were determined by weighted co-expression analysis of the differentially expressed genes using the Weighted Gene Correlation Network Analysis ( WGCNA ) program [19] . The ‘topology overlapping’ ( TO ) within the cluster tree was used in the program to predict the expression modules by a dynamic hybrid cutting method [20] . It is based on pair-wise positive and negative correlations ( Pearson correlation matrix ) among the differentially expressed genes . The connection strength ( connectivity ) among the genes was calculated from the absolute value of the matrix raised to a predefined power and genes with similar patterns of connection strengths ( or topological overlap ) are identified . Using topological overlap values , hierarchical clustering was performed to identify modules of highly interconnected genes . A trait file of input genes was used for analysis in WGCNA . It contained binary numbers ( 0 or 1 ) for each gene depending on if the gene was up-regulated or down-regulated in MS or in MR strain , respectively . The WGCNA program was also used to generate the heat maps of gene expression in each module . Genetic networks of the responsive genes were constructed by using the ‘GeneNet’ package implemented in R [21] . The program uses graphical Gaussian models ( GGMs ) to represent multivariate dependencies of genes based on expression data . The algorithm estimates a partial correlation ( pcor ) matrix that is then used to calculate shrinkage covariance estimators of gene expression [22] . Once the shrinkage estimators of pcor values are generated , the program performs GGM selection by multiple testing of false discovery rates that are used to define the nodes and edges of the association network by an empirical Bayes approach [23] . Although graphical GGMs are generally applied to independent and identically distributed data , GeneNet incorporates provisions for small scale datasets , where the observation time points may be unequally spaced . We used the expression data at 3 hr and 18 hr time points for the 2 , 455 responsive genes to estimate partial correlations . The input data for this estimation was generated by use of the ‘longitudinal’ program included in the GeneNet package . The graphical view of the networks was either created by using Graphviz 2 . 18 ( http://www . graphviz . org/ ) or the pair-wise pcor values were extracted for further analyses . The interacting gene pairs identified from GeneNet program were obtained in a tabular form and were compared with networking genes predicted by WGCNA to determine the inter- and intra-modularly interacting genes . In order to understand the functional characteristics of the modular gene expression patterns , we determined if genes of specific pathway ( s ) are over-represented in the modularly expressed genes . The A . aegypti pathway genes were obtained from KEGG ( Koyota Encyclopedia of Genes and Genomes , Japan; http://www . genome . jp/kegg/ ) in October , 2008 . The biased representation of pathway genes was determined by mapping KEGG pathway genes annotated for A . aegypti ( http://www . genome . jp/kegg-bin/show_organism ? org=aag ) . The observed numbers of pathway genes in each of the expression modules ( predicted from array data ) were counted . The cumulative value of gene counts representing a KEGG module was obtained by determining the total number of genes representing each pathway included in the KEGG module . We assumed a null hypothesis where the annotated KEGG pathways had a non-biased representation to the expression modules identified from our array data . Under this assumption , the expected number of genes representing each KEGG module relative to each of our predicted expression modules was the mean number of genes per KEGG module . A test of goodness-of-fit was conducted from the observed and expected gene counts for each expression module using Pearson's Chi-square method . Expression levels of randomly selected responsive genes ( n = 5 ) from the microarray analyses were validated using SYBR Green dye technology ( Applied Biosystems ) by quantitative real-time PCR ( qRT-PCR ) . The qRT-PCR assays were performed with RNAs from MS and MR strains as well as from D2S3 and MD females infected with DENV-2 JAM1409 . The D2S3 and MD strains were infected with DENV-2 JAM1409 as described above . DENV infections and RNA extractions were performed in triplicate from each strain at 3 hr post-infection . Mid-guts were dissected and the blood meal removed as previously described from ∼10 infected individuals each for D2S3 and MD females and RNA was isolated using TRIzol Reagent ( Invitrogen: http://www . invitrogen . com/ ) . Control RNA was isolated from a pool of ∼30 mid-guts each from the uninfected blood fed MD and D2S3 females . First strand cDNA synthesis was performed using Superscript II Reverse Transcriptase ( Invitrogen ) according to manufacturer's instructions . Primer Express Software version 3 . 0 ( Applied Biosystems , Foster City , CA ) was used to design primers . All amplifications and fluorescence quantification were performed using an ABI 7500 Fast System Sequence Detector System ( Applied Biosystems ) and the Sequence Detector Software version 1 . 3 ( Applied Biosystems , Foster City , CA ) . The reactions were performed in a total volume of 25 µl containing 12 . 5 µl of SYBR Green PCR Master Mix , 10 ng of template , 300 nmol of each primer , and nuclease free water . Reactions were performed with the following conditions: 50°C for 2 min , 95°C for 10 min followed by 40 cycles of denaturation at 95°C for 15 s , annealing and extension at 60°C for 1 min . PCR efficiency was determined by amplifying serially diluted cDNA with each primer pair using the identical conditions . The log values of the template concentration versus the threshold cycle ( CT ) were used to plot the growth curve for the amplified products corresponding to each dilution . The slope of the curve was determined to quantify the efficiency of PCR . Efficiency greater than 0 . 95 was ensured for each qRT-PCR . The CT value of each test gene relative to the reference gene , ribosomal protein S17 ( RpS17 ) , was used to determine the delta CT values of infected sample and uninfected control . The RpS17 gene was chosen as the reference gene based on previous results [24] and because it showed no changes in expression in our microarray data . Relative expression values were obtained using the delta-delta cycle threshold ( ΔΔCT ) method [25] . The P-values for testing differences in ΔCT values between susceptible and refractory strains were derived using the nonparametric Wilcoxon two group test [26] . The null hypothesis assumed that ΔΔCT was equal to 0 , P-values <0 . 05 were considered significant . The A . aegypti genes responsive to the critical early stages of DENV infection were identified by a genome-wide transcriptome assay carried out using a NimbleGen oligonucleotide microarray format in MOYO-S ( MS , susceptible to DENV ) and MOYO-R ( MR , refractory to DENV ) females , upon challenging them with the JAM1409 strain of DENV ( serotype-2 ) . We chose to analyze gene expression at 3 hr and 18 hr post-exposure to DENV as the eventual susceptibility status of individual mosquitoes is likely defined during the first 24 hr . That is , DENV is known to rapidly enter vertebrate and insect cells via endocytic pathways [27] , [28] and in vivo studies have reported that successful infection of midgut epithelial cells was already evident in ∼30% of midguts from three susceptible A . aegypti strains ( including D2S3 ) by two days post-infection and thereafter spread laterally to infect neighboring cells [29] . The DENV-specific transcription response genes were identified by comparing each of the test samples ( MS at 3 hr , MS at 18 hr , MR at 3 hr and MR at 18 hr ) with the pooled uninfected control . An initial set of 6 , 339 genes were identified by Statistical Analysis of Microarray ( SAM ) [16] with significant ( δ = 1 . 61 , fdr = 0 . 52% ) differential expression between the infected samples and uninfected control . Because we used a pooled reference microarray design , some of these differentially expressed genes were undoubtedly not related to DENV infection but instead were likely to be associated with developmental changes of the mosquitoes between 3 hr and 18 hr and/or differences between MS and MR strains related to feeding behavior , aging and other factors . To identify differentially expressed genes that were specifically responsive to Ae . aegypt-DENV interaction irrespective of time or strain differences , we selected for differentially expressed genes that were either up-regulated in both strains and time points or were down-regulated in both strains and time points in comparison to the common pool control . Using this pooled reference strategy , a total of 2 , 454 DENV responsive genes were identified from the initial set of 6 , 339 genes . Among the various commonly used strategies for microarray design [30] , we felt that the pooled reference approach offered the most efficient and biologically relevant approach to uncovering only those genes directly associated with DENV infection . Based on the observed expression levels of these significant genes , eight groups [ ( 2 strain ×2 time points ×2 patterns of expression ( up-regulation or down-regulation ) ] of non-overlapping genes were identified that constituted the DENV-specific transcription response genes ( Table 1 ) . All the genes within the individual groups showed differential expression levels corresponding to strain and time point . The levels of expression , however , varied from gene to gene as described in more detail below . Hierarchical clustering based on weighted gene co-expression network analysis ( WGCNA ) [19] identified extensive modular network patterns of Ae . aegypti genes in response to DENV infection . It was found that a total of 1 , 331 genes of the 2 , 454 responsive genes ( 54 . 2% ) were involved in this global network , but in a modular manner . A total of seven ‘modules’ ( designated as ‘A’ through ‘G’ ) of gene expression were predicted using a ‘dynamic hybrid cutting’ method [20] ( Figure 1 ) . A given gene in this network made interactions with as many as 52 other responsive genes as evident from the pair-wise gene interactions . The pair-wise gene interactions within and between these modules were predicted based on partial correlations ( pcor ) of gene expression by an empirical Bayes approach [21] using the GeneNet program ( Table S1 ) . The genes interacting within a module showed elevated average partial correlations among each other compared to that of genes interacting between modules ( data not shown ) . Each module represented a group of genes with correlated expression patterns ( Figure 2 ) . For example , genes belonging to modules B and C have significantly similar expression patterns among the samples as compared to the genes belonging to modules D , E and F . Thus , B and C modules are clustered within one branch of the network cluster tree whereas modules D , E and F are clustered within the other branch of the tree ( Figure 1 ) . Module A and G , on the other hand , represent genes whose expression variations are quite distinct and hence are localized at the distal ends of the cluster tree . The observed patterns of pair-wise interactions within and between the predicted expression modules showed that about two-thirds of all the genes in the network interacted in an inter-modular manner indicating extensive cross-talk among the modules . The numbers of interacting genes that formed the connectivity among these modules were highly variable ( Figure 3 ) . The genes that responded coordinately to DENV infection in the susceptible MS strain were predominantly clustered in one specific expression module ( module E; susceptible response module or SRM ) , whereas in the refractory MR strain they were distributed in modules A , C , D and F ( refractory response module or RRM ) ( Figure 1 ) . Modules B and G represent genes that show time specific differential expression ( between 3 hr and 18 hr post infection ) in both the susceptible and refractory strains ( see Figure 2 ) . These genes may be involved in triggering a common host response ( core response module , CRM ) in the early stages of infection in both strains . A list of responsive genes specific to individual predicted modules is provided in Table S2 . In order to analyze the functional attributes of these expression modules , we made use of A . aegypti KEGG pathways to determine if the modularly expressed genes represented specific pathway ( s ) . Of all the KEGG pathways that were mapped to each expression module , genes of specific pathway ( s ) were predominant within the predicted expression modules ( Table 2 ) . All the predicted pathways associated with the seven expression modules ( except module E where p<0 . 1 ) were significantly ( p<0 . 05 ) enriched with genes representing specific pathways . The susceptible response module ( module E ) showed enrichment with genes related to energy metabolism and DNA replication and repair . The over-representation of genes of the DNA replication pathway in the susceptible mosquitoes may be related to activities associated with the DENV infection process . Indeed , it has been shown that the cell cycle environment in C6/36 cells influences the course of DENV infection wherein DENV replication is enhanced in S-phase cells [31] , and blood feeding alone activates cellular metabolism and is known to induce S-phase in multiple tissues in adult females [32] , [33] . The refractory response modules ( A , C , D and F ) showed significant enrichment with different metabolism pathway genes that included cytochrome P450 genes , genes involved in DDT [1 , 1 , 1-Trichloro-2 , 2-bis ( 4-chlorophenyl ) ethane] degradation ( mostly including the short-chain dehydrogenase , amino acid decarboxylase as well as glutathione-s-transferase theta ( gst ) coding genes ) and also genes associated with cell growth and death , such as cell division and apoptosis . Several p53 signaling genes , caspase genes and phosphatidylinositol 3-kinase signaling genes were up-regulated in the refractory response modules ( Table 3 ) . These pathways are known to modulate apoptosis in response to viral infections from other studies . Because such evidence is presently restricted to vertebrate cells [34]–[39] , further studies are needed to confirm their roles in insects . The common core response modules ( B and G ) were enriched with genes related to signal transduction , as well as sorting and degradation . The mechanisms of how the gene networks transduce signals to trigger the appropriate host action in A . aegypti against DENV infection are important aspects of vector competence . Genes related to important signal transduction pathways such as the Wnt , MAPK , mTOR and JAK-STAT pathways were predominant among all the responsive signal transduction pathway genes ( Figure 4 ) . We identified several genes associated with the JAK-STAT pathway among the responsive genes ( Table 4 ) . Significant differential expression of A . aegypti genes of this pathway may be involved in the activation of the STAT in response to induction by JAK ( Figure 5 ) . In addition to the activation of STAT , JAK induction may mediate the recruitment of other molecules such as the MAP kinases which results in the activation of additional transcription factors . It is possible that the JAK-STAT signaling pathway may be involved in activating the MAPK cascade [40]–[41] or in regulating apoptosis as shown in Drosophila [42] . The interface of these core response genes with genes involved in susceptible response and refractory response module ( s ) suggests their important roles in the global-cross talk among the host factors during these early infection periods that could trigger the appropriate host action in the susceptible and refractory mosquitoes . Vacuolar trafficking upon DENV entry into a mosquito mid-gut epithelial cell may be an important cellular process associated with mosquito-virus interaction . Such mechanisms have been described wherein pH-dependent vacuolar trafficking modulates flaviviral entry into human as well as mosquito cells [43]–[46] . Consistent with this likelihood , our data suggests the potential for differential expression of the endocytic pathway between susceptible and refractory strain in response to dengue infection . We observed that genes coding for V-ATPases are differentially expressed between the susceptible and refractory strain . That is , of the 17 differentially expressed V-ATPases genes , 15 genes were up-regulated in the susceptible strain ( Table 5 ) . It is plausible that differential expression of these genes could influence the endocytic pathway , possibly by differential acidification of endosomes [44] . Our data also revealed that cuticle protein ( CP ) genes may play important roles in A . aegypti response to dengue infection . We observed that while a total of 28 responsive cuticle protein ( CP ) genes were down-regulated among the strains upon DENV infection ( Table 6 ) , the extent of down-regulation of these genes was much more severe in the MS strain ( n = 25 genes ) than the MR strain ( n = 3genes ) ; thus these genes showed proportionally higher overall expression levels in MR mosquitoes than the MS mosquitoes . Although the function of cuticle protein in dengue infection is not known , it is possible that they may play a role in development of anatomical barriers for virus dissemination as an additional innate defense mechanism in the refractory mosquitoes . To validate the microarray data and to determine if genes differentially expressed between the MS and MR strains in response to DENV infection are also involved in driving similar transcriptional responses in other A . aegypti strains upon DENV infection , we performed quantitative real time PCR ( qRT-PCR ) assays . Five genes were randomly chosen from the microarray data set ( Table S3 ) and subjected to qRT-PCR in the MS and MR strains , and two additional A . aegypti strains ( D2S3: DENV susceptible and Moyo-in-Dry or MD: DENV refractory ) . After challenge with DENV-2 JAM1409 , D2S3 and MD samples at 3 hr post-infection were quantified by qRT-PCR for all five genes and compared with results for both microarray and qRT-PCR with the MS and MR strains ( Figure 6 ) . The qRT-PCR data showed statistically significant ( P<0 . 05 ) up-regulated or down-regulated expression patterns for each gene between the MS and MR strains as well as between the D2S3 and MD strains with respect to the uninfected control . Comparisons of the microarray and qRT-PCR results showed consistent trends in variation ( R2>0 . 9 and P<0 . 05; Figure S1 ) . These observations indicate that the DENV responsive genes may have similar susceptibility-specific host responses to DENV infection in different A . aegypti strains and may play important roles in vector competence to DENV infection at the critical early infection stages . We conducted a comparative genome-wide survey of gene expression patterns observed in response to DENV infection among A . aegypti females known to be susceptible or refractory to infection . Our results show that 2 , 454 DENV responsive genes interact in well-defined patterns that distinguish the two response phenotypes . The observed transcriptional network establishes global cross-talk among the DENV response genes that may subsequently trigger the appropriate host actions at the critical early time points following exposure to and infection by DENV . Our data revealed that 293 genes were responsive in both susceptible and refractory strains , whereas most DENV responsive genes had expression patterns that were specific to either susceptible or refractory genotypes ( Figure 3 ) . It also showed that the refractory phenotype involved a much greater number of genes that acted in tightly correlated manners relative to the number of genes which were associated with the susceptible phenotype . This indicated that A . aegypti may utilize more complex defense mechanisms compared to those required to host the virus . This is also clearly seen from data shown in Table 1 . Based on the correlated expression levels of the responsive genes , we identified several candidate pathways that may determine the compatible or non-compatible interaction between A . aegypti and DENV . Most of these pathway genes were associated with expression fold-changes ranging from 0 . 5 to 2 . 2 ( Tables 3–6 ) indicating that some important pathways show relatively modest transcriptional responses to DENV infection . We did , however , observe other DENV responsive genes with much higher transcriptional responses ( Table S4 ) . The length of the extrinsic incubation ( EIP , total time from viral entry to transmission stage ) period likely varies depending upon the host and viral genotype , environmental factors such as temperature and humidity , as well as other unknown intrinsic factors [47] . Recent evidence demonstrates that , depending on genetic background of the A . aegypti population , the EIP can be completed in as few as 4 days in contrast to 7 to 10 days as most commonly observed [29] . As observed in a recent study that examined transcriptome response to DENV infection in a susceptible A . aegypti strain at seven days post-infection [48] , the involvement of the JAK-STAT pathway in controlling virus infection was evident in our study . Another recent study identified candidate genes in a susceptible A . aegypti strain at 10 days post-infection [49] , and determined that the Toll and the JAK/STAT pathways play important roles in controlling DENV infection in A . aegypti . In Drosophila , the JAK-STAT pathway has been shown to be necessary but not sufficient for triggering an anti-viral defense in Drosophila to Drosophila C virus ( DCV ) infection [50] . Our data , that focused on the critical early time points for infection and compared DENV susceptible and refractory strains , revealed that genes of the JAK-STAT pathway were up-regulated in response to DENV challenge in both the MS and MR strains at these early periods , suggesting that while it does play a role in determining DENV infection its significance in defining vector competence remains unclear . We observed activation of JNK and p53 related genes as well as caspase genes in response to DENV infection in the MR strain . This suggests possible induction of programmed cell death events in the refractory strain following DENV infection . Programmed cell death is an efficient host survival mechanism in insects where the infected cells undergo apoptosis to prevent viral infections [51] . Moreover , a role for apoptosis has been implicated in mosquito response to infection by several arboviruses and orthologs to apoptosis-associated genes in Drosophila have been identified and are expressed in A . aegypti [52] . Although a caspase-dependent role in apoptosis induction has been suggested in dengue virus infection in animal cells [36] , [38] , [53] , further studies are however required to determine if programmed cell death is one of the mechanisms of controlling DENV infection in A . aegypti . We also identified several DENV responsive genes that were previously reported to play important roles in modulating viral infections in animal cells . For example , a trypsin gene ( AAEL010195 ) was significantly down-regulated gene in both strains ( data not shown ) . In A . aegypti , midgut trypsins have been shown to influence the rate of DENV-2 infection and dissemination [54] . In addition , another serine protease ( AAEL005753 ) was similarly significantly down-regulated in the infected mosquitoes of both strains . Serine proteases play an important role in proteolytic digestion of blood meal proteins in mosquitoes , and results have shown that some midgut serine proteases may play a role in DENV-2 infectivity of A . aegypti [55] . We also identified a Toll-like receptor gene ( AAEL015018 ) as significantly up-regulated in both strains . Toll-like receptors are well known genes that have been shown to invoke anti-viral innate immune responses or to ameliorate viral infection in various host cells including A . aegypti in response to DENV [49] , [56] . Additionally , two furin-like genes ( AAEL010725 and AAEL002317 ) were also found over-expressed in the MS strain only that may be involved in efficient maturation of newly synthesized virions by cleavage of the DENV precursor membrane protein , prM [57] . We observed that while cuticle protein genes were significantly down-regulated following DENV infection in both the MS and MR strains , the majority ( 25 of 28 genes ) were down-regulated in the MS strain . It is plausible that the generally higher expression levels of these genes in the MR strain may be associated with enhanced anatomical barriers in these mosquitoes , possibly by the tracheal system , to limit DENV escape from the mid-gut epithelium . In this regard , it is interesting to note that the reported anatomical barriers to DENV transmission in A . aegypti include a mid-gut escape barrier ( MEB ) , wherein the virus may be able to successfully invade and replicate in the mid-gut epithelium but is blocked from disseminating to other tissues [47] . The tracheal system makes intimate contact with mid-gut epithelial cells and has been identified as a dissemination conduit for several insect/virus systems , including A . aegypti and DENV [29] . Tracheae do contain a cuticular lining that could limit virus dissemination [58] and , therefore , dissemination from the mid-gut could be impacted by differential induction of cuticle proteins in the MR strain . Our data suggest , based on numbers of genes and the diversity of metabolic pathways involved , that defending against viral infection reflects greater evolutionary complexity . The resilient nature of A . aegypti as the primary vector for dengue transmission is apparent [1] and thus , understanding gene expression patterns across various natural populations is needed to provide insights on genome-wide networking of DENV responsive genes , and the ultimate impact on population-specific vector competence . It is expected that comparative analyses at the population level may identify genes within differential network patterns critical to a susceptible or refractory response . Such variation may uncover key points in metabolic pathways for development of novel intervention strategies . Future efforts need to be directed toward better clarification of the specific roles of individual pathways and well as identification of key points for their interactions using systems biology approaches [59] .	Dengue virus is primarily transmitted by Aedes aegypti mosquitoes . Control of the vector mosquito is the major practice to prevent dengue . However , it is not well known how the virus can infect some mosquito strains but fail to do so with other refractory strains . To address that question , we conducted whole genome microarray based gene expression studies between susceptible and refractory strains of A . aegypti to identify gene expression patterns following challenge with dengue virus . Our analysis of the early infection periods reveals that a large number of genes are involved in a highly coordinated manner either to host or defend against the virus . Genes responding to dengue infection were clustered in seven expression modules . Genes associated with susceptibility to infection were largely clustered in one expression module , while those associated with refractoriness were distributed in four distinct modules . A common core set of genes expressed in both susceptible and refractory individuals were clustered in two expression modules . We identified genes and specific pathways that potentially regulate compatible or non-compatible interactions between dengue virus and the mosquito , most notably energy metabolism and DNA replication in the susceptible response in contrast to cell growth and death in the refractory response .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[ "genome", "expression", "analysis", "emerging", "viral", "diseases", "microbiology", "host-pathogen", "interaction", "viral", "vectors", "mechanisms", "of", "resistance", "and", "susceptibility", "emerging", "infectious", "diseases", "viral", "immune", "evasion", "medical", "microbiology", "gene", "expression", "biology", "mosquitoes", "heredity", "vector", "biology", "virology", "genetics", "genomics", "gene", "networks", "genetics", "and", "genomics", "complex", "traits" ]	2011	Global Cross-Talk of Genes of the Mosquito Aedes aegypti in Response to Dengue Virus Infection
Differentiating pluripotent cells from fibroblast progenitors is a potentially transformative tool in personalized medicine . We previously identified relatively greater success culturing dura-derived fibroblasts than scalp-derived fibroblasts from postmortem tissue . We hypothesized that these differences in culture success were related to epigenetic differences between the cultured fibroblasts by sampling location , and therefore generated genome-wide DNA methylation and transcriptome data on 11 intrinsically matched pairs of dural and scalp fibroblasts from donors across the lifespan ( infant to 85 years ) . While these cultured fibroblasts were several generations removed from the primary tissue and morphologically indistinguishable , we found widespread epigenetic differences by sampling location at the single CpG ( N = 101 , 989 ) , region ( N = 697 ) , “block” ( N = 243 ) , and global spatial scales suggesting a strong epigenetic memory of original fibroblast location . Furthermore , many of these epigenetic differences manifested in the transcriptome , particularly at the region-level . We further identified 7 , 265 CpGs and 11 regions showing significant epigenetic memory related to the age of the donor , as well as an overall increased epigenetic variability , preferentially in scalp-derived fibroblasts—83% of loci were more variable in scalp , hypothesized to result from cumulative exposure to environmental stimuli in the primary tissue . By integrating publicly available DNA methylation datasets on individual cell populations in blood and brain , we identified significantly increased inter-individual variability in our scalp- and other skin-derived fibroblasts on a similar scale as epigenetic differences between different lineages of blood cells . Lastly , these epigenetic differences did not appear to be driven by somatic mutation—while we identified 64 probable de-novo variants across the 11 subjects , there was no association between mutation burden and age of the donor ( p = 0 . 71 ) . These results depict a strong component of epigenetic memory in cell culture from primary tissue , even after several generations of daughter cells , related to cell state and donor age . DNA methylation ( DNAm ) at CpG dinucleotides plays an important role in the epigenetic regulation of the human genome , contributing to diverse cellular phenotypes from the same underlying genetic sequence . For example , DNAm levels at particular genomic loci can accurately classify different tissues [1] and even underlying cell types within tissues [2] . These stable cell type- and tissue-discriminating loci appear to represent only a subset of "dynamic" CpGs , approximately 21 . 8% , actively involved in regulation of gene expression [3] . Changes in these epigenetic patterns across aging have been extensively studied [4] , particularly in large studies of whole blood [5–7] , but subsets of these age-associated CpGs appear tissue-independent [8] . These epigenetic barcodes also play an important role in cellular reprogramming ( the conversion of somatic cells to pluripotent stem cells ) , a powerful and promising experimental system in biology , genetics and personalized medicine [9] . This epigenetic reprogramming of somatic cells to induced pluripotent stem cells ( iPSCs ) induces demethylation [10] followed by specific patterns of subsequent DNA methylation that can reflect the original somatic tissue [11] . Fibroblasts are one of the most popular cell types for generating iPSCs [12] , particularly from skin , given the relative ease of access to these cells , although other skin-derived cell types such as keratinocytes from the same individual generate similar iPSC lines [13] . Skin , however , is perhaps the most susceptible tissue source in the body to environmentally induced insult , particularly through sunlight and chemical exposures , which can induce changes in epigenetic patterns [14] . The epigenetic “memory” of source tissue for iPSC characterization has been well characterized [11] . In our previous work , we successfully cultured fibroblast lines from the dura mater of postmortem human donors , a source location largely protected from environmental insult with slowly dividing cells [15] . We compared these cultured fibroblast lines to those derived from scalp samples from the same individuals , and found that the rate of culture success was higher for dura-derived fibroblasts; in some cases only the dura fibroblasts from an individual would culture . While the resulting cultured cells from these two sampling locations were largely morphologically indistinguishable ( see Figure 1 in Bliss et al , 2012 [15] ) , we hypothesized that increased culture success might have a strong epigenetic component . Previous research has shown that dermal fibroblasts from different locations in the body have distinct gene expression profiles [16] , including compared to some non-dermal sources [17] , and previous reports have indicated that cultured cells have largely stable epigenomes , with the exception of a small number of loci [18] . We therefore sought to characterize the methylomes and transcriptomes of fibroblasts from these two sampling locations–scalp and dura–from donors across the lifespan . Here we identify several components of epigenetic “memory” in cultured fibroblasts after multiple passages ( i . e . splitting and continuing to grow ) where primary tissue originated from two locations in the body . The strongest epigenetic memory was related to sampling location in the body , as we identified widespread DNAm differences at local and regional spatial scales preserved through identical culturing processes . We further find increased stochastic epigenetic variability in cultured fibroblasts from the scalp compared to dura . This increased variability manifested in significant increased quantitative pairwise methylome-wide distances in a combined analysis with publicly available DNAm data on skin fibroblasts [19] , pure cell populations from peripheral blood [20] , and cells from the dorsolateral prefrontal cortex [21] . Another component of epigenetic memory was related to the age of the donor , including a subset of CpGs that displayed location-dependent changes through aging . The epigenetic differences between these fibroblasts appear to occur largely through epigenetic-dependent mechanisms , as there were few differences in coding sequence across the fibroblasts from the two locations within the same individual . These results demonstrate the effect of epigenetic memory in cultured fibroblasts by sampling location and donor age in morphologically indistinguishable cells . We first characterized differences in DNAm levels from cultured fibroblasts derived from different locations ( scalp versus dura ) . Many probes , targeting individual CpGs , were differentially methylated between scalp- and dura-derived fibroblasts– 101 , 989 ( 22% ) at genome-wide significance ( false discovery rate , FDR < 5% , see Methods ) . These significant DNAm differences between cultured fibroblasts from the scalp and dura were large in magnitude , with 57 , 704 probes having differences in DNAm levels greater than 10% , and 23 , 752 with differences greater than 20% ( Fig 1A ) . The directionality of these DNAm differences was balanced , with approximately equal proportions of CpGs showing increased versus decreased methylation in cultured fibroblasts from scalp compared to dura . These differentially methylated probes ( DMPs ) were widely distributed across the genome , as 18 , 551 genes ( defined by UCSC knownGene database ) had at least one DMP within 5 kilobases ( kb ) , as did 33 , 247 transcripts ( see Methods ) . These widespread single CpG differences manifest as the largest component of variability in the entire dataset , as the first principal component ( Fig 1B , explaining 38% and 62 . 3% of the variability before and after surrogate variable analysis , SVA [24] ) represents the sampling location of these cultured fibroblasts , suggesting a strong epigenetic memory of original cell location . Since these differentially methylated CpGs tended to cluster in a smaller number of genes , we further identified 697 differentially methylated regions ( DMRs ) at stringent genome-wide significance ( family-wise error rate , FWER < 10% ) –these regions were identified based on adjacent probes showing directionally-consistent differences in DNAm > 10% between groups [25] ( see Methods ) . For example , we identified a region of 24 contiguous probes hypermethylated in scalp-derived fibroblasts within the gene RUNX3 –a tumor suppressor that plays an integral role in regulating cell proliferation and the rate of apoptosis [26] ( Fig 1C , see S2 Fig and S2 Table for all significant DMRs ) . Regional differences , particularly in CpG island shores , previously have been shown to better distinguish tissues and cell types [1] and correlate with neighboring gene expression levels [23] than individual CpGs . Unlike at the single CpG level , which had balanced directionality of differential methylation , the majority of DMRs had higher DNAm levels in fibroblasts derived from scalp compared to those derived from dura ( N = 414 , 59 . 4% ) . Using gene sets defined by biological processes [27] , these neighboring genes ( within 5 kb ) were strongly enriched for morphogenesis ( including morphogenesis of the epithelium ) , developmental processes , cell differentiation , and epithelium and connective tissue development , among other more general gene sets ( all p < 10−8 , S3 Table ) . In addition to the extensive differential methylation at both the CpG and regional level , we identified 243 long-range regions with consistent significant methylation change ( FWER < 10% ) , called “blocks” [28] , using an algorithm adapted from whole genome bisulfite sequencing ( WGBS ) data to Illumina 450k [23] . A representative significant block is shown in Fig 1D ( see S3 Fig for all significant blocks at FWER < 10% ) . Blocks have now been identified across many cancer types [29] , and tend to associate with higher order chromatin structure including nuclear lamin-associated domains ( LADs ) [30] and large organized chromatin K9 modification ( LOCKs ) [28] . The 243 significant blocks in our data represent 41 Mb of sequence and contain 298 annotated genes . These blocks contain 41 of the significant DMRs that differentiate sampling location of the fibroblasts , and more interestingly , every block overlaps at least one “dynamic” cell/tissue DMR identified using WGBS data from Ziller et al ( 2013 ) [3] . While these cultured fibroblasts were several generations/passages removed from the primary tissue and morphologically indistinguishable , we nevertheless found widespread epigenetic differences by sampling location of the primary fibroblasts at varying spatial scales , suggesting a strong epigenetic memory of the original cell location . We next sought to determine the functional correlates of the widespread epigenetic differences identified between scalp- and dura-derived fibroblasts by performing RNA sequencing ( RNA-seq ) on polyadenylated ( polyA+ ) mRNA from the same cultured samples ( see Methods ) . Briefly , we aligned the reads to the transcriptome using TopHat [31] and generated normalized gene counts ( as fragments per kb per million mapped reads , FPKM ) based on the Illumina iGenome hg19 annotation using the featureCounts software [32] . A median of 88 . 0% ( interquartile range , IQR: 85 . 5%– 88 . 8% ) of reads mapped to the genome , of which a median of 84 . 7% ( IQR: 84 . 4%–85 . 5% ) mapped to the annotated transcriptome ( see S1 Table for sample-specific percentages ) . We identified 11 , 218 expressed genes with average FPKM expression greater than 1 . 0 . Initial clustering of the gene FPKM values separated the fibroblast samples by location in the first principal component ( PC ) , which explained 35 . 4% of the variance ( S4 Fig ) , mirroring the first principal component of the DNAm data ( Fig 1B ) . We could further cluster our samples by sampling location using a set of 337 genes ( of which 210 were in our dataset ) that were previously identified by Rinn et al [17] to group largely dermal fibroblasts by their anatomical sites of origin ( S5 Fig ) –these genes better clustered the samples by sampling location than random sets of 210 genes ( p<0 . 001 , see Methods ) . Differential expression analysis of the RNA-seq data , independent of the results from the epigenetic analyses above , identified many genes that differed by the source of the primary fibroblast– 5 , 830 genes at FDR < 5% . Both scalp- and dura-derived fibroblasts expressed high levels of Fibroblast Specific Protein-1 ( FSP-1 ) and this gene was more highly expressed scalp-derived fibroblasts ( fold change = 5 . 5 , FDR = 5 . 6x10-6 ) in line with increased higher proliferation rates in the scalp-derived versus dura-derived fibroblasts [15] . The differentially expressed genes were strongly enriched for signaling and cell communication , cell proliferation , apoptotic processes , and epithelium development and morphogenesis via gene ontology ( GO ) analysis ( all p < 10−8 , S4 Table ) –these gene sets were similar , and much more significant , to those identified comparing gene expression profiles across positional-identity genes in dermal fibroblasts [17] . We next used the gene expression data as a functional readout of the differentially methylated loci identified between fibroblasts cultured from scalp and dura . The majority of significant DMPs ( 76 , 971/101 , 989 , 75 . 47% ) were inside or near ( within 5kb of ) a UCSC annotated gene , and 28 . 2% ( 21 , 742/76 , 971 ) were significantly associated with gene expression levels ( at p < 0 . 05 ) . This percentage of DMPs with significant expression readout was elevated ( 34 . 9% ) among those DMPs with larger DNAm differences by sampling location ( greater than 10% difference in DNAm levels ) . These DMPs were strongly significantly enriched among the CpG sites that associated with expression levels at the p < 0 . 05 ( 48 , 062 probes within 5kb of genes , odds ratio , OR = 3 . 99 , p < 2 . 2x10-16 ) and FDR < 0 . 05 ( 6 , 559 probes within 5kb of genes , OR = 19 . 54 , p < 2 . 2x10-16 ) significance thresholds . Surprisingly , we found that the DNAm levels at the majority of these expression-associated differentially methylated CpGs tended to be positively associated with gene expression , regardless of overall methylation levels ( un- , partially- , or highly-methylated ) or their location in the gene ( islands , shores and shelves ) –these biases towards positive associations were statistically significant for many of these comparisons ( see S5 Table , panels A and B ) . We hypothesize these positive correlations could be due to the probe design of the Illumina 450 ( the majority of probes are in lowly methylated regions ) combined with the majority of genes having low expression ( 38 . 75% had mean FPKMs < 1 ) . We identified similar associations using transcript-level expression data using the Sailfish program [33] ( see Methods ) on the above transcriptome– 76 . 5% ( 77 , 981/101 , 989 ) of the DMPs were within 5 kb of a transcript , and 30 . 4% of them ( 23 , 672/77 , 981 ) correlated with expression ( at p < 0 . 05 ) . 33 , 247 unique transcripts overlapped or were within 5 kb of DMPs , and of them , 27 . 0% ( 8 , 981/33 , 247 ) exhibited significant correlation between DNAm and expression ( at p < 0 . 05 ) . The 33 , 247 transcripts proximal to the DMRs corresponded to 18 , 699 genes , the majority of which ( 84 . 3% , 15 , 761/18 , 699 ) contained more than one transcript . Interestingly , these associations often appear in a transcript-specific manner—6 , 190 genes ( 39 . 3% ) had ≥ 1 transcript with significant correlation between DNAm and expression ( at p < 0 . 05 ) , with ≥ 1 transcripts that were not associated with nearby CpG levels . These results suggest that genes , and their underlying transcripts , can functionally validate many of the differentially methylated CpGs for sampling location . Moving beyond individual CpGs , 587/697 ( 84 . 2% ) DMRs were in or near ( <5kb ) genes , and many had DNAm levels that were significantly associated with gene expression levels ( 306/587 , 52 . 1% at p < 0 . 05 ) . For instance , a DMR overlapping an intronic sequence of the SIM1 gene ( Fig 2A ) was unmethylated with low corresponding expression of the gene in the cultured fibroblasts from dura , and highly methylated with corresponding high expression levels of the gene in the scalp-derived fibroblasts ( Fig 2B and S2 Table ) . This is in line with previous reports suggesting that gene body methylation levels positively associate with local gene expression [34] , unlike CpG island shore methylation that tends to be negatively associated with gene expression levels [1] . Of the 478 unique genes in or within 5kb of DMRs , the expression of 235 ( 49 . 2% ) of them was significantly correlated with DNAm ( p < 0 . 05 ) . These 235 unique genes tended to exhibit stronger differential expression between the scalp- and dura-derived fibroblasts ( median fold change = 1 . 59 , IQR = 1 . 23–2 . 68 ) than individual CpG results , in line with previously published findings [23] . GO analysis on expression-associated genes proximal to DMRs revealed enrichment for multiple important biological processes such as connective tissue development , epithelium morphogenesis and development , cell differentiation ( specifically including epithelial cell differentiation ) , and cell proliferation ( including epithelial cell proliferation ) , among other more general sets ( all p < 10−8 , see S6 Table ) . Unlike at the single CpG-level , we found that the majority of DMRs in and around the transcriptional start sites of genes ( CpGs islands and shores ) were negatively correlated with gene expression ( S5 Table ) , in line with previous research [1] . We observed similar methylation-expression associations using transcript-level expression measurements– 312/599 DMRs ( 52 . 1% ) near ≥ 1 transcripts associated with expression , and like at the single CpG level , found evidence for transcript-specific epigenetic regulation of expression ( among 28 . 9% of genes containing multiple transcripts and associated with DNAm levels within the DMRs ) . Lastly , we found that the majority of differentially methylated blocks contained at least one gene and transcript differentially expressed between scalp- and dura-derived fibroblasts . The majority of blocks contained at least one gene ( N = 188/243 , 77 . 4% ) ; 63 . 8% ( N = 120/188 ) had at least one gene and 66 . 66% ( N = 124/186 ) at least one transcript that was differentially expressed ( at p < 0 . 05 ) . As a representative example , one of the blocks , hypermethylated in scalp-derived fibroblasts , overlaps the HOXB gene cluster ( Fig 3A ) , which has previously been shown to be play a role in the position identities of fibroblasts [17] . In this block , expression levels of the HOXB genes are significantly greater in fibroblasts cultured from scalp than those from dura ( Fig 3B ) , which contrasted previous microarray-based data showing these genes were not expressed in dermal samples taken from the head [17] highlighting the improved precision of RNA-sequencing data to quantify expression levels . Similarly , the 188 significant blocks contained 298 unique genes , and 126 of them ( 42 . 3% ) were differentially expressed ( at FDR < 0 . 05 ) which is a higher proportion than the rest of the transcriptome ( 0 . 42 vs . 0 . 32 , p = 3 . 79x10-9 ) . Given the strong association between DNAm levels and local expression levels , we sought to more fully examine the epigenetic states of these sampling location-associated DNAm differences . We downloaded chromatin state data ( 18 states ) from the NIH Roadmap Epigenomics Consortium on the four available fibroblast samples ( 2 primary foreskin , 1 adult dermal , and 1 lung ) [35] , and mapped our DMPs , DMRs , and blocks for fibroblast sampling location onto these states ( S7 Table ) . The CpGs differentially methylated by sampling location were largely enriched for enhancer chromatin states , including preferential enrichment of genic ( EnhG2 ) and active ( EnhA1 ) enhancer states and depleted for active transcriptional start site ( TSS ) states ( TssA ) . At the region level , DMRs were largely enriched for bivalent TSS ( EnhBiv ) and repressive polycomb ( ReprPC ) states and depleted for transcription ( Tx ) genic enhancer ( EnhG2 ) states , and blocks were strongly enriched for quiescent ( Quies ) and heterochromatin ( Het ) states and depleted for transcriptional states . These enrichments were relatively conserved across the four Roadmap fibroblast samples , further suggesting distinct epigenetic states in scalp- compared to dura-derived fibroblasts . These results suggest that epigenetic memory related to original cell location manifests in genomic state differences and largely reads out in the transcriptome , particularly among regional changes in DNAm related to fibroblast sampling location . We hypothesized that scalp-derived fibroblasts might have more variable levels of DNAm than dura-derived fibroblasts , given the chronic exposure to environmental factors ( e . g . sunlight , chemicals ) in the primary tissue across the lifespan . At the individual CpG level , we tested for differences in variance between the scalp- and dura-derived fibroblasts independent of the underlying mean methylation levels [36] ( see Methods section ) . While only two probes reached genome-wide significance ( at FDR < 0 . 05 ) for differences in variance , at marginal levels of significance ( p < 0 . 05 ) , fibroblasts cultured from scalp had more variable DNAm levels than fibroblasts cultured from dura ( N = 13 , 169/16 , 330 , 80 . 6% ) . We next sought to characterize methylome-wide patterns of DNAm across these fibroblasts in the context of other diverse cell types . After downloading and normalizing Illumina 450k data from sorted blood [20] and frontal cortex [21] , as well as skin-derived fibroblasts [19] and melanoma samples ( SKCM ) from the Cancer Genome Atlas ( TCGA ) [37] , we computed methylome-wide Euclidean distances between and across each of the 11 cell types ( see Methods section ) . We noted that these cell types largely cluster by tissue source ( brain , blood , and fibroblasts in the first two principal components and largest dendrogram splits , S6 Fig ) . The inter-individual epigenomic distances , and their variability , were much greater in the scalp-derived ( as well as skin-derived ) fibroblasts than dura-derived fibroblasts ( p = 1 . 34x10-9 and p = 1 . 77x10-14 respectively , see Fig 4 ) . The distances within scalp- and skin-derived fibroblasts were significantly larger than those calculated within pure blood and cortex cell types ( p-values range from 1 . 04x10-21 to <10−100 ) . Interestingly , the inter-individual distances between fibroblasts cultured from scalp samples were greater than the distances between different cell types within a blood cell lineage ( e . g . natural killer cells versus CD4+ T-cells ) which were previously suggested for different dermal fibroblasts [16] and instead more similar to distances across lineage ( e . g . natural kill cells versus monocytes ) . Note that comparing inter-individual distance between two cell types ( e . g . scalp- versus dura-derived fibroblasts ) reflects the extensive differential methylation between these two cell types ( see Fig 1 ) —the inter-individual distances are large but the variability in distances was low . As another example , the distances across scalp-derived fibroblasts were lower than the inter-individual variability between neurons and non-neurons ( via NeuN+ sorting ) , which reflects the extensive methylation differences between these two cell types [21] . As expected , we found the greatest methylome-wide distances and largest inter-individual variability in the melanoma samples [28 , 36] , which highlights the relative scale of these methylome-wide distances ( ranging from pure cell types to cancer ) . These increased epigenomic distances may relate to the rate of cell division , which is non-existent in neuronal cells [38] and infrequent in T-lymphocytes at the population level [39] . The increased epigenetic variability in the scalp samples was further not associated with differences in donor age ( p > 0 . 05 , S7 Fig ) , suggesting increased epigenetic stochastic variability in scalp- ( and skin- ) derived fibroblasts . We hypothesized that a subset of this increased variability might result from age-related divergence in DNAm at individual loci that were differential by sampling location , such that young donors would have lesser difference in DNAm levels , and older donors would have larger differences in DNAm . By fitting linear models on the difference in DNAm levels across sampling location as a function of donor age ( see Methods ) , we identified 7 , 265 CpGs associated with diverging DNAm levels across aging ( at FDR < 10% , S8 Fig ) . These loci appeared to be clustered into representative patterns of their age-related changes ( Fig 5 ) . The majority of these CpGs had significant age-related changes in fibroblasts derived from the scalp ( 64 . 0% ) , but not dura ( 17 . 4% ) , and the magnitude of change across age was larger in scalp-derived fibroblasts–the average change in percent DNAm per decade of life was 3 . 13% ( IQR = 1 . 81%-4 . 29% ) in fibroblasts derived from scalp compared to 1 . 13% ( IQR = 0 . 295%-1 . 61% ) in those from the dura mater . A subset of these CpGs showing sampling location-dependent age-related changes associated with nearby gene expression levels . Most of the probes ( N = 5 , 185/7 , 265 , 71 . 4% ) were annotated to 3 , 553 unique genes ( within 5kb ) and 21 . 8% of these ( N = 775/3 , 555 ) showed significant correlation between DNAm and gene expression ( p < 0 . 05 ) . These DNAm associated genes were enriched for multiple general developmental processes including cell development , morphogenesis , and differentiation ( all p<10−8 , S8 Table ) . Several of the age-related CpGs showing expression association were within genes that are involved in cell proliferation and apoptosis . For instance , DNAm levels at two significant probes inside the gene TEAD1 , which regulates notochord development and cell proliferation [40] , were significantly associated with gene expression levels ( p = 8 . 60x10-4 and 0 . 045 , respectively ) . Another significant DNAm-expression pair ( p = 0 . 02 ) involved AVEN , a gene shown to inhibit Caspase activation in apoptosis [41] . Interestingly , while we identified a large number of age-related CpGs , “DNA methylation ages” [8] were very similar to the chronological ages of the samples ( see Methods and S9 Fig ) –these associations did not differ by sampling location ( p = 0 . 72 ) and there was further no association between “DNA methylation age” and sampling location alone ( p = 0 . 96 ) . The age-associated CpGs identified here therefore suggest that altered regulation of DNAm levels across aging occurs primarily in fibroblasts derived from scalp but not from dura , perhaps through altered cell proliferation and apoptosis , and possibly reflecting greater exposure to environmental agents that can affect the methylome . Lastly , we characterized the expressed sequences of the scalp- and dura-derived fibroblasts within each individual to examine the extent of genetic mosaicism , which may contribute to differences in DNAm through changing the underlying genetic sequence in the fibroblasts taken from scalp . De novo variants were called directly from the RNAseq data , and after filtering by many quality metrics ( see Methods ) we identified 64 high-confidence candidate variants that were discordant by sampling location in at least a single individual ( S9 Table ) , including 22 annotated coding variants ( 13 synonymous and 9 non- synonymous ) [42] . We found no association between coding variant burden and subject age ( p = 0 . 71 , S10 Fig ) . These results suggest that many of the location- and age-associated DNAm differences are not due to somatic mosaicism and likely arise through epigenetic mechanisms that are maintained through cell culture and multiple passages . Here we interrogated the methylomes and transcriptomes of pairs of fibroblasts cultured from scalp and dura mater taken from the same individual , in a subject cohort that ranges in age across the human lifespan . These cultured fibroblasts , generations removed from the primary tissue of origin , and with indistinguishable morphology , still maintained strong components of epigenetic “memory” related to sampling location ( scalp versus dura ) and differential changes in DNAm levels across aging . The widespread differences in DNAm levels by sampling location were identified at many spatial scales , including single CpGs , differentially methylated regions , blocks , and globally . Furthermore , many of these differences in DNAm levels manifested in the transcriptome , showing significant corresponding differences in expression for genes most proximal to these epigenetic changes . The genes with differences in expression and DNAm levels by sampling location were previously implicated in processes relating to cell proliferation and apoptosis , which likely relate to the function of the fibroblasts in the primary tissue . One might have predicted this outcome , as fibroblasts in the scalp , including those that are cultured , turnover much more rapidly than those in the dura mater [15] , which we confirmed here with increased FSP-1 expression in the scalp-derived fibroblasts . Another component of epigenetic memory in these cultured fibroblasts was related to ages of the donors , where age-related changes occurred differentially by sampling location . These age-associated loci can be clustered into general patterns of epigenetic changes by age and location , all showing significant interaction between donor age and sampling location . While some patterns were expected , such as divergence in DNAm levels from similar levels at birth ( clusters 1 , 4 , 5 , and 7 ) , several other clusters showed an unexpected convergence in DNAm across aging ( clusters 2 and 3 ) . We do note that the elderly donor ( age 85 ) is influential in both the statistical discovery at individual loci and in some of the subsequent clusters–larger sample sizes can hopefully further define and replicate these observations . Also , while the fibroblasts were analyzed from some subjects with psychiatric disorders , almost all comparisons between scalp and dura sampling locations , and differential changes with age were naturally matched within an individual , reducing the potential impact of diagnostic confounding . Furthermore , a larger sample size would likely identify significant age-related divergence in DNAm at the region level–while we found 7 , 265 individual CpGs , we found very few DMRs at global significance ( 6 and 11 DMRs at FWER ≤ 10% and 20% respectively ) . The region-finding approach has been shown to be statistically conservative [25] and the identification of these differential age-related changes by sampling location was based on number of donors ( N = 10 ) , not the number of observations ( N = 21 ) . Lastly , while proliferation rates were not measured for these particular fibroblast samples , analyses in a much larger skin biopsy sample ( N = 298 ) showed no association between proliferation rates and donor age [43] , which was our sampling location with the greater number of age-related changes in DNAm levels . These age-related changes in cultured fibroblasts are one of the first examples , to our knowledge , of genome-wide significant age-related changes in a pure cell population that is many mitoses and passages from the original donor cells . Many papers have identified widespread age-related changes in heterogeneous cell populations , like blood [5 , 7] , brain [44] , and other tissue types [8] , which may result in false positives when the underlying cellular composition changes across aging [4] . Other papers have used individual cell populations to validate age-associated loci identified in homogenate tissue at marginal significance [45] or have identified age-related changes in targeted approaches at limited numbers of loci [46] . Similarly , these fibroblasts cultured from the scalp and dura mater were the first example , again to our knowledge , of morphologically indistinguishable cells with vastly different epigenomic profiles . Using epigenomic distances , these two cohorts of fibroblasts were more different in their DNAm patterns than different lineages of blood cells , while less different that neuronal versus non-neuronal cells from the frontal cortex ( Fig 4 ) ; the cells underlying each comparison have very different morphologies and cellular function . Furthermore , the majority of differences in DNAm levels between scalp- and dura-derived cultured fibroblasts appeared to be determined early in development , prior to early infancy in this sample , and remained stable throughout the lifespan . Of the 101 , 989 significant DMPs for sampling location , 98 , 461 ( 96 . 5% ) were not associated with differential age-related changes . These findings demonstrate strong components of epigenetic memory related to cell location and aging in fibroblasts cultured from the scalp and dura mater from postmortem human donors . There are important implications from this study for the field of regenerative medicine . If fibroblasts are going to be the source for iPSCs , and ultimately differentiated tissues , the source of these fibroblasts , and their epigenetic characteristics , may be an important consideration . For example , these differences in cellular states in cultured fibroblasts may relate to the number of cell divisions , as skin and scalp fibroblasts have a much quicker turnover than fibroblasts in the dura [15] . The extent of cell division could relate to the epigenomic distances between and across the diverse cell types we have analyzed . Analyses in larger samples of skin biopsy-derived fibroblasts suggest that while donor age does not appear to associate with proliferation rates of fibroblasts , the cultured cells derived from younger donors reprogrammed more readily [43] , which presumably has a strong epigenetic component . Further research may better determine the extent of epigenetic memory of cell state of fibroblasts cultured from different locations after the generation of iPSCs and subsequent differentiation into new cell types . As the field of regenerative medicine advances , our study demonstrates that deciding upon the source of fibroblasts from an individual to generate new tissues and organs may be an important consideration . While it was shown that transcriptional variability by tissue of origin was low in iPSCs ( 13 ) , it was also demonstrated that the DNAm landscape in iPSCs differs greatly by tissue or origin , and this phenomenon may explain the propensity of iPSCs derived from different somatic tissues to differentiate into different lineages ( 11 ) . Human dural and scalp fibroblasts on which the methylation and gene expression studies were performed were obtained from fibroblast lines derived from human post mortem scalp and dura mater tissues . For this study , tissues from 11 individuals were used , with the ages of individuals ranging from 0 . 1 to 85 years of age ( see S1 Table for additional demographics ) . The post-mortem tissues from 2 of the subjects were collected by the Lieber Institute for Brain Development ( LIBD ) and the tissues from the remaining 9 subjects were collected by National Institute for Mental Health ( NIMH ) ( Clinical Brain Disorders Branch ( CBDB ) , Division of Intramural Research Programs ( DIRP ) ) . The NIMH tissues were collected from two medical examiners ( Washington , DC office and Commonwealth of Virginia , Northern District office ) ; the LIBD tissues were obtained the Office of the Chief Medical Examiner ( Baltimore , MD ) . A preliminary neurological or psychiatric diagnosis was given to each case after demographic , medical , and clinical histories were gathered via a telephone screening on the day of donation . For each case , the postmortem interval ( PMI ) ( the time ( in hours ) elapsed between death and tissue freezing ) was recorded . ( See S1 Table for PMIs and demographics for every subject used in this study ) . Every case underwent neuropathological examinations to screen for neurological pathology . Additionally , the medical examiner’s office performed toxicology analysis of every subject’s blood to screen for drugs . Dura and scalp tissue were collected at the time of autopsy . From the autopsy room , the tissues were transported in separate bags: one containing cerebral dura mater and the other a 1 in X 1 in scalp segment with hair attached . Both bags were transported on wet ice to the lab , where the culture procedure was immediately started . The dura culture medium was prepared out of 1X DMEM ( Ref#11960–044 , GIBCO ) with 10% by volume fetal bovine serum , 2% by volume 100X GlutaMAX ( Cat#: 35050 , GIBCO ) , 1% by volume Penicillin-Streptomycin/Amphotercin solution ( Ref# 15140–122 , GIBCO ) , and 1% by volume Gentamicin solution ( Cat# 17105–041 , Quality Biological ) . This culture medium was used in all subsequent steps of the dura culturing procedure . The scalp culture medium used for all subsequent steps of the scalp culturing procedure was made the same way except without the 1% Gentamycin . A rinsing solution was prepared out of 1X PBS ( pH 7 . 2 ) ( Ref# 21-040-CV , Corning Life Sciences ) , 1% by volume Penicillin-Streptomycin/Amphotericin solution ( Ref# 15140–122 , GIBCO ) , and 1% by volume Gentamicin ( Cat# 17105–041 , Quality Biological ) . The dissected scalp sample was washed with the rinsing solution three times , the fat tissues were cut away , and all hair was plucked out with forceps . The scalp sample was then placed epidermis side down on a dish and floated with Dispase II enzyme solution ( 2 . 4 units of the Dispase II enzyme per mL of PBS , Dispase II enzyme: Cat#17105–041 , GIBCO ) . ( Dispase II enzyme is a proteolytic enzyme used to separate the dermis from the epidermis by cleaving the zone of the basement membrane . ) The dish was covered with parafilm and foil , and placed in a 37°C incubator for 24 hours . After the 24-hour period , the epidermis was peeled away from the dermis . The dermis was washed with the rinsing solution , dried , and cut into 2–3 mm2 pieces . The pieces were placed in a Falcon Easy Grip tissue culture 35×10 mm dish and one drop of scalp culture medium was added to each piece of scalp . The dish was placed in the incubator at 37°C and 5% CO2 for culturing . A similar procedure was followed for the dura samples . Dura samples were washed with the rinsing solution three times . Then , a few 2–3 mm2 pieces were cut from the dura mater and placed together in an Easy Grip cell culture 35×10 mm dish . One drop of dura culture medium was added to each dura piece . The culture dish was then placed in an incubator ( at 37°C and 5% CO2 ) for culturing . The medium of each culture was changed to fresh medium 2–3 times per week to promote growth of the fibroblasts . On average , fibroblast cells started to proliferate at 7–14 days , however some samples took longer ( up to 3 weeks ) . The dura and scalp tissue cultures were monitored under a phase-contrast microscope . When the fibroblast growth reached 90–95% confluence , 1 mL of a 0 . 25% trypsin solution ( Cat#T4049 , Sigma ) was added to each culture dish , and the cells were incubated for 5 to 8 min . Then , 1mL of media was added to each dish stop the enzymatic reaction . Next , the contents of each culture dish were transferred into separate 15 mL Falcon conical tubes and 8mL of media was added to each tube . The conical tubes were centrifuged for 5 min at 1100 rpm . The supernatant was discarded , 5mL of fresh media was added to each conical tube , and the contents of the tubes were transferred onto separate 25 cm3 cell culture Easy Flasks ( Thermo Scientific , Cat# 156367 ) , where they were kept in cultures for 3–5 days in an incubator ( at 37°C and 5% CO2 ) . When the cells reached 90–95% confluence , the cells from each 25 cm3 flask were transferred onto two 75 cm3 cell culture easy flasks ( Thermo Scientific , Cat# 156499 ) and kept in cultures for continued growth . When the cells reached 90–95% confluence , they were incubated with 3 mL of 0 . 25% trypsin solution for 5 to 8 min , after which 3mL of fresh culture media was added to stop the enzymatic reaction . Then , the contents of the flasks were transferred into separate 15 mL Falcon conical tube and 4mL of media was added to each tube . The tubes were centrifuged ( 5 min , 1100 rpm ) , the supernatant was discarded and the pellets containing the fibroblasts were removed from the centrifuge tubes and transferred to cryoTube vials ( Cat#375418 , Thermo Scientific ) . 0 . 5 mL of recovery cell culture freezing medium ( Cat#12648–010 , GIBCO ) was added to each vial , after which the vials were insulated with Styrofoam and placed into a -80°C freezer . Later , the tubes were transferred to a -152°C liquid nitrogen freezer . These frozen dura and scalp fibroblast cells were then used generate DNA methylation and gene expression levels . Genomic DNA was extracted from approximately 3 million cultured human fibroblast cells using the AllPrep DNA/RNA/miRNA Universal Kit ( Qiagen ) . Bisulfite conversion was performed on 600 ng genomic DNA was done with the EZ DNA methylation kit ( Zymo Research ) . DNA methylation landscapes of the dura- and scalp-derived fibroblasts were analyzed using the Illumina HumanMethylation 450 BeadChip array ( “450k” ) . The 450k array interrogates >485 , 000 DNA methylation sites ( probes ) and measures the proportion DNA methylation at each target site ( the 450k array interrogates both CpG and CH sites ) . The microarray preparation and scanning were performed in accordance with the manufacturer’s protocols . The resulting data from the 450k consists of R ( ed ) and G ( reen ) intensities using two different probe chemistries [22] , which we converted to M ( ethylated ) and U ( nmethylated ) intensities using the minfi Bioconductor package [23] , version 1 . 14 . 0 using with R version3 . 2 . One dura sample had lower median probe intensities and was removed prior to normalization and downstream analyses . After quality control ( QC ) , the M and U intensities were normalized separately across samples using stratified quantile normalization [23] . Probes containing common SNPs ( based on dbSNP 142 ) at the target CpG or single base extension site , and probes on the sex chromosomes were removed , leaving 456 , 513 probes on 21 samples for analysis . We determined differential methylation using linear modeling on the normalized DNAm levels , using the model: yij=αi+βiLocj+ζiSVsj+εij ( 1 ) where yij is the normalized proportion methylation at probe i and sample j , αi is the proportion methylation in the fibroblasts sampled from the dura mater , βj is the difference in methylation in the scalp-derived fibroblast , and Locj is the sampling location represented by a binary variable ( Dura = 0 , Scalp = 1 ) . These statistical models were adjusted for surrogate variables ( 6 SVs ) estimated using surrogate variable analysis ( SVA ) [24] . Differentially methylation probes ( DMPs ) were identified by fitting Eq 1 to each probe , and obtaining the corresponding moderate t-statistic and p-value using the limma package [47] . P-values were adjusted for multiple testing using the false discovery rate ( FDR ) [48] and significant probes were called were FDR < 0 . 05 . Principal component analysis ( PCA ) was performed after regressing out the surrogate variables from the DNAm levels of each probe , preserving the effect of fibroblast sampling location . Finding differentially methylated regions ( DMRs ) involves identifying contiguous probes where β ≠ 0 using the bumphunter Bioconductor package ( version 1 . 6 . 0 ) [25] , here requiring \|β\| > 0 . 1 ( argument: cutoff = 0 . 1 ) and assessing statistical significance using linear modeling bootstrapping with 1000 iterations ( argument: nullMethod = ‘bootstrap’ and B = 1000 ) . DMRs were called statistically significant when the family wise error rate ( FWER ) ≤ 0 . 1 . We identified blocks using the same model as above using the blockFinder function in the minfi package [23] , which collapses nearby CpGs into a single measurement per sample , and then fits Eq . 1 above , only here j represents probe group , not probe . Here we again required at least a 10% change in DNAm between groups and assessed statistical significance using the FWER based on 1000 iterations of the linear model bootstrap . RNA was extracted from the cultured dura and scalp fibroblasts with the RNeasy kit ( Qiagen ) , in accordance with the manufacturer’s protocol . RNA molecules were treated with DNase , polyadenylated ( polyA+ ) RNA was isolated , and resulting sequencing libraries were constructed using the Illumina TruSeq RNA Sample Preparation Kit ( v2 ) and sequenced on an Illumina HiSeq 2000 . We note that while all samples were run on the same flow cell , the samples were somewhat imbalanced by lane–however , the first PC of the expression data did separate perfectly by sampling location . Sample-specific information on reads and alignments are available in S1 Table . Resulting reads were mapped to the genome using TopHat2 [31] using the paired-ends procedure ( we used the option—library-type fr-firststrand ) . Gene counts relative to the UCSC hg19 knownGene annotation were calculated using the featureCounts script of the Subread package ( version 1 . 4 . 6 ) [32] . There were 23 , 710 genes in this annotation , and we dropped 305 genes that were annotated to more than 1 chromosome . Of the remaining 23 , 405 genes , 18 , 316 genes had non-zero expression counts in at least one sample . Counts were converted to FPKM ( fragments per kilobase per million reads mapped ) values to allow comparisons across genes with different lengths and libraries sequenced to different depths . These FPKMs were transformed prior to statistical analysis: log2 ( FPKM + 1 ) . The log transformed FPKM values were used in all subsequent gene-level analyses . Next , we used the Sailfish software ( 33 ) , version 0 . 7 . 6 , to quantify isoforms from our RNA-seq reads . As a result , we obtained TPM ( transcripts per million ) values for each isoform , which we log transformed: log2 ( TPM + 1 ) . The log transformed TPM values were used in all subsequent transcript-level analyses . Differential expression for sampling location was identified using Eq 1 above , where yij represents transformed expression ( rather than DNAm ) levels , and different SVs ( N = 4 ) were calculated from the expression data . To test whether we could use a subset of genes to cluster our fibroblasts by sampling location , like reported by Rinn et all [17] , we took the 337 genes published by the authors , which they found to group fibroblasts by anatomical location . Of these 337 , we used only 210 genes , since a subset of the tabulated genes did not contain gene symbols , another subset was not interrogated by our RNAseq , and yet another subset was not expressed in any of our samples . We then perfumed Euclidean distance computations and clustering analysis by first using these 210 genes and then repeating the analysis 1000 times using 210 randomly chosen genes . We carried out gene ontology analysis on the differentially expressed genes with the GOstats package [49] . Transformed FPKMs were next used to assess functional significance of differentially methylated features . We mapped the DMPs to genes in the UCSC knownGenes ( hg19 ) and determined which DMPs exhibit correlation between DNAm and gene expression with the MatrixEQTL package [50] . We used Pearson's Chi-squared test with Yates' continuity correction to examine whether DMPs are more likely to exhibit correlations between DNAm and gene expression than non-DMPs . We then mapped significant DMRs to genes expressed in the RNA-seq data ( e . g . showing non-zero expression levels in ≥ 1 samples ) , and correlated the average DNAm level within the DMR to the transformed expression level . When multiple genes were within or near a DMR , we retained the gene ( and its correlation ) with the largest absolute correlation . We carried out gene ontology analysis for the genes proximal to DMRs with the GOstats package . For each significant block , we found the UCSC annotated gene ( s ) containing within the block and their evidence for differential expression as calculated above . We used Pearson's Chi-squared test with Yates' continuity correction to test whether differentially expressed genes were enriched in blocks compared to the rest of the transcriptome . Finally , we analyzed the directionality of DNAm—expression correlations for DMPs and DMRs , as a function of DMR/DMP positions relative to genes . We used the binomial test to access the significance of distributions between positive and negative correlations of DNAm and gene expression . In addition to gene-level analysis , we studied transcript-level expression and its correlation with DNAm . We carried out the same analysis for isoform expression as for gene-level expression , with the exception that here we used relative isoform abundance values that we obtained with the Sailfish software ( see above ) . The 18-chromatin state data , derived using hidden Markov models ( HMMs ) , was obtained for 4 fibroblast samples: samples E055 and E056 ( foreskin primary fibroblasts ) , E126 ( adult dermal fibroblast ) , and E128 ( lung fibroblsts ) in the Epigenome Roadmap project22 ( http://egg2 . wustl . edu/roadmap/web_portal/chr_state_learning . html ) . The chromatin states overlapping DMPs , DMRs , and blocks were obtained , and compared to a background of all 450k probes , considered probe groups , and collapsed probe groups respectively . Overlap was assessed based on the total coverage ( in base pairs ) of the chromatin states . Fold changes for enrichment of > 1 . 5 fold were highlighted . Prior to carrying out the enrichment analysis , the sex chromosomes and the mitochondrial chromosome were dropped . We performed a second larger data processing and normalization procedure on our scalp- and dura-derived fibroblasts after adding data from skin fibroblasts ( GSE52025 ) [19] , pure populations of blood [20] and prefrontal cortex cells [21] from the FlowSorted . Blood . 450k and FlowSorted . DLPFC . 450k Bioconductor packages respectively , and then melanoma data from TCGA [37] . The M and U channels were combined across all experiments and then normalized with stratified quantile normalization as described above . We then dropped the probes on the sex chromosomes as well as probes that are common SNPs ( based on dbSNP 142 ) as described above . Within the normalized data , we then calculated all pairwise Euclidean distances on the proportion methylation scale , and selected specific comparisons to display in Fig 4 . We calculated differential variability between scalp and dura CpG DNAm levels using the Levene test [51] and subsequent p-values were adjusted for multiple testing using the FDR . We filtered out the 101 , 989 genome-wide significant probes showing mean methylation differences by sampling location , as there is a strong mean-variation relationship in DNAm data due to being constrained within 0 and 1 ( e . g . gaining methylation from an unmethylated state or losing methylation from a fully methylated state increases variance ) . We tested for probes that showed differential age-related divergence in DNAm by fibroblast sampling location . First , we calculated the difference in DNAm between scalp- and dura-derived fibroblasts from the same individual at every probe ( creating a 456 , 513 probe by 10 individual matrix ) . We then computed 3 surrogate variables ( the number estimated by the SVA algorithm ) for a statistical model with donor age , and fit the following linear model: Δyij=γi+δiAgej+ζSVsj1+εij ( 2 ) where Δyji is the difference in DNAm levels between scalp and dura for probe i and individual j , γi is the difference in DNAm levels at birth , Agei is the age of the donor , and δi is the change in the difference of DNAm per year of life . We then generated a Wald statistic and corresponding p-value for δi and adjusted for multiple testing via the FDR . Post hoc age-related changes , e . g . the change in DNAm levels per year of life , were calculated within the scalp and dura samples . We then associated expression of nearby genes ( within 5kb ) with the DNAm levels at the probes showing significant age-by-location effects and performed gene ontology on the significant genes with the GOstats package [49] . We lastly computed the “DNAm age” of our scalp and dura samples using the R code published by S . Horvath , ( available at https://labs . genetics . ucla . edu/horvath/dnamage/ ) and fit a linear model containing main effects of biological age and sampling location , and an interaction term between these two variables on “DNAm age” . We called expression variants directly from the RNA sequencing alignments using samtools ( version 1 . 1 ) and mpileup across all samples [52] . We then filtered variants in the resulting variant call format ( VCF ) file based on coverage ( <20 ) , variant distance bias ( p<0 . 05 ) , read position bias ( p<0 . 05 ) , mapping quality bias ( p<0 . 05 ) , base quality bias ( p<0 . 05 ) , inbreeding coefficient binomial test ( p<0 . 05 ) , and homozygote bias ( p>0 . 05 ) . The resulting 64 high quality variants were annotated with SeattleSeq138 [42] . For every subject from whom the post-mortem tissues were collected , informed consent was obtained verbally from the legal next-of-kin using a telephone script , and was both witnessed and audiotaped , in accordance with the IRB approved NIMH protocol 90-M-0142 and the Department of Health and Human Services for the State of Maryland ( protocol # 12–24 ) . DNA methylation data in both raw and processed forms are available on the Gene Expression Omnibus ( GEO ) : GSE77136 . RNA sequencing reads ( raw data ) are available on the Sequencing Read Archive ( SRA ) : SRP068304 ( BioProject: PRJNA286856 ) and the genes and transcript counts ( processed data ) are available on GEO at the above accession number ( GSE77136 ) .	Regenerative medicine specialists have been using a type of cell commonly found in the skin called the fibroblast because it is easily obtained from skin samples , grows well in culture , and can be manipulated in the laboratory to de-differentiate into a primordial state known as the induced pluripotent stem cell . These primitive stem cells can then be transformed into mature tissues , such as liver or pancreas cells . Here we show that fibroblasts , coming from different locations in the same individual , vary significantly in epigenetic marks called DNA methylation , which are involved in the regulation of gene expression . In addition to location-specific patterns of DNA methylation , we also find that fibroblasts from different anatomical locations respond differently in epigenetic patterns related to aging . As the field of regenerative medicine advances , our study demonstrates that deciding upon the source of fibroblasts from an individual to generate new tissues and organs may be an important consideration .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods", "and", "Materials" ]	[ "sequencing", "techniques", "medicine", "and", "health", "sciences", "scalp", "biological", "cultures", "fibroblasts", "connective", "tissue", "cells", "genome", "analysis", "epigenetics", "dna", "molecular", "biology", "techniques", "rna", "sequencing", "dna", "methylation", "chromatin", "research", "and", "analysis", "methods", "genomics", "animal", "cells", "chromosome", "biology", "gene", "expression", "connective", "tissue", "biological", "tissue", "chromatin", "modification", "dna", "modification", "head", "cell", "lines", "molecular", "biology", "biochemistry", "cell", "biology", "anatomy", "nucleic", "acids", "genetics", "transcriptome", "analysis", "biology", "and", "life", "sciences", "cellular", "types", "cultured", "fibroblasts", "computational", "biology" ]	2016	Strong Components of Epigenetic Memory in Cultured Human Fibroblasts Related to Site of Origin and Donor Age
Identifying transcription factors ( TF ) involved in producing a genome-wide transcriptional profile is an essential step in building mechanistic model that can explain observed gene expression data . We developed a statistical framework for constructing genome-wide signatures of TF activity , and for using such signatures in the analysis of gene expression data produced by complex transcriptional regulatory programs . Our framework integrates ChIP-seq data and appropriately matched gene expression profiles to identify True REGulatory ( TREG ) TF-gene interactions . It provides genome-wide quantification of the likelihood of regulatory TF-gene interaction that can be used to either identify regulated genes , or as genome-wide signature of TF activity . To effectively use ChIP-seq data , we introduce a novel statistical model that integrates information from all binding “peaks” within 2 Mb window around a gene's transcription start site ( TSS ) , and provides gene-level binding scores and probabilities of regulatory interaction . In the second step we integrate these binding scores and regulatory probabilities with gene expression data to assess the likelihood of True REGulatory ( TREG ) TF-gene interactions . We demonstrate the advantages of TREG framework in identifying genes regulated by two TFs with widely different distribution of functional binding events ( ERα and E2f1 ) . We also show that TREG signatures of TF activity vastly improve our ability to detect involvement of ERα in producing complex diseases-related transcriptional profiles . Through a large study of disease-related transcriptional signatures and transcriptional signatures of drug activity , we demonstrate that increase in statistical power associated with the use of TREG signatures makes the crucial difference in identifying key targets for treatment , and drugs to use for treatment . All methods are implemented in an open-source R package treg . The package also contains all data used in the analysis including 494 TREG binding profiles based on ENCODE ChIP-seq data . The treg package can be downloaded at http://GenomicsPortals . org . The specificity of transcriptional initiation in the genomes of eukaryotes is maintained through regulatory programs entailing complex interactions among transcription factors ( TF ) , epigenetic modifications of regulatory DNA regions and associated histones , chromatin-remodeling proteins , and the basal transcriptional machinery [1] . High-throughput sequencing of immuno-precipitated DNA fragments ( ChIP-seq ) provides means to assess genome-wide expression regulatory events , such as TF-DNA interactions [2] . Sophisticated statistical methodologies have been developed for identifying TF binding events in terms of “peaks” in the distributions of ChIP-seq data [3]–[8] . The evidence provided by ChIP-seq binding data that a gene's expression is regulated by a TF is a function of the number of peaks , their intensity and proximity to the transcription start site ( TSS ) [9] . Furthermore , binding of a transcription factor in a gene's promoter alone does not always result in transcriptional regulation . In the case of highly studied pleiotropic regulator ERα , transcriptional regulation depends on the presence of specific co-factors as well as on the type of activating ligand [10] , [11] . Therefore , the identification of true regulatory TF-gene relationships requires per-gene summaries/scores measuring the totality of the evidence in ChIP-seq data , integrated with measurements of gene expression levels . Current approaches to summarizing binding peaks in order to correlate TF binding with transcriptional changes range from simple summaries in proximal gene promoter ( e . g . maximum peak height within a narrow region around the promoter ) [12]–[14] to weighted sums of peak heights where weights are inversely proportional to the distance of the peak to the gene's TSS [9] , [15] . Currently used distance-based weights are dependent on TF-specific tuning constants established through ad-hoc examination of the distribution of the peaks [9] , [12] , [13] . Dysregulation of transcriptional programs is intimately related to the progression of cancer [16] , [17] and other human diseases [18] , [19] . Modulating the behavior of specific TFs is a popular strategy for developing new disease treatments [20]–[23] . Genome-wide transcriptional profiles associated with a disease phenotype provide indirect evidence of TF involvement in the etiology of the disease . The most common strategy of implicating TF involvement is by computational analysis of genomic regulatory regions of differentially expressed genes [24]–[27] . However , such strategies are not effective when the search needs to include distant enhancers and when concurrent activity of multiple regulatory programs lead to “messy” transcriptional signatures . ERα-driven proliferation is one such case where the involvement of ERα regulatory program has been difficult to identify in resulting transcriptional profiles using the DNA binding motif analysis [27] . We have developed a comprehensive statistical framework for assessing True REGulatory ( TREG ) TF-gene interactions by integrated analysis of ChIP-seq and gene expression data . In the first step we introduce a novel two-stage mixture generative statistical model for summarizing “peaks” within 2MB window centered around a gene's TSS . Fitting this two-stage model yields scores and associated probabilities of regulation based on ChIP-seq data alone ( ie TREG binding profile ) . We show that our approach produces effective summaries for a TF with binding sites clustered in close proximity of TSS ( E2F1 ) and a TF known to exhibit regulation through binding to distant enhancers ( ERα ) . In the second step we integrate the TREG binding profile with a differential gene expression profile to create an integrated TREG signature of TF regulatory activity . We use TREG signatures to detect faint signals of ERα regulation in “messy” transcriptional signature , and demonstrate how such analysis can yield better drug candidates than simply correlating transcriptional signatures of the disease and the drug activity [28]–[30] . We assume that observed peaks consist of two populations: Functional peaks that are more likely to occur closer to TSS and whose distance to TSS is distributed as an exponential random variable; and , Non-functional peaks that are randomly occurring throughout the 2 million base pair genomic region centered around the TSS , and whose distances to TSS are distributed as a uniform random variable . The distances to TSS of all peaks are then distributed as a mixture of the exponential and the uniform distribution ( Fig . 1 , Eq1 ) , where π is the proportion of functional peaks among all observed peaks . We define the TREG binding score for gene g as the logarithm of the weighted average of peak intensities , using the probability of the peak belonging to the population of “functional peak” as weights ( Fig . 1 Eq3 ) . We assessed the effectiveness of the TREG binding score by comparison to the simple scoring method based on the maximum peak intensity ( MPI ) within a window of specific size around TSS . The two types of scores were evaluated by comparing the enrichment of genes with high evidence of TF binding among genes differentially expressed in appropriately matched experiments . For gene expression data , we identified genes differentially expressed ( two-tailed FDR<0 . 01 ) 24 h after treating MCF-7 cell line with estradiol ( E2 ) with and without pre-treating the cell line with Cycloheximide ( CHX ) [27] . CHX is an inhibitor of protein biosynthesis in eukaryotic organisms . Treatment with E2 after pre-treatment with CHX ( E2+CHX ) resulted in differential expression of genes presumed to be directly regulated by ERα; whereas after E2 treatment without CHX , the majority of differentially expressed genes were secondary target genes functionally enriched for cell-cycle genes and reflective of the rapid proliferation resulting from the E2 treatment [27] . For the TF binding data , we used ChIP-seq analysis of the key proliferation regulator E2f1 in growing mouse embryonic stem ( ES ) cells [32] , and ERα binding 1 h after treating MCF-7 cells with estradiol [10] . ChIP-seq data at 1 h hour after treatment with E2 is correlated with gene expression changes 24 h after treatment because of the expected time-delay between ERα binding to a gene promoter and the observable change in the gene's expression level . Among differentially expressed genes , enrichment of genes with high TREG binding scores was statistically significant for both E2F1 and ERα in both experiments ( Table 1 ) . Fig . 2 shows the relative levels of enrichment for maximum peak intensity ( MPI ) score over the range of window sizes around TSSs in comparison to the TREG binding score . Simple MPI scores never attain the level of statistical significance of enrichment attained by TREG binding scores . Furthermore , the performance of the simple score is heavily dependent on the specific size of the window used , and expectedly , the optimal windows are TF–specific . The optimal window size for E2f1 and ERα is around 1 kb and 50 kb respectively , with maximum statistical significance of enrichment attained for the simple score reaching 42% and 80% of the TREG binding score significance , respectively . Similar results were obtained using unweighted sum and linear-weighted sum of TF binding peak intensity scores ( supplementary results in Text S1 and Fig . S1 ) . This indicates that TREG binding scores not only provide the best correlation with expression changes , but they also obviate the need of knowing the right window size to use in deriving the summary measure of TF binding . The calculation of TREG binding scores does not include any free parameters that need to be specified in ad-hoc fashion , such as the length of the genomic region around TSS for simple scores , or the ad-hoc weighting parameters used in similar scores before [9] , [15] . Having constructed gene-specific TREG binding score , our goal was to estimate gene-level probabilities of “functional interaction” between a TF and a gene based on these scores . The histogram of the TREG binding scores ( Fig . 1C ) clearly shows two populations of TREG binding scores . One population with a majority of TREG binding scores being close to zero , representing genes with low likelihood of functional TF-gene interaction , and the other populations with TREG binding scores distributed in bell-shaped form around the mean slightly higher than 2 , representing functional interactions . Therefore , we assume that TREG binding scores come from two populations: Scores significantly greater than zero representing functional TF-gene interactions which are distributed as a Normal random variable; and , scores close to zero representing non-functional interactions which are distributed as an exponential random variable . Assuming that the proportion of TREG binding scores corresponding to functional interactions is η , the distribution of all TREG binding scores is a mixture of Normal and exponential probability distribution functions ( Fig . 1 Eq4 ) . The probability that a TREG binding score for gene g ( Sg ) is functional is defined as the probability of Sg belonging to the normal component ( Fig . 1 Eq5 ) . The set of TREG binding scores and associated probabilities of the score indicated functional TF-gene interaction for all genes in the genome ( Sg , pg ) , g = 1 , … , G , is the TREG binding profile . Identifying genes that both have high probability of “functional” TF binding and are differentially expressed is complicated by the need to set arbitrary thresholds for statistical significance . We have previously developed a method , based on the Generalized Random Set ( GRS ) analysis that obviates the need for such thresholds when assessing concordance of two differential gene expression profiles [31] . Here we apply the GRS framework to assess the concordance between the TREG binding profile and the differential gene expression profile ( Fig . 1 Eq6 ) ( details in Text S1 ) , and to identify genes with statistically significant concordance . The results ( Table 2 ) of the analysis generally followed the results based on designating differentially expressed genes ( Table 1 ) with the levels of statistical significance being orders of magnitude higher in the GRS concordance analysis . We demonstrate that GRS is producing expected distribution of p-values under the null hypothesis by systematically examining empirical cumulative distribution functions ( ECDFs ) of p-values after randomly permuting gene labels in TREG binding profile before GRS analysis ( supplementary results in Text S1 , Fig . S2 ) . We also compared the results of GRS analysis with the thresholding approach based on TREG binding probability where gene was placed in the “regulated” group if the corresponding TREG probability ( pg ) was greater than 0 . 95 . Results were similar to the GRS analysis ( supplemental results Text S1 ) . However , we also show that in the situations when binding signal is relatively “faint” , GRS is likely to outperform thresholding approach ( Text S1 , Fig . S3 ) . Since these are situations in which the method of concordance analysis will make the difference , the GRS is still likely the better default choice for performing the concordance analysis . Finally , we integrate at the gene level TREG binding profiles with differential gene expression profiles as the contribution of an individual gene to the overall concordance in the GRS concordance statistics eg ( Fig . 1 Eq7 ) . The statistical significance of gene-level GRS statistics is assessed by associated resampling-based p-values ( see methods ) which define gene-specific TREG concordance scores ( tg , Fig . 1 , Eq8 ) . The vector of such scores for all genes represents the TREG signature of TF activity ( Fig1 Eq9 ) . We examined the ability of TREG binding profiles and TREG signatures to identify genes regulated by ERα and E2f1 . Fig . 3A contrasts the statistical significance of the enrichment by the computationally predicted ERα targets from MSigDB database [33] based on E2+CHX differential gene expression profile ( Diff Exp ) , ERα TREG binding scores ( TREG bind ) and integrated TREG signature ( TREG sig ) . In this setting , MSigDB targets provide a “noisy” gold standard since the perfect gold standard does not exist . While all three data types provided statistically significant enrichment , the integrated TREG signature showed the highest statistical significance of the enrichment . The overall relationship between the TREG binding scores , statistical significance of differential gene expression ( −log10 ( p-value ) E2+CHX ) and the statistical significance of TREG concordance scores ( ERα TREG score ( sg ) ) is shown in Fig . 3B . The “statistically significant” ( p-value<0 . 001 ) TREG concordance scores ( red dots in Fig . 3B ) required both , a high TREG binding score and a high statistical significance of differential expression . Similar analysis of the E2f1 TREG signature showed a similar pattern ( Fig . 3C and D ) , although the overall statistical significance of enrichment was much higher for all three data types . These results show that integrated TREG signatures are more informative of the regulatory TF-gene relationships than expression or TF binding data alone . TREG binding scores , gene specific concordance statistic , and TREG concordance scores for all genes are given in the Table S1 . We further examined ERα and E2F1 TREG signatures to determine molecular pathways and biological processes regulated by these two TFs and to evaluate benefits of such integrated signatures . We assessed the enrichment of genes with high TREG concordance scores in lists of genes related to the prototypical function of ERα and E2F1 . For the ERα signature the list consisted of genes associated with the Gene Ontology term “cellular response to estrogen stimulus” , and for the E2F1 with the term “regulation of mitotic cell cycle” . In both cases , integrated TREG signatures showed significantly higher statistical significance of enrichment than either TREG binding scores or differential gene expressions ( Fig . 4 ) . Unsupervised enrichment analysis of the two signatures revealed that biological processes specifically associated with ERα signature were related to the development of the mammary gland ( Fig . 5A ) . Moreover , significant associations between ERα-regulated genes and some key developmental processes could not have been established using either TF binding or gene expression alone . Likewise , processes related to mitotic cell cycle were most highly associated with E2f1 signature ( Fig . 5B ) . Results of enrichment analysis for all GO terms are provided in Table S2 . To assess the reproducibility and specificity of our results , we constructed TREG binding signatures for all 494 TF ChIP-seq datasets in the Genome Browser ENCODE tables [14] , [34] . Two gene expression profiles in our analysis ( E2+CHX and E2 ) were then systematically compared with 494 ENCODE TREG binding profiles . Top 10 most concordant profiles are shown in Fig . 6 . Results show that ENCODE ERα binding profiles correlates equally well with E2+CHX profile as did our original TREG profile ( Fig . 6A ) . Furthermore , all five ENCODE ERα binding profiles correlated better with E2+CHX profile than any other ENCODE profile . Similarly , ENCODE binding signatures most concordant with E2 profile ( Fig . 6B ) included E2F4 , E2F1 and MYC which are all known to be important cell cycle regulators . The statistical significance of the concordance was again similar to the levels we observed with the E2f1 binding profile in mouse embryonic stem cells . These results indicate that reproducibility of TREG results across different ChIP-seq datasets and its ability to identify key transcriptional regulators for a given profile . Results of the concordance analysis for all ENCODE TREG profiles are in Table S3 . The ultimate goal of the TREG framework is to facilitate identification and characterization of signatures of TFs regulating disease-related differential gene expression profiles ( DRGEP ) . Here we demonstrate the power of TREG signatures and TREG binding scores in elucidating the faint signals of ERα activity in two complex DRGEPs , the response of MCF-7 cell line 24 hours after treatment with E2 [27] and differences between ER− and ER+ breast tumors [35] . In both of these DRGEPs , the signal of direct ERα regulation is “drowned out” by the strong secondary proliferation-related transcriptional signature , and the standard enrichment analysis of computationally predicted ERα targets in MSigDb fails to find evidence of ERα regulation ( Fig . 7 ) . However , the GRS concordance analysis with both TREG binding scores and TREG signatures are highly statistically significant , and the TREG signature which integrated binding and transcriptional evidence again shows the highest statistical significance of concordance ( Fig . 7 ) . Additional discussion of these results is provided in supplementary results ( Text S1 ) . We used the ERα TREG signature to mine a collection of differential gene expression profiles in GEO datasets ( GDS signatures ) , and differential gene expression profiles of small drug perturbations ( CMAP signatures ) [29] , for evidence of ERα regulatory activity . Fig . 8 shows differential gene expression levels of top 10 GEO profiles and top 10 drug perturbations based on the statistical significance of the concordance between the ERα TREG signature and each differential gene expression profiles . In both situations the top transcriptional profiles are obviously related to the ERα activity demonstrating the precision of the TREG signature in this setting . Additional results related more specifically to disease-associated GEO profiles are given in the supplementary results ( Text S1 ) . The problem of identifying functional TF targets that regulate gene expression , in a specific biological context , requires joint considerations of both TF DNA-binding data and the target gene's expression changes . We described a statistical framework for quantifying the evidence of TF-gene interaction from ChIP-seq data , and integrating them appropriate gene expression data to construct genome-wide signatures of TF activity . Two main findings of our study are that 1 ) TREG binding scores derived from ChIP-seq data alone are more informative than simple alternatives that can be used to summarize ChIP-seq data; and 2 ) TREG signatures that integrate the binding and gene expression data are more sensitive in detecting evidence of TF regulatory activity than available alternatives . We show that this advantage of TREG signatures can make the difference between being able and not being able to infer TF regulatory activity in complex transcriptional profiles . This increased sensitivity also showed to be critical in establishing connections between disease and drug signatures that would not be possible using currently available strategies . Identifying the role of specific TFs in producing disease-related transcriptional profiles is of vital importance for understanding the molecular mechanisms underlying disease phenotype . Although it is possible to obtain direct measurements of TF activity in disease samples [45] , such ChIP-seq profiling is technically challenging and systematic profiling of many different TFs is not feasible . Therefore , the ability to infer the role of a TF from the transcriptional profiles remains challenging . The most common strategy of implicating TF involvement is by computational analysis of genomics regulatory regions of differentially expressed genes [24]–[27] , or by searching for enrichment of known targets among differentially expressed genes [46] . Here we present an alternative strategy relying on direct concordance analysis between TREG signatures of TF activity and disease-related transcriptional profiles . When searching for evidence of regulation by the TF with functional binding sites in distant enhancers , such as ERα , and “messy” transcriptional signatures resulting from activity of multiple regulatory programs , our approach dramatically improves the precision of the analysis . Our results indicate that TREG signatures derived from in-vitro experiments ( ERα; MCF-7 cells ) , and even from a different organism ( E2f1; mouse ) provide effective means for analyzing transcriptional profiles derived from human tissue samples . This would indicate that TF binding profiles coming from any biological system under which TF shows signs of activity might be sufficiently informative to construct TREG signatures . In this context the recently released ENCODE project data [14] , [34] may be turned into a powerful tool for detecting TF activity . As a step in this direction , we have created 494 TREG binding profiles using the ENCODE ChIP-seq data and made it available from the support web-site ( http://GenomicsPortals . org ) . Complementary gene expression data generated by directly perturbing specific TFs , such as shRNA knock-downs and overexpression experiments can be used to construct TREG signatures . For example , transcriptional signatures of such systematic perturbations that is being generated by NIH LINCS project ( http://LincsProject . org ) could provide complementary transcriptional profiles for ENCODE ChIP-seq data . Our methods are complementary to methods used to analyze the recently released ENCODE project data [14] , [34] . For some experimental conditions , the ENCODE project provides additional data types that can be used in assessing the functionality of TF binding peaks , such as distribution of specific epigenetic histone modifications . For discussion on how to possibly incorporate this additional information within TREG methodology , please see supplemental discussion ( Text S1 ) . Up-regulated expression of proliferation genes is a hallmark of neoplastic transformation and progression in a whole array of different human cancers [47] . While the core transcriptional signature of proliferation is recognizable in a wide range of biological systems and diseases , the events and pathways that drive the transcriptional program of proliferation vary widely . Increased expression of proliferation-associated genes has been associated with poor outcomes in breast cancer patients [48]–[54] . However , the driver mechanisms in many aggressive cancer types are poorly understood . Inhibiting known driver pathways , such as ERα signaling in breast cancer often leads to treatment resistant tumors due to activation of alternative , poorly understood driver pathways [55] , [56] . Using the signatures of such “driver events/pathways” we can identify candidate drugs capable of inhibiting them . In our analysis of ERα activity in ER+ breast cancers we showed that such an approach can highlight connections between disease and drug candidates that would be missed by simply correlating disease and drug transcriptional signatures [28]–[30] . We assume that observed peaks consist of two populations: Functional peaks that are more likely to occur closer to TSS and whose distance to TSS is distributed as an exponential random variable with the parameter λ; and , non-functional peaks that are randomly occurring throughout the 2 million base pair genomic region centered around the TSS , and whose distances to TSS are distributed as a uniform random variable . The distances to TSS of all peaks are then distributed as a mixture of the exponential and the uniform distribution ( Fig . 1 , Eq1 ) , where π is the proportion of functional peaks among all observed peaks , a is the distance of a peak to the gene's TSS , is the probability density function ( pdf ) of the exponential random variable ( rv ) with the location parameter λ , and is the pdf of a uniform rv on the interval ( −106 , 106 ) . We use the standard Expectation-Maximization ( EM ) algorithm [57] to estimate the parameters of this mixture model ( π , λ ) for each TF . Given the estimates we calculate the posterior probability for peak i with distance ai from a TSS to belong to the population of “functional peaks” ( Fig . 1 Eq2 ) . Suppose now that for a gene g , ng is the number of peaks within the 1MB window around its TSS ( 1MB upstream to 1MB downstream ) , is the peak intensity ( ie , the maximum number of overlapping reads over all positions within the peak ) , and is the distance to TSS of the kth such peak ( k = 1 , … , ng ) . We define the TREG binding score for the gene g as the logarithm of weighted average of peak intensities , using the probability of the peak belonging to the population of “functional peak” ( ) as the weight ( Fig . 1 Eq3 ) . We assume that TREG binding scores come from two populations: Scores significantly greater than zero representing functional TF-gene interactions which are distributed as a Normal random variable; and , scores close to zero representing non-functional interactions which are distributed as an exponential random variable ( histogram in Fig . 1B ) . Assuming that the proportion of TREG binding scores corresponding to functional interactions is η , the distribution of all TREG binding scores is a mixture of Normal and exponential probability distribution functions ( Fig . 1 Eq4 ) , where S is the TREG binding score , is pdf of the exponential random variable with the location parameter ψ , and is the pdf of a Normal random variable with mean μ and variance σ2 . We again use the standard EM algorithm to estimate the parameters of this mixture model ( η , ψ , μ , σ2 ) for each TF . Given the estimates , the probability of a TREG binding score for gene g ( Sg ) being functional is defined as the probability of Sg belonging to the normal component ( Fig . 1 Eq5 ) . The set of TREG binding scores and associated probabilities of the score indicated functional TF-gene interaction for all gene in the genome ( Sg , pg ) , g = 1 , … , G , is the TREG binding profile . Additional discussion of motivations for the choice of specific distributions is provided in supplemental methods ( Text S1 ) . Details of the EM algorithm are provided in supplemental methods ( Text S1 ) . The enrichment of genes with high TREG and MPI scores among differentially expressed genes ( Table 1 , Fig . 2 ) was performed using the logistic regression-based LRpath methodology [58] . LRpath does not require thresholding on binding scores but uses such scores as the continuous variable that explains the membership of a gene in the “differentially expressed” category . Similarly , LRpath was used to analyze enrichment of differentially expressed genes among genes associated with GO terms in Fig . 5 and 6 . When performing concordance analysis between TREG binding profiles and the two differential gene expression profiles of interest ( E2+CHX and E2 ) ( Table 2 ) and constructing TREG signatures in Fig . 4 , 5 , and 6 , we used two-tailed p-values not distinguishing between induction and repression activity . When comparing TREG signatures with other DRGEPs ( Table 3 , and Fig . 7 and 8 ) , we account for directionality of gene expression changes by using single-tailed p-values for increase in gene expression . This is necessary to account for the directionality of the concordance between the TREG signature and the DRGEPs . The ERα TREG signatures for this analysis was constructed by the GRS concordance analysis ( Fig . 1D ) between ERα TREG binding profile and the single tailed p-values for statistically significant up-regulation of gene expression after E2+CHX treatment of MCF-7 cell line . The genes used for plotting heatmaps in Fig . 8 were then selected based on the gene-specific p-values of concordance ( p-value ( eg ) , Fig . 1D ) being <0 . 001 ( Table S5 ) . The concordance between this ERα TREG signature , and GEO/CMAP transcriptional signatures was performed again using the GRS analysis . The description , location and processing of the ChIP-seq and gene expression datasets are provided in supplemental methods ( Text S1 ) . All computational methods are implemented in the R package treg which can be downloaded from our web site ( http://GenomicsPortals . org ) . The package also contains processed ChIP-seq data for ERα [10] , E2f1 and 15 other transcription factors [32] , as well as TREG signatures for ERα and E2f1 , and transcriptional signatures derived from GEO GDS datasets and CMAP drug signatures . We have previously described derivation of CMAP signatures [31] . All functional enrichment analyses were performed using the LRpath methodology [58] as implemented in the R package CLEAN [59] .	Knowing transcription factors ( TF ) that regulate expression of differentially expressed genes is essential for understanding signaling cascades and regulatory mechanisms that lead to changes in gene expression . We developed methods for constructing gene-level scores ( TREG binding scores ) measuring likelihood that the gene is regulated based on the generative statistical model of ChIP-seq data for all genes ( TREG binding profile ) . We also developed methods for integrating TREG binding scores with appropriately matched gene expression data to create TREG signatures of the TF activity . We then use TREG binding profiles and TREG signatures to identify TFs involved in the disease-related gene expression profiles . Two main findings of our study are: 1 ) TREG binding scores derived from ChIP-seq data are more informative than simple alternatives that can be used to summarize ChIP-seq data; and 2 ) TREG signatures that integrate the binding and gene expression data are more sensitive in detecting evidence of TF regulatory activity than commonly used alternatives . We show that this advantage of TREG signatures can make the difference between being able and not being able to infer TF regulatory activity in complex transcriptional profiles . This increased sensitivity was critically important in establishing connections between disease and drug signatures .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[]	2013	Genome-Wide Signatures of Transcription Factor Activity: Connecting Transcription Factors, Disease, and Small Molecules
Dengue laboratory diagnosis is essentially based on detection of the virus , its components or antibodies directed against the virus in blood samples . Blood , however , may be difficult to draw in some patients , especially in children , and sampling during outbreak investigations or epidemiological studies may face logistical challenges or limited compliance to invasive procedures from subjects . The aim of this study was to assess the possibility of using saliva and urine samples instead of blood for dengue diagnosis . Serial plasma , urine and saliva samples were collected at several time-points between the day of admission to hospital until three months after the onset of fever in children with confirmed dengue disease . Quantitative RT-PCR , NS1 antigen capture and ELISA serology for anti-DENV antibody ( IgG , IgM and IgA ) detection were performed in parallel on the three body fluids . RT-PCR and NS1 tests demonstrated an overall sensitivity of 85 . 4%/63 . 4% , 41 . 6%/14 . 5% and 39%/28 . 3% , in plasma , urine and saliva specimens , respectively . When urine and saliva samples were collected at the same time-points and tested concurrently , the diagnostic sensitivity of RNA and NS1 detection assays was 69 . 1% and 34 . 4% , respectively . IgG/IgA detection assays had an overall sensitivity of 54 . 4%/37 . 4% , 38 . 5%/26 . 8% and 52 . 9%/28 . 6% in plasma , urine and saliva specimens , respectively . IgM were detected in 38 . 1% and 36% of the plasma and saliva samples but never in urine . Although the performances of the different diagnostic methods were not as good in saliva and urine as in plasma specimens , the results obtained by qRT-PCR and by anti-DENV antibody ELISA could well justify the use of these two body fluids to detect dengue infection in situations when the collection of blood specimens is not possible . Dengue virus ( DENV; family Flaviviridae , genus Flavivirus ) , is a mosquito-borne , enveloped , single stranded positive-sense RNA virus . Until recently , four serologically related but antigenically and genetically distinct dengue viruses ( DENV-1 , -2 , -3 , and -4 ) causing disease in human were known , but a fifth serotype was recently discovered in Malaysia [1 , 2] . Over the last decades , DENV has become the most important arthropod-borne virus affecting humans . Several factors such as rapid urbanization , failure to control mosquitoes and rapid progress in air transportation have contributed to the emergence of endemic dengue in 128 countries in the world [3 , 4] . In 2009 , the World Health Organization ( WHO ) estimated that 2 . 5 billion people were living in areas at high risk for infection , among which 50 million were infected annually [1] . However due to weak disease surveillance , low level of reporting of cases and difficulties in diagnosis , the true incidence and burden of dengue are very likely under estimated . Using new modeling approaches , Bhatt et al . estimated that approximately 3 . 97 billion people were living in areas where DENV is circulating and that almost 400 million cases of dengue occurred every year worldwide [4] . While most infections are asymptomatic or result in a mild febrile illness , the virus is capable of producing life-threatening dengue hemorrhagic fever , dengue shock syndrome , and non-specific complication of systemic diseases ( e . g . encephalitis , hepatitis ) [1] . The course of dengue illness can be divided in three main phases: the febrile phase , the critical phase and the recovery phase . Severe clinical disease manifestations such as fluid leakage , bleeding and shock occurs during the critical phase which begins around day 4–7 after the onset of fever , when the temperature drops under 38°C , and lasts usually between 48 to 72 hours . During the critical phase , the condition of patients experiencing a severe form of the disease worsens as a result of plasma leakage whereas the condition of patients with a non-severe disease improves . Direct diagnosis of DENV infection is based on virus isolation , detection of the viral genome by reverse transcription polymerase chain reaction ( RT-PCR ) or detection of NS1 antigen . Indirect diagnosis using serological methods to detect anti-DENV IgM and IgG is commonly employed , while IgA tests remain less commonly used . The selection of diagnostic methods depends greatly on the time-point of the sample collection during the course of the disease . RT-PCR and virus isolation require a blood sample collected during the early febrile phase of the disease ( 0–5 days after the onset of fever ) [1] . A sample obtained during the early phase is also preferred for NS1 detection , but in patients experiencing a primary infection the NS1 antigen remains detectable for nine days or more after the onset of fever [5 , 6] . Serological methods can be used later during the course of the disease . IgM and IgA , however , can persist in blood during several weeks or even months after the infection while IgG may persist for decades . Differentiating between an acute and a recent DENV infection by these methods may therefore be challenging . Currently almost all laboratory diagnostic methods require a blood sample that may be difficult to obtain in children , the population which is most commonly affected by dengue in endemic regions , especially in field conditions or outbreak settings . Saliva and urine samples could be used as surrogates for blood as the collection of these body fluids is non-invasive and better accepted by patients , does not require medically-trained staff and the samples are easier to process as they require only limited laboratory facilities . Diagnosis using urine and saliva samples has already been explored for several other viral infections such as HIV , Hepatitis A , B and C and rubella [7–13] . Urine specimens seem suitable for the diagnosis of infection by West-Nile virus ( WNV ) and Zika virus ( ZIKV ) –two other flaviviruses—as viral genome of both viruses can be detected longer in urine than in serum [14 , 15] and infectious WNV and ZIKV can be recovered from urine [16–18] . Previous studies conducted on a limited number of samples have demonstrated that urine and saliva could be used as an alternative to blood for serological or virological diagnosis of dengue [19–29] . The purpose of this study was to describe the excretion profiles of the anti-DENV antibodies ( IgG , IgM and IgA ) , the NS1 antigen , the viral genome and of the infectious virus in plasma , saliva and urine specimens obtained from a large number of Cambodian children with a confirmed dengue infection . In this study , we aimed to evaluate the possibility of using saliva and urine as alternatives to serum or plasma samples . We also aimed to assess if urine and/or saliva specimens could contribute to predict progression towards a severe form of the disease , as suggested by Chuansumrit et al . [21] , or if the use of such specimens increases the time window for DENV infection confirmation as suggested by several previous studies [20 , 22 , 24] . Plasma , saliva and urine samples used in this study were collected in Cambodia in 2013 , during the DENFREE ( DENgue research Framework for Resisting Epidemics in Europe ) study [30] . Samples were prospectively obtained from children hospitalized with clinically-suspected dengue in one of the three hospitals of the Kampong Cham province participating to the DENFREE study . Dengue infection was confirmed in patients through a combination of: viral RNA detection by quantitative RT-PCR ( qRT-PCR ) ; isolation of the virus in cell culture; detection of the NS1 protein in the blood; by evidencing a IgM seroconversion in paired plasma and/or a fourfold increase of the antibody titer measured by hemagglutination inhibition assay ( HIA ) in paired plasma collected at least seven days apart . Disease severity was assessed for each patient according to the 1997 WHO criteria and to the 2009 WHO guidelines [1 , 31] using clinical , biological and paraclinical examination data recorded at admission and throughout the entire hospitalization period . Patients were classified into 2 groups: non-severe dengue ( dengue fever ( DF ) according to the 1997 guidelines and DF without warning signs according to the 2009 WHO guidelines ) and severe dengue ( dengue hemorrhagic fever ( DHF ) and DSS ( dengue shock syndrome ) according to the 1997 guidelines and dengue with warning signs as well as severe dengue according to the 2009 WHO recommendations ) . Urine and saliva were collected daily during hospitalization and then one week , two weeks , three weeks , one month and three months after the discharge from hospital . Plasma specimens were collected at three time-points during hospitalization: on admission , during hospitalization and at the time of hospital discharge . If the patient or guardians gave consent , blood was also collected 30 and 90 days after leaving the hospital . Urine specimens were collected in 50 ml sterile tubes . Due to the large volume of saliva required for the analyses , the saliva specimens were obtained by direct spitting into 1 . 8 ml sterile tubes . If possible , children were asked to spit into two separate tubes , an empty one and a pre-filled one with 100µl of viral transport medium ( VTM ) , until minimal volumes of 100 µl and 200 µl were reached in the empty and the pre-filled tubes , respectively . All clinical samples were stored at -80°C prior to testing . In addition to prevent degradation of the RNA and of the virus infectivity , the freezing process was also beneficial to reduce the viscosity of the saliva . Only patients with well-documented medical records allowing a clear assessment of the disease severity were included in this study . The clinical specimens obtained from 20 patients hospitalized for a non-dengue febrile illness were also used to assess the specificity of the different diagnostic methods in urine and saliva samples . These controls showed no biological evidences of on-going DENV infection ( DENV qRT-PCR negative , NS1 test negative and no anti-DENV antibodies in paired plasma [admission and discharge] ) . The viral RNA was extracted from 140 µl of plasma , 280 µl of saliva mixed with VTM and 280 µl of concentrated urine , using the QIAmp Viral RNA Mini kit ( Qiagen , Germany ) . Urine samples were concentrated using the 100K Microsep ultrafiltration device ( Pall , USA ) to convert an initial volume of 5 ml urine to a final volume of 280 µl of concentrated urine . After extraction , the DENV nucleic acids present in plasma were detected by a serotype-specific quantitative multiplexed real-time RT-PCR ( qRT-PCR ) as previously described by Hue et al . [32] . For urine and saliva , the multiplex qRT-PCR was replaced by a monoplex qRT-PCR as the serotype was already identified in the corresponding plasma . Results were expressed in equivalents to complementary DNA ( cDNA ) copies per ml . The limit of detection in plasma was 350 copies/ml for DENV-1 , 75 copies/ml for DENV-2 , 350 copies/ml for DENV-3 and 715 copies/ml for DENV-4 . The limit of detection was 200 copies/ml , 50 copies/ml , 200 copies/ml and 350 copies/ml for DENV-1 , -2 , -3 and -4 , respectively , with the qRT-PCR in urine and plasma specimens spiked with quantified synthetic plasmids . As DENV viremia lasts only a few days after the onset of fever ( DAOF ) , the samples selected for the detection of the viral genome were mainly those collected during hospitalization . Additional urine and saliva specimens collected one to three weeks after discharge from the hospital were also added as some data previously published suggested that the DENV genome could be detected in urine until day 16 after the onset of fever [22] . C6/36 cells were used in attempt to isolate DENV from urine and saliva samples . Before inoculation , urine samples were dialyzed and concentrated in PBS using the 100K Microsep ultrafiltration system ( Pall , USA ) and then filtered through 0 . 2-μm membrane ( Nalgene Thermo Scientific , USA ) . Saliva diluted in VTM as well as urine samples were diluted 1/2 with L15 Leibovitz Medium ( Sigma Aldrich , Germany ) containing 2% of fetal calf serum ( Gibco Life Technology , USA ) . Final volumes of 300 µl of diluted specimens , or controls , were inoculated into 12-well plates containing 100% confluent C6/36 cells and incubated for 1 hour at 28°C . After incubation , 1 . 7 ml of medium was added to each well and the plates were incubated at 28°C . After 7 days , cells were harvested and DENV was detected by an immunofluorescence assay using serotype-specific monoclonal antibodies as described previously [33] . For samples that tested negative , two additional passages on C6/36 were performed before concluding that DENV did not replicate . Our positive controls consisted of urine and saliva specimens obtained from healthy individuals as well as VTM spiked with infectious virus at a final concentration of 3 to 7 log10 cDNA copies/ml . Spiked urine controls were dialyzed and concentrated as for the patient’s urine samples . A capture ELISA was used to detect NS1 in plasma , non-concentrated urine and undiluted saliva . In plasma and urine , the NS1 protein was detected by the two-step ELISA method described by Alcon et al . [6] but after replacing polyclonal antibodies by monoclonal antibodies . For the saliva specimens , a one-step capture ELISA was used . All the protocols are detailed in S1 Table . Briefly , in the one-step ELISA developed for NS1 detection in saliva , the sample and the conjugated antibody were incubated together for 2 hours . In the two-step ELISAs ( used for plasma and urine specimens ) , samples were first incubated for one hour and then after a washing step , the conjugated antibody was added and incubated for one hour . For the patients infected with DENV-1 , the NS1 protein was quantified by ELISA using serial two-fold serial dilutions of a solution containing a known concentration of affinity-purified DENV-1 recombinant NS1 protein expressed in HEK293T cells ( kindly provided by Dr . Marie Flamand , Institut Pasteur , Paris , France ) . The standard curve was linear between 0 . 5 ng/ml and 16 ng/ml for all three NS1 assays . A result was considered positive if the OD was greater than twice the mean OD value measured in 20 negative controls ( 20 healthy children enrolled during the community study of DENFREE project ) . Two positive controls , a weak one and a strong one , and one negative control were used to validate the results of each plate . Inter- and intra-precision of the different ELISAs are shown in S4 and S5 Tables . The samples used for the NS1 protein detection were mainly those collected between DAOF 1 and 10 in hospitalized patients . In addition , 29 urine and 7 saliva specimens collected between DAOF 10 and 16 were tested . In-house capture ELISAs were used to detect anti-DENV IgM ( MAC-ELISA ) and IgA ( AAC-ELISA ) in plasma , non-concentrated urine and undiluted saliva specimens . For plasma , MAC-ELISA and AAC-ELISA were performed as described previously [33 , 34] . For urine and saliva samples , minor modifications to MAC-ELISA and AAC-ELISA protocols were adopted . A 100 µl/well format was used for plasma and urine samples , but as only a limited volume of saliva was available , a 50 µl/well format was used for this body fluid . The protocols used are detailed in S2 and S3 Tables . One negative control , one weakly positive control and one strongly positive control were tested in each plate in order to validate each plate of tests . Inter- and intra-precision of the different assays are shown in S4 and S5 Tables . Results were expressed as ΔODs ( OD of the sample incubated in the presence of antigen—OD of the sample incubated in the absence of antigen ) . A result was considered as positive when the ΔOD was higher than 0 . 05 for the saliva-based AAC-ELISA and higher than 0 . 1 for all the other ELISAs . The cut-offs were determined by testing the samples from 20 healthy children who showed no biological evidences of a previous DENV infection ( DENV qRT-PCR negative , NS1 ELISA negative and HI titer = 0 ) . Cut-off values are shown in S1 Fig . In-house indirect ELISAs were used to detect anti-DENV IgG in plasma , non-concentrated urine and undiluted saliva specimens [35] . A 100 µl/well format was used for the plasma and urine-based ELISAs , while a 50 µl/well format was used for the saliva-based ELISA , due to the limited volume of samples available . The protocols used here are described in S6 Table . Results were expressed as ΔODs . Results were considered as negative when the ΔOD was below 0 . 2 for the ELISA in plasma ( cut-off determined by comparing ELISA results to those obtained by HIA with the same samples ) and below 0 . 1 for the ELISAs in urine and saliva . Cut-off values were determined by testing the samples obtained from 20 healthy children and are described in S1 Fig . For the evaluation of the antibody detection assays , the samples tested were collected during the hospitalization as well as during the follow-up period after the patient was discharged from the hospital . The panel contained samples collected as early as the day of the onset of fever as well as specimens obtained up to 103 days after the beginning of the disease . Statistical analysis was performed using STATA version 11 . 0 ( StataCorp , USA ) . A significance was assigned at P<0 . 05 for all parameters and were two-sided . The statistical differences between various categorical groups were detected using Chi-squared test . Correlation coefficients between two continuous variables were calculated using Spearman’s rank correlation test . Uncertainty was expressed by 95% confidence intervals ( CI95 ) . In order to identify explanatory variables associated with the detection of the viral genome , the NS1 antigen or the anti-DENV antibodies in the three different fluids , multivariate analysis using a boosted regression tree ( BRT ) approach was used . BRT is a method combining regression trees with weak individual predictive performances into a single model with higher performance . We used BRT because it is capable of dealing with complex responses , including non-linear relationships and interactions between explanatory variables [36] , and we were expecting such complex relationships between our explanatory and response variables . As part of the final model , the BRT assesses the relative importance ( RI ) of each explanatory variable based on the number of times a variable is used in all trees and its contribution to the final model improvement . A higher RI of a predictor indicates a stronger influence on the response in question ( i . e . the detection of each diagnostic marker , in this study ) . The effect of each explanatory variable on the response can be visualized by the use of partial dependence plots . For each model with a binary response variable , a Receiver-Operating Characteristic ( ROC ) curve was constructed and the area under the ROC curve ( AUC ) was calculated . The AUC ranges between 0 . 5 for random prediction to 1 , the higher the AUC the better the model performance . For models with a continuous response variable , the performance of the model was estimated by a correlation coefficient between observed and predicted datasets . This correlation coefficient ranges between 0 ( no correlation ) and 1 ( perfect correlation ) , with values superior to 0 . 5 suggesting a high correlation . The analysis was carried out using the “dismo” package ( version 1 . 0–5 ) [37] and the “gbm” package ( version 2 . 1 ) [38] under the R statistical environment ( R Foundation , Vienna , Austria ) . A learning rate of 0 . 001 , a bag fraction of 0 . 5 and a tree complexity of 3 were used . The explanatory variables used are summarized in Table 1 . The dengue serotype was not included in the analyses as most of the patients included in this study were infected with a DENV-1 strain during the 2013 epidemic in Cambodia . The children’s legal representatives signed a written consent before the enrolment of the patient . The DENFREE project was approved by the Cambodian National Ethics Committee for Health Research ( authorization no . 063NECHR ) . Overall , 85 . 4% ( 323/378 , CI95 = [81 . 5–88 . 8] ) of the plasma specimens obtained from 144 patients with a confirmed acute DENV infection and sampled from the day of onset of fever until the 10th day after fever onset , tested positive . Out of the 442 urine and the 562 saliva samples obtained between the day of fever onset until the 28th day after the onset of fever ( DAOF 28 ) from 118 and 132 patients , respectively , 41 . 6% ( n = 184 , CI95 = [37 . 0–46 . 4] ) and 39% ( n = 219 , CI95 = [34 . 9–43 . 1] ) tested positive by qRT-PCR . The qRT-PCR results are presented based on days of sampling after the onset of fever ( Fig 1 , S7 Table , S9 Table ) . The proportion of urine samples positive by qRT-PCR increased with time , starting from a minimum of 13 . 6% during the first 2 days after fever onset to a maximum of 66 . 6% at DAOF 9–10 . At the latest time-points , the proportion of urine samples positive by qRT-PCR was approximately the same as the proportion of RNA-positive plasma specimens ( 71 . 4% and 62 . 5% at DAOF 9–10 , respectively ) . DENV RNA remained detectable in one urine sample at DAOF 16 . The detection rate of viral RNA by qRT-PCR in saliva samples followed the decreasing curve of the RNA detection rate observed in the plasma specimens . The highest proportion of saliva samples that tested positive for DENV RNA was observed between the first day and the 4th day of fever ( 60 . 5% at DAOF 0–2 , 63 . 9% at DAOF 3 and 58 . 5% at DAOF 4 ) whereas plasma samples collected at the same time-points were RNA-positive in 96% to 100% of the cases . The RNA detection rate in saliva decreased thereafter ( 44 . 2% , 30 . 9% , 23 . 9% , 9 . 9% , 16% and 8 . 3% at DAOF 5 , 6 , 7 , 8 , 9 and 10 , respectively ) , and all samples tested negative after DAOF 10 . By comparison with the percentage of plasma samples positive by qRT-PCR , the sensitivity of qRT-PCR in saliva was on average 46 . 2% lower , with a minimal difference of 36 . 1% at DAOF 3 and a maximal difference of 55 . 4% at DAOF 9 . In total , 57 . 6% ( 68/118 ) of the patients had at least one of their urine samples that tested positive by qRT-PCR and 68 . 9% of them ( 92/132 ) had at least one saliva sample that tested positive . When only considering the 243 matched plasma , urine and saliva samples ( i . e clinical specimens collected from the same patients at the same time-points ) obtained between DAOF 1 and DAOF 12 , the proportion of specimens that tested positive by qRT-PCR was 90 . 5% ( 220/243 , CI95 = [86 . 1–93 . 9] ) , 39 . 9% ( 97/243 , CI95 = [33 . 7–46 . 4] ) and 49 . 8% ( 121/243 , CI95 = [43 . 3–56 . 3] ) for plasma , urine and saliva , respectively ( Table 3 ) . Three urine samples that tested positive had a corresponding matched plasma sample that tested negative . These 3 urine samples were all collected at the time of hospital discharge: one at DAOF 6 and the two others at DAOF 10 . Testing by qRT-PCR the urine and the plasma specimens led to an increase of the diagnostic sensitivity: 90 . 5% of sensitivity if the plasma was tested alone , 91 . 8% when both urine and plasma specimens were tested in parallel . The diagnostic sensitivity increased from 85% for plasma samples alone to 86 . 7% for the plasma and urine combination with samples collected at DAOF 6–7 and from 70 . 7% to 75 . 6% during the second week after fever onset . All subjects whose saliva samples tested positive also had detectable RNA levels in their corresponding plasma samples . If both the urine and the saliva samples collected at the same time-points were screened for DENV-RNA the diagnostic sensitivity increased to 69 . 1% ( 168/243 , CI95 = [62 . 9–74 . 9] ) . If saliva and urine specimens were tested in parallel instead of testing the plasma , the overall diagnostic sensitivity decreased by 21 . 4% ( by 16 . 7% if the samples were collected at DAOF 0–1 or at DAOF 6–7 , by 22 . 4% at DAOF 3–4 , by 23% at DAOF 4–5 , and by 24 . 4% for samples obtained during the 2nd week after the onset of fever ) . All clinical specimens obtained from 20 patients hospitalized for a non-dengue febrile illness tested negative . On average , 8 . 2 log10 cDNA copies/ml ( min = 2 . 4 log10 and max = 9 . 9 log10 cDNA copies/ml ) were detected in plasma samples collected between DAOF 0 and 10 , 3 . 5 log 10 cDNA copies/ml ( min = 1 log 10 and max = 5 . 2 log10 cDNA copies/ml ) were measured in urine specimens obtained between DAOF 1 and 16 , and 4 . 6 log10 cDNA copies/ml ( min = 1 . 1 log10 and max = 6 . 1 lon10 cDNA copies/ml ) were detected in saliva specimens collected between DAOF 1 and 10 . The results from 261 , 167 and 164 sequential positive plasma , urine and saliva samples collected between DAOF 1 and 10 and obtained from 103 , 57 and 54 patients , respectively , were used to estimate the mean viral load according to the DAOF . The results are presented in Fig 2 . Briefly , before the 4th day after the onset of fever , the mean viral load in plasma was at its maximum , at approximately 8 . 5 log10 cDNA copies/ml . Thereafter , the mean viral load in those samples that tested positive progressively decreased over time . The mean viral load in urine increased from 2 . 1 to 3 . 8 log10 cDNA/copies from the 2nd day until the 5th day after onset of fever . Then it decreased very slowly until DAOF 9 before significantly dropping at DAOF 10 . Until DAOF 6 , the mean viral load in saliva specimens that tested positive by qRT-PCR was approximately 3 . 5 log10 lower than in plasma . From DAOF 7 , the mean viral load in saliva continued to only very slightly decrease and at DAOF 9 reached the same level than in plasma . The viral load measured in saliva correlated with the viral load measured in plasma ( Spearman coefficient = 0 . 51 , p-value<0 . 001 ) whereas there was no correlation between viral loads in plasma and in urine ( p-value = 0 . 28 ) . In total , 15 urine and 15 saliva samples that tested positive by qRT-PCR with a high viral load , i . e . between 3 . 5 and 5 log10 cDNA copies/ml , were inoculated onto C6/36 mosquito cells . Even after 3 passages , all cell cultures remained negative by immunofluorescence assay . DENV was isolated only from VTM , urine and saliva controls spiked with virus at a concentration equal to or greater than 4 log10 cDNA copies/ml but not when the initial virus concentration was lower . No cell death was observed in the wells inoculated with the three different negative controls i . e . , cell culture medium , dialyzed urine samples and saliva samples obtained from healthy donors . A total of 856 urine and 688 saliva samples obtained from 193 and 197 patients with confirmed dengue were tested . In total , 14 . 5% ( n = 124 , CI95 = [12 . 2–17 . 0] ) of the urine samples and 28 . 3% ( n = 195 , CI95 = [25–31 . 9] ) of the saliva samples tested positive by NS1 ELISA . By comparison , 63 . 4% ( 338/533 , CI95 = [59 . 2–67 . 5] ) of all plasma samples obtained from hospitalized patients and included in this study tested positive for NS1 ( S7 Table , S9 Table ) . The positivity rates of NS1 antigen detection in plasma , urine and saliva samples by day of sampling after the onset of fever are presented in Fig 3 and S7 Table . The NS1 protein was detected in 15 . 4% of the saliva and in 3 . 2% of the urine specimens at the beginning of the disease . This percentage increased to reach a maximum of 42 . 6% in saliva samples and 24 . 3% in urine specimens at DAOF 4 . The proportion of NS1-positive samples then decreased slowly until DAOF 10 and no sample tested positive after that time-point . In plasma , the proportion of samples that tested positive for NS1 was the highest ( 88 . 2% ) at DAOF 3 and then progressively decreased over time . At DAOF 9 , the last time-point at which the plasma specimens were tested , 22 . 2% of the samples tested positive by NS1 ELISA . In total , 46 . 2% ( 91/197 ) and 37 . 8% ( 73/193 ) of the patients tested positive in NS1 capture ELISA in at least one of their saliva and urine samples . Among the 91 patients that had at least one NS1-positive saliva sample , the first specimen collected at admission tested positive in 67% ( n = 61 ) of the cases . Among the 73 patients that had at least one of their urine samples that tested positive , only 41 . 1% ( n = 30 ) had a positive urine sample at the time of hospital admission . Using the results obtained by testing the sequential biological samples collected throughout the hospitalization , the median first day of the NS1 detection in urine was determined for 30 patients at DAOF 5 ( mean = 5 . 1 ) , with a minimum at DAOF 3 and a maximum at DAOF 9 . Based on a subset of 17 patients , we were able to establish that the median duration of NS1 antigen detection in the urine of a DENV-infected patient was 1 day ( mean = 1 . 5 ) , with a maximum of 3 days . The NS1 antigen was detected in 70 . 9% ( 202/285 ) , 27 . 4% ( 46/168 ) and 47 . 7% ( 73/153 ) of the plasma , urine and saliva specimens that also tested positive by qRT-PCR . The NS1 ELISA was also positive in 20 . 9% ( 9/43 ) , 10 . 7% ( 21/196 ) and 20 . 7% ( 42/203 ) of the plasma , urine and saliva samples that tested negative by qRT-PCR . NS1 ELISA was used to test a total of 314 plasma , urine and saliva samples obtained from the same patients at the same time-points , between DAOF 1 and DAOF 12 . The results were positive in 69 . 4% ( 218/314 , CI95 = [64 . 0–74 . 5] ) , 15 . 9% ( 50/314 , CI95 = [12 . 1–20 . 4] ) and 30 . 6% ( 96/314 , CI95 = [25 . 5–36 . 0] ) of the plasma , urine and saliva samples , respectively ( Table 3 ) . An overall diagnostic sensitivity of 34 . 4% ( 108/314 , CI95 = [29 . 2–39 . 9] ) was obtained by combining the NS1 results of urine and saliva . If saliva and urine samples were tested together instead of plasma , diagnostic sensitivity decreased to 16 . 7% , 34 . 9% , 49 . 5% , 32 . 2% and 10 . 9% , at day 0–1 , 2–3 , 4–5 , 6–7 and during the second week after the onset of fever , respectively , while sensitivity in plasma was 83 . 3% , 91 . 6% , 79% , 55 . 9% and 29 . 1% at the same time-points . Each time a saliva or a urine sample tested positive , the NS1 antigen was also detected in the corresponding plasma specimen . All clinical specimens obtained from 20 patients hospitalized for a non-dengue febrile illness tested negative . Between DAOF 1 and 10 , the average concentration of NS1 protein detected in the clinical specimens was 889 ng/ml ( min = 0 . 8 ng/ml and max = 8 µg/ml ) in plasma , 7 . 2 ng/ml ( min = 0 . 5 ng/ml and max = 60 ng/ml ) in urine and 3 . 8 ng/ml ( min = 0 . 5 ng/ml and max = 41 . 5 ng/ml ) in saliva . The mean NS1 concentration in the plasma specimens increased from 800 ng/ml at DAOF 1–2 to reach a maximal mean concentration of 1 . 2 µg/ml at DAOF 5–6 . It then decreased to reach a concentration of 400 ng/ml at DAOF 8–10 ( S2 Fig ) . The mean NS1 concentration in urine was stable between DAOF3 and 5 ( 7 ng/ml ) and increased at DAOF 6 ( 14 ng/ml ) and DAOF 7 ( 26 ng/ml ) ( S2 Fig ) . No significant variation in NS1 concentration was observed over time in saliva samples ( S2 Fig ) . A total of 100 urine specimens were initially tested by MAC-ELISA and since all of them were negative we decided not to include additional samples to the evaluation panel . A total of 1493 and 1483 urine samples were used to evaluate the performances of the IgG and IgA ELISAs in this biological fluid . Out of the 1489 saliva samples available for serological evaluation , 1123 , 1395 and 1101 were used for the IgG , IgM and IgA assays , respectively . A total of 766 , 678 and 778 plasma samples were tested by indirect IgG ELISA , MAC-ELISA and AAC-ELISA , respectively . Anti-DENV IgM were detected in 36% ( 244/678 , CI95 = [32 . 4–39 . 7] ) and 38 . 1% ( 531/1395 , CI95 = [35 . 5–40 . 7] ) of all the plasma and saliva samples , respectively , while none of the urine specimens ( 0/100 ) tested positive by MAC-ELISA . Overall 37 . 4% ( 291/778 , CI95 = [34 . 0–40 . 9] ) of the plasma , 26 . 8% ( 397/1483 , CI95 = [24 . 5–29 . 1] ) of the urine samples and 28 . 6% ( 315/1101 , CI95 = [26 . 0–31 . 4] ) of the saliva samples tested positive by AAC-ELISA . In total , 54 . 4% ( 417/766 , CI95 = [50 . 8–58 . 0] ) of the plasma samples , 38 . 5% ( 575/1493 , CI95 = [36 . 0–41 . 0] ) of the urine samples and 52 . 9% ( 594/1123 , CI95 = [49 . 9–55 . 8] ) of the saliva specimens tested positive by indirect IgG ELISA . Fig 4 and S7 Table show the percentage of positive results obtained for anti-DENV IgM , IgA and IgG detection in urine , saliva and plasma samples according to DAOF . The positivity rates obtained for IgM , IgA and IgG antibodies detection in the saliva samples were slightly lower than the ones observed for these antibodies in plasma specimens , but both kinetics were in general very similar . MAC-ELISAs sensitivity increased until DAOF 7 , to reach 86% and 73% in plasma and saliva , respectively . The sensitivity remained stable during the second week of the disease and then decreased gradually over the following weeks . Six weeks after the onset of the fever , 16% and 6% of the plasma and saliva samples tested positive . At three months , less than 2% of the plasma and saliva samples still contained detectable levels of anti-DENV IgM . The profiles of the sensitivity kinetic curves for anti-DENV IgA were similar in plasma , urine and saliva specimens . The sensitivity increased until the second week after the onset of fever to reach a maximum of 90% in the plasma and 65% in both the urine and the saliva specimens . The detections rates in urine and saliva then decreased and six weeks after the beginning of the infection only 3% of the urine and saliva samples still tested positive . The anti-DENV IgA antibody was detected in 30% and 10% of the plasma samples collected six weeks and three months after the onset of fever , respectively . All of the three kinetic curves of the anti-DENV IgG antibodies detection also had a similar overall profile . The proportion of plasma samples that tested positive for IgG quickly increased after the onset of the disease and between the three last collection points the values varied only slightly from 82% at week 2 to 74% at month 3 after fever onset . The percentage of urine samples that tested positive for IgG during the first week was on average 11% lower compared to the percentage of plasma samples that were IgG-positive . Globally both kinetic curves followed the same trend . After a peak at 70 . 2% during the second week of the disease , the curve demonstrated a regular decrease . In total , 25 . 9% of the samples tested positive one month after the infection and only 10% after three months . The IgG detection rate in saliva increased continuously from the 1st day of fever until the second week when it reached a plateau around 80% . The sensitivity of the IgG test in saliva samples during the first two weeks of the disease was on average 7% lower compared to the sensitivity of the assay in plasma specimens . From week 5 , the detection rate began to decrease and three months after the infection , 46% of the saliva samples remained positive for anti-DENV IgG . A total of 514 matched plasma , urine and saliva samples collected between DAOF 1 and DAOF 103 were tested by AAC-ELISA . The results were positive in 39 . 9% ( 205/514 , CI95 = [35 . 6–44 . 3] ) , 23 . 3% ( 120/514 , CI95 = [19 . 8–27 . 2] ) and 24 . 1% ( 124/514 , CI95 = [20 . 5–28 . 1] ) of the plasma , urine and saliva samples , respectively ( Table 3 ) . Concurrent testing of urine and saliva specimens by AAC-ELISA increased sensitivity by 6 . 7% and 5 . 9% compared to testing the urine or saliva samples alone and resulted in an overall decrease of the sensitivity to detect anti-DENV IgA by 10% , compared to AAC-ELISA in plasma ( decrease by 2 . 3% for samples collected at day 2–3 , by 7 . 5% at day 4–5 , by 15 . 1% at day 6–7 , by 7 . 3% during the 2nd week , by 25 . 8% six week after and by 5 . 5% three months after the onset of fever ) . A total of 563 plasma , urine and saliva samples obtained from the same patients at the same time-points , between DAOF 1 and DAOF 103 , were tested by IgG indirect ELISA . Anti-DENV IgG were detected in 56 . 8% ( 320/563 , CI95 = [52 . 6–61 . 0] ) , 32 . 1% ( 181/563 , CI95 = [28 . 3–36 . 2] ) and 42 . 8% ( 241/563 , CI95 = [38 . 7–47 . 0] ) of the plasma , urine and saliva samples , respectively ( Table 3 ) . An overall detection rate of 49 . 7% ( 280/563 , CI95 = [45 . 5–53 . 9] ) was obtained by combining the IgG results of paired urine and saliva specimens . The difference between the sensitivity of the IgG test in plasma and in urine and saliva samples varied from a minimum of 9 . 1% ( DAOF 2–3 ) and 3 . 4% ( DAOF 2–3 ) to a maximum of 63 . 6% ( month 3 ) and 30 . 6% ( month 3 ) , respectively . The difference between the sensitivity of the assay in plasma and the one obtained by combining results of paired urine and saliva was below 10% at all time-points ( minimal difference: 0% at DAOF 6–7; maximal difference: 7 . 5% at DAOF 4–5 ) , except at DAOF 0–1 and 3 months after the onset of fever when the difference reached 18 . 2% and 28 . 4% , respectively . The BRT analyses demonstrated that the three explanatory variables ( “daof” , “status” and “classif” ) used in the study had a significant effect on the detection of DENV genome in plasma by qRT-PCR . The variable “daof” ( number of days after the onset of fever ) had the higher relative importance ( RI = 57 . 4% ) , followed by the immunological status ( primary or secondary infection ) ( RI = 25 . 8% ) and then by the severity of the disease ( RI = 16 . 8% ) ( Table 4 ) . The partial dependence plots suggested that there was a higher probability of detecting DENV-RNA in the plasma specimens obtained during the early febrile phase of the infection , in the samples obtained from patients with a primary infection and in patients presenting with a severe form of the disease ( Fig 5 ) . Models with four variables ( “daof” , “status” , “classif” and “logvir” [viral load in plasma] ) were applied for the analysis of the qRT-PCR results in saliva and urine samples . The variable “logvir” was the main explanatory variable ( RI saliva = 64 . 2% and RI urine = 52 . 6% ) in both models , followed by “daof” ( RI saliva = 18 . 5% and RI urine = 26 . 8% ) and “status” ( RI saliva = 14 . 1% and RI urine = 14 . 4% ) ( Table 4 ) . The model applied for saliva specimens demonstrated a positive association between the RNA load measured in plasma and the detection of the dengue genome in saliva . With the model used for urine specimens , a more complex relationship with two maxima was observed: the first one for a viral load between 4 and 6 log10 cDNA copies/ml and the second one for a viral load greater than 8 log10 cDNA copies/ml . The detection of the virus genome in saliva and urine samples was better during primary infections ( Fig 5 ) . The RI for the variable “classif” was only 3 . 2% and 6 . 2% in saliva and urine specimens , respectively ( Table 4 ) . The partial dependence plots suggested a slightly better detection of DENV-RNA in the urine of patients experiencing a severe form of the disease ( Fig 5 ) . The best BRT model for NS1 detection in plasma was obtained by using the variables “daof” , “classif” and by replacing the variable “status” by the continuous variable “IgG” ( level of IgG antibodies in the plasma specimens estimated by the ELISA OD value ) . The relative importance of the variable “IgG” was 89 . 3% and the partial dependence plot demonstrated a nearly linear decreasing association with the detection of NS1 protein ( Table 5 , Fig 6 ) . The variables “daof” and “classif” had a RI of 8 . 0% and 2 . 7% , respectively . The severity of the disease ( variable “classif” ) can be considered as having a negligible effect on the detection of the NS1 protein in the plasma samples . The BRT models used for NS1 antigen detection in saliva and urine specimens were built with the variables “daof” , “status” , “classif” and “logNS1” ( NS1 protein concentration in plasma ) . The higher RI in both models was obtained for the continuous variable “logNS1” ( RI = 67 . 3% and 64 . 6% for the saliva and urine models , respectively ) ( Table 5 ) . The partial dependence plots demonstrated a positive association between NS1 protein concentration in plasma samples and detection of NS1 antigen in saliva and urine specimens ( Fig 6 ) . The second main explanatory variable identified was “daof” ( RI = 20 . 9% and 25 . 4% for the saliva and the urine models , respectively ) with an optimal detection of the NS1 protein being between the 3rd and the 8th day after the onset of fever , in both the saliva and the urine samples ( Table 5 , Fig 6 ) . The contribution to the models of the variables “status” and “classif” was very low ( Table 5 ) . In all of the BRT models generated for anti-DENV antibodies detection in the three different body fluids , the impact of the explanatory variable “classif” ( severe versus non-severe dengue ) was negligible . The detection of IgG in plasma specimens was explained by the variables “daof” and “status” with a close relative importance ( 47 . 2% and 49 . 3% , respectively ) ( Table 6 ) . The partial dependence plots showed that the detection of IgG in plasma was associated with secondary infection and that IgG antibodies were detected from one week until the end of the follow-up period ( Fig 7 ) . For the detection of IgM in plasma specimens , the variable “daof” appeared to be the only explanatory variable and was associated with a RI of 95 . 4% . The RIs of the variables “status” and “classif” were negligible at 3 . 3% and 1 . 4% ( Table 6 ) . The detection of anti-DENV IgA antibodies in plasma samples was mainly explained by the variable “daof” ( RI = 69 . 8% ) and was the highest between one and four weeks after the onset of fever ( Table 6 , Fig 7 ) . The immune status of the patient ( secondary infection ) also contributed to explain the ability to detect IgA antibodies ( RI = 26 . 2% ) ( Table 6 , Fig 7 ) . The detection of antibodies in saliva samples was linked mainly to the level of the corresponding class of immunoglobulin in plasma ( IgG detection: RI = 50 . 8%; IgM detection: RI = 68 . 5%; IgA detection: RI = 58 . 4% ) . Similarly , the detection of IgA antibody in urine samples correlated with the titer of IgA in plasma ( RI = 60 . 2% ) . The relative importance of plasmatic IgG level appeared to be only 20 . 3% for the detection of IgG in urine specimens . The main variables associated with the presence of IgG in urine samples were the immunological status ( RI = 42 . 1% in secondary infections ) and the date of sampling ( RI = 36% ) ( Table 6 , Fig 7 ) . Two BRT models were used to identify factors influencing NS1 concentration ( NS1 model , variable “logns1” ) and the RNA load ( RNA model , variable “logvir” ) in plasma samples during the acute febrile phase of the disease ( DAOF 0–5 ) . Both models were built with the explanatory variables “daof” , “classif” and “igg” . In both models the higher RI was obtained for the variable “igg” ( RI = 85 . 2% and 77 . 3% for the NS1 and the viral RNA models , respectively ) , followed by the variable “daof” ( RI = 8 . 5% and 20 . 9% for the NS1 and the viral RNA models , respectively ) and the variable “classif” ( RI = 6 . 3% and 1 . 8% for the NS1 and viral RNA models , respectively ) ( S8 Table ) . The partial dependence plot analysis demonstrated that higher NS1 concentrations and RNA loads were associated with very low level of anti-DENV IgG ( OD<0 . 15 ) ( S3 Fig ) . One of the objectives of this study was to assess the possibility to use saliva or urine specimens instead of blood for the diagnosis of dengue infections in specific situations . The evaluation included the assays most routinely used for dengue diagnosis: viral genome detection by qRT-PCR , virus isolation in mosquito cell lines , NS1 antigen and anti-DENV antibody detection by ELISA methods . In order to generate the strongest data possible , a large number of samples obtained from children hospitalized in Cambodia for dengue of varying degrees of severity were tested . This is the first study in which plasma , urine and saliva obtained from the same patients , at the same time-points , were tested in parallel and the results compared to estimate the performances of the various dengue diagnostic assays . Saliva and urine specimens are obtained through non-invasive procedures that require the patient’s participation . If devices for urine sampling in infants exist and if saliva can be collected with small specific swabs , one may consider that in the youngest children the collection of capillary blood on filter paper after a finger-prick could be a better alternative to venipuncture than urine or saliva sampling . This is the first study that aimed to detect the DENV genome in such a large number of saliva samples and to comprehensively describe the kinetics of DENV genome detection in this body fluid . Previously , Anders et al . attempted to detect the DENV genome in saliva swab specimens obtained from six patients with confirmed DENV infection but all samples tested negative [26] . The DENV genome was successfully detected from saliva samples in only few instances and described in some cases reports [20 , 23] as well as in one very recent study performed on only 14 patients [19] . We demonstrated here that , as in the plasma , the sensitivity of the genome detection in saliva specimens was the highest during the early acute phase of the infection ( i . e . , 63 . 9% at day 3 post-fever; approximately 60% during the four first days of the disease ) and then decreased over time before the RNA became undetectable 10 days after the onset of fever . The curves of the sensitivity of the RNA detection in plasma and saliva specimens had very similar profiles over time . Nevertheless , the sensitivity of the qRT-PCR in saliva was on average 40% lower than in plasma when the specimens were tested between DAOF0 and DAOF8 . The BRT analysis indicated that the higher the RNA load in plasma , the higher the probability of detecting the viral RNA in saliva samples was . Some authors showed in a limited number of patients and samples that the DENV genome could be detected in urine specimens [19 , 20 , 22–24] . Hirayama et al . described the results obtained with 77 urine samples collected from 53 patients and described DENV genome excretion kinetics similar to the one we report here , i . e . a delayed excretion of the viral genome in urine by comparison to blood [22] . In some cases , the DENV genome was still detectable in urine samples while the serum already turned negative by qRT-PCR [20 , 22 , 24] . In our study , we observed this pattern in three patients whose plasma collected at the time of hospital discharge ( up to10 days after onset of fever ) tested negative , whereas their urine collected at the same time was positive . The sensitivity of the confirmation of the dengue infection by viral genome detection improved from 79 . 6% to 82 . 6% when qRT-PCR was performed in both urine and plasma samples collected after the 5th day of fever , compared to qRT-PCR in plasma alone . Nevertheless , even after DAOF 5 , plasma remained the best biological fluid to use for dengue diagnosis . Indeed , out of the 116 pairs of urine and plasma samples collected during the early convalescent phase of the disease from the same patients , 63 . 8% were concordant ( i . e . , both urine and plasma samples tested positive or both urine and plasma samples tested negative ) . Only three pairs demonstrated discordant results with an advantage for urine over plasma whereas 39 pairs gave discordant results in favor of plasma ( i . e . the urine sample tested negative whereas the corresponding plasma sample tested positive ) . After discharge from the hospital , the urine sample of three other patients still tested positive ( two patients at DAOF 13 and one patient at DAOF 16 ) but the comparison with plasma was not possible as blood samples were unavailable for these time points . All patients whose saliva samples tested positive also had detectable levels of RNA in the plasma samples collected at the same time-points . Thus , testing saliva in addition to the plasma samples did not significantly improve diagnostic sensitivity . When it is not possible to get blood samples , testing concurrently both urine and saliva specimens by qRT-PCR could offer an interesting alternative as the overall diagnostic sensitivity was 76 . 1% during the acute phase of the disease ( DAOF 0–5 ) and 59 . 4% during the early convalescent phase ( DAOF 6–12 ) . Alternatively , the analysis of sequential urine or saliva specimens could also provide , in specific situations , some acceptable results as 57% of the patients had at least one positive urine sample and almost 69% of them had a least one saliva sample that tested positive in the course of the disease . To date , DENV has never been isolated from urine or saliva . Two unsuccessful attempts to isolate the DENV from urine and one attempt to isolate the virus from saliva were previously reported by Hirayama et al . and Korhonen et al . [19 , 22] . The low RNA loads in urine and saliva observed in our study could provide one explanation to the inability to isolate the virus after inoculation of the samples onto C6/36 or Vero E6 cells [39] . Nevertheless , the absence of infectious DENV particles from urine and saliva specimens will be attested only after unsuccessful inoculation of the RNA-positive samples into mosquitoes , as this method is recognized as the most sensitive for DENV isolation [40] . Direct toxicity on C6/36 or Vero E6 cells of some urine and saliva components could also explain the inability to isolate the virus from those samples . In our study , urine samples were dialyzed before inoculation in order to eliminate potentially toxic components and saliva samples were inoculated after dilution with culture media to avoid excessive direct toxicity . Furthermore , the controls included in the series of culture plates provided evidence that a premature cell death was not responsible for the absence of virus detection . It rather appeared that a low viral load was at the origin of an absence of detection of infectious virus in saliva and urine specimens . But it is also possible that the DENV RNA detected by qRT-PCR in urine was “free” viral RNA that was able to go through the glomeruli while virus particles , especially those embedded in large immune complexes , were stopped . The detection of NS1 protein in saliva and/or urine samples has already been reported in three studies bearing on a limited number of patients [19 , 21 , 26] and recently in a larger study including urine samples from 96 patients [41] . Here , we demonstrate that the sensitivity of NS1 detection in saliva and urine samples follow a similar trend over time . Diagnostic sensitivity increased progressively until DAOF 4 to reach a maximum of 25% in urine and 40% in saliva samples , and then decreased until NS1 became undetectable , ten days after fever onset . The sensitivity NS1 antigen detection in urine and saliva samples observed in our study was lower than the one reported previously . Anders et al . reported 64 . 7% ( 55/85 ) of saliva samples positive for NS1 protein when testing patients with a positive NS1 antigenemia [26] . In our study , only 25 . 8% ( 99/383 ) of the paired plasma and saliva samples tested both positive , at the same time-point . Chuansumrit et al . , Korhonen et al . and Saito et al . previously evaluated the sensitivity of NS1 detection in urine using the same commercial ELISA kit . Chuansumrit et al . and Korhonen et al reported sensitivities of 65 . 6% and 54 . 2% , respectively [19 , 21] . Saito et al . described a positive detection rate of NS1 ranging from 13% to 43% according to days after disease onset . The highest detection rate was obtained on days 6–10 after disease onset [41] . As observed for the RNA load , NS1 protein concentrations measured in urine and saliva samples were much lower than those measured in plasma specimens . The multivariate analysis using BRT identified the NS1 concentration in plasma as the main factor explaining the possibility of detecting the viral protein in urine and saliva . Korhonen et al . also showed that the NS1 concentration in urine positively correlated with the concentration in urine of the total proteins [19] . In their study , Chuansumrit et al reported a higher NS1 detection rate in urine samples obtained from patients with DHF than in patients presenting with a mild DF [21] . Our statistical analysis performed in a large patient series did not evidence a higher probability of detecting NS1 in the urine of patients experiencing a severe form of the disease ( DHF and DSS ) than in those presenting with a mild infection . We decided not to concentrate urine specimens before testing for NS1 in order to keep the ELISA protocol as easy and cheap as possible to meet the “real-life” conditions of most endemic countries which are also often developing countries . Saito et al . investigated the benefit of urine concentration prior to NS1 capture by ELISA with 37 paired concentrated and non-concentrated urine samples . Concentration allowed the detection of NS1 antigen in three samples that tested negative before concentration [41] . For well-equipped laboratories that may be willing to use urine as a replacement to blood for DENV infection confirmation , it would be important to further develop of a fast and simple method for NS1 protein concentration . Our study also shows that 2/3 of the plasma samples that tested positive by qRT-PCR also tested positive for NS1 detection; that half of the saliva samples that tested positive by qRT-PCR also tested positive by NS1 capture ELISA , but that NS1 was detected in less than 30% of the urine samples that tested positive by qRT-PCR . The discrepancy between the RNA and NS1 detection in urine samples obtained from patients with detectable viremia suggests that these two biological markers of dengue infection are potentially released in urine through different mechanisms that remain to be clarified . Similarly to Vasquez et al . , we were unable to detect anti-DENV IgM in urine specimens [25] . The sensitivity of the IgM and the IgG assays in saliva was close to those obtained in plasma samples . The sensitivity of the IgA serology in both urine and saliva was 65% at the peak point when it was almost 90% in plasma . This is in agreement with the data previously described by Vasquez et al . as well as by Balmaseda et al . [25 , 27 , 42] . Multivariate analysis demonstrated that the probability of detecting antibodies in saliva specimens depended essentially on antibodies titers in the plasma . Cuzzubbo et al . reported previously that salivary IgG levels correlated well with serum HI titer [43] . This study also showed that IgG detection in saliva could be used to distinguish between primary and secondary DENV infection . Our BRT analysis confirms this result . Balmaseda et al . investigated the use of IgG and IgM detection in saliva to evaluate DENV infection incidence during a serological survey and demonstrated that anti-DENV IgG detection in saliva was a good tool to measure the incidence of dengue in a community cohort [27] . The anti-DENV IgG antibody urinary excretion curve follows the same trend than that of IgG in plasma during the first week of the disease . Subsequently , urinary excretion of IgG decreases over time while it remains stable in plasma . Three months after disease onset , 76 . 1% of the patients had detectable anti-DENV IgG in their plasma whereas only 12 . 5% of them still tested positive in their urine . The present study is the first to document the kinetics of anti-DENV antibodies until the third month after fever onset . Multivariate analysis showed that the detection of anti-DENV IgG in urine was not primarily associated with antibody titers in plasma but with the immune status of the patient and with the time of sampling after the onset of the disease . Antibodies are high-molecular-weight ( HMW ) proteins . In physiologic conditions , these proteins are almost completely restricted from filtration by the glomerular barrier and only a very small quantity of it can normally be detected in urine . The presence of macromolecules such as IgG and IgA in urine could reflect an alteration of the glomerular barrier , with or without tubular cells impairment . IgM is a pentameric immunoglobulin of very high molecular weight ( 950 kDa ) and its detection in urine would reflect a severe glomerular injury [44] . This probably explains why IgM were not detected in the patients included in this study as none of them had any record of severe renal impairment during the course of their illness . In this study , as in most of those reported previously , saliva specimens were collected by direct spitting . Only Anders et al . collected saliva using swabs and reported results significantly different from ours [26] . These authors were unable to detect DENV genome by RT-PCR but obtained a higher sensitivity for NS1 ( i . e . , 64 . 7% ) . Michishige et al . demonstrated that saliva samples collected by different methods of sampling were not equivalent in terms of total proteins and secretory IgA concentrations [45] . These different methods of saliva sampling should be compared in order to better define which one is the most suitable to perform each test . The early identification of an ongoing DENV infection and a rapid implementation of specific clinical management procedures are key to ensure the best clinical outcome [1] . The current lack of simple and reliable prognostic marker to predict the risk of evolution of the patient towards a severe form of the disease makes the clinical management of the DENV infections extremely challenging , especially during epidemics and in countries with weak health systems . Among all the parameters evaluated within this study , the NS1 protein would theoretically be the ideal candidate for early diagnosis as this antigen is a marker of acute dengue infection that is easy and fast to detect . Anti-DENV antibodies usually appear only after the patient has already progressed toward the critical phase of infection . Conflicting results have been reported on a possible association between the concentration of free NS1 circulating in plasma and the disease severity . Although several studies reported a positive association [46 , 47] , other studies have not [48–50] . In our study , which included a large number of patients experiencing primary and secondary infections , the multivariate analysis using BRT suggested that disease severity does not influence NS1 concentration in plasma during the acute phase of the disease but that plasmatic titers of anti-DENV IgG was the most important factor to explain free NS1 protein concentration in blood . Indeed , undetectable or very low levels of anti-DENV IgG antibodies were associated with higher concentrations of free NS1 , the only fraction that is detected by the assays , while the NS1 fraction that is trapped into the immune-complexes is not captured in the diagnostic tests . A potential association between the DENV-RNA load measured during the acute phase of the disease and disease severity has also been suggested , but the scientific literature reports conflicting findings [46 , 48 , 51 , 52] . Our BRT analysis suggests that the severity of infection was not associated to the RNA load in plasma during the acute phase of the disease ( DAOF 0–5 ) . The level of anti-DENV IgG was the main parameter explaining the RNA load in the plasma specimens during the acute phase of the disease , with an inverse relationship like the one observed for NS1 . The present report is the first to investigate the relationship between the NS1 concentration and the RNA load in plasma and the severity of the disease by using a multivariate analysis that included the anti-DENV IgG level . These findings require to be confirmed during other epidemics involving more patients and other infecting serotypes as this factor may affect both the RNA load and the NS1 concentration [53 , 51] . Our study shows that urine and saliva could be considered for the diagnosis of dengue infection in some situations when optimal sensitivity is not necessarily required . These two body fluids provide different type of information . The saliva essentially mirrors what happens in the blood , which is not surprising as approximately 2 . 5% of the whole saliva is composed of gingival fluid , a protein-rich serum transudate , therefore reflecting the protein pattern in the serum . Antibodies titers in saliva are thus a few orders of magnitude lower compared to those in serum [54] . It was estimated that the IgA , IgG and IgM levels in saliva specimens were approximately 1/10 , 1/800 and 1/400 of those measured in serum [55] . In urine , the detection of macromolecules such as IgGs ( 150 kDa ) , IgAs ( 320 kDA ) and NS1 protein ( 300 kDa ) most likely reflects an alteration of the glomerular filtration barrier which has three major components: the glomerular endothelial cells ( GECs ) , the glomerular basement membrane and the podocytes . One of the characteristics of GECs is the presence of numerous fenestrations , which in theory should allow the passage of large molecules . GECs , however , are coated by a glycocalyx layer composed of glycosaminoglycans , which in normal conditions impedes the leakage of macromolecules . Abnormal disruption of the glycocalyx restores the passage of big molecules through the fenestration of the GECs [56] . High-molecular-weight proteins cannot pass through the podocytes barrier in normal conditions , but the accumulation of such macromolecules probably induces an alteration of this structure , which in turn leads to a proteinuria [57] . It has been suggested that during dengue infection , impairment of the glycocalyx layer on endothelial cells could occur . Both the virus itself and the NS1 antigen are known to fix heparan sulfate , a major glycosaminoglycan of the glycocalyx [58 , 59] . Moreover Wills et al . observed an increased urinary heparan sulfate excretion in children with severe dengue infection compared to healthy control subjects , suggesting that an alteration of the glycocalyx layer may occur in severe dengue infections [60] . The presence of circulating immune complexes formed to eliminate the virus and the NS1 antigen may provide another possible explanation to the alteration of the glomerular barrier . Cases of glomerulonephritis with glomerular immune complex-type deposits have been reported during DENV infections [61] . Jessie et al . also described the presence of viral antigen suggestive of the presence of immune complexes in the kidney tubular cells from dengue-infected patients [62] . Further investigations are needed on the potential interactions of dengue virus , immune complexes , and other components of the immune response with heparan sulfate and other glycosaminoglycan of the glycocalyx but also with the other kidney cells to determine if subclinical kidney lesions do not occur more often during DENV infection . In this study we were not able to evaluate the performances of the different diagnostic methods for all four dengue virus serotypes as it was conducted using well-characterized and sequential clinical samples prospectively collected during a DENV-1 epidemic , when the DENV-2 and DENV-4 were circulating at lower level and no DENV-3 was detected . The results of our study demonstrate that the diagnosis of dengue infection in urine and saliva specimens is possible but we believe that there is certainly still a room for improvement of these assays . Existing commercial rapid diagnostic tests ( RDTs ) designed to be used with blood samples would probably require significant adjustments , as urine and saliva specimens are very different from whole blood , plasma or serum . One of the main differences resides in the huge difference in protein concentration . A Singaporean group has recently developed a rapid test for the detection of anti-DENV IgG in saliva . The device gave good results in saliva samples spiked with IgG but requires further optimization to detect IgG from the clinical samples of dengue-infected patients [63] . Recently , saliva samples were tested using a commercial ELISA initially developed for IgG detection in blood and demonstrated 100% sensitivity and specificity [64] . In conclusion , although the performances of the different diagnostic methods evaluated here were not as good in saliva and urine as in plasma , results obtained with qRT-PCR and with antibody detection could justify the use of these two body fluids for the diagnosis of dengue infection for instances such as outbreak investigations or in young children ( once they are old enough to comply to instructions ) , in addition to the situations when blood cannot be easily collected ( e . g . , lack of phlebotomist , refusal of the procedure , etc . ) . The disadvantage resulting from a slight decrease in the diagnostic confirmation performances when using molecular and serological tests in urine and saliva samples instead of blood is partially balanced by the ease to obtain these specimens and by the better compliance of the patient or by the number of individuals that can be investigated during studies when non-invasive sample collection methods are used . Moreover , the modest sensitivity of the test can be offset by high prevalence during the peak of an outbreak and/or when guided by competent clinicians . It results in high predictive positive value , making this test a useful addition for biological diagnosis in field conditions .	Dengue is the most important arthropod-borne disease affecting humans and represents a huge public health burden in affected countries . Symptoms are often non-specific hence the need for an early , sensitive and specific diagnosis of dengue for appropriate management as well as for early epidemic detection . Currently , almost all laboratory diagnostic methods require a blood specimen that may be sometimes be difficult or inconvenient to obtain . In this study , we assessed the possibility to use saliva and urine samples as alternatives to blood specimens in dengue diagnosis . We demonstrated that the performances of the different diagnostic methods ( RT-PCR , NS1 antigen detection and anti-DENV IgM/IgG/IgA ELISAs ) were in general not as good in saliva and urine as in plasma , but that the use of these body fluids obtained by non-invasive methods could be of value in certain circumstances such as outbreak investigations or in young children ( once they are old enough to comply to instructions ) , in addition to the situations when blood cannot be easily collected ( e . g . , lack of phlebotomist , refusal of the procedure , etc . ) .	[ "Abstract", "Introduction", "Materials", "and", "Methods", "Results", "Discussion" ]	[]	2015	Value of Routine Dengue Diagnostic Tests in Urine and Saliva Specimens
Sepsis is a frequent complication in critical illness . The mechanisms that are involved in initiation and propagation of the disease are not well understood . Scavenger receptor A ( SRA ) is a membrane receptor that binds multiple polyanions such as oxidized LDL and endotoxin . Recent studies suggest that SRA acts as a pattern recognition receptor in the innate immune response . The goal of the present study was to determine the role of SRA in polymicrobial sepsis . SRA deficient ( SRA−/− ) and C57BL/6JB/6J ( WT ) male mice were subjected to cecal ligation and puncture ( CLP ) to induce polymicrobial sepsis . NFκB activity , myeloperoxidase activity , and co-association of SRA with toll like receptor ( TLR ) 4 and TLR2 was analyzed in the lungs . Spleens were analyzed for apoptosis . Serum cytokines and chemokines were assayed . Blood and peritoneal fluid were cultured for aerobic and anaerobic bacterial burdens . Long term survival was significantly increased in SRA−/− septic mice ( 53 . 6% vs . 3 . 6% , p<0 . 05 ) when compared to WT mice . NFκB activity was 45 . 5% lower in the lungs of SRA−/− septic mice versus WT septic mice ( p<0 . 05 ) . Serum levels of interleukin ( IL ) -5 , IL-6 , IL-10 and monocyte chemoattractant protein −1 were significantly lower in septic SRA−/− mice when compared to septic WT mice ( p<0 . 05 ) . We found that SRA immuno-precipitated with TLR4 , but not TLR2 , in the lungs of WT septic mice . We also found that septic SRA−/− mice had lower bacterial burdens than WT septic mice . SRA deficiency had no effect on pulmonary neutrophil infiltration or splenocyte apoptosis during sepsis . We conclude that SRA plays a pivotal , and previously unknown , role in mediating the pathophysiology of sepsis/septic shock in a murine model of polymicrobial sepsis . Mechanistically , SRA interacts with TLR4 to enhance the development of the pro-inflammatory phenotype and mediate the morbidity and mortality of sepsis/septic shock . The critically ill patient frequently develops a complex disease spectrum that may include acute respiratory distress syndrome , systemic inflammatory response syndrome , sepsis syndrome and/or septic shock [1] . Current wisdom implies that following severe injury or infectious challenge , the host responds by over-expressing inflammatory mediators resulting in a systemic inflammatory response that culminates in severe shock , multi-organ failure and death [2] , [3] , [4] . At present , we do not understand the cellular and molecular mechanisms that are involved in the initiation and propagation of septic injury; nor do we understand the physiologic mechanisms that attempt to maintain homeostasis and promote survival in the septic patient . The macrophage scavenger receptor A ( SRA , CD204 , Entrez gene Msr1 ) is a type II membrane receptor [5] . SRA is primarily expressed by macrophages , though evidence suggests it may also be expressed by bone marrow derived and splenic dendritic cells [5] . SRA is a multi-functional receptor which binds endogenous ligands including oxidized LDL and apoptotic cells [5] , [6] , [7] and pathogen associated molecular patterns including endotoxin , lipoteichoic acid , and fungal glucans [7] , [8] , [9] , [10] , [6] . Evidence for direct intracellular signaling by SRA is limited and conflicting . However , several reports indicate that SRA interacts with Mer receptor tyrosine kinase [10] , Lyn kinase [11] and PTK ( Src ) /Rac1/Jnk [12] . Additionally , phosphorylation of SRA may facilitate the interaction of the SRA transmembrane domain with signaling components [13] . It has also been reported that SRA induces activation of MyD88 dependent toll like receptor ( TLR ) 4 signaling and inhibits TLR4 dependent IRF3 activation in response to endotoxin or fucoidan [14] . Finally , it has been demonstrated that SRA interacts with TRAF6 upon exposure to endotoxin and prevents its degradation , thus limiting the inflammatory response to endotoxin [15] . When taken together , these data suggest that SRA does participate in intracellular signaling in response to ligand interaction . SRA plays a role in several important pathological processes , including atherosclerosis [6] . SRA has also been identified as a pattern recognition receptor in the innate immune system [16] and thus is involved in the immune response to infectious disease [17] , [18] , [19] , [20] , [21] , [22] . However , the role of SRA in response to infection is complex and not well understood . By way of example , SRA has been reported to be protective in Listeria monocytogenes infection [23] , herpes simplex-1 infection [23] , Neisseria meningitides septicemia [22] and pneumococcal pneumonia [20] . In contrast , SRA deficient mice challenged with Pneumocystis carinii infection cleared the organisms from the lung more efficiently when compared to wild type controls [21] , suggesting that SRA contributes to the pathophysiology of P . carinii infection . However , these studies were all performed in single organism infections . The role of SRA in a clinically relevant model of polymicrobial sepsis has not been investigated . In the present study , we found that SRA plays a central and previously unknown role in mediating the pathophysiology of sepsis/septic shock in a murine model of polymicrobial sepsis . Specifically , SRA interacts with TLR4 thereby facilitating the development of the pro-inflammatory phenotype and mediating the morbidity and mortality of sepsis/septic shock . In response to CLP sepsis , we found that SRA−/− mice showed a much longer median survival time ( 300 hrs vs 43 hrs ) than did WT mice . Of greater importance , SRA deficient mice , with CLP sepsis , showed a significant increase in long term survival ( 53 . 6% vs . 3 . 6% , p<0 . 001 ) when compared to WT mice ( Figure 1 ) . That is to say that 53 . 6% of the SRA−/− mice with sepsis went on to survive indefinitely while only 3 . 6% of the WT septic mice survived . These data strongly suggest that SRA contributes to the mortality associated with fulminating polymicrobial sepsis . To determine if differences in the inflammatory response were responsible for the differences in survival , NFκB activity was measured in WT and SRA−/− mice in response to CLP by EMSA ( Figure 2 ) . In WT mice , CLP resulted in a significant increase in pulmonary NFκB activity ( 132 . 29% vs . WT control and 59 . 22% vs . WT sham; p<0 . 05 ) . In contrast , NFκB activity was not significantly increased in the lungs of SRA−/− mice with polymicrobial sepsis ( 14 . 12% and 5 . 05% vs . SRA−/− control and sham mice , respectively; p>0 . 05 ) , and was significantly less than that of the lungs from WT CLP mice ( 45 . 49%; p<0 . 05 ) . These data illustrate that the inflammatory response to CLP induced sepsis is blunted in SRA−/− mice . To further define the role of SRA in the inflammatory response in sepsis , serum cytokine levels were assayed in WT and SRA−/− mice 16 hrs after CLP ( Figure 3 ) . CLP resulted in a significant increase in serum interleukin ( IL ) -6 ( p<0 . 05 ) when compared to sham operated or control animals in both WT and SRA−/− mice . However , IL-6 levels were decreased in SRA−/− septic mice when compared to septic WT mice ( ↓72 . 4%; p<0 . 05 ) . IL-10 was also increased in both WT and SRA−/− septic animals compared to sham operated animals ( p<0 . 05 ) . Again , the increase in serum IL-10 was significantly blunted in septic SRA−/− animals compared to WT CLP mice . ( p<0 . 05 ) . Sepsis resulted in a dramatic increase in monocyte chemoattractant protein-1 ( MCP-1 ) when compared to sham control mice ( Figure 3 ) . Serum MCP-1 was also increased in SRA−/− mice , when compared to sham ( p<0 . 05 ) , but the magnitude of the increase was significantly less than that observed in WT CLP mice . Specifically , serum MCP-1 levels were 95 . 4% less in septic SRA−/− mice compared to septic WT mice ( p<0 . 05 ) . Finally , CLP resulted in a significant increase in circulating IL-5 levels in both WT ( ↑370%; p<0 . 05 ) and SRA−/− ( ↑65 . 5%; p<0 . 05 ) mice compared to sham controls . As with the other cytokines , serum levels of the Th2 cytokine IL-5 were less in SRA−/− CLP mice than in WT CLP mice ( p<0 . 05 ) . We did not detect significant differences in levels of the other cytokines assayed . These data demonstrate that the inflammatory phenotype , as viewed from the perspective of circulating cytokines and chemokines , is significantly attenuated in SRA−/− mice . Additionally , the significant attenuation of Th2 associated cytokines in septic SRA−/− mice indicates an overall maintenance of the Th1 phenotype in these mice . To further characterize the differences in the inflammatory response to sepsis in WT and SRA−/− mice , neutrophil infiltration in the lung was determined by measuring myeloperoxidase ( MPO ) activity ( Figure 4 ) . Though there was a significant increase in MPO activity in CLP mice vs . sham mice , there was no difference between septic WT and SRA−/− mice . To confirm these data , paraformaldehyde fixed lungs were sectioned and stained with hemotoxylin and eosin . The stained tissues were evaluated for inflammation and neutrophil infiltration by two independent pathologists . There was no difference detected in the lung tissues harvested from WT or SRA−/− mice subjected to CLP ( data not shown ) . Specifically , there was no evidence of increased neutrophils in the lungs , no foci of inflammation and no margination of neutrophils in blood vessels in any of the lung tissues examined ( data not shown ) . These data indicate that SRA deficiency has no significant effect on pulmonary neutrophil infiltration or inflammation in sepsis . Apoptosis of splenic lymphocytes plays a role in the immunosuppression associated with sepsis [24] . To determine if there is a difference in splenocyte apoptosis in septic WT and SRA−/− mice , spleens were lysed and caspase 3 and 7 activity was measured ( Figure 4 ) . Though there was a significant increase in caspase activity in septic SRA−/− spleens compared to sham controls , there was no difference between WT sham and CLP nor was there a significant difference between WT and SRA−/− in septic spleens . The data above indicate that SRA contributes to the inflammatory response in CLP sepsis , and that SRA deficiency correlates with improved survival . Since TLRs are known to play a major role in the inflammatory response to sepsis [25] , [26] , [27] , we sought to determine if SRA interacts with TLRs during CLP sepsis . Protein from WT murine lungs was precipitated with SRA antibody and blotted for TLR2 and 4 . We found that TLR2 was not precipitated with SRA in the control , sham , or CLP lungs ( Figure 5 ) . On the other hand , TLR4 was precipitated with SRA in CLP lungs and to a far lesser extent in control and sham lungs ( Figure 5 ) . These data indicate that SRA is exerting its inflammatory effect , in part , by interacting with TLR4 , but not TLR2 in polymicrobial sepsis . The previous experiments revealed that SRA deficiency results in decreased inflammation in polymicrobial sepsis . We sought to determine if SRA deficiency had an impact on bacterial burden and/or the composition of bacterial species post-CLP . Blood and peritoneal fluid were harvested from WT and SRA−/− septic mice and cultured for aerobic and anaerobic bacteria . The bacterial burdens were quantified by quadrants positive for growth , and the species cultured were identified . One of four SRA−/− showed no bacteria in the blood , while two others showed only non-pathogenic Lactobacillus ( Table 1 ) . The remaining SRA−/− blood sample also showed Lactobacillus as well as E . coli ( Table 1 ) . In contrast , all of the WT mice showed positive bacterial cultures ( Table1 ) . One contained only Lactobacillus , while the remaining three were culture positive for a pathogen , i . e . Group D Enterococcus or Campylobacter gracilis ( Table 1 ) . Culture of peritoneal fluid from WT and SRA−/− mice showed a much greater number and diversity of micro-organisms ( Table 2 ) . Four of the five SRA−/− mice had 1+ or less bacterial growth from their peritoneal fluid , and these bacteria were primarily non-pathogenic ( Table2 ) . However , one mouse in this group did show E . coli in the blood and higher levels of peritoneal growth with 3+ E . coli and 2+ Staphylococcus ( Table 2 ) . As with the blood , WT mice showed greater levels of bacteria than SRA−/− mice ( Table 2 ) . Each WT mouse had at least one bacterial species with 2+ growth , and all mice had at least 1+ growth of the pathogenic Group D Enterococcus in their peritoneal fluid . These data suggest that SRA plays a role in the clearance of bacteria in sepsis . Several important observations emerged from this study . Our data indicate that SRA deficient mice are much more resistant to fulminating polymicrobial sepsis , as demonstrated by increased long term survival . In association with the improved survival , SRA deficient mice showed an attenuated inflammatory phenotype as determined by decreased organ NFκB activity and attenuation of sepsis induced serum cytokine levels . In addition , SRA−/− mice showed lower bacterial burdens and fewer pathogenic bacteria when compared to WT mice . In order to elucidate the mechanisms by which SRA facilitates the morbidity and mortality of sepsis , we discovered that SRA co-associates with TLR4 during sepsis and that this interaction is closely correlated with tissue NFκB activation , development of a pro-inflammatory phenotype and mortality in sepsis . When considered as a whole , our data suggest that SRA plays a key role in the morbidity and mortality of sepsis via its interaction with TLR4 . In the present study , SRA deficiency resulted in a significant attenuation of sepsis-induced NFκB activation , which correlated with improved survival outcome . Whether this is a cause-and-effect relationship cannot be established by the present data , but we and others have shown that NFκB dependent signaling plays a major role in the morbidity and mortality of CLP sepsis [28] , [29] , [30] . How SRA contributes to NFκB activation is not clear , since the ability of SRA to transduce an intracellular signal remains controversial [31] , [12] , [10] . SRA lacks a signaling motif in the cytoplasmic tail suggesting that it does not signal [31] , though treatment of cells with known SRA ligands does result in signal transduction [12] . One explanation for these data would be that SRA interacts with other receptors that do have the ability to induce signal transduction . Indeed , our data support this explanation by demonstrating that SRA interacts with TLR4 in CLP sepsis . TLR4 is known to play an important role in septic inflammation and is a well known inducer of NFκB activity and inflammatory cytokine production [25] , [26] , [27] . Our data suggest that the interaction of SRA with TLR4 amplifies the signal generated by TLR4 in response to the bacterial and endogenous ligands released during polymicrobial sepsis . Loss of the interaction with SRA would then result in loss of the amplification and a lower overall inflammatory response . In fact , we demonstrated that sepsis-induced cytokine/chemokine expression was attenuated in SRA deficient mice . Of specific interest , both IL-6 and IL-10 levels , which strongly correlate with survival outcome in sepsis [32] , [33] , were significantly attenuated in SRA deficiency . When taken together these data suggest that in response to sepsis , SRA interacts and/or cooperates with TLR4 to enhance NFκB activation and cytokine/chemokine expression with a concomitant increase in inflammatory phenotype and mortality . Thus , our data indicate that SRA is required for a maximal TLR4/NFκB response to CLP sepsis . These data are consistent with previous reports which indicate that SRA can function as a co-receptor [14] . Interestingly , SRA does not appear to play a role in all aspects of the response to CLP sepsis . Lung neutrophil sequestration and splenocyte apoptosis have been implicated in the pathophysiology of sepsis [34] , [35] . Tissue neutrophil infiltration , adhesion and degranulation are thought to play a prominent role in tissue damage during sepsis [34] . Surprisingly , we found that SRA does not appear to play a significant role in sepsis induced pulmonary neutrophil infiltration and sequestration , despite differences in circulating inflammatory cytokines . It is possible that pulmonary neutrophil infiltration/inflammation does not play a significant role in survival outcome in our model of acute sepsis . Splenocyte apoptosis is thought to play a central role in late immune dysfunction in sepsis and may contribute to sepsis associated multiple organ failure [35] . However , we did not detect any significant difference in splenic apoptosis between WT and SRA−/− mice in response to sepsis . Lymphocytes are a major portion of the splenic leukocyte population; therefore these data primarily reflect apoptosis in lymphocytes . Why SRA deficiency would not have an effect on lymphocyte apoptosis while still improving survival outcome is unclear . It may be that splenocyte apoptosis does not play a significant role in the pathophysiology in this model of acute sepsis , but may contribute to later septic sequelae . We also observed that SRA deficiency correlated with decreased bacterial burden in response to CLP sepsis . This was particularly true for the peritoneal cavity . It is not clear why SRA−/− mice would have a decreased bacterial burden in response to CLP sepsis . Decreased bacterial burden in septic SRA−/− septic mice might seem counterintuitive since SRA is known to facilitate the uptake of bacterial products such as LPS and LTA , and it has been reported to mediate the non-opsonic phagocytosis of bacteria [36] , [20] , [8] , [22] . Indeed , SRA deficient mice have been reported to show higher levels of bacteremia in Listeria monocyotgenes and Neisseria meningitides septicemia when compared to WT mice [22] . However , in Pneumocystis carinii infection , SRA deficient mice cleared the organisms from the lung more efficiently than wild type controls [21] . These data suggest that the response of SRA to infection is complex and may be dependent on the pathogen ( s ) encountered . In the case of a polymicrobial infection such as CLP sepsis , it appears that SRA contributes to bacterial burden , particularly in the peritoneal cavity . The mechanisms responsible are unclear , but one possible mechanism is that SRA−/− mouse macrophages and/or neutrophils are better able to kill microbes , perhaps due to an overall maintenance of the Th1 immune response . Future studies are warranted to determine how SRA deficient mice are better able to clear the bacteria associated with polymicrobial sepsis . This is the first report documenting the role of SRA in polymicrobial sepsis . In 2005 , Cotena and colleagues reported on the role of SRA in a sterile peritonitis model which employed the injection of zymosan to elicit peritoneal inflammation [37] . These investigators reported that SRA is a non-activating receptor that serves to counter the activities of pro-inflammatory receptors and attenuates the production of specific chemokines to ensure an inflammatory response of the appropriate magnitude [37] . However , Cotena et al did not study the role of SRA in a model of infection , such as CLP [37] , so comparisons between our results and theirs must be made with caution . Kobayashi [17] , Chen [38] and colleagues have reported that SRA deficient mice are more resistant to LPS challenge , suggesting that SRA plays a role in endotoxic shock . However , Yu et al have recently reported that SRA attenuates TLR4 induced NFκB activation in an endotoxemia model by directly inhibiting ubiquitination of TRAF6 [15] . In addition , Yu et al reported that SRA−/− mice are more susceptible to endotoxin challenge [15] . The reasons for the differences between the work of Kobayashi [17] , Chen [38] and colleagues and Yu et al [15] are not readily apparent . However , it is important to note that treatment of cultured cells with endotoxin , or injection of animals with endotoxin , is not a reliable surrogate for polymicrobial sepsis/septic shock [3] . Indeed , numerous reports have clearly delineated the differences between endotoxemia and a clinically relevant in vivo model of sepsis , such as CLP [3] , [39] , [40] , [41] . Consequently , the effect of SRA in endotoxemia may not be indicative of the role that SRA plays in a fulminating polymicrobial sepsis model , such as CLP , or in clinical sepsis for that matter [40] , [41] . We believe that by examining SRA in a clinically relevant model of sepsis , such as CLP , we will have a more accurate assessment of the role of this receptor in septic disease . In conclusion , our data indicate that the scavenger receptor class A plays a key role in mediating the pathophysiology of fulminant sepsis/septic shock . To the best of our knowledge , this is the first report documenting the deleterious effects of SRA in a clinically relevant model of polymicrobial sepsis . Specifically , SRA appears to be necessary for maximal development of the pro-inflammatory phenotype , in part , through interaction and co-operativity of SRA with the TLR4/NFκB signaling pathway . These data advance our knowledge of the in vivo mechanisms of sepsis and , of potentially greater importance , suggest that modulation of SRA activity may be a viable approach to the management of sepsis syndrome . All animal procedures were conducted in strict compliance with the National Institutes of Health “Guide for the Care and Use of Laboratory Animals” . The animal protocol was reviewed and approved ( protocol number 101201 ) by the University Committee on Animal Care at the James H . Quillen College of Medicine , East Tennessee State University under the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care , US Department of Agriculture , and the Public Health Service guidelines for the care and use of animals as attested by the National Institutes of Health . All efforts were made to minimize suffering . The SRA knock-out ( SRA−/− ) mouse was originally generated by Suzuki et al [23] . Breeding pairs of SRA−/− mice were kindly provided by Siamon Gordon , University of Oxford . WT control mice , C57BL/6J , were purchased from Jackson Labs ( Bar Harbor , ME ) . The animals were maintained on standard laboratory chow and water ad libitum with a 12-hour light/dark cycle . Serologic testing confirmed that the mice were virus free . Age and weight matched male mice underwent cecal ligation and puncture ( CLP ) as previously described to induce polymicrobial sepsis [28] , [42] , [43] . Surgery was performed under isoflurane anesthesia . Briefly , the cecum was exteriorized , the contents were massaged distally , and the cecum was ligated distal to the ileocecal junction . The cecum was punctured once with a 20 gauge needle in an avascular region near the distal end , and a bleb of cecal contents was extruded from the puncture . Sham surgery ( laparotomy alone ) mice were used as a control for surgery and anesthesia , and animals that underwent no surgery or anesthesia were employed as negative controls . Mice were sacrificed at 16 hr post-operatively . Lungs , spleens , and sera were harvested , flash frozen , and stored in liquid nitrogen . Blood and peritoneal fluid were harvested and immediately cultured for bacterial growth . Parallel groups were followed for survival . Mice were terminated upon becoming moribund . Approximately 1 mg of lung cellular proteins were immunoprecipitated with 2 µg of antibodies to SRA ( Santa Cruz Biotechnology , Santa Cruz , CA ) for 1 h at 4°C followed by the addition of 15 µl of protein A/G-agarose beads ( Santa Cruz Biotechnology ) as previously described [44] . The precipitates were washed four times with lysis buffer and subjected to immunoblotting with the appropriate antibodies . Precipitated pulmonary proteins were immunoblotted as described previously [45] , [46] . Briefly , the proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred onto Hybond ECL membranes ( Amersham Pharmacia , Piscataway , NJ ) . The ECL membranes were incubated with anti-TLR4 or TLR2 ( Santa Cruz Biotechnology ) , followed by incubation with peroxidase-conjugated secondary antibodies ( Cell Signaling Technology , Danvers , MA ) . The signals were detected with the ECL system ( GE Healthcare , Piscataway , NJ ) . To control for lane loading , the same membranes were probed with anti-GAPDH ( glyceraldehyde-3-phosphate dehydrogenase , Biodesign , Saco , Maine ) after being washed with stripping buffer . The signals were quantified by scanning densitometry using a Bio-Image Analysis System ( Bio-Rad , Hercules , CA ) . Nuclear proteins were isolated from lung samples as previously described [45] , [46] . NFκB binding activity was examined by EMSA in a 15 µl binding reaction mixture containing 15 µg of nuclear proteins and 35 fmols of [γ-32P] labeled double-stranded NFκB consensus oligonucleotide . Tissues were processed as directed by the MPO Fluorometric Detection Kit ( Assay Designs , Ann Arbor , MI ) . Specifically , 50 mg of tissue was weighed out into 1× assay buffer containing 10 mM N-ethylmaleimide ( Sigma ) . The samples were homogenized by using a Polytron homogenizer . After pelleting , the cells were lysed using 0 . 5% hexadecyltrimethylammonium in 1× assay buffer . Following homogenization by Polytron and sonication at 50% power for 3–10 s pulses , the homogenates were subjected to two freeze thaw cycles . After clearing of cell debris by centrifugation , the lysates were stored at −80°C until assayed . The samples were assayed according to kit directions , and the fluorescence was measured using the Modulus Microplate fluorescent plate reader after 30 min incubation ( Turner Biosystems , Sunnyvale , CA ) . The lungs were fixed in formalin , put into paraffin by an automated tissue processor , cut at 8 µm and stained with hematoxylin and eosin by standard methods . The resultant tissue sections were examined by two pathologists . Serum cytokine levels were assayed with an Invitrogen murine 20 plex cytokine assay ( Carlsbad , CA ) on a Luminex 100 instrument . Specifically , we assayed the serum for FGF basic , GM-CSF , IFN-γ , IL-1α , IL-1β , IL-2 , IL-4 , IL-5 , IL-6 , IL-10 , IL-2p40/p70 , IL-13 , IL-17 , IP-10 , KC , MCP-1 , MIP-1α , MIG , TNFα and VEGF . Cytokine levels were established by comparison to a standard curve as per the manufacturer's instructions . Caspase 3 and 7 activities were measured in splenic lysates using the Caspase-Glo 3/7 Assay from Promega according to manufacturer's directions . Briefly , whole spleens were disrupted in TBS with a Polytron homogenizer . After centrifugation , the pellets were suspended in hypotonic lysis buffer ( 10 mM HEPES , 10 mM KCL , 0 . 1 Mm EDTA , 0 . 1 mM EGTA , and protease inhibitors ) and homogenized using a Polytron . After incubation on ice for 1 h , 10% NP-40 was added and the lysates were vortexed at high speed for 1 min . The insoluble fraction was removed by centrifugation , and the protein concentration of the lysates was measured by BCA . The protein concentration was adjusted to 10 µg/ml , and the lysates were mixed with an equal volume of kit reagent . Luminescence was measured at 2 h of incubation with a Modulus Microplate reader . Peritoneal fluid and blood ( ∼300 µl ) were harvested aseptically from mice 16 h after CLP . Blood samples were inoculated into fresh thioglycollate broth ( 6 ml ) and incubated at 37°C for 48 hr at which time 0 . 025 ml aliquot of broth was streaked to produce isolated colonies on 5% sheep blood agar plates . An anaerobic blood agar plate was also streaked and incubated . In parallel , two 0 . 025 ml of peritoneal fluid was directly streaked and incubated in the same manner . Blood plates were incubated in a candle extinction jar at 37°C for 72 hr . The anaerobic blood agar plates were incubated in an anaerobe jar ( BD Gas Pak EZ Anaerobe Container System , BD Biosciences , San Diego , CA ) at the same temperature for a comparable amount of time . Growth was estimated by scoring the number of plate quadrants covered with colonies , i . e . 1+ = one quadrant , 2+ = two quadrants , etc . Isolated colonies were separated by colony morphology and Gram stained . Each colony was purified and subjected to differential tests including oxidase , catalase , indol , etc . as appropriate for their Gram stain morphology . Aerobes were identified by standard methods . Survival trends ( Figure 1 ) were plotted with Kaplan-Meier technique and compared with the log-rank test . Continuous measurements of study groups were summarized with the mean and sem ( Figures 2–5 ) ; group mean levels were compared with the 1-way analysis of variance followed by the least significant difference comparison strategy . A probability level of 0 . 05 or smaller was used to indicate statistical significance .	Trauma and other critical illnesses can progress to septic shock . The mechanisms that result in this progression are not understood . For this reason , there are no proven treatments available , and the mortality rate from sepsis remains quite high . We have found that mice that lack a certain cell surface protein , scavenger receptor A , have a higher rate of survival from a surgically induced sepsis than those that have the receptor . Previously , this receptor has been found to play a role in atherosclerosis , and more recently , to play a role in the immune response to infection . In this study we have found that in addition to improved survival , mice without scavenger receptor A have fewer bacteria in their abdominal cavities and in their blood . They also have lower levels of inflammation . We demonstrated that scavenger receptor A interacts with another protein involved in inflammation and infection , toll like receptor 4 . This interaction might be one mechanism for the effects seen in mice without scavenger receptor A . These studies provide a better understanding of the underlying mechanisms of sepsis . Drugs that target scavenger receptor A could result in better therapies for sepsis .	[ "Abstract", "Introduction", "Results", "Discussion", "Materials", "and", "Methods" ]	[ "medicine", "immunology", "biology", "microbiology", "critical", "care", "and", "emergency", "medicine" ]	2012	Scavenger Receptor Class A Plays a Central Role in Mediating Mortality and the Development of the Pro-Inflammatory Phenotype in Polymicrobial Sepsis
Recent reports suggest that NF-Y transcription factors are positive regulators of skotomorphogenesis in Arabidopsis thaliana . Three NF-YC genes ( NF-YC3 , NF-YC4 , and NF-YC9 ) are known to have overlapping functions in photoperiod dependent flowering and previous studies demonstrated that they interact with basic leucine zipper ( bZIP ) transcription factors . This included ELONGATED HYPOCOTYL 5 ( HY5 ) , which has well-demonstrated roles in photomorphogenesis . Similar to hy5 mutants , we report that nf-yc3 nf-yc4 nf-yc9 triple mutants failed to inhibit hypocotyl elongation in all tested light wavelengths . Surprisingly , nf-yc3 nf-yc4 nf-yc9 hy5 mutants had synergistic defects in light perception , suggesting that NF-Ys represent a parallel light signaling pathway . As with other photomorphogenic transcription factors , nf-yc3 nf-yc4 nf-yc9 triple mutants also partially suppressed the short hypocotyl and dwarf rosette phenotypes of CONSTITUTIVE PHOTOMORPHOGENIC 1 ( cop1 ) mutants . Thus , our data strongly suggest that NF-Y transcription factors have important roles as positive regulators of photomorphogenesis , and in conjunction with other recent reports , implies that the NF-Y are multifaceted regulators of early seedling development . Plants utilize multiple properties of light , such as intensity , quality , and direction , to guide growth and development [1] . The effects of light on plant development are exemplified by the transition of seedlings from dark growth ( where they exhibit skotomorphogenesis ) to light growth ( photomorphogenesis ) . This transition is crucial for plant viability and is characterized by the inhibition of hypocotyl elongation , the expansion of cotyledons , and the accumulation of photosynthetic pigments . In Arabidopsis thaliana , several different classes of photoreceptors mediate light perception , including the phytochromes , which perceive red and far red light , cryptochromes , phototropins , and LOV ( Light , Oxygen , or Voltage ) domain proteins , which are blue light receptors , and UV RESISTANCE LOCUS 8 , the most-studied morphogenic photoreceptor for UV-B light [2–6] . Of these receptors , the photomorphogenic transition is primarily controlled through the actions of the phytochromes and the cryptochromes [7] . Through their combined actions , signaling cascades are initiated that significantly modify the expression of at least two thousand genes in Arabidopsis [8] . While sustained photomorphogenic growth requires the actions of multiple phytochromes and cryptochromes , the initial signaling cascade is established primarily through phyA , which accumulates to high levels in darkness [9] . Upon activation by far red light , phyA is imported into the nucleus through interactions with FAR RED ELONGATED HYPOCOTYL 1 ( FHY1 ) and FHY1-LIKE ( FHL ) [10] . The physical interaction of phyA with FHY1/FHL is also necessary for phyA to bind further downstream transcription factors that regulate light signaling [11 , 12] . The function of phyA , as well as the photomorphogenic downstream transcription factors , is modulated at multiple levels , including through phyA-mediated protein phosphorylation and targeted , proteasome-mediated degradation . The ubiquitination and targeting of many photomorphogenic proteins for proteasome degradation is regulated through the actions of CONSTITUTIVE PHOTOMORPHOGENESIS 1 ( COP1 ) . In the light and in response to the initiation of phytochrome-mediated signal transduction , COP1 protein is excluded from the nucleus , allowing the accumulation of photomorphogenesis-promoting transcription factors [13–15] . One of the most significant targets of COP1 is HY5 , a relatively small bZIP transcription factor that regulates photomorphogenesis by activating a large number of further downstream transcription factors [16–18] . HY5 has also been identified as an integrator of pathways not directly related to light signaling , including hormone signaling ( abscisic acid ( ABA ) and brassinosteroids ) , apoptosis , and temperature acclimation [19–22] . Unlike many other genes involved in the light signaling cascade , the phenotypes of hy5 mutants are not wavelength-specific [18 , 23] . Other COP1 targets , including the bHLH transcription factor LONG HYPOCOTYL IN FAR-RED 1 ( HFR1 ) and the MYB transcription factor LONG AFTER FAR-RED LIGHT 1 ( LAF1 ) , function in a wavelength-dependent manner; while both hfr1 and laf1 mutants have reduced responses to far red light , only hfr1 has a visible phenotype in blue light , and neither mutant exhibits phenotypes in red or white light [24–28] . Further elucidation of the light-signaling cascade has revealed a handful of other transcription factors whose function is necessary for normal photomorphogenic growth and are also regulated by COP1 , including the B-box ( BBX ) containing proteins SALT TOLERANCE HOMOLOG 2 ( STH2/BBX21 ) and LIGHT REGULATED ZINC FINGER1/STH3/BBX22 , the bHLH proteins PHYTOCHROME RAPIDLY REGULATED 1 ( PAR1 ) and PAR2 , and the Mutator transposase-like FAR-RED ELONGATED HYPOCOTYL 3 ( FHY3 ) and FAR-RED IMPAIRED RESPONSE 1 ( FAR1 ) [29–32] . Thus , light perception , and the associated photomorphogenic signaling cascades , converges at a small suite of transcription factors just downstream of a COP1-mediated hub . From these terminal transcription factors it appears that the signal cascade immediately fans out to thousands of light-regulated genes [33 , 34] . While significant progress has been made identifying and characterizing transcription factors functioning at this COP1-mediated hub , there are likely to be undiscovered pieces in this puzzle . No combination of downstream transcription factor mutants that can phenocopy the phyA mutant has been identified—e . g . , when grown in far red light , the hypocotyls of hy5 hfr1 laf1 triple mutants are ~60% as long as the phyA mutant [35] . While this triple mutant has significantly longer hypocotyls than any of the single mutants or double mutant combinations , residual far red light perception is clearly still present . This is in contrast to fhy1 fhl double mutants which appear phenotypically identical to both phyA mutants and dark grown plants for hypocotyl elongation [10 , 36 , 37] . One explanation is that the downstream transcription factor components are already known , but the right combination of mutations has yet to be assembled in a single genotype . For example , as with hy5 hfr1 laf1 plants , hy5 sth2 sth3 mutants are also additively defective in light perception [30] , but no hy5 hfr1 laf1 sth2 sth3 higher order mutant has been reported . Alternatively , additional transcription factor components may remain unknown . Following two recent publications [38 , 39] , we report here additional strong evidence for the involvement of NUCLEAR FACTOR Y ( NF-Y ) transcription factors in light perception . NF-Y transcription factors consist of three unique proteins , called NF-YA , NF-YB , and NF-YC , and each is encoded by a small family of ~10 genes in Arabidopsis ( this expansion is mirrored in other sequenced plant species , including monocots and dicots; [40 , 41] ) . None of the NF-Y subunits is thought to regulate transcription independently; instead , the mature NF-Y transcription factor is composed of one of each subunit type and all three subunits contribute to DNA binding . NF-YB and NF-YC initially form a dimer in the cytoplasm that translocates to the nucleus where a trimer is formed with NF-YA [42–46] . Thus , regulation of any one NF-Y subunit can alter the function of the entire complex . Following nuclear assembly of the mature complex , NF-Ys bind DNA at CCAAT-containing cis regulatory elements and are typically positive regulators of gene expression [47] . Although the generalized characterization of NF-Ys ( largely from animal and yeast systems ) describes them as binding DNA in the proximal regions of promoters , recent data suggests that they also bind more distal regions of promoters to regulate gene expression [48 , 49] . In the animal lineage , each NF-Y is usually encoded by only one or two genes and the functional consequences of expanded NF-Y gene families in the plant lineage remains only modestly explored . Nevertheless , much progress has been made in recent years describing the roles of individual NF-Y subunits in the control of specific processes , especially the control of photoperiod-dependent flowering through interactions with CONSTANS ( CO ) [50–52] , various functions in the development of nitrogen-fixing root nodules in legume species [53–55] , and abscisic acid signaling during germination and early seedling establishment , often mediated by interactions with bZIP transcription factors [38 , 39 , 56–58] . Relevant to NF-Y roles in photomorphogenesis and light perception , little is currently known . However , NF-YA5 and NF-YB9 were previously implicated in regulating blue light-dependent transcript accumulation for LIGHT-HARVESTING CHLOROPHYL A/B BINDING PROTEIN [59] and NF-Y complexes were also shown to bind and regulate the expression of the spinach photosynthetic gene AtpC [60] . Further , the promoters of a number of light signaling components were bound by LEC1/NF-YB9 ( LEAFY COTYLEDON 1 [61 , 62] ) in chromatin immunoprecipitation experiments , including light harvesting and chlorophyll binding proteins ( e . g . , LHCA1 and LHCB5 ) and transcriptional regulators of light perception ( e . g . , HY5 , HY5 HOMOLOG ( HYH ) , and HFR1 ) [38] . Finally , alterations in hypocotyl elongation resulting from both NF-YB loss of function and inducible overexpression have been observed [38 , 63] , including the recent report that LEC1/NF-YB9 regulates skotomorphogenesis through physical interaction with PHYTOCHROME-INTERACTING FACTOR 4 [39] . Previous work in our lab identified physical interactions between NF-YC and HY5 , as well as other bZIP proteins [57] . Here we extend these initial observations to show that these same NF-YC proteins ( NF-YC3 , 4 , and 9 ) are broad spectrum regulators of light perception . Interestingly , in the same way that HY5 , HFR1 , and LAF1 can physically interact , but still appear to signal through independent pathways , hy5 nf-yc mutants also show additive—even synergistic—light perception defects . This manuscript characterizes several photomorphogenesis-related phenotypes of nf-yc mutants and proposes that NF-Y complexes constitute a novel component of the light signaling cascade , functioning at least partially independent of HY5 , HFR1 , and LAF1 . We further demonstrate that nf-yc mutants can partially suppress several cop1 mutant phenotypes and that proteasome regulation of NF-Y complexes during light perception is mitigated through NF-YA subunits . Similar to the multiple regulatory roles of HY5 in light perception and abscisic acid ( ABA ) signaling , our cumulative research on these three NF-Y proteins demonstrates that they have essential roles in photoperiod-dependent flowering , ABA perception , and light perception . We initially observed slightly elongated hypocotyls in plate grown nf-yc3-1 nf-yc4-1 nf-yc9-1 triple mutants ( hereafter nf-yc triple , [51] ) . These visual differences primarily appeared in plants grown for shorter day lengths . To quantify these observed differences , we compared nf-yc triple mutants to their parental Columbia ( Col-0 ) ecotype under continuous white light ( cWL ) , short day ( SD , 8hrs light/16hrs dark ) , and continuous dark ( cD ) conditions ( Fig 1A–1D ) . While cWL and cD grown nf-yc triple mutants were not significantly different from Col-0 , SD grown seedlings had moderately elongated hypocotyls ( ~50–60% longer , Fig 1B and 1C ) . To de-convolute the contributions of individual NF-YC genes , we additionally examined hypocotyl elongation for the six possible single and double mutants from the three mutant alleles . Modest differences were observed for only the nf-yc3 nf-yc9 double mutant ( Fig 1B ) , although we note that this mutant phenotype was inconsistent in additional experiments . Overall , the data suggested that NF-YC3 , NF-YC4 , and NF-YC9 were collectively necessary for the suppression of hypocotyl elongation . Supporting the genetic data showing overlapping functions , all three genes were strongly expressed in the hypocotyl with peak expression in the vascular column ( S1 Fig ) . To determine whether the hypocotyl elongation defects were wavelength specific , we additionally examined the same suite of mutants grown in continuous blue ( cB ) , far red ( cFR ) , and red ( cR ) light conditions ( Fig 2A–2C ) . In cFR conditions , no significant differences were observed . However , in cR and cB light the nf-yc triple mutants were ~50% longer than Col-0 . Additionally , significant hypocotyl elongation defects were observed in some single and double mutants ( ranging from ~18–29% longer that Col-0 ) . As with the SD white light measurements , differences in the single and double mutants in cB and cR light were less robust between repeated experiments than for the nf-yc triple mutants . Interestingly , longer hypocotyls were always associated with the presence of the nf-yc9 mutant allele—somewhat unexpected as the nf-yc3-1 and nf-yc4-1 alleles are strong knockdowns while the nf-yc9-1 allele retains ~20–25% normal expression levels ( S2 Fig and [51] ) . Collectively , these data demonstrate that NF-YCs are broad spectrum regulators of light perception . NF-Y complexes are known to associate with bZIP transcription factors in both plants and animals [46 , 64–66] . Relevant to light perception , we previously reported a modest yeast two-hybrid ( Y2H ) interaction between NF-YC4 and NF-YC9 with HY5 and a non-interaction with NF-YC3 , although we assumed the inability to detect an NF-YC3 interaction was likely due to its autoactivation problems in the Y2H system [57] . To further confirm this Y2H data , we performed transient interaction assays in Nicotiana benthamiana ( Fig 3A–3C ) . We utilized fluorescence lifetime imaging ( FLIM ) and fluorescence recovery after photobleaching ( FRAP ) to detect fluorescence resonance energy transfer ( FRET ) between epitope-tagged NF-YC and HY5 proteins . In these experiments , HY5 or NF-YB2 ( positive control for interaction with NF-YC ) were translationally fused to enhanced yellow fluorescent protein ( YFP , [67] ) and assayed for FRET against NF-YC3 , 4 , and 9 fused to modified cerulean 3 ( mCer3 , [68] ) . By comparing fluorescence lifetimes of the donor ( mCer3 ) , pre- and post-photobleaching of the acceptor ( YFP ) , we could infer whether or not chosen protein pairs were closely physically associated . Direct physical interaction between proteins was indicated by a significant increase in the lifetime of mCer3 upon YFP photobleaching [69] . Fluorescence recovery of both mCer3 and YFP was monitored during acceptor photobleaching , and was used as an internal control for balancing the destruction of YFP signal and the preservation of mCer3 signal during experimentation ( Fig 3B ) . After identifying a proper photobleaching regimen , in pairs of known interacting proteins we observed that mCer3 signal would increase over the course of the initial photobleaching event , but not over subsequent treatments ( Fig 3B ) . This was consistent with what is expected when observing FRET , as a significant majority of the acceptor ( YFP ) is destroyed in the initial photobleaching event , and further photobleaching events have a reduced effect on the already diminished pool of YFP . As a positive interaction control , we initially tested NF-YB2:YFP by NF-YC3 , 4 , and 9:mCer3 and were able to consistently detect significantly increased mCer3 fluorescence lifetimes after YFP photobleaching ( Fig 3C ) . This is consistent with previous publications showing strong Y2H and in vivo NF-YB by NF-YC interactions [51 , 52 , 70–72] . As a negative control for each interaction test , we demonstrated that when NF-YB2 lacked the YFP fusion , mCer3 lifetimes were not altered after a photobleaching treatment ( Fig 3C ) . Substituting HY5:YFP for NF-YB2:YFP demonstrated that NF-YC9 could consistently physically interact with HY5; however , no FLIM-FRET interaction was detected between NF-YC3 or NF-YC4 and HY5 . Thus , it remains possible that HY5 only interacts with a subset of the light perception-regulating NF-YC proteins described here ( see Discussion ) . With the knowledge that at least some NF-YCs can physically interact with HY5 , we generated nf-yc triple hy5 mutants and examined them for hypocotyl elongation phenotypes in both SD and cWL ( Fig 4A–4C ) . Surprisingly , in SD conditions the nf-yc triple hy5 mutants were considerably longer than either parental mutant line , suggesting that the previously observed NF-YC roles in hypocotyl elongation were at least partially independent of HY5 . Even more striking was the strongly synergistic increase in hypocotyl elongation in cWL in the nf-yc triple hy5 mutants over both mutant parents ( Fig 4C ) . Compared to parental Col-0 , dark grown plants showed no differences in hypocotyl elongation for any of the mutant genotypes ( Fig 4D ) . Rescue assays confirmed that each gene ( NF-YC3 , 4 , 9 and HY5 ) was capable of significantly suppressing the nf-yc triple hy5 elongated hypocotyl phenotype ( S3 Fig ) . Collectively , these data demonstrated that the presence of HY5 masked the effects of the nf-yc triple mutant on hypocotyl elongation , especially in cWL conditions . These results are not trivially explained by cross regulation between NF-YC and HY5 as their transcription levels are only altered in their own mutant backgrounds ( S2 Fig ) . Finally , because some commercial white light sources contain contaminating UV radiation , we additionally examined hypocotyl lengths of cWL-grown plants grown under Mylar to filter out UV light . In accordance with previous work , hy5 mutants were longer in the absence of UV [73]; however , no difference was detected in the nf-yc triple mutant , and while not statistically significantly different , a minor difference observed in nf-yc triple hy5 mutants can be completely accounted for by the loss of HY5 ( S4 Fig ) . To further dissect the genetic relationship between NF-YC and HY5 , we compared the transcriptome profiles of seven day old , cWL grown nf-yc triple , hy5 , and nf-yc triple hy5 mutant seedlings using RNA Sequencing ( RNA-Seq , NCBI GEO accession GSE81837 ) . When compared to wild type , hy5 mutants had 1 , 368 up-regulated and 941 down-regulated genes , whereas analysis of differentially expressed genes in the nf-yc triple mutant showed a smaller set of 645 up-regulated genes and 493 down-regulated genes ( at least 1 . 5 fold , adjusted p < 0 . 05 , S1 Table ) . Direct comparison of the hy5 and nf-yc triple down-regulated genes showed substantial overlap , with approximately 40% of the nf-yc triple down-regulated genes being contained in the hy5 data set ( Fig 5A ) . Gene-ontology ( GO ) analysis for genes down-regulated in both the nf-yc triple and hy5 mutants identified enrichment in many categories involved in photomorphogenesis and early seedling development , including response to light stimulus and pigment biosynthetic processes ( S2 Table ) . Comparison between up-regulated gene sets yielded similar results with ~50% shared between the nf-yc triple and hy5 ( Fig 5B ) . GO enrichment analyses of genes up-regulated in both nf-yc triple and hy5 yielded categories in cellular stress responses and cellular responses to hormones , including ethylene , salicylic acid , and jasmonic acid ( S3 Table ) . To further investigate the regulatory relationship between HY5 and the NF-YCs , we analyzed genes that were either significantly up-regulated or down-regulated in nf-yc triple hy5 mutants ( S1 Table ) . These data sets were then sub-divided into four groups based on the level of differential gene expression in the nf-yc triple hy5 mutant relative to both nf-yc triple and hy5: Group I ) Genes differentially expressed more in the nf-yc triple hy5 mutant than both nf-yc triple and hy5; Groups II-III ) Genes differentially expressed more in the nf-yc triple hy5 mutant compared only to the nf-yc triple ( II ) or hy5 ( III ) ; and Group IV ) Genes not differentially expressed compared to either the nf-yc triple or hy5 ( i . e . , still differentially expressed in the quadruple mutant relative to wild type , but no change from nf-yc triple and hy5 ( Fig 5C and 5D ) ) . GO enrichment analyses of these four groups represent putative biological processes that NF-YCs and HY5 regulate cooperatively ( genes more differentially expressed in nf-yc triple hy5 than parental lines ) and independently ( genes not differentially expressed in nf-yc triple hy5 relative to nf-yc triple and/or hy5 ( S2 and S3 Tables ) ) . Analysis of genes significantly more down-regulated in the nf-yc triple hy5 mutant ( relative to its parental genotypes ) identified several over-represented categories , including flavonoid biosynthesis and polyol metabolic processes ( S2 Table ) . Among genes up-regulated to a similar level in the nf-yc triple hy5 , nf-yc triple , and hy5 data sets , and consistent with the synergistic hypocotyl phenotype of the nf-yc triple hy5 mutant , was a significant enrichment for genes involved in cell wall organization , cell wall biogenesis , and cell wall macromolecule metabolic processes ( S3 Table ) . Taken together , these data identify putative targets and biological processes regulated both cooperatively and independently by NF-YCs and HY5 , solidifying the existence of a complex functional relationship . Previous research established that the elongated hypocotyls in hy5 mutants are directly related to increased epidermal cell length [18]; therefore , we additionally examined individual files of epidermal cells along the hypocotyls of nf-yc triple hy5 mutants for total cell number and mean cell length ( Fig 6A–6C ) . The mean length of individual epidermal cells in hy5 ( 82μm ) was ~90% greater than Col-0 ( 43μm ) , while cells in the nf-yc triple mutant measured only ~15% longer than Col-0 . Reflecting the synergistic hypocotyl elongation phenotypes of nf-yc triple hy5 mutants , the epidermal cells of the quadruple mutant ( 158μm ) were ~270% longer than those measured in Col-0 . Total epidermal cells in the quadruple mutant were also increased >60% compared to Col-0 . Therefore , the very long hypocotyls of nf-yc triple hy5 mutants can be explained by a combination of comparatively modest increases in cell count and highly increased cell elongation . HY5 regulates photomorphogenesis regardless of wavelength , whereas HFR1 and LAF1 are more specific to FR light responses [24–28] . To better compare the spectrum of nf-yc mutant defects to these other transcription factors , we first examined both the nf-yc triple and nf-yc triple hy5 lines in cB , cFR , and cR over a gradient of light intensities ( Fig 7A–7C ) . Under all but the lowest cB fluence rates , the nf-yc triple hy5 mutants had significantly longer hypocotyls than all other lines ( Fig 7A ) . Considering our previous observation that nf-yc triple light perception defects were only apparent in SD conditions ( Fig 1B ) , it was somewhat surprising to find that nf-yc triple hy5 defects in cB were most pronounced at the highest light intensities ( Fig 7A ) . The nf-yc triple hy5 mutants were significantly longer than their nf-yc triple and hy5 parental lines under all cFR conditions ( Fig 7B ) . One possible cause for defects in FR light perception could be differential expression of HY5 , HFR1 , or LAF1 in nf-yc mutants—i . e . , NF-Y complexes could control the expression of these genes . S2 Fig shows that HY5 is not differentially expressed in an nf-yc triple background in cWL and we additionally examined the expression of HY5 , HFR1 , and LAF1 in cFR grown plants . Consistent with previous reports [35 , 74] , modest differences in HY5 , HFR1 , and LAF1 were either insignificant or not reproducible in repeated expression analyses in the various mutant backgrounds ( Fig 7D ) . We conclude that nf-yc mutant phenotypes are not likely related to simple changes in the expression of these well-known regulators of light perception . Further , as discussed below , nf-yc triple mutants appear to have a different spectrum of light defects than either hfr1 or laf1 mutants . Under low fluence rate cR and cFR , the nf-yc triple mutant alone had significantly longer hypocotyls than Col-0 , which is similar to hy5 ( Fig 7B and 7C ) . However , hfr1 and laf1 are only reported to have defects in cFR ( hfr1 and laf1 ) and cB light ( hfr1 ) . To confirm these previous reports with our experimental conditions , we directly compared hfr1 and laf1 to nf-y mutants under low fluence rate cR ( Fig 7E and 7F ) . As previously reported , hfr1 and laf1 appeared identical to wild type Col-0 plants , whereas the nf-yc triple mutants were consistently ~40% longer than Col-0 and similar to hy5 mutants . We additionally compared the nf-yc triple to hfr1 and laf1 in SD conditions ( Fig 7G and 7H ) . As expected , the hfr1 and laf1 mutants appeared phenotypically identical to Col-0 , while the nf-yc triple , hy5 , and nf-yc triple hy5 mutants were all significantly longer . Collectively , our data suggests that NF-YCs regulate hypocotyl elongation via an independent pathway ( s ) from HY5 , and at least partially independent of HFR1 and LAF1 , with broad roles in light perception at variable fluence rates . In the dark , HY5 , HFR1 , and LAF1 are all targeted for degradation by the proteasome in a COP1-dependent manner [28 , 75 , 76] . COP1 mutants ( cop1 ) have short hypocotyls and other photomorphogenic phenotypes even when grown in the dark , and these phenotypes are partially suppressed in hy5 cop1 , laf1 cop1 , and hfr1 cop1 double mutants [75 , 77 , 78] . Therefore , we examined if the nf-yc triple mutation could also suppress the short hypocotyl phenotype of dark-grown cop1-4 mutants . Similar to cop1 hy5 , an nf-yc triple cop1 mutant had ~80% longer hypocotyls than the cop1 single mutant when grown in constant darkness ( Fig 8A ) . Because cop1-4 mutants are known to have reduced rosette diameters ( dwarf phenotype ) and early flowering [79] , we further characterized these phenotypes in nf-yc triple cop1 mutants . For rosette diameter , the nf-yc triple mutant was once again able to partially suppress cop1 ( Fig 8B ) . One possibility is that this suppression is simply a function of nf-yc triple mutants being late flowering—i . e . , because they are later flowering , the rosettes have time to achieve a greater diameter prior to the phase change to reproductive growth . However , a control cross between cop1 and an even later flowering constans mutant ( the alternatively named co-sail or co-9 allele [80] ) had no impact on rosette diameter when crossed to cop1-4 . This suggests that nf-yc loss of function alleles genuinely suppress the small cop1 rosette diameter phenotype and that this particular NF-YC function is genetically separable from its role in flowering time . Finally , we also tested whether the nf-yc triple mutant could suppress the early flowering phenotype of cop1 ( Fig 8C ) . The nf-yc triple cop1 mutant plants flowered moderately , but significantly , later than Col-0 , intermediate to the early flowering cop1-4 mutant and late flowering nf-yc triple mutant . This result is not surprising as an important role for COP1 in flowering is to suppress CO function via protein degradation [79 , 81] . Because CO and NF-Y function together to regulate photoperiod-dependent flowering [49 , 52 , 57 , 82] , the basis of early flowering in cop1 is largely a function of CO protein ( and potentially NF-Y , see below ) hyper-accumulation [79 , 81] . Measurements of FLOWERING LOCUS T ( FT ) expression—the regulatory target of CO and NF-Y function in flowering time—perfectly correlated with expectations from the flowering time measurements ( Fig 8D ) . NF-YC regulation of light perception appears to share many parallels with HY5 , HFR1 , and LAF1 , including the suppression of cop1 mutant phenotypes [75 , 77 , 78] . As with these other photomorphogenic transcription factors , it is tempting to speculate that NF-YC proteins might be targets of COP1-mediated proteasome degradation in the dark . However , this does not appear to be the case as native antibodies to both NF-YC3 and NF-YC4 show modest fluctuations , but largely stable accumulation throughout both short day and long day cycles ( S5 Fig—recall also that expression of any one NF-YC from a native promoter rescues the mutant phenotype , S3 Fig ) . Nevertheless , NF-YC proteins function within the context of a heterotrimeric complex and reduction of the NF-YA or NF-YB components could also disrupt activity . In this regard , overexpression of most NF-YAs leads to small , dwarf phenotypes that are not unlike those observed for cop1 mutants [58 , 83] . In fact , when we examined NF-YA overexpressing plants ( 35S promoter driven; previously described in [58] ) , they were found to have significantly shortened hypocotyls in both cD and cR conditions ( Fig 9A ) . While shortened hypocotyls in cD is a classic constitutive photomorphogenic phenotype , expressing p35S:NF-YA hypocotyl lengths in cR as a percentage of cD growth additionally showed that most of these plants were specifically , additionally defective in red light perception ( Fig 9A ) . While it is unknown which of the 10 Arabidopsis NF-YAs is natively involved in hypocotyl elongation , two recent publications suggested that NF-YA2 may be found in complex with NF-YC3 , 4 , and 9 [84 , 85] . Therefore , using qPCR , we examined the expression of NF-YA2 in 24hr cWL or after 24-48hrs of cD and found that expression was strongly down-regulated in cD ( Fig 9B ) . At the same time we compared NF-YA2 expression to a subset of other NF-YA genes—NF-YA1 , 7 , 9 , and 10 . NF-YA10 is the most closely related paralog to NF-YA2 ( encoding 63% identical full length proteins , [86] ) and it showed the same pattern of down-regulation in cD . However , the less related NF-YA1 and 9 genes ( proteins are 42% identical to each other , but only 23–24% to NF-YA2 ) remained stably expressed in cD , while NF-YA7 was actually up-regulated . Thus , expression of the NF-YA gene family in response to cD is quite variable , and suggests potential for light regulated accumulation and depletion . To determine if NF-YA proteins might be targets of degradation in the dark , we examined the accumulation of NF-YA2 and NF-YA7 expressed from constitutive 35S promoters . We chose to use a constitutive promoter to differentiate between changes in protein accumulation due to reduced gene expression ( see Fig 9B ) versus active protein degradation processes . NF-YA2 protein accumulation was strongly reduced in cD conditions , even when expression was driven from the 35S promoter , suggesting an active degradation process ( Fig 9C ) . This was in stark contrast to NF-YA7 which maintained stable protein accumulation in the dark . To determine if the proteasome was involved in the process , we additionally performed cell-free protein degradation assays ( as previously described [87] ) and determined that NF-YA2 was rapidly degraded ( Fig 9D ) . However , the addition of the proteasome inhibitor MG132 strongly reduced the apparent degradation of NF-YA2 protein . We note that NF-YA7 also degraded in an MG132 dependent manner in these cell-free assays , suggesting that it can also be targeted by the proteasome for degradation , even if darkness may not be the driving force ( Fig 9B and 9C ) . Collectively , these data suggest that NF-YAs can also regulate light perception and are targeted for proteasome mediated degradation , perhaps controlling the overall stability of the NF-Y complexes necessary to suppress hypocotyl elongation in the light . Unexpectedly , while we were able to detect a physical interaction between NF-YC9 and HY5 through FRET-FLIM analyses , we were not able to detect an interaction between NF-YC3 or NF-YC4 and HY5 . This is surprising because the histone fold domains of NF-YC3 , 4 , and 9 are nearly identical ( in fact , NF-YC3 and NF-YC9 are identical [86] ) ; however , the amino- and carboxy-terminal regions are more divergent and could be involved in the NF-YC by HY5 physical interaction . Because of the extreme spatial constraints required for FRET to occur , it is not valid to conclude from these experiments that NF-YC3 and NF-YC4 cannot interact with HY5 [88 , 89] , and Y2H analyses did previously show a positive interaction between NF-YC4 and HY5 [57] . The question also remains whether or not the ability of NF-YCs to physically interact with HY5 is of biological importance relative to their specific functions in light signaling . Both HY5 and the NF-YCs are also regulators of ABA signaling [19 , 57 , 58] and it is possible that a physical interaction between them is only related to this or another undefined pathway . This possibility is supported by the additive , and even synergistic , mutant phenotypes of the nf-yc triple hy5 plants—i . e . , if these proteins are physically interacting in a linear pathway or at a common hub in light signaling , how does the quadruple nf-yc3 nf-yc4 nf-yc9 hy5 mutation result in these synergistic phenotypes ? Alternatively , arguing for the relevance of physical interactions in light signaling , there is clearly a significant amount of overlap in the putative regulatory targets of NF-YC3 , 4 , 9 and HY5 . Similarly , it was also previously suggested that subsets of photomorphogenic responses might be regulated by combinations of both overlapping ( where physical interactions were relevant ) and non-overlapping functions between HY5 , HFR1 , and LAF1 [35] . In future experiments , it will be informative to examine the stability of NF-YA proteins in the presence or absence of these other transcriptional regulators as it was previously shown that HFR1 and LAF1 were co-dependent for their protein stability ( i . e . , they required each other to avoid proteasome-mediated degradation; [74] ) . Ultimately , deciphering these putative cooperative versus individual roles remains an exciting challenge for future research . While we have demonstrated that NF-YC3 , 4 , and 9 function at least partially independently of HY5 in light perception , one pressing question is whether the NF-Y complex is functioning through other known light-responsive transcription factors , such as HFR1 or LAF1 . Directly addressing this hypothesis would require the creation of nf-yc triple hfr1 and nf-yc triple laf1 quadruple mutants; however , because HFR1 is linked to NF-YC9 , traditional crossing techniques would be prohibitively difficult . Therefore targeting of these loci in the nf-yc triple mutant with CRISPR-Cas9 is currently underway . To support that the NF-YCs are functioning separately , or at least differently , from HFR1 or LAF1 , we identified strong phenotypes in the nf-yc triple mutant in SD- and low cR-grown seedlings , where hfr1 and laf1 showed no mutant phenotype . Additionally , we found no differential expression of HFR1 or LAF1 in cFR-grown nf-yc triple mutants . These data suggest that the function of NF-YCs in light perception is at least partially separable from the functions of HFR1 and LAF1 . Similar to the NF-YCs , both STH2/BBX21 and STH3/BBX22 function as photomorphogenesis-activating transcription factors over a broad range of light conditions and are also able to physically interact with HY5 [29 , 30] . It is possible that the NF-YCs are functioning in an STH2/STH3-dependent manner; however , phenotypes of sth2 sth3 double mutants and sth2 sth3 hy5 triple mutants suggest that this might not be the case . In contrast to the genetic relationship between the nf-yc triple and hy5 , the hypocotyls of sth2 sth3 hy5 triple mutants were not longer than hy5 in cR . Further , we observed the most severe nf-yc triple phenotypes in low-intensity light while the sth2 sth3 mutant phenotypes were only observed in high-intensity light [30] . Nevertheless , these observations do not preclude genetic interactions for subsets of shared functions , similar to what we have already suggested with HY5 . When examining the protein domains of STH2/BBX21 and STH3/BBX22 , it is tempting to speculate that there may be indirect physical interactions with the NF-YC proteins as part of a larger light perception complex . STH2 and STH3 have B-box domains , thus their alternate BBX21 and BBX22 designations [90] , and these domains are necessary for direct physical interactions with HY5 [29 , 30] . BBX proteins also often have so-called CO , CO-LIKE , and TIMING OF CAB ( CCT ) domains [90 , 91] . For example , CO ( BBX1 ) is a BBX-CCT protein and mutations in either of these domains impacts its ability to regulate flowering time [92] . It is well-established that NF-YC3 , 4 , and 9 can all physically interact with CO and the CCT domain is both necessary and sufficient for this interaction [49 , 51 , 52] . While STH2 and STH3 do not have a CCT domain , recent evidence demonstrated that BBX proteins can heterodimerize with other BBX proteins [93] . Therefore , it is possible that NF-Y complexes may interact with BBX-CCT proteins via the CCT domain and recruit other non-CCT containing BBX proteins , such as STH2 and 3 , to these complexes . In contrast to our genetic evidence showing that NF-Ys act as suppressors of hypocotyl elongation , NF-YB9/LEC1 appears to have the opposite role . This idea comes from recent evidence demonstrating that inducible overexpression of NF-YB9/LEC1 also resulted in elongated hypocotyls , suggesting that NF-YB9/LEC1 might actually function as an enhancer of hypocotyl elongation [38 , 94] . Consistent with this finding , embryonic hypocotyls are shortened in lec1 mutants [62] . Further , recent data shows that the hypocotyls of both light and dark grown lec1 mutants are significantly shorter than wild type plants [39] . Interestingly , overexpression of a repressor of photomorphogenesis—PHYTOCHROME-INTERACTING FACTOR 4—results in elongated hypocotyls , but this phenotype is partially dependent on the presence of LEC1 [39] . These results raise a few interesting questions: why would NF-YB9 act opposite to the NF-YA and NF-YC members of the complex ( as reported here ) and what mechanism would allow this result ? Considering these questions , it is important to remember that each NF-Y subunit—A , B , and C—is part of a 10 member family [41] . Thus , many unique NF-Y complexes could theoretically form and , depending on their composition , actually act to competitively suppress or enhance a given process . In this scenario , some members of a given NF-Y family might enter a complex , but render it inactive , while other members of the same family would have the opposite effect . Our previous research on ABA-mediated seed germination provides some precedence for the above idea . We demonstrated that members of the same NF-Y family can act in opposing manners , either enhancing or delaying germination when their expression is altered ( both NF-YA and NF-YC examples exist; [57 , 58] ) . Similarly , some BBX proteins also show these opposing functionalities . For example , BBX24 and 25 are hypothesized to interfere with BBX22 function by entering into non-functional complexes with HY5 [95] . Fitting this scenario nicely , NF-YB9/LEC1 , and its closest relative NF-YB6/LEC1-LIKE , are quite unique and very different from the other eight NF-YB proteins in Arabidopsis . This includes 16 amino acid differences in their highly conserved histone fold domains that are completely unique to only this pair [96] . However , an alternative hypothesis must be considered related to the most recent lec1 data [39]: lec1 shortened hypocotyls may not be developmental patterns related to loss of skotomorphogenesis or post embryonic in nature , but , instead , are lasting patterns laid down during embryogenesis . This possibility is supported by both the modest magnitude of the effects and the finding that lec1 plants are short in all conditions ( dark and light ) . This is not the case for the nf-yc mutants reported here as they are indistinguishable from wild-type plants in the dark and elongated in the light , clearly defining them as positive regulators of photomorphogenesis . While the functional relationship between the NF-YCs and HY5 is similar to that observed with many other photomorphogenic transcription factors , the NF-YCs do not appear to be transcriptionally or translationally regulated in a manner consistent with light-responsive proteins; however , because the NF-YCs act in the larger context of an NF-Y trimer , the physical properties and regulatory components of the functional unit can be spread across multiple proteins . We showed that several NF-YA subunits with photomorphogenic phenotypes are regulated by light , and that NF-YA2 is targeted for degradation by the proteasome . Regulation of the NF-YA subunit establishes NF-Y complexes as possessing all of the properties generally expected of photomorphogenic transcription factors , including DNA-binding capacity , the ability to physically interact with other photomorphogenic factors , and a light-regulated mechanism to modulate function and abundance . While specific NF-YA subunits have not been conclusively identified to natively regulate the inhibition of hypocotyl elongation , a recent publication showed that over-expression of NF-YA2 led to earlier flowering [84] . This suggests that NF-YA2 could be integrated into an NF-Y complex containing NF-YC3 , 4 , and/or 9 , as each is also redundantly involved in photoperiod-dependent flowering [51]; finally , an NF-YA2/NF-YB2 , 3/NF-YC9 trimer has been identified through yeast three-hybrid analyses , and further verified through two-way interaction assays ( including BiFC and co-IP , [84] ) . The identity of photomorphogenic NF-YB proteins remains unknown and it will interesting to determine which , if any , non-LEC-type NF-YB will be positive regulators of photomorphogenesis . The data presented here firmly establishes NF-Y complexes as positive regulators of photomorphogenesis , significantly extending recent findings [38 , 39] . Future research on the potential regulation of NF-YA proteins by the proteasome and the identity of photomorphogenic NF-YA and NF-YB will improve our current understanding . Although only discussed at a cursory level here , research on NF-Y roles in flowering time demonstrate important interactions with the BBX protein CONSTANS and suggest that NF-Y by BBX interactions may be generalizable [49–52 , 97] . Given the numerous roles for BBX proteins in light perception [29 , 30 , 95 , 98–100] , we predict that future studies will uncover BBX by NF-Y interactions that are essential for light perception . This would be an exciting finding , significantly extending the regulatory reach and capacity of the four interacting families of proteins . All plants were of the Col-0 ecotype and were grown at 22C . Prior to starting germination on plates or soil , seeds were cold-stratified in a 4C walk-in cooler in the dark for 2–3 days . Plants grown in cWL were grown in a Conviron ATC13 growth chamber or a custom walk-in growth chamber . Plants in single wavelength light experiments were grown in a Percival E30-LED growth chamber after initial exposures to 4 hours of white light to induce germination . Plants used in flowering-time experiments , rosette diameter measurements , and GUS staining were grown in a previously-described soil mixture [51] . All other plants were grown on 0 . 8–2% agar plates supplemented with Gamborg B-5 Basal Medium ( PhytoTechnology Laboratories , product #G398 ) . Nicotiana benthamiana plants were grown under long-day conditions ( 16h light/8h dark ) at 22C in a Conviron ATC13 growth chamber . For UV experiments , plants were grown in cWL for 5 days under a Mylar filter ( Professional Plastics , catalog #A736990500 ) or mock-filter . GUS staining was performed as previously described on 5 day old soil-grown seedlings [86] , and images were taken on a Leica dissecting stereoscope . Flowering time was measured as the total number of rosette and cauline leaves present shortly after bolting , and all genotypes exhibited similar developmental rates . To quantify rosette diameter , plants were photographed from above at the time of bolting , and the Feret’s diameter was measured , anchored at the tip of the longest rosette leaf . To measure hypocotyl elongation , seeds were sown onto B5 supplemented plates with 2% agar and cold-stratified at 4C for 2 days . Before transfer to specific light conditions , all plates were set at room temperature in continuous white light for 4 hours . Plates were grown vertically for the duration of the experiments . Germination rates for the nf-y mutants under study were previously shown to be the same in B5 media and confirmed for the experiments reported here [57] . To facilitate proper measurement , all plants were straightened on the plate before taking pictures . Pictures were processed , and all individual hypocotyls were traced and measured in FIJI [101] . For individual cell length and total cell number measurements of single files of hypocotyl cells , seedlings were grown on plates as described above for 5 days in cWL . Seedlings were fixed in an FAA solution and dehydrated through sequential 30-minute incubations in 90% and 100% ethanol [102] . Fixed and dehydrated seedlings were individually mounted in the clearing agent methyl salicylate [102] , and immediately taken for measurement on a Nikon Eclipse NI-U compound microscope . Using differential interference contrast ( DIC ) optics , individual cell files were identified and measured manually through the NIS-Elements BR software ( Nikon ) , and pictures were taken at the hypocotyl-cotyledon junction for every seedling measured . Total RNA was isolated from 7-day-old seedlings grown under cWL conditions and from 5-day-old seedlings grown under cFR using the Omega Biotek E . Z . N . A Plant RNA Kit ( catalog #R6827-01 ) , and was DNase treated on-column with Omega Biotek’s RNase-free DNase set ( catalog #E1091 ) . First-strand cDNA synthesis was carried out with Invitrogen’s SuperScript III Reverse Transcriptase ( catalog #18080–044 ) and supplied oligo dT primers . qRT-PCR was performed on a Bio-Rad CFX Connect Real-Time PCR Detection System ( http://www . bio-rad . com/ ) , using Thermo Scientific’s Maxima SYBR Green/ROX qPCR Master Mix ( catalog #K0222 ) . Each genotype was assayed with three independent biological replicates , consisting of approximately 100mg of starting tissue each . White light grown seedlings were normalized to At2g32170 , while far red light grown seedlings were normalized to At3g18780 and At1g49240 . Statistical analysis and comparisons between samples was performed in the Bio-Rad CFX Manager Software ( http://www . bio-rad . com/ ) through use of the 2 ( −ΔΔCT ) method . The leaves of 4- to 6-week old N . benthamiana were co-infiltrated with Agrobacterium tumefaciens GV3101 strains harboring either a YFP-fused protein or an mCer3-fused protein , in addition to the Agrobacterium strain C58C1 harboring the viral silencing suppressor helper complex pCH32 [103] . Before infiltration into Nicotiana leaves , Agrobacterium cultures grown overnight were treated with 200uM acetosyringone in a modified induction buffer for 4 hours [104] . This induced culture was re-suspended in 10mM MES 10mM MgSO4 and directly infiltrated into young leaves . All downstream analyses were conducted 2–4 days after initial infiltration . FLIM data was acquired through time-correlated single photon counting ( TCSPC ) on a Lecia TCS SP8 confocal laser scanning microscope using an HC PL APO 40x/1 . 10 water immersion objective . Fluorescent protein excitation was achieved through use of a titanium-sapphire multiphoton laser ( Chameleon , Coherent ) operating at 120 femtosecond pulses of 858nm infrared light . Fluorescence emissions were detected by non-descanned hybrid detectors ( HyDs ) . Fluorescence lifetimes of entire nuclei were fit to a single-exponential model through the SymphoTime 64 ( www . picoquant . com ) software , and comparison of the fluorescence lifetimes before and after FRAP was used to detect FRET . For FRAP analyses , YFP photobleaching was accomplished with a high-intensity Argon laser line at 514nm for 15 seconds , followed by recovery imaging of both mCer3 ( excited at 458nm ) and YFP ( excited at 514nm ) every second for 5 seconds . Descanned HyDs were used to detect mCer3 emission from 459nm to 512nm , and a Photomultiplier Tube ( PMT ) was used to detect YFP emission from 512 to 562nm , with a 458/514 notch filter in place . This process was performed a total of 3 times for each nucleus , and both mCer3 and YFP intensities were calculated relative to initial fluorescence intensity . FRAP was conducted as an internal control during FLIM measurements , allowing us to assess the level of YFP photobleaching and ensure that relatively little mCer3 was inadvertently photobleached . Total protein was extracted from 14-day-old plants by grinding in lysis buffer ( 20 mM Tris , pH 8 . 0 , 150 mM NaCl , 1 mM EDTA , pH 8 . 0 , 1% Triton X-100 , 1% SDS with fresh 5 mM DTT , and 100 μM MG132 ) . NF-YA-CFP/HA was probed with high affinity anti-HA primary antibody ( cat#11 867 423 001; Roche ) and goat anti-rat secondary antibody ( cat#SC-2032; Santa Cruz Biotechnology ) . NF-YC3 and NF-YC4 were detected by previously described native antibodies [51] . The Bio-Rad ChemiDoc XRS imaging system was used for visualizing the protein blot after incubations with ECL plus reagent ( cat#RPN2132; GE Healthcare ) . Equivalent loading and transfer efficiency was determined by staining the protein blot with Ponceau S ( cat#P3504; Sigma-Aldrich ) . Seedlings grown for seven days on B5 media in continuous white light . Total RNA was isolated using the E . Z . N . A . Plant RNA Kit from ( Omega Biotek , Cat#R6827 ) . To ensure low levels of contaminating ribosomal RNA , two rounds of poly-A mRNA purification were performed using the μMACS mRNA Isolation Kit ( Miltenyi Biotech , Cat#130-090-276 ) . Indexed RNA-Seq libraries were prepared from 100 ng of poly-A RNA starting material using the NEXTflex Illumina qRNA-Seq Library Prep Kit ( Bioo Scientific , Cat#5130 ) . Sequencing of 150 bp paired end reads was performed on an Illumina HiSeq 2500 in rapid output mode at the Texas A&M Agrilife Research Facility ( College Station , TX ) . Sample de-multiplexing was performed using CASAVA software v1 . 8 . 2 and bcl2fastq was performed using conversion software v1 . 8 . 4 . Resulting sequences were trimmed and quality checked using the pipeline detailed at the iPlant Collaborative Discovery Environment ( http://www . iplantcollaborative . org ) . Sequences were mapped to the TAIR 10 representative gene models set using Burrows-Wheeler Aligner [105 , 106] within iPlant . Differential gene expression was determined using the Bioconductor package edgeR [107] . Gene Ontology over-representation analyses were performed in AmiGO 2 version 2 . 3 . 2 [108 , 109] . Raw sequencing data and the final differentially expressed gene lists were deposited with NCBI’s Gene Expression Omnibus , accession number GSE81837 . All image processing and figure construction was performed in either FIJI , Photoshop ( www . adobe . com ) , or Prism ( www . graphpad . com ) . Mutant lines used in this study , including references for their original derivation and description in the literature , are reported in S4 table [19 , 24 , 51 , 110–114] . AGI identifiers for all genes reported are also described in S4 Table .	Light perception is critically important for the fitness of plants in both natural and agricultural settings . Plants not only use light for photosynthesis , but also as a cue for proper development . As a seedling emerges from soil it must determine the light environment and adopt an appropriate growth habit . When blue and red wavelengths are the dominant sources of light , plants will undergo photomorphogenesis . Photomorphogenesis describes a number of developmental responses initiated by light in a seedling , and includes shortened stems and establishing the ability to photosynthesize . The genes regulating photomorphogenesis have been studied extensively , but a complete picture remains elusive . Here we describe the finding that NUCLEAR FACTOR-Y ( NF-Y ) genes are positive regulators of photomorphogenesis—i . e . , in plants where NF-Y genes are mutated , they display some characteristics of dark grown plants , even though they are in the light . Our data suggests that the roles of NF-Y genes in light perception do not fit in easily with those of other described pathways . Thus , studying these genes promises to help develop a more complete picture of how light drives plant development .	[ "Abstract", "Introduction", "Results", "Discussion", "Methods" ]	[ "plant", "anatomy", "plant", "growth", "and", "development", "gene", "regulation", "plant", "embryo", "anatomy", "regulatory", "proteins", "dna-binding", "proteins", "light", "microscopy", "plant", "physiology", "developmental", "biology", "plant", "science", "luminescent", "proteins", "microscopy", "yellow", "fluorescent", "protein", "transcription", "factors", "seedlings", "plants", "research", "and", "analysis", "methods", "plant", "embryogenesis", "plant", "development", "proteins", "gene", "expression", "fluorescence", "recovery", "after", "photobleaching", "biochemistry", "photomorphogenesis", "phenotypes", "embryogenesis", "genetics", "biology", "and", "life", "sciences", "hypocotyl", "organisms", "fruit", "and", "seed", "anatomy" ]	2016	NUCLEAR FACTOR Y, Subunit C (NF-YC) Transcription Factors Are Positive Regulators of Photomorphogenesis in Arabidopsis thaliana

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.