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
|
|---|---|---|---|---|---|
Developing physiologically-based pharmacokinetic ( PBPK ) models for chemicals can be resource-intensive , as neither chemical-specific parameters nor in vivo pharmacokinetic data are easily available for model construction . Previously developed , well-parameterized , and thoroughly-vetted models can be a great resource for the construction of models pertaining to new chemicals . A PBPK knowledgebase was compiled and developed from existing PBPK-related articles and used to develop new models . From 2 , 039 PBPK-related articles published between 1977 and 2013 , 307 unique chemicals were identified for use as the basis of our knowledgebase . Keywords related to species , gender , developmental stages , and organs were analyzed from the articles within the PBPK knowledgebase . A correlation matrix of the 307 chemicals in the PBPK knowledgebase was calculated based on pharmacokinetic-relevant molecular descriptors . Chemicals in the PBPK knowledgebase were ranked based on their correlation toward ethylbenzene and gefitinib . Next , multiple chemicals were selected to represent exact matches , close analogues , or non-analogues of the target case study chemicals . Parameters , equations , or experimental data relevant to existing models for these chemicals and their analogues were used to construct new models , and model predictions were compared to observed values . This compiled knowledgebase provides a chemical structure-based approach for identifying PBPK models relevant to other chemical entities . Using suitable correlation metrics , we demonstrated that models of chemical analogues in the PBPK knowledgebase can guide the construction of PBPK models for other chemicals .
Developing physiologically-based pharmacokinetic ( PBPK ) models for chemicals can be resource-intensive , as the formulation of new PBPK models is dictated by multiple factors . These factors include intended use for the model , target organism , target subpopulation ( e . g . , life stage , gender ) , endpoint of interest ( which can affect which organs are modeled individually and which are lumped together ) , routes of exposure , dosing regimen or exposure scenarios , and availability of relevant data for model calibration or evaluation . Among these factors , collecting chemical-specific data for parameterizing and calibrating a PBPK model is often the most resource-intensive task . For example , tissue-blood partition coefficients , or data that can be used to estimate these coefficients ( e . g . , log KOW ) , can be missing . Several in silico models are available for predicting tissue-specific partition coefficients based on chemical structure or properties [1–7] . On the other hand , computational tools for predicting a chemical’s metabolic pathway and rates of metabolism have been more difficult to develop due to widely variable interspecies ( e . g . , rat vs . human ) , intraspecies/interindividual ( i . e . , fast vs . slow metabolizers ) , and intra-individual ( i . e . , liver vs . kidney ) variation in metabolic activities [8] . Many environmental chemicals and most pharmaceuticals are metabolized in the body , and metabolites often exhibit drastically different pharmacokinetic properties and toxic effects than the parent compound or alternative metabolites of the parent compound [9] . While progress has been made toward increasing the accuracy of in silico predictions of metabolic parameters [10–12] , experimental data is always preferable , but much more costly to obtain . In addition to a dearth of chemical-specific data , time course measurements of tissue concentrations , along with dose-response measurements that reflect the disposition of a chemical and its metabolites inside the body , often do not exist , further impeding model validation . Chemical-specific parameters or in vivo pharmacokinetic data are unavailable for the vast majority of chemicals in commerce . Previously published PBPK articles are great resources to search for well-parameterized and thoroughly-vetted models that can inspire the structural design , code implementation , parameter optimization and experimental validation of models for additional chemicals . Incremental improvements , adaptations or modifications of existing models are common strategies used in the PBPK field to extrapolate chemical effects from laboratory animals to humans [13–17] , to incorporate additional exposure routes or life stages [18–22] , to link to pharmacodynamic endpoints [23–27] , or to build new models for similar chemicals [28–33] . While adapting a model to use for a different chemical has been demonstrated previously [4 , 34–41] , the actual process of selecting the most suitable published PBPK model for use as a starting template is not trivial . One strategy is to identify existing PBPK models that describe chemical analogues of the chemical entity of interest . This approach also works when adapting only a portion of the model ( i . e . , its compartmental structure ) for chemicals that are similar to previously modeled chemicals [42–44] . A simple , yet efficient way to identify analogous chemicals is by conducting a similarity search in a comprehensive knowledgebase . Many online tools , such as PubChem ( https://pubchem . ncbi . nlm . nih . gov/ ) and ChemSpider ( http://www . chemspider . com/ ) provide similarity-searching capabilities on generic sets of chemicals , but currently no repository exists specifically for PBPK models . Thus , the objective of this study is to compile a knowledgebase that contains PBPK modeling-related literature annotated with respective chemical structures along with several easily accessible molecular descriptors for these chemicals . These molecular descriptors can then be used to build a correlation matrix for each unique chemical in the knowledgebase . The knowledgebase can be queried by inputting the structure of the chemical of interest so that existing PBPK-related literature containing that chemical’s close analogues might be found . Illustration of this approach involved two case studies . In the first , the PBPK knowledgebase and correlation matrix were applied in the development of a new PBPK model for ethylbenzene using parameter values from six chemicals . These new models were then evaluated by comparing model-simulated blood concentrations of ethylbenzene against measured literature values . In the second case study , a published model of gefitinib was used to predict blood concentrations of its close-analogues and non-analogues categorized using the PBPK knowledgebase and correlation matrix . In addition to enhancing the efficiency of analogue-based PBPK model construction for additional chemicals , the power of the PBPK knowledgebase lies in its compilation of a wealth of information related to these PBPK chemicals , such as time course tissue concentration data , dose-response data , the authors’ assumptions about the model , limitations and applications of the model , and cited material . The PBPK knowledgebase directs users to published knowledge describing a specific chemical in order to aid in the development of new PBPK models for additional chemicals of interest .
A compilation of all Supplementary Tables from the current study were summarized in a separate web repository in csv format ( https://sites . google . com/site/pbpkknowledgebase/supplementary-materials ) . An open-source web interface is currently under development to provide intuitive navigation to data of interest for users . An abstract-based PBPK corpus was created to provide a comprehensive composition of PBPK-related literature using PubMed ( http://www . ncbi . nlm . nih . gov/pubmed ) . Query parameters included: “pbpk OR ( “physiologically based” AND ( pharmacokinetic OR toxicokinetic ) ) ” . URL: http://www . ncbi . nlm . nih . gov/pubmed/ ? term=pbpk+OR+ ( %22physiologically+based%22+AND+ ( pharmacokinetic+OR+toxicokinetic ) ) Additional search filters included “Abstract/title only . ” No publication date boundaries were set for the query . Search results returned articles that were available only as early as 1977 . All search results were saved and exported as a text file ( S1 Table ) . The PBPK abstract corpus ( as a text file ) was loaded into Google sites to be processed through www . chemicalize . org ( developed by ChemAxon ) , which is a public web resource that uses chemical named-entity recognition ( NER ) and a chemical taxonomy mark-up utility to identify unique chemical structures from text . The entire corpus was subdivided into smaller sections ( ~400 abstracts per set ) to accommodate the processing capability of chemicalize . org . The marked-up page source was copied into Microsoft Excel 2007 , parsed , and filtered so that the only entries remaining were chemical names and PubMed manuscript ID ( PMID ) ( a unique database-designated index for cataloging purposes ) . This process identified 795 abstracts containing specific chemical names; results are summarized elsewhere ( S2 Table ) . Because many chemicals have more than one abstract associated with each chemical name , the CAS registry number and SMILES string for these chemicals were obtained from other databases ( e . g . , ACToR [http://actor . epa . gov/actor/faces/ACToRHome . jsp ) , DSSTox [http://www . epa . gov/ncct/dsstox/] , and ChemSpider [http://www . chemspider . com/] ) . Duplicates and synonyms were removed based on the CAS registry number and SMILES strings . For quality control purposes , two authors manually curated the chemical list to ensure that the knowledgebase contains only specific chemical entities ( e . g . , “ethyl” was excluded ) and that PBPK models exist for these chemicals ( e . g . , existing studies measuring kinetic data for a specific chemical that could be used to build a PBPK model ) . After the manual curation , 307 unique chemicals remained . Their chemical names , CAS registry numbers and SMILES strings are provided elsewhere ( S3 Table ) . The 795 abstracts ( S2 Table ) and corresponding 307 unique chemicals ( S3 Table ) are referred to as the “PBPK knowledgebase” throughout this article . The abstracts in the PBPK knowledgebase were analyzed in order to identify the presence or absence of PBPK-associated word-stems . The purpose of this analysis was to improve our knowledge of the type of PBPK model information that could be expected from a publication . The PBPK-associated word-stems selected for our analyses were as follows: Species included “rat , rats , mouse , mice , human , pig , cow , goat , guinea pig , hamster , marmoset , monkey , rabbit , rhesus , rodent , sheep , bird , chicken , fish , pony , swine , turkey , and whale . ” Life stages included “adult , pregnant , children , lactating , fetus , infant , dam , neonate , pediatric , pup , child , fetal , neonatal , and maternal . ” Gender included “female , male , man , woman , men , and women . ” Compartmental organs included “cutaneous , venous , arterial , carcass , body , fin , skin , lungs , heart , adipose , fat , brain , kidney , liver , bone , placenta , testes , ovary , breast , hepatic , blood , urine , plasma , plasma , feces , fecal , renal , milk , and hair . ” Mining for these terms in each chemical name-containing abstract was performed using the open-source statistical program R ( R Foundation for Statistical Computing , Vienna , Austria ) to create a presence ( 1 ) or absence ( 0 ) vector ( summarized in S2 Table ) . Absorption , distribution , metabolism and elimination ( ADME ) of chemicals are largely governed by their physicochemical properties [2 , 3 , 45–48] . For each of the chemicals identified in the PBPK abstract corpus , eight easily obtainable 2D physicochemical molecular descriptors were calculated using the proprietary software Molecular Operating Environment ( MOE ) ( Chemical Computing Group Inc . , Montreal , QC , Canada ) . These descriptors include molecular weight ( MW ) , hydrogen bond acceptor count ( hba ) , hydrogen bond donor count ( hbd ) , number of rotatable bonds ( nRotB ) , polar surface area or topological polar surface area ( PSA ) , octanol:water partition coefficient ( logP ) , log transformation of solubility ( logS ) and area of van der Waals surface ( vdw_area ) . Descriptor values are summarized in S3 Table . These descriptors are commonly accepted by the research community as correlated with chemicals’ pharmacokinetic properties . MW , hba , PSA , logP and logS have been associated with human intestinal absorption [49–51] . MW hba , hbd , nRotB and PSA can be used to predict clearance and volume of distribution [52] . MW , logP , hba , hbd , PSA , nRotB and logS have been associated with percent binding to plasma and liver microsomal proteins [53 , 54] . MW , hba , hbd , PSA , logP were included in the in silico identification of cytochrome P450 isoform-specific substrates [55 , 56] . Because other studies have shown that increasing the number of descriptors does not necessarily increase the predictive power from descriptor to PK properties [47 , 57] , and to limit descriptors to those that are easily accessible to the public , no additional descriptors were calculated for this study . In this study , the similarity between chemicals was calculated as correlation coefficients based on the eight descriptors described above . Since the scientific community lacks consensus on the weight of importance for each descriptor toward a chemical’s pharmacokinetic properties , each descriptor was considered to contribute equally to the calculation of correlation coefficients . Six of the eight descriptors , hba , hbd , nRotB , PSA , vdw_area and MW , have values 0 or above and are positively skewed to the right . Thus , a log transformation was conducted to normalize these descriptors . The remaining two descriptors , logP and logS , exist in their transformed states . All the log-transformed descriptors were converted to standard normal distribution ~N ( 0 , 1 ) , based on Eq 1 . Where is the normalized value of the kth descriptor for chemical i , X_k_i is the value of the log-transformed kth descriptor for chemical i , and μ_k and σ_k are the respective mean and standard deviation values of the log-transformed kth descriptor for all chemicals . Normalized molecular descriptors for all chemicals in the PBPK knowledgebase are summarized in S4 Table . A correlation coefficient was then calculated based on the normalized molecular descriptors , as shown in Eq 2 . Where Cij represent the correlation coefficient between chemicals i and j in the knowledgebase . XST_k_i and XST_k_j represent the normalized kth descriptor of chemical i and j , respectively . The pairwise correlation coefficients matrix for each chemical in the PBPK knowledgebase is summarized elsewhere ( S5A Table ) . Each cell in the matrix represents the correlation coefficient between two chemicals ( column and row names ) . This matrix has been further flattened into chemical-pairs , and then ordered by rank based on their correlation coefficient values . The rank-ordered correlation coefficients of chemical pairs are provided in S5B Table . To demonstrate the utility of the PBPK knowledgebase in finding analogous chemicals with existing PBPK models that could act as a starting template to build a new model , ethylbenzene was used as a case study . Six chemicals with varying structural similarities towards ethylbenzene were selected from the PBPK knowledgebase , and the equations/parameters from their existing models were used for the construction of an ethylbenzene PBPK model . The simulation results from these newly constructed models were compared to the experimental data on ethylbenzene [29] . Models for the anti-cancer drug gefitinib [61] were coded and executed in Matlab in order to predict blood concentrations for seven other chemicals of varying similarity to gefitinib . Predicted blood concentrations for these seven structural analogues were then compared against measured values [26 , 61–67] . All entries in the PBPK knowledgebase were first ranked based on their similarity toward gefitinib , as described above ( S8 Table ) . Close- and non-analogues of gefitinib were selected from the top and bottom of the ranking list . Because some of the top or bottom ranked chemicals do not have published experimental data , only entries that are associated with experimental data were kept as examples . Four close-analogues ( itraconazole , cocaine , diclofenac and 3 , 3'-diindolylmethane ) and 3 non-analogues ( perchlorate , phosphorothioate oligonucleotide , and melamine ) of gefitinib were selected for experimental data extraction ( Table 3 ) . The existing gefitinib model [61] was used to simulate the blood concentrations for these selected example chemicals . Dose and body weight ( BW ) were obtained from references for each chemical [26 , 61–67] . The volume of distribution ( V1 , V2 ) for each new chemical was linearly scaled by body weight . For example , V1_itraconazole = ( BW_itraconazole /BW_ gefitinib ) * V1_ gefitinib . Other parameters were not altered from the original gefitinib model ( Table 4 ) . Predicted blood concentrations were then compared to published experimental concentrations for each chemical . Calculation of Chi Square statistics ( χ2 ) as an indication of goodness-of-fit was performed as described above .
The 2 , 039 PBPK-related articles were assigned to one of three categories ( Fig 1A ) : publications on unique chemicals that appeared for the first time ( likely to be a newly-developed PBPK model ) ; publications on chemicals that appeared in previous publications ( likely to be an application or refinement of a previously-developed PBPK model ) ; and reports on general PBPK concepts , methods , commentaries , perspectives , or reviews . Regression analysis was performed for the three categories ( Fig 1B ) . Linear relationships between the number of publications and the year of publication were calculated to help identifying the growth rates . The growth rates for publications are 14/year , 36/year , and 78/year for the first , second , and third categories , respectively . These trends reflect the difficulty in developing a new PBPK model due to the great quantity of experimental data required . While our search suggests ongoing development and expansion of PBPK-related modeling methodologies , the low output of new PBPK models limits the utility of PBPK modeling for examination of health risks resulting from chemical exposures . When comparing the two gender keywords , “male” appeared much more frequently than “female” ( 66% vs . 34% ) . “Human” and “rat” were the most frequently mentioned species , and these key words appeared at about three times the frequency of “mouse” ( Fig 2A ) . “Dog , ” “rabbit , ” “monkey , ” “fish , ” and “pig” comprised the 4th to 8th most common animal species mentioned in the abstracts , but with noticeably lower frequency than the top 3 species , which accounted for >94% of the total ( Fig 2A ) . The most frequently mentioned life stage was “adult , ” appearing three times more frequently than the second-most frequent term “pregnant” ( Fig 2B ) . For the top 9 life stage terms mentioned in PBPK-related literature , the key words “pregnant , ” “dam , ” and “lactating” refer to the reproductive cycle of the female parent; the key words “fetus , ” “neonate , ” “infant , ”“pediatric , ” and “children” refer to growth and developmental stages of offspring ( Fig 2B ) . “Blood” and “liver” were the two most frequent organs incorporated into PBPK models , with appearance frequencies twice as high as the 3rd most frequent organ , “fat ( adipose ) ” ( Fig 2C ) . “Brain , ” “kidney , ” “lungs , ” “gut ( intestine ) , ” “skin , ” “heart , ” and “spleen” comprised the 4th to 10th most frequent organs mentioned in these publications . The means and standard deviations of hba , hbd , nRotB , logP and logS were much smaller than those of PSA , vdw_area and MW ( Fig 3A ) . After normalization of the physicochemical molecular descriptors for these chemicals using Eq 1 above to reduce bias in calculation of correlation coefficients , the mean and standard deviation of each descriptor was set as 0 and 1 , respectively ( Fig 3B ) . Each cell in the correlation matrix contains the correlation coefficient of one chemical toward another chemical in the knowledgebase . The top five correlation coefficients were equal to 1 , because those five chemical pairs contained identical molecular descriptors . These chemicals were either chiral isomers or isotopically-labeled compounds . The remaining chemical-pair combinations exhibited a maximum correlation coefficient of 0 . 999990409 ( between 1 , 2 , 4-trimethylbenzene and 1 , 2 , 3 , 5-tetramethylbenzene ) and a minimum correlation coefficient of 0 . 589788342 ( between ethylene and methyl mercury ) . Simulated blood concentrations from the PBPK model based on an “exact match” ( ethylbenzene ) [58] aligned extremely well with the experimental data [29] ( Fig 4A ) . Simulated blood concentrations from PBPK models based on “close-analogues” ( xylene , toluene and benzene ) [58] deviated slightly from the ethylbenzene data ( Fig 4B and 4C and 4D ) . In contrast , simulated blood concentrations from the models based on “non-analogues” ( dichloromethane and methyl iodide ) [59 , 60] exhibited significant deviations from the experimental data on ethylbenzene ( Fig 4E and 4F ) . The PBPK model based on an “exact match” ( ehtylbenzene ) resulted in the highest χ2 goodness-of-fit p-value of 0 . 9991; PBPK models based on “close-analogues” had p-values of 0 . 8603 , 0 . 5789 and 0 . 1479 for xylene , toluene , and benzene , respectively; PBPK models based on “non-analogues” ( dichloromethane and methyl iodide ) resulted in much lower p-values of <6 × 10−215 . Fig 5 shows published experimental observed blood concentrations for each example chemicals compared to their predicted values from the each of the accommodated gefitinib models ( dose , BW , V1 , V2 adjusted ) . Chi Square statistics ( χ2 ) were calculated and stored in S9 Table . The gefitinib model fit the best with its own experimental data , with χ2 test p-values equal to 0 . 999 . The predictive ability of the gefitinib model for the structural analogues cocaine and 3 , 3'-diindolylmethane were high , with χ2 goodness-of-fit p-values of 0 . 994 and 0 . 898 , respectively . The χ2 test p-values for the other two structural analogues itraconazole and diclofenac were not as high , but still better than those of non-analogues , with χ2 test p-values of 2 . 81 × 10−21 and 5 . 71 × 10−14 , respectively . The χ2 test p-values for non-analogues were all zero .
Researchers have used molecular modeling approaches ( e . g . , quantitative structure-activity relationships ) to predict parameters , such as volume of distribution and clearance rate , to fill data gaps when building new PBPK models for chemicals lacking these data [52 , 68–70] . This approach can be labor intensive and requires background knowledge in computational chemistry and statistics . Utilizing pre-existing models with well-calibrated parameters to help new model construction is a more efficient approach and has been widely implemented [28–33] . However , reviewing or sorting through publications for relevant information often would be an overwhelming task for investigators . The PBPK knowledgebase presented in this current work serves as an effective means for finding analogues whose publications contain necessary model information and/or data that can aid the construction or validation of new PBPK models for new chemicals . One-compartment ( whole body ) and two-compartment ( blood and the remainder of the body ) models are simplest classical pharmacokinetic models . In these models , the organ-structure and mathematical equations remain the same for different chemicals , so extracting parameters from pre-existing models would be reasonable approach [61] . When a classical PK model extends beyond two compartments , grouping and integration of organs into hypothetical compartments often occurs [71] . For example , the liver might be integrated into one compartment ( e . g . , compartment 3 of a five-compartment model ) for one chemical but integrated into an entirely different compartment ( e . g . , compartment 4 ) for another chemical . This difference in integration could lead to changes in not only chemical-specific parameters , but also in physiological parameters ( e . g . compartmental volume , protein content , metabolic capacity ) for compartment 3 and 4 , respectively . Therefore , extraction from pre-existing models would not be an appropriate parameterization approach for high-dimension classical PK models . Compartments in PBPK models correspond to real biological organs ( tissues ) , so the physiological parameters of compartments ( e . g . , blood flow , organ volume , and protein content ) are highly conserved among PBPK models [72] . Chemical-specific parameters for each organ , such as tissue:blood partition coefficients and fraction of protein binding , are related to organ structure and the compound’s physicochemical properties . These characteristics enable researchers to use parameters from an analogue’s pre-existing model for a new PBPK model of new chemicals . However , mathematical equations in PBPK models are determined on a case-by-case basis , with variation in the number of compartments , type of compartments ( flow-limited or diffusion-limited ) , chemical-specific elimination routes , active transport of the parent chemical and/or metabolites , and other factors [73] . Therefore , borrowing parameters from other PBPK models would be a case-by-case practice and require extensive browsing and reading of relevant publications . The PBPK knowledgebase not only ranked the publications based on structural similarity ( S5A Table ) , but also summarized much essential information , such as chemical names , organs , genders and species ( S2 Table ) . It will help expedite selection and reading through relevant publications for locating appropriate parameter values . The current estimate for the number of chemicals in commerce in the United States is nearly 100 , 000 , with 500 to 1000 new chemicals being produced each year [74] . The current rate of PBPK model development ( ~14 chemicals/year; Fig 1B ) will likely never catch up with rate of new chemical production . Just covering the 1 , 800 chemicals found in consumer products would take more than 100 years . The new strategy presented in this current work is designed to facilitate the generation of provisional pharmacokinetic models as chemical inventories continue to expand . The PBPK knowledgebase can be used to gauge the chemical space of existing PBPK models , as well as developing a methodology to search for existing PBPK models for structural analogues of chemicals of interest . It is up to the discretion of future investigators whether to use our proposed approach based on eight molecular descriptors to select an analogue or to use a different similarity testing method based on the intended purpose of the new model . Identifying structural analogues is an ongoing area of research . Commonly accepted numerical measurements of chemical similarity are distance coefficients based on chemical descriptors , such as binary ( 0 or 1 ) values , indicating the absence or presence of some particular feature , topological indices , physicochemical properties ( sometimes estimated using in silico approaches ) , or on/off indications for molecular fingerprints [75] . A large number of similarity calculations have been defined and used in the literature , including Euclidean [75 , 76] , Hamming [75 , 76] , Minkowsky [76 , 77] , Correlation [76 , 78] , Tanimoto [75 , 76] , Molecular Access System ( MACCS ) [79 , 80] , and Artificial Neural Network ( ANN ) [81 , 82] . Currently , there is no consensus regarding the best practices in selecting molecular descriptors and similarity calculation methods . Pearson’s correlation coefficient has previously been used in similarity calculations [76 , 78] . We chose this metric because it equally-weighted the descriptors and also can easily be calculated through a simple R programming language script . The eight molecular descriptors ( Fig 3 ) presented here were selected based on their relationship with pharmacokinetic properties [47 , 49 , 52 , 57 , 83] and their ease of accessibility to the general public . Route of administration , as well as molecular properties of chemicals , can influence chemical behavior entering into a biological system . Route of administration determines the bioavailability , peak blood/tissue concentrations ( Cmax ) , time of peak concentrations ( tmax ) , biological half-life ( t1/2 ) , and other pharmacokinetic characteristics [84 , 85] . When intravenously injected , ethylbenzene is 100% bioavailable and reaches tmax at time 0 . When exposed through oral administration , the bioavailability of ethylbenzene is influenced by metabolic degradation in gut tissue , by gut lumen bacteria , and by liver hepatocytes [86–88] . If exposed through inhalation , ethylbenzene’s bioavailability and rate of absorption is determined by the air:blood partition coefficients and gas exchange rate of the lung [89 , 90] . The physiological parameters of the model change with a given animal species . Although extrapolation between species is a common practice in PBPK modeling [91] , the original model is more appropriate when using the same species as that used to derive the experimental data . Therefore , in our ethylbenzene case study which used model predictions comparable to measured time course data obtained from a rat inhalation study [29] , “inhalation” and “rats” were used as filters for the PBPK knowledgebase before chemical selection . The three categories of “exact match , ” “close-analogue , ” and “non-analogue” in our case study represent the major scenarios that can aid researchers in model development using the PBPK knowledgebase . For a given chemical , if an “exact match” entry is found , the existing model should have the best predictive capability . Although the same information may be retrievable through a PubMed search for the chemical , our PBPK knowledgebase can be used to search existing models that might have been built based on alternatives for a chemical ( e . g . , synonyms , chiral isomers or isotopically-labeled compounds ) . For example , ChemSpider lists 21 synonyms for ethylbenzene . The PBPK knowledgebase provides a more efficient and more precise solution , especially for those without a background in chemistry: any synonyms , chiral isomers or isotopically-labeled compounds would have a calculated correlation coefficient of 1 ( S5B Table ) . Besides being able to search for an “exact match , ” the power of the PBPK knowledgebase lies in its ability to detect “close-analogues” of a chemical simply by searching for the highest correlation-ranked entries . Searching for “close-analogues” without the knowledgebase could potentially be achieved by a two-step process of ( 1 ) searching public chemical databases for structural analogues; and then ( 2 ) searching PubMed for existing PBPK models for each analogue . This two-step process , however , is unnecessarily time-consuming . Although many publications exist that contain PBPK-related models and information , this number pales in comparison to the hundreds of thousands of chemical structures found in public chemistry databases ( e . g . , ChemSpider , PubChem ) [92] . It is more efficient to start the search of analogues found within the PBPK knowledgebase , which contains 307 entries rather than with the entire universe of chemicals . For example , in a structural-similarity search using ethylbenzene on ChemSpider , more than 10 , 000 results were retrieved with a Tanimoto score >99% . Xylene and toluene , which ranked 1st and 3rd in the PBPK knowledgebase , were not included in the top 100 of this list of chemical analogues for ethylbenzene identified in ChemSpider . Studies have shown that not all structural properties are associated with chemicals’ PK properties [2 , 3 , 45–48] . Using all available molecular descriptors , such as the structural analogue algorithm in the public chemical database , may result in a less accurate estimation of the desired PK property analogue . The two “non-analogues” of ethylbenzene were selected from lower correlation-ranked entries in our ethylbenzene case study to confirm that the parameter values for non-analogues differ most from parameter values for “exact matches” ( Table 2 ) , and that simulations from PBPK models of non-analogues deviate most from experimental data ( Fig 4E and 4F ) . Since experimental data for the target chemical , ethylbenzene , was available in our case study , it was straightforward for us to categorize the knowledgebase entries as “close-analogues” or “non-analogues” by comparing model predictions with data . For a chemical of interest lacking experimental data , there is no clear way to select a threshold for similarity rankings . A proposed rule of thumb is to select three to five chemicals from the first ten correlation-ranked entities , and then the “best” published model is picked from this shortlist . We caution that this recommendation is subjective , and the choice of the best model should be rooted in the quantity and applicability of the data that is available from published research to calibrate the model . For example , a model that was calibrated using time course data in multiple tissues in animals and evaluated against human data would be considered a better model than one that was calibrated using only urinary metabolite data and was not evaluated against any human data . Our second case study with gefitinib further demonstrates the utility and versatility of the PBPK knowledgebase . A pre-existing PBPK model of gefitinib was accommodated to predict the experimental observations of other chemicals , selected from the top and bottom of similarity-ranked PBPK knowledgebase entries . The simulated drug kinetics in Fig 5 and calculated χ2 test p-values in S9 Table , demonstrated that the gefitinib model gave better predictions for the closer structure analogues ( cocaine , 3 , 3'-diindolylmethane ) than to non-analogues ( perchlorate , phosphorothioate oligonucleotide , melamine ) . These results supported the theoretical assumption for using close-analogue’s existing parameters for a new chemical’s model construction [4 , 34] . The PBPK knowledgebase not only contains the chemical names , animal species , route of administrations , and tissue compartments ( all of which were extracted and used for indexing and searching in the current manuscript ) , but also can facilitate the discovery of corresponding experimental data that can be easily extracted from publications . Such extracted data was used to test the appropriateness of gefitinib’s model in our second case study . These data can also serve additional needs and interests of knowledgebase users . Commercial software packages such as SimCyp , Gastroplus or PKSim are used extensively in the pharmaceutical industry , as well as in academia , to support rapid PBPK model development [93–96] . We wish to emphasize several fundamental differences between these commercial software packages and our PBPK knowledgebase . Firstly , values of chemical-specific parameters in the knowledgebase are either measured or optimized against experimental data; while in SimCyp , Gastroplus and PKSim , chemical-specific parameters are often QSAR-based predictions . Secondly , our knowledgebase provides more information than merely model code and parameter values . Through its abstract corpus compilation , the knowledgebase also refers users to relevant articles containing time course tissue concentration data , dose-response data , the authors’ assumptions , limitations and applications of the model , and cited resources , in regards to chemicals of interest . Thirdly , our knowledgebase is free to the public and acts as a central location for abstract information relevant to chemicals of interest . Users accessing this information can contact the authors of the publications for additional information pertaining to model code or related data , while commercial software packages require licensing fees . Finally , the model structures in the published literature , which can be easily located from abstract information provided in the knowledgebase , were constructed based on the authors’ expert judgement and modeling philosophy ( e . g . , top-down vs . bottom-up ) ; in SimCyp , Gastroplus , and PKsim , the model structure is primarily generic . Although a modifiable , generic model structure is easy to use , more experienced users may prefer to construct their own models based on data availability and purposes of the study . The knowledgebase herein provides abstracts for previously published PBPK articles , beginning from 1977 onwards . With extracted information from these published articles , the knowledgebase can aid users in identifying the most relevant publications . Use of this extracted information is highly dependent on the scientific questions and problems of interest and is applicable mostly on a case-by-case basis . The two case studies presented here represent two specific circumstances addressing different scientific queries . Future users should select a strategy that meets their individual needs , based on data availability and study purpose , when using the knowledgebase . In summary , a PBPK knowledgebase was compiled that contains a thorough documentation of the chemical space of PBPK models . This knowledgebase provides scientists with a structure-based approach to identify provisional nearest-neighbor chemicals that are described by existing PBPK models and whose existing models might be used as a template to construct new models for chemicals of interest . These comprehensive dataset initiatives can be coupled with in vitro or in vivo chemical and biological data curated and accessible from other sources ( e . g . , STITCH 4 . 0 , CSS Dashboard , Comparative Toxicogenomics Database ) , or with more recent methods such as in silico multi-target profiling in DockScreen [97] , in order to pave the way for more rapid PBPK model development . Such approaches can complement efforts to rapidly develop PBPK/PD models that are designed to act as supporting computational tools in modern risk assessment .
|
Physiologically-based pharmacokinetic ( PBPK ) models are complex kinetic models describing the absorption , distribution , metabolism and excretion of chemicals in humans or animals in vivo . They can be utilized in many applications , such as dosimetry testing , toxicological investigations , and chemical risk assessment . De novo construction of PBPK models can be very challenging when chemical data are limited . Previously developed PBPK models from structurally similar chemicals can provide valuable insight in the construction of a new model . We compiled a PBPK knowledgebase that contains the chemical space covered by existing PBPK models . This knowledgebase indexes PBPK publications with chemical names , animal species , routes of administration , and model compartments . The knowledgebase aids new PBPK model generation by providing a structure-based approach to identify literatures related to provisional nearest-neighbor chemicals . Such approaches can complement efforts to develop de novo PBPK models that might act as supporting computational tools in modern risk assessment .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"chemical",
"compounds",
"organic",
"compounds",
"toluene",
"physiological",
"parameters",
"aromatic",
"hydrocarbons",
"pharmacology",
"drug",
"metabolism",
"hydrocarbons",
"chemistry",
"iodides",
"hematology",
"xylenes",
"pharmacokinetics",
"blood",
"anatomy",
"organic",
"chemistry",
"benzene",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
] |
2016
|
Developing a Physiologically-Based Pharmacokinetic Model Knowledgebase in Support of Provisional Model Construction
|
We present a theory of decision-making in the presence of multiple choices that departs from traditional approaches by explicitly incorporating entropic barriers in a stochastic search process . We analyze response time data from an on-line repository of 15 million blitz chess games , and show that our model fits not just the mean and variance , but the entire response time distribution ( over several response-time orders of magnitude ) at every stage of the game . We apply the model to show that ( a ) higher cognitive expertise corresponds to the exploration of more complex solution spaces , and ( b ) reaction times of users at an on-line buying website can be similarly explained . Our model can be seen as a synergy between diffusion models used to model simple two-choice decision-making and planning agents in complex problem solving .
Decision-making has been studied in great detail relying on binary choices by the Two-Alternative Forced-Choice paradigm ( 2AFC ) . In 2AFC tasks , choices are made between two alternatives with limited information while speed and accuracy are registered . In addition , for simplicity , in the vast majority of the experiments , the decision variable is a single scalar ( for example the luminosity of patches , the number of dots in a comparison task or the pitch in an auditory discrimination task ) . This paradigm has a great benefit for computational and theoretical understanding of decision making . It can be fully expressed in a set of equations which have analytic solution [1] . Also , the functional dependence of behavioral observables such as response times , error rates or confidence can be described in detail by low dimensional models ( e . g . [1–5] ) . While models differ , many of them rely on a stochastic search process , in which the accumulation of the evidence is integrated over time , and whose crossing of a boundary represents the event of reaching a conclusion or making a decision [6–8] . The Drift Diffusion Model ( DDM ) [2 , 9 , 10] has been shown to be , under some conditions , an optimal model for 2AFC decision-making [7 , 11] . The discrete analogue of DDM consists of a random walk on an 1-D interval , with one extreme as origin and the other as the decision boundary . Several variants of this model have been proposed depending on how the threshold is set , whether the integrated signal decays in time , and whether the two choices are represented by two competing and possibly interacting signals . More recently , the Ising Decision Maker has been presented as a new formal model for 2AFC , showing increased performance compared to Ratcliff diffusion model [12] . Over many years , this research program has shown with exquisite detail how humans and other animals reach decisions with a small number of options . However , most real-life decisions are made of a large and often virtually infinite number of choices , relying on heuristics and a relatively shallow search process in decision trees with complex geometries . Beyond some simple scenarios where classic diffusion models can be extended to more dimensions , this class of models can hardly adequately describe decision-making in multiple choices [13 , 14] . For instance , Usher and McClelland introduce the leaky , competing accumulator model extending this framework to multiple-decision tasks [5] . This model proposes several leaky integrators of signals—based on Ornstein-Uhlenbeck equations—which compete and inhibit each other until a decision is made by reaching a threshold . This model has a greater number of parameters ( increasing the degrees of freedom ) and show similar patterns of RT curves as DDM , which lead to ( small ) systematic gains ( in terms of quadratic error in fitting ) compared to DDM results . Lee and colleagues have successfully used DDM in sequential 2AFC tasks where context conditions are changed during time , and showed how diffusion models with non-homogeneous information focus on search for evidence and explain adaptation on search termination [15] . Another example is the Multi-modal Processing Tree ( MPT ) model class , a framework for developing and testing quantitative theories based on observations in categorical data [16] . It provides a data-analysis tool capable of disentangling and measuring separate contributions of different cognitive processes , measuring latent processes that are confounded in observable data . However , as MPT models are explanatory they require a detailed description of cognitive processes behind the behaviour under study , which may be impossible to propose when investigating high-level phenomena . In the artificial intelligence literature , techniques from operations research have been used to attack problems of choosing actions by planning agents in partially observable stochastic domains [17–19] . These methods build a policy tree—where actions are selected to optimize reward—and resemble the Monte-Carlo search approach used to model the dynamics of decision-making ( see Fig 1b ) . The objective of the present work can be viewed as an effort to bring together diffusion models in binary decisions and planning agents in complex problem solving , two very influential but largely disconnected literatures of decision-making . We theorize that the defining characteristic of a generic decision process—in humans and computer algorithms—is the presence of entropic barriers [20] , i . e . paths that are diffusively explored and usually lead to dead ends or sub-optimal solutions . We investigate here whether the distribution of reaction times ( RT ) in multiple-alternative decision-making can be modeled by a diffusion process on a space with topological traps , as we intend to capture the essence of a stochastic search process involving the exploration of “dead ends” and the concomitant back-tracking . In other words , we hypothesize that a random walk in a 2-D grid with obstacles may represent the decision-making process as trajectories of tree search algorithms ( see Fig 1b ) . Each position in the grid represents different board states , radius of bound reflects search depth , obstacle density represents amount of pruning and time per step represents processing speed ( see Fig 1c ) . We emphasize that while the model aims to integrate in its simplest form the notion of diffusion to a boundary with the idea of exploring branches in a tree , it is not intended to be an exact correspondence , as the topology of all the decisions problems in the search tree cannot be embedded in a one to one fashion in the 2 dimensional grid . A traditional limitation of behavioral and cognitive modeling has been the mismatch between the complexity of the models and the availability of experimental data to validate them . The advent of big data , however , has turned this difficulty on its head [21] . To examine our hypothesis , therefore , we capitalized on a vast corpus of decision-making obtained from on-line chess servers [22] ( see Fig 1a ) . This database has several virtues: ( 1 ) in any given move , players have to opt among a large number of options [23] , ( 2 ) options grow exponentially ( three steps down on decision tree typically results in more than a billion alternatives ) and hence the search process becomes rapidly intractable without heuristics , ( 3 ) as in real-life , in chess choices have to be made with a finite time budget ( 4 ) the quality of the decision-maker is particularly easy to calibrate in chess ( players have an ELO which indicates the quality of their decisions ) , and ( 5 ) it contains detailed information about more than 1 billion decisions , a volume that would be unthinkable to reproduce in a classical laboratory setting . We show that this simple theoretical construct , a natural extension of diffusion processes with the inclusion of obstacles ( or entropic barriers ) provides a remarkably accurate description of the data that was not captured with previous models . We also show that this obstacles model may characterize individual RT distributions and predict similarity with other players based on their time-to-move distributions . Moreover , we test the model in a complete different scenario of an online buying website where , suggesting that this model may describe a broad class of decision-making processes ranging from simple binary decision-making to complex decision in problem solving with heuristics not only in chess playing .
The Drift Diffusion Model ( DDM ) is the continuous analogue of a random walk with a direction bias , and is considered the optimal model for simple two-choice decision processes [11] . The model assumes that decision is made by the integration of evidence in a noisy process over time . Our implementation consists of a particle in a 2-D mesh walking to a threshold . Steps may be in any of four directions ( up , down , left , right ) equally distributed . Bias is inserted by accepting the step following the conditions: Z → = ( x - x 0 , y - y 0 ) Z p → = ( x p - x , y p - y ) C = 1 - Z → | Z → | · Z p → | Z p → | 2 p a c c e p t a n c e = e - C / ϵ where Z → is the current position , Z p → is the intended next position , C is the cost of putative move , p the probability of accepting the move and ϵ the noise in the acceptance decision . Thus , the parameters of the simulation are the radius of the boundary ( in units of the grid step ) R , the decision noise ϵ and the time step Δt . Several alternatives to DDM model appear in the literature , both for 2AFC and multiple-choices extensions ( e . g . [5 , 12] ) . These models show similar RT patterns and increase the number of free parameters , which permit small but systematic gains in fitting RT distributions . However , in our complex chess scenario , patterns of RT distribution differ significantly from these classic models . Thus , for sake of clarity we will compare our obstacles model to DDM . A one-dimensional random walk [24] , representing a blind decision-making process , initiates its exploration at a fixed position . The walker ends its exploration when it reaches a threshold value , after which the process may be re-initiated . The time to reach this threshold is the first-passage time ( FPT ) , and characterizes stochastic models of RT . In continuous time and space , the dynamics are expressed as: x ˙ ( t ) = η ( t ) ⟨ η ( t ) ⟩ = 0 ⟨ η ( t ) η ( t + τ ) ⟩ = δ ( τ ) The transition probability P ( x|x′ , t ) of finding the walker in x at time t given that it was in x′ at time 0 satisfies the Fokker-Planck equation ∂ t P + ∂ x J = 0 J = - ∂ x P ( 1 ) on which the initial and boundary conditions are easily expressed: assuming the origin of the walker at x0 , the initial condition implies P ( x|x0 , 0 ) = δ ( x − x0 ) , a crossing threshold at xT is equivalent to an absorbing boundary , P ( xT|x0 , t ) = 0 , whereas a reflecting boundary at xR corresponds to J ( xR|x0 , t ) = 0 . The probability distribution for the FPT is W ( t ) = - ∫ x 1 x 2 ∂ t P ( x T | x 0 , t ) d x ( 2 ) where x1 and x2 are the boundaries ( one is the absorbing , the other may be absorbing , reflecting or located at ∞ ) , and x1 < x0 < x2 . The analytic solution to this problem is known , and the limiting cases of reflecting boundary at finite and at infinite distance result in exponential and power-law tail distributions , respectively . The extension to higher-dimensional Euclidean geometries of the analytic results demonstrate a similar behavior for the tail distributions [25] ( see Fig 1d and Suppl . Mat . ) . These kind of models have been successfully used to model 2AFC and other tasks . However , we show that these models fail to represent the full distribution in more complex scenarios such as RT in chess and online buying . We propose that including entropic barriers in a 2D space would add the neccesary complexity to the model so that it would replicate the full distribution . Unfortunatley , at present there are not analytic solutions of the FPT problem for this type of models . Thus , we simulate the model by a random walk , discretizing the 2D space into a mesh where some of the nodes are considered reflecting entropic barriers . In the obstacles model , we use a square grid with a circular absorbing boundary , and reflecting nodes scattered randomly with a given density . We simulate trajectories in the 2-dimensional plane , since this corresponds to the minimum number of dimensions in which obstacles do not necessarily disconnect space . The process moves randomly and without bias in one of the four possible grid directions at each step . The parameters of the simulation are the radius of the boundary ( in units of the grid step ) R , the smoothness of the space ρ ( i . e . the number of obstacles in the grid ) , and the time step Δt . The update rule is then for position x → = ( x 1 , x 2 ) at time t is: x → ( t + Δ t ) = x → ( t ) + η → ( 3 ) where the random drift η → = ( η 1 , η 2 ) is defined formally by ηi = ±1 , 〈ηi〉 = 0 , 〈ηiηj〉 = δij . The smoothness is related to the probability that a grid point is an obstacle , p ( o , x → ) = 1 - ρ , 0 ≤ ρ ≤ 1 , so that ρ = 1 is a pure Euclidean space . The obstacles are defined by the constraint on the probability current , J → ( o ) = 0 . Fig 1 shows a walker that starts in the center and wanders through the labyrinth with obstacles until it finds a passage point . Simulations of models were implemented in Python . Each execution of a model with fixed parameters returns a number of steps taken to reach the threshold which represents a response time of a single decision . Parameters fitting of both models ( DDM and obstacles model ) was performed by exhaustive search in discrete parameters ( distance ) and Nead-Melder method [26] for non-linear optimization over the continuously parameter space ( step size and number of obstacles ) . For each parameter combination , we executed 75 , 000 simulations , i . e . 75 , 000 simulated decisions . In the case of the DDM , the parameter space range was: R ∈ [1 , 10] in grid-steps units , ϵ ∈ [0 , 0 . 25] and Δt ∈ [10 , 150] in milliseconds . On the other hand , the obstacles-model parameter space range was: R ∈ [1 , 10] in grid-steps units , ρ ∈ [0 . 2 , 0 . 65] and Δt ∈ [10 , 150] in milliseconds . To fit the human RT distribution , for each parameter combination we calculated the Jensen-Shannon divergence ( JSD ) [27] between the human ( hRT ) and the model ( mRT ) distributions . Thus , the optimization method consists on the parameters lookup in the simulated-decisions distributions which minimizes: ( R , ρ , Δ T ) = arg min ( R , ρ , Δ T ) 1 2 D ( P | | M ) + 1 2 D ( Q | | M ) where M = 1/2 ( P + Q ) , D ( P||M ) the Kullback-Leibler divergence [28] , P and Q the distributions to compare . To compare the relative goodness of fit between our model and DDM , we also use JSD instead of the usual AIC , following [29] .
We fitted parameters for both models ( DDM and obstacles model ) for the RT distribution obtained at each instant of the game . That is , given a remaining time , we estimated the best parameters which make the model fit more accurately the RT distribution at that instant of the game . The remaining time was divided into 0 . 1 seconds bins , i . e . 3000 bins in 300 seconds of total time available per game . Once this parameter fitting was performed for each instant of the game , we calculated the estimated median of the RT distribution ( Fig 2b ) . At every instant of the game , both models—with obstacles ( blue line ) and classic DDM ( red line ) —show an extremely precise median ( KS Test , no significant difference between distributions at every instant ) . However , if we compare the goodness-of-fit of the complete distribution we find that obstacles model fits significantly better than classic DDM . As an example of this , we show the parameter fitting of RT distribution at 50% of game using a single DDM model with only 3 free parameters: R the threshold distance , ϵ noise in the acceptance decision , and ΔT the time per step ( see Methods section for fitting details ) . Standard diffusion models ( in particular , DDM without obstacles ) produce distributions of FPT with exponential decay or power-law regardless of the dimensionality ( [32] , see also Suppl . Mat . ) . As expected , this fit is inconsistent with our observations of RT distributions which display a similar initial distribution but a super-exponential tail . We observe the best fit for RT distribution at the middle of the game using a single DDM ( ϵ = 0 . 2 , R = 8 and ΔT = 60ms , see Fig 2c , red line ) which is numerically very distant to the data ( Kolmogorov-Smirnoff statistics , Dn = 0 . 754 , p < 10−6 ) and , moreover , shows a qualitative different behavior with a tail that decays more rapidly . Instead , fitting with the 2-D diffusion model with obstacles ( see Fig 2c , blue line ) results in a very accurate description of the data ( Kolmogorov-Smirnoff statistics Dn = 0 . 014 , p > 0 . 95 ) . For this representative fit , ρ = 0 . 45 ( 45% of grid points are obstacles ) , with a radius of R = 4 and ΔT = 90ms . We repeated this procedure for each instant of the game and calculated the goodness of fit of both models , estimated by the Jensen-Shannon divergence . We observe that model with obstacles outperforms classic DDM ( Fig 2d ) . Model without obstacles shows values of JS divergence much higher than obstacles models in every instant of the game . The model is reminiscent of “comb” geometries , consisting of a main backbone with the origin and threshold in each extreme , and side branches where the diffusing particle may get trapped . This configuration is more restrictive in the topology of search process , but a comparative advantage is that analytic solutions for this problem have been developed . However , the best comb solution does not fully capture the observed RT distributions ( see Supp . Mat . for details ) . The results above show that diffusion with obstacles model provides an accurate description of RT distributions obtained from a vast corpus of human decisions . However these distributions aggregated decisions from many different players . Hence a possible and alternative explanation is that the non-exponential tails we observed in the data resulted from an addition of different exponentials ( assuming that players may have different characteristic decision times ) . To test this hypothesis , we selected individual players with more than 20 , 000 games each ( 17 players in total ) and performed the model fitting considering entropic barriers for each individual player . The model could fit very accurately the distribution of RTs of each individual player , reflected in small KS-statistics and p-values close to 1 , which indicate that the model and the data are not distinguishable ( KS test , p > 0 . 99 , Dstatistic < 10−8 in all cases; see fits in Supp . Mat . ) . This shows that the inability to fit the distribution of all decisions was not a matter of aggregating different individuals: instead the model with entropic barriers provides a very accurate fit of the RT distribution of individual players . To compare models , we implemented a cross-validation scheme . For each player with high number of games , we selected the RT distribution at 50% of the game and splited into 80%-20% sets for training both models and testing the fits , respectively . Using the first set , we fitted both models parameters and evaluated in the test set , obtaining 2 JSD measures , one for each model . We repeated this procedure 1 , 000 times and calculated the average JSD values for each model . Obstacles model showed better performance than DDM model in average JSD for all players ( see Supp . materials for details ) . Then , we evaluated the prediction capabilities of the model . We proposed that players who show similar RT distributions ( in terms of JSD values ) , should obtain similar JSD values for the fitted distributions of the model . Thus , given both RT distributions of a particular player ( the real data , and the fitted model ) , sorting other players based on similarity of their RT distribution to the real and fitted ones must be correlated . We fitted each player individually and calculated JSD values of real data and the fitted model to every other player . We found that correlation between sortings is significantly better using the Obstacles model ( Spearman rank-correlation r > 0 . 99 , p < 10−10 in all cases ) than DDM ( Spearman rank-correlation r ≈ 0 . 6 , p < 10−2 ) . Our next aim is to test the Obstacles model , examining specificity , i . e . whether it can be used to obtain useful discriminations and generality , i . e . whether the model is valid across different contexts . To test specificity we investigated whether the Obstacles model can identify meaningful differences in the decision process between strong and weak players . A comparative advantage of studying decision-making in chess is that the quality ( i . e . rating ) of players can be precisely determined , representing the player’s strength in a very confident manner . The rating of a player is determined based on their past results with other players , and updated after each game played . Players receive rating points in proportion to the difference in their strength . A strong player would increase a very small amount of rating points when winning a game versus a weak player . In contrast , a weak player would win rating points even in the case of a draw playing to a stronger opponent [31] . Decision-time distributions of strong and weak players are comparable , although stronger players make slightly slower moves during middle-game and faster during the opening and the end-game [22] . We reasoned that this subtle difference in RT distributions between strong and weak players may rely on different decision processes . Specifically , in line with notions of expertise [33] we hypothesized that strong players ( expert decision-makers ) discard a larger number of states which through heuristic which in our model accounts to having a larger number of obstacles ( i . e ρstrongplayers > ρweakplayers ) [34] which is compensated by a faster navigation of the decision-tree ( i . e ΔTstrongplayers < ΔTweakplayers ) . As for the radius ( the equivalent to the depth of search ) , different chess theories differ on whether search depth increases with quality or remains constant . To examine these hypotheses we splited the population into quintiles groups ( i . e . each group has the same number of played games ) and analyze distinctively the parameters of the model . We performed the parameter fit to data corresponding to each one of the five groups . The correlation between each parameter ( obstacles rate ( ρ ) , time per step ( ΔT ) and threshold distance ( R ) ) and player ratings are presented in Fig 3 . With higher ratings , players show higher obstacles ratio ( Fig 3a , Pearson correlation r = 0 . 89 , p < 0 , 04 ) and shorter distances ( Fig 3c , Pearson correlation r = −0 . 97 , p < 0 , 006 ) . Time per step showed no correlation with player ratings . Obstacle density shows higher values for strongest players compared with weakest players ( KS Test , p < 10−17 ) , showing more complex grids to transit during the random walk . On the other hand , time per step shows similar patterns between both classes of players ( KS Test , p = 0 . 20 ) , suggesting that this parameter may be representing a canonical aspect of decision making which does not differ among player classes . As strong players show more complex scenarios with similar time per step , distances to threshold show smaller values in strong players than in weak players ( KS Test , p < 10−7 ) . These results suggest that all players spend the same time for checking each step , but strong players explore more difficult decision spaces ( more obstacles ) , with shorter distances to threshold . To test how the model generalizes , first we performed independent fits at each instant of the game , which correspond to different fractions of time available , and exhibit different RT distributions . Every distribution was fitted with the proposed model , obtaining R , ρ and ΔT parameters , and compared by their median value and a two-sample Kolmogorov-Smirnoff test of the complete distribution . In Fig 2b , we show the median value obtained both from players and model data at each instant of the game . We observe an almost perfect match to the proposed model with obstacles . To quantify this observation , at each instant of game , we verified the goodness of fit comparing the distribution of human ( black line ) and simulated data ( red line: no obstacles model; blue line: obstacle model ) by analyzing JS divergence between distributions ( see Fig 2d ) . Results show that all RT distributions are indistinguishable from the best fit of model with obstacles ( Fig 2d , blue dots ) ; a KS test between obstacles model and human data is non-significant in every instant of the game . In contrast , when testing the DDM versus human data with the KS test , the null hypothesis is rejected ( p < 10−3 ) indicating that the basic model without obstacles cannot reproduce human RT behavior . In accordance to this result , JS divergence show much higher values than obstacles model ( Fig 2d ) . We conclude that our obstacle model is universal to the decision-making process when playing on-line chess , regardless the instant of the game . We further investigated whether this model extends to different cognitive domains . For this , we analyzed navigation logs of users in MercadoLibre , the biggest on-line buying website of Latin-America . Logs consist of response times of users after doing a product search , i . e . how long users take to decide for ( more information of ) a product . Fig 4 shows the RT in this website; again , with a super-exponential tail . The obstacle model is plotted in blue line showing an almost perfect match ( KS test , Dn = 0 . 018 ) . This result suggests that the model may transfer to very different domains ( from chess to on-line buying ) , and that it is reflecting a intrinsic decision-making process .
Different models of accumulation to a boundary have been the hallmark of two-choice decision-making [1 , 35 , 36] . A few studies have investigated experimentally ( e . g [37 , 38] ) how they extend to more than two alternatives . For instance , Churchland and colleagues studied a four-choice paradigm [37] . Their results show that an urgency signal was more prominent on the two-choice paradigm than four-choice , consistent with longer response times on the latter experiment . Theoretical investigations have generalized competing integrators of higher number of alternatives [37 , 39 , 40] . Others have modeled by a competition between neural pools [5 , 13] . These studies have succeeded in accounting for a range of behavioral data in conscious multi-choise tasks [41] . In chess , as in many other domains of human problem solving , it is well known that participants do not exhaustively search all alternatives [42] . Actually , studies show that board evaluation is rather small , and only a few moves simulations are evaluated by players [43] . Chess players search in a decision tree , but halting this process by evaluations or heuristics which dictate that a given state is not desirable [44] . How decisions emerge without an exhaustive exploration of move alternatives is still an open question . Current chess models suggest that cognitive architecture should concentrate on relevant pieces or positions in the board , and may search for the best move by analogy with previous studied/plausible positions [45] . Classic DDM models are unable to capture the RT distributions produced by chess plays . Our model is designed to capture this process by presenting entropic barriers which can be seen as stop points in a search path , i . e . a moment in which there is sufficient visibility to discard the path based on an evaluation function . However , this interpretation must not be overstated . In a 2D grid the number of branches at each state is bound to 4 ( all possible directions ) . This limit if obviously not true in chess or other multiple-choice tasks , but resembles the idea of having multiple trajectories available at each iteration in the decision process . Using chess as a model of expertise in decision-making , Gobet and colleagues have demonstrated that stronger players do not necessarily outperform the maximum search depth of decision trees than weaker ones [46] . Instead , they cover deeper searches in average , showing a more exhaustive coverage of branches and a very efficient cutting threshold . In agreement with this view , we show that fitting RT distributions with our three-parameter entropic model achieves a remarkable precision . Our model refines this idea indicating that better players explore solutions in a more punctuated space . We interpret the model’s entropic barriers as stop points in a search process . The fact that stronger players show an increased number of obstacles indicates a more efficient cut algorithm for discarding sub-optimal branches . Expert players compensate the increase in time due to a navigation in a more punctuated space by a smaller distance to the threshold . The model of decision-making we present here departs from traditional approaches by explicitly incorporating the presence of such entropic barriers in a stochastic search process . Using rapid chess big data as an unprecedented , high-throughput experimental laboratory , we show that our model provides a remarkable fit to the response time statistics ( i . e . the distribution of times-to-move ) at different stages of the game , not only first and second order moments but for the entire probability distribution over several response-time orders of magnitude . While at present we do not have the tools to investigate the potential mapping of the formal solutions to mental processes , it is expected that traces of them , in particular the presence of entropic barriers , should be found in the evidence that is readily available in psychophysical measurements .
|
Decision-making has been studied in great detail relying on binary choices , modeled as the noisy accumulation of a decision variable to a threshold . We show that it breaks down when used to describe real-life human decision involving multiple options . We show instead that including obstacles in the diffusion model ( a natural conceptual extension ) can describe the data with great degree of accuracy . We evaluate this new model by capitalizing on the advent of big data , analyzing a vast corpus of decision making obtained from on-line chess servers . The present manuscript resolves a conflict between current theories of decision-making and concrete data , it solves this data with a concrete theoretical proposal and analyzes specific predictions of the model .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"recreation",
"decision",
"theory",
"decision",
"making",
"engineering",
"and",
"technology",
"applied",
"mathematics",
"social",
"sciences",
"mathematical",
"models",
"neuroscience",
"problem",
"solving",
"decision",
"analysis",
"cognitive",
"psychology",
"mathematics",
"statistics",
"(mathematics)",
"management",
"engineering",
"cognition",
"research",
"and",
"analysis",
"methods",
"random",
"walk",
"decision",
"trees",
"games",
"behavior",
"mathematical",
"and",
"statistical",
"techniques",
"psychology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cognitive",
"science"
] |
2018
|
An entropic barriers diffusion theory of decision-making in multiple alternative tasks
|
Kinetic models provide the means to understand and predict the dynamic behaviour of enzymes upon different perturbations . Despite their obvious advantages , classical parameterizations require large amounts of data to fit their parameters . Particularly , enzymes displaying complex reaction and regulatory ( allosteric ) mechanisms require a great number of parameters and are therefore often represented by approximate formulae , thereby facilitating the fitting but ignoring many real kinetic behaviours . Here , we show that full exploration of the plausible kinetic space for any enzyme can be achieved using sampling strategies provided a thermodynamically feasible parameterization is used . To this end , we developed a General Reaction Assembly and Sampling Platform ( GRASP ) capable of consistently parameterizing and sampling accurate kinetic models using minimal reference data . The former integrates the generalized MWC model and the elementary reaction formalism . By formulating the appropriate thermodynamic constraints , our framework enables parameterization of any oligomeric enzyme kinetics without sacrificing complexity or using simplifying assumptions . This thermodynamically safe parameterization relies on the definition of a reference state upon which feasible parameter sets can be efficiently sampled . Uniform sampling of the kinetics space enabled dissecting enzyme catalysis and revealing the impact of thermodynamics on reaction kinetics . Our analysis distinguished three reaction elasticity regions for common biochemical reactions: a steep linear region ( 0> ΔGr >-2 kJ/mol ) , a transition region ( -2> ΔGr >-20 kJ/mol ) and a constant elasticity region ( ΔGr <-20 kJ/mol ) . We also applied this framework to model more complex kinetic behaviours such as the monomeric cooperativity of the mammalian glucokinase and the ultrasensitive response of the phosphoenolpyruvate carboxylase of Escherichia coli . In both cases , our approach described appropriately not only the kinetic behaviour of these enzymes , but it also provided insights about the particular features underpinning the observed kinetics . Overall , this framework will enable systematic parameterization and sampling of enzymatic reactions .
Since the seminal work of Michaelis and Menten [1] , enzyme kinetics theory has been developed for most catalytic mechanisms , captured as a series of elementary reactions representing events at molecular level , i . e . binding and release of reactants from enzyme intermediates and catalysis . The particular catalytic pattern of the enzyme dictates its mathematical representation , which can be obtained by solving equations for the enzyme intermediate concentrations [2] . A quasi-steady-state assumption for these intermediates is commonly employed to this end [3] , yielding a final expression function of microscopic rate constants and reactant concentrations . The whole process can be automated using King-Altman’s method [4] . Some of the rate constants are related to the apparent equilibrium constant of the overall reaction by the Haldane relationships , directly linking kinetics with thermodynamics [5] . Moreover the rate constants can be converted into macroscopic kinetic constants [6] , which can be measured and estimated from subsequent enzymatic assays . The classical approach including various graphical representations works well for studying regular enzymes up to a couple of substrates and a couple of products . With more complex reaction mechanisms , the number of parameters becomes so large that a combination of model reduction and/or computational sampling is required for study . Models can be reduced by introducing simplifying assumptions , e . g . , steps being at equilibrium , lumped elementary steps , etc . Multiple expressions have been proposed to reasonably approximate kinetic behaviour using available data while maintaining essential thermodynamic consistency [7–12]; however they always incur some loss of generality . Computational sampling enables us to determine emergent enzyme properties from high dimensional parameter spaces . Several sampling strategies have been proposed , for example uniform parameter sampling within the S-formalism [13 , 14] , elasticity sampling [15 , 16] and relative enzyme saturation sampling [17 , 18] . All use simplified kinetic expressions ( loss of generality ) and most ignore intrinsic thermodynamic constraints between kinetic parameters , hence they will sample infeasible parameter sets . Current parameterization and sampling approaches also fail to accurately capture allosteric regulation . Even though mass inhibition/activation can account for some degree of regulation , metabolic control is mostly achieved through allosteric and transcriptional regulatory interactions [19] . Modelling allosteric behaviour requires the inclusion of conformational information , which enables description of both allosteric and cooperative interactions . The two most famous of such models are the symmetry model of Monod , Wyman and Changeux ( MWC ) [20] and the sequential model of Koshland , Némethy and Filmer ( KNF ) [21] . Both models are based on the assumed equilibrium between two conformational states of the enzyme; a relaxed ( R ) and tense ( T ) state . The main difference between these theories rests in how conformational transitions proceed upon binding of ligands . While the MWC model demands maintenance of conformational symmetry , the KNF model does not and instead requires strict induced fit . Although these models can be considered as special cases of more general models , these generalizations have so far not proven to be useful . In fact , the MWC model is regarded the model of choice for describing allosteric and cooperative interactions [22] . Moreover , it has been demonstrated that the MWC model can be cast in a convenient mathematical form [23] , which can be combined with the elementary reaction formalism . In the current work , we present a General Reaction Assembly and Sampling Platform ( GRASP ) capable of parameterizing and sampling the kinetics of any oligomeric enzyme by using minimal reference and biochemical mechanistic data . Parameterization combines the generalized MWC model for modelling the kinetics of oligomeric enzymes with the elementary reaction formalism for deriving thermodynamically consistent catalytic expressions . By employing a convenient normalization at the elementary reaction level [24] , we are able to describe reaction kinetics provided a reference flux and the thermodynamic affinity of the reaction at the reference point . Using an accurate parameterization necessarily requires a large number of parameters . An advantage is that the parameterization retains all intrinsic thermodynamic constraints between kinetic parameters . We designed an efficient sampler producing thermodynamically consistent parameters obeying the principle of microscopic reversibility . Using an efficient Monte Carlo sampling technique that exploits the shape of the sampling space , we ensure high parameter quality and low parameter rejection rates . The framework is demonstrated through exploration of the full kinetic space of reactions , assessing the impact of thermodynamic affinity , reaction molecularity and mechanisms , as well as modelling complex kinetic behaviours .
In this article , we shall describe a general framework that enables parameterization and sampling of kinetic parameters consistent with thermodynamic constraints . The framework employs the MWC model , which provides the basis for modelling most cooperative and allosteric behaviours of multimeric enzymes [20] . The classical formulation including detailed assumptions and limitations are described in S1 Text . Our framework is based on the recasting of the model developed by Popova and Sel’kov [23 , 25–27] , in which the velocity of reaction of any oligomeric enzyme is expressed as the product of two independent functions , v=Φcatalytic⋅Ψregulatory ( 1 ) where Φcatalytic represents a rate law function for the protomers in the so-called relaxed ( R ) conformation , and Ψregulatory denotes a regulatory function describing the conformational mechanism of transition from a so-called tense ( T ) to the relaxed conformation . Equation 1 provides a general and simple interpretation of the kinetics of an oligomeric enzyme . Firstly , the shape of the catalytic function is determined only by the mechanism of elementary interactions between substrates and products with one active site of the enzyme ( catalytic mechanism ) . Secondly , the regulatory function is invariant with respect to the action mechanism of the catalytic sites . Thus , if one has information about the conformational mechanism of the enzyme , i . e . number of subunits , transitions , allosteric sites and effectors , and possesses an expression representing the catalytic mechanism for the conversion of substrates to products , the catalytic and regulatory functions can be written as follows [23] , Φcatalytic=n⋅vR ( 2 ) Ψregulatory=1+ ( vT/vR ) Q1+Q ( 3 ) where vR and vT represent the velocity of reaction for the R and T conformations of the oligomeric enzyme , respectively ( both states follow the same reaction mechanism as protomers are identical ) , and Q is a function that determines the current ratio between the R and T states ( see later ) . The generalized MWC model enables parameterization of the kinetics of any oligomeric enzyme by decomposing the reaction velocity into two independent functions . An important feature of this parameterization is that it enables inclusion of fundamental thermodynamic relations between kinetic parameters , as it is compatible with the elementary reaction formalism . Some of these relations are lost when using other parameterizations . A complete overview of the proposed framework is depicted in Fig . 1A . In the following , we present a systematic method for parameterizing and sampling thermodynamically consistent kinetics . Before considering complex cooperative or allosteric mechanisms , we will consider a simple non-allosteric enzyme , i . e . n = 1 and L = 0 ( no tense state ) or L = ∞ ( no relaxed state ) , the resulting flux at any reference state is purely due to the catalytic function . Every enzymatic reaction can be broken into simple reversible steps called elementary reactions . Using the law of mass action , the rate of each elementary reaction is written as [2] , velem=k⋅x⋅e for binding stepsvelem=k⋅e for dissociation steps ( 5 ) where k is a microscopic rate constant , x is the reactant concentration and e is the concentration of enzyme intermediate involved in the elementary step . Typically , the absolute values of the metabolite and total enzyme concentrations are not known ( although physiological ranges can be estimated ) . To overcome this limitation , normalization of all the variables around a reference point ( steady-state flux ) is a convenient strategy . Following the scaling procedure employed by Tran et al . [24] , normalization of these variables yields , velem= ( k⋅etotalref⋅xref ) ⋅ ( xxref ) ⋅ ( eetotalref ) =k˜⋅x˜⋅e˜ for binding stepsvelem= ( k⋅etotalref ) ⋅ ( eetotalref ) =k˜⋅e˜ for dissociation steps ( 6 ) The normalized metabolite concentrations are unitary at the reference point ( x˜ref=1 ) , which is used extensively in developing an efficient sampling strategy . Rate laws can be derived from the enzyme mechanism and the microscopic rate constants using the King-Altman’s method [4] . In order to sample kinetics , we can in principle sample the rate constants directly . However , this leads to an inefficient sampler , where thermodynamic constraints can only be validated after sampling resulting in a high rejection rate . Instead , we design the sampling procedure to incorporate constraints directly without the need of assuming any particular distribution for the rate constants . Returning to the generalised MWC model , Equations 1–3 , we note that the catalytic and regulatory contributions are confounded in the case of allosteric enzymes . Even if the reaction flux is known in a particular state , the particular values of both functions are unknown . To resolve this , we need to elucidate the contributions of the relaxed and tense conformations . The regulatory function is always less than or equal to 1 , as the catalytic activity of the T state is less than the R state [32] , and we introduce the activity ratio ( aref ) of a tense protomer at the reference state and sample this uniformly . Given aref we can rearrange Equations 2 and 3 to calculate the contributions of each state vRref=Φcatalyticrefn=vrefnΨregulatoryrefvTref=aref⋅vRref ( 22 ) The catalytic mechanism is assumed identical for the R and T state and we sample this mechanism twice using the two different reference points to generate feasible parameterisations for the R and T state . The final step is to generate a sample of parameters for the Q function that satisfy Equation 3 in the reference point . The Q function can be expressed as [33] , Q=L0⋅ ( e˜R0 ( x , kR ) e˜T0 ( x , kT ) ⋅∏i=1m∑j=1r ( 1+xF , i , j/KT , i , j ) ∏i=1m∑j=1t ( 1+xF , i , j/KR , i , j ) ) n ( 23 ) where L0 is the allosteric constant between the R and T states in the absence of ligands , xF , i , j represent effector concentrations binding to specific allosteric sites , KR , i , j and KT , i , j denote the effectors dissociation constants for each state , e˜R0 and e˜T0 are the free enzyme fractions in both conformational states as function of the respective rate parameters ( kR , kT ) and reactant concentrations ( x ) , m represents the number of allosteric sites , and finally , r and t are the number of positive and negative effectors binding to the allosteric sites in the R and T states , respectively . Here we have assumed that the allosteric activators and inhibitors only bind to the R and T , respectively [32] , although this constraint can be relaxed . In the reference point , Q does not depend on the reactant and effector concentrations as they are defined as unitary at the reference point . We can determine all parameters in the Q function based on sampled enzyme abundances ( refer to Equation 8 ) . In the case of the conformational transitions , the change of the Gibbs free energy of conformations between the R and T states is constrained by the ligand affinity of the two states . These transitions point to a higher relative abundance of the free enzyme in the T state in the absence of substrate , whereas in the presence of substrate the R state is more favoured [32] . Ultimately , the latter yields L>1 and a ratio of affinity constants for both states KRaff/KTaff>1 or equivalently KT/KR > 1 in the case of dissociation constants ( Fig . 1D ) . Colosimo et al . [34] have derived a simple expression to determine the allosteric constant in the absence of ligands assuming symmetric binding for the two states ( Equation 24 ) . L0= ( KTKR ) n/2 ( 24 ) To estimate the dissociation constants in Equation 24 , we can make use of the equilibrium assumption between the R and T states and the unitary ligand concentration normalized at the reference state . L0= ( e˜T0ref/e˜T1refe˜R0ref/e˜R1ref ) n/2 ( 25 ) In Equation 25 , ( e˜0ref ) and ( e˜1ref ) denote the enzyme fractions free and bound to the ligand for the R and T states at the reference state . The same principle can be used to calculate the effectors dissociation constants . Calculation of each constant is achieved using the following general formula for the allosteric site m , and effector r binding to the R state and effector t binding to the T state . K˜R , m , r=eR0ref/etotal⋅xF , m , rrefeR1ref/etotal=e˜R0refe˜R1ref⋅xF , m , rrefK˜T , m , t=eT0ref/etotal⋅xF , m , trefeT1ref/etotal=e˜T0refe˜T1ref⋅xF , m , tref ( 26 ) In particular , the absolute concentrations of the allosteric effectors ( xm , rref , xm , tref ) need not to be known , as they are scaling factors for the absolute effector concentrations in Equation 23 which are unitary at the reference state . The sampling procedure generates only feasible parameter sets . Since this parameterization is built upon a reference point , this can be validated by confirming that the parameter set produces the reference flux at the reference point , i . e . when the normalized metabolite concentrations are unitary . We control the numerical accuracy of parameter sets by accepting only sets achieving the reference flux within a tolerance , e . g . ε = 10-8 , |Φcatalyticref⋅Ψregulatoryref−vref|<ε ( 27 ) In general , rejected instances are insignificant and normally represent less than 0 . 01% of the sampled models . If the reference values for the total enzyme and metabolite concentrations are known , they can be used to transform the set of scaled rate constant into absolute constants . Microscopic rate constants are commonly difficult to measure; therefore standard macroscopic constants are preferred to parameterize rate expressions . Transformation of rate constants into macroscopic rate constants can be performed following Cleland’s rules [6] , which are consistent with the Haldane relationships . In this manner , estimated macroscopic constants can readily be compared with available experimental data . In order to determine the impact of reactant perturbations on the reaction rate , we estimated the reaction elasticities upon an infinitesimal variation in the concentration of substrates and products [2] . The partial derivatives were calculated using a central difference approximation , εx˜v=x˜v∂v∂x˜| ( vref , x˜ref , k˜ ) ≈x˜refvref⋅v ( x˜ref+Δh , k˜ ) −v ( x˜ref−Δh , k˜ ) 2Δh ( 28 ) where x˜ref represents the perturbed normalized reactant concentration , k˜ denotes the rate constants vector and Δh is the size of the perturbation . Given that the perturbations for reactant concentrations are performed in the vicinity of the reference state , i . e . x˜ref=1 , a uniform step size of 10-2 equivalent to a 1% change was employed for all calculations . The implementation and execution of this workflow was performed in MATLAB 2013a ( The MathWorks , Natick , MA ) . Definition of the reversibility matrix was achieved using appropriate MATLAB functions from the Bioinformatics toolbox . Automated derivation of rate laws was achieved using King-Altman’s method for finding valid reaction patterns [4] . Qi et al . [35] have recently reported an efficient algorithm ( KAPattern ) employing topological theory of linear graphs for accomplishing this goal . By employing this algorithm , we derived the enzyme intermediates abundance functions which we then assembled to build the final velocity rate . All computations were run on a Dell OptiPlex 990 Desktop ( Intel Core i5-2400 , 4 GB ram , Microsoft Windows 7 , x86-based architecture ) .
The connection between reaction thermodynamics and kinetics can be normally found in the form of the Haldane relationships [5] . Depending on the mechanism of reaction , these relations relate the values of the macroscopic kinetic constants , i . e . dissociation and catalytic constants , with the apparent equilibrium constant of the reaction . It would be interesting to derive similar relations under non-equilibrium steady-state conditions , provided that biological systems operate in this regime . It has been demonstrated that the analysis of the relation between the thermodynamic affinity ( represented by ΔGr/RT ) and the reaction velocity can still be performed as if it were at equilibrium , by displacing the reference point from the equilibrium to another appropriate reference state [36] . As such , relations analogous to the Haldane relationships can be derived at this reference point ( S1 Text ) . For example , for a Uni-Uni reaction converting a substrate A into a product P the following relation can be derived , exp ( ΔGrRT ) =k˜−1k˜−2k˜−3k˜1k˜2k˜3=k˜cat , -⋅K˜Ak˜cat , +⋅K˜P ( 29 ) where k˜i denote the rate constants at the reference state , K˜A and K˜P represent the normalized dissociation constants for A and P at the reference state , respectively , and k˜cat , + and k˜cat , - are the scaled catalytic constants for the forward and reverse reactions . It can be demonstrated that the following generalized equation holds true for any reaction mechanism , ΔGrRT=ln ( k˜cat , -k˜cat , + ) catalytic+∑is∑jpln ( K˜iK˜j ) binding ( 30 ) where s and p denote the number of substrates and products , respectively . We have designated the first term on the right-hand side of Equation 30 the catalytic ( turnover ) term , while the second is called the binding ( saturation ) term . Notably , the catalytic term is independent of the reactant concentrations at the reference state , as opposed to the saturation term . In fact , it can be shown that K˜A=KA/Aref , which can be regarded as the degree of saturation of the enzyme for reactant A . In this way , Equation 30 enables the energetic analysis of any reaction by decomposing it in two contributions: ( 1 ) catalysis ( how efficient is the enzyme converting substrates into products at the reference state ) and ( 2 ) binding ( how saturated is the enzyme at the reference state ) . More importantly , this relation imposes a natural trade-off in enzyme catalysis . Greater contributions of the catalytic term suggest a higher enzymatic efficiency in the conversion of substrates to products , but a suboptimal saturation of the enzyme , i . e . KA/Aref high . On the contrary , high saturation contributions suggest lower conversion efficiency of substrates into products , i . e . k˜cat , -/k˜cat , + is relatively higher compared to the binding term . To explore the consequences of this trade-off in enzymatic reactions , we uniformly sampled the kinetic space of three ordered mechanisms following different molecularities: Uni-Uni , Bi-Bi and Ter-Ter , under different Gibbs free energy differences and a constant reference flux . S1 Table shows the definition of the dissociation and catalytic constants in terms of the rate constants using Cleland’s definitions . We performed the analysis considering conditions close to equilibrium ( -1 kJ/mol ) to far from equilibrium ( -80 kJ/mol ) ( Fig . 2 ) . Notably , the average conversion and saturation contributions ratio remains constant for different ΔGr/RT for all the kinetics sampled . As expected , the higher the molecularity of the reaction , the higher the contribution of the binding term . The Uni-Uni mechanism exhibits a low average binding contribution ( 32% ) , which increases for the Bi-Bi mechanism ( 60% ) and further for the Ter-Ter mechanism ( 71% ) . These results suggest that catalysis in multi-substrate enzymes is a process strongly driven by the degree of saturation of the enzyme and to a lesser extent by the actual conversion of substrates into products . Uniform sampling of the kinetic space enables an unbiased appraisal of the relation between thermodynamics and kinetics . For the aforementioned cases , the sums of the blue and red areas represent the 95% confidence region of all the thermodynamically feasible parameter sets . Interestingly , for all the tested mechanism , the feasible area increases with higher driving force ( ΔGr/RT more negative ) . Such increased area point to a greater diversity of feasible parameter sets under more thermodynamically favourable conditions . On the contrary , more homogeneous parameter sets can be found closer to equilibrium , i . e . different parameter sets have similar energetic contributions . The latter suggests that sampled parameter sets might have similar kinetic behaviours closer to equilibrium . The next analysis will provide additional support for this assertion . In the previous section , we assessed the effect of the reaction molecularity on enzyme catalysis . We next analysed the impact of thermodynamics on enzyme kinetics . To that end , we uniformly sampled the kinetic space of reactions with same molecularity . Since the majority of enzymes found in nature catalyse bimolecular reactions [2] , we focussed our attention on the most representative bimolecular mechanisms . There are two general mechanisms for bimolecular reactions , the ping-pong mechanism in which a product is released before both substrates have reacted with the enzyme , and the ternary complex ( sequential ) mechanism in which the enzyme combines with both substrates before products are formed [6] . Sequential mechanisms can be further divided into random and ordered mechanisms . In random-order mechanisms , reactants can bind in either order , while in the ordered type sequential process one reactant always binds to a certain site before a second reactant binds to the other site [37] . The representation of all three mechanisms is shown in Fig . 3A . The impact of thermodynamics was evaluated in each case by calculating the reaction sensitivities for substrates and products , i . e . substrate and product elasticities , at different Gibbs free energy differences ranging from -1 ( close to equilibrium ) to -80 kJ/mol ( practically irreversible ) . The 95% confidence regions for the calculated reaction elasticities for the above mentioned reaction mechanisms are shown in Fig . 3B-C . The substrate and product elasticities strongly depend on the chosen thermodynamic reference state . Within the range of ΔGr analysed , two reaction elasticity regions can be distinguished: a variable elasticity region ( 0>ΔGr>-20 kJ/mol ) and a constant elasticity region ( ΔGr<-20 kJ/mol ) . The former region can be further subdivided in two: a linear regime with a steep slope ( 0>ΔGr >-2 kJ/mol ) and a transition regime ( -2>ΔGr>-20 kJ/mol ) . Notably , previous works have demonstrated an almost perfect linear correspondence between the thermodynamic affinity and the reaction flux close to equilibrium , i . e . 0>ΔGr > -1 . 5 kJ/mol [38 , 39] . Such relationship has been shown to hold for ΔGr>-7 kJ/mol with less than 15% error [40] . Our simulation results are thus in line with what would be expected to be the kinetic behaviour close to equilibrium . A remarkable feature of our sampling strategy is that it not only enabled analysis of the reaction elasticities behaviours close to equilibrium but also far from it . In the case of the reaction elasticities close to equilibrium , our sampling results show the substrate and product elasticities are respectively monotonically decreasing and increasing , reaching their respective maximum and minimum at equilibrium ( Fig . 3B ) . This behaviour is consistent with previous analyses on the behaviour of the reaction elasticities in this region [2] and supports our intuition: reactions operating close to equilibrium are more susceptible to slight changes in the reactant concentrations , exhibiting greater changes upon these perturbations . On the other hand , our analysis far from equilibrium revealed both substrate and product elasticities reach a plateau at highly negative thermodynamic affinities for all the reaction mechanisms analysed . In the case of the substrate elasticity , this plateau stabilizes close to zero for highly negative ΔGr . Notably , the average substrate elasticity for thermodynamically favoured reactions is approximately 0 . 22 and not 0 as it would be expected . The latter is a direct result of our sampling strategy , which seeks to uniformly sample all the possible parameter sets that are consistent with the Haldane relationships at a chosen reference point . On the contrary , in the case of the product elasticity , the average elasticity consistently approaches zero for all the mechanisms analysed under favoured conditions . Furthermore , our sampling results suggest that product inhibition is almost negligible for thermodynamically favoured reactions , i . e . ΔGr<-30 kJ/mol ( Fig . 3C ) . S1 Text provides an illustrative explanation of the asymptotic behaviour of the reaction elasticities near and far from equilibrium . Additionally , it is also important to highlight that the allowable elasticity regions becomes tighter as we move closer to equilibrium ( Fig . 3B-C ) . The latter supports our earlier findings of sampling more homogenous parameter sets close to equilibrium . Indeed , the closer to equilibrium the tighter the feasible parameter space becomes , and thus , a more similar response upon reactant perturbations is obtained . Finally , reaction mechanisms have similar elasticity behaviours across the analysed ΔGr range . However , the average response of ordered and random mechanisms is more similar than the response of the ping-pong mechanism . This is most readily appreciated when comparing the substrate elasticities ( Fig . 3B ) . As previously mentioned , the ping-pong mechanism executes a fundamentally different mechanism in which the release of the first product takes place before binding of the second substrate . The latter explains the slightly lower average substrate elasticity . Interestingly , negative substrate elasticities can be encountered in the random order mechanism for high thermodynamic affinities ( approx . ΔGr<-18 kJ/mol ) , i . e . the addition of substrate decreases the velocity of reaction ( elasticities below the red dashed line in Fig . 3B ) . This behaviour has been demonstrated previously for random-order mechanisms [41] . Theoretical analysis of these mechanisms has shown that they can give rise to apparent substrate inhibition or substrate activation [2] . For example , depending on the kinetic parameter values and the substrate concentrations , the reaction rate for the enzymes following this mechanism can display an apparent cooperative behaviour ( sigmoidal reaction rate ) upon addition of one substrate maintaining the other constant , while in the opposite situation they can exhibit substrate inhibition ( the reaction rate pass through a maximum ) [41] . Although this behaviour is not very common , there is evidence of such in the literature [42] . More importantly , our framework enabled unbiased sampling of all feasible kinetic behaviours , thereby revealing the impact of the thermodynamic affinity on the reaction rate . ATP-mediated phosphorylation of glucose represents the first step of the glycolysis . In mammals , this step is performed by four different isozymes ( EC 2 . 7 . 1 . 1 ) located in different tissues . Hexokinases I-III are mainly located in the brain , muscle and erythrocytes [43] , while hexokinase IV ( glucokinase ) is primarily located in the liver and pancreatic β-cells [44] . In the latter tissues , glucose phosphorylation is the rate-limiting step of glucose metabolism , and more importantly , it is ultimately responsible for the release of insulin into the bloodstream which maintains glucose homeostasis in the body . Unlike hexokinases I-III , glucokinase has unique kinetic features that enable its regulatory function . It displays a sigmoidal response upon increasing glucose concentration , i . e . positive cooperativity , but has a hyperbolic behaviour upon increasing MgATP2- , i . e . Michaelis-Menten kinetics [45] . This positive cooperativity for glucose is intriguing as the enzyme is monomeric , which contradicts standard cooperativity models and thus requires a conceptually different explanation . One of the simplest models capable of explaining this behaviour is the mnemonic model [46] . Briefly , this model proposes that the conformation of an enzyme following product release could be different from the initial enzyme state , i . e . the enzyme has memory . In the case of glucokinase , this model suggests a slow conformational transition from a low-affinity state ( E* ) to a high-affinity state ( E ) , which ultimately carries out both catalysis and product release ( Fig . 4 ) . This transition can be enhanced with increasing glucose concentrations , yielding the observed positive cooperative behaviour on this substrate ( kinetic cooperativity ) . Other possible explanations like a reaction mechanism with random addition of substrates has been discarded , as isotope exchange studies have demonstrated an ordered mechanism with glucose binding first [47] . One question that can be addressed using the current framework is how likely it is to encounter a particular kinetic behaviour . In this case , we are interested in estimating the frequency of positive cooperativity for glucose of the glucokinase . To this end , we uniformly sampled the kinetic space for this enzyme and counted the frequency of parameter sets displaying positive cooperativity for glucose ( estimated Hill coefficient , nH>1 ) . In order to mimic enzymatic assays conditions , i . e . high substrate concentrations and almost no products , a reference Gibbs free energy difference of -100 kJ/mol was used . Very similar results were found using down to -50 kJ/mol of ΔGr . The latter was expected as reaction elasticities were found to reach a plateau for ΔGr<-30 kJ/mol ( Fig . 3B-C ) . Nevertheless , in order to ensure initial velocity conditions , we chose the former difference . The reference steady-state flux under this condition was set to the experimental value of 0 . 064 mM/min found in the literature [48] . Approximately 93% of the sampled kinetics displays positive cooperativity for glucose ( Fig . 5A ) . However , a small portion of the models ( ~7% ) displayed an apparent negative cooperative behaviour . The latter is possible due to the existence of competing parallel pathways in the reaction mechanism , i . e . E → E-glc → E-glc-atp → E-g6p-adp → E-adp → E and E → E* → E-glc → E-glc-atp → E-g6p-adp → E-adp → E ( Fig . 4 ) . As previously mentioned , reaction mechanisms with branched steps can exhibit apparent substrate inhibition depending on the kinetic parameter values and the substrate concentrations . In this case , this will depend on the value of the branching factor for the E → E-glc and E → E* steps . In particular , if a large elementary net flux is sampled for the E → E* step at the reference state , then the successive addition of glucose will inhibit the velocity of the reaction as opposed to enhancing it . The great majority of the models displaying cooperativity for glucose are consistent with a slow transition from the low to the high-affinity enzyme state ( 98 . 1% ) ( Fig . 5B ) . This kinetic behaviour has been extensively studied and there is abundant supporting evidence [49] . Remarkably , the sampled kinetics contained the experimentally observed Hill coefficient for this enzyme ( nH , real = 1 . 70 ±0 . 1 [49] ) , within one standard deviation ( nH , sampled = 1 . 77±0 . 5 ) , which confirms the suitability of the mnemonic model for modelling this kinetic behaviour . Fig . 5C shows a comparison of the fitted kinetic model by Storer and Cornish-Bowden [48] and the sampled kinetics using our framework . Better agreement between the models is encountered at high glucose concentrations ( >20 mM ) , i . e . where the cooperative behaviour is less pronounced , while at lower glucose concentrations ( <4 mM ) slightly higher discrepancies can be observed . As expected , a perfect match between both models is found at the reference state , as by construction the proposed framework builds the parameterization at this point ( vref = 0 . 064 mM/min ) . Interestingly , the variability of the sampled kinetics is not proportional to the glucose concentration as opposed to fitted model ( Fig . 5C ) . For some glucose concentration regions , the kinetic behaviour of the sampled models displays a relative greater variability for low glucose concentrations ( <20 mM ) and a relative lower variability for higher concentrations . The latter shows that thermodynamically consistent parameter sampling can provide means for effectively accounting for the feasible kinetic space without the need for excessive data . Moreover , the addition of extra data will further reduce the feasible kinetic space for this particular mechanism . Taken together , these results show that by using this framework , exploration of the kinetic space for complex reactions can be achieved provided that the reaction mechanism is known and by setting the right thermodynamic constraints . Phosphoenolpyruvate carboxylase ( EC 4 . 1 . 1 . 31 ) is one of the CO2-fixing enzymes present in the carbon central metabolism of many photosynthetic organisms as well as most non-photosynthetic bacteria and protozoa [50] . In glucose-limited Escherichia coli cultures , PEPC is involved in the only active anaplerotic reaction replenishing pools of intermediary metabolites of the tricarboxylic acid ( TCA ) cycle [51] ( Fig . 6A ) . This enzyme catalyses the conversion of phosphoenolpyruvate ( PEP ) and bicarbonate into oxaloacetate and orthophosphate in the presence of Mg2+ . This reaction is highly exergonic ( ΔGro = -43 . 2kJ/mol[52] ) , making it practically irreversible under physiological conditions . The catalytic mechanism of this enzyme has been elucidated with an ordered binding of phosphoenolpyruvate first and releasing of oxaloacetate as the final step ( Fig . 6B ) [50] . The regulatory mechanism behind the operation of this enzyme is much more complex . The enzyme is allosterically activated by acetyl-CoA , long-chain fatty acids and their acyl coenzyme A derivatives , fructose 1 , 6-bisphosphate ( FBP ) and the nucleotides guanosine-5’-triphosphate and cytidine-5’-diphosphate , whereas it is inhibited by L-aspartate and L-malate [53–56] ( Fig . 6A ) . In particular , the mechanism of activation of this enzyme upon the combined action of FBP and acetyl-CoA has been studied in detail . These two activators bind to different allosteric sites , synergistically activating the tetrameric active form [57] . Alone FBP has been shown to exert little activation of the enzyme upon binding of PEP; however acetyl-CoA alone can greatly enhance the activation induced by FBP [58] . A plausible synergistic model capable of explaining this behaviour has been proposed by Smith et al . [57] and it is shown in Fig . 6C . Notably , this model is a hybrid between that of Monod et al . [20] and that of Koshland et al . [21] in that the mechanism of transitions requires the presence of the activators and/or ligand ( PEP ) for the enzyme to be active ( induced fit ) , however it also assumes isomerization of all subunits to describe cooperative interactions ( concerted model ) . The particular kinetic features of the PEPC have been shown lately to play a key role in the regulation of anapleurosis in E . coli . Xu et al . [58] have demonstrated that E . coli is able to turn off PEP consumption quickly upon glucose removal thanks to the ultrasensitive response of the PEPC upon FBP depletion . Following glucose removal , depletion of FBP from 15 mM to 0 . 45 mM has been shown to almost entirely turn off PEP consumption [58] . This rapid response enabled accumulation of PEP for the rapid import of glucose when it becomes available again . Using the present framework , we sampled the complex regulatory behaviour using the mechanistic information depicted in Fig . 6B-C . For the regulatory mechanism , however , we also considered the tense form to be capable of performing the reaction , although with a lower activity compared to the relaxed form . Indeed , the model of Smith et al . [57] can be regarded as a special case of our parameterization with aref = 0 . In this case , we are interested in describing the ultrasensitive response of PEPC in the presence/absence of acetyl-CoA under changing FBP concentrations . To this end , experimental data from Xu et al . [58] was used as reference data to build and sample feasible kinetics . The thermodynamic reference state was chosen to ensure initial velocity as done for the mammalian glucokinase . To assess the performance of our framework , we compared our results against an empirical model developed by Lee et al . [59] and calibrated using data collected under similar conditions [55] . The empirical model was adjusted to the same reference state used during sampling to ensure a fair comparison . Our sampling strategy accurately described the kinetic behaviour of the PEPC for different FBP concentrations in the presence of acetyl-CoA at physiological concentrations ( Fig . 7A ) . Moreover , it exhibited a slightly better performance than the model of Lee et al . [59] under the same condition . However , a worse performance of our approach is observed in the absence of acetyl-CoA ( Fig . 7A ) . This was expected as our framework builds kinetic models around one reference state , which in this case was set to 0 . 63 mM acetyl-CoA and 2 mM PEP . When acetyl-CoA is absent , a larger diversity of plausible kinetics is predicted by our sampling approach , although displaying a more sigmoidal behaviour . In order to improve the fitting to this condition , we can perform a rejection step during the sampling so that every accepted parameter set agrees with the experimental data under this condition . This strategy is typically used in Bayesian Inference by Approximate Bayesian Computation ( ABC ) methods [60] . In particular , the above method corresponds to the simplest ABC method , the rejection sampler [61] . There are other more efficient samplers implemented within the ABC setting to compute the parameters’ distributions [62 , 63]; however we opted for the simplest as a-proof-of-principle to demonstrate our strategy . As it can be observed in Fig . 7B , the inclusion of additional experimental data further constraints the plausible kinetic space . Interestingly , even with the addition of extra information during the sampling , the most accurate description for this kinetics displays a sigmoidal behaviour as opposed to the observed hyperbolic kinetics . Indeed , PEPC activation is a fairly complex phenomenon and involves the synergistic interplay of two classes of effectors ( type I , e . g . FBP , and type II , e . g . acetyl-CoA ) [57] . In particular , PEPC activation can be achieved by the sole action of FBP or combined with acetyl-CoA . This behaviour has been previously attributed to play a key role in the rapid adaptation of E . coli from normal-growing culture conditions to carbon-starvation or acetate switch conditions [58] . To explore this feature more thoroughly , we generated the Hill curves for PEPC upon PEP binding using the most accurate sampled parameter set in the presence/absence of FBP and acetyl-CoA upon binding of PEP ( Fig . 7C , see S1 Text for generation of Hill curves ) . In order to compare our results , we also show experimentally measured Hill coefficients reported by Izui et al . [55] obtained under similar conditions . Firstly , our results are consistent with the theory as they show bell-shaped Hill curves reaching an asymptotic value of unity at either PEP → 0 or PEP → ∞ [64] . Furthermore , they suggest acetyl-CoA is the most powerful activator . As the curves move to the left , the apparent affinity constants increase . The largest displacements are observed in the presence of acetyl-CoA , which supports its role as an invariant activator for the synergistic co-activation of PEPC [57] . Another consequence of the latter is the increase of the enzyme forms in the relaxed state . As the apparent affinity increases , more enzyme forms transition from the tense to the relaxed state . In terms of the prediction of the cooperativity under the different conditions studied , the sampled kinetics is in close agreement with the experimental data . Notably , a theoretical higher cooperativity for PEPC upon PEP binding is observed in the presence of FBP rather than acetyl-CoA ( max ( nH , Acetyl-CoA ( - ) , FBP ( + ) ) = 2 . 05 , max ( nH , Acetyl-CoA ( + ) FBP ( - ) ) = 1 . 4 ) ( Fig . 7C ) . These results agree with previous reports indicating low cooperativity for acetyl-CoA alone ( 1 . 07< nH < 1 . 2 for 0 . 02 < acetyl-CoA < 1 . 0 mM and 0 . 1 < PEP < 50 mM [55 , 57] ) , whereas significant cooperativity for the FBP interaction [53 , 55 , 65] . Moreover , our results predict a maximal Hill coefficient of approx . 1 . 15 for PEPC in the absence of both FBP and acetyl-CoA which is close to the reported value of 1 . 2 reported by Izui et al . [55] . To further understand the synergistic contributions of the FBP and acetyl-CoA on the PEPC , we explored the impact of acetyl-CoA on the regulatory function ( Ψregulatory ) using again the most experimentally consistent sampled parameter set . Under carbon-starvation conditions ( low FBP ) and at acetyl-CoA physiological concentrations , PEPC was only activated in ~10% ( Fig . 7D ) . The presence of both effectors at physiological concentrations as seen in glucose-limited cultures enhanced PEPC activity to almost 99% . Altogether , our results predict a drop of ~90% in the activity of PEPC upon shifting from normal culture conditions to carbon starvation . Experimental values determined for this enzyme simulating these conditions yielded an approx . 94% decrease in its activity [58] , close to the one predicted by our model . More importantly , the employed parameterization provides the means to understand how this enzyme is being regulated . As such , this framework is not only capable of sampling and modelling complex kinetic behaviours , but it also provides useful insights into the regulatory mechanisms underpinning those kinetics .
We have presented a general framework for parameterizing and sampling almost any reaction kinetics . Parameterization relies on the integration of the generalized MWC model for modelling the kinetics of oligomeric enzymes with the elementary reaction formalism for deriving the catalytic rates and thermodynamic constraints between rate parameters . As a result , this framework enables exploration of the feasible kinetic space of models following a particular reaction mechanism , displaying a given reference flux under a specific thermodynamic condition . Exploration of the kinetic space in this way provided the necessary tools for evaluating the impact of thermodynamics on enzyme kinetics , as well as the consequences of particular regulatory features on the kinetic behaviour . Uniform sampling of the kinetic space of reactions enabled inspection of energetic contributions for different reaction molecularities . To that end , a simple relation between conversion terms ( catalytic constants ) and saturation terms ( dissociation constants ) was derived . As expected , higher molecularities increase the relative importance of the saturation term on average compared to the conversion term . The latter suggest that higher impacts on the catalysis of multi-substrate enzymes would be expected by modifying its ligand binding and releasing properties rather than the conversion rate of substrates into products . In fact , most common approaches for multi-substrate enzyme engineering involve modifying substrate specificity and selectivity [66] , although the success of these strategies will ultimately depend on the particular reaction mechanism and enzyme . Notably , our approach could be employed to sample the rate constants of a particular reaction and determine the flux control coefficient distributions of each step in the mechanism [67] . The latter would provide a broader picture of the pattern regulation , enabling the identification of rate-limiting steps whose modification would most likely improve desired kinetic properties . This framework was also useful for revealing the impact of the thermodynamic reference state on enzyme kinetics . It has been shown that the reaction elasticities strongly depend on this variable . Our analysis distinguished three elasticity regions as a function of the thermodynamic affinity independent of the bimolecular mechanism analysed: a linear sensitivity region with a steep slope ( 0> ΔGr>-2 kJ/mol ) , a transition region ( -2> ΔGr>-20 kJ/mol ) and a constant sensitivity region ( ΔGr<-20 kJ/mol ) . Notably , substrate and product elasticities reached their highest absolute values close to equilibrium , which agrees with the tendency of substrate elasticities to approach infinity at equilibrium [2] . This framework also enabled analysis far from equilibrium . In all reaction mechanism tested , both substrate and product elasticities reached a plateau , reflecting the saturation state of the enzyme . In terms of the reaction mechanisms , all of them displayed elasticities across the thermodynamic reference states , although a particular behaviour could be distinguished for the random mechanism . The latter suggests that knowledge of the thermodynamic state can be more valuable than exact determination of the reaction mechanism at least for non-random reaction mechanisms . Exploration of complex kinetic behaviours can also be achieved by employing this framework . To illustrate this , we sampled the kinetic behaviours of the mammalian glucokinase and the phosphoenolpyruvate carboxylase of E . coli . In the case of the mammalian glucokinase , we were able to model its positive cooperativity for glucose by employing the mechanistic mnemonic model . Random sampling of the kinetic behaviour for this enzyme showed that the experimental Hill coefficient for this enzyme lies surprisingly fairly close to the average Hill coefficient sampled , highlighting the importance of the architecture of its reaction mechanism . Moreover , detailed analysis of the rate constants for the transition from the low- to the high- affinity state confirmed that positive cooperativity for glucose is the result of a slow transition between these two states . In particular , application of our framework to other monomeric proteins exhibiting allosteric behaviours , e . g . thrombin , myoglobin [68 , 69] , might be a valuable tool to better understand the mechanisms underpinning their kinetics . In the case of PEPC , exploration of the kinetic space provided valuable information related to its regulation . Not only were we able to describe the complex co-activation behaviour of this enzyme by FBP and acetyl-CoA , but also to determine the overall impact of this co-activation on its cooperativity for PEP . Furthermore , our strategy offered insights into the ultrasensitive regulation of PEPC upon binding of FBP and acetyl-CoA . As in the case of mammalian glucokinase , our results emphasises the importance of using mechanistic/phenomenological models for describing and interpreting kinetic behaviours . We have shown a diverse set of examples illustrating the capabilities of this framework; however our approach holds a vast number of possible further applications . The sampling of enzymatic reactions within metabolic pathways seems particularly promising . The difficulty of parameterizing metabolic pathways is widely recognised with the main difficulty being the large number of parameters and the relative few data available . Rather than grossly simplifying kinetics to enable fitting , this work suggests that it is possible to build feasible , accurate kinetic parameterizations with limited data by integrating phenomenological models with efficient sampling techniques . In particular , we believe that experimentalists could greatly benefit from this framework in those cases where the fitting is difficult or requires large amounts of data . It will be therefore our next step to deploy GRASP as a software package to assist in such complicated tasks .
|
Kinetic models enable understanding and prediction of the dynamic behaviour of enzymatic reactions . Different frameworks have been proposed to parameterize enzymatic reactions using approximate expressions while maintaining thermodynamic consistency . Approximate expressions have been particularly sought and used , as kinetic expressions typically require large amounts of data to fit their parameters . The latter however ignores real kinetic behaviours and incurs in loss of generality . To overcome these limitations , here we present a novel framework GRASP for exploring the kinetic behaviour of enzymatic reactions under uncertainty based on parameter sampling . By formulating the appropriate thermodynamic constraints and using minimal biochemical reference data , our framework is capable of parameterizing the kinetics of any oligomeric enzyme without sacrificing complexity . To this task , we integrated the generalized MWC model and the elementary reaction formalism , providing a thermodynamically-safe and easy-to-sample parameterization . Application of our framework provided valuable insights into how reactions are regulated under non-equilibrium conditions . We also showed how our approach can be used to describe and understand complex kinetic behaviours of enzymes involved in key regulatory steps of cell metabolism . Overall , this framework enables systematic exploration of the feasible kinetic behaviour of enzymes .
|
[
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
] |
[] |
2015
|
A General Framework for Thermodynamically Consistent Parameterization and Efficient Sampling of Enzymatic Reactions
|
Adeno-associated viruses ( AAV ) are Dependoparvoviruses that have shown promise as recombinant vectors for gene therapy . While infectious pathways of AAV are well studied , gaps remain in our understanding of host factors affecting vector genome expression . Here , we map the role of ring finger protein 121 ( RNF121 ) , an E3 ubiquitin ligase , as a key regulator of AAV genome transcription . CRISPR-mediated knockout of RNF121 ( RNF121 KO ) in different cells markedly decreased AAV transduction regardless of capsid serotype or vector dose . Recombinant AAV transduction is partially rescued by overexpressing RNF121 , but not by co-infection with helper Adenovirus . Major steps in the AAV infectious pathway including cell surface binding , cellular uptake , nuclear entry , capsid uncoating and second strand synthesis are unaffected . While gene expression from transfected plasmids or AAV genomes is unaffected , mRNA synthesis from AAV capsid-associated genomes is markedly decreased in RNF121 KO cells . These observations were attributed to transcriptional arrest as corroborated by RNAPol-ChIP and mRNA half-life measurements . Although AAV capsid proteins do not appear to be direct substrates of RNF121 , the catalytic domain of the E3 ligase appears essential . Inhibition of ubiquitin-proteasome pathways revealed that blocking Valosin Containing Protein ( VCP/p97 ) , which targets substrates to the proteasome , can selectively and completely restore AAV-mediated transgene expression in RNF121 KO cells . Expanding on this finding , transcriptomic and proteomic analysis revealed that the catalytic subunit of DNA PK ( DNAPK-Cs ) , a known activator of VCP , is upregulated in RNF121 KO cells and that the DNA damage machinery is enriched at sites of stalled AAV genome transcription . We postulate that a network of RNF121 , VCP and DNA damage response elements function together to regulate transcriptional silencing and/or activation of AAV vector genomes .
Adeno-associated viruses ( AAVs ) rely on co-infection of the host cell by a helper virus as well as several host factors for replication [1] . The 4 . 7kb single stranded DNA AAV genome contains two open reading frames flanked by two inverted terminal repeats ( ITRs ) packaged into an icosahedral capsid measuring 25nm in diameter [1 , 2] . The only required cis-packaging signal for generating recombinant AAV vector genomes are the two ITRs [1 , 2] . The AAV infectious cycle begins with binding of attachment factors on the cell surface , with different serotypes binding distinct glycan moieties , which have been linked to different tissue tropisms in vivo [1 , 3] . Following endocytic uptake , AAV traffics through endosomes and the golgi network to the nucleus [4] . Further , AAV is thought to enter the nucleus through nuclear pores with the capsid intact , only uncoating once inside this environment . The uncoated ssDNA AAV genome undergoes second-strand synthesis and is then transcribed , although the impact of host factors on the latter event remains poorly understood [5 , 6] . Understanding post-second strand synthesis events in AAV biology could shed light on AAV vector genome silencing that has been observed in gene therapy clinical trials [6 , 7] . Several high-throughput screening based studies have facilitated the discovery of previously unknown host factors that influence AAV transduction . For instance , high-throughput proteomic screens probing AAV2 and AAV8 host binding partners revealed a wealth of novel interactors , including CDK2/cyclinA kinase [8] . An siRNA screen elucidated host restriction factors involved in DNA damage response and cell cycle checkpoint activation [9] . A different siRNA screen demonstrated that the SUMOylation pathway restricts AAV transduction prior to nuclear entry and vector genome release [10] . In another siRNA screen , the U2 snRNP spliceosome was identified as a key host factor that limits transcription of the AAV vector transgene and specific components of the spliceosome validated as AAV capsid binding partners [11] . More recently , a haploid screen for essential AAV host factors uncovered the importance of a new universal AAV receptor , KIAA0319L or AAVR[12] . In the same study , amongst other hits for endosomal and Golgi related genes implicated as essential host factors , RNF121 or Ring Finger Protein 121 was identified as a potential AAV host factor . In the current study , we dissect the previously uncharacterized role of RNF121 in AAV transduction . We utilize CRISPR knockout cell lines to demonstrate RNF121 depletion causes near ablation of AAV transgene expression , regardless of serotype , cell line , or vector dose . Interrogation of the importance of RNF121 at each step of the AAV infectious cycle is used to demonstrate that RNF121 is unnecessary for AAV cellular uptake , nuclear entry , uncoating , and production . A battery of RNA biology-focused assays including qRT-PCR , mRNA stability , and Chromatin Immunoprecipitation studies establish that RNF121 is essential for AAV genome transcription . Further , through interrogation of ubiquitination and degradation pathways , proteomics as well as transcriptome analysis , we discover that RNF121 is essential to prevent transcriptional arrest of AAV genomes potentially mediated by a complex network involving DNA-PKCs , VCP/p97 and the DNA damage machinery .
CRISPR guide RNAs targeting RNF121 , as well as a nontargeting Scrambled guide control , were cloned into the LentiCrisprV2 backbone and used to produce recombinant lentivirus . We transduced Huh7 hepatocarcinoma cells with these constructs and selected with puromycin to produce stable knockout lines . Clonal lines were isolated via serial dilution and knockout validated via western blot and high-throughput sequencing of indel sites ( Fig 1A , S1 Fig ) . RNF121 knockout ( RNF121 KO ) lines show a highly significant , 2-log-fold knockdown of AAV2-luciferase transgene expression in comparison to non-targeting ( Scr ) control line ( Fig 1B ) . This phenotype was recapitulated in the HEK293 human embryonic kidney and U87 glioblastoma cell lines , which were chosen in order to interrogate the consequences of RNF121 KO in divergent cell lines derived from other tissues ( Fig 1C , S1 Fig ) . RNF121 KO mediated decrease in transduction was observed for multiple serotypes tested , including AAV1 , AAV6 , and AAV9 ( Fig 1B ) . RNF121 KO and Scr Huh7 cells were also transduced with AAV2-luciferase at a range of 100–100 , 000 vector genomes per cell . A 2-log decrease in transgene expression was observed independent of vector dose ( Fig 1D ) . We next generated an RNF121 cDNA plasmid and validated it by western blot after transfection ( Fig 1A ) . Transfection of RNF121 cDNA in Scr HEK293 cells did not enhance transduction over an empty vector control , suggesting that RNF121 is not typically a limiting factor in tissue culture models ( Fig 1E ) . However , overexpression partially restored AAV infectivity to previously resistant RNF121 KO cells ( Fig 1E ) . Given the role of Adenovirus as a helper virus for AAV replication , we investigated whether RNF121 was an essential host factor for Adenovirus transduction . We transduced RNF121 KO Huh7 with recombinant human Adenovirus 5 expressing GFP ( hAd5-GFP ) [13] . Flow cytometric analysis demonstrated similar GFP expression in RNF121 KO cells relative to non-targeting control after hAd5-GFP transduction ( Fig 2A ) . Because Adenovirus enhances AAV transduction through more efficient endosomal trafficking and transcriptional activation [14–16] , we postulated that helper virus genes could rescue AAV transduction in RNF121 KO cells . However , coinfection of wild-type human adenovirus 5 with AAV2 luciferase did not fully restore AAV transduction in the Huh7 knockout line , though coinfection did significantly enhance transduction in Scr cells , as determined by two-way ANOVA analysis ( Fig 2B ) . To further investigate the role of RNF121 in the later steps in the AAV infectious pathway , we utilized Scr control and RNF121 KO HEK293 to produce AAV2 vectors packaging a luciferase transgene cassette by triple plasmid transfection [17] . We found no difference in AAV vector yield between the two cell lines , and the vectors produced displayed similar infectivity as well ( S2A and S2B Fig ) . We interrogated the different steps in the infectious pathway of AAV in RNF121 KO cells . Viral binding and uptake assays were performed as previously described [17] . Briefly , cells were pre-chilled and incubated with virus at 4°C to prevent cellular uptake . Following three PBS washes , vector genome DNA was harvested to quantify binding . To monitor cellular uptake , cells were moved to 37°C for one hour , trypsinized to remove bound , but not internalized virions , washed , then vector genome DNA harvested for quantitation . As seen in Fig 2C and 2D , binding and uptake assays demonstrate no defect in these early steps of the AAV infectious pathway in RNF121 KO cells , regardless of AAV serotype . RNF121 KO cells demonstrated a slight yet significant increase in binding and uptake of AAV1 ( Fig 2C and 2D ) . However , this modest change is unlikely to account for the large decrease observed in transduction efficiency . Given the previously characterized importance of AAVR in AAV cellular entry , we then investigated AAVR expression in RNF21KO cells via western blot [12] . Whole cell lysate from Scr and RNF121 KO HEK293 had similar levels of AAVR ( Fig 2E ) . Characterization of the distinct interaction of each AAV serotype with AAVR has demonstrated that some AAVs , e . g . , AAV4 infect cells through an AAVR-independent pathway [18] . To investigate whether AAVR independence can rescue transduction in the absence of RNF121 , we infected RNF121 KO cells with AAV4 . However , AAV4 displayed a transduction defective phenotype in RNF121 KO cells mirroring that of other serotypes ( Fig 2F ) . Next , cytoplasmic and nuclear fractions were harvested from RNF121 KO and control Scr cells following AAV transduction . AAVs have previously been characterized to enter the nucleus through nuclear pores with the capsid intact , only uncoating following nuclear entry [19 , 20] . The distribution of viral genomes was similar for both cell lines , indicating that both cytoplasmic and nuclear entry is not affected in the context of RNF121 KO ( Fig 2G ) . Further , AAV uncoating was interrogated as previously described [21] . Briefly , nuclear extracts were harvested from Scr and RNF121 KO cells transduced with AAV . One half of each extract was subject to Benzonase treatment , whereby uncoated genomes would be degraded , while the second aliquot of nuclear extract was untreated . Both fractions were subject to qPCR to assess the relative number of vector genomes . No statistically significant difference was observed between the percentage of uncoated vector genomes in RNF121 KO vs Scr cells , and the percentage of uncoated vector genomes was consistent with previous publications ( Fig 2H ) [21] . Second-strand synthesis of the ssDNA AAV genome typically represents a rate limiting step in AAV gene expression; however , this step can be circumvented with the use of self-complementary vectors , which contain an inverted repeated genome that can self-anneal without host DNA synthesis machinery [1 , 22] . Transduction of Scr and RNF121 KO Huh7 cells with single-stranded and self-complementary AAV2-GFP vectors demonstrated no measureable GFP transgene expression with either vector in RNF121 KO cells ( Fig 3A ) . To further probe this step we transduced control and RNF121 KO cells with a self-complementary GFP vector , and analyzed GFP expression via flow cytometry . Flow cytometric quantification of rAAV2-scGFP expression confirmed our observed ablated expression of self-complementary vectors ( Fig 3B ) , suggesting RNF121 KO cannot be rescued with vectors that can circumvent second-strand synthesis . To assess the effect of different promoter elements on AAV genome expression in RNF121 KO , we transduced Scr and RNF121 KO Huh7 with AAV2-GFP cassettes driven by the liver specific promoters hAAT and TTR and the MVM intron . We found GFP transgene driven by these promoters was also deficient in RNF121 KO cells ( Fig 3C ) . Additionally , we transduced Scr and RNF121 KO HEK293 with rAAV2-CMV-Luciferase , finding CMV-Luciferase transduction recapitulated the phenotype seen with CBA driven vectors ( Fig 3D ) . These data confirmed that altering rAAV promoter cassettes was not sufficient to rescue transduction . To determine whether AAV genomes were affected in RNF21 KO cells in a capsid-independent manner , we purified single-stranded , vector genomic DNA from AAV capsids packaging the luciferase transgene flanked by ITRs , and transfected capsid-free vector genomes into RNF121 KO cells . No effect in transgene expression relative to Scr was seen ( Fig 3E ) . To determine whether AAV capsids in trans were capable of repressing transduction in the context of RNF121 KO , KO and control cells were transfected with the vector genome , following which with the cells were infected with AAV capsids packaging a different ( human factor FIX ) transgene cassette . Again , no change in luciferase transgene expression was observed , demonstrating that capsid presence in trans does not repress expression of transfected AAV genome in RNF121 KO cells ( Fig 3E ) . These results suggest that RNF121 KO specifically affects gene expression from vector genomes packaged into AAV capsids . We then evaluated the impact of RNF121 KO on transcription of AAV vector genomes . Scr and RNF121 KO cells were transduced with AAV2-GFP , and transgene-derived mRNA levels were quantitated by qRT-PCR relative to mRNA levels of the housekeeping gene , GAPDH . As shown in Fig 4A , transgene mRNA was reduced by over two orders of magnitude in RNF121 KO cells relative to Scr control . Further , we also infected Scr and RNF121 KO cells with AAV capsids packaging the endogenous wild type genome ( wtAAV ) in the presence and absence of human Ad5 to assess the impact of RNF121 on transcription of wtAAV genes . A robust decrease in wtAAV genomic mRNA was observed as well , measured using qRT-PCR with primers against the AAV Rep gene ( Fig 4B ) , which was determined to be significant via two-way ANOVA in the case of wtAAV and adenoviral coinfection . We then carried out a chromatin immunoprecipitation ( ChIP ) assay to determine whether AAV vector genomes were differentially modified in the context of RNF121 KO . Following transduction of Scr and RNF121 KO HEK293 with AAV2-Luc for 48 hours , we analyzed epigenetic marks near the transcriptional start site ( TSS ) of the AAV vector genome . First , we probed H3K27 acetylation ( H3K27ac ) , a marker of transcriptional activation . Previously , inhibition of histone deacetylases has been shown to enhance AAV genome expression , suggesting that acetylation may be important for transcriptional potency of AAV genomes [23 , 24] . At 180 bp downstream of the TSS , Scr cells had a 1 . 51 fold increase in H3K27ac relative to RNF121 KO , while 550 bp downstream of the TSS Scr cells showed a 1 . 48 fold increase in H3K27ac ( Fig 4C ) . However , it is unlikely that this modest difference is a major contributor to the robust transcriptional phenotype of RNF121 KO . We also investigated changes in the recruitment of transcriptional machinery using an antibody against serine 5 phosphorylated RNA polymerase II ( phosphoRNApolII ) , an activated form of RNA polymerase [25] . AAV genomes demonstrated statistically significant 3-fold depletion of phoshpoRNApolII both 120 and 380 bp upstream of the TSS in RNF121 KO cells relative to Scr control ( Fig 4D ) . As shown in S3A and S3B Fig , we did not observe altered recruitment of these modifications in the host GAPDH gene , suggesting that RNF121 KO specifically reduces RNA pol II association with the AAV genome . Because RNF121 KO profoundly affects levels of AAV genomic mRNA , we also assayed mRNA stability in the absence of RNF121 to determine whether the lack of AAV mRNA in RNF121 KO is due in part to mRNA degradation . We performed an Actinomycin D time course to assay mRNA stability , transducing RNF121 KO cells with 100 , 000 viral genomes/cell to achieve modest transcription of the AAV genome relative to a lower dose ( 10 , 000 vg/cell ) on Scr cells . Briefly , qRT-PCR analysis was carried out targeting the AAV vector genome ( GFP transgene ) , as well as GAPDH , TATA-Binding Protein ( TBP ) , and c-myc , three host genes with decreasing mRNA half-life ( Fig 4E and 4F ) . Scr and RNF121 KO cells demonstrated similar sustained presence of GAPDH , a highly stable transcript , as well as similar depletion of c-myc and TBP transcripts , which are known to be quickly degraded [26] . Approximately 80% of AAV GFP mRNA was present at 30hrs post-infection in both Scr and RNF121 KO cells indicating that reduced stability of the AAV mRNA transcript cannot explain the large decrease in steady-state AAV mRNA levels seen in RNF121 KO cells . These data are consistent with the AAV genome having a deficit in transcription in RNF121 KO , rather than altered kinetics or mRNA half-life . Scr and RNF121 KO cells were transduced with AAV2 vectors and subject to confocal microscopy to assess immunofluorescence-based colocalization , probing for both the AAV capsid ( via A20 antibody ) and RNF121 . A20 binds intact AAV capsids , typically localizing around the perinuclear space as AAVs traffic through the endosome and Golgi apparatus [27] . Intracellular localization of AAV capsids was similar between the Scr and RNF121 KO cell lines and demonstrated a perinuclear pattern ( Fig 5A ) . Further , RNF121 demonstrated a perinuclear pattern in Scr cells , consistent with earlier reports and as expected , did not show any fluorescent signal in KO cells . ( Fig 5A ) [28 , 29] . RNF121 also demonstrated marked colocalization with AAV capsids in Scr cells ( Fig 5A ) . To assess whether RNF121 may be directly interacting with AAV capsids , AAV2 capsids were biotinylated as previously described [19] , and RNF121 KO and Scr cells transduced with biotinylated vector were subject to immunoprecipitation ( IP ) using an anti-biotin antibody ( Fig 5B ) . Blotting of input whole cell lysate confirmed no difference in capsid protein in transduced Scr vs RNF121 KO cells , and anti-biotin immunoprecipitation of these lysates successfully pulled down capsid in both Scr and RNF121 KO samples relative to an unlabeled virus control ( Fig 5B ) . However , we were unable to pull down RNF121 with the capsid above background levels ( Fig 5B ) . Because many E3 ligases interact transiently with their substrates , we sought alternative methods to determine whether RNF121 ubiquitinates AAV capsid proteins . Using recombinant human RNF121 , purified AAV capsids , E1 activating and E2 conjugating enzymes ( UBE2N , UBE2E3 ) , we performed cell-free in vitro ubiquitination assays of AAV capsids . We probed capsids incubated with these enzymes with antibodies against denatured AAV capsid proteins ( VP1/2/3 ) as well as ubiquitin ( Fig 5C and 5D ) . While we were able to detect AAV capsid VP ubiquitination using an anti- ubiquitin antibody and higher molecular weight bands via anti-VP antibody upon treatment with UBE2N/2E3 E2 conjugating enzymes , ubiquitination was still detectable in the absence of recombinant RNF121 E3 ligase . The addition of recombinant RNF121 failed to alter the presence of high molecular weight AAV capsid bands and ubiquitin signal ( Fig 5C and 5D , lanes 2 vs 3 and 7 vs 8 ) , with E1 and E2 alone producing the same ubiquitination pattern . Therefore , we hypothesized that the capsid proteins VP1/2/3 may not be direct substrates of RNF121 . To understand whether the E3 ligase activity of RNF121 is essential for AAV genome transcription in the cellular context , we generated two catalytic mutants of RNF121; C226A/C229A , described previously , as well as V228A , based on bioinformatic prediction of the RNF121 catalytic site [30] . These plasmids were transfected into RNF121 KO HEK293 cells and assessed for their ability to rescue AAV2-Luciferase expression relative to the WT RNF121 plasmid construct or a negative control plasmid ( Fig 5E and 5F ) . C226A/C229A failed to rescue , but also did not express at wild-type levels; however , V228A RNF121 , which was expressed at the same level of WT RNF121 also failed to fully rescue AAV genome expression . The latter mutant only achieved under 50% transduction as seen when rescuing the defective phenotype with the WT RNF121 plasmid . ( Fig 5E and 5F ) . First , since E3 ubiquitin ligases such as RNF121 , typically act in conjunction with E1 activating and E2 conjugating enzymes , we evaluated PYR-41 , a pan-inhibitor of E1 activating enzymes [31 , 32] ( Fig 6A ) . PYR-41 treatment increased transduction in both Scr and RNF121 KO by roughly one order of magnitude . Further , we blocked ubiquitination by transfecting a dominant negative ubiquitin construct , which prevents ubiquitin chain elongation by adding a terminal ubiquitin without lysine residues [33] . As shown in Fig 6B , transfection of the dominant negative ubiquitin increased transduction in Scr cells by over 3-fold , while transduction in RNF121 KO cells was not affected . Thus , while ubiquitination status might affect AAV transduction in general , blocking this process did not rescue the RNF121 KO phenotype . Next , proteasomal degradation of AAV particles is known to inhibit transduction , and treatment of cells with proteasome inhibitor MG132 has been previously shown to increase AAV transduction [34] . We treated both Scr and RNF121 KO cells with MG132 prior to transduction ( Fig 6C ) . While Scr cells demonstrated a 5-fold increase in transduction , RNF121 KO cells demonstrated a modest increase of approximately 2-fold , which did not rescue the ablated expression caused by RNF121 KO ( Fig 6C ) . We next investigated the potential role of the segregase/unfoldase , Valosin containing protein ( VCP ) , also known as p97 , a AAA+ ATPase known to target substrates to ubiquitin-proteasome pathways and orchestrate the DNA Damage response [35–37] . Blocking VCP with the selective and reversible inhibitor , DBeQ did not substantially alter transduction in Scr cells , but resulted in over 500 fold enhancement of transduction in RNF121 KO cells ( Fig 6D ) . This increase in transduction selectively and almost completely rescued the RNF121 KO phenotype ( Fig 6D ) . In light of the transcriptional arrest of AAV vector genomes in RNF121 KO cells , we carried out transcriptome analysis to identify mRNAs that were deregulated in general . Briefly , we performed RNA seq on Scr and RNF121 KO HEK293 , before or after AAV2-Luciferase transduction ( 10 , 000 vg/cell ) . These studies were carried out in biological triplicates and differential gene expression analyzed between samples ( Fig 7A , Supplemental Dataset 1 ) . The most significantly upregulated host transcript in RNF121 KO cells was PRKDC , which encodes the catalytic subunit of DNA Protein Kinase ( DNA-PKcs ) . DNA-PKcs has previously been shown to be important for the cellular response to DNA damage . In particular , DNA-PKcs has been shown to recruit VCP/p97 through phosphorylation to DNA damage sites [38] . Moreover , the mechanism by which DNA-PKcs results in transcriptional arrest by reducing the association of RNA pol II at double strand break sites is well known [38 , 39] . Given the demonstrated importance of the capsid in transcriptional repression of AAV genomes in RNF121 KO cells , we also performed affinity-purification mass spectrometry to better understand capsid binding partners in Scr and RNF121 KO cells . Scr and RNF121 KO cells were transduced with biotinylated AAV2-Luciferase , and a control population of cells was transduced with unlabeled AAV2-Luciferase . These cells prepared in biological triplicate and subject to immunoprecipitation with a biotin antibody , and the efficiency of this pull-down was confirmed as shown in Fig 5B . Scoring of Mass spectromic hits revealed enrichment of known host transcriptional inhibitors associating with the AAV capsid in RNF121 KO cells , including SF3B ( Supplemental Dataset 2 ) [11] . Ingenuity Pathway Analysis of significant hits enriched in RNF121 KO capsids revealed augmented representation of genes associated with RNA processes ( green ) , translation ( blue ) , viral infection ( lime ) , and DNA damage ( purple ) , with some overlap of hits involved in these processes ( Fig 7B , Supplemental Dataset 3 ) . For example , while U2 snRNP spliceosome components SF3A3 and SF3B are implicated in multiple RNA processes , these genes were also implicated in viral infection , translational processes , and DNA homologous recombination ( Supplemental Dataset 3 ) . Notably , proteomic analysis did reveal enhanced association of both PRKDC and VCP with the AAV capsid in RNF121 KO cells ( Supplemental Dataset 2 ) . To dissect the roles of different pathways highlighted in our transcriptome and proteome results , we performed pharmacological inhibition of notable pathways and assessed the effect on AAV genome expression . U2snRNP inhibition via PladB increased transduction in Scr cells nearly 10 fold , while transduction in RNF121 KO cells was less robustly enhanced ( Fig 7C ) . Inhibition of DNAPKcs did not substantially affect AAV transduction in Scr or RNF121 KO cells ( Fig 7D ) . Conversely , ATM/ATR kinase inhibition resulted in a modest 2 . 7 fold increase in Scr AAV2 transduction , consistent with previous publications , while RNF121 KO cells experienced a selective , nearly 30 fold increase in transduction with this treatment ( Fig 7E ) [40] . Given the proven importance of VCP and PRKDC in DNA Repair pathways , we performed a StringDB interactome analysis of factors involved in the enriched DNA repair pathway of our proteomics ( p = 2 . 78e-14 ) , demonstrating multiple DNA damage repair factors with enhanced association with the capsid in RNF121 KO ( Fig 7F ) . Many of these factors have previously characterized roles in inhibiting AAV genome expression [41–43] . Based on these data , we hypothesize that the absence of RNF121 catalyzes enhanced DNA Damage sensing of the AAV capsid genome complex , resulting in VCP mediated transcriptional silencing .
The current study began by exploring the role of RNF121 , an E3 ubiquitin ligase that was previously identified as an essential host factor for AAV infection[12] . CRISPR KO of RNF121 causes a robust decrease in AAV genome expression , irrespective of capsid serotype , cell line , transgene , or vector dose . RNF121 does not appear to be essential for early steps in the AAV infectious cycle , including cell surface binding and uptake . Immunofluorescent colocalization with RNF121 suggests that the AAV capsid may associate with this ubiquitin ligase during trafficking through the perinuclear space . However , this phenomenon did not correlate with nuclear entry , uncoating , or second-strand synthesis , which were unaffected . Nevertheless , we considered the possibility that RNF121 might play a role in regulation of the universal AAV receptor , AAVR [12] . However , western blotting demonstrated no difference in AAVR protein levels in RNF121 KO , and transduction of RNF121 KO cells by an AAVR independent serotype ( AAV4 ) yielded a similar transduction deficient phenotype . Collectively , these results suggested that RNF121 might be essential for an infectious event further downstream . Subsequent efforts were focused on assessing potential mechanisms that might lead to silencing of AAV gene expression . Specifically , H3K27 acetylation and phosphorylated RNA Pol II-ChIP experiments demonstrated that the RNF121 KO mediated defect in AAV transduction is due to transcriptional arrest . We observed loss in transduction efficiency regardless of the different promoter elements incorporated into the AAV genome cassette . In addition , actinomycin D mRNA stability studies further confirmed a defect in mRNA synthesis for AAV vector genomes . Relatedly , a particularly interesting observation in the current study was that RNF121 KO does not alter transgene expression following transfection of plasmid DNA or purified AAV vector genomes in a capsid-independent setting; however , expression from AAV vector genomes associated with the capsid is robustly silenced . Further , addition of capsids in trans does not affect transgene expression following transfection of vector genomes , suggesting that repression of AAV genome transcription in RNF121 KO cells occurs only with capsid present in cis . Our results are further corroborated by previous studies indicating that the AAV capsid plays a role in the transcription of its genome , where capsid mutants were derived that could traffic effectively to the nucleus and uncoat , but displayed severe transcriptional deficiency mirroring the phenotype observed in RNF121 KO cells [21 , 44] . These observations together suggest that the transcriptional role of the AAV capsid occurs in cis , and that the vector genome must remain associated with the AAV capsid in order to observe the RNF121 related phenotype . Given the critical role of the AAV capsid outlined above , we carried out in vitro ubiquitination assays of the capsid proteins ( VP1/2/3 ) with recombinant RNF121 . Unfortunately , we did not observe direct RNF121 mediated ubiquitination of the capsid in vitro . Furthermore , immunoprecipitation studies did not reveal a direct interaction between RNF121 and the AAV capsids . While it is plausible that the transient nature of interactions preclude observation of direct RNF121-AAV interactions , we posed the larger question as to whether the E3 ligase activity of RNF121 was essential . To this end , we generated catalytically dead RNF121 mutants , which should demonstrate decreased ligase activity [30] . Attempts to rescue AAV genome expression by transfecting catalytically dead RNF121 constructs demonstrated that the E3 ligase activity of RNF121 is indeed essential for AAV genome expression . We then adapted a pharmacological approach to interrogate the role of the ubiquitin-proteasome system in AAV transduction in RNF121 KO cells . Inhibition of ubiquitination with a dominant negative ubiquitin construct increases AAV genome expression in Scr , but not RNF121 KO cells . Inhibition of E1 activating enzymes in the ubiquitin pathway increased transduction in Scr and RNF121 KO cells by the same order of magnitude . While proteasome inhibition increases AAV transduction in Scr cells , it does not substantially increase transduction in the context of RNF121 KO . In sharp contrast to these observations , pharmacological inhibition of Valosin containing protein ( VCP ) /p97 produced essentially restored AAV genome expression in RNF121 KO cells compared to that of control Scr cells . These results suggested that in the absence of RNF121 , VCP , a segregase [35] , is involved in inactivating the AAV genome transcription machinery . To further expand on these findings , we applied high-throughput approaches to understand the impact of RNF121 KO on the AAV genome transcription complex . First , transcriptomic profiling of the host genes upregulated in RNF121 KO cells revealed PRKDC , the transcript encoding DNAPKcs , as our lead hit [39] . As further corroboration , we identified both DNAPKcs and VCP as host proteins that were enriched during proteomic analysis of the AAV genome transcription complex in RNF121 KO cells . Interestingly , DNAPKcs is known to phosphorylate VCP recruiting this segregase to double strand breaks [38 , 45] . Notably , VCP has been shown to function in various chromatin associated degradation processes [36 , 38 , 46] . Further , VCP has also been implicated in sites of stalled transcription and RNA Pol II turnover [47] . In addition , DNAPKcs has been shown to cause transcriptional arrest via decreased association of RNA pol II at DNA breaks [39 , 45] . These observations are consistent with our data showing substantially decreased association of RNA pol II with the AAV genome in RNF121 KO cells . We postulate that VCP functions to inactivate the AAV genome transcription complex in the absence of RNF121 . Moreover , our mass spectrometric data identified proteins from the ubiquitin-proteasome pathway such as UBE2N , OTUB1 , PARK7 , USP10 in addition to VCP associating with the AAV genome transcription complex . These targets warrant further investigation . Consistent with the notion that DNA damage related processes are involved , we observed enrichment of Rad50 , MRE11 , XRCC5/6 ( Ku86/70 ) , DDB1 , PCNA as binding partners of the AAV genome transcription complex in RNF121 KO cells . Several of these host proteins have previously been characterized as host restriction factors that recognize the AAV genome ITRs , while others warrant further investigation [41 , 42 , 48] . Furthermore , DNA-PKcs has been demonstrated to localize with and orchestrate the cellular DNA damage response to wild-type AAV replication centers [49 , 50] . Notably , although Adenovirus type 5 ( hAd5 ) is known to dampen the DNA Damage response through degradation of factors such as Mre11 , hAd5 coinfection was not sufficient to rescue wild-type or recombinant AAV genome expression [51] . The failure of hAd5 coinfection or direct pharmacological inhibition of DNAPKcs to rescue the RNF121 KO phenotype , but the ability of an ATM/ATR inhibitor to partially restore AAV gene expression suggests that only certain selective components of the DNA damage machinery might be directly involved . Overall , we postulate that RNF121 , VCP and DNAPKcs are essential components of a regulatory network that controls AAV genome transcription ( Fig 7G ) . Upregulation of DNAPKcs and recruitment of inhibitory host factors such as VCP in the absence of RNF121 appears to inactivate RNA Pol II mediated AAV genome transcription . Because pharmacological inhibition of both VCP and ATM/ATR kinases restores AAV gene expression in RNF121 KO more robustly than in wild-type cells , we believe that DNA damage mediated repression of AAV genomes is augmented when RNF121 is absent . These data are corroborated by enhanced expression of DNA-PKcs , a kinase known to cooperate with ATR and ATM in activating the DNA damage checkpoint , though inhibition of DNA-PKcs alone was not sufficient to rescue AAV transduction in RNF121 KO cells [45 , 52] . Further studies are warranted to identify and dissect the role of specific RNF121 substrates impacting AAV genome transcription . Nonetheless , our data highlights a novel role for RNF121 in regulating genome transcription for both recombinant and wild type AAVs . Further investigation of the interplay between the ubiquitin-proteasome and DNA damage machinery could shed light on mechanisms underlying transcriptional silencing of AAV genomes with implications for gene therapy .
Guides against RNF121 were designed and ordered as single stranded oligonucleotides from IDT with sequences 5’-GGATCATTGAGAACACGTAT-3’ ( Guide 1 ) and 5’–GTTCCAGAACTCCATAGTAG-3’ ( Guide 2 ) . Guides were phosphorylated , annealed , and ligated into BsmBI digested LentiCrisprV2 ( a gift from the Feng Zhang lab; Addgene plasmid # 52961 ) [53] . To generate the RNF121 overexpression construct , RNF121 cDNA was generated by reverse transcription PCR from total RNA isolated from HEK293 cells ( adherent human embryonic kidney 293 cell line obtained from the University of North Carolina Vector Core ) using Trizol following manufacturer’s protocol ( Invitrogen , Carsbad , CA ) . RNF121a DNA was amplified by PCR ( Phusion HF , NEB , Ipswich , MA ) and cloned into pCDNA3 . 1+ using EcoRI and NotI sites . Catalytic mutants were introduced into this backbone via Site Directed Mutagenesis . pRK5-HA-Ubiquitin-KO was a gift from the Ted Dawson lab ( Addgene plasmid # 17603 ) [33] . The TTR enhancer and promoter , as well as the ApoE enhancer and hAAT promoters , were isolated from gblocks ( IDT ) via restriction enzyme digestion , and cloned along with the MVM intron into an AAV GFP expression cassette . Mouse anti-actin ( ab3280 ) and mouse anti-KIAA0319L ( AAVR ) ( ab105385 ) were obtained from Abcam ( Cambridge , MA ) . Goat anti-rabbit-HRP ( 111-035-003 ) was obtained from Jackon ImmunoResearch ( West Grove , PA ) . Anti-Ubiquitin antibody P4D1 ( sc-8017 ) was purchased from Santa Cruz Biotechnology ( Dallas , TX ) . Rabbit anti-RNF121 ( PA5-61136 ) , Goat anti-Biotin ( 31852 ) , and Goat anti-mouse-HRP ( 32430 ) were obtained from ThermoFisher . Anti-capsid protein antibody B1 [54] was used to blot for capsid protein , while anti-capsid antibody A20 [55] was used for immunofluorescence . MG132 ( 10012628 ) was purchased from Cayman Chemical ( Ann Arbor , MI ) . PYR-41 ( N2915 ) , DBeQ ( SML0031 ) , as well as ATM/ATR ( 118501 ) and DNAPK ( 260960 ) inhibitors were purchased from Sigma-Aldrich . PladB was a kind gift from the lab of Yasuhiro Ikeda ( Mayo Clinic , Rochester , Minnesota ) . Recombinant AAV vectors packaging a chicken β-actin ( CBA ) promoter driven firefly luciferase cassette , self-complementary AAV ( scAAV ) vectors packaging a hybrid CBA ( CBh ) promoter driving GFP , and vectors packaging hAAT and TTR GFP cassettes described above were generated using triple plasmid transfection in HEK293 cells described previously [17] . Viral titers were obtained as previously indicated [17] . Briefly , viral preparation samples were subject to DNAse ( 10mg/mL ) treatment to remove unencapsidated viral genomes . Following inactivation of DNAse with 0 . 5 M EDTA , samples were subject to Proteinase K digestion to release encapsidated viral genomes . Viral DNA was then measured against vector core standards by quantitative PCR with primers against the ITRs using a Roche Lightcycler 480 ( Roche Applied Sciences , Pleasanton , CA ) . rAAV2-CMV-Luciferase was purchased from the UNC Chapel Hill Vector Core . Recombinant lentivirus packaging guides against RNF121 or Scr control guides was produced via triple plasmid transfection with psPax2 and VSVG glycoprotein for pseudotyping in HEK293 cells , as previously described [53] . Media supernatant was harvested at 30 and 48 hours post transfection and filtered . Recombinant Ad-GFP vectors were a kind gift from the Xiao Xiao lab at UNC Chapel Hill and wild type human Ad5 was obtained from ATCC ( VR-1516 ) . Huh7 and U87-MG ( human hepatocarcinoma and human glioblastoma cells obtained from the University of North Carolina Lineberger Tissue Culture Facility ) and HEK293 ( human embryonic kidney cells obtained from the University of North Carolina Vector Core ) . Cells were maintained in Dulbecco’s Modified Eagle’s Medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) , 100U/ml penicillin , 100ug/ml streptomycin . Cells were maintained in 5% CO2 at 37°C . For generation of stable cell lines , Huh7 , HEK293 , or U87 were seeded at 1e5-3e5 cells/well in a 6 well plate and incubated with media containing recombinant lentivirus and 8ug/mL polybrene for 10 minutes , followed by spinoculation for 30 minutes at 400g at room temperature . Following spinoculation , cells were incubated for 2–4 hours after which the media was removed and replaced with media containing lentivirus without polybrene . 48 hours post spinoculation cells were selected with puromycin for 7 days , after which clonal lines were generated by serial dilution of cells . Cells were counted and seeded at equal density ( 3e4 for Huh7 , 1e5 for U87 and HEK293 ) on 24 well plates and allowed to adhere overnight . Cells were then transduced with AAV-CBA-Luciferase vectors at a dose of 2000 vg/cell unless otherwise indicated . 48 hours post transduction , cells were harvested in Passive Lysis Buffer and lysate combined with Luciferin substrate from Promega ( Madison , WI ) . Luciferase signal was then quantified by a VictorX plate reader from PerkinElmer ( Waltham , MA ) . Scr and RNF121 KO Huh7 cells ( representing single Huh7 clonal cell lines stably integrated with scrambled guide RNAs or targeted guide RNAs that knockout the RNF121 gene , respectively ) were seeded in 12 well plates at a density of 5e4 cells/well 18 hours prior to experiment . Cells were first pre-chilled at 4°C for 30 minutes , and then incubated with rAAV-CBA-Luciferase at 4°C for 1 hr , followed by three washes with ice-cold 1x phosphate buffered saline ( 1x PBS ) to remove unbound virions . 300 uL of ddH20 was then added to each well and cells were subject to three freeze-thaw cycles prior to extraction of total genomic DNA using the IBI Mini Genomic DNA Kit ( IBI , Dubuque , IA ) . Quantification of viral genomes per cell was determined via qPCR of DNA samples with primers against the Luciferase transgene and the host Laminin gene . For cellular uptake studies , following removal of unbound virions cells were immersed in warm DMEM + 10% FBS 1% P/S and moved to a 37°C incubator to synchronize virus internalization . 1hr post incubation , cells were treated with 300uL/well 0 . 05% trypsin for 5 minutes to dissociate cell-surface associated virions . Trypsin was quenched with 300uL/well DMEM + 10% FBS 1% P/S , followed by three washes of the cell pellet with cold 1x PBS . Total genomic DNA was then extracted as previously described [17] and vg/cell determined by qPCR . For trafficking studies , Scr and RNF121 KO Huh7 cells were transduced with rAAV2-luciferase for 18 hours , then cytoplasmic and nuclear fractions were harvested with the NE-PER kit ( Thermo Fisher ) . Fractions were subject to qPCR as previously described to quantify vector genomes . Cells were washed with 1X PBS , trypsinized , and resuspended in full media . Resuspended cells were then washed twice with cold 1X PBS and stained for viability with Zombie Violet for 30 min at 4 degrees in the dark , per manufacturer protocol ( Biolegend ) . Zombie Violet staining was quenched with IFWB buffer , then cells were fixed with paraformaldehyde and filtered to remove clumps prior to analysis on a CyAn ADP ( Beckman Coulter ) . ChIP was performed as previously described [56] with HEK293 cells using 10 million cells per replicate and experimental condition . H3K27ac ( Abcam , ab4729 ) and phosphorylated RNA pol2 ( Active Motif , 61085 ) were used . Following ChIP , qPCR was performed with SYBR Green ( Roche , 19317900 ) . Enrichment levels were normalized to GAPDH . For each experimental type , 3 biological replicates were analyzed . The following luciferase primers ( 5’-3 ) were used: ( - ) 380 ( F’ gcagccattgccttttat , R’ gctccgcacagatttgg ) ; ( - ) 120 ( F’ taaccatgttcatgccttct , R’ caaaatgatgagacagcaca ) ; ( + ) 180 ( F’ cctggaacaattgcttttac , R’ gtttcatagcttctgccaac ) ; ( + ) 550 ( F’ atacgattttgtgccagagt , R’ gcagaccagtagatccagag ) . Cells were washed once with 1x PBS then harvested in Trizol Reagent using the manufacturer protocol ( Invitrogen , Waltham MA ) . Following DNAse treatment , cDNA was generated from total RNA via RT-PCR . cDNA samples were then analyzed with qPCR using primers against AAV genomic mRNA and GAPDH housekeeping gene . For mRNA stability work , 2e5 cells/well were seeded in triplicate into 6-well plates and transduced with 1e5 vg/cell ( Scr ) or 1e6 vg/cell ( RNF121 KO ) of the indicated rAAV2 vectors . At 3 days post transduction , media was removed and replaced with fresh media for half an hour . This media was removed and replaced with pre-warmed and equilibrated media containing 5 ug/mL Actinomycin D ( Sigma-Aldrich , St . Louis MO ) . Cells were treated for 30 min , then media was removed and replaced with fresh media . Cells were harvested in Trizol at 0- , 4- , 8- , 12- , and 24-hour time points . RNA was extracted and analyzed by qRT-PCR as described above . Scr and RNF121 KO cells were seeded on slide covers in 24-well plates at a density of 4e4 cells/well and allowed to adhere overnight . Cells were then transduced with AAV2 for 12 hours , then fixed with 4% paraformaldehyde for 30 minutes and permeabilized with 0 . 1% Triton X-100 for 30 minutes . Following 30 minutes of blocking with 5% Normal Goat Serum , cells were stained with A20 and RNF121 primaries for one hour , washed 3x with PBS , and then stained with fluorescent secondaries . Cells were subject to 5 minutes staining with DAPI , and then mounted in Prolong Diamond ( Invitrogen ) and imaged using a Zeiss 710 scanning confocal microscope . AAV capsid was labeled with NHS-PEG4-Biotin ( Thermo Fisher #21330 ) as previously described [19] . Briefly , NHS-PEG4-Biotin was diluted in molecular biology grade water to 2 nM , combined with purified rAAV2-GFP , and incubated for 30 minutes at room temperature . Unbound biotin was removed by buffer exchange using ZebaSpin Desalting Columns ( Thermo Fisher #89882 ) . For immunoprecipitations , Scr and RNF121 KO cells were seeded overnight , and transduced with AAV2 vector with or without biotinylation . 24 hours post transduction , cells were washed with ice cold 1X PBS and harvested in RIPA with 1x Halt Protease Inhibitor ( Thermo Fisher ) for 30 minutes shaking on ice . Lysates were spun at max speed for 10 minutes at 4 degrees , and supernatant was subject to pre-clearing by 1 hour incubation with 25uL Protein G Magnetic beads ( GE ) on a mutator at 4 degrees . Goat anti-Biotin was bound to Protein G Magnetic beads for 1 hour at 4 degrees , after which pre-cleared lysate was added to antibody bound beads and rotated overnight at 4 degrees Celsius . Immunoprecipitations were eluted in 10mM DTT and 1X LDS , and analyzed via western blot . In vitro ubiquitination reactions were performed using rAAV2-Luc ( 2 . 9e12 vg/ml ) as a substrate . The ubiquitin thioester/conjugation kits were purchased from Boston Biochem ( K-995 and K-980B ) . Purified recombinant RNF121 was purchased from Abcam ( ab163056 ) . The reaction mixture contained 3 μM purified RNF121 and 4 . 8nM rAAV2-Luc . Concentrations of other components were used following the manufacturer’s instructions . Reactions were performed at 37°C for 2h . Samples were terminated with 4X non-reducing LDS sample buffer ( Invitrogen NP0007 ) supplemented with 10 mM DTT . Samples were boiled at 95°C for 5min , then subjected to SDS-PAGE ( NuPAGE 4–12% Bis-Tris Gel ) and transferred onto nitrocellulose membrane ( ThermoScientific ) . The signal was visualized with SuperSignal West Femto maximum sensitivity substrate ( ThermoScientific ) according to the manufacturer’s instructions . Scr and RNF121 KO HEK293 were seeded and transduced with 10 , 000 vg/cell rAAV2-Luciferase in biological triplicate . Samples were harvested in trizol and subject to phenol-chloroform extraction , and total RNA samples were submitted to Genewiz ( Morrisvile , NC ) for mRNA library preparation and high-throughput sequencing . Briefly , raw reads were trimmed and aligned with BBMAP , quantified with SubRead , Compared with DeSeq2 , and graphed in R version 3 . 2 . 4 [57 , 58] . Two 15 cm plates of Scr or RNF121 KO HEK293 cells were counted and seeded for each of three biological replicates , and allowed to adhere overnight . Cells were then transduced with biotinylated vector or an unlabeled control at a dose of 2000 vector genomes per cell . 24 hours post transduction , cells were washed with ice cold 1X PBS and harvested in RIPA with 1x Halt Protease Inhibitor ( Thermo Fisher ) , then subject to immunoprecipitation as described above . In lieu of elution with LDS and DTT , beads were washed three times with ice cold 1X PBS , then subject to on-bead digestion and affinity purification mass spectrometric analysis with the UNC Proteomics Core . Raw spectral count data was subject to analysis with the MiST ( mass spectrometry interaction statistics ) system , as previously described [59 , 60] . Targets were analyzed for pathway representation with Ingenuity Pathway Analysis ( Qiagen ) , and graphed with R version 3 . 2 . 4 . Mass spectrometric binding partners represented in the DNA Repair pathway and enriched in RNF121 KO capsids were mapped using StringDB [61] . All data are expressed as mean with error bars representing standard error . A two-tailed unpaired student t test was used or two-way ANOVA where indicated , calculated with GraphPad Prism Version 6 . p values less than 0 . 05 were considered significant . Asterisks are used to denote p values as follows: * , p<0 . 05; ** , p<0 . 01; *** , p<0 . 005 .
|
Recombinant AAV vectors are at the forefront of clinical gene therapy . There is a need to better understand the mechanisms dictating AAV transduction in the host . Here , we identify a network of host proteins involving RNF121 , p97 and the DNA damage machinery as potent factors regulating AAV genome transcription . Our study sheds light on an understudied aspect of AAV biology with implications for gene therapy .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"transfection",
"luciferase",
"enzymes",
"messenger",
"rna",
"microbiology",
"enzymology",
"dna",
"damage",
"immunoprecipitation",
"dna",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"genome",
"complexity",
"genomics",
"proteins",
"oxidoreductases",
"ubiquitination",
"viral",
"packaging",
"viral",
"replication",
"molecular",
"biology",
"precipitation",
"techniques",
"biochemistry",
"rna",
"nucleic",
"acids",
"post-translational",
"modification",
"virology",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] |
2019
|
Ring finger protein 121 is a potent regulator of adeno-associated viral genome transcription
|
Little is known of the direct microbicidal activity of T cells in leprosy , so a lipopeptide consisting of the N-terminal 13 amino acids lipopeptide ( LipoK ) of a 33-kD lipoprotein of Mycobacterium leprae , was synthesized . LipoK activated M . leprae infected human dendritic cells ( DCs ) to induce the production of IL-12 . These activated DCs stimulated autologous CD4+ or CD8+ T cells towards type 1 immune response by inducing interferon-gamma secretion . T cell proliferation was also evident from the CFSE labeling of target CD4+ or CD8+ T cells . The direct microbicidal activity of T cells in the control of M . leprae multiplication is not well understood . The present study showed significant production of granulysin , granzyme B and perforin from these activated CD4+ and CD8+ T cells when stimulated with LipoK activated , M . leprae infected DCs . Assessment of the viability of M . leprae in DCs indicated LipoK mediated T cell-dependent killing of M . leprae . Remarkably , granulysin as well as granzyme B could directly kill M . leprae in vitro . Our results provide evidence that LipoK could facilitate M . leprae killing through the production of effector molecules granulysin and granzyme B in T cells .
The introduction of multidrug therapy in 1982 and the WHO campaign for the ‘elimination of leprosy as a public health problem’ , have contributed greatly to the decrease in the prevalence rate over the past three decades . But leprosy still remains to be a public health problem in some countries , and the number of new cases detected during the last three years , remain steady [1] . The disease presents as a clinical spectrum that correlates with the level of the immune response to the pathogen [2] . Patients with lepromatous form of the disease have poor cellular immunity , resulting in extensive intracellular proliferation of Mycobacterium leprae bacilli in the skin and nerves . On the other hand , patients with the tuberculoid form of the disease are relatively resistant to the bacilli , so that few , if any , demonstrable bacilli are seen in the lesions [2] , [3] . For patients with abundant bacilli , whose lesions are characterized by type-2 cytokines , there is a need to up-regulate the T-cell mediated type 1 immune responses , by immunotherapeutic means to kill the bacilli . We have previously identified a lipoprotein of M . leprae , a 33-kD lipoprotein ( ML0603 ) [4] . Truncated protein , having the N-terminal 60 amino acids of 33-kD lipoprotein , had cytokine inducing ability in human monocytes , in contrast to the C-terminal 192 amino acids having no such ability [5] . In this study , we synthesized the lipopeptide ( LipoK ) having the N-terminal 13 amino acids of the 33-kD M . leprae lipoprotein linked to tri-palmitoylated portion of a lipid . Since GC mass spectrometry of mycobacterial lipoproteins provided evidence for the presence of three fatty acids ( either palmitic , stearic or tuberculostearic acid ) , we assumed that tri-palmitoylated peptide would represent the natural lipoprotein of M . leprae [6] , [7] . Further , N-acyl transferase ( Lnt ) activity was identified in mycobacteria , which transfers the amide-linked acyl group to the N-terminal cysteine residue [6] . This presence of Lnt activity would indicate the presence of triacylated lipoproteins in mycobacteria , although the exact lipid structure of M . leprae lipoprotein is still to be determined . Previously , it was observed that hexameric peptides with tri-palmitoyl modification , corresponding to 19-kD and 33-kD lipoproteins of M . leprae , partially activates cells through TLR2-TLR1 heterodimers [8] , [9] . Since dendritic cells ( DCs ) are the most potent antigen presenting cells capable of bacilli uptake , antigen presentation and initiating acquired immune responses , DCs were used as target antigen presenting cells , in the present study [10] , [11] . As expected , it was found that LipoK , delivered signals through TLR2 , and activated M . leprae infected DCs to produce abundant IL-12 , although , LipoK does not produce IL-12 , in non-infected DCs . Several mechanisms are known to be involved in the clearance of intracellular bacteria , including interferon gamma ( IFN-γ ) release , apoptosis induction of the host cells and anti-microbial activity of CD8+ cytotoxic T lymphocytes ( CTL ) [12]–[15] . CTL mediated killing of mycobacteria , was demonstrated in tuberculosis by Thoma-Uszynski et al . They showed that CD8+ CTL-mediated killing of M . tuberculosis was dependent on granule exocytosis [16] . In the present study , we analyzed whether M . leprae infected DCs , activated through LipoK could undergo functional changes and subsequently induce type 1 T cell activation to kill the bacilli . We observed that LipoK is a potent inducer of T cells equipped with cytolytic function , which can largely contribute to the killing of M . leprae in host cells .
Peripheral blood was obtained from healthy Japanese individuals under informed consent . But no information of the donor ( exposure to bacilli ) was provided . In Japan , BCG vaccination is compulsory for children ( 0∼4 years old ) . Monocyte-derived DCs were differentiated from monocytes using GM-CSF and IL-4 as described earlier [17] , [18] . Animal studies were carried out in strict accordance with the recommendations from Japan's Animal Protection Law . The protocol was approved by the Experimental Animal Committee , of the National Institute of Infectious Diseases , Tokyo ( Permit Number: 210001 ) . M . leprae ( Thai-53 strain ) is passaged in athymic nu/nu mice ( Clea Co , Tokyo ) [19] . At 8 to 9 months post-infection , the footpads were processed to recover M . leprae [20] . For all experiments , M . leprae was freshly prepared . The multiplicity of infection ( MOI ) was determined based on the assumption that DCs were equally susceptible to infection with M . leprae [21] , and immature DCs were infected with M . leprae at MOI 50 in all experiments . Human cells without the bacilli was cultured at 37°C , but when infected with the bacilli , the cells were cultured at 35°C , which is the minimal temperature at which the cells can survive in in-vitro experiments . LipoK having the structure Palmitoyl-Cys ( ( RS ) -2 , 3-di ( palmitoyloxy ) -propyl ) -Leu-Pro-Asp-Trp-Leu-Ser-Gly-Phe-Leu-Thr-Gly-Gly-OH , was synthesized by Bachem ( Bubendorf , Switzerland ) . Using LAL assay ( QCL-1000 , Lonza ) , endotoxin was undetectable in original LipoK preparation ( 50 µg/ml ) . Therefore , any contaminating LPS in the synthesized product could be ruled out . Monoclonal Ab to TLR2 was kindly provided by Genentech , and mAb to mannose receptor and DC-SIGN were obtained from BD Biosciences . Parthenolide obtained from Santa-Cruz was used at a concentration of 2 and 5 µM . CD40L ( Pepro Tech ) was used at the concentration of 1 µg/ml , whenever needed . Immature DCs were stimulated with M . leprae and/or LipoK for 48 hours . The expression of cell surface antigens on DCs , were analyzed using FACSCalibur flow cytometer ( BD Biosciences ) . Dead cells were eliminated from the analysis by staining with 7-amino actinomycin D stain . For analysis of cell surface Ag , the following mAb were used: FITC-conjugated mAb against HLA-ABC ( G46-2 . 6 ) , HLA-DR ( L243 ) and CD86 ( FUN-1 ) , purchased from PharMingen , and CD83 ( HB15a , Immunotech ) . The ability of DCs to produce IL-12 on stimulation with either LipoK and/or M . leprae , was assessed . DCs were stimulated with the Ags on day 4 after the start of culture from monocytes . After 24 hours , OptEIA Human IL-12 ( p70 ) ELISA Set ( BD Biosciences ) was used to determine the concentration of IL-12 p70 in the culture supernatant . The ability of M . leprae-infected DCs to stimulate T cells was assessed using an autologous DC-T cell co-culture . CD4+ T cells and CD8+ T cells were purified using respective T cell enrichment Set ( BD IMag ) from freshly thawed PBMCs . The purity of CD4+/CD8+T cells was determined to be more than 95% . The purified responder cells ( 1×105 per well ) were plated in 96-well round-bottom tissue culture plates , and mitomycin C-treated DCs which were pulsed with Ag , were added to give the indicated DC: CD4+ or CD8+ T cell ratio . Supernatants of DC-T cell co-cultures were collected on day 4 , and IFN-γ production was measured by ELISA , using Opt EIA Human IFN-γ ELISA Set ( BD Biosciences ) . In other experiments , Ag-pulsed DCs were treated with mAb to HLA-ABC ( W6/32 ) , HLA-DR ( L243 ) , CD86 ( IT2 . 2 ) or normal mouse IgG . For obtaining naïve T cells , anti-CD45RO mAb ( Dako ) and anti-mouse IgG Ab Dynabeads M-450 ( Invitrogen ) were used to negatively select the cells . Since BCG is compulsory for children in Japan , it is likely that naïve T cells respond to M . leprae antigens , some of which are cross reactive to M . bovis BCG . DCs stimulated with Ags were co-cultured with the CFSE labeled total T cells . CFSE ( Molecular Probes ) was added at the concentration of 1 µM and incubated at 37°C for 10 min and stabilized according to the manufacturers' protocol . A total of 1×106 cells/well were seeded in a 24-well plate at a DC∶T cell ratio of 1∶6 . After 8 days co-culture , cells were co-stained with PE conjugated anti-CD4 mAb and APC conjugated anti-CD8 mAb ( BD Biosciences ) . CFSE signal of gated T cells were analysed . Imaging of cells was performed using laser scanning microscope LSM5-Exciter ( Carl Zeiss ) . DCs grown on a 13-mm coverglass in a 24-well plate , were infected with M . leprae and/or stimulated with LipoK for 48 hours . T cell from the same donor was purified using the Dynal T cell isolation kit , and co-cultured with DCs for additional 3 days , after washing out extracellular bacilli . Cells were fixed in 2% paraformaldehyde , and the bacilli stained with 0 . 01% auramine O as described [22] . Anti-M . leprae membrane ( minus LAM ) polyclonal antibody was kindly provided by Dr . John S . Spencer through the NIH/NIAID Leprosy Research Support ( N01 A1-25469 ) . Fixed cells were blocked with normal human IgG ( 10 µg/ml ) , and stained with the above polyclonal antibody ( 1 µg/ml ) for 30 min in PBS containing 0 . 1% saponin and 0 . 5% BSA , and the secondary antibody used was Alexa Fluor 633-conjugated goat anti-rabbit IgG ( Molecular Probes ) , and images were recorded on fluorescent confocal microscope using a 63× oil objective , 488-nm and 633-nm lasers . Data was processed using the LSM software ZEN 2007 . All bacilli observed were not surface attached as observed by section scanning ( Z-stack Navigation ) . After 7 days co-culture of purified T cells with DC pulsed with M . leprae and/or LipoK , intracellular detection of cytolytic effector molecules was performed . Briefly , GolgiStop ( BD Biosciences ) was added to the media for the last 12 hours of culture . Cells were first surface stained , fixed , permeabilized , and finally stained with FITC conjugated anti-perforin mAb or anti-granzyme B mAb or isotype control IgG2a ( BD Biosciences ) . For the determination of intracellular levels of granulysin , the procedure was followed as for the intracellular stain of perforin , except that the surface stain used was FITC conjugated-CD4 and APC conjugated anti-CD8 mAb ( BD Biosciences ) , and subsequently PE conjugated granulysin ( eBioscience , GmbH , Germany ) was used to determine the percentage of granulysin producing cells . Since M . leprae cannot be cultured in vitro , we measured the viability of the bacilli , by the measurement of radioactive CO2 production from oxidation of palmitic acid as described previously [23] . DCs were infected with M . leprae with or without LipoK , and co-cultured with T cells in some cases . Six days later , cells were harvested and washed 3 times in PBS , and centrifuged , so that M . leprae that might have escaped from the DCs into the media could be eliminated from our assay . Cell lysates were prepared as follows: 0 . 1 N NaOH solution was added to the cells for few minutes and then neutralized with the equal volumes of 0 . 1 N HCl solution . Subsequently , equal volume of 2 times concentrated Middlebrook 7H9 broth supplemented with ADC was added . 14C labeled palmitic acid was added to the lysates of DCs and cultured at 33°C . Seven days later , the amount of 14CO2 evolved and trapped on the filter paper was measured using a Packard 1500 TRI-CARB liquid scintillation analyzer . In a likewise manner , direct effect of M . leprae killing was observed by incubation of the bacilli with 3 µg/ml of granulysin ( R&D systems ) or granzyme B ( Calbiochem ) for a period of 3 days at 33°C , and then 14C labeled palmitic acid was added to determine the viability as described above . The unpaired student's t test was used to find the significance of the two sets of data . Differences were considered as statistically significant if p<0 . 05 . All experiments were performed at least 3 times with different blood donors , unless otherwise stated , and the reproducibility of the experiment was evaluated . In some cases , ANOVA was used for probability calculation .
We investigated the effect of LipoK stimulation on human monocyte derived DCs . All DCs were CD1a positive and CD14 negative [21] . When LipoK was used as a stimulant for immature DCs , maturation of DCs was observed as shown in Fig . 1 . Up-regulation in the expression of CD83 ( maturation marker of DCs ) and CD86 ( co-stimulatory molecule ) was observed in LipoK stimulated DCs , the level of which , was similar to that of M . leprae infected DCs . M . leprae was used at the multiplicity of infection ( MOI ) : 50 in all the experiments . The expression of the CD83 and CD86 molecules was more pronounced when LipoK was used to stimulate M . leprae infected DCs . The expression of HLA-ABC and HLA-DR molecules was not significantly different in LipoK stimulated M . leprae infected DCs from non-infected DCs , after 48 hours . Although , at earlier time points ( 18 hours after stimulation with antigen ) , a higher expression of HLA-ABC and HLA-DR is observed in LipoK stimulated M . leprae infected DCs compared to non-stimulated . Alternatively , when the IL-12 p70 secreted by DCs was measured , increasing dose of LipoK on M . leprae infected DCs produced the cytokine , with maximal cytokine production at LipoK concentration of 0 . 3 µg/ml ( Fig . 2A ) . LipoK alone did not produce statistically significant amounts of IL-12 at the concentration of 0 . 3 µg/ml compared to the non-stimulated DCs . Another TLR-2 agonist , peptidoglycan could produce IL-12 ( data not shown ) , probably due to the heterogeneous nature of the peptidoglycan which contains long peptide linkages . LipoK probably need other protein/peptide molecules to activate IL-12 production in DCs . Also , M . leprae infection alone did not produce IL-12 in DCs . When CD40 ligand ( CD40L ) was used to stimulate M . leprae infected DCs , IL-12 production was negligible . As could be expected , TLR-2 antagonistic Ab completely blocked IL-12 production , whereas mannose receptor Ab did not , suggesting that IL-12 production from LipoK stimulated M . leprae infected DCs was TLR-2 dependent ( Fig . 2B ) . When DCs were pre-treated with parthenolide , which is known to inhibit NF-kB activity [24] , it was found that both 2 µM and 5 µM could significantly inhibit the production of IL-12 in a dose-dependent manner ( Fig . 2C ) , indicating that NF-kB is involved in the IL-12 production from these LipoK stimulated DCs . To investigate the effect of LipoK on T cell responses , purified CD4+ and CD8+ T cells from autologous donors were cultured with activated DCs . IFN-γ release was measured as correlates of T cell activation . When the IFN-γ levels were compared , DCs activated with M . leprae and LipoK produced significantly higher dose of IFN-γ from CD4+ T cells , when compared to that produced by DCs stimulated with M . leprae or LipoK alone , or M . leprae in presence of CD40L ( Fig . 3A ) , at both high ( T∶DC = 20∶1 ) and low ( T∶DC = 40∶1 ) dose of DCs . Note that M . leprae-infection or LipoK-stimulation alone was not efficient in stimulating T cells . Similarly , secretion of IFN-γ was also observed from CD8+ T cells but at lower level compared to that from CD4+ T cells . Again there was significantly high production of IFN-γ from CD8+ T cells co-cultured with LipoK stimulated M . leprae-infected DCs compared to that from CD40L stimulated M . leprae-infected DCs ( Fig . 3A ) . Although the IL-12 p70 production differed in LipoK stimulated M . leprae-infected DCs and CD40L stimulated DCs , no IL-12 production was observed from these mitomycin treated DCs which were co-cultured with T cells . In addition , as shown in Fig . 3B , although normal murine IgG did not affect the T cell stimulating activity of both CD4+ and CD8+ T cells , mAbs to HLA-ABC and HLA-DR , inhibited CD8+ T cells and CD4+ T cell activation of LipoK-stimulated M . leprae-infected DCs' respectively . The results indicated that the activation of these T cells were MHC Class II- and Class I-dependent in CD4+ T cell and CD8+ T cells respectively . The inhibition was comparable to that of inhibition of IFN-γ production by mAb to co-stimulatory molecule CD86 . Proliferation of these LipoK activated CD4+ and CD8+ T cells , was confirmed by the CFSE labeling of T cells . The labeling experiment was preferable because it could measure proliferation of individual T cell subsets even in the presence of the other subsets . M . leprae stimulation of DCs resulted in proliferation of 39 . 7% of total CD4+ T cells , but stimulation with both LipoK and M . leprae resulted in proliferation of 67 . 5% of total CD4+ T cells . LipoK stimulation alone did not induce any significant proliferation of CD4+ T cells ( Fig . 3C ) . The profiles of flow cytometric analyses showed that 25 . 3% of CD8+ T cells proliferated by stimulation with M . leprae alone , but higher number of cells proliferated ( 38 . 9% ) in presence of LipoK stimulus . Subsequently , we examined the response of naïve T cells to LipoK activated DCs . When naïve CD4+ T cells were cultured with DCs activated with M . leprae and LipoK , significantly higher dose of IFN-γ was produced in comparison to those cultured with DCs stimulated with M . leprae alone or LipoK alone . Production of IFN-γ was low from those activated with M . leprae and CD40L ( Fig . 3D ) . It was observed that the IFN-γ production from naïve CD8+ T cells , co-cultured with DCs stimulated with M . leprae and LipoK was meager . When M . bovis BCG was used for infecting DCs , the MOI of the bacilli had to be lowered to almost 1∼10 , because higher MOI ( 50 ) would kill the DCs in in-vitro culture . BCG when infected at MOI:1 produced 156 pg/ml of IFN-γ from CD8 T cells , but when LipoK was used to stimulate BCG infected DCs , the amount of IFN-γ increased to 380 pg/ml , indicating that LipoK could lead to further T cell activation of BCG infected DCs . It is also likely that LipoK stimulation could increase the production of perforin and granulysin in M . tuberculosis infected host cells . To determine whether cytotoxic effect could be induced in highly activated T cells , we analysed the intracellular production of perforin and granzyme B in DC co-culture system with unseparated T cells . As seen in Fig . 4A , 15 . 8% of activated CD8high T cells produced perforin and 24 . 9% produced granzyme B when stimulated with DCs activated with M . leprae and LipoK , in comparison to those co-cultured with DCs activated with M . leprae , showing 1 . 4% of perforin and 1 . 8% of granzyme B-producing T cells . Thus , prominent enhancement of both perforin and granzyme B producing CD8+ T cells was observed . Recently , since CD4+ T cells are also known to possess direct cytotoxic potential [25] , we measured the percentage of CD4+ T cells producing perforin and granzyme B . When LipoK and M . leprae stimulated DCs were co-cultured with T cells , 12 . 7% of CD4high T cells produced perforin and 14 . 6% of those cells produced granzyme B , whereas in presence of M . leprae stimulated DCs , 6 . 6% produced perforin and 8 . 3% produced granzyme B ( Fig . 4B ) . These data indicated that in addition to CD8+ T cells , CD4+ T cells also had the capacity to produce significant amounts of perforin and granzyme B . Nevertheless , the percentage of CD8+ T cells producing these cytolytic proteins was 1 . 2∼1 . 7 fold higher than CD4+ T cell . Then , we examined , whether CD8+ T cells alone without the direct contact with CD4+ could have the same capacity . When CD4+ T cells were allowed to culture in inserts , so that there was no direct contact between CD8+ and CD4+ T cells , there was decreased production of both perforin ( 7 . 3% v/s 15 . 8% ) and granzyme B ( 9 . 5% v/s 24 . 9% ) producing CD8+ T cells ( Fig . 4A ) . So , a direct contact of CD4+ and CD8+ T cells was necessary for sufficient production of cytolytic proteins . When we examined whether exogenous IL-2 could substitute the action of CD4+ T cells , we found that addition of 50 U/ml of IL-2 ( excess amount ) to CD8+ T cells , could produce both perforin and granzyme B equivalent to that of CD8+ T cells co-cultured with LipoK stimulated , M . leprae infected DCs in the presence of CD4+ T cells . However such high levels of IL-2 cannot be produced from host cells , in our experimental setting . The intracellular level of another cytolytic protein , granulysin , was then examined . Enhancement of granulysin producing CD8+ T cells was observed when co-cultured with DCs activated with M . leprae and LipoK . As seen in Fig . 4C , 18 . 9% of activated CD8high T cells and 28 . 4% of activated CD4high T cells produced granulysin when co-cultured with DCs activated with M . leprae and LipoK , in comparison to those co-cultured with DCs activated with M . leprae , ( 1 . 7% of CD8high T cells and 0 . 6% of CD4high T cells ) . To examine the fate of M . leprae in activated DCs , the cells were stained with anti-M . leprae membrane polyclonal antibody . Confocal microscopy revealed rod shaped M . leprae as observed by auramine-O stain , and membrane components seem to be rather localized in the region where M . leprae are present ( Fig . 5 ) . Strickingly , those DCs stimulated with LipoK for 48 hours and co-cultured with T cells for additional 3 days showed membrane staining at the periphery of the DCs ( Fig . 5 arrowheads shown ) , probably due to processing of the bacilli in activated DCs . We determined the viability of M . leprae in DCs after stimulation with LipoK in the presence of autologous CD4+ and CD8+ T cells . Since M . leprae is uncultivable in vitro , the viability of M . leprae in DCs , after co-culture with the T cells for a week , was determined by the radiorespirometric assay . The amount of radioactive CO2 evolved which reflects the rate of 14C-palmitic acid oxidized by M . leprae , was measured by the scintillation counter . No significant reduction in 14CO2 production was observed , from DCs , not co-cultured with T cells , even in the presence of LipoK stimulation ( Fig . 6A ) . But , when the bacilli were recovered from DCs stimulated with LipoK and co-cultured with T cells , 14CO2 production were significantly lower ( p<0 . 001 ) than those recovered from DCs not stimulated with LipoK or T cells . The result indicates that approximately 50% reduction in the viability of M . leprae was observed in LipoK activated DCs and co-cultured with T cells compared to those obtained from DCs not stimulated with LipoK ( Fig . 6B ) , indicating that T cells were essential and LipoK stimulation to DCs , was necessary to kill M . leprae in DCs . To further determine whether the cytolytic granules namely , granulysin and granzyme B could directly kill M . leprae , the bacilli was incubated with human granulysin or granzyme B for a period of 3 days at 33°C . Statistically significant reduction of 14CO2 was observed when the bacilli were incubated with granulysin as well as granzyme B ( Fig . 6C ) .
In the present study we investigated the role of M . leprae-derived synthetic lipopeptide ( LipoK ) , which consists of N-terminal 13 amino acids of the 33-kD M . leprae lipoprotein ( Accession no . ML0603 ) linked to Palmitoyl-Cys ( ( RS ) -2 , 3-di ( palmitoyloxy ) -propyl group in the induction of intracellular killing of M . leprae through immuno-activation . Previously , we observed that the 33-kD lipoprotein and the truncated form of the protein induced the production of IL-12 in human peripheral blood monocytes [4] , [5] . Although human DCs are potent inducers of acquired immune responses , when DCs were exposed to M . leprae , they are inefficient in activating T cells [21] , [26] . It is generally recognized that , stimulation of T cells by intracellular pathogens , such as mycobacteria , is achieved by the coordinated processing of the antigens in the phago-lysosome of APCs and the expression of the antigenic determinants on APCs . Furthermore , CD40-CD40L interaction on immature DCs , are known to contribute to cell mediated responses in leprosy [27] , [28] . In fact , when M . leprae infected DCs were stimulated with CD40L , up-regulation of CD83 and CD86 molecules was observed ( not shown ) . However , we found that CD40L failed to induce the production of IL-12 p70 in M . leprae infected DCs . In contrast to CD40L stimulation , LipoK stimulation on M . leprae infected DCs induced significant production of IL-12 . Further , the expression of CD40 on DCs was not enhanced by stimulating M . leprae infected DCs with LipoK . It was evident that IL-12 inducing ability of these matured DCs was mediated by TLR2 , and not by other receptors such as mannose receptor or DC-SIGN , as observed in DCs exposed to M . tuberculosis or M . bovis BCG [29] , [30] , [31] . The TLR2 antagonistic antibody could almost totally inhibit the IL-12 production from DCs , as well as the T cell activating function of DCs ( not shown ) , probably through blocking of the classical NF-kB pathway . Indeed , parthenolide , one of the major sesquiterpene lactones , known to inhibit NF-kB activity [24] , inhibited the IL-12 production from DCs stimulated with M . leprae and LipoK . Also , IL-12 was efficiently produced when M . leprae was viable and not dead . Thus , although the exact mechanisms remain to be elucidated , some cell surface molecules and secreted components of M . leprae are responsible for the production of IL-12 , which further modulates type 1 T cell responses [32] , [33] . A number of mechanisms are known to be involved in the clearance of intracellular bacteria , such as IFN-γ release , apoptosis induction of the macrophages and anti-microbial activity of CTL [12] , [15] . Production of IFN-γ could boost the ability to kill pathogens in host cells . In fact , it was found that LipoK activated M . leprae infected DCs , highly stimulated both memory CD4+ and CD8+ T cells , as well as naïve CD4+ to produce IFN-γ , and further assisted in the proliferation of both T cell subsets ( Fig . 3 ) . Inhibition of MHC class I and class II molecules on DCs , indicated that the activation of these T cells were MHC class II- and class I-dependent in CD4+ T cell and CD8+ T cells respectively . Further , proteolytic processing of M . leprae antigens was probably enhanced by LipoK treatment of DCs , since incubation with anti-M . leprae membrane Ab showed positive staining at the periphery of DCs , when co-cultured with T cells ( Fig . 5 ) . In addition , preliminary results showed that expression of MHC class I and II molecules on LipoK activated DCs , were elevated in those co-cultured with T cells . Thus , LipoK could probably assist in the processing and presentation of M . leprae antigens , and thereby , highly activate T cells . The other important parameter , for the clearance of mycobacteria from the host cell , is their potential to activate antimicrobial effector mechanisms in human T cells . DCs have been shown to be involved in CTL induction following uptake of antigenic particles [25] , [34] , [35] , [36] . CD8+ T cells co-cultured with LipoK stimulated M . leprae-infected DCs , through CD4+ T cells' help produced increased amount of cytolytic effector molecules: perforin and granzyme B . Adequate production of these cytolytic proteins from CD8+ T cells required direct contact with CD4+ T cells . Recently , there are studies that certain types of CD4+ T cells possess direct cytotoxic potential [25] , [37] , [38] . We observed a portion of CD4high T cells ( activated T cells ) , have the capacity to produce cytotoxic granules . Bastian et al . demonstrated that native M . tuberculosis heterogenous lipopeptides are potent immunogens for primary human T cells , and those T cells were CD4+ and MHC class II restricted , challenging the current concepts that cytotoxic T cells were restricted to CD8+ T cell subset [25] . Another lytic molecule , present in cytotoxic granules of T cells , is granulysin , which is reported to have direct anti-bacterial activity . Reports have shown the ability of T cells to secrete granulysin at the site of M . leprae infection , which provides evidence that anti-microbial activity of granule containing T cells is a mechanism of host defense in leprosy [39] , [40] . We observed that LipoK stimulated , M . leprae infected DCs , highly enhanced the production of granulysin from CD8+ T cells . Unexpectedly , we observed that the percentage of CD4+ T cells producing granulysin was higher than CD8+ T cells . But , this fact was in lines with the earlier data , which showed co-localization of granulysin and CD4+ T cells in tuberculoid leprosy lesions [39] , [40] . Thus , granulysin release by LipoK-mediated activation process , may lead to a direct anti-microbial effector pathway of host defense . These data demonstrated that both CD4+ T cells and CD8+ T cells , contribute to the induction of intracellular killing of M . leprae . These speculations were further supported by the fact that 50% of the phagocytosed bacilli were killed when infected DCs stimulated with LipoK , were co-cultured with T cells . This is the first observation of killing of M . leprae in an ex vivo system using human DCs and T cells . To further provide evidence of the effector mechanism at work during M . leprae killing by CTL , the direct effect of granulysin on M . leprae killing in vitro was analyzed . Results indicated that about 40% of M . leprae was killed by granulysin . Granulysin could probably lyse M . leprae by binding to the lipidic cell wall , through the same mechanism by which M . tuberculosis is destroyed by granulysin . Since , perforin is an essential molecule in the killing of intracellular M . tuberculosis [16] , similar operation may be involved in intracellular M . leprae killing since perforin was effectively produced by T cells in our CTL culture system . On the other hand , direct killing of mycobacteria by granzymes is not known . But the viability of M . leprae was significantly lowered by granzyme B . Since granzyme B is one of the serine proteases that can target cytosolic and nuclear substrates to induce host cell death through mitochondrial perturbation , it may be involved in destroying the cell wall architecture of M . leprae by still unknown mechanism [41] , [42] . The contribution of the cytotoxic granules to killing of bacteria remains to be of interest for further investigation . Together , the results indicate that LipoK could contribute to protective host response against leprosy and eventually kill the bacteria , through the production of perforin , granulysin and granzyme B in T cells .
|
We observed that LipoK ( Mycobacterium leprae lipopeptide with 13 amino acids ) is capable of inducing a good immune response in M . leprae infected human dendritic cells ( DCs ) . These activated DCs had up-regulated expression of costimulatory molecule CD86 as well as CD83 ( well known maturation marker ) on their surface , and secreted IL-12 , which is an important cytokine involved in the host defense against pathogens . Importantly , these mature DCs were capable of further driving type 1 responses by stimulating CD4+ T cells and CD8+ T cells for proliferation and interferon-gamma production . Further , both subsets of T cells were capable of producing cytotoxic granules: granulysin and granzyme B . In vitro experiments proved that these molecules are capable of killing M . leprae directly . It is the first report of the type , which proves that granulysin as well as granzyme B could partially kill M . leprae . LipoK would facilitate in inducing the immune responses in patients' harboring M . leprae .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"immunology",
"biology",
"immune",
"response"
] |
2011
|
A Lipopeptide Facilitate Induction of Mycobacterium leprae Killing in Host Cells
|
The translocation of single-stranded DNA ( ssDNA ) across membranes of two cells is a fundamental biological process occurring in both bacterial conjugation and Agrobacterium pathogenesis . Whereas bacterial conjugation spreads antibiotic resistance , Agrobacterium facilitates efficient interkingdom transfer of ssDNA from its cytoplasm to the host plant cell nucleus . These processes rely on the Type IV secretion system ( T4SS ) , an active multiprotein channel spanning the bacterial inner and outer membranes . T4SSs export specific proteins , among them relaxases , which covalently bind to the 5' end of the translocated ssDNA and mediate ssDNA export . In Agrobacterium tumefaciens , another exported protein—VirE2—enhances ssDNA transfer efficiency 2000-fold . VirE2 binds cooperatively to the transferred ssDNA ( T-DNA ) and forms a compact helical structure , mediating T-DNA import into the host cell nucleus . We demonstrated—using single-molecule techniques—that by cooperatively binding to ssDNA , VirE2 proteins act as a powerful molecular machine . VirE2 actively pulls ssDNA and is capable of working against 50-pN loads without the need for external energy sources . Combining biochemical and cell biology data , we suggest that , in vivo , VirE2 binding to ssDNA allows an efficient import and pulling of ssDNA into the host . These findings provide a new insight into the ssDNA translocation mechanism from the recipient cell perspective . Efficient translocation only relies on the presence of ssDNA binding proteins in the recipient cell that compacts ssDNA upon binding . This facilitated transfer could hence be a more general ssDNA import mechanism also occurring in bacterial conjugation and DNA uptake processes .
Agrobacterium tumefaciens is a Gram-negative pathogenic bacterium able to transfer and integrate up to 150 , 000-bases-long single-stranded DNA ( ssDNA ) into the infected cell nuclear genome [1] . In Agrobacterium pathogenesis , the sequence of ssDNA to be transferred ( T-DNA ) and the genes encoding the virulence ( Vir ) proteins required for transfer of T-DNA into the host are localized on a large plasmid called the tumor-inducing plasmid [2] . Some virulence proteins have a function in the bacterium , namely the 11 VirB proteins and VirD4 , which compose the Type IV secretion system ( T4SS ) machinery . T4SS exports T-DNA and effector proteins out of the bacterium [3–5] . The effectors are proteins , which are synthesized in the bacterium but exert their function in the recipient cell . The export signal of the effector proteins is localized at their C terminus and is recognized by VirD4 [6] . Among the effectors , the relaxase VirD2 binds covalently to the 5′ end of the ssDNA . The combined action of the three NTP-binding/hydrolysing proteins VirB4 , VirB11 , and VirD4 has been proposed to energize the transfer of the proteins and VirD2-T-DNA through the T4SS [7] . How the T-DNA then crosses the plasma membrane of the host remains unknown , but the effector protein VirE2 might be involved . In vitro , VirE2 was shown to form channels , which transport ssDNA , and VirE2 was hence proposed to mediate transfer of T-DNA through the eukaryotic plasma membrane [8–10] . VirE2 is a necessary , multifunctional protein [11] and another important function of VirE2 is to bind cooperatively T-DNA in the host cytosol . The interaction of VirE2 with T-DNA mediates its import into the nucleus . As evidenced by scanning transmission electron microscopy ( STEM ) , the VirE2–ssDNA complex consists of a helical structure in which 19 nucleotides are bound per VirE2 monomer [12] . This structure prevents exonuclease degradation in vitro [13] . Moreover , recent in vitro experiments demonstrate the microtubule-guided transport of such DNA–VirE2 complexes [14] . Using single-molecule technology , we measured the binding properties of VirE2 to ssDNA , and we suggest here that VirE2 binds to ssDNA nucleotides in a zipper-like mode . This property was confirmed biochemically with the ability of the VirE2 protein to bind to a shorter oligonucleotide than its footprint of 19 nucleotides . We also show that cooperative VirE2 binding compacts the ssDNA against high loads ( 50 pN ) , which could , in vivo , help to actively pull the T-DNA into the recipient cell . Using cell biology detection techniques , VirE2 was localized at the plant cell periphery , an ideal localization for VirE2-mediated pulling of the incoming T-DNA . Altogether , a combination of very different techniques allowed the emergence of a completely new view on T-DNA transfer energetics upon translocation into the host plant cell .
Binding of VirE2 to ssDNA was studied using different optical tweezers modes ( Figure 1B , inset ) . First , the VirE2 binding rate was determined at a pre-set force that was kept constant by a feedback system ( force-feedback experiment , [Figure 1B , inset]; VirE2 concentration of 20 μg/ml , 330 nM ) . Polymerization of VirE2 dramatically affected the length of the tethered ssDNA ( Figure 1A ) . From a detailed analysis of the time traces [Figure S1 , showing a plot where the transition takes place; Text S1 , section: Rate of polymerization ( experimental determination ) ] , we found a value of ( 1510 ± 200 ) nm/s ( n = 5 ) for the polymerization rate originating from a single nucleation site ( 5 pN ) . For the 4 , 502-bases-long DNA and taking 19 as the number of nucleotides bound per VirE2 monomers ( as determined by scanning transmission electron microscopy ) [12] , this yields a binding rate of ∼10 VirE2/ms at a VirE2 concentration of 20 μg/ml ( 5 pN ) . These force-feedback experiments were performed at different forces . For forces ≤22 pN , the normalized extension at full polymerization was found to be about 0 . 11 ± 0 . 02 ( n = 21 ) . Compaction also occurred even when the ssDNA was forced to remain in an extended form ( 50 . 5 pN; Figure 1A ) . At this force , the polymerization rate was found to be considerably slower ( ∼ 50 nm/s , Figure S1 ) . Force-feedback experiments performed at high forces ( >22 pN ) yield a normalized extension at full VirE2 coverage of 0 . 66 ± 0 . 05 ( n = 11 ) , much longer than the one observed at low force ( about 0 . 1 ) ( Figure 1A ) . This 6-fold difference in normalized extension ( observed at full coverage ) indicates that VirE2 filaments adopt a different global structural arrangement depending on the preset force . This point will be discussed in details below ( section: Global Structural Arrangement of ssDNA upon VirE2 Binding ) . Force-feedback measurements at different forces allow the force dependence of the polymerization rate k ( f ) ( originating from a single polymerization front , see Figure S1 ) to be determined ( Figure 1B ) . Using the Arrhenius law , k ( f ) is described by k ( f ) = k0 exp ( -<w ( f ) >/kBT ) , where <w ( f ) > is the work produced by VirE2 per locally bound single nucleotide [15] , k0 isthe rate at zero force , kB is Boltzmann's constant , and T is temperature . In a local model , <w ( f ) > is approximated to f ( LSS<cosθ> – LV ) , ( Text S1 , section: Rate of polymerization ( theory ) , and Figures S2 and S3 , showing a detailed analysis of the model ) , where LV ( LSS ) is the DNA base-to base backbone distance in the presence ( absence ) of VirE2 . From structural data , a value of 0 . 7 nm is found for LSS [16] . In a freely jointed chain ( FJC ) model , <cosθ> follows the Langevin formula [17] , yielding an analytical expression for k ( f ) . The local model gives a good description of the experimental data when the base-to-base distance of the VirE2-bound ssDNA LV equals 0 . 41 nm ( Figure 1B and Figure S3 ) . This value ( 0 . 41 nm ) is estimated from electron microscopy ( EM ) studies ( assuming the ssDNA to lie concentrically within the protein helix [18] , Figure 2A ) and is in good agreement with a statistical analysis of the compaction steps ( Figure S4 , probability density function ( PDF ) analysis of the time trace at 50 . 5 pN ) . The good description of the experimental data by the local model suggests that VirE2 monomers bind one nucleotide at a time in a zipper-like motion and that the probability of binding a nucleotide is site-independent , a prerequisite for the local model . Such a model predicts that VirE2 could bind stably to less than 19 nucleotides . Experimental proof was provided by a gel-shift assay ( Figure 2B ) , which demonstrated that VirE2 can bind a 12-bases-long oligonucleotide . Finally , the good description of the force-feedback measurements ( Figure 1B ) by the local model suggests that the base-to-base distance of ssDNA bound to VirE2 is force independent . Standard force-versus-extension curves ( pulling and relaxing the tethered DNA molecule without any feedback; “pull and relax” ) ( Figure 1B , inset ) at lower protein concentration ( 6 μg/ml , 100 nM ) were recorded ( Figure 3A ) . These curves show the progressive compaction of bare ssDNA ( red ) as coverage with VirE2 proteins occurs , up to a state where the filament adopts a stable conformation ( black ) . This final conformation ( for which subsequent pulls did not noticeably change the shape of the force-versus-extension curves ) yields an average compaction factor of 9 . 7 ± 2 . 0 ( n = 15 ) . Previous EM studies have reported a compaction factor of 11 . 9 for a perfect VirE2-ssDNA helical structure ( Text S1 , section: Length reduction upon protein binding ) . This suggests that the final state we observe ( also confirmed by distance-clamp experiments , Figure S6 ) corresponds to a conformation where the VirE2 proteins rearrange into a helix ( Figure 2A ) . As seen in Figure 3A , the final state ( black curve ) is extremely stiff ( as compared to ssDNA ) . Curves recorded at intermediate stages of polymerization ( green , blue ) can be fitted with a FJC model considering the ssDNA compaction factor upon VirE2 binding of 11 . 9 , the persistence length of bare ssDNA and a normalized contour length l = λlssDNA + ( 1 - λ ) /11 . 9 ( 0 < λ < 1 ) , where lssDNA is the normalized contour length of ssDNA ( Figure 3A , gray lines ) . Therefore , partially coated ssDNA-VirE2 filaments ( blue and green curves ) exhibit two domains . First , the flexible , uncoated ssDNA of length λlssDNA , and second the almost nondeformable , fully VirE2-coated domain of length ( 1 - λ ) /11 . 9 . Through the sequential binding and subsequent release of VirE2 proteins from bare ssDNA molecule ( red curve ) , the final helical conformation is obtained ( black curve ) . In Figure 3A , the intermediate state of polymerization ( blue curve , normalized extension of about 0 . 2 ) shows the detachment of a large amount of VirE2 proteins at ∼50 pN ( yielding a decrease in the VirE2 coverage , i . e . , an increase in the fraction of bare ssDNA in the filament from λ = 0 . 27 to λ = 0 . 46 ) . When the force applied to ssDNA was relaxed , VirE2 molecules bound to ssDNA again , achieving a more stable coverage , since almost no VirE2 was driven off upon restretching of the DNA–protein complex up to 70 pN ( green curve ) . These findings correlate nicely with the binding mode of VirE2 . VirE2 is a non–sequence specific ssDNA binding protein , and the interaction of VirE2 with ssDNA ( at a low protein concentration of 6 μg/ml ) leads to multiple nucleation sites . This yields a number of different VirE2-ssDNA helical domains , which might not be in register ( i . e . , yielding a nonperfect helical structure over the whole length of the ssDNA , Figure 3B ) . When the VirE2-ssDNA filament is pulled , short VirE2 domains seem to progressively detach from the ssDNA molecule . When the tension is relaxed , VirE2 proteins bind again . Subsequent pulls yield an increase in the average length of VirE2 helical domains , which then resist higher forces ( green curve ) . The final state therefore corresponds to an extremely stable conformation in which no VirE2 release from the filament is observed when proteins were removed from the fluid chamber . The fully polymerized nucleoprotein complex was unusually stiff ( Figure 3A , black curve ) . From the critical force for buckling |FB| ∼ 3 . 5 pN ( obtained while compressing the filament; Figure 3A , arrow ) , we estimate a persistence length ( |FB|l2/4π2kBT ) [19] of ∼14 μm , about 4 orders of magnitude larger than that of bare ssDNA [20] . Because binding of ssDNA to VirE2 in a zipper-like way requires some initial protein flexibility , the high stiffness measured in the final ( fully covered ) VirE2-ssDNA filaments suggests that VirE2 , the DNA , or both are stiffened by their interaction . A similar increase in stiffness upon DNA binding has been observed for other ssDNA binding proteins such as RecA [21] . As mentioned in the Text S1 ( section: Mechanical properties of ssDNA-VirE2 filaments: Helix model ) , a mechanical model consisting of a pure helical structure gives a value of 110–2 , 200 nm for the persistence length , much lower than reported here . This difference could be attributed to the presence of axial interactions along the protein helix ( reported by EM studies [18] ) that could considerably stiffen the structure ( Figure 2A , circles ) . For force-clamp experiments performed at low forces ( ≤22 pN , Figure 1 ) , the value for the normalized extension at full coverage was estimated to be at 0 . 11 ± 0 . 02 ( value obtained from a total of 21 experiments performed between 2 and 22 pN ) . This value for the normalized extension is in good agreement with that of EM studies for a perfect helical arrangement ( 0 . 084 or 1/11 . 9 [18] ) , suggesting that the helical VirE2-ssDNA structure can even form against loads up to ∼20 pN . This helical conformation was not achieved when performing force-feedback experiments at >22 pN . For these forces and at full coverage , the normalized extension was found to be 0 . 66 ± 0 . 05 ( estimated from a total of eight experiments performed at 30 , 36 , 45 , and 50 . 5 pN ) . This normalized extension corresponds to an average base-to-base distance of ssDNA ( projected along the direction of the applied force ) of 0 . 46 ± 0 . 04 nm ( Figure S5 , showing typical force versus extension curves of both ss- and double-stranded ( ds ) DNA ) , in close agreement with that found from EM studies ( 0 . 41 nm , Figure 2A ) [18] . From this result , we deduce that the rearrangement of the VirE2-ssDNA filament into a helix ( Figure 2A ) cannot proceed against large forces and that the normalized extension reduction observed at forces >22 pN corresponds to the sole binding of VirE2 on ssDNA . The small discrepancy between the expected value and the experimental observation , although significant , can be attributed to the large footprint of VirE2 ( 19 nucleotides ) as well as the possible loss of cooperativity at high forces . The local model ( Figure 1B and Figure S3 ) was shown to give a good description of the force dependence of the rate of polymerization . However , this model only considers the zipper binding mode of VirE2 to ssDNA and does not take into account the rearrangement into a helical structure . This suggests that the helical rearrangement is much faster than the local binding of VirE2 to ssDNA . Hence , the binding of VirE2 to ssDNA is the rate-limiting step of the overall polymerization process and dominates the kinetics . Note finally that we did not observe any compaction of the ssDNA molecule for force-clamp experiments performed at low protein concentrations ( <1 μg/ml ) . This correlates with gel-shift retardation experiments ( Figure S7 ) , which demonstrate that binding of VirE2 to 170-bases-long ssDNA occurs over a small range of protein concentration without intermediate bands . In vivo , the VirE2 protein exerts its role in the plant . It is sufficient to express the VirE2 protein in the plant to restore full virulence: transgenic plants expressing VirE2 allow efficient T-DNA transfection by nearly avirulent virE2-null-Agrobacterium [22] . If the VirE2 proteins accumulate at the periphery of the plant , then the interaction of VirE2 and ssDNA would not only protect the T-DNA from exonuclease degradation but also greatly facilitate the import of the T-DNA thanks to the capability of VirE2 to work against large forces when binding to ssDNA ( see above ) . Because localization of VirE2 protein originating from the bacterium has proven to be extremely challenging and has so far not yielded a usable result , we chose to use the fact that , when VirE2 is expressed in the plant , it is active . Hence , we transiently expressed VirE2HA in tobacco BY-2 cells . VirE2HA is a biologically active fusion ( Figure 4A ) and was used to perform immunofluorescence experiments . Figure 4B demonstrates the localization of VirE2 around the nucleus , at the cell periphery , in cytosolic strands , and in a few cytoplasmic spots . This non-nuclear cytoplasmic localization is supported by VirE2-GFP localization ( also an active fusion protein when expressed in plant cells , S . Gelvin , personal communication ) . On the contrary , β-glucuronidase ( GUS ) -VirE2 fusion protein was reported to localize in the nucleus [22] . This controversial results might be explained by the fact that the GUS-VirE2 fusion protein mimics the conformation of VirE2 when bound to ssDNA and hence get imported into the nucleus ( V . Citovsky , personal communication ) . Also , it is widely accepted by the community that the VirE2-ssDNA complex already forms in the host cytoplasm , allowing subsequent nuclear import of the nucleo-protein complex into the nucleus [2] . Hence , there must be some free VirE2 proteins in the host cytosol , which is consistent with our localization data ( Figure 4B ) .
The Agrobacterium pathogenesis mechanism allows for the efficient transfer of long ssDNA molecules into eukaryotic cells [2] . The VirE2 protein is involved in this process by protecting the ssDNA from nuclease degradation and by mediating nuclear import [2] . Here , based on new experimental findings , we propose that VirE2 is an effector that is transported into the host cytoplasm at an early stage to actively pull the T-DNA into the host and protect it from nuclease degradation from the very first moment it enters the cell . In a first step , a single VirE2 protein binds to T-DNA as it enters the plant cell . This binding , occurring in a zipper-like motion , is mainly limited by thermal fluctuations of T-DNA . In a second step , the fast cooperative binding of VirE2 facilitates the formation of a helical structure and actively pulls T-DNA into the plant cytosol ( Figure 5 ) . This model has indirect assumptions . First , VirE2 and T-DNA should not interact in the bacterium , even though they are both synthetized there . Indeed , in Agrobacterium's cytoplasm , VirD2-T-DNA and VirE2 do not interact , and VirE2 only binds to the T-DNA once it is in the plant cytosol [23] . Second , the VirE2 protein should be present at the site of entry of the T-DNA , namely at the periphery of plant cells . This was evidenced by immunofluorescence experiments ( Figure 4B ) , suggesting that VirE2 is properly localized to assist T-DNA pulling as it enters the plant cytosol . Finally , the interaction between VirE2-bound ssDNA and the rigid microtubule network could provide an anchor point that would facilitate the VirE2 mediated-force transduction at an early stage of the translocation process [14] . According to our model , which identifies VirE2 as an essential factor that pulls T-DNA into the plant cytoplasm , the free energy released upon the formation of the nucleoprotein complex allows VireE2 proteins to work against large forces , which might be required to translocate T-DNA into the host ( see below ) . The production of mechanical energy occurs solely through the free energy gain during the binding of VirE2 to ssDNA without the need for an external source of energy , e . g . , nucleotide hydrolysis . To our knowledge , this is the first time that a glimpse at forces involved in ssDNA translocation into the recipient cell is obtained . Their magnitude compares to forces produced by dsDNA translocating molecular motors ( see , e . g . , [24] ) . Other competing mechanisms might tend to pull the DNA back out of the host cytosol . For instance , during conjugation , pili can retract after binding to the host cell [25] . Moreover , during DNA transfer into the host , the Type IV pilus of Neisseria gonorrhoeae can undergo a series of extension and retraction cycles , generating retraction forces up to a few tens of pN [25] . Thus , binding of a protein to the transferred ssDNA to form a complex that prevents recoiling of the ssDNA in the T4SS by such forces would be a great advantage . Tato et al . have proposed that the coupling protein TrwB of the Escherichia coli R388 conjugative system acts as an ATP-driven ssDNA transporting molecular motor [26] . This analogue to VirD4 is located at the bacterial inner membrane and is thought to pump ssDNA through the Type IV secretion channel . Considering the short persistence length of ssDNA ( ∼0 . 7 nm ) and the large distance between the coupling protein and the host membrane ( at least 30 nm [27] ) , just pushing the flexible ssDNA through the T4SS would be inefficient . Transfer would be facilitated if it were also actively pulled through by VirE2 present in the host . Single-molecule experiments have shown that the “final” VirE2-ssDNA helical filament obtained is a very stable and stiff structure . Washing the complex with buffer without VirE2 protein does not destabilize the complex . But in vivo , the uncoating of the filament is necessary for the integration of the T-DNA into the nuclear genome of the recipient cell . Hence , the question is how the rigid VirE2-ssDNA complex is freed from VirE2 . Indeed the very tight interaction between VirE2 and the ssDNA and between VirE2 molecules seem to need a specific mechanism of degradation to remove the VirE2 protein . It was shown recently that VirE2 is specifically targeted for degradation by the VirF-containing Skp1-Cdc53-cullin-F-box complex for proteolysis [28] . The critical role of proteasomal degradation in Agrobacterium-mediated genetic transformation was also evident from the inhibition of T-DNA expression by a proteasomal inhibitor . In summary , our findings and these data correlate nicely and explain why such a specific degradation mechanism would be needed . The unique mechanism that Agrobacterium exploits to translocate any ssDNA molecule has paved the way for genetic engineering of plants and fungi but also offers novel possibilities for gene transfer into mammalian cells [2] . However , the Agrobacterium pulling mechanism proposed here might be more general . It does not rely on VirE2 but needs the following: ( i ) an ssDNA binding protein compacting ssDNA upon interaction and ( ii ) occurrence of this single-strand binding ( SSB ) activity only in one compartment . In bacterial conjugation and DNA-uptake processes , SSB proteins are also present and might have an important funtion . For instance , the SBB homologs ( YwpH ) accumulate preferentially at the cell poles of B . subtilis [29] . Hence these proteins could be , as is VirE2 , capable of generating a force without external source of energy and pull the ssDNA into the recipient compartment .
VirE2-His6 proteins were expressed in E . coli and purified as described in [30] , with the addition of glycerol ( final concentration 20% w/v ) to the sample buffer ( 50 mM NaH2PO4 , pH 8 , 300 mM NaCl ) before storage of the protein at −80 °C . Two types of DNA handles were prepared and used either for force-feedback ( type A ) or pull and relax ( type B ) experiments . Type A: DNA molecules were prepared by PCR amplification ( Taq DNA Polymerase , Roche , http://www . roche . com ) of the pTYB1 plasmid ( 7 , 477 bp ) [New England Biolabs ( NEB ) , http://www . neb . com] using 5′-Thiol- TGG TTT GTT TGC CGG ATC AAG AGC −3′ and 5′-TCC TAA GCC AAC AAT AGC GTC CCA-3′ as forward and reverse primers , respectively . The 4 , 927-bp PCR fragment was digested with HindIII ( NEB ) . Finally , the main fragment was end-filled with Klenow Exo- ( NEB ) with one dATP and one biotin-14-dGTP ( Invitrogen , http://www . invitrogen . com ) , yielding a 4 , 502-bp-long dsDNA . Type B: DNA molecules were prepared by PCR amplification ( Expand Long Template PCR System , Roche ) of the pPIA plasmid ( 15 , 071 bp ) using 5′-thiol-TAT CGT CGC CGC ACT TAT GAC TGT-3′ and 5′-TAT GTC GAT GTA CAC AAC CGC CGA-3′ as forward and reverse primers , respectively . The resulting 14 , 107-bp PCR fragment was digested with EagI ( NEB ) . After digestion , the longest fragment ( 13 , 883 bp ) was end-filled with Klenow Exo- ( NEB ) with two dGTPs and two biotin-14-dCTPs ( Invitrogen ) . DNA molecules were covalently coupled to 2 . 17-μm amino-modified beads ( Spherotech , http://www . spherotech . com ) using sulfo-SMCC ( Sigma ) as a cross-linker [21] . The experimental apparatus for optical tweezers experiments has been described [31] . DNA beads were trapped by the laser and the free biotinylated DNA end was attached to a 2 . 20-μm streptavidin bead ( Spherotech ) , which was held by suction on a micropipette . The bead-to-bead distance was determined from both the movement of the micropipette ( controlled with a closed-loop piezoelectric element ) and the deflection of the laser producing the optical trap ( monitored by a two-dimensional , position-sensitive detector ) . The pipette bead was moved away from the trapped bead at a constant velocity of 0 . 8 nm/ms . At this rate , complete force-extension curves were recorded within a few seconds . Forces were obtained from the direct measurement of the change in light momentum flux [31] . All signals ( distance , force ) were low-pass filtered at 159 Hz . Force curves were measured in assembly buffer ( 50 mM NaH2PO4 , pH 8 . 0 , 150 mM NaCl , and 5% w/v glycerol ) . To obtain ssDNA molecules , dsDNA was exposed to 150 mM NaOH . Subsequently , the chamber was rinsed with assembly buffer and VirE2 proteins were injected . Prior to injection , proteins were centrifuged at 14 , 000g for 20 min . The supernatant was kept at 4 °C and injected at a protein concentration ranging from 6 to 20 μg/ml in assembly buffer . Forces were monitored in a constant VirE2 flow . Experiments were performed at room temperature . The force-clamp mode uses a digital “P” ( proportional gain ) -like feedback that runs at 150 Hz ( taking into account the time for the acquisition , some CPU time for the calculations , and communication with the different instruments ) . In details , the feedback works as follows: if the change in force |Δf| is smaller than 0 . 7 pN , we do not feedback at all; for larger changes in force , the pipette if moved by ±5 nm ( |Δf|≤ 2 pN ) or Δf × 7 nm ( |Δf| > 2 pN ) . During a force-clamp operation , the data are only recorded and plotted when |Δf|≤ 0 . 7 pN . In that case , an additional ∼6 ms is required to process the different routines of the software . The oligonucleotide 5′-ACA TTG ACC CCT-3′ was radioactively labeled at the 5′ terminus by incubating 100 pmoles of oligonucleotide with 20 units of polynucleotide kinase ( Roche ) and 30 mCi of 32P γ-ATP ( Pharmacia ) for 30 min at 37 °C . The amount of incorporated radioactivity was measured using a TRI-CARD 2100 TR Liquid Scintillation Analyzer . Five pmoles ( 5 , 000 cpm ) of the 12-nucleotides-long 32P 5′-labeled oligonucleotide were added to the VirE2 protein in 50mM NaH2PO4 , pH 8 , 300 mM NaCl , and the reaction was incubated on ice for 1 h . The mixture was then loaded on a native , 10% acrylamide gel and run in 0 . 25× TBE at 100 mV for 2 h at 4 °C . The gel was dried and exposed on a Kodak x-ray film . See Figure 2B . See Figure S7 and [30] for details . To clone VirE2H6 into pCAMBIAmod [32] , the entire open reading frame ( ORF ) of pET-VirE2H6 [8] was amplified by PCR at 43 °C . A BamHI site was added at the 5′ terminus using the primer 5′-CGC GGA TCC TTT AAC TTT AAG AAG GAG ATA TAC-3′ and a PstI site was added to the 3′ terminus using the primer 5′-AAG ACG TCC TCA GTG ATG GTG ATG GTG ATG AAA GC-3′ . The PCR product was cloned into pGEMT ( Promega ) , cut with BamHI and PstI , and cloned into pCAMBIAmod that had been digested with the same enzymes , resulting in pCAMBIA-VirE2H6 . pCAMBIA-VirE2HA was generated by digesting pCAMBIAmod with BamHI and XbaI and inserting the VirE2HA gene extracted from pcDNA3 . 1-VirE2HA ( see below ) with the same enzymes . Cloning of pcDNA3 . 1-VirE2HA was performed using the primers 5′-TCA TGG ATC CAC CAC CAT GGA TCT TTC TGG CAA TGA GAA A-3′ ( adding a BamHI site and the Kozak sequence on the 5′ of VirE2 ) and 5′-ACT CTC TAG ATC AAG CGT AAT CTG GAA CAT CGT ATG GGT AAA AGC TGT TGC TTT GGC T-3′ ( adding an hemaglutinin ( HA ) tag to the 3′ terminus of VirE2 as well as an XbaI site ) were used to generate VirE2HA by PCR amplification of the VirE2 gene using pET- VirE2H6 as a template [8] . The PCR product was cut with BamHI/XbaI and ligated into pcDNA 3 . 1 ( Invitrogen ) cut with the same enzymes . The resulting construct was named pcDNA3 . 1-VirE2HA . For production of transgenic tobacco plants expressing VirE2 or mutants and to prevent expression of VirE2H6 and VirE2HA in Agrobacterium , an intron of potato ST-LSI [33] was inserted into pCAMBIAmod VirE2H6 and VirE2HA as a BamHI/BglII fragment . The resultant plasmids were named pCAMBIAmod VirE2H6-int and VirE2HA-int . The plasmids were subsequently electroporated into electrocompetent Agrobacterium strain GV1301 ( pPM6000 ) cells using a GenePulser ( Biorad ) at 2 . 5 kV , 200 Ω , 25 μFd . Transgenic plants expressing VirE2H6 , VirE2HA were obtained by transforming tobacco ( SR1 ) leaf discs with Agrobacterium GV1301 ( pPM6000 , pCAMBIAmodVirE2H6-int/ VirE2HA-int ) . Control plants were generated by transformation with the empty vector pCAMBIAmod . The selection was performed on Murashig and Skoog ( MS ) medium supplemented with BAP ( 4 μM ) , naphthalene acetic acid ( NAA ) ( 0 . 5 μM ) , cefotaxime ( 500 mg/l ) , timentin ( 150 mg/l ) , and hygromycin ( 20 mg/l ) . Individual plants were regenerated , and five plants from each category were transferred to soil for seed production . WT-VirE2 expressing plants were obtained from the laboratory of Andrew Binns [34] . Seeds from transgenic plants ( VirE2HA , VirE2H6 ) were sterilized and allowed to germinate on MS medium supplemented with hygromycin ( 50 mg/l ) . Fourteen-day-old seedlings were infected with diluted Agrobacterium GV1301 ( pPM6000E , pCAMBIA 2201; Agrobacterium strain where the virE2 gene has been deleted ) , cocultivated for 48 h to an optical density of 1 , followed by extensive washing with MS medium [35] . For the last wash the medium was supplemented with timentin ( 150 mg/l ) . The histochemical GUS staining was performed as described [35] . Virulence was quantified as GUS positive spots per 100 seedlings . ssDNA fragments ( M13 ) were incubated with VirE2 as described in [30] ( Figure 4B ) . Tobacco BY-2 cells were plasmolyzed on MS-agar plates with 0 . 25 M mannitol/sorbitol ( Merck ) for 3 h . DNA of pCAMBIA-VirE2HA and pCAMBIA-GFP was precipitated on 1-μm-diameter gold particles ( Biorad ) . The particles coated with DNA were bombarded on the plasmolyzed BY-2 cells with a PDS-1000/He Biolistic Particle Delivery System ( Biorad ) . VirE2HA protein was transiently coexpressed with green fluorescent protein ( GFP ) after particle bombardment of tobacco BY-2 cells with plasmids pCAMBIA-VirE2HA and pCAMBIA-GFP . GFP was used as a positive marker for transformation . After 16 h recovery , the cells were fixed for 1 h in 3 . 7% paraformaldehyde ( Sigma ) in MSB/Gly buffer ( 50 mM Pipes , pH 6 . 9 , 5 mM EGTA , 1 mM MgCl2 , 2% glycerol ) . The cells were then washed three times with MSB/Gly buffer and deposited on polylysine-coated slides ( polylysine L , Sigma ) . The cell wall was digested for 5 min with the following mix of enzymes from Yakult Honsha ( Pectolyase 0 . 02% , Macerozyme 0 . 1% and Caylase 0 . 3% ) diluted-10 fold in digestion buffer ( 25 mM MES , pH 5 . 5 , 8 mM CaCl2 , and 600 mM Mannitol ) . The cells were permeabilized with 0 . 1% Triton ( Merck ) in PBS ( phosphate-buffered saline ) for 5 min . Unspecific binding of antibody was prevented by incubation of the cells with 5% normal goat serum ( Calbiochem ) . The rat monoclonal anti-HA antibody ( Boehringer ) was diluted 1:100 and the reaction carried out overnight at 4 °C . After washing the cells in PBS , the secondary antibody ( goat anti-rabbit TRITC , Jackson Immuno Research Laboratories ) , was added at 1:30 dilution for 1 h at room temperature . DAPI ( 4′ , 6-diamidino-2-phenylindole , Calbiochem ) , a nucleic acid stain , was added to the cells at 1 mM concentration and incubated for 5 min . Following a PBS wash , fading of the fluorescent signal was minimized by fixing the cells in Vectashield ( Vector Laboratories ) . The cells were observed using a Leica DMRD fluorescence microscope , at 430 nm for DAPI , 488 nm for GFP , and 543 nm for rhodamine . Signals were recorded sequentially using PL APO x63 / 1 . 32 oil / PH3 */ 0 . 17/ D oil immersion objectives equipped with a filter for Nomarski . The VISIOLAB 200 program and a Sony 3CCD color video camera “Power HAD” were used for image processing .
|
The importation of genetic material into cells is a common and fundamental mechanism occurring in bacterial conjugation , DNA uptake , and Agrobacterium plant infection and is , for instance , responsible for antibiotic resistance spread . Previous studies suggested that this process relied only on the activity of complex molecular machines pumping the single-stranded DNA ( ssDNA ) into the recipient cell . Here , we show that proteins provided by the pathogenic organism and translocated prior to the arrival of ssDNA into the recipient cell also play a fundamental role . These proteins not only bind to ssDNA to protect it but also rearrange ssDNA into a compact helix , thus generating a contractile force that pulls the DNA into the host . Interestingly , the production of mechanical energy occurs solely through the free-energy gain during the binding of VirE2 to ssDNA without the need for an external source of energy , such as nucleotide hydrolysis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"microbiology",
"biophysics"
] |
2008
|
VirE2: A Unique ssDNA-Compacting Molecular Machine
|
Recognition of intracellular pathogenic bacteria by members of the nucleotide-binding domain and leucine-rich repeat containing ( NLR ) family triggers immune responses against bacterial infection . A major response induced by several Gram-negative bacteria is the activation of caspase-1 via the Nlrc4 inflammasome . Upon activation , caspase-1 regulates the processing of proIL-1β and proIL-18 leading to the release of mature IL-1β and IL-18 , and induction of pyroptosis . The activation of the Nlrc4 inflammasome requires the presence of an intact type III or IV secretion system that mediates the translocation of small amounts of flagellin or PrgJ-like rod proteins into the host cytosol to induce Nlrc4 activation . Using the Salmonella system , it was shown that Naip2 and Naip5 link flagellin and the rod protein PrgJ , respectively , to Nlrc4 . Furthermore , phosphorylation of Nlrc4 at Ser533 by Pkcδ was found to be critical for the activation of the Nlrc4 inflammasome . Here , we show that Naip2 recognizes the Shigella T3SS inner rod protein MxiI and induces Nlrc4 inflammasome activation . The expression of MxiI in primary macrophages was sufficient to induce pyroptosis and IL-1β release , which were prevented in macrophages deficient in Nlrc4 . In the presence of MxiI or Shigella infection , MxiI associated with Naip2 , and Naip2 interacted with Nlrc4 . siRNA-mediated knockdown of Naip2 , but not Naip5 , inhibited Shigella-induced caspase-1 activation , IL-1β maturation and Asc pyroptosome formation . Notably , the Pkcδ kinase was dispensable for caspase-1 activation and secretion of IL-1β induced by Shigella or Salmonella infection . These results indicate that activation of caspase-1 by Shigella is triggered by the rod protein MxiI that interacts with Naip2 to induce activation of the Nlrc4 inflammasome independently of the Pkcδ kinase .
Recognition of intracellular pathogenic bacteria by members of the nucleotide-binding domain and leucine-rich repeat containing ( NLR ) family triggers immune responses against bacterial infection [1] , [2] . A major response against several pathogenic Gram-negative bacteria , including Salmonella , Legionella , and Shigella is the activation of caspase-1 via Nlrc4 in macrophages [1] , [3] . Upon bacterial stimulation , Nlrc4 mediates the formation of a multi-protein complex termed the inflammasome that induces the activation of caspase-1 leading to the proteolytic maturation of pro-IL-1β and pro-IL-18 as well as the induction of pyroptotic cell death in macrophages [4]–[6] . Many Gram-negative bacteria encode a type III secretion system ( T3SS ) with conserved structural features that promote virulence by injecting bacterial effector proteins directly into the cytosol of host cells [7] , [8] . In macrophages infected with Salmonella , the cytosolic delivery of flagellin or the bacterial rod protein PrgJ through the T3SS is recognized by Nlrc4 leading to inflammasome activation [9] . Recently , Naips ( NLR family , apoptosis inhibitory proteins ) have been shown to act as adaptor molecules that connect flagellin or the bacterial rod protein PrgJ to Nlrc4 [10] , [11] . Specifically , Naip5 and Naip6 associate with flagellin to promote Nlrc4 oligomerization and inflammasome activation , whereas Naip2 links PrgJ to Nlrc4 [10]–[12] . These findings suggest a model in which certain Naips specifically recognize flagellin or PrgJ to mediate Nlrc4 inflammasome activation . Recent studies , however , have revealed that the activation of Nlrc4 is more complex in that phosphorylation of Nlrc4 at Ser533 was found to be critical for the activation of the inflammasome [13] . Furthermore , it was suggested that Pkcδ is the major Nlrc4 kinase responsible for Nlrc4 phosphorylation and inflammasome activation [13] . Shigella are non-flagellated bacterial pathogens that contain highly evolved invasion systems that enable them to invade host cells and colonize the epithelium of the large intestine , which ultimately leads to a severe form of colitis called bacillary dysentery [14] . After uptake of Shigella by intestinal macrophages , the bacterium delivers a subset of effector proteins via the T3SS apparatus into the host cytosol [7] , [8] , [15] . The inner rod of the T3SS needle complex forms a conduit for protein transport through the periplasm which is assembled by the polymerization of PrgJ in Salmonella and its homologue MxiI in Shigella [16] , [17] . Because of the homology of Salmonella PrgJ with Shigella MxiI , it can be predicted that Shigella induces activation of Nlrc4 via the sensing of MxiI by host macrophages . Consistent with this notion , the T3SS of Shigella is required to induce IL-1β secretion and pyroptosis via the Nlrc4 inflammasome [18] . Furthermore , ectopic expression of MxiI reduced the viability of macrophages and this was inhibited in the absence of Nlrc4 [9] . However , the mechanism by which Shigella MxiI induces activation of the Nlrc4 inflammasome remains unknown . In this study , we provide evidence that MxiI mediates the activation of the Nlrc4 inflammasome through interactions with Naip2 . Furthermore , we demonstrate that Naip2 , but not Naip5 , is critical for the interaction of MxiI with Nlrc4 and the activation of the inflammasome in macrophages infected with Shigella . Finally , we show that Pkcδ is dispensable for Nlrc4 activation .
In the case of flagellated pathogenic bacteria , flagellin is a major and potent stimulator of the Nlrc4 inflammasome . In addition , Salmonella T3SS rod protein PrgJ is sensed by Nlrc4 to activate caspase-1 . Because Shigella are unflagellated bacteria , we hypothesized that the Shigella T3SS rod protein MxiI , a homologue of Salmonella PrgJ , induces the activation of the Nlrc4 inflammasome . To test this hypothesis , we expressed MxiI in wild-type ( WT ) and Nlrc4-deficient bone marrow-derived macrophages ( BMDM ) using a MSCV-IRES-GFP retroviral vector and assessed cell viability by the numbers of viable green fluorescence protein ( GFP ) -positive cells . After overnight culture , the viability of WT macrophages was dramatically decreased by MxiI-GFP expression when compared to expression of GFP ( Figure 1A ) . Importantly , the decrease in cell viability was inhibited in Nlrc4−/− macrophages ( Figure 1A ) . Consistently , expression of MxiI-GFP , but not GFP , induced the release of IL-1β in WT macrophages , which was abolished in macrophages lacking Nlrc4 ( Figure 1B ) . These results indicate that expression of MxiI induces the activation of the Nlrc4 inflammasome . We next tested whether the rod protein MxiI interacts with Naip2 or Naip5 in macrophages . Because expression of MxiI in macrophages causes cell death ( Figure 1A ) , we used macrophages from caspase-1-deficient mice to assess the interaction of MxiI with Naip proteins by immunoprecipitation . In these experiments , we expressed T7-tagged MxiI in the presence of HA-tagged Naip2 , HA-tagged Naip5 or control plasmid . Immunoprecipitation analysis showed that MxiI associated with Naip2 , but much less with Naip5 as revealed by immunoblotting with anti-HA antibody ( Figure 2A ) . Next , we investigated the interaction between Nlrc4 and Naip2 in Shigella-infected macrophages . To assess this , we expressed T7-tagged Nlrc4 and HA-tagged Naip2 or Naip5 , or control empty vector in uninfected or caspase-1-deficient macrophages infected with WT or an isogenic Shigella strain deficient in the T3SS ( S325 ) . Immunoprecipitation analysis revealed that Naip2 interacts with Nlrc4 in macrophages infected with WT Shigella ( Figure 2B ) . However , Naip2 did not associate with Nlrc4 in uninfected macrophages or macrophages infected with the mutant bacterium lacking a functional T3SS that are unable to release MxiI into the host cytosol ( Figure 2B ) . Furthermore , infection with Shigella preferentially promoted the interaction of Nlrc4 with Naip2 relative to Naip5 ( Figure 2B ) . MxiI is secreted into the culture medium by Shigella which relies on the presence of a functional T3SS [19]–[21] . Therefore , MxiI is presumably leaked into the host cytosol via the T3SS to activate Nlrc4 , as it was suggested for Salmonella PrgJ [22] , [23] . Therefore , we next asked whether expression of MxiI promotes the association of Naip2 with endogenous Nlrc4 in uninfected macrophages . Immunoprecipitation experiments showed that expression of MxiI induced the interaction of Naip2 with endogenous Nlrc4 ( Figure 2C ) . Collectively , these results indicate that MxiI interacts preferentially with Naip2 and promotes the interaction between Naip2 and Nlrc4 . We next performed additional studies to verify that Shigella infection promoted the activation of Nlrc4 via Naip2 . To confirm the preferential effect of Naip2 on Nlrc4 activation , we performed reconstitution experiments by expressing Nlrc4 , Asc , caspase-1 , pro-IL-1β and Naip2 or Naip5 in 293T cells . One day after transfection , cells were infected with WT or T3SS-deficient Shigella for 3 hrs and inflammasome activation was analyzed by immunoblotting with an antibody specific for mature IL-1β p17 . In the absence of exogenous Naip2 or Naip5 , infection with WT Shigella enhanced the processing of pro-IL-1β into IL-1β p17 ( Figure S1A ) . The formation of IL-1β p17 was further enhanced by Naip2 , but inhibited by Naip5 in Shigella-infected cells ( Figure S1A ) . In this reconstitution system , the enhancement of IL-1β p17 formation by Naip2 in cells infected with WT Shigella required Nlrc4 , Asc and caspase-1 ( Figure S1B ) . Shigella infection stimulates Nlrc4- and Asc-dependent inflammasome activation in macrophages [18] . However , Shigella was also shown to induce macrophage cell death via Nlrp3 after 2–6 hrs of infection at a bacteria/macrophage ratio of 50∶1 [24] . To verify these seemingly contradictory results , we reassessed the role of Asc , Nlrc4 and Nlrp3 in Shigella-induced caspase-1 activation . In these experiments , LPS-primed BMDM were infected with the Shigella WT or S325 ( T3SS-deficient mutant ) at a bacteria/macrophage ratio of 10∶1 for 30 min . As expected , WT , but not mutant Shigella , induced processing of procaspase-1 into the p20 subunit of caspase-1 ( Figure S2A ) . The inability of the mutant bacterium to activate caspase-1 could not be explained by reduced uptake by macrophages ( Figure S3 ) . Importantly , caspase-1 activation , IL-1β release , and pyroptosis required Nlrc4 and Asc , but not Nlrp3 ( Figure S2A–C ) . Because previous studies showed that Asc was not required for pyroptosis induced by Shigella in BMDM differentiated for 5 days [18] , we assessed cell death induced by Shigella in BMDM differentiated for 3 , 4 and 5 days in culture ( Figure S4 ) . Consistent with previous studies [18] , Asc was not required for pyroptosis in macrophages differentiated for 5 days ( Figure S4 ) . In macrophages differentiated for 3 or 4 days , however , cell death induced by Shigella was enhanced in WT macrophages and impaired in Asc-deficient macrophages ( Figure S2C and S4 ) which is in line with the results presented in Figure S2C . Next , we investigated the role of Naip2 and Naip5 in caspase-1 activation induced by Shigella . We used siRNA-mediated knockdown to reduce the expression of Naip2 and Naip5 in macrophages ( Figure 3A ) . Notably , caspase-1 activation induced by Shigella was attenuated by inhibiting the expression of Naip2 , but not Naip5 ( Figure 3B ) . Importantly , the ability of individual siRNA to inhibit caspase-1 activation correlated with reduction of Naip2 expression ( Figure 3A , B ) . In addition , knockdown of Naip2 , but not Naip5 , reduced the release of IL-1β and IL-18 induced by Shigella infection at 1 or 2 hrs post-infection ( Figure 3B , C ) . In control experiments , knockdown of Naip2 did not affect the production of IL-6 or CXCL2 in macrophages infected with WT or S325 mutant Shigella ( Figure 3C ) . These results suggest that Shigella induces Nlrc4-dependent inflammasome activation via Naip2 in macrophages . The Asc pyroptosome is a molecular platform that is thought to be important for the recruitment and activation of caspase-1 [25]–[27] . Infection of macrophages with WT , but not T3SS-deficient , Shigella induced the formation of the Asc pyroptosome which was detected in the cell cytoplasm by staining with an antibody that recognizes Asc ( Figure 4A , B ) . The Asc pyroptosome induced by Shigella infection co-localized with FLICA staining that labels activated caspase-1 ( Figure 4A ) . Importantly , knockdown of Naip2 by siRNA reduced Asc pyroptosome formation whereas Naip5 did not ( Figure 4C , D ) . To provide direct biochemical evidence that the Asc pyroptosome is formed , we cross-linked the insoluble Asc protein complexes from Shigella or Salmonella infected macrophages and subjected them to immunoblotting with anti-Asc antibody . Immunoblotting analysis revealed that infection with WT Shigella or Salmonella induces prominent Asc dimer formation in WT , but not Asc-deficient macrophages ( Figure 5A , upper panel ) . The induction of Asc dimers correlated with IL-1β release in culture supernatants ( Figure 5A , lower panel ) . In contrast , Shigella deficient in T3SS and the fliA-deficient Salmonella mutant were impaired in the induction of Asc dimer formation ( Figure 5A ) . Notably , expression of MxiI was sufficient to induce the formation of Asc dimers in caspase-1-deficient macrophages in the absence of Shigella infection ( Figure 5B ) . Furthermore , knockdown of Naip2 by siRNA , but not Naip5 , inhibited Asc dimer formation ( Figure 5C ) . These results indicate that Shigella MxiI and Naip2 are important in Asc pyroptosome formation which is associated with inflammasome activation . Recent studies reported that Nlrc4 phosphorylation by Pkcδ is critical for inflammasome activation induced by Salmonella infection [13] . Thus , we assessed whether inflammasome activation caused by Shigella infection also requires Pkcδ . In these experiments , LPS-primed BMDM from WT and Pkcδ-deficient mice were infected with WT or S325 ( T3SS-deficient mutant ) Shigella , and IL-1β release was evaluated at different time points and bacterial/macrophage ratios after infection . As expected , expression of Pkcδ was induced by LPS stimulation in WT , but not Pkcδ-deficient macrophages ( Figure 6A ) . Importantly , Pkcδ was not required for IL-1β secretion induced by Shigella or Salmonella ( Figure 6B–D ) . In fact , Pkcδ deficiency enhanced IL-1β secretion in response to Shigella and Salmonella infection ( Figure 6B–D ) . Furthermore , Pkcδ-deficient macrophages produced higher amounts of IL-1α and CCL5 , but not CXCL2 than WT macrophages in response to infection ( Figure 6D ) . The increased production of cytokines in Pkcδ-deficient macrophages was not associated with enhanced NF-κB or MAPK activation after Shigella infection ( Figure S5 ) . Notably , induction of apoptosis in Shigella-infected macrophages was inhibited in macrophages deficient in Pkcδ ( Figure S5 ) . Furthermore , treatment with z-DEVD-fmk , a cell permeable caspase-3 inhibitor , increased the production of IL-1β in WT macrophages infected with Shigella ( Figure S5 ) , suggesting that increased production of IL-1β in Pkcδ-deficient macrophages is mediated , at least in part , by inhibition of apoptosis in Shigella-infected macrophages . Importantly , caspase-1 activation induced by Shigella or Salmonella was unimpaired in macrophages deficient in Pkcδ ( Figure 6E ) , whereas it was abolished in macrophages deficient in Nlrc4 ( Figure 6E ) . These results indicate that Pkcδ is not essential for inflammasome activation induced by Shigella or Salmonella infection .
The intracellular sensing of flagellin is the major trigger for the activation of the Nlrc4 inflammasome in macrophages infected with Salmonella [4] . Because Shigella is non-flagellated , the current studies were aimed at understanding the mechanism by which Shigella induces the activation of Nlrc4 in macrophages . We show here that Shigella induces the activation of the Nlrc4 inflammasome through MxiI , an inner rod protein of the T3SS . MxiI associated with Naip2 and was sufficient to induce Nlrc4-dependent IL-1β secretion and the interaction with Nlrc4 . Importantly , inhibition of Naip2 expression impaired the activation of the Nlrc4 inflammasome and IL-1β/IL-18 release in Shigella-infected macrophages . Because IL-1β secretion induced by Shigella was not abolished by Naip2 knockdown , it is possible that Shigella also activates another inflammasome pathway that is minor and only unmasked by the inhibition of the Naip2-Nlrc4 pathway . Alternatively , it is possible that the partial inhibition of IL-1β secretion reflects residual Naip2 protein expression in macrophages . Our work is consistent with a model in which the T3SS inner rod proteins including PrgJ in Salmonella and MxiI in Shigella are recognized by Naip2 and this interaction leads to the recruitment and activation of Nlrc4 . Consistent with this model , we show that expression of MxiI promotes the association of Naip2 with Nlrc4 and induces the oligomerization of Asc in macrophages . Furthermore , WT , but not T3SS-deficient Shigella , enhances the association of Naip2 and Nlrc4 in macrophages . The failure of mutant Shigella to induce the interaction between Naip2 and Nlrc4 is presumably explained by the inability of the T3SS mutant to release MxiI into the host cytosol . A measure of inflammasome activation is the formation of Asc oligomers [25]–[27] . Importantly , Asc oligomerization induced by MxiI was observed in caspase-1-deficient macrophages , indicating that this critical event is not a secondary event of caspase-1 activation . MxiI is composed of 97 amino acids and is predicted to be a soluble protein using publically available tools ( http://www . psort . org/psortb ) . It has been shown that MxiI is secreted into the culture medium by Shigella in a T3SS dependent manner [19]–[21] . Thus , as it was suggested for Salmonella PrgJ [9] , [22] , we propose that small amounts of MxiI are leaked into the host cytosol via the T3SS during Shigella infection to induce the activation of Nlcr4 . Recent studies showed that Nlrc4 phosphorylation was induced by Salmonella and was found to be critical for inflammasome activation [13] . Furthermore , it was proposed that Pkcδ was the major kinase responsible for phosphorylation of Nlrc4 [13] . In contrast to the latter finding , we found that IL-β secretion and caspase-1 activation induced by Shigella and Salmonella infection were not impaired in Pkcδ-deficient macrophages . Notably , the production of several cytokines including IL-1β was enhanced in infected Pkcδ-deficient macrophages . A possible mechanism to account for the enhanced production of cytokines in Pkcδ-deficient macrophages is the observation that Pkcδ regulates phagosomal production of ROS [28] which is known to inhibit pro-inflammatory responses including cytokine production [29] . However , we did not observe enhanced NF-κB or MAPK activation in Pkcδ-deficient macrophages infected with Shigella . Pkcδ has been shown to regulate the induction of apoptosis [30]–[32] . Consistently , apoptosis induced by Shigella infection was impaired in Pkcδ-deficient macrophages and treatment with a caspase-3 inhibitor enhanced IL-1β secretion in WT macrophages . These results suggest that the increased production of cytokines observed in Pkcδ-deficient macrophages might be due , at least in part , to suppression of apoptosis in infected macrophages . Regardless of the mechanism involved , our results clearly show that caspase-1 activation induced by Shigella or Salmonella infection is not impaired in Pkcδ-deficient macrophages . We do not have a clear explanation for the difference in results between our studies and previous results by Qu et al . These authors showed that in addition to Pkcδ , Pak2 was capable of phosphorylating Nlrc4 at the critical Ser533 , although the results suggested that Pak2 was a minor Nlrc4-phosphorylating kinase [13] . Thus , it is conceivable that the difference in results could be explained by kinase redundancy and subtle variation in the expression of Nlrc4-phosphorylating kinases in different macrophage preparations . Regardless of the explanation , findings within this investigation clearly show that Pkcδ is dispensable for Nlrc4 activation . Thus , our results challenge the notion that Pkcδ is critical for inflammasome activation and indicate that further work is needed to understand the mechanism and role of Nlrc4 phosphorylation in inflammasome activation . Shigella MxiI associates with Naip2 to induce the interaction of Naip2 with Nlrc4 , which presumably leads to Nlrc4 oligomerization and inflammasome activation . In the Salmonella system , cytosolic flagellin binds to Naip5 and induces the association of Naip5 with Nlrc4 [10]–[12] . Reconstitution experiments with purified flagellin , Naip5 and Nlrc4 revealed that these components are sufficient to induce the formation of a disk-like complex composed of 11 or 12 proteins including Nlrc4 and Naip5 , although the exact ratio of Naip5 and Nlrc4 in the complex remains unclear [12] . Based on the latter observations , we suggest that Shigella MxiI induces the oligomerization of Nlrc4 via their interaction with Naip2 . Consistent with this model , we found that MxiI induced the interaction of Naip2 with Nlrc4 and the oligomerization of Asc . Furthermore , Naip2 , but not Naip5 , was critical for caspase-1 activation , pyroptosome formation , Asc oligomerization and IL-1β secretion . Collectively , these results support a model in which distinct Naip family members act as sensors of flagellin and T3SS inner rod proteins and oligomerized Nlrc4 provides a platform for the recruitment and activation of caspase-1 . While Naip2 knockdown reduced inflammasome activation , Naip5 knockdown had the opposite effect in response to Shigella infection . Although further work is needed to understand the role of Naip5 , one possibility is that there is competition between Naip2 and Naip5 protein complexes and inhibition of Naip5 enhances the Naip2-Nlrc4 inflammasome pathway . Nlrc4 and caspase-1 contain CARD domains and they could interact directly via homotypic CARD-CARD interactions . However , the adaptor Asc is essential for the activation of caspase-1 in response to Salmonella and Shigella [18] , [33] . These results suggest that Asc is somehow required for the interaction between Nlrc4 and caspase-1 or that Asc is critical for another step which is important for inflammasome activation .
All animal experiments were conducted according to the U . S . A . Public Health Service Policy on Humane Care and Use of Laboratory Animals . Animals were maintained in an AAALAC approved facility and all animal studies followed protocol 09716-2 that was approved by the Animal Care and Use Committee of the University of Michigan ( Ann Arbor , MI ) . Mice deficient in Nlrc4 , Nlrp3 , Asc and caspase-1/11 have been previously described [4] , [34] , [35] . All mice were crossed at least 5 times on a C57BL/6 background . Bone marrow samples from Prkcd−/− mice in C57BL/6 background were provided by Hee-Jeong Im Sampen ( Rush University Medical Center , Chicago , IL ) . Shigella flexneri strain YSH6000 [36] was used as the WT strain , and S325 ( mxiA::Tn5 ) [37] was used as the T3SS–deficient control . The WT S . enterica serovar Typhimurium SR-11 χ3181 and the isogenic fliA::Tn10 were provided by H . Matsui ( Kitasato Institute for Life Science , Tokyo , Japan ) [38] . ΔfliA Salmonella mutant is impaired in the expression of flagellin [18] . cDNAs encoding mouse Naip2 , Naip5 , Nlrc4 , Asc , caspase-1 , and bacterial MxiI were amplified by PCR and cloned into the pCMV based mammalian expression vector or the MSCV-IRES-GFP retroviral expression vector ( Addgene ) . Human pro-IL-1β clone ( RDB6666 ) was provided by RIKEN BRC which is participating in the National Bio-Resource Project of the MEXT , Japan . BMDMs were prepared from the femurs and tibias of mice and cultured for 3–7 days in 10% FCS IMDM ( Gibco ) supplemented with 30% L-cell supernatant , non-essential amino acids , sodium pyruvate and antibiotics ( Penicillin/Streptomycin ) . 293T cells were cultured on Dulbecco's Modified Eagle's medium ( Sigma ) containing 10% FCS and antibiotics ( Penicillin/Streptomycin ) . The rabbit anti mouse caspase-1 p20 and anti-mouse Nlrc4 antibodies were produced in our laboratory by immunizing rabbits with mouse caspase-1 ( p20 subunit ) and mouse Nlrc4 ( amino acids 1–152 ) recombinant proteins [39] . Anti–IL-1β p17 ( #2021 ) and anti-Pkcδ ( #2058 ) antibodies were from Cell Signaling . Mouse monoclonal anti-β-actin antibody was from Sigma . HRP-conjugated goat anti–rabbit ( Jackson Laboratories ) or anti–mouse IgG ( Sigma ) or anti-rat ( Jackson Laboratories ) , or AP-conjugated goat anti-rabbit ( Santa Cruz Biotechnology Inc . ) or anti-mouse IgG ( Santa Cruz Biotechnology Inc . ) antibodies were used as secondary antibodies for immunoblotting . Macrophages were seeded in 24-well plates at a density of 3×105 cells per well . Cells were stimulated with or without 0 . 1 µg/ml LPS ( from E . coli O55:B5 , Sigma ) for 6 h and then infected with Shigella or Salmonella . Bacterial strains were pre-cultured overnight in Mueller-Hinton broth ( Difco ) at 30°C , then were inoculated into brain heart infusion broth ( Difco ) and incubated for 2 h at 37°C prior to infection . The cells were infected with Shigella at a bacteria/macrophage ratio of 10∶1 , or with Salmonella at a bacteria/macrophage ratio of 1∶1 unless otherwise stated . The plates were centrifuged at 700 g for 5 min to synchronize the infection , and gentamicin ( 100 µg/ml ) and kanamycin ( 60 µg/ml ) were added after 20 min . At the indicated times after infection , cytokines were measured in culture supernatants by enzyme-linked immunoabsorbent assay ( ELISA ) kits ( R and D Systems ) . RNA was isolated with E . Z . N . A . TM total RNA kit ( Omega Biotek ) according to the manufacturer's instructions . RNA was reverse transcribed using the High Capacity RNA-to cDNA kit ( Applied Biosystem ) and cDNA was then used for RT-PCR . For immunofluorescence studies , the infected cells were fixed and immunostained , and then analyzed with a confocal laser-scanning microscope ( LSM510; Carl Zeiss ) or fluorescence microscopy ( Olympus ) . Carboxyfluorescein FLICA ( Immunochemistry Technologies , LLC ) was added 1 hr before bacterial infection . Apoptosis was measured by the AnnexinV ( Roche ) and TUNEL ( Promega ) assays using fluorometric protocols according to the manufacture's recommendations . For the caspase-3 inhibitor studies , the cells were treated with 200 µM z-DEVD-fmk ( Calbiochem ) for 1 h before bacterial infection . 293T cells were seeded in 6-well plates at a density of 5×105 cells per well and incubated overnight . Then , the cells were transfected with or without 1 µg T7-tagged Nlrc4 , 1 µg T7-tagged Asc , 0 . 4 µg HA-tagged caspase-1 , and 0 . 4 µg FLAG-tagged proIL-1β [40] , and 1 µg HA-tagged Naip2 or Naip5 , using FuGENE 6 ( Roche ) . Cells were infected one day after infection . Intensities of casp1 p20 or IL-1β p17 bands were quantified by densitometry , the values normalized to the β-actin protein levels and results were analyzed with ImageJ software . The Shigella MxiI gene was cloned into the MSCV-IRES-GFP retrovirus vector , which contains an IRES-GFP element to track retroviral infection . WT or Nlrc4−/− BMDMs were immortalized using the J2 virus to increase nucleofection efficiency [41] . Then , cells were nucleofected with MSCV-IRES-GFP or MSCV-IRES-GFP encoding Shigella MxiI using an Amaxa nucleofector system ( Nucleofector kit V and the D-032 program ) . After 20 hrs , cell survival in the GFP-positive cell population was analyzed by fluorescence microscopy . The LDH activity in the culture supernatants of infected cells was measured using the CytoTox 96 assay kit ( Promega ) according to the manufacturer's protocol . Assays were performed in triplicate for each independent experiment . The invasion efficiency of Shigella strains was evaluated using a gentamicin/kanamycin protection assay . Briefly , cells were infected for 20 min and then incubated for 20 min at 37°C in medium containing gentamicin ( 100 µg/ml ) and kanamycin ( 60 µg/ml ) to kill extracellular bacteria . The infected cells were then washed in PBS , lysed in 0 . 5% TritonX-100/PBS , and serial dilutions were plated on LB agar plates to determine the number of intracellular bacteria . DNA and siRNAs specific for Naip2 and Naip5 were transfected into macrophages using an Amaxa nucleofector system ( Y-001 program for primary macrophages or D-032 program for cell lines ) according to the manufacturers' instructions . siRNA pools for mouse Naip2 ( 17948; D-044151-01-04 ) and Naip5 ( 17951; D-044141-01-4 ) and non-targeting siRNAs were purchased from Dharmacon or synthesized by Sigma and targeting the sequences CTTACACTGAATCACAAGA ( naip2 ) or GTGCCTTTTTAGTCCTTGT ( naip5 ) . Primer sets for RT-PCR were naip2-forward ( AGGCTATGAGCATCTACCACA ) , naip2-reverse ( AAGACATCAATCCACAGCAAA ) , naip5-forward ( TGCCAAACCTACAAGAGCTGA ) , naip5-reverse ( CAAGCGTTTAGACTGGGGATG ) , actin-forward ( CATGTACGTTGCTATCCAGGC ) and actin-reverse ( CTCCTTAATGTCACGCACGAT ) . To compare caspase-1 p20 levels in immunoblotting experiments , the bands were quantified by densitometry , analyzed with ImageJ software , and normalized to the β-actin protein levels . Cell ware lysed in IP buffer [CelLytic M Cell Lysis Reagent ( Sigma ) , 0 . 1 mM PMSF , and a complete protease inhibitor cocktail-EDTA ( Roche ) and clarified lysates were mixed with anti-T7 antibody–conjugated agarose beads ( Novagen ) or anti-HA conjugated sepharose beads ( Covance ) for 1 hr at 4°C with gentle rotation in IP buffer . Beads were washed with PBS , mixed with SDS-sample buffer and subjected to immunoblot analysis . Cells were fixed with 4% paraformaldehyde and 0 . 1% NP40 , washed and stained with anti- Asc antibody and FITC-conjugated anti–rat antibody ( Sigma ) as a secondary antibody . Imaging analysis was performed using fluorescence microscopy ( Olympus ) , and percentage of cells containing Asc pyroptosomes was determined by counting at least 300 cells in 5 separate fields . The Asc dimerization assay was previously described [25]–[27] . Briefly , cells were lysed ( 20 mM HEPES-KOH , pH 7 . 5 , 150 mM KCl , 1% NP-40 , 0 . 1 mM PMSF , and Complete protease inhibitor cocktail ( Roche ) ) and forced onto a 21-gauge needle 10 times . The cell lysates were centrifuged at 6000 rpm for 10 min at 4°C to isolate the insoluble fraction in the pellet . The pellets were washed twice with PBS , resuspended in 500 µl of PBS and cross-linked with fresh 2 mM disuccinimidyl suberate ( DTT , Sigma ) for 30 min . The cross-linked pellets were isolated by centrifugation at 13000 rpm for 10 min and resuspended in 20 µl of SDS sample buffer for immunoblotting with anti-mouse Asc antibody . Mouse cytokines in culture supernatants were measured by ELISA kits ( R&D Systems ) . Assays were performed in triplicate for each independent experiment . Statistical analyses were performed using the Mann–Whitney U test . Differences were considered significant at p<0 . 05 .
|
Shigella are bacterial pathogens that are the cause of bacillary dysentery . An important feature of Shigella is their ability to invade the cytoplasm of host epithelial cells and macrophages . A major component of host recognition of Shigella invasion is the activation of the inflammasome , a molecular platform that drives the activation of caspase-1 in macrophages . Although Shigella is known to induce the activation of the Nlrc4 inflammasome , the mechanism by which the bacterium activates Nlrc4 is largely unknown . We discovered that the Shigella T3SS inner rod protein MxiI induces Nlrc4 inflammasome activation through the interaction with host Naip2 , which promoted the association of Naip2 with Nlrc4 in macrophages . Expression of MxiI induced caspase-1 activation , Asc oligomerization , pyroptosis and IL-1β release which required Naip2 , but not Naip5 . Significantly , caspase-1 activation induced by Shigella infection was unaffected by deficiency of the Pkcδ kinase . This study elucidates the microbial-host interactions that drive the activation of the Nlrc4 inflammasome in Shigella-infected macrophages .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"immunity",
"innate",
"immunity",
"immunology",
"biology"
] |
2014
|
Shigella Type III Secretion Protein MxiI Is Recognized by Naip2 to Induce Nlrc4 Inflammasome Activation Independently of Pkcδ
|
Deep sequencing technologies have the potential to transform the study of highly variable viral pathogens by providing a rapid and cost-effective approach to sensitively characterize rapidly evolving viral quasispecies . Here , we report on a high-throughput whole HIV-1 genome deep sequencing platform that combines 454 pyrosequencing with novel assembly and variant detection algorithms . In one subject we combined these genetic data with detailed immunological analyses to comprehensively evaluate viral evolution and immune escape during the acute phase of HIV-1 infection . The majority of early , low frequency mutations represented viral adaptation to host CD8+ T cell responses , evidence of strong immune selection pressure occurring during the early decline from peak viremia . CD8+ T cell responses capable of recognizing these low frequency escape variants coincided with the selection and evolution of more effective secondary HLA-anchor escape mutations . Frequent , and in some cases rapid , reversion of transmitted mutations was also observed across the viral genome . When located within restricted CD8 epitopes these low frequency reverting mutations were sufficient to prime de novo responses to these epitopes , again illustrating the capacity of the immune response to recognize and respond to low frequency variants . More importantly , rapid viral escape from the most immunodominant CD8+ T cell responses coincided with plateauing of the initial viral load decline in this subject , suggestive of a potential link between maintenance of effective , dominant CD8 responses and the degree of early viremia reduction . We conclude that the early control of HIV-1 replication by immunodominant CD8+ T cell responses may be substantially influenced by rapid , low frequency viral adaptations not detected by conventional sequencing approaches , which warrants further investigation . These data support the critical need for vaccine-induced CD8+ T cell responses to target more highly constrained regions of the virus in order to ensure the maintenance of immunodominant CD8 responses and the sustained decline of early viremia .
A major challenge to the development of effective vaccines against highly variable viruses is their ability to adapt to evade host immune responses [1]–[4] . During HIV-1 infection , for example , immune escape mutations develop which impair the ability of both CD8+ T cell responses and neutralizing antibodies to maintain immune control [5]–[9] . However , some CD8+ T cell escape mutations have been shown to dramatically impair viral replication capacity , which may slow viral escape and contribute significantly to the ability of some responses to effectively control HIV-1 [10]–[13] . The outcome of this dynamic interplay between immune responses functioning to eliminate infected cells , emerging escape variants that evade these responses , and the impact of these variants on viral replication , critically influences early immune control of HIV-1 . The majority of studies on HIV-1 evolution have relied on bulk Sanger sequencing to define the major genetic variants that arise during infection . These studies have demonstrated that upwards of 50% of mutations observed over the course of infection may be associated with viral adaptations to CD8+ T cell responses [5] , [14] . Unfortunately , bulk Sanger sequencing is insufficient to detect low frequency variants that are particularly important during the acute phase of infection when viral escape occurs rapidly . The application of single genome amplification and sequencing ( SGA or SGS ) has increased the sensitivity for detecting and quantifying low frequency viral variants [7] , [15] , [16] but high cost and poor scalability limit its broader application . As a result , a sensitive and comprehensive understanding of the genetic pathways and kinetics of viral adaptation to acute phase immune selection pressures across the entire HIV-1 genome , likely a critical determinant of the success or failure of both natural and vaccine-elicited immune responses , is lacking . Next-generation sequencing ( NGS ) or deep sequencing approaches such as 454 pyrosequencing [17] have the potential to transform the study of HIV-1 and other highly variable pathogens by providing a rapid and cost-effective approach for the sensitive characterization of the complex and rapidly evolving intra-patient viral quasispecies . Recent studies have applied deep sequencing approaches to HIV-1 and SIV to detect low frequency drug resistance variants [18]–[22] and CD8+ T cell escape variants [23]–[26] , although these studies have largely been limited to the analysis of discrete regions of interest . Here we report an approach for routine whole genome sequencing of HIV-1 that combines deep sequencing with novel algorithms for de novo sequence assembly and for accurate quantification of low frequency viral variants . This new platform not only provides the capacity to rapidly sequence across the whole HIV-1 genome for population-scale genetic analyses of large cohorts of HIV-1 infected individuals , but also the sensitivity to comprehensively characterize the earliest stages of viral immune adaptation during the critical initial interactions with the host immune response . The application of this whole genome deep sequencing platform to longitudinal samples from a single subject during acute HIV-1 infection reveals the speed and complexity of the simultaneous adaptation of HIV-1 to multiple host immune responses , and suggests that early , low frequency escape variants to dominant acute-phase CD8+ T cell responses may have a significant impact on the early immune control of HIV-1 .
Prior studies utilizing deep sequencing to more critically examine HIV-1 and SIV sequence diversity and evolution have focused predominantly on short , specific regions of the virus where evolution was known or expected to occur . To apply a deep sequencing approach that can interrogate diversity across the complete genome we designed primers that amplify four overlapping PCR amplicons spanning the entire protein-coding region of the HIV-1 genome ( HXB2 nt 779–9551; Figure 1A ) and validated them against a set of 89 HIV-1 clade B ( HIV-1B ) clinical samples from subjects in the acute and chronic phase of infection , as well as low-viremia controllers ( Table S1 in Text S1 ) . To reduce costs , we pooled the four amplicons from each individual sample prior to acoustic shearing and subject-specific molecular bar-coding , and then batched bar-coded samples from multiple subjects prior to performing emulsion PCR and pyrosequencing . In contrast to traditional Sanger sequence data , the 454 sequence data provides deep read coverage ( sequencing reads per site ) where each individual base and the context in which it occurs in the read can be leveraged to inform the consensus assembly . As such , we developed AssembleViral454 ( AV454 ) , a module in the ARACHNE17 , 18 assembly tool kit ( see Supplementary Methods in Text S1 ) , which takes advantage of deep sequence coverage and the knowledge that continuous RNA viral genomes do not generally contain large repetitive sequences to correctly assemble all reads . As shown in Figure 1B , AV454 consensus assemblies captured on average 96 . 3%±11 . 3% ( s . d . ; n = 89 ) of patient-specific reads into a single contig ( Table S1 in Text S1 ) , significantly outperforming the other assemblers . While both AV454 and Newbler captured >98% of the target genome by all contigs assembled , AV454 captured a significantly greater percentage of the genome in a single continuous contig than any other assembler ( see Supplementary Methods in Text S1; Wilcoxon , p<0 . 001 , n = 67 ) and exhibited a much tighter distribution of results . These data demonstrate the ability of this sequencing and assembly strategy to reproducibly generate genome-wide sequence assemblies from a wide variety of different HIV-1B clinical isolates . A major challenge to the utility of deep sequence data is distinguishing true genetic polymorphisms from process errors [22] , [23] , [26] , [27] . We addressed this problem by developing an analysis pipeline designed to: ( i ) maximize the read data retained for analysis , ( ii ) optimize read alignments , and ( iii ) leverage phase information to improve the sensitivity and specificity of variant calling . First , all read alignments are made to the sample's AV454 consensus assembly . A comparison of read alignments to the AV454 de novo assembly versus an HIV-1B reference sequence demonstrated that use of the AV454 assembly retained more reads and bases for analysis and significantly reduced the number of insertions and deletions that result in alignments with frame shifts ( Table S2 in Text S1; Wilcoxon , p<0 . 0001 ) , an important consideration for variant calling . Second , ReadClean454 ( RC454 ) applies a Neighborhood Quality Standard ( NQS ) base filter [28] , corrects reads for common process errors such as homopolymer and carry-forward-incomplete-extension ( CAFIE ) miscalls ( see Supplementary Methods and Figure S1 in Text S1 ) , and further optimizes read alignments using coding frame information . As shown in Figure 1C , RC454 significantly reduces the average process read error rate from 1 . 3×10−2 to 0 . 5×10−4 errors per base as determined by the sequencing of infectious HIV-1 clones . Next , V-Phaser distinguishes true variants from sequencing errors by defining the frequency at which a nucleotide polymorphism must be observed to be considered a true variant . This is accomplished through the application of an error probability model initially defined by a uniform empirical process read error rate and then refined by the inclusion of variant nucleotide phasing information i . e . correlated sequence changes ( see Supplementary Methods in Text S1; Macalalad et al , manuscript submitted ) . Lastly , V-Profiler calculates the frequency of each triplet codon composed of nucleotides accepted by V-Phaser . When applied to samples of known composition , this pipeline quantified variants with high sensitivity ( 100% ) and specificity ( 97% ) , and implementation of the phasing-based approach achieved detection of 1 . 0% variants when ≥200-fold shared sequence coverage ( sequencing reads per site ) was attained; this represents a three-fold decrease in required coverage over non-phase based methods; Figure 1D; see Supplementary Methods in Text S1 ) . The application of these algorithms provides the ability to rapidly characterize intra-patient HIV-1 genetic diversity , and facilitate the routine handling of deep sequencing data for whole genome assembly and variant detection , as shown in Figure S2 in Text S1 for all HIV-1 proteins from the array of 89 HIV-1B clinical samples . The whole HIV-1 genome 454 deep sequence platform was validated by comparison to bulk Sanger sequencing , cloning and sequencing , and SGA . First , we compared full length consensus HIV-1 sequences for four longitudinal samples from a single subject ( 9213 ) generated by bulk Sanger sequencing and by the 454 platform ( 35 , 093 total nucleotides compared; see Supplementary Results in Text S1 ) . Overall , the Sanger and 454 consensus sequences differed at only six nucleotides and one insertion/deletion ( InDel ) , and in each case the discrepancy resulted from a differential consensus call at a highly polymorphic position ( Table S3A and S3B in Text S1 ) . Next , we extensively compared variant quantification across a highly variable 1544 nucleotide region spanning from vif to tat in a single sample ( subject 9213 ) by deep sequencing ( average 566-fold high quality sequencing reads per site ) , traditional PCR cloning and sequencing ( 768 clones ) , and single genome amplification ( 87 single genomes ) . We observed 95 . 6% concordance between the three methods in the detection of invariant/variant sites ( see Supplementary Results in Text S1 ) , and the calculated variant frequencies were highly correlated between methods as shown for deep sequencing vs cloning and sequencing in Figure 2 . These data confirm the ability of this high-throughput , deep sequencing platform to profile HIV-1 quasispecies diversity as accurately as conventional cloning and sequencing or SGA . Recent studies utilizing deep sequencing to more sensitively assess early , low frequency variants within specific CD8 epitopes reveal that viral escape from CD8+ T cell responses can occur very rapidly [23] , [25] , [26] , even as soon as 17 days following SIV infection of macaques [23] . To further explore the dynamics of HIV-1 evolution and immune adaptation during acute infection , we conducted a comprehensive and sensitive assessment of early viral evolution , without bias towards previously studied epitopes , by producing longitudinal genome-wide 454 sequence data from longitudinal samples from a single subject identified as HIV-1 infected prior seroconversion . Subject 9213 presented with a baseline viral load of 9 . 3 million copies/ml ( day 0 post-presentation ) that peaked at 21 million copies/ml on day 3 ( Figure 3A ) . A negative Western blot on day 0 supported likely infection within 15–20 days of first sampling , i . e . Fiebig stage II–III [29] . We captured genetic diversity data for the entire open reading frame of HIV-1 at six time points over the first 4 years of infection ( day 0 , 3 , 59 , 165 , 476 , 1543 ) at an average number of sequencing reads per site of 535±325 reads ( Figure 3B , Table S1 in Text S1 ) . Codon diversity , defined as the frequency of codons that differed from the consensus codon at baseline ( day 0 ) , was calculated for each position of the HIV-1 proteome . As illustrated in Figure 4A and 4B , there was strikingly little codon diversity present in the viral population during peak viremia , with less than 2% and 5% of all positions exhibiting detectable diversity at day 0 and day 3 , respectively , and of those positions that did vary the majority varied by less than 2% . The low genetic diversity of the viral quasispecies during early acute infection , which would not have been discernable using traditional bulk sequencing approaches , confirms that infection in this subject was founded by a single genetic lineage , in line with recent reports suggesting that most sexually transmitted HIV-1 infections arise from a single founder virus [15] , [16] , [30]–[33] . The first evidence of HIV-1 evolution was observed at day 59 , when 11% of all codons exhibited detectable diversity ( Figure 4C ) . However , still only a minor subset of 21 codons exhibited any substantial ( >10% ) degree of variation from baseline at this time point when peak viral loads were observed to dramatically decline to 298 , 000 copies/ml ( Figure 3A ) . Although as expected the number of evolving codons continued to increase over time , with 38 exhibiting detectable diversity at day 165 ( Figure 4D ) , it is notable that over half of the day 59 sites exhibiting substantial variation ( >10% ) declined in variation by day 165 ( Figure 5 ) . These data reveal complexities in the early evolution of the viral quasispecies that are not typically observed by traditional sequencing methods . Moreover , as shown in Figure 5 even by day 165 no single codon had yet mutated towards fixation ( >95% ) , suggesting that the substantial early decline in peak viremia in subject 9213 was not associated with any dramatic turnover of the viral population . Given that CD8+ T cell responses represent a major driving force of viral evolution following acute HIV-1 infection [5] , [7] , [34] , we examined the extent to which these early , low frequency mutations might represent viral adaptation to cellular immune responses . Here we compared amino acid divergence from baseline within described CD8+ T cell epitopes restricted by subject 9213's HLA alleles to the amino acid divergence at all other positions across the proteome . We observed that the majority of early viral evolution at days 59 and 165 was indeed shaped by cellular immune responses , with significantly greater diversity observed within restricted epitopes ( Wilcoxon , p = 0 . 016; Figure 6A; Table S4 in Text S1 ) . At day 59 , this was most pronounced in Vif and Nef , with Env and Pol also exhibiting diversity preferentially within restricted CD8 epitopes by day 165 . These data suggest that rapid adaptation to cellular immune responses was the major driving force for the early , low frequency viral evolution observed in subject 9213 . To better understand the early immune adaptation of HIV-1 in subject 9213 , we characterized the breadth and magnitude of CD8+ T cell responses to all 19 described CD8+ T cell epitopes by IFN-gamma ELISPOT assay using autologous peptides . Acute phase ( day 59 ) responses were detected against six epitopes , with the two most dominant responses directed against the Vif B38-WI9 ( 2744 SFC/Mill PBMC ) and Nef A24-RW8 ( 2584 SFC ) epitopes , while weaker subdominant responses were directed against the Pol B44-EW9 ( 814 SFC ) , Rev A01-IY9 ( 734 SFC ) , Gag B44-AW11 ( 444 SFC ) , and Gag A01-GY9 ( 144 SFC ) epitopes ( Table S5 in Text S1 ) . The deep sequencing data revealed evidence of viral adaptation , i . e . , escape , within four of the six epitopes ( Figure 7 and Table S6 in Text S1 ) . The escape phenotype of the observed genetic variants was confirmed by the impaired recognition of each of the variant's peptides when tested in IFN-gamma ELISPOT assays ( Table S5 in Text S1 ) . Viral immune adaptation was most rapid in the dominantly targeted Vif B38-WI9 and Nef A24-RW8 epitopes , with estimated escape rates of 0 . 0987 day−1 and 0 . 0976 day−1 , respectively ( Figure 6B ) . Interestingly , we observed distinct adaptive pathways by which the virus evaded each of these dominant , early responses . In the Vif B38-WI9 epitope , by day 59 56 . 6% of the viral population expressed one of four intra-epitope mutations ( Figure 6C ) , and/or three flanking mutations likely affecting antigen processing [35] , [36] . This initial apparent exploration of multiple escape pathways resolved over time with over 98% of sequenced reads from the population now comprising just three variant “haplotypes” by day 165 , before fixing on the I87V mutation at the C-terminal HLA-class I epitope anchor residue by day 476 ( Figure 7 ) . In contrast , immune adaptation in the Nef A24-RW8 epitope followed a more restricted pathway with 54 . 7% of the quasispecies expressing a single escape mutation ( F148L ) at day 59 , followed by the emergence of a second escape mutation ( T147M ) at day 165 which together comprised >99% of the total population ( Figure 6D ) . Interestingly , the original F148L mutation at position 6 of the epitope was replaced by day 476 with the Y144F mutation , a position 2 HLA-anchor mutation that is likely a more potent escape mutation . This approximately 50–50 mixed population of position 2 Y144F and position 5 T147M escape variants remained stable out to day 1543 ( Figure 7 ) . Thus , deep sequencing during the acute phase of infection revealed rapid viral escape from the two most dominant acute phase CD8+ T cell responses , in some cases through the combined effects of multiple low frequency variants that would be missed by traditional bulk sanger sequencing . Interestingly , in both cases the early escape mutations were ultimately replaced in the viral population by HLA anchor position mutations that more efficiently escaped immune recognition ( Table S5 in Text S1 ) , presumably through reductions in MHC-I:peptide binding at the cell surface . Viral escape was also observed in the Pol B44-EW9 and Gag A01-GY9 epitopes ( Figure 7 and Table S6 in Text S1 ) that were targeted by subdominant acute-phase CD8+ T cell responses of 814 SFC and 144 SFC , respectively ( Table S5 in Text S1 ) . Here , the lower magnitude of these responses was associated with slower estimated escape rates of 0 . 0133 and 0 . 0036 day−1 ( Figure 6B ) . In both cases , early , low frequency mutations at positions 4 and 6 of these epitopes at day 165 , likely T cell receptor ( TCR ) escape mutations , were subsequently replaced by HLA-anchor mutations at position 2 or 9 . These data provide insight into possible mechanisms underlying the transient variation of some residues observed in Figure 5 , whereby early mutations are being out competed by more effective secondary mutations . Finally , the two other epitopes that were targeted during acute infection , Rev A01-IY9 ( 734 SFC ) and Gag B44-AW11 ( 444 SFC ) , exhibited no evidence of immune escape over the course of infection despite the higher sensitivity of deep sequencing ( Figure 7 and Table S6 in Text S1 ) . In addition to the six epitopes targeted during acute infection , weak CD8+ T cell responses were also detected against four other epitopes during the chronic phase of infection ( day 476 ) : Env Cw4-SF9 ( 190 SFC ) , Env A24-RL9 ( 84 SFC ) , Env A01-RY9 ( 80 SFC ) , and Nef A01-YT9 ( 70 SFC ) ( Table S5 in Text S1 ) . There was evidence of viral escape in the three Env epitopes ( Figure 7 and Table S6 in Text S1 ) , but similar to the epitopes targeted by subdominant acute phase responses , the rate of escape in these chronically targeted epitopes was slow at 0 . 0067 , 0 . 0087 , and 0 . 0026 day−1 respectively ( Figure 6B ) . Overall , the virus escaped from four of the six CD8+ T cell responses mounted during the acute phase of infection and three of the four CD8+ T cell responses mounted during chronic infection , with highly variable rates of escape observed for different epitopes . In subject 9213 , we found that the rate of immune escape from acute phase CD8+ T cell responses correlated with the magnitude of these responses ( p = 0 . 01 ) , reflecting the differential selective pressure imposed on the viral population by distinct CD8+ T cell responses . Interestingly , greater than 50% of the viral population had escaped the dominantly targeted Vif B38-WI9 and Nef A24-RW8 epitopes by 59 days post-presentation , which corresponds temporally to the plateauing of the precipitous decline from peak viremia and the subsequent equilibration of viral load ( Figure 3A ) . Thus , these data from a single subject suggest that the rate at which the virus escapes from critical acute phase immunodominant responses , in some cases through the combined effects of multiple low frequency mutations , may influence the magnitude of the drop from peak viremia and duration of effective early immune control , and by extension set-point viral load . We have previously observed that CD8+ T cell responses can arise that are capable of recognizing CTL escape variants [37]–[39] , demonstrating that the immune system is at least partially able to contend with immune escape . To investigate the kinetics of such variant-specific responses , and whether they might be triggered by early , low frequency mutations arising during the acute phase of infection , we screened for responses against the most frequent escape variants in the rapidly escaping Vif B38-WI9 and Nef A24-RW8 epitopes . As early as day 59 , strong responses were detected against two of the primary escape variants in the Vif B38-WI9 epitope , despite the fact that the S86A and I87V mutations comprised less than 15% of the viral quasispecies ( Figure 8A; Table S5 in Text S1 ) . These variant-specific responses persisted out to day 476 , and as we have previously observed in chronic infection were equal in magnitude to the autologous wild-type response [38] . However , fixation in the epitope of the C-terminal I87V mutation , likely impairing MHC-I binding and presentation , ultimately coincided with a significant ( >10-fold ) decline of both wild-type and the variant-specific responses by day 661 . We also detected early responses to escape variants in the Nef A24-RW8 epitope , albeit at much lower magnitudes , and similarly the emergence of an HLA anchor position escape mutation ( Y144F ) ultimately abrogated responses against both wild-type and variant peptides ( Figure 8B; Table S5 in Text S1 ) . This early recognition of the low-frequency escape variants , followed by loss of responses upon outgrowth of HLA anchor mutations , suggests partial cross-recognition of early escape variants by the wild-type-specific response [40] rather than development of de novo variant-specific responses [37] . Thus , eventual loss of the wild-type sequence , required for continuous expansion of these wild-type-specific responses , results in the eventual decline of all responses . Thus , these data extend earlier reports of the ability of CD8+ T cell responses to recognize viral escape mutations [37]–[40] by illustrating the ability of early responses to recognize low frequency escape mutations and providing a mechanism for the observed substitution of early escape mutations with more potent secondary HLA-anchor mutations . Apart from the evolution in the targeted CD8 epitopes described above , we also observed substantial evolution in four other non-targeted CD8 epitopes restricted by subject 9213's HLA alleles ( Nef A01-WH10 , Env B44-AY10 , Gag A24-KW9 , and Gag B44-LY9; Table S7 in Text S1 ) . Responses were never detected against these epitopes during either acute ( day 59 ) or chronic ( day 476 and day 661 ) infection despite testing with autologous peptides matching the founder virus ( Table S5 in Text S1 ) . Each of these evolving epitopes was found to contain one or more transmitted mutations at baseline ( day 0 ) , with the observed evolution consistent with the reversion of these transmitted mutations back towards the HIV-1B consensus sequence . Reversions in the Nef A01-WH10 and Env B44-AY10 epitopes occurred with estimated rates of 0 . 0722 and 0 . 0887 day−1 ( Figure 9A ) , respectively , nearly equaling those of the most rapidly escaping Vif B38-WI9 and Nef A24-RW8 epitopes ( Figure 6B ) . Reversion in the Gag A24-KW9 epitope was actually the result of the transmission and reversion of a K28Q mutation that is a well-described escape mutation in the overlapping A03-RK9 epitope [35] , [41] . Interestingly , founder virus mutations reverted in three additional HLA-A03 epitopes ( Pol-ATK9 , Pol-QK9 , and Vif-RK10 ) , and two HLA-B57 epitopes ( Vif B57-IF9 and Nef B57-YY9 ) , suggesting that the founder virus in subject 9213 had previously adapted to both HLA-A03 and B57 immune responses ( Figure 9A; Table S8 in Text S1 ) . Reversion of other well-described escape mutations such as I293T in the Pol B51-TI8 epitope [41] , [42] and I63T in the Vpr A02-AL9 epitope [43] was also detected . In total , 15% ( 56/373 ) of all transmitted , non-consensus mutations exhibited evolution consistent with reversion over the four years of follow-up ( Figure 9B ) . Thus , the increased sensitivity afforded by the longitudinal deep sequencing data revealed that not only is reversion of transmitted mutations a significant contributor to the evolution of HIV-1 , but that these mutations revert at vastly different rates implying significantly different impacts of each mutation on viral replication capacity . As a result of these findings we undertook a closer examination of the evolution within the ten targeted CD8 epitopes . Transmitted mutations at baseline were in fact present in five of these epitopes ( Table S6 in Text S1 ) , with evolution in two of the chronically targeted epitopes consistent with the reversion of transmitted mutations . In the Env A01-RY9 epitope , despite the fact that CD8+ T cell responses were not detected until day 476 , as early as day 59 low frequency mutations developed at the residues containing transmitted mutations ( Figure 7 , Table S6 in Text S1 ) . In line with the hypothesis that the early evolution in this epitope may have represented reversion of transmitted mutations , we first detected low magnitude ( 70 SFC ) immune responses against this Env A01-RY9 epitope at day 476 following partial outgrowth of the HIV-1B consensus residue ( R794 at 9%; Tables S5 and S6 in Text S1 ) . Similarly , in the other late-targeted Env A24-RL9 epitope we observed partial reversion ( 20% ) towards consensus of another transmitted mutation ( K593R ) at day 476 , which was also associated with the late development of a low magnitude ( 84 SFC ) response against the wild-type form of the epitope ( Table S5 and S6 in Text S1 ) . Thus , while the transmission of mutations within some CD8 epitopes restricted by subject 9213 prevented the mounting of early immune responses to these epitopes , the reversion of transmitted mutations , even at very low frequencies , was sufficient to enable the priming of immune responses to these epitopes .
We have established a high-throughput deep sequencing platform to assess HIV-1 sequence diversity across the entire HIV-1 genome . As the result of developing novel sequence assembly and variant detection algorithms , we were able to rapidly produce deep sequence data for a diverse set of 89 clade B clinical isolates and to dissect the evolutionary dynamics of HIV-1 during the earliest stages of acute infection . Our results from an in-depth analysis of a single subject reveal that the majority of early , low frequency mutations arising during the acute phase of infection reflect adaptation to host CD8+ T cell responses . Moreover , the temporal link observed between interruption of the decline in peak viremia and escape from the most immunodominant CD8+ T cell responses through low-frequency mutations suggests that the rate of escape from a few key acute phase CD8+ T cell responses may strongly influence primary control of HIV-1 , and potentially viral set point . Thus , immune control during acute HIV-1 infection may be substantially influenced by early viral adaptations not detected by conventional sequencing approaches . The central role of cellular immune responses in the early control of HIV-1 is highlighted by our findings that across the viral proteome the majority of early , low frequency adaptive mutations in subject 9213 were associated with CD8+ T cell responses . These data support the substantial selective pressure exerted upon HIV-1 by these responses early after infection . While limited sample availability precluded an analysis of CD4+ T cell responses , none of the rapidly evolving sites in subject 9213 arose exclusively within described CD4 T cell epitopes . While this does not exclude the possibility of CD4 escape , our data were not able to directly identify any evidence of CD4 escape . Recent studies have illustrated that HIV-specific CD8+ T cell responses are guided by distinct immunodominance hierarchies , whereby certain responses consistently arise more rapidly during the acute phase of infection , and can even dominate responses restricted by other HLA alleles [44] , [45] . The two most immunodominant described B38- and A24-restricted epitopes , Vif B38-WI9 and Nef A24-RW8 , were also found to be immunodominant in subject 9213 . Moreover , they also represented the most rapidly escaping epitopes , with the kinetics of viral escape in subject 9213 corresponding in general to the hierarchy of all CD8+ T cell responses at baseline . These data support a strong link between the strength of a response and the relative selection pressure exerted , in line with recent data by Ferrari et al . [46] . More importantly , the observation that cessation of the rapid decline from peak viremia in subject 9213 was temporally coincident with viral escape from these two most immunodominant CD8+ T cell responses suggests that the duration of effectiveness of such immunodominant responses may be critical to the successful containment of early viral replication and prolonged viral load decline . Thus , the rate at which the earliest immunodominant CD8+ T cell responses are lost through viral escape may substantially influence the establishment of viral load set point , and thus progression to AIDS [47] . Unfortunately , with the exception of a few protective HLA alleles , the majority of immunodominant CD8 epitopes occur within more variable regions of the virus that would be expected to escape rapidly because they impart little or no viral fitness cost . As such , our data revealing that combinations of low frequency adaptive amino acid mutations may critically impact early control of HIV-1 by subverting the key CD8+ T cell responses may help to explain the inability of most HLA alleles to fully suppress early viral replication . Characterization of the molecular pathways of viral escape is central to the rational design of a durable T-cell based vaccine . The sensitivity of our approach revealed a common pattern of evolution within the majority of escaping epitopes , including both immunodominant and subdominant responses , in which combinations of multiple low frequency escape mutations were replaced over time by HLA-anchor mutations . CD8+ T cell responses specific for the earlier escape variants were associated with selection of these “secondary” escape mutations that were substantially more effective in abrogating CTL recognition . These data , and prior reports of variant-specific responses [37]–[40] , reveal the efficacy of these variant-specific responses , and suggest a potentially more important role for these responses in the control of HIV-1 . It is important to note , however , that while some studies have carefully demonstrated the ability of the immune response to recognize CTL escape variants using tetramers and peptide dilutions [37]–[39] , other studies have found that the high peptide concentrations often used to detect cross-reactive responses to variants can be misleading since the peptide levels are often substantially higher than physiological levels [48] , [49] . Unfortunately , a lack of sample availability at the early time points when these responses were robust prevented the testing of mutant and autologous peptides at additional dilutions . Therefore , it will be important in future studies to examine the recognition of these types of early CTL escape mutations using physiological peptide concentrations of peptides , or point-mutant strains of HIV-1 so that the mutant epitopes can be naturally processed and presented at the cell surface at physiological levels . Nonetheless , these data exemplify the continuous nature of host-virus co-adaptation and suggest the need to consider these early transient escape mutations when designing vaccine immunogens . For example , mosaic immunogen approaches [50] , [51] , designed to impede viral escape by inducing responses against early escape mutations , may benefit from inclusion of these transient low frequency variants that are likely absent from the larger chronic sequence datasets upon which mosaic vaccine antigens are based . Similarly , these deep sequence data provide greater insight into the critical role of compensatory mutations , whereby viral escape within structurally interacting regions of a protein requires one or more co-evolving secondary mutations to retain protein structure and function [10] , [52] , [53] . In the Nef A24-RW8 epitope , eventual development of the position 2 HLA-anchor mutation ( Y144F ) was tightly linked to an upstream I142T mutation , exclusively present on the haplotype expressing the escape mutant ( Figure 6D ) . Thus , supplementing existing HIV-1 sequence databases with deep sequence data from both acute and chronically infected individuals may help to identify regions of HIV-1 which require co-evolving sites to escape [54] and thus would be most susceptible to immune targeting [55] , [56] . Transmitted escape mutations can also influence the course of infection both by impairing the induction of CD8+ T cell responses [35] , [43] , [57] , but also by attenuating viral replication capacity [58] , [59] . Importantly , the rate at which transmitted mutations revert may serve as a more accurate in vivo measurement of the relative impact of these mutations on viral fitness , as compared to in vitro viral fitness measurements [10] . The range of reversion rates of transmitted mutations observed in this genome-wide study ( 0 . 0887 to 0 . 0015 day−1 ) , including some that were very rapid , supports a significant impact of some of these mutations on viral replication capacity . The ability to more accurately determine the true rates of genome-wide reversions using the more sensitive deep sequencing data provides the unique opportunity to now systematically quantify the contribution of transmitted mutations on viral fitness , which may provide additional insight into the potentially significant contribution of viral genotype to HIV-1 set-point viral load [60] . The deep sequencing approach presented here yields results consistent with those of traditional cloning or SGA . A recent study by Jordan et al illustrates similar results for sequence diversity detection between standard PCR/cloning and SGA [61] . While improving upon the sensitivity of these methods , and providing the ability to simultaneously assay genetic diversity across all residues in the genome , our variant detection methods achieve a sensitivity and specificity of >97% at a substantially reduced cost as compared to SGA or cloning . Despite this high accuracy , as with other sequencing approaches , deep sequencing has its own set of limitations . First , despite efforts to optimize read alignments , mis-alignments can occur especially at the ends of amplicons and reads and lead to false positives; V-phaser is designed to limit false positives and Macalalad et al . ( manuscript submitted ) have shown that the variant detection methods described here achieve a positive predicted value ( PPV ) of 98% . Second , 454 deep sequencing is constrained in its ability to identify long-range linked mutations beyond a single read length of approximately 400 bp . When compared to SGA , this may limit its utility to understand more complex haplotype interactions , such as whether escape mutations in two simultaneously escaping eptiopes are arising upon the same viral haplotype [7] , or upon distinct viral haplotypes which later recombine [38] . Here , deep sequencing approaches and SGA may well serve to complement their respective individual strengths . Third , the bulk amplified PCR products used for this deep sequencing approach may be more prone to in vitro recombination events than the single-template amplifications used during SGA [62] . While this is unlikely to alter the frequency of variants detected by deep sequencing , it could limit the ability to accurately assess in vivo recombination rates and longer viral haplotypes . However , since both bulk amplification and SGA approaches rely on the bulk reverse transcription ( RT ) of RNA to cDNA , which itself may be prone to in vitro recombination [63] , both deep sequencing and SGA approaches may still be susceptible to recombination events . Finally , given the ability to routinely sequence the viral quasispecies at near unlimited depth the issue of template resampling may be a concern . In subject 9213 we quantified the number of input RNA template molecules used for each cDNA synthesis . In each case the number of template molecules ( >1000 RNA copies ) was greater than the fold depth of sequence data achieved ( 535±325 reads ) , arguing against template resampling having unduly influenced our findings . Supporting this conclusion is the congruence in variants and variant frequencies observed across 454 , clonal , and SGA data sets ( see Figure 2 , and Supplementary Results in Text S1 ) . The accuracy of the deep sequencing methods described here to identify variable and conserved sites are further confirmed by comparison of the diversity detected within individual patients to that observed in the global HIV-1 population . As shown in Figure S2 in Text S1 , which illustrates diversity plots for the 89 clade B clinical isolates , consistent diversity “hotspots” were observed in each protein , including the 5′ ( p17 ) and 3′ ( p15 ) regions of Gag and the V1–V3 loops of Env . Notably , sites that frequently exhibited high intra-patient diversity were more likely to be highly polymorphic in consensus sequences of circulating strains when compared both across the whole genome ( Wilcoxon , p<0 . 0001 ) and within any gene ( Wilcoxon , p<0 . 01 ) . Conversely , 28 residues were entirely conserved in both the intra-patient and global datasets . Such data support the accuracy of the deep sequencing methods and also provide a comprehensive view of the extent of genome-wide intra-host sequence diversity achieved during chronic HIV-1 infection , revealing that sites commonly susceptible to intra-host diversity contribute directly to the diversity observed between circulating strains . The development of a robust genome-wide HIV-1 deep sequencing approach provides both the means to rapidly produce whole genome data for large cohorts and a unique opportunity to sensitively and globally profile HIV-1's earliest adaptations to host immune pressures . Genome-wide diversity profiles may serve as a sensitive and effective readout of host immunity during both natural infection , but also following vaccination such as in the case of breakthrough subjects from the HIV-1 STEP trial [64] . Our analysis of early sequence evolution in a single subject indicates that a small number of early specific CD8+ T cell responses represent the major selective force being evaded when peak HIV-1 viremia first comes under control . Extending these results to larger cohorts of individuals , especially in subjects naturally controlling HIV-1 following acute infection , would support a critical role for the maintenance of a few key CD8+ T cell responses in the critical control of HIV-1 . If so , vaccine strategies aimed at triggering immunodominant responses against critical regions of the virus may prove more effective than efforts attempting to maximize the breadth or polyfunctionality of vaccine-elicited CD8+ T cell responses [55] .
All subjects gave written informed consent and the study was approved by the Massachusetts General Hospital Review Board . Plasma samples were obtained from HIV-1 cohorts at the Massachusetts General Hospital in Boston , Massachusetts , the Jessen-Praxis in Berlin , Germany and the HIV Swiss Cohort . Subject 9213 was identified during primary HIV-1 infection ( Western Blot negative; Fiebig II–III ) [29] , and time points are defined from day of presentation with symptomatic acute HIV-1 infection . High and intermediate-resolution HLA class I genotyping was performed by sequence-specific PCR and direct sequencing according to standard procedures . See Text S1 for a detailed description of sample preparation , library construction , and sequencing protocols , as well as a description of the genome assembly and variant detection algorithms and their validation .
|
The ability of HIV-1 and other highly variable pathogens to rapidly mutate to escape vaccine-induced immune responses represents a major hurdle to the development of effective vaccines to these highly persistent pathogens . Application of next-generation or deep sequencing technologies to the study of host pathogens could significantly improve our understanding of the mechanisms by which these pathogens subvert host immunity , and aid in the development of novel vaccines and therapeutics . Here , we developed a 454 deep sequencing approach to enable the sensitive detection of low-frequency viral variants across the entire HIV-1 genome . When applied to the acute phase of HIV-1 infection we observed that the majority of early , low frequency mutations represented viral adaptations to host cellular immune responses , evidence of strong host immunity developing during the early decline of peak viral load . Rapid viral escape from the most dominant immune responses however correlated with loss of this initial viral control , suggestive of the importance of mounting immune responses against more conserved regions of the virus . These data provide a greater understanding of the early evolutionary events subverting the ability of host immune responses to control early HIV-1 replication , yielding important insight into the design of more effective vaccine strategies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"immune",
"cells",
"clinical",
"immunology",
"immunity",
"hiv",
"immunology",
"biology",
"viral",
"diseases"
] |
2012
|
Whole Genome Deep Sequencing of HIV-1 Reveals the Impact of Early Minor Variants Upon Immune Recognition During Acute Infection
|
While an increasing number of conserved small regulatory RNAs ( sRNAs ) are known to function in general bacterial physiology , the roles and modes of action of sRNAs from horizontally acquired genomic regions remain little understood . The IsrK sRNA of Gifsy-1 prophage of Salmonella belongs to the latter class . This regulatory RNA exists in two isoforms . The first forms , when a portion of transcripts originating from isrK promoter reads-through the IsrK transcription-terminator producing a translationally inactive mRNA target . Acting in trans , the second isoform , short IsrK RNA , binds the inactive transcript rendering it translationally active . By switching on translation of the first isoform , short IsrK indirectly activates the production of AntQ , an antiterminator protein located upstream of isrK . Expression of antQ globally interferes with transcription termination resulting in bacterial growth arrest and ultimately cell death . Escherichia coli and Salmonella cells expressing AntQ display condensed chromatin morphology and localization of UvrD to the nucleoid . The toxic phenotype of AntQ can be rescued by co-expression of the transcription termination factor , Rho , or RNase H , which protects genomic DNA from breaks by resolving R-loops . We propose that AntQ causes conflicts between transcription and replication machineries and thus promotes DNA damage . The isrK locus represents a unique example of an island-encoded sRNA that exerts a highly complex regulatory mechanism to tune the expression of a toxic protein .
The first systematic searches for bacterial sRNA were based on bio-computational identification of conserved genes in intergenic regions [1] . The subsequent characterization of these conserved sRNAs identified them as important players in many adaptive and physiological responses . These conserved core-genome encoded regulatory sRNAs comprise many antisense RNAs of which a subset are cis-encoded whereas the majority acts on trans-encoded target mRNAs by limited base complementarity [2] . Most trans-acting base-pairing sRNAs of enteric bacteria require the RNA chaperone protein Hfq for both intracellular stability and for efficient annealing to target mRNAs [3] . However , the chromosomes of these bacteria are mosaics , composed of conserved collinear regions interspersed with unique genetic islands that were acquired horizontally via once-mobile genetic elements . Therefore , the early searches based on sequence conservation generally disregarded unique horizontally acquired sRNAs . However , subsequent global cDNA cloning , comparative genomic based expression screens , Hfq-bound and global transcriptomic screens detected short RNA species in non-conserved horizontally acquired regions as well as highly abundant short RNA species from the UTRs of protein-coding genes [4–8] . The function of these non-conserved sRNAs remains enigmatic yet promising , as they may inform new regulatory principles . Members of the genus Salmonella carry numerous genomic islands that were acquired by horizontal transfer of phages , plasmids and transposons . These islands carry fitness and virulence genes that are integral to Salmonella pathogenicity , enabling the bacteria to adapt to different niches , invade intestinal cells and multiply within cells of the immune response [9] . In a previous study , we screened the horizontally acquired genomic islands of Salmonella typhimurium for non-conserved sRNA genes . Our analysis led to identification of 19 unique island-encoded sRNAs including the Gifsy-1 prophage encoded IsrK RNA [6] . The chromosome of Salmonella is lysogenic for a number of phages including Gifsy-1 and Gifsy-2 , both of which carry genes implicated in Salmonella virulence [9 , 10] . Bacteriophage Gifsy-2 carries the sodCI gene encoding a periplasmic superoxide ( Cu , Zn ) -dismutase , with a proposed role in Salmonella defense against killing by macrophages as well as a number of genes encoding type III secreted effectors [11–13] . Gifsy-1 carries multiple virulent factors such as gipA , which is involved in bacterial colonization of small intestine [14 , 15] . To coordinate expression between core and island genes , bacteria often recruit sRNAs [16 , 17] . For example , InvR from the major SPI-1 island represses the production of the core-encoded major outer membrane protein OmpD [18] . IsrE , the paralogue of RyhB , represents another example of a cross talk between genes of core and islands [6] . The island-encoded sRNA IsrE is regulated by Fur , a core-encoded repressor in response to iron-deplete conditions and contributes to control of the core-encoded iron regulon [6 , 19] . Here we show that IsrK of Gifsy-1 prophage controls the expression of its genetic locus leading to growth arrest of Salmonella by acting as small and messenger RNA , in an Hfq-independent manner . The growth inhibition is caused by an increase in expression of a Q-like antiterminator protein ( here denoted AntQ ) that is encoded on the same locus . AntQ belongs to Q proteins’ family of lambdoid phages . Bacteriophage λ Q protein is an operon specific transcription anti-termination factor required for expression of the phage late genes . Q protein joins the elongation complex at early stages of transcription and enables RNA polymerase to read-through the terminator located upstream of the phage late genes [20–23] . To join the elongation complex , Q interacts with a specific DNA sequence element as well as with RNA polymerase that is paused during early elongation . The binding of Q alters the functional properties of the transcription elongation complex interfering with termination signals [24 , 25] . We find that although Q proteins are known to bind specific sites within phages , the function of the Gifsy-1 Q-like protein AntQ is not limited to Salmonella or phage DNA . By contrast AntQ promotes transcription elongation of core genome transcripts resulting in growth arrest and ultimately cell death .
The gene encoding IsrK sRNA is located within Gifsy-1 prophage . Upstream of IsrK is SL2579 encoding a Q-like anti-terminator protein ( here denoted AntQ ) . The transcription start-site of antQ was mapped 1 , 600 bases upstream of antQ [8] . Downstream of isrK we noticed a putative ORF of 45 amino acids , followed by SL2578 encoding a predicted anti-repressor-like protein ( denoted AnrP ) and the previously identified sRNA encoding gene , isrJ ( Fig 1A ) [6] . To examine whether transcription of the downstream gene anrP is linked to IsrK , we constructed isrK-orf45-anrP-lacZ transcriptional fusions with and without the isrK promoter . The assays showed that IsrK promoter directs transcription of the downstream gene anrP . Interestingly , the corresponding translation fusion demonstrated that although the operon is transcribed , there is no translation of anrP mRNA ( Table 1A ) . We noticed that plasmid borne constitutive expression of isrK is toxic . Salmonella cells fail to yield any colonies when transformed with plasmids expressing isrK , constitutively ( Fig 1B ) . To further investigate the toxic phenotype , the isrK gene was cloned under the inducible PBAD promoter and we followed bacterial growth in wild type and a strain deficient of the chromosomal isrK promoter . High levels of IsrK expressed in trans result in growth arrest of wild type Salmonella ( Fig 1C ) . The growth inhibition is not observed when the chromosomal isrK promoter is deleted , suggesting that the genetic regulation leading to toxicity requires expression of the isrK locus in both cis and trans . In addition , we examined whether the IsrK-dependent toxicity involves lysis of the host by Gifsy-1 induction . To this end , we deleted Gifsy-1 genes SL2575 and SL2576 encoding proteins of phage lysis and phage lysozyme superfamily , respectively . Both mutant strains failed to yield any colonies when transformed with plasmids expressing isrK constitutively , indicating that the growth arrest phenotype does not involve Gifsy-1 induction and lysis of the host ( S1 Fig ) . Deletion mapping at the isrK locus to identify the cause of toxicity demonstrated that strains deleted for sequences upstream ( antQ ) or downstream ( anrP ) of isrK showed no growth inhibition , forming normal size colonies when transformed with a plasmid expressing constitutive high levels of IsrK , indicating involvement of both genes ( S2A Fig ) . To define whether any of the above-mentioned genetic elements is toxic when expressed alone , in the absence of IsrK , we transformed strains deleted of the locus including the lysis genes up to antQ with plasmids expressing antQ or anrP from Ptac promoter under the control of the lacI repressor . Whereas cells transformed with plasmids expressing anrP formed regular colonies ( S2B Fig ) , cells expressing antQ fail to grow , indicating that AntQ is sufficient for toxicity . The growth arrest data indicated that toxicity involves IsrK RNA present in cis and expressed in trans , hence , we monitored the effect of IsrK expressed in trans on transcription and translation at the isrK locus . A northern blot probed with an isrK specific primer , detected a long transcript of ~ 900 nucleotides when isrK was expressed in trans ( S3A Fig ) . Probing the northern blot with an isrJ specific primer demonstrated that the long transcript encompasses anrP , suggesting that transcription starting at the isrK promoter reads-through the isrK Rho-independent transcription termination signal downstream into orf45 and anrP . Analysis of shorter RNA species using isrK specific primer supports expression of the plasmid encoded isrK gene ( S3B Fig , lanes 4–9 ) , as well as chromosomally encoded short isrK form , indicating that transcription starting at the isrK promoter produces the IsrK sRNA as well as isrK operon mRNA ( S3B Fig , lanes 1–3 ) . In addition , a lacZ-transcription fusion starting at the isrK promoter ( PisrK-isrK-orf45-anrP'-lacZ ) showed that isrK expressed in trans increased the levels of the long transcript by ~8 fold . The anrP translation fusion also showed that isrK expressed in trans activates translation of anrP by 140 fold ( Table 1B ) . Together these data demonstrate that RNA polymerase partially reads-through the isrK Rho-independent transcription termination signal downstream into orf45-anrP and that IsrK sRNA acting in trans causes a slight increase in the levels of the downstream polycistronic mRNA and activates anrP translation . Gifsy anti-repressor proteins bind to and inactivate the lysogenic repressor , thereby leading to transcription of phage operons [26] . To learn about the correlation between increased levels of IsrK sRNA , the anti-repressor protein AnrP and the anti-terminator AntQ , we monitored antQ mRNA levels upon expression of isrK and anrP , using quantitative Real-Time PCR . This analysis showed that in trans expression of isrK resulted in increased antQ mRNA levels . Similarly , in trans expression of anrP led to higher antQ transcript levels ( Fig 2A and 2B ) . To learn about the activity of the anti-repressor protein AnrP , we measured RNA levels of SL2581 , the second gene of antQ operon . Similarly to antQ , SL2581 levels increased upon expression of isrK or anrP , indicating that AnrP activates transcription of the antQ operon most likely through activated transcription activity ( S4 Fig ) . Together , the results indicate that IsrK activates expression of anrP , which in turn leads to AntQ synthesis . Furthermore , we monitored Gifsy-1 prophage induction upon expression of isrK and anrP ( S5 Fig ) . Phage plating on a susceptible strain demonstrates that Gifsy-1 phage induction by IsrK requires an intact isrK locus , whereas Gifsy-1 induction by AnrP is independent of isrK locus . These results further support the regulatory cascade we present for the isrK locus and the biological relevance of this locus to phage development . We also observed oxidative stress dependent general phage induction ( Materials and Methods and S5C Fig ) , upon which the levels of IsrK sRNA increase during the first minutes of exposure to hydrogen peroxide while antQ and SL2581 mRNA levels increase gradually ( S3D Fig and S6 Fig ) . To visualize the translation pattern of the downstream operon including orf45 and anrP , we integrated the coding sequence of the sequential peptide affinity ( SPA ) tag [27] into orf45 and anrP to generate C-terminal fusion proteins , in two separated strains . The western blot showed that IsrK expressed in trans increases translation of both orf45 and anrP ( Fig 3 ) . orf45 carries two in frame initiation codons and a stop codon overlapping the initiation codon of anrP ( AUGA ) . Nucleotide and amino acid conservation analysis demonstrated that the nucleotide sequence of orf45 is conserved among the enterobacteria ( S7 Fig ) , whereas the amino acid sequence of orf45 varies ( S8 Fig ) . The proximity of orf45 to anrP prompted us to examine their potential translation coupling . By mutating the initiation codons we found that translation starting at the first initiation codon leads to translation of anrP . Moreover , insertion of a stop codon proximal to the translation initiation site reduced anrP translation activation by IsrK ( Table 1C ) . Together , our data indicate that isrK expressed in trans and orf45 translation are required to stimulate anrP translation . To investigate the mechanism of the translational regulation of isrK-orf45-anrP by IsrK , we induced random mutations at this locus using PisrK-orf45-anrP-'lacZ translation fusion plasmid and screened for high-level expression mutants in the absence of in trans isrK . Two high-level expression mutants were found to carry mutations within isrK ( G28A and G31A ) . Structural prediction analysis using RNA fold program ( http://rna . tbi . univie . ac . at/ ) show that the wild type transcript ( 1–180 nt ) forms one conformation ( A ) having a ΔG of -85 . 66 kcal/mol , whereas the mutated RNA forms an alternative conformation ( B ) with a predicted ΔG value of -84 . 21 kcal/mol ( Fig 4A and 4B ) . Functional studies of translation fusions carrying mutations G28A or G31A showed that G28A and G31A , which are predicted to form the alternative structure B , increase the basal level of AnrP translation ( Table 1D ) , indicating that structure B is translationally active , whereas structure A is translationally silent . To visualize the two isoforms and to learn about their ratios in the wild type RNA , we examined the RNAs on nondenaturing polyacrylamide gels . The native gels demonstrate that wild type RNA is found almost exclusively in one structure , while G31A and G28A mutant RNAs display two conformers of which one resembles the wild type conformation and the other represents the alternative structure B ( Fig 4C ) . In the inactive structure formed by wild type RNA ( A ) , the middle part of IsrK ( in purple ) base pairs with ~ 30 nt long sequence of orf45 ( in blue ) forming helix b-I ( Fig 4A and 4B ) . In this structure the ribosome-binding site of orf45 forms hairpin d-I . In the alternative structure ( B ) , the middle part of IsrK forms an alternative hairpin ( b-II ) , whereas the RBS of orf45 forms a new helix by pairing with its 3’-end ( d-II ) . Mutations G28A and G31A are likely to destabilize structure A by disrupting the middle helix , but are predicted to have no effect on structure B . To examine base pairing , we modified helix b opposite to G28A and G31A to carry the corresponding complementary mutations C162U and C159U , respectively and when combined , would restore formation of the helix . RNA mutants carrying G28A/C162U or G31A/C159U exhibit one conformation , the same as wild type RNA ( Fig 4A , 4B and 4C ) , indicating that G28A basepairing with C159U and G31A/C162U basepairing form structure A . Functional studies of translation fusions carrying C162U and C159U showed that like mutations G28A or G31A , C162U and C159U mutants exhibited a high basal level of anrP translation ( Table 1D ) . The basal levels decrease when these mutations are combined with the corresponding complementary mutations further indicating that structure A is translationally inactive , whereas structure B is translationally active . Because mutations C162U and C159U affect the stability of structure B in addition to A , they exhibit a higher basal level of translation than that observed for the opposite mutations ( see below ) . To affirm the differences between the two structures , we constructed mutations G114A and the corresponding complementary mutation C175U , both predicted to destabilize structure B with no effect on structure A . Given that wild type RNA is found almost exclusively in conformation A , these mutations only a mildly affected translation of anrP ( S1 Table ) . Fig 4A shows that in structure A , the middle part of the cis-encoded IsrK ( in purple ) binds a long sequence of orf45 ( in blue ) . Given the complementarity between cis-encoded IsrK and orf45 and the influence of in trans expression of isrK on downstream translation , we explored the functional and structural consequences of IsrK binding to structures A and B , in trans . In binding to structure A , IsrK is predicted to compete with its own sequence for the binding of the middle helix ( Fig 5 ) . Binding of structure B by IsrK is predicted to destabilize the helix d-II that sequesters the RBS of orf45 ( Fig 5 ) . Mutational analysis supported binding of in trans IsrK to the cis-encoded isrK-orf45-anrP target mRNA . Mutations C159U and C162U are predicted to affect base paring with IsrK by replacing CG pairs with UG pairs ( Fig 5 ) . Functional studies of translation fusions of C159U and C162U mutant RNAs demonstrate that wild type IsrK weakly affected translation of anrP , indicating that stable binding of IsrK in trans is important for anrP translation activation and that destabilization of the helix formed between in trans IsrK and cis-encoded orf45 abrogates anrP translation control by IsrK ( Table 1D ) . Similarly , mutation G173A is predicted to affect base paring with wild type IsrK by replacing a GC pair with an AC pair ( Fig 5 ) . Translation fusions studies of orf45 carrying G173A mutation demonstrated that wild type IsrK RNA is less effective in activation of anrP translation than an IsrK mutant carrying the corresponding complementary mutation C18U ( Table 1E ) . Likewise , mutation C175U replaces CG pair with UG pair and anrP translation activation by wild type IsrK is less productive . Moreover , because of imperfect basepairing , IsrK activation of G173A is lower than that of C175U mutant ( S1 Table ) . The effect of IsrK acting in trans is visible in native gels . Incubation of wild type RNA with IsrK at 37°C results in minimal binding of structure A to IsrK ( S8 Fig ) . In accordance , binding of structure A of isrK-orf45G31A RNA by IsrK is indistinct as opposed to binding of the structure B of this RNA mutant ( S8 Fig ) . Pre-incubation of the target RNAs at 70°C to denature the structures , prior to their incubation with IsrK , facilitates binding by IsrK . Under these conditions , IsrK binds structure A , characteristic of wild type RNA and both structure A and B that are characteristic of isrK-orf45G31A ( Fig 6A ) . No binding can be detected when isrK-orf45 wild type RNA is incubated with isrKG31A mutant further confirming that this mutant is inactive as supported by our cultivation experiments ( Fig 6B , S10 Fig ) . Since IsrK binds orf45 adjacent to its RBS , detection of 30S binding in real time ( toe printing ) is inconceivable . However , 30S binding at the RBS would protect the neighboring upstream and downstream sequences . To probe the accessibility of the sequence surrounding the RBS , we used dimethyl sulfate ( DMS ) , which methylates unpaired adenosine and cytidine residues at N1 and N3 positions , respectively . Samples were incubated with and without 30S and/or IsrK prior to the addition of DMS . The modified sites were detected by primer extension after phenol extraction . A few nucleotides that surround the RBS of orf45 are susceptible to DMS modification in the presence of 30S ribosomes or IsrK ( Fig 7 lanes 2 , 4 ) . The same nucleotides are protected from DMS in presence of both IsrK and 30S ( Fig 7 lane 6 ) , indicating that wild type IsrK facilitates 30S binding to the RBS of orf45 . 30S protection from DMS decreases much in the presence of isrKG31A mutant that is unable to bind orf45 ( S11 Fig ) . On the one hand , IsrK is part of the target isrK-orf45-anrP mRNA , and as such mutations in IsrK affect the structure this mRNA forms . For instance , G28A disrupts helix b in structure A , shifting the equilibrium towards the translationally active structure B . Therefore , mRNA carrying isrKG28A-orf45-anrP exhibits a high basal level of anrP translation . On the other hand , the short IsrK acts in trans to destabilize the cis-encoded translationally inactive target mRNA leading to anrP translation . Therefore , one mutation in isrK gene is predicted to yield two different phenotypes . We examined the effect of isrKG28A in cis and in trans and found that whereas , cis-encoded isrKG28A-orf45-anrP exhibits a high basal level of anrP translation; isrKG28A acting in trans is unable to activate anrP translation ( Table 1F ) . In accordance , high levels of isrKG28A or isrKG31A expressed from the PBAD promoter have no effect on growth of wild type Salmonella ( S10 Fig ) . We have shown that toxicity involves expression control of anrP by IsrK that in turn induces transcription of the anti-terminator protein AntQ ( Fig 2A and 2B ) . Considering the origin of antQ , i . e . , Gifsy-1 prophage , we examined whether its toxicity is specific to Salmonella and/or phage genes . Accordingly , the influence of high levels of AntQ was investigated in two E . coli K-12 strains; wild type ( MG1655 ) and MDS42 that is deleted of all genetic islands including prophages and insertion elements [28] . antQ expression repressed growth and decreased survival of both strains . At 40 minutes of induction , survival of wild type and MDS42 decreased by ~15 and ~30 fold , respectively , indicating that toxicity of the Q-like anti-terminator protein is not specific to Salmonella or phage DNA ( S12 Fig ) . Moreover , the results suggest that AntQ protein has natural recognition sites within the core genome of these strains . Bacteriophages Q antiterminator proteins interfere with transcription termination by binding specified sites at promoter regions and forming a persistent complex with RNA polymerase . This complex of RNA polymerase and Q protein can bypass terminators [24] . We examined changes in protein expression pattern of Salmonella upon exposure to AntQ using one-dimensional SDS-PAGE . Gel areas showing differences in the pattern of proteins because of AntQ were isolated and subjected to mass spectrometry ( S13 Fig ) . Two proteins , whose expression increased , were selected for further analysis because of their score and annotation; the transcription termination factor Rho and the DEAD-box-containing ATP-dependent RNA helicase SrmB [29–32] . We suspected that expression of rho and srmB increased in response to transcription elongation related stress . Thus , we examined whether co-expression of antQ with these genes would abolish the toxic effect of AntQ . Survival assays show that co-expressing rho with antQ help to rescue cells from AntQ-mediated toxicity . After 40 minutes of induction , survival of cells expressing antQ alone dropped to ~8% of their original amount , whereas cells expressing both antQ and rho managed to maintain a high CFU count , indicating that Rho can halt the toxicity inflicted by AntQ . Likewise , survival assays in which srmB and antQ were co-expressed demonstrated that SrmB prevented AntQ toxicity ( Fig 8A and 8B ) . Together the results show that proteins that harbor RNA helicase activity impede the toxic effects of transcription anti-termination . Unregulated transcription elongation increases formation of DNA-RNA hybrids upstream of RNA polymerase ( R-loops ) . The resulting R-loops may initiate DNA replication independently of oriC , leading to DNA damage [33 , 34] . RNase H is an evolutionary conserved helicase that resolves R loops , thus protecting genomic DNA from breaks [30 , 35] . Survival rates of cells co-expressing antQ and rnhA encoding RNase H were 10 fold higher than those expressing antQ alone , suggesting that the toxic effects of AntQ result from DNA damage due to the creation of R-loops ( Fig 8C ) . It is well documented that exposure of bacteria to detrimental stressful conditions impairing protein synthesis or causing DNA damage , triggers genome condensation [36] . We visualized the effect of antQ expression on chromatin morphology by florescence microscopy . Images of Salmonella expressing a control plasmid display , as excepted , chromatin spread over the entire cytoplasm . In contrast , images of Salmonella expressing antQ reveal condensed chromatin morphology . Likewise , exposure of Salmonella cells to nalidixic acid ( NA ) , a pleiotropic drug that inflicts diverse DNA lesions ( nicks , gaps , and DSBs ) [36] resulted in genome condensation ( Fig 9 ) . Furthermore , E . coli cells expressing antQ or exposed to NA exhibit genome condensation , similarly to Salmonella , indicating that antQ toxicity is mechanistically conserved . The E . coli UvrD protein is a DNA helicase/translocase that functions in methyl-directed mismatch repair ( MMR ) nucleotide excision repair ( NER ) and more broadly in genome integrity maintenance [37] . Recent studies in E . coli have shown that UvrD can act as an accessory replicative helicase that resolves conflicts between the replisome and transcription complexes [37–39] . Using uvrD-yfp [40] , we demonstrate in vivo localization of the fluorescently tagged uvrD to the nucleoid upon expression of antQ and upon exposure to DNA damaging agents ( Fig 10 ) . Together our data show that the function of the phage antiterminator protein is wide-ranging causing changes in bacterial chromatin morphology .
In this study we show that a subset of the IsrK sRNA transcripts reads through its transcription terminator to form a translationally inactive bi-cistronic mRNA . Concomitantly , short IsrK RNA acts in trans , interacting with the inactive transcript to promote formation of a translationally active structure , in which orf45 translation leads to anrP translation by translational coupling ( model Fig 11 ) . In bacteria , translational coupling provides a mechanism to coordinate expression of multiple proteins with adjacent or overlapping coding sequences . Ribosomes terminating translation of upstream ORF dissociate and re-initiate translation at the downstream RBS [41 , 42] . Re-initiation is enabled due to ribosomes elongating along the upstream ORF to unfold mRNA structures that sequester the downstream ribosome-binding site . Such an example is PhrS sRNA that activates translation of pqsR mRNA by interaction with a sequence sequestering the RBS of an ORF upstream of pqsR [43] . We find that inserting a stop codon proximal to the translation initiation site of orf45 reduces IsrK-controlled anrP translation activation , indicating that translational coupling is necessary for AnrP synthesis . Given that the structural changes caused by IsrK and/or ribosome binding at the orf45 RBS do not seem to involve structural changes in the RBS of anrP , we suggest that the translational coupling between orf45 and anrP requires ribosome elongation from the RBS of the orf45 downstream to anrP . A structural homolog of IsrK is SeqA RNA of P4-like phages [44] . In the lysogenic state P4 prevents expression of its own replication genes by premature transcription termination . The factor responsible for efficient termination is CI RNA that is generated by processing of a primary untranslated transcript . CI RNA acting as an antisense RNA leads to transcription termination by pairing with two complementary sequences , seqA and seqC located upstream and downstream of CI , respectively [45–47] . In Salmonella , transcriptome analysis revealed the existence of a stable non-coding RNA species downstream of IsrK ( STnc1160 ) [8] . Our RNA analysis detected STnc1160 in wild type cells but not in a strain deleted of the isrK promoter , suggesting that STnc1160 is generated by processing of the readthrough transcript initiating from the isrK promoter ( S3 Fig ) . It is possible that STnc1160 similarly to CI modulates transcription termination at IsrK Rho-independent terminator . Since IsrK and STnc1160 share complementary sequences , IsrK binding of STnc1160 renders it inactive as a termination factor leading to transcription readthrough and thus to grow arrest . However , in experiments of co-expression of isrK and orf45 ( STnc1160 ) in which STnc1160 was constitutively expressed , IsrK mediated growth arrest was even more pronounced ( S14 Fig ) . In S3B Fig , we present a northern blot showing the levels of chromosomally and plasmid encoded isrK . It is interesting to note that in addition to short IsrK , our analysis revealed a stable transcript ( isrK-orf45’ ) that is generated by processing of the long polycistronic transcript . isrK-orf45’ species is observed upon expression of plasmid-encoded isrK in wild type cells ( lane 6 ) , as well as upon expression of chromosomally-encoded isrK ( see lane 1–3 ) , indicating that the pattern detected with high level expression is valid with chromosomally-encoded isrK . In addition , the results demonstrating that Gifsy-1 phage induction by IsrK requires an intact isrK locus , whereas Gifsy-1 induction by AnrP is independent of isrK locus , further substantiate the regulatory cascade we present for the isrK locus and signify the biological relevance of this locus to phage activation . Moreover , we show that wild type cells grown in minimal medium to stationary phase exhibit prophage induction , whereas isrK promoter deletion mutant ( ΔPisrK::frt ) fails to produce phages ( S15 Fig ) . These findings indicate that IsrK is an important player in initiating prophage induction . Concerning the conditions inducing isrK expression , we find that IsrK levels increase at stationary phase and under low Mg2+ conditions ( S15 Fig ) . Salmonella global transcriptome analysis carried out by Kröger et al [8] shows that IsrK levels increase during conditions such as low Fe2+ shock , oxygen shock and growth in InSPI2 medium . In addition they find that the levels of the transcript encoding orf45 resulting by transcription elongation through the isrK transcription terminator ( STnc1160 ) increase during low Fe2+ shock , InSPI2 and late stationary phase . Together , the data indicate that isrK short and long forms are produced under a variety of environmental conditions . The majority of the sRNA genes are encoded within intergenic regions acting in trans to control expression of physically unlinked target genes . However , it is now increasingly appreciated that in addition to intergenic regions , many sRNAs originate from the 5’ or 3’ regions of coding mRNAs . Such examples are 3’ UTR derived sRNAs that are generated either by internal processing of the related mRNA , as in the case of RybD or produced as a primary transcript like MicL and DapZ [7 , 48] . Generated from within protein coding loci , these sRNAs act in trans controlling expression of unlinked target mRNAs . Likewise , SreA and SreB originate from 5’ UTRs of two S-adenosylmethionine ( SAM ) riboswitches , and base pair with the unlinked prfA mRNA to repress translation [49] . In Staphylococcus aureus , SprA1AS is transcribed from the strand opposite to SprA1 target mRNA encoding pepA1 ORF . The antisense RNA SprA1AS acts in trans by base pairing with the 5’ domain of SprA1 to repress pepA1 translation by occluding its RBS [50] . Somewhat different is the archaeal RNA regulator , sRNA162 . sRNA162 masks the RBS of MM2441 by binding MM2440-MM2441 mRNA internally [51] . Biochemical studies demonstrated that in addition to in trans binding of MM2441 RBS , encoded opposite of MM2442 , the 5’-end of sRNA162 targets the 5’-untranslated region of the cis-encoded MM2442 mRNA . However , the regulatory outcome of this interaction is as yet unknown . The mechanism of expression regulation of the isrK locus is unique , representing the first example of an RNA that acts as a small RNA on its own mRNA . On the one hand , IsrK is part of a translationally inactive target mRNA , whereas on the other hand the short RNA species acts in trans to enable translation of the target mRNA . Therefore one mutation within IsrK RNA yields two different phenotypes; when located in the long target mRNA it increases translation whereas the short mutant IsrK RNA can no longer activate translation . Increasing evidences indicate that in prokaryotes and eukaryotes , common transcription-replication encounters lead to blockage of replication that is often accompanied by DNA damage and genome instability . In bacteria , because replication and transcription proceed simultaneously on the same template DNA , yet DNA replication forks move 10 to 30 times faster than do RNA polymerases , both co-directional , and head-on collisions appear to be unavoidable [34] . Transcription-replication conflicts may also result from stalled transcription elongation complexes , as they form stable barriers to the replication machinery . These complexes increase the production and/or the length of DNA-RNA hybrid structures within the transcription bubble , causing the region of complementary single-stranded DNA to loop out . The resulting R-loops may initiate DNA replication independently of oriC , leading to DNA damage [34] . We find that expression of the antiterminator protein results in bacterial growth arrest and ultimately cell death . Co-expression of antQ with transcription termination factor Rho rescues cells from the toxic effects of the antiterminator protein . Likewise , survival rates of cells co-expressing antQ and rnhA encoding RNase H were 10 fold higher than those expressing antQ alone . Given that RNase H protects genomic DNA from breaks by resolving R-loops , suggests that the toxic effects of AntQ result in part from DNA damage due to the creation of R-loops . In accordance , we find that expression of antQ affects bacterial chromatin morphology . Fluorescence microscopy images of E . coli and Salmonella expressing the antiterminator protein reveal condense chromatin morphology . Fluorescently tagged UvrD localizes to the nucleoid upon expression of the antiterminator protein as well as upon exposure to NA , a DNA-damaging agent . uvrD is a member of DNA helicase superfamily 1 and part of the SOS regulon . During SOS response the intracellular level of UvrD increases approximately three fold . UvrD functions in methyl-directed mismatch repair ( MMR ) nucleotide excision repair ( NER ) and genome integrity maintenance . Recent studies have demonstrated that UvrD contributes to genomic integrity by resolving conflicts between transcription and DNA repair complexes [37–39] . UvrD binds RNA polymerase in blocked transcription elongation complexes , forcing it to slide backwards along the DNA . This backwards sliding exposes DNA lesions that are out of reach allowing the nucleotides excision repair enzyme to access the site of damage . A question remains , are the massive effects on chromatin structure detected upon IsrK/AntQ overproduction physiologically relevant ? Quantitation analysis of antQ mRNA levels indicates that at around two hours upon exposure to in trans IsrK , during the time that the long transcript isrK-orf45-anrP is detected , the copy number of antQ is increased by 6 fold ( 2 . 5x10-2 per 16S rRNA ) . During the first nine minutes of exposure to hydrogen peroxide the copy number of antQ increases by 4 fold ( 1 . 2x10-2 per 16S rRNA ) . Thus , it is conceivable that the phenotype observed with overproduction of antQ is biologically relevant . Regardless , the toxic phenotype of antQ is intriguing , a phage encoded antiterminator protein known to facilitate transcription antitermination at a specific phage terminator , is in fact , wide-ranging capable of affecting bacterial core genome sites . Given the morphological appearance of cells expressing the antiterminator protein and that the toxic phenotype is reversed by opposing functions , including transcription termination and elimination of R loop , AntQ , by affecting transcription elongation causes DNA damage . Whether the phage exploits antQ to modulate vital bacterial machineries or bacterial cells use the Q-like antiterminator core genome native sites for self-inhibition upon phage infection remains to be addressed in further studies .
Salmonella Typhimurium SL1344 cells were grown at 37°C ( 200 rpm ) in LB medium ( pH 6 . 8 ) . Ampicillin ( 100 μg/ml ) , Chloramphenicol ( 20 μg/ml ) and kanamycin ( 40 μg/ml ) were added where appropriate . Induction of PBAD promoter was obtained with arabinose ( 0 . 2% ) , whereas Ptac promoter was induced with IPTG as indicated . ( List of strains , plasmids and DNA primers used in this study appear in S2–S4 Tables ) Gene deletion mutants were generated using the gene disruption method as described [52] . For construction of deletion mutants , chloramphenicol or kanamycin cassettes were amplified from plasmids pKD3 and pKD4 , respectively [53] . The PCR product ( 5–10 μg ) purified using the Wizard SV PCR clean-up system ( Promega , Madison , WI ) was introduced into arabinose treated LB5010 cells [54] carrying pKD46 cells [52] and chloramphenicol or kanamycin-resistant colonies were selected . The deletion mutation was transferred into a wild type SL1344 genetic background by transduction using the P22 bacteriophage . The resistance gene was eliminated using pCP20 [53] . In ΔanrP::kan , the chromosomal region flanked by genome coordinates 2759753 and 2760316 ( GenBank entry CBW18679 . 1 ) was replaced by the kan gene using primers 1571 and 1572 . anrP gene disruption was examined by PCR using flanking primers , 1528 and 1529 . In ΔantQ::cat , the chromosomal region flanked by genome coordinates 2760589 and 2761266 ( GenBank entry CBW18680 . 1 ) was replaced by the cat gene using primers 1573 and 1458 . antQ gene disruption was examined by PCR using flanking primers , 1542 and 1486 . In Δ ( SL2575-SL2576 ) ::cat and Δ ( SL2575-SL2576 ) ::kan the chromosomal region flanked by genome coordinates 2756646 and 2757568 ( GenBank entry CBW18677 . 1 and CBW18676 . 1 ) was replaced by the cat gene using primers 1615 and 1616 , or kan cassette using primers 2149 and 2150 . ( SL2575-SL2576 ) gene disruption was examined by PCR using flanking primers , 1617 and 1618 . In SL1344 antQ::cat Δ ( SL2575-SL2576 ) ::kan double deficient mutant , a P22 lysate generated from ΔantQ::cat was transferred into Δ ( SL2575-SL2576 ) ::kan by transduction and the resistant genes were later eliminated using pCP20 to generate a mutant deficient of the entire region SL1344 antQ to ( SL2575-SL2576 ) ::frt . The disruption was examined by PCR using flanking primers , 1761 and 1618 . To construct strains carrying SPA tags in the chromosome; SL1344 orf45-SPA-kan and SL1344 anrP-SPA-kan , primers were designed to amplify the sequential peptide affinity ( SPA ) tag together with the kanamycin resistance cassette from the plasmid pJL148 , and flanked by 45 nt of sequence homologous to the insertion region , as described before . SL1344 orf45-SPA-kan and SL1344 anrP-SPA-kan were constructed using primers 2122 , 2123 and 2209 , 2210 respectively . The PCR products were purified from gels and used to transform SL1344 cells carrying pKD46 plasmid [52] . Insertions were confirmed by sequencing of PCR products generated using primers 1987 , 2227 ( 198 nt of orf45-SPA-kan ) and 839 , 2227 ( 196 nt of anrP-SPA-kan ) . The products were sequenced using primer 2227 . uvrD-yfp fusion [40] was transferred into strain RW118 [55] by P1 transduction . To construct Ptac-isrK and PBAD-isrK , isrK sequence from its transcription start site plus 37 nt downstream of its transcription terminator was PCR amplified from SL1344 chromosomal DNA using primers 1364 and 1365 and cloned into the EcoRI and HindIII restriction sites of pRI and pJO244 respectively . To construct Ptac-antQ-lacI , antQ sequence from its ATG plus 22 nt downstream of its stop codon was PCR amplified from SL1344 chromosomal DNA using primers 1544 and 1510 and cloned into the EcoRI and SalI sites of pKK177-3-lacI . In this plasmid , antQ translation is directed by an artificial translation initiation signal found in the right position in pKK177-3-lacI . To construct Ptac-anrP-lacI , anrP sequence from its ATG plus 40 nt downstream of the stop codon was PCR amplified from SL1344 chromosomal DNA using primers 1893 and 1897 and cloned into the EcoRI and PstI sites of pKK177-3-lacI . In this plasmid , anrP translation is directed by an artificial translation initiation signal found in the right position in pKK177-3-lacI . To construct Prho-rho ( p15A origin ) , a DNA fragment including both the promoter of Rho and the rho gene ( -200 to 112 nt downstream of the stop codon ) was PCR amplified from SL1344 chromosomal DNA using primers 1872 and 1862 and cloned into the BglII and HindIII sites of pACYC184 . To construct PBAD-srmB ( p15A origin ) , a DNA fragment containing srmB from nucleotide146 upstream of the ATG to nucleotide 2 downstream of the stop codon was PCR amplified from SL1344 chromosomal DNA using the primers 1884 and 1885 and cloned into the PstI and HindIII sites of pEF21 . In this plasmid , srmB translation is directed by its own translation initiation signal . To construct PrnhA-rnhA ( p15A origin ) , a DNA fragment including both the promoter of rnhA and the rnhA gene ( -112 to 89 nt downstream of the stop codon ) was PCR amplified from SL1344 chromosomal DNA using primers 1907 and 1908 and cloned into the XbaI and BamHI sites of pACYC184 . To construct lacZ fusions in single copy plasmids ( pBOG551 and pBOG552 ) , the origin of replication of pRS551 and pRS552 [56] was replaced by the origin of replication of pZS*24 ( pSC101* ) [57] . The origin of replication was amplified using primers 2032 and 2042 and cloned into the PstI and SalI sites of pRS551 and pRS552 plasmids . To construct wild type fusions PCR fragments carrying PisrK-orf45-anrP were amplified from genomic DNA using primers 1512 and 1703 , digested and cloned into the EcoRI and BamHI sites of pGEM3 . The PisrK-orf45-anrP fragment was sub-cloned into pBOG551 and pBOG552 using the EcoRI and BamHI sites . All fusion mutants were constructed by transferring mutated PisrK-orf45-anrP fragments from pGEM3 into pBOG plasmids , as described above . To construct PLtetO-1-orf45 ( p15A origin ) , we first deleted the luc gene of pZA31-luc by PCR using primers 1989 ( KpnI ) and 1990 ( phosphorylated ) . Thereafter , orf45 sequence from its second ATG to its stop codon was PCR amplified from SL1344 chromosomal DNA using primers 1987 ( KpnI ) and 1988 ( phosphorylated ) . The two PCR fragments ( orf45 and pZA31 ) where then ligated . To carry out random mutagenesis , 1 volume ( 1 . 5 μg ) of plasmid DNA carrying PisrK-orf45-anrP-'lacZ ( pSA81 ) was mixed with 5 volumes of phosphate solution ( 0 . 5 M NaH2PO4 , 1 mM EDTA , adjusted to pH 6 with NaOH ) , and 4 volumes of hydroxylamine solution ( 1 M hydroxylamine hydrochloride ( Fluka ) in phosphate solution , adjusted to pH 6 with NaOH ) . The mixture was incubated at 65°C for 2 hours , then dialyzed overnight against TE buffer ( 10 mM Tris-HCl at pH 8 , 1 mM EDTA ) at 4°C and again for 2 . 5 hours . The mutagenized plasmid was used to transform MC4100 cells . Dark blue colonies were picked from LB plates containing 40 μg/ml 5-bromo-4- chloro-3-indolyl-β-D-galactopyranoside ( x-gal ) ( Inalco ) . Plasmid inserts from selected colonies were sequenced [56] . Mutations A107C , AU119-120UA , G28A , C162U , G31A , C159U , G114A , G173A and C175U were generated by PCR using plasmid carrying PisrK-orf45-anrP ( pSA77 ) and two tail-to-tail divergent primers of which one carried the desired mutation . The PCR product was gel purified , subjected to blunt end ligation and the mutated plasmid was digested with EcoRI and BamHI for sub-cloning to pBOG551 and pBOG552 . The double mutants A107C/G121A , G28A/C162U , G31A/C159U and G114A/C175U were constructed using PisrK-orf45A107C-anrP ( pSA77A107C ) , PisrKG28A-orf45-anrP ( pSA77G28A ) , PisrKG31A-orf45-anrP ( pSA77G31A ) and PisrK-orf45G114A-anrP ( pSA77G114A ) , respectively as template and two tail-to-tail divergent primers of which one carried the desired second mutation . The PCR product was gel purified , subjected to blunt end ligation and the mutated plasmid was digested with EcoRI and BamHI for sub-cloning to pBOG551 and pBOG552 [56] . Overnight cultures of S . typhimurium SL1344 or SL1344 carrying plasmids were grown from fresh transformation plates . Each strain was grown in duplicates . Starters were diluted 1/100 in 15 ml LB ( 125 ml Erlenmeyer flasks ) and grown at 37°C ( 200 rpm ) . Arabinose ( 0 . 2% ) was added at the time of dilution where indicated . One hour after dilution , IPTG was added to a final concentration of 0 . 2 mM to induce expression of antQ . Samples were taken prior to , 20 and 40 minutes after the addition of IPTG , diluted in 1X PBS and plated . Each sample was plated twice . Colonies were counted and percentage of survival rate was calculated . Overnight cultures were diluted 1/100 in 20 ml LB medium supplemented with ampicillin and kanamycin , and grown to OD600 ~ 1 . 0 . To induce IsrK , arabinose ( 0 . 2% ) was added at the time of dilution . β-galactosidase activity was assayed as described [58] . To detect Gifsy-1 phage induction by AnrP , overnight cultures of S . typhimurium wild type , isrK promoter deletion mutant ( ΔPisrK::frt ) , and Gifsy-1 lysis proteins deletion mutant ΔSL2575-SL2576::frt carrying Ptac-lacI or Ptac-anrP-lacI were diluted ( 1/100 ) in LB medium supplemented with 10 mM MgSO4 and ampicillin and grown at 37°C to OD600 ~ 0 . 3 . Thereafter , IPTG ( 0 . 2 mM ) was added to induce expression of anrP . At 2 hr after induction the cultures were treated with chloroform to release phage particles . 5μ of the supernatant were plated on LT2 ( lambda sensitive ) bacterial lawn made with soft agar . To detect Gifsy-1 phage induction by IsrK , wild type and ΔPisrK::frt mutant carrying PBAD and PBAD-isrK were diluted , grown and their phages plated as above . Arabinose ( 0 . 2% ) was added at the time of dilution for induction of IsrK . To detect oxidative stress dependent phage induction [15 , 59] H2O2 ( 0 . 1 and 0 . 5 mM ) was added at OD600 ~ 0 . 3 and phages were plated as above . To measure the effect of AnrP on expression at antQ locus , ΔSL2575-SL2576::frt cells carrying Ptac-lacI or Ptac-anrP-lacI were grown to OD600 ~ 0 . 5 . Total RNA was extracted prior to and at 25 min of induction with IPTG ( 1 mM ) . To measure the effect of IsrK on expression at antQ locus , wild type cultures carrying PBAD and PBAD-isrK ( treated with arabinose ( 0 . 2% ) at the time of dilution ) were grown and total RNA was extracted at time points as indicated . To measure expression at antQ locus upon phage induction , H2O2 ( 0 . 1 mM ) was added at OD600 ~ 0 . 3 . Total RNA was extracted prior to and upon exposure to H2O2 ( as indicated ) and cDNA was prepared for real time PCR . To monitor expression of chromosomally encoded-isrK , wild type and ( ΔPisrK::frt ) mutant strains were grown in LB medium to OD600 ~ 0 . 3 , 0 . 6 , 1 . 0 and for 8 hours or in low MgCl2 N-minimal medium to OD600 ~ 0 . 3 [6] . Overnight cultures of S . typhimurium SL1344 or SL1344 carrying plasmids were diluted and grown as described before . To isolate total RNA , the cultures were pelleted and re-suspended in 50 μl 10 mM Tris–HCl ( pH 8 ) containing 1 mM EDTA . Lysozyme was added to 0 . 9 mg/ml and the samples were subjected to three freeze-thaw cycles . Total RNA was purified using TRI reagent ( Sigma ) according to the manufacturer’s protocol . RNA concentrations were determined using a NanoDrop machine ( NanoDrop Technologies ) . DNA was removed by DNase treatment according to the manufacturer’s instructions ( RQ1 RNase free DNase , Promega ) . About 1 μg DNA-free total RNA was used for cDNA synthesis using MMLV reverse transcriptase and random primers ( Promega ) . Quantification of cDNA was performed by real-time PCR using SYBR-green mix ( Absolute SYBR GREEN ROX MIX , ABgene ) with Rotor gene 3000A ( Corbett ) according to manufacturer’s instructions . Specific primer pairs were designed according to the Guidelines for Amplicon and Primer Design ( http://www . tamar . co . il/tamar-laboratory-supllies/guidelines-amplicon-primer-design/ ) . The level of 16S rRNA ( rrsA ) was used to normalize the expression data for each target gene . The relative amount of cDNA was calculated using the standard curve method . A standard curve was obtained from PCR on serially diluted genomic DNA as templates and was analyzed using Rotor-gene analysis software 6 . 0 . Overnight cultures of wild type cells carrying a control plasmid ( Ptac ) or an AntQ expressing plasmid were diluted 1/100 and grown at 37°C . IPTG ( 1 mM ) was added at OD600 ~ 0 . 2 for AntQ induction . Samples were taken 30 and 60 minutes after exposure to IPTG , pelleted and then fluidized in 1X Laemmli sample buffer , heated at 95°C for 5 min and centrifuged for 5 min . 20 μl of each sample were analyzed by 12% SDS-PAGE ( 9 mA , 1X running buffer for 24 hours at 4°C ) . To visualize the proteins , the gel was stained for 30 minutes at 37°C ( coomassie blue staining ) . The mass spec data of the band ( see S13 Fig ) were analyzed based on coverage , which represents the percentage of the protein that was sequenced; area , which describes the fraction of the specific protein out of the sample , and the number of unique peptides , found to only match this specific protein . The high scored proteins were considered for further analysis based on their relevant function in transcription elongation or termination . Overnight cultures of SPA-tagged strains carrying control plasmid ( PBAD ) or IsrK expressing plasmid ( PBAD–IsrK ) were diluted 1/100 and grown shaking at 37°C . Arabinose ( 0 . 2% ) was added at the time of dilution . Samples were taken at indicated OD600 , pelleted and then fluidized in 1X Laemmli sample buffer , heated at 95°c for 5 min and centrifuged for 5 min . Samples of 3x107 cells were analyzed on SDS-PAGE ( 12% and 15% for ORF45-SPA and AnrP-SPA , respectively ) [60] . The proteins were transferred to a nitrocellulose membrane ( Invitrogen ) , the blots were blocked with skim milk ( 2 . 5% for 1 hour ) and probed with FLAG M2-AP monoclonal antibody ( Sigma-Aldrich ) according to the manufacturer’s protocol . The tagged proteins were visualized using secondary antibody Anti-Mouse IgG-Alkaline Phosphatase ( Sigma-Aldrich ) based on Alkaline Phosphatase development protocol . The RNAs; isrK-orf45 wild type ( from transcription start site to nucleotide 217 within orf45 or to nucleotide 785 at the end of anrP ) and mutants: isrKG31A-orf45 ( 217 nt ) , isrKG31A-orf45C159U ( 217nt ) , isrKG28A-orf45 ( 217nt ) , isrKG28A-orf45C162U ( 217nt ) as well as IsrK sRNA ( 90 nt ) wild type and G31A mutant were synthesized with phage T7 RNA polymerase ( 25 units; New England Biolabs ) in 50 μl reactions containing 40 mM Tris-HCl ( pH 7 . 9 ) , 6 mM MgCl2 , 10 mM diothiothreitol ( DTT ) , 20 units RNase inhibitor ( CHIMERx ) , 500 μM of each NTP and 200 ng of purified PCR templates carrying the sequence of the T7 RNA polymerase promoter . Synthesis was allowed to proceed for 2 hours at 37°C , and was terminated by phenol/chloroform extraction and ethanol precipitation in the presence of 0 . 3M ammonium acetate . RNA samples ( 20 μg for detection of isrK and 30 μg for detection of STnc1160 and orf45 ) were denatured for 5 min at 65°C in 98% formamide loading buffer , separated on 8 M urea-6% polyacrylamide gels and transferred to Zeta Probe GT membranes ( Bio-Rad Laboratories ) by electroblotting . To detect IsrK RNA , the membranes were hybridized with [32P]-end-labeled isrK primer ( 1197 ) in modified CHURCH buffer [6] . STnc1160 and orf45 were detected using anti-STnc4100 labeled riboprobe synthesized using PCR template ( 2316 and 2317 ) as previously described [6] . Riboprobe hybridization buffer contained 50% formamide , 3 . 5% SDS , 250 mM NaCl , 82 mM Na2HPO4 , 40 mM NaH2PO4 at pH 7 . 2 . After 2 hours at 50°C , the membranes were treated for 20 min at 50°C in 2X SSC , 1% SDS , 20 min at 55°C in 1X SSC , 0 . 5% SDS and 20 min at 60°C in 0 . 5X SSC , 0 . 1% SDS . To detect isrK-orf45-anrP full length RNA , samples ( 20 μg ) were denatured for 10 min at 65°C in MOPS loading buffer , separated on 1 . 2% agarose gels and transferred to Zeta Probe GT membranes by capillary transfer [6] . The membranes were hybridized in modified CHURCH buffer using end-labeled isrK ( 1197 ) or isrJ ( 1471 ) specific primers . To detect binding of IsrK to its templates in vitro synthesized RNAs wild type isrK-orf45 and mutants ( 217 nt , 0 . 2 pmol ) were incubated in 10 μl of Native Buffer ( 6 . 7 mM Tris-acetate ( pH 7 . 4 ) , 3 . 3 mM Na-acetate , 1 mM DTT and 10 mM MgCl2 ) for 3 minutes at 70°C and chilled on ice . Thereafter , in vitro synthesized IsrK or IsrKG31A RNAs were added as indicated in the Fig and incubated for 15 minutes at 37°C . The RNA samples were analyzed on 5% non-denaturing polyacrylamide gels ( 19:1 ) run at 50 volts in 20 mM Tris-HCl ( pH 7 . 5 ) , 60 mM KCl and 10 mM MgCl2 for 5–6 hours at 4°C as described before [56] . After its transfer to nylon membrane by electroblotting , the RNA was detected by probing with end-labeled orf45 specific primer ( 1948 ) . To detect RNA conformations the template RNAs wild type and mutants ( 217 nt , 0 . 2 pmol ) were incubated in Native Buffer ( as above ) for 15 minutes at 37°C and then analyzed as describe above . The RNAs were also analyzed on denatured gels ( see Northern analysis ) using end-labeled orf45 specific primer that detects isrK-orf45 templates ( 1948 ) . Template RNA synthesized in vitro ( 0 . 7–1 pmol ) was incubated in 50 mM Na-cacodylate ( pH 7 . 4 ) , 10 mM magnesium acetate , 100 mM NH4Cl and 2 . 5 mM β-mercaptoethanol for 3 minutes at 70°C and chilled on ice for 10 min . Thereafter , in vitro synthesized wild type IsrK or IsrKG31A ( 7–10 pmol ) was added and the mixture was incubated for 15 minutes at 37°C . Pre-activated ( 30 minutes at 37°C ) 30S ribosomal subunits ( 1 . 2–2 . 4 pmol ) were added for 5 minutes prior to the addition of uncharged fMet-tRNA ( 12 pmol ) . The binding reactions were incubated for 15 min before DMS ( 0 . 5 μl , diluted 1:10 in ethanol ) was added . The modification reaction was allowed to proceed for 5 minutes . Reactions were stopped with phenol/chloroform and precipitated with ethanol in the presence of 0 . 3 M sodium acetate , 1 μl of Quick-Precip ( Edge BioSystems ) and 20 μg yeast RNA . The modification sites were detected by primer extension using MMLV reverse transcriptase ( Promega ) and end-labeled primer ( 1948 ) . Fluorescence microscopy was carried out as described previously [61] . In brief , 1–2 ml cells were centrifuged , washed with 1X phosphate buffered saline ( PBS ) and finally re-suspended in 10–100 μl of PBS . The membrane was stained with FM4-64 ( Molecular Probes , Invitrogen ) at a final concentration of 10 μM or 1 μg/ml , respectively . DNA was stained with DAPI ( Sigma-Aldrich ) at a final concentration of 2 μg/ml . Cells were washed twice before microscopic examination . Cells were visualized and photographed using Nikon Eclipse Ti-E inverted microscope equipped with Perfect Focus System ( PFS ) and ORCA Flash 4 camera ( Hamamatsu photonics ) . Images were processed using NIS Elements-AR software .
|
As the function of conserved core-genome-encoded small RNAs ( sRNA ) reflects the basic lifestyle of bacteria , the function of non-conserved island-encoded sRNAs remains enigmatic . The island-encoded sRNA IsrK belongs to Gifsy-1 prophage of Salmonella . Here , we report a complex mechanism in which the IsrK RNA functions as both sRNA and mRNA to control the production of the toxic AntQ protein . The isrK promoter directs the synthesis of two distinct RNA species: a full-length translationally inactive target mRNA and the correctly terminated , shorter IsrK sRNA . IsrK sRNA binds the full-length inactive mRNA producing an antiterminator protein , AntQ , which interferes with transcription termination . Expression of antQ results in bacterial growth arrest and ultimately cell death . Fluorescence microscopy of E . coli and Salmonella expressing antQ revealed condensed chromatin morphology as observed upon exposure to DNA-damaging agents . We propose that expression of the phage antiterminator protein results in conflicts between transcription and replication machineries and thus facilitates DNA damage . In summary , the RNA regulator IsrK presents a new regulatory principle in which a horizontally acquired sRNA controls genome integrity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"nucleic",
"acid",
"synthesis",
"pathology",
"and",
"laboratory",
"medicine",
"bacteriophages",
"pathogens",
"messenger",
"rna",
"microbiology",
"dna",
"transcription",
"viruses",
"bacterial",
"diseases",
"enterobacteriaceae",
"molecular",
"biology",
"techniques",
"rna",
"synthesis",
"bacteria",
"bacterial",
"pathogens",
"chemical",
"synthesis",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"artificial",
"gene",
"amplification",
"and",
"extension",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"denaturation",
"molecular",
"biology",
"salmonella",
"biosynthetic",
"techniques",
"biochemistry",
"rna",
"nucleic",
"acids",
"polymerase",
"chain",
"reaction",
"protein",
"translation",
"genetics",
"biology",
"and",
"life",
"sciences",
"rna",
"denaturation",
"organisms"
] |
2016
|
Gifsy-1 Prophage IsrK with Dual Function as Small and Messenger RNA Modulates Vital Bacterial Machineries
|
The spirochete Borrelia recurrentis is the causal agent of louse-borne relapsing fever and is transmitted to humans by the infected body louse Pediculus humanus . We have recently demonstrated that the B . recurrentis surface receptor , HcpA , specifically binds factor H , the regulator of the alternative pathway of complement activation , thereby inhibiting complement mediated bacteriolysis . Here , we show that B . recurrentis spirochetes express another potential outer membrane lipoprotein , termed CihC , and acquire C4b-binding protein ( C4bp ) and human C1 esterase inhibitor ( C1-Inh ) , the major inhibitors of the classical and lectin pathway of complement activation . A highly homologous receptor for C4bp was also found in the African tick-borne relapsing fever spirochete B . duttonii . Upon its binding to B . recurrentis or recombinant CihC , C4bp retains its functional potential , i . e . facilitating the factor I-mediated degradation of C4b . The additional finding that ectopic expression of CihC in serum sensitive B . burgdorferi significantly increased spirochetal resistance against human complement suggests this receptor to substantially contribute , together with other known strategies , to immune evasion of B . recurrentis .
B . recurrentis , the causative agent of louse-borne relapsing fever is transmitted to humans by contamination of abraded skin with either hemolymph from crushed , infected lice ( Pediculus humanus humanus ) or excreted feces thereof [1] , [2] . The last century has seen multiple epidemics of louse-borne relapsing fever in Europe , with high mortality rates of up to 40% . Louse-borne relapsing fever has been epidemic in Africa throughout the 20th century with foci persisting in the highlands of Ethiopia [3] , [4] . Clinically , louse-borne relapsing fever is characterized by a 5- to 7-day incubation period followed by one to five relapses of fever , and spirochetemia [5] , [6] . Spontaneous mortality remains as high as 2–4% despite antibiotics , with patients suffering from distinctive hemorrhagic syndrome and/or Jarish-Herxheimer reactions [7] . To survive in human tissues , including blood , B . recurrentis has to escape innate and adaptive immune responses . Complement is a major component of first line host defense with the potential to eliminate microbes . However , pathogens have evolved strategies to evade complement-mediated lysis , either indirectly , by binding host-derived regulators to their surface or directly , by expressing endogenous complement inhibitors [8] , [9] . In fact , we and others have recently demonstrated that tick- and louse-borne pathogens , i . e . B . hermsii and B . recurrentis , specifically bind complement regulatory proteins , i . e . CFH and CFHR-1 , via their outer surface lipoproteins FhbA , BhCRASP-1 and HcpA , respectively [10]–[14] . Surface bound CFH was shown to interfere with the alternative complement pathway by inhibiting complement activation via accelerating the decay of the C3 convertase and inactivating newly formed C3b [15] , [16] . However , complement may also attack pathogenic bacteria via the classical pathway , i . e . by interacting with previously bound antibodies , resulting in deposition of the membrane attack complex on the surface of bacteria and their final death [17] . The classical pathway is initiated by the binding and activation of the C1 complex , consisting of C1q , C1r and C1s . C1q can bind to clustered IgG and IgM bound to the surface of bacteria , and also directly to many bacteria through lipoteichoic acids or other structures [18] , [19] . When C1q binds , its associated proteases , C1r and C1s , become activated and form the activated C1 complex , which cleaves C4 and C2 to generate the C3 convertase . The lectin pathway is initiated when mannose-binding lectin ( MBL ) or ficolins bind carbohydrates on the surface of a microbe [20] . A key endogenous regulator of the classical and lectin pathway is serum-derived C4b-binding protein ( C4bp ) . C4bp is a cofactor in factor I-mediated cleavage of C4b to C4d and interferes with the assembly and decay of the C3-convertase ( C4bC2a ) of the classical and lectin pathway [21] , [22] . It was recently shown that acquisition of the regulators CFH and C4bp on the surface of B . recurrentis and B . duttonii contributes to serum resistance in vitro [17] . However , the respective receptors on the spirochetal surface have not been identified . It was thus the aim of the present study to identify and characterize the putative receptor for C4bp of B . recurrentis and B . duttonii . Here , we show for the first time that B . recurrentis and B . duttonii express a novel potential outer surface lipoprotein , which specifically binds C4bp and in addition C1-Inh . The finding that pathogen-bound C4bp retains its co-factor activity suggests that this process contributes to the exceptional resistance of the two spirochetes species to bactericidal activity of human serum .
Relapsing fever spirochetes B . recurrentis strains A1 and A17 , B . hermsii ( ATCC35209 ) strain HS1 , B . duttonii strain LA , B . parkeri RML , B . turicatae RML ( provided by Tom Schwan , Rocky Mountain Laboratories ) and the Lyme disease spirochete B . burgdorferi strains ZS7 and B313 , a clonal mutant of B31 lacking all linear and circular plasmids with the exception of cp32-1 , cp32-2 , cp32-4 , cp26 and lp17 [23] , [24] , were cultivated in BSK-H complete medium ( Bio&Sell , Feucht , Germany ) supplemented with 5% rabbit serum ( PAN Biotech , Freiburg , Germany ) at 30°C . Bacteria were harvested by centrifugation and washed with phosphate-buffered saline . The density of spirochetes was determined using dark-field microscopy and a Kova counting chamber ( Hycor Biomedical , Garden Grove , CA ) . E . coli JM109 were grown at 37°C in LB medium . All human plasma and serum samples used in this study were purchased from the Heidelberg University blood bank . Human plasma obtained from 20 healthy , anonymous blood donors without known history of spirochetal infections were pooled and used as source for C4bp . Nonimmune human serum ( NHS ) was acquired from healthy donors with no prior history of Borrelia spp . infection . Factor B-depleted human serum was purchased from Complement Technology , Inc . ( http://www . ComplementTech . com ) . C4bp protein was purified from pooled human plasma by barium citrate precipitation as described [25] . Briefly , following extensive dialysis the solution was subjected to ion exchange chromatography using Q-Sepharose ( GE Healthcare ) and proteins were eluted with a gradient of 0 – 2 M NaCl . C4b , C1-Inh and factor I were purchased from Calbiochem . Purified C4bp , C1-Inh and BSA were conjugated to biotin with No-Weigh Biotin-NHS ( Pierce Biotechnology ) . Isolation of the C4bp binding protein of B . recurrentis was carried out by co-immunoprecipitation . Whole cell lysates of B . recurrentis were prepared as described elsewhere with minor modifications [10] . Briefly , cultures were grown at 33°C in modified BSK medium to the late-log phase and harvested by centrifugation at 6 . 000×g for 10 min at 4°C . The resulting pellets were washed twice with PBS , resuspended in ice-cold 50 mM Tris-HCl ( pH 7 . 5 ) , 25 mM KCl , 5 mM Mg2Cl , 1mM EGTA , 0 . 5% NP40 and rotated for 1 h at 4°C . Cell debris were removed by centrifugation and for pre-clearing lysates were incubated with protein G sepharose ( GE Healthcare ) for 1 h at 4°C . For immunoprecipitation pre-cleared B . recurrentis lysates were incubated with protein G Sepharose previously loaded with anti-C4bp antibody and purified human C4bp for 12 h at 4°C with gentle agitation . After washing in 50 mM NaH2PO4 , 300 mM NaCl , 10 mM imidazole ( pH 8 ) bound proteins were eluted with 2x SDS sample buffer ( Serva ) and subjected to 14% Tris/Tricine SDS-PAGE under reducing conditions . Immunoprecipitates were separated by SDS-PAGE and visualized by staining with colloidal Coomassie ( Pierce/Thermofisher , Bonn , Germany ) . The selected protein band of 40 kDa was cored from the gel and subjected to MALDI mass spectrometric analysis as previously described [26] . Recently , the genome of the selected B . recurrentis strain A1 was sequenced [27] . The identified peptide matched an open reading frame of 1071 bp of the B . recurrentis A1 genome , named cihC . The gene encoding CihC was amplified by PCR using primers CihC F and CihC R ( Table 1 ) , cloned into pGEM-T Easy vector ( Promega , Mannheim , Germany ) and sequenced by using the BigDye terminator cycle sequencing kit ( PE Applied Biosystems ) . The resulting plasmid pGEM-BrCihC was used as template for construction of expression plasmids by PCR amplification . For recombinant full-length CihC protein , primers CihC Bam and CihC HincII were used . For N- and C-terminal deletion mutants , these primers were applied in combination with CihCΔ83F , CihCΔ122F , CihCΔ160F , CihCΔ149R , CihCΔ190R , CihCΔ260R , and CihCΔ294R ( Table 1 ) resulting in recombinant proteins CihCΔ20–260 , CihCΔ83–294 , CihCΔ122–294 , CihCΔ20–190 , CihCΔ83–149 , and CihCΔ160–294 , respectively . The ORF encoding CihC of B . duttonii ( CihCBD ) was amplified using genomic DNA of B . duttonii strain La in combination with oligonucleotides CihC Bam and CihC Hinc . After digestion with restriction enzymes BamHI and HincII , PCR fragments were ligated in frame into the His6-tag encoding sequence into vector pQE-30Xa ( Qiagen , Hilden , Germany ) . For expression of N-terminal His-tagged fusion proteins , the plasmids were transformed into E . coli strain JM109 and recombinant proteins were purified as recommended by the manufacturer ( Qiagen ) . Monoclonal antibody BR2 , directed against CihC was generated by immunization of Balb/c mice with whole cells of B . recurrentis A1 according to a method described elsewhere [28] . All animal research was approved in advance by the Laboratory Animal Committee of the University of Heidelberg ( RP Karlsruhe 35-9185 . 82/A-25/07 ) . The animals were kept in a filter cabinet and given food and water ad libitum , with all maintenance performed according to German animal welfare guidelines . To prepare whole cell lysates Borrelia were centrifuged and washed three times with PBS . Cells were resuspended in BugBuster Master Mix ( Merck ) and lysed for 5–10 min on ice . Borrelial whole cell lysates ( 15 µg ) or purified recombinant CihC proteins ( 200 ng ) were subjected to Tris/Glycine-SDS-PAGE under reducing conditions and transferred to nitrocellulose as previously described [29] . Briefly , after transfer of proteins onto nitrocellulose , nonspecific binding sites were blocked using 5% ( w/v ) dried milk in TBS ( 50 mM Tris-HCl pH 7 . 4 , 200 mM NaCl ) for 2 h at room temperature ( RT ) . Subsequently , membranes were rinsed two times in TBS and incubated for 1 h at RT with NHS ( 1:1 diluted in TBS ) or purified C4bp . Membranes were washed four times with 50 mM Tris-HCl pH 7 . 5 , 150 mM NaCl , 0 . 2% Tween20 ( TBST ) and incubated for 1 h with either peroxidase-conjugated anti-C1-Inh ( Linaris ) or anti-C4bp antibody ( Quidel , San Diego ) . Following four washes with TBST , blot strips were incubated with a secondary peroxidase-conjugated anti-mouse IgG antibody ( Dako , Glostrup , Denmark ) for 1 h at RT . Detection of bound antibodies was performed using the enhanced chemiluminescence ECL Western blotting detection reagent and ECL Hyperfilms ( GE Healthcare , Amersham ) . For Western blot analysis , membranes were incubated for 1 h at RT with either anti-C4c antiserum ( Dako ) , anti-C1s ( Atlantic antibody ) , anti-CihC ( mAb BR2 ) or anti-flagellin ( mAb LA21 ) monoclonal antibodies [30] . For detection of purified recombinant CihC full-length protein and deletion mutants , the anti-His6-tag ( Calbiochem ) antibody was employed . Southern blotting of total genomic DNA was done as previously described [31] . Briefly , 250 ml of Borrelia cultures were centrifuged , washed twice in PBS and resuspended in 9 ml of TE ( 10 mM Tris pH 7 . 5 , 1 mM EDTA ) buffer . Subsequently , 20% SDS ( 1 ml ) and 20 mg/ml proteinase K ( 50 µl ) was added and incubated for 1 h at 37°C . NaCl ( 5 M ) and Hexadecyl-trimethyl-ammonium-bromide ( 10% ) was added followed by incubation for 10 min at 65°C . DNA was extracted twice with phenol-chloroform-isoamyl ethanol ( 25∶24∶1 ) and DNA was precipitated with 0 . 6 volume of isopropanol . The precipitates were washed with 70% ethanol and resuspended in H2O . 10 µg of total genomic borrelial DNA was prepared as agarose blocks , loaded into the agarose gels and fractionated by pulse-field gel electrophoresis ( PFGE ) in combination with the CHEF-DR II System ( Bio-Rad , Germany ) . Hybridization with a random primed cihC gene probe was conducted as described [32] . Spirochetes ( 1×107 ) were washed with Tris buffer ( 30 mM Tris , 60 mM NaCl , pH 7 . 4 ) and incubated with mAb directed against CihC ( mAb BR2 ) or flagellin ( mAb LA21 ) for 1 h at RT . Spirochetes were then washed with Tris buffer/0 . 1% BSA , spotted on coverslips and allowed to air-dry for 1 h . After methanol fixation , samples were dried for 15 min and incubated for 1 h in a humidified chamber with Cy3-labeled rabbit anti-mouse IgG ( 1/200 , Dianova ) . Cells were visualized at a magnification of 1000x using a Nikon Eclipse 90i upright automated microscope and images were obtained using a Nikon DS-1 QM sensitive black and white CCD camera at a resolution of 0 . 133 µm/pixel . Cells of B . recurrentis strain A1 were treated with proteases using a modified , previously described method [33] . Briefly , intact borrelial cells were incubated with either proteinase K or trypsin to a final concentration of 0 -12 . 5 µg/ml . Borrelial cells were then lysed and equal volumes ( 20 µl ) were separated by SDS-PAGE ( 13% ) . Proteins were visualized by Western blotting using specific monoclonal antibodies . Functional activity of C4bp was analyzed by measuring factor I-mediated conversion of C4b to iC4b . Either 100 µl of CihC ( 0 . 5 µg/well ) or intact B . recurrentis A1 spirochetes were coated onto microtiter plates ( MaxiSorp , Nunc ) and incubated with purified human C4bp ( 50 µg/ml ) for 1h at RT and after washing , C4b ( 4 µg/ml ) and factor I ( 2 µg/ml ) were added and incubated at 37°C for up to 2 h . Supernatants were removed from the wells , subjected to SDS-PAGE ( 10% ) under reducing conditions and transferred to a nitrocellulose membrane . Degradation of C4b was evaluated by using a rabbit anti-C4c antibody ( DAKO ) followed by a peroxidase-conjugated goat anti-rabbit IgG . The protease inhibitory activity of C1-Inh bound to the borrelial surface was examined by detection of SDS-insoluble complexes of C1-Inh and C1s protease . To opsonize cells with C1 , 108 B . recurrentis cells were incubated with 10% NHS for 1 h at 30°C . After washing , cells were treated with 1 µg biotinylated C1-Inh for 1 h at 30°C . Following three washes , cells were lysed and the borrelial whole cell preparations were subjected to SDS-PAGE ( 7 . 5% ) under non-reducing conditions . Proteins were transferred to nitrocellulose membranes and probed with either peroxidase-conjugated streptavidin or goat anti-C1s ( Atlantic antibody ) followed by a peroxidase-conjugated rabbit anti-goat IgG . The CihC encoding cihC gene including its native promoter region was amplified by PCR using primers CihC Prom and CihC SphI . The resulting amplicon was cloned into pBSV2 yielding shuttle vector pCihC . Transformation of B . burgdorferi B313 and characterization of transformants was previously described [11] . Expression of CihC of transformed B . burgdorferi B313 was determined by Western blot , whole cell ELISA and immunofluorescence analysis , using mAb BR2 . High-passage , non-infectious B . burgdorferi strain B313 were grown in 100 ml BSK medium and harvested at mid exponential phase ( 108 cells/ml ) . Electrocompetent cells were prepared as described previously [26] with slight modifications . Briefly , 50 µl aliquots of competent B . burgdorferi strain B313 cells were electroporated at 12 . 5 kV/cm in 2-mm cuvettes with 10 µg of plasmid DNA . For control purpose B . burgdorferi strain B313 cells also were transformed with pBSV2 vector alone . Cells were immediately diluted into 10 ml BSK medium and incubated without antibiotic selection at 30°C for 48 to 72 h . Bacteria were then diluted into 100 ml BSK medium containing kanamycin ( 30 µg/ml ) and 200 µl aliquots were plated into 96-well cell culture plates ( Corning ) for selection of transformants . After three weeks , wells were evaluated for positive growth by color change of the medium , confirmed by dark-field microscopy for the presence of motile spirochetes . The cihC gene of transformed B . burgdorferi B313 was detected by PCR using oligonucleotides CihC F and CihC SphI . Ectopic CihC expression was analyzed using immunofluorescence microscopy and ELISA in combination with mAb BR2 . In addition , ectopically expressed CihC was analyzed by ligand affinity blotting and flow cytometry with regard to its capacity to acquire C4bp and C1-Inh . Briefly , 107 B . recurrentis A1 , B . duttonii La , B . burgdorferi B313/vc and B313/pCihC cells were washed twice with PBS , blocked for 15 min at RT with PBS/10% BSA , and incubated with 10 µg/ml of biotinylated C4bp or C1-Inh in FACS-buffer ( PBS/1% BSA ) for 1 h at RT . As a negative control , spirochetes were incubated with the same concentration of biotinylated BSA . Cells were washed three times , stained with phycoerythrin ( PE ) labeled streptavidin ( Bio-Rad ) and were then fixed with 1% paraformaldehyde overnight and analyzed using a FACS-Calibur and the CellQuest software ( BD Biosciences ) . The serum susceptibility of mock-transformed B . burgdorferi B313 ( B . burgdorferi B313/vc ) and transformed B . burgdorferi B313 ( B . burgdorferi B313/pCihC ) was assessed using a survival assay . Cells grown to mid-logarithmic phase were harvested , washed and approximately 3×107 spirochetes were resuspended in BSK-H medium supplemented with either 50% factor B-depleted human serum ( NHS-B ) or 50% heat inactivated factor B-depleted human serum ( hiNHS-B ) . Cells were incubated in Eppendorf tubes at 30°C for 2 days . At day 0 , 1 , and 2 , cells were washed in 0 . 85% NaCl , transferred to microtiter plates and incubated with SYTO9 ( Molecular Probes , Invitrogen ) as recommended by the manufacturer . Subsequently , relative growth of spirochetes as compared to day 0 was determined by measuring the fluorescence intensity at 530 nm ( excitation 485 nm ) on a microtiter plate reader ( Victor2 plate reader , Perkin Elmer ) . For whole cell ELISA , approximately 108 spirochetes ( B . burgdorferi B313/vc and B313/pCihC ) were washed twice , resuspended in PBS and immobilized on microtiter plates overnight at 4°C . The wells were washed with PBS/0 . 05%Tween , blocked with PBS/5% BSA and were then incubated with the CihC-specific mAb BR2 or the flagellin-specific mAb LA21 followed by a peroxidase-conjugated sheep anti mouse IgG . Substrate reaction was performed with o-phenyldiamine dihydrochloride ( Sigma-Aldrich ) and absorbance was measured at 492 nm . Statistics were analyzed with the unpaired Student's t-test , P values less than 0 . 05 were considered significant . The cihC gene sequences of B . recurrentis and B . duttonii reported in this paper have been deposited in the EMBL/GenBank data bases under the following accession numbers: FN552439 and FN552440 , respectively .
To verify acquisition of C4bp onto the outer surface B . recurrentis and B . duttonii spirochetes were incubated with biotinylated human C4bp and analyzed by flow cytometry . Both strains were found to acquire C4bp onto their surfaces ( Fig . 1A ) . By applying ligand affinity blot analysis for detection of C4bp-binding molecules , a protein of about 40 kDa was identified in B . recurrentis and B . duttonii , but not in B . hermsii and B . burgdorferi ( Fig . 1B ) . In addition , B . recurrentis and B . duttonii are capable of binding the complement regulator C1-Inh ( Fig . 1 ) . To isolate and characterize the receptor for C4bp , cell lysates of B . recurrentis A1 were incubated with C4bp and added to Protein G Sepharose coupled anti-C4bp immune serum . The co-precipitating protein of approximately 40 kDa was analyzed by mass spectrometry and the peptides generated matched an open reading frame of 1071 bp on the genome of B . recurrentis A1 [27] . The open reading frame encoded for a putative lipoprotein with a calculated molecular mass of 40 . 4 kDa . The encoding gene was designated cihC ( complement inhibition via C4bp ) . Pulse-field gel electrophoresis and hybridization analysis revealed that the cihC gene represents a single genetic locus in B . recurrentis and B . duttonii that maps to a larger plasmid of approximately 200 kb ( Fig . 2 ) [27] , [34] . Neither the tick-borne relapsing fever strains of B . parkeri , B . hermsii and B . turicatae nor B . burgdorferi , the causal agent of Lyme disease , hybridized with the cihC probe ( data not shown ) . Isolation of the homologous B . duttonii gene revealed 91% amino acid sequence similarity with that of B . recurrentis ( Fig . 3 ) . Lescot et al . previously identified the cihC gene of B . duttonii as a p35-like antigen ( BDU_1 ) exhibiting similarity to the B . burgdorferi fibronectin-binding lipoprotein BBK32 . In contrast to our observation the homologous gene in B . recurrentis was not detected [27] . Interestingly , our preliminary studies indicated that recombinant CihC of B . duttonii and B . recurrentis binds fibronectin ( unpublished ) . A BLAST search failed to detect any other protein with significant homology , indicating that the two genes/proteins are restricted to these highly related species of Borrelia . To determine whether CihC is surface exposed , immunofluorescence assays were performed using mAb BR2 specific for CihC . B . recurrentis spirochetes were incubated sequentially with mAb BR2 and rabbit anti-mouse Cy3-conjugated antibody ( Fig . 4A ) . Epifluorescence microscopy revealed that B . recurrentis organisms expressed CihC on their outer surface in a patch-like manner . Controls incubated with the secondary antibody alone were negative ( not shown ) . To further confirm surface localization of CihC , B . recurrentis organisms were treated with either proteinase K or trypsin and subjected to Western blot analysis . As shown in Figure 4B , a significant reduction was observed for CihC after 2 h of incubation with proteinase K at concentrations ≥3 µg/ml . Upon treatment of the spirochetes with trypsin , a more restricted protease , only higher amounts ( ≥6 µg/ml ) yielded complete degradation of CihC . The mouse mAb LA21 directed against the periplasmic FlaB protein was used in this experiment as a internal control to confirm that the fragile spirochetal outer membrane was not damaged ( Fig . 4B , lower panels ) . These data indicate that CihC is exposed at the outer surface of B . recurrentis . To localize the putative domain ( s ) of CihC that bind to C4bp and C1-Inh , a number of CihC deletion mutants with distinct N- or C-terminal truncations were constructed ( Fig . 5A ) . Protein expression was confirmed by using a His-tag specific antibody and all recombinant proteins exhibited the predicted size . Screening for C4bp and C1-Inh binding by ligand affinity blotting revealed that from the polypeptide preparations tested , full-length CihC ( residues 20 to 356 ) and all truncated versions employing the central protein domain ( amino acid residues 145 – 185 ) similarly retained C4bp and C1-Inh binding activity ( Fig . 5B ) . These results suggest that CihC contained a central region that bound to both human complement regulators . Inactivation of complement component C4b occurs by factor I mediated cleavage of the C4b alpha chain . To assess whether C4bp maintains this cofactor activity when attached to the surface of intact B . recurrentis spirochetes were coated with purified human C4bp and incubated with C4b and factor I . The supernatant was subjected to SDS-PAGE and C4b alpha chain degradation products were detected by immunoblot analysis . As shown in Figure 6 ( left panel ) , binding of C4bp to the cell surface resulted in α4 and α3 degradation products of 15 kDa and 25 kDa , respectively . In contrast , B . recurrentis spirochetes alone did not promote factor I-mediated cleavage of C4b demonstrating that louse-borne relapsing fever spirochetes lack endogenous C4b cleaving activity . Similarly , C4bp bound to immobilized recombinant CihC protein efficiently mediated C4b processing via factor I , as indicated by the appearance of a α4 fragment ( Fig . 6 , right panel ) . B . recurrentis or CihC preincubated with C4bp and C4b in the absence of factor I did not promote cleavage of C4b ( data not shown ) . These findings demonstrate that CihC-associated C4bp retains its cofactor activity and may lead to accelerated disintegration of the C3 convertase ( C4bC2a ) of the classical complement activation pathway . The protease inhibitor C1-Inh reacts with its complement target proteases such as C1s and C1r to form high molecular weight SDS resistant complexes [35] . We examined the formation of these covalent C1-Inh-protease complexes as an index for the protease inhibitory activity of CihC-associated C1-Inh . Intact B . recurrentis cells were preincubated in NHS as source for C1 and biotinylated C1-Inh was applied . Subsequently , cells were washed extensively to remove unbound C1-Inh , lysed and subjected to immunoblotting . As shown in Figure 7A , biotinylated C1-Inh acquired by B . recurrentis formed complexes on the spirochetal surface as indicated by the occurrence of a high molecular weight band at >170 kDa . To identify the constituent protease of these complexes , immunoblot analysis using C1s-specific antiserum was employed revealing C1s is a component of the >170 kDa large complex ( Fig . 7B ) . Exogenously applied biotinylated C1-Inh as well as serum-derived C1-Inh formed the respective complexes with C1s . These data suggest that C1-Inh bound to the surface of B . recurrentis retains its functional activity and thus , by inactivating C1s protease , exhibits complement inhibitory activity . To test whether CihC of B . recurrentis plays an important role in mediating complement resistance , the serum-sensitive B . burgdorferi B313 mutant strain was transformed with the shuttle vector pCihC containing the complete cihC gene ( B . burgdorferi B313/pCihC ) ; for control , the pBSV2 vector alone ( B . burgdorferi B313/vc ) was employed . Expression and surface localization of CihC was determined by whole cell ELISA ( Fig . 8A ) and immunofluorescence ( Fig . 8B ) analyses using the CihC-specific mAb BR2 . Moreover , to ascertain whether the ectopically expressed CihC is capable of recruiting C4bp and C1-Inh to the surface of B313/pCihC flow cytometry was performed ( Fig . 8C ) . The B313/pCihC-transformed isolate but not the mock-transformed B313/vc isolate of B . burgdorferi strongly expressed CihC and acquired both complement regulators . To compare the susceptibility of B313/pCihC and B313/vc to complement-mediated killing , both specimens were subjected to a human serum sensitivity assay . In order to avoid killing of Borrelia strains via the alternative pathway of complement activation a factor B-depleted human serum was employed . Accordingly , spirochetes were incubated in factor B-depleted human serum ( NHS-B ) or heat-inactivated factor B-depleted serum ( hiNHS-B ) and spirochetal growth was monitored by uptake of a nucleic acid dye . B313/pCihC and the mock-transformed strain multiplied during the 48 h time interval when incubated with heat-inactivated factor B-depleted serum ( Fig . 8D ) . However , when exposed to NHS-B only B313/pCihC spirochetes could replicate indicating that ectopic expression of CihC renders serum-sensitive B . burgdorferi B313 more resistant to complement-mediated lysis . These data suggest a decisive role for CihC in serum resistance of B . recurrentis and B duttonii .
Bacteria have evolved multiple strategies to interfere with complement-mediated clearance of pathogens by blocking distinct steps of the lytic cascade . Recently , we provided evidence that the louse-borne relapsing fever spirochete B . recurrentis selectively inhibits activation of the alternative complement pathway by specifically binding the endogenous complement inhibitor CFH via its lipoprotein HcpA [11] . We now demonstrate for the first time that B . recurrentis also expresses a surface receptor specific for C4bp and C1-Inh , two major serum-derived inhibitors of the lectin and classical complement pathways , termed CihC . Genetic and molecular analyses revealed that CihC of B . recurrentis is a potential lipoprotein and that B . duttonii harbors a homologue of CihC [27] . Upon binding to the pathogen's surface or to recombinant CihC , C4bp retained its cofactor activity for factor I-mediated C4b inactivation . Together with the fact that B . recurrentis also expresses HcpA , the presented data suggest that the potential of louse-borne relapsing fever spirochetes to interfere with both , classical and alternative pathways , contributes to their high resistance and pathogenicity in humans . The correlation between serum resistance of bacteria and cell surface binding of functionally active C4bp has been reported before for a number of pathogenic microorganisms , including the spirochetes B . recurrentis , B . duttonii and B . burgdorferi s . s . ( strain IA ) , the causative agent of Lyme disease [17] . Moreover , when incubated with human serum , Yersinia enterocolitica , Bordetella pertussis , Neisseria gonorrhoeae , Candida albicans , Moraxella catarrhalis , Escherichia coli K1 , Streptococcus pyogenes and Yersinia pestis were also shown to acquire C4bp [36]–[43] . However , the respective receptors for C4bp have only been identified for some bacteria , e . g . Streptococcus pyogenes , Yersinia enterocolitica , and Moraxella catarrhalis , but not for B . recurrentis , B . duttonii and B . burgdorferi . The present data provide evidence that the receptor for C4bp of B . recurrentis , CihC is a surface exposed putative lipoprotein . Preliminary Southern Blot analysis and BLASTN search on databases revealed a putative homologue of cihC only in B . duttonii but not in other spirochetal species suggesting that the gene encoding C4bp receptor is unique to these two Borrelia species . To determine whether the cihC gene is located either on the chromosome or any of the linear plasmids , PFGE and Southern blotting were performed . The cihC gene was localized to a 190 kb linear plasmid adjacent to the previously identified factor H binding hcpA gene . Similarly , the B . hermsii gene encoding the factor H binding protein FhbA maps to the large linear plasmid of 220 kb [13] . However , further studies are required to resolve this issue for other bacterial pathogens . To localize the peptide domains of CihC relevant for binding of C4bp and C1-Inh , truncated N- and C-terminal deletion mutants were generated and used for functional analyses . C4bp and C1-Inh binding was not abrogated by N-terminal ( amino acid residues 20–121 ) or the C-terminal ( amino acid residues 191–356 ) deletion mutants of CihC indicating that both , C4bp and C1-Inh , bind to the central domain of CihC . In related studies , Streptococcus pyogenes was previously shown to bind C4bp through the N-terminal highly variable region of M-proteins Arp and Sir and similar results were also obtained with the FHA receptor for C4bp of Bordetella pertussis [38] , [39] , [41] , [44] . However , the reason for the differential binding domains of the various pathogen receptors for C4bp is not known at present . Preliminary data indicate that binding of C4bp to CihC ectopically expressed by B . burgdorferi B313 cells is independent of ionic strength suggesting a hydrophobic interaction between the receptor and its ligand . Similar findings have been reported before for other pathogens . Thus , interaction of the Y . enterocolitica Ail receptor with C4bp was also found to be less sensitive to salt [45] . Moreover , C4bp receptors like Por1A of N . gonorrhoeae [42] , [46] , UspA1/2 of M . catarrhalis [36] , OmpA of E . coli [43] and the M-proteins of S . pyogenes bind C4bp in a nonionic fashion . However , further analyses including C4bp deletion constructs are required to solve this issue for CihC . The present study adds another facet on the versatility of relapsing fever spirochetes to persist in human blood and to evade innate and adaptive immunity . The best-known immune evasion strategy of relapsing fever Borrelia is antigenic variation , i . e . the ability to respond to newly generated specific antibodies with a switch to an altered variable major outer surface protein ( Vmp ) . Essentially , the pathogen always stays one step ahead of antibodies . However , while antigenic variation is restricted to Vmps , other surface exposed proteins are stable and antigenic , e . g . the surface-exposed lipoprotein FhbA of B . hermsii [12] , [13] , [47] , [48] . In this context it could be speculated that upon binding to CihC , C4bp and C1-Inh inhibit the lectin and classical complement pathway , including the formation of the lytic membrane attack complexes . In addition to the observed anti-complement activity , B . recurrentis-exposed C4bp may exhibit another biological activity relevant for spirochetal serum resistance . This is indicated by the fact that C4bp circulates in plasma as a complex with protein S that in turn binds to negatively charged phospholipids on membranes [49]–[51] . Thus , it is possible that C4bp also promotes adhesion and subsequently hematogenous dissemination by simultaneously binding to B . recurrentis and endothelial cells . This assumption is supported by the recent observation that the related fibronectin and glycosaminoglycan binding protein , BBK32 , of B . burgdorferi mediates endothelial interactions in vivo , thereby facilitating microvascular interactions [52] . Similarly , CihC of B . recurrentis and B . duttonii bound fibonectin and thus could also be involved in the dissemination process of relapsing fever spirochetes . The assumption that CihC of B . recurrentis and probably also CihC/BDU_1 of B . duttonii are critically involved in their escape from complement-mediated lysis is further supported by the present finding that ectopic expression of CihC in the serum-sensitive B . burgdorferi strain B313 led to a significant increase in resistance to complement mediated lysis . Moreover , binding of C1-Inh , the major inhibitor of several pathways of inflammation in humans , to CihC could be observed . However , the actual role of CihC in the pathogenesis of louse-borne relapsing fever will only be elucidated by in vivo studies in a relevant mouse model [53] . Complement resistance in cihC transformed B . burgdorferi strain was detected in the presence of non-immune factor B-depleted human serum indicating that the lectin/classical pathway of complement activation may be triggered by Borrelia structures other than specific antibodies . Indeed , we have shown that C1q and the C1 complex can bind to the surface of B . recurrentis in the absence of specific antibodies . Moreover , recognition molecules specific for the lectin pathway ( i . e . MBLs and ficolins ) could also bind to borrelial carbohydrates and activate MASPs [20] , [54]–[58] . MASP-2 is the enzyme component that , like C1s in the classical pathway , cleaves the complement components C4 and C2 to form the C3 convertase C4bC2a , common for activation of both the lectin and the classical pathways . However , it remains to be determined whether C4bp and C1-Inh binding significantly increases B . recurrentis spirochetes resistance against complement attack in humans . In summary , this study is the first to show that B . recurrentis and most probably B . duttonii express a potential lipoprotein receptor , which selectively binds C4bp and C1-Inh , the endogenous regulators of the classical and lectin complement pathway . Together with the fact , that both spirochetal species also carry a specific receptor for the serum-derived complement inhibitor of the alternative pathway , CFH , the present data emphasize the versatility of B . recurrentis and B . duttonii to evade lectin/classical and alternative pathways of complement activation . Elucidating the pathological processes underlying relapsing fever will be helpful to design novel regimens for therapeutic treatment of spirochete-induced relapsing fever and to develop potential vaccine candidates .
|
Borrelia recurrentis , the causal agent of louse-borne relapsing fever is transmitted to humans via infected body lice . Infection with B . recurrentis has been achieved only in humans and is accompanied by a systemic inflammatory disease , multiple relapses of fever and massive spirochetemia . A key virulence factor of B . recurrentis is their potential to undergo antigenic variation . However , for survival in the blood during the early phase of infection and for persistence in human tissues , spirochetes must be endowed with robust tools to escape innate immunity . We have recently shown that B . recurrentis acquires the serum-derived regulator factor H , thereby blocking the alternative complement pathway . Here , we show that B . recurrentis expresses in addition a novel outer surface lipoprotein that selectively binds serum-derived C4b-binding protein and C1 esterase inhibitor , two endogenous regulators of the classical and lectin pathway of complement activation . The combined data underscore the versatility of B . recurrentis to effectively evade innate and adaptive immunity , including serum resistance . Thus , the present study elucidates a new mechanism of B . recurrentis important for its evasion from complement attack and will be helpful for the development of new drugs against this fatal infection .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/bacterial",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"immunology/immunity",
"to",
"infections",
"immunology/innate",
"immunity"
] |
2010
|
Human Complement Regulators C4b-Binding Protein and C1 Esterase Inhibitor Interact with a Novel Outer Surface Protein of Borrelia recurrentis
|
For making decisions in everyday life we often have first to infer the set of environmental features that are relevant for the current task . Here we investigated the computational mechanisms underlying the evolution of beliefs about the relevance of environmental features in a dynamical and noisy environment . For this purpose we designed a probabilistic Wisconsin card sorting task ( WCST ) with belief solicitation , in which subjects were presented with stimuli composed of multiple visual features . At each moment in time a particular feature was relevant for obtaining reward , and participants had to infer which feature was relevant and report their beliefs accordingly . To test the hypothesis that attentional focus modulates the belief update process , we derived and fitted several probabilistic and non-probabilistic behavioral models , which either incorporate a dynamical model of attentional focus , in the form of a hierarchical winner-take-all neuronal network , or a diffusive model , without attention-like features . We used Bayesian model selection to identify the most likely generative model of subjects’ behavior and found that attention-like features in the behavioral model are essential for explaining subjects’ responses . Furthermore , we demonstrate a method for integrating both connectionist and Bayesian models of decision making within a single framework that allowed us to infer hidden belief processes of human subjects .
A typical problem that humans encounter , in our complex environment , is to identify those environmental features that are relevant for achieving a desired outcome in a given task . This is computationally difficult because the real-world environment displays a large number of environmental features . In addition , the relevance of the features can change over time and the observations do not always reflect the relevance of specific features . For example , to increase the chance of catching a fish , a fisherman has to consider various features ( e . g . time of the day , lightening conditions , water transparency , etc . ) . Depending on the fishing place ( e . g . pond , lake , or river ) only some of these features will be relevant . To perfectly solve such tasks all possible features should be taken into account simultaneously . However , due to an apparent limitation in their cognitive resources , humans dynamically attend only to the most relevant environmental features when deciding what action to pursue [1 , 2] . Our goal here is to develop a computational model to analyze behavioral data and understand better how attention modulates the update of beliefs about the relevance of features in such complex environments . An ideal test bed to address these questions is the Wisconsin card sorting task ( WCST ) , as it provides an experimental environment with multiple visual features , in which at any moment of time only a single feature is relevant for correctly solving the task . The WCST was originally designed to test for the damage or dysfunction of the prefrontal cortex , which regulates executive functions [3–6] . More recently it was employed in various behavioral models as a paradigm with which one can investigate computational mechanisms of higher cognitive functions [7] . Here we will focus on the computational mechanisms that underlie update of beliefs about the relevance of various visual features . However , inferring the hidden belief states of subjects performing the standard WCST is difficult , as the only expression of an internal , multidimensional belief space are the behavioral choices [1 , 8–10] . To address this issue we designed a probabilistic variant of WCST in which we solicited subjects’ beliefs [11] , that is , we requested from subjects to bet an amount of money proportionally to their beliefs about the relevance of each visual feature . Importantly , various sources of uncertainty made the environment of WCST probabilistic and made the task more difficult , thus allowing us to measure smooth belief trajectories that evolve over single trials . This fine-grained measure provides more direct access to subjects’ hidden belief states and thus allowed for improved inference , compared to the standard WCST . Using this novel variant of the WCST , we were able to develop a probabilistic model for the analysis of behavioral data to provide novel insights into the hidden learning mechanism , which drives human behavior [12–14] . Previous computational models for the WCST can be divided into three groups based on the assumed computational principle that were used to capture human behavior and cognition: ( i ) functional cognitive models [10] , which are motivated by algorithmic properties of the task; ( ii ) connectionist models [9 , 15–19] , which are motivated by the evidence that the brain is an active and distributed system that constantly generates hypotheses about its environment and tests for their validity [20–25]; and probabilistic Bayesian models [1] , which further assume that the brain combines prior knowledge and present sensory information based on their relative precision , that is , in a Bayes-optimal manner [26–32] . The classical connectionist approach provides an elegant framework for defining attention formation in a distributed and dynamical manner . A potential limitation is that one requires additional and rather ad-hoc assumptions to describe the interaction of prediction errors with internal dynamics of beliefs . This issue can be addressed by the Bayesian approach which provides a framework for defining optimal interaction between prediction errors and current belief states . Furthermore , the Bayesian framework provides a computational account of attention [33–36] , which the connectionist approach lacks . Here we build upon these past views of attention within the Bayesian framework , with an attentional focus mechanism that relies on competitive and self-organized dynamical principles that guide spontaneous formation of attention . We will fuse the winner-take-all ( WTA ) dynamics [37–43] with a Bayesian formalism of decision making . With this combined approach we can investigate , at the same time , the influence of attention and the influence of probabilistic aspects of the environment on the evolution of beliefs during decision making . In addition , this framework allows us to relate our investigation to previous findings of a presumed hierarchical representation in the brain [12 , 14 , 44–48] . Importantly , the introduction of such an attentional focus mechanism within a Bayesian framework takes the model away from the rational Bayesian observer that is fully informed about the structure of the probabilistic WCST and which updates beliefs about all features independent of their relevance . However , we expect an attentional focus mechanism to provide a better account for experimentally observed human behavior . To test whether subjects’ behavior reflects the assumption that the update of beliefs is modulated by attentional focus we compared multiple variants of the behavioral models , both with and without an attentional focus mechanism , in their ability to generate behavioral data . In particular , we used a recently described meta-Bayesian approach , the so-called ‘Observing the observer’ ( OTO ) framework to infer the hidden belief states and their influence on behavioral responses of human subjects [49 , 50] . Importantly , using the OTO framework enabled us to put perception and action ( i . e . , subjects’ responses ) into a single behavioral model and to compare various variants of both perceptual and response models . Each variant of the perceptual model tested for different assumptions about the mechanisms that underlie the update of beliefs . Similarly different variants of the response model tested for evidence regarding sub-optimality in human decision making , caused by a potentially stochastic representation of posterior beliefs in the brain [51–53] . In what follows , we will first describe the experimental paradigm , briefly introduce the OTO framework , and derive the update equations of several variants of the behavioral models . Then we will describe the data analysis technique that relies on Bayesian model selection using a random effects metric [54 , 55] , and present the results of the analysis that we performed on a behavioral , multi-subject , data set obtained from a probabilistic WCST paradigm . In the last section of the article we discuss the relevance of the proposed attentional-focus mechanism and its relation to past works .
The experiment was approved by the Caltech Institutional Review Board and all subjects gave informed consent before participating in the study . We designed the experimental task with the aim to access the hidden belief states of the subjects . For this purpose we instructed the subjects to infer , by observing a series of an experimenter's choices , which one of the three different visual features is relevant for the current choice , and to report their beliefs about the relevance of each of the features . Participants in the experiment were all healthy volunteers recruited from the Caltech student population . The visual stimuli that we presented to subjects consisted of a pair of cards ( top and bottom ) , where each card contained three visual features ( color , motion , shape ) . In turn , each visual feature was represented by one of the two possible exemplars ( red-green , left-right , circle-square ) . As each card had to contain a distinct exemplar , there were eight distinct configurations of card pairs . Thus at each experimental trial the visual stimulus was randomly selected from one of the eight configurations ( e . g . , a red right-moving circle and green left-moving square; see Fig 1A ) . Each out of n = 22 pre-trained subjects ( 14 male and 8 female ) was exposed to an experimental session divided into six blocks consisting of T = 40 trials each . In three randomly selected blocks the relevant feature remained fixed ( no-switch condition ) , whereas in the other three blocks the relevant feature would change with a probability p = 0 . 35 ( switch condition ) . After each switch the relevant feature would remain constant for 8 trials before another switch could occur . Importantly , to make the otherwise quite simple task more difficult for healthy subjects we introduced observation uncertainty: the experimenter would select a wrong card ( a card not containing the relevant exemplar ) with probability ε = 0 . 2 in the no-switch condition , and with probability ε = 0 . 3 in the switch condition . The error rate ε was set to values that induced the most distinct behavioral responses between two experimental conditions , while rendering the switch condition informative enough to induce betting responses in subjects . At the beginning of each experimental block we informed the subjects about the block type , but we did not inform them about the exact values of the error rates ε or switch probabilities; they had to infer these probabilities during the training phase . Each subject went through three training sessions , where each subsequent session slightly increased the difficulty of the task in the following manner: In the first session subjects were exposed to a no switch environment with error rate of experimenters choices set to zero . In the second session the switches in the selection rule where announced with error rate still being set at zero . The third session consisted of the no-switch environment with ε = 0 . 2 . Afterwards , we explained to subjects the condition in the final switch environment with non-zero error rate . During a single trial subjects were first exposed to one of the eight possible visual stimuli ( see Fig 1A ) . After one second the presentation program would select a card containing the relevant exemplar with probability 1 − ε ( see Fig 1B ) . After observing the selected choice for 5 seconds subjects had a 4 second period to respond by distributing 20$ on the three visual features depending on their belief about the relevance of each feature for the selection process . The response was generated by moving a cursor within a triangle presented on the screen ( see Fig 1C ) . The closer the cursor was to one of the corners of the triangle the more money was assigned to the corresponding visual feature . Importantly , subjects were told that at the end of the experiment a single trial will be randomly selected and that subjects will gain the amount of money that they assigned to the relevant feature in that trial . This ensured that participants were motivated to provide an accurate rendering of their beliefs over the features . For clarification of the task we present at this point some of the key behavioral results ( see Fig 2 ) . We quantified the performance of subjects as the median amount of their money bets on a truly relevant visual feature over an experimental block . The maximal performance would correspond to betting the full amount of 20$ to the truly relevant feature at each trial . As expected , the median of subjects’ performance was higher during the no-switch condition ( Kruskal-Wallis test , p <10−14 ) , whereas the median reaction times were lower ( Kruskal-Wallis test , p <10−12 ) during the same experimental condition which reflects the increased difficulty of the switch condition . Our goal is to infer , from the behavioral data , the hidden belief states of each subject that are conditioned on the past sequence of visual stimuli and experimenter choices . By deriving an adequate mapping of observations onto internal belief states ( the perceptual model ) and the mapping of the internal belief states onto desired responses ( response model ) , we can define a generative model of the whole observation-response process [49 , 50] as ( see Fig 3 for a graphical representation ) : p ( r→t , γ , θ|e→t , m ( p ) , m ( r ) ) =p ( r→t|bt ( bt−1 , e→t , γ ) , θ , m ( p ) , m ( r ) ) p ( γ , θ|m ( r ) , m ( p ) ) , ( 1 ) where p ( r→t|bt ( bt−1 , e→t , γ ) , θ , m ( p ) , m ( r ) ) denotes the probability of observing a response r→t given the hidden belief states bt ( that depend on past beliefs , current sensory observations e→t , and a set γ of free parameters of the perceptual model m ( p ) ) and a set θ of free parameters of the response model m ( r ) . The last term p ( γ , θ|m ( r ) , m ( p ) ) in Eq ( 1 ) denotes a prior distribution over the space of free parameters . Thus , to infer the hidden belief states of a subject we have to invert the generative model ( Eq ( 1 ) ) for the given set of behavioral responses r1…t and sensory stimuli e1…t , and compute the posterior distribution over the model parameters p ( γ , θ|e1…t , r1…t ) = p ( γ , θ ) ∏k=1tp ( r→k|γ , θ , e1…k ) p ( r1…t|e1…t ) , ( 2 ) where we omitted m ( r ) , m ( p ) for better readability . Knowing the posterior distribution one can either compute the most likely belief state at trial t as bt ( γ^ ) — where γ^ denotes the mode of the posterior—or an expected belief state at trial t , as b¯t=Ep ( γ|e1…t , r1…t ) [bt ( γ ) ] . To test the hypothesis that subjects focus their attention on a subset of environmental features when updating their beliefs about the features' relevance , it is essential to compare multiple models in their ability to replicate the behavioral data and select the most appropriate model . Bayesian model comparison uses model evidence , that is , marginal likelihood p ( r1…t|e1…t ) , to estimate the probability that a specific model has generated the data . The advantage of such a procedure , compared to standard goodness of fit approaches , is that more complex models are penalized automatically . The model evidence , for any pair of perceptual and response models , is given as p ( r1…t|e1…t , m ( p ) , m ( r ) ) = ∫dγdθp ( γ , θ ) ∏k=1tp ( rk , |γ , θ , e1…k , m ( p ) , m ( r ) ) . ( 3 ) To estimate the model evidence and obtain the posterior distribution over model parameters p ( γ , θ|e1…t , r1…t ) any approximate inference scheme can be applied . In particular , Daunizeau et . al . [49 , 50] proposed the use of a variational scheme where the model log-evidence is approximated with the variational free-energy and the posterior distribution over the model parameters is selected as the maximizer of the free-energy obtained through variational calculus . However , this method requires the computation of the gradients of the log-joint probability distributions ( natural logarithm of the joint probability distribution given in Eq ( 1 ) ) , which in our case are not obtainable analytically as the derivatives affect the parameters of the non-linear equations of the belief process . Furthermore , a small change in the parameters of the update equations of beliefs ( Eq ( 11 ) , see below ) can have a large influence on the shape of the trajectory , thus the log-joint probability distribution can be ill-conditioned with respect to model parameters . Therefore , even if the gradient , with respect to model parameters , would be computable at every point of the trajectory , a gradient ascent method would have difficulties to converge to a global mode of the joint probability distribution , as the underlying landscape might have a multimodal , non-linear , and non-convex structure . Thus , we use a numerical gradient-free scheme to find the mode of the log-joint probability distribution and apply a numerical method to compute the Hessian matrix at that mode [56–58] . With the obtained values of the mode and the Hessian we compute the Laplace approximation to the model evidence [59] . We will discuss the specifics of the numerical estimates in the final subsection of the methods . In what follows we will first introduce the behavioral models . For the model comparison , we have paired all the full variants of the Bayesian perceptual models with the two variants of the response model; the reduced variants of the Bayesian models and all the variants of the non-Bayesian perceptual models were paired only with the reduced response model , as the posterior uncertainty about the visual features Σt ( f ) is set to constant values in this cases . In addition , we have defined a simple baseline model . Hence in total we consider 17 behavioral models denoted as: To summarize the motivation for these different variants of the perceptual model ( see Methods above for details ) : the structure-free model variants test for the possibility that the structured representation is not required for describing the behavioral data; the model variants without the final level of the hierarchy ( rw , rd ) test for the possibility that the final level of hierarchy is redundant for describing the behavior; the non-Bayesian variants of the perceptual test for the possibility that the Bayesian observer assumption is not required for describing the behavior . Each model variant is defined using a set of free parameters {γ , θ} for the perceptual and response models . To be able to define prior and posterior distributions in the same functional form of multivariate normal distributions , we transform all parameters so that they have the same domain of real numbers . Note that such a transformation does not change the value of model evidences , as to compute the model evidence one integrates over all the free parameters of a generative model . Let us denote by χ→ the vector of perceptual and response parameters transformed to real space , then χ→= ( ϑ ( γ ) , ϑ ( θ ) ) , where ϑ ( z ) ={ln ( z ) , if z ∈{α , κe , f , qe , f , wdist , σe , f0 , θ1 , θ2 , θ3}ln ( 2z1−2z ) , if z=ε ln ( 2z−12 ( z−1 ) ) , if z=τe , f z , if z∈{μ→e0 , μ→f0} . Thus , we can define the prior distribution over model parameters as a multivariate normal distribution N ( χ→;η→0 , soI ) . The log-joint probability distribution can then be written as l ( χ→ ) =∑k=1Tlnp ( r→k|bk ( e→k , ϑ−1 ( χ→γ ) ) , ϑ−1 ( χ→θ ) ) +lnN ( χ→;η→0 , soI ) , ( 17 ) where T denotes the number of trials within a single experimental block . The Laplace approximation to the log-evidence is obtained as lnp ( r1…t|e→1…t ) =l ( β→ ) +12ln|2πS| , ( 18 ) where β→ denotes the mode of l ( χ→ ) and S=−∂χ→ , χ→l ( χ→ ) −1|χ→=β→ , i . e . S is the negative inverse of the Hessian matrix at the mode β→ . To find the mode of l ( χ→ ) we applied the so-called Covariance Matrix Adaptation Evolution Strategy ( CMA-ES ) . CMA-ES is a numerical optimization method , which has been applied successfully in various research areas [74–77] and is particularly useful for ill-conditioned and multimodal objective functions . In short , CMA-ES is a stochastic derivative-free method for numerical optimization of non-linear optimization problems [56 , 57] . We used a freely available Matlab toolbox that implements the algorithm [Hansen , Nikolaus ( 2004 ) . ( https://www . lri . fr/~hansen/cmaes_inmatlab . html#matlab ) , Version 3 . 61] . Once the mode of the log-joint probability distribution ( Eq ( 17 ) ) is found , we have to estimate the curvature at the mode , that is , the Hessian matrix . We estimated the Hessian matrix by numerical differentiation [58] , where we used the following toolbox [D’Errico , John ( 2006 ) . ( http://www . mathworks . de/matlabcentral/fileexchange/13490 ) , MATLAB Central File Exchange . Retrieved 10 . November 2013] . Because of the stochastic nature of the CMA-ES algorithm we repeated the stochastic search N = 50 times per experimental block for each model . For each of the N solutions we estimated the Hessian matrix and computed the Laplace approximation to the log-evidence . Finally , we kept the solution with the largest log-evidence , therefore increasing the probability of finding the maximal lower bound to the log-evidence and thus the most likely model of a subject’s behavior . The numerically obtained β→ and S are used as the mean and the covariance matrix of the approximate posterior distribution N ( χ→;β→ , S ) . Note that in this way we obtain the full covariance matrix without the need for a mean field approximation , which would neglect any existing correlations between parameters . All data processing was performed using MATLAB [version 8 . 1 , The MathWorks Inc . , Natick , Massachusetts] . We first estimated the log model evidence of the 17 generative models described above for each experimental block . To obtain a total per-subject log-evidence for each experimental condition , we summed the estimated log-evidences over experimental blocks of a single experimental condition . This gives us the log model evidence of each generative model for each subject per experimental condition . We used the obtained log-evidences to apply the hierarchical Bayesian model selection approach described in [54 , 55] . By using hierarchical Bayesian model selection we assumed that the identity of the best-fitting model may vary across subjects . This requires treating the posterior model probability ( the posterior belief that a given model has generated the data ) as a random variable . Thus , the two computed quantities of interest are the expected probability ( EP ) and the exceedance probability ( XP ) of each model: The EP is defined as the probability that a given model generated the behavioral data of a randomly selected subject ( see [55] for a detailed mathematical description ) ; The exceedance probability XP tells how likely it is that a given model will have the largest probability in a random sample from the posterior distribution . Importantly , the XP can be seen as a degree of confidence in the difference between posterior model probabilities [55] . Thus , when presenting the results of a model comparison we will only report the XP of the corresponding model or model family , as large XP at the same time implies significantly larger EP . Importantly , we will only consider recently proposed “protected” exceedance probability , which takes into account the null hypothesis that assumes that all the models are equally likely ( see [55] for details ) . We will consider that the EP of a single generative model is significantly larger than the EP of other generative models , if the model’s XP is above threshold value set at 0 . 95 . Although , this threshold value was selected in the analogy to classical statistical tests that rely on p-values , its relation to the statistical power is not equivalent ( see [55] ) . We used the MATLAB implementation of the random-effect Bayesian model selection [ ( https://sites . google . com/site/jeandaunizeauswebsite/code/rfx-bms ) , retrieved January 2014] . In what follows we will describe the results obtained by applying the Bayesian model selection to the set of behavioral models that we used to approximate subjects’ behavior in the probabilistic WCST .
In Figs 6 and 7 we present the results of the random-effects Bayesian model comparison at the group-level . We have separated the model comparison between the two experimental conditions , switch and no-switch . We estimated the per-subject log-evidence for each experimental condition as the sum of log-evidences across the three corresponding experimental blocks . The top graph in both Figs 6 and 7 depicts the model attributions to the behavioral responses of each subject , that is , the posterior probability that a given model has generated the behavioral responses of each subject , for each condition separately . The bottom graphs show the corresponding XP for each of the 17 models . The direct comparison of behavioral models is inconclusive , as the highest XP is in both cases below the threshold value . Note that this is a typical issue when the model comparison set contains groups of closely related models [78] . The solution here is that instead of trying to answer which of the models provides the best description of behavioral data , we should ask which of the features of the perceptual and the response model are the most relevant for generating the data [78] . Note that in both figures we observe clustering of high model probabilities ( top graphs ) within closely related perceptual models ( e . g . B w1 , w2 , w3f ) which only differ in the type of the connectivity matrix ( see subsection Structured models in Methods ) . Thus , to determine which of the features of the perceptual and the response model are the most relevant for generating the behavioral data , we have performed four so-called family-wise model comparisons [78] . To test whether non-Bayesian or Bayesian model variants better describe the behavioral data , we grouped all models into baseline ( BM = {BM} ) , non-Bayesian ( NB={NBrw , rd , d , w1 , w2 , w3r} ) and Bayesian ( B={Brw , rd , d , w1 , w2 , w3f , r} ) model families . Similarly , to test whether a hierarchical representation of feature relevance is truly necessary we have grouped the models into BM , reduced perceptual ( RP={NBrw , rdr , Brw , rdr} ) , and full perceptual ( FP={NBd , w1 , w2 , w3r , Bd , w1 , w2 , w3f , r} ) model families . Finally , to test whether the attractor dynamics contributes to an explanation of the behavioral data , we have grouped models into the BM , structure-free ( SFM={NBrd , dr , Brdr , Bdf , r} ) , and structured ( SM={NBrw , w1 , w2 , w3r , Brwr , Bw1 , w2 , w3f , r} ) model families . In addition to separating behavioral models based on the features of perceptual model , we have grouped them based on the features of the response model , for which we considered only two model families , a model family with the reduced response model ( RR={BM , NBrw , rd , d , w1 , w2 , w3r , Brw , rd , d , w1 , w2 , w3r} ) and a family with the full response model ( FR={Brw , rd , d , w1 , w2 , w3f} ) . From the results of the four family-wise model comparisons , shown in Fig 8 , we can conclude with high confidence ( XP above the threshold level of 0 . 95 ) that the Bayesian formulation of the perceptual model is essential for generating behavioral data in both experimental conditions ( see Fig 8A and 8B ) . To understand the difference between NB and B model families in their ability to predict subjects’ behavior we tested how well the behavioral models within each of these families predict subjects’ performance . We computed the mean model performance by first estimating the expected performance per trial . To do this , we fixed model parameters to the mode β→ of the posterior parameter distribution and computed the expected model response; hence the expected performance per trial corresponds to the mean fraction of money assigned to the truly relevant visual feature at that trial . We averaged the per-trial expected model performance over a whole experimental block to obtain the mean model performance per experimental block . We then estimated the Pearson correlation coefficient between the mean model performance and mean subjects’ performance across blocks and both experimental conditions . In Fig 9 we illustrate , with a box plot , the distribution of the estimated correlation within NB and B model families . The correlation coefficient shows that , on average , the NB model family has significantly lower correlation with subjects’ performance , or in other words , the NB model family provides a worse fit to subjects’ behavior compared to the Bayesian model family . Interestingly , within the NB family the models with consistently low correlation , in both conditions , are the structure-free model variants NBdr and NBrdr ( see S1 Fig ) , whose update equation correspond to what is typically used in classical reinforcement learning models . On the other hand , the non-Bayesian model variants with attractor dynamics , namely NBrw , w1 , w2 , w3r , show consistently high correlation with subjects’ performance in both conditions ( with one exception being model NBw2r ) . This indicates that even only within the NB model family the attentional focus mechanism plays a critical role in replicating subjects’ behavior . Importantly , from the results of the family-wise model comparison we can also conclude with high confidence that the full variant of the perceptual model ( including both the 2nd and 3rd level of the hierarchy , see Reduced structured and structure-free models for details ) is an essential feature in both experimental conditions ( see Fig 8C and 8D ) . The structured family of the perceptual model shows an XP above the threshold level only in the no-switch condition ( Fig 8F ) , whereas in the switch condition the XP is slightly below the confidence threshold level ( Fig 8E ) , but still high enough to be considered a trend . One possible explanation for the slightly reduced confidence in the structured model family ( Fig 8E ) is that in the switch condition one expects high levels of posterior uncertainty about the relevance of visual features . This is due to an increased difficulty in assigning contradicting evidence either to an experimenter’s error or a change in the selection rule . Thus , in such an environment one does not expect that a subject can form strong beliefs about the relevance of each visual feature . Hence the attractor dynamics would not show strong advantages in generating the data , when compared to the structure-free model family . Finally , when comparing model families with the full against the reduced variant of the response model we get mixed results across conditions . The full response model seems to be relevant for generating behavioral data only in the no-switch condition ( Fig 8H ) , whereas in the switch condition the evidence is inconclusive ( Fig 8G ) . This discrepancy between the confidence levels in the two experimental conditions may be caused by the increased difficulty of the switch task , which effectively introduced a higher variability in subjects’ responses . Most of this variability may be explained simply by a high but constant level of response noise as formulated in the reduced response model . To illustrate the dynamics encountered under the most likely types of behavioral model ( Bw1 , w2 , w3f in the no-switch condition and Bw1 , w2 , w3r in the switch condition ) we have plotted the measured and modeled responses of a representative subject ( #9 ) , see Fig 10 . The modeled response was averaged over posterior model probability ( see top graphs of Figs 6 and 7 ) . Note that for the selected subject only the B w3f ( in the no-switch condition ) and Bw2r ( in the switch condition ) have posterior model probabilities close to one and therefore contributed to the shown modeled responses . Importantly , one can see that the expected model responses appropriately track the subject’s responses in all six experimental blocks , and that the deviations of the subject’s responses from the expected response are mostly explained by the response variability , as indicated by the shaded area .
The variant of the WCST used here can be seen as a simple but representative task to which humans are often exposed , namely making decisions in situations where the relevant features of the environment are not obvious but need to be inferred first . What makes the WCST simpler when compared to natural environment is the reduced number of possible pre-learned hypotheses . However , the dynamic complexity is comparable to real world situations: ( i ) the rules of the environment can change , and ( ii ) in the specific WCST used here the experimenter occasionally ‘makes a mistake’ just as in the natural environment one often cannot know something with certainty . For the WCST task , these two naturally occurring sources of uncertainties make the necessary inference sufficiently complex to compute the subject’s uncertainty about the relevance of visual features . To better infer the hidden internal beliefs and uncertainties of subjects , we used belief solicitation in a form of a betting assignment , which reflect a subject’s hidden beliefs over the space of possible hypotheses . To our knowledge , such belief solicitation was not previously used in a WCST task , although similar experimental designs were used for simpler tasks [11 , 79] . To incorporate attentional-focus within the perceptual part of the behavioral model we modeled the dynamics of the hidden states of a probabilistic generative model with a winner-take-all ( WTA ) dynamics . This is a well-known type of dynamics applied to artificial neural networks [37–40 , 80–82] and used as a part of connectionist models of decision making and planning [19 , 25] . In addition , WTA network dynamics have been reported to capture a wide range of experimental findings [48 , 83–86] . For our purposes , the WTA neuronal network implemented a dynamic and self-regulated attention formation at the top level of a hierarchical representation of environmental features . In comparison to the classical connectionist approach , e . g . [25] , the main advantage of using the WTA dynamics within a Bayesian framework is that the adaptive coupling between the intrinsic network dynamics and external input ( see Eqs ( 11 ) and ( 12 ) ) is derived automatically as part of the update equations . These update equations provide Bayes-optimal behavior of the model by setting the connection weights to their optimal value . Although the optimization technique used by the brain may be different , such weight optimization may be assumed as a guiding computational principle of information processing in the brain . Our finding—that competitive inhibitory WTA dynamics as a model of attentional focus is required for describing the hidden update process of subjects’ beliefs—is in agreement with previous findings of Wilson and Niv [1] . This suggests that in a WCST task humans actively track only the evidence corresponding to features they pay attention to , that is , the ones they found potentially relevant for the current task . Importantly , as a safe-guard against over-fitting the data with a complex WTA dynamics , we employed simpler ( with a reduced number of free parameters ) variants of the perceptual model . The fact that the less complex behavioral models have lower model evidence suggests that the WTA dynamics has indeed adequate complexity to describe the behavioral data . The WTA dynamics introduces the following features in the evolution of beliefs: ( i ) faster convergence of beliefs to the working hypothesis; ( ii ) the beliefs are more inert to frequent changes in the environment , that is , to switch between the hypotheses sufficient amount of contradicting evidence has to accumulate . ( iii ) The beliefs change faster if the changes in the environment are rare , as after the fixed point is reached beliefs do not evolve further . In contrast , the diffusive dynamics of the SFM variants of the perceptual model is not bounded within finite volume of the belief space . Hence , as the posterior beliefs about a hypothesis’ relevance can be strongly separated if the environment is stable for a long period of time and , once the switch occurs it would take a very long time to adjust the beliefs as nothing constrains the separation of the posterior expectations . Consequently , as our results suggest , the proposed attractor dynamics modulate expectations . This would predict the following effects on behavior: ( i ) Even small amount of evidence can have a big impact on beliefs , ( ii ) if changes in the environment are too frequent they will have smaller impact on beliefs than expected from the diffusive dynamics , and ( iii ) if changes in the environment are rare it will take less contradicting evidence to change the working hypothesis than predicted by the diffusive and unconstrained dynamics . Although various studies have demonstrated that human behavior can approximate a Bayesian observer [26–28 , 60–62 , 87] , human subjects can also behave sub-optimally when exposed to sufficiently complex tasks [28] . In recent work Acerbi et al . [51] have demonstrated that the response variability ( deviation from expected response ) is proportional to posterior uncertainty . Such a deviation from optimal responses can be explained if one assumes a stochastic representation of the posterior beliefs by the human brain [52 , 53] . Thus , to account for potential dependence of response variability on posterior uncertainty we considered two variants of the response model . In the first variant we assume that the response variability is constant over an experimental block . In the second variant we additionally allow for the variability of the modeled responses proportional to the posterior uncertainty ( see Eqs ( 15 ) and ( 16 ) ) , which accounts for the potential stochastic representation of posterior beliefs . Depending on the experimental condition both variants of the response model provide good accounts for the deviation of subjects’ responses from the optimal response . In the no-switch condition ( the relevance of visual feature is unchanged during the block , see Fig 8H ) we found that the response variability is indeed proportional to the posterior uncertainty; in the switch condition ( Fig 8G ) the evidence is inconclusive although in favor of the assumption that the response variability is fixed and independent of the posterior uncertainty . A reason for this inconclusive result may be the increased difficulty of the experimental task in the switch condition . An increased difficulty makes the behavioral responses noisier ( responses deviate more from the optimal response compared to the no-switch condition , see Fig 10 ) . As the average response variability increases , there is less information about the dependency of response variability on experimental trials . Hence , most of this additional variability may be explained simply by a rather high but constant level of response noise as formulated in the reduced response model . Earlier work on the computational role of attention in the processing of sensory information suggested that attention can be understood as prior expectations about the sensory stimuli [88 , 89] . This rather simple view of attention as a prior has recently been extended to account for both selective and integrative attentional phenomena [34–36] . This extended view suggests that due to the computational complexity of the exact probabilistic inference and the limited amount of available cognitive resources , the human brain has to rely on approximations to efficiently solve perceptual tasks . In other words , the role of attention is to assign limited cognitive resources to the relevant part of the sensory stimuli , which provides local refinement of the internal representation of the hidden states of the environment . However , this view on attention as an approximation to the exact Bayesian inference has been recently challenged . Under the free-energy principle [90]—which suggests that perception , attention , and action are all aimed toward suppressing the perceptual surprise about future sensory stimuli—attention is viewed as a sampling of only those parts of sensory stimuli that have high-precision in relation to the predictions of the internal model of the world [33] . Importantly , if the model of the world also predicts the precision of different parts of sensory stimuli , then that prediction is what Friston and colleagues propose to be associated with attention . Our work presented here can be related to both assumptions about the computational role of attention , and as such cannot reconcile this dispute . Note , that the competitive attractor dynamics can be seen both as an approximation to the exact inference ( the attractor dynamics regulates the update of beliefs by assigning the computational resources only to the most relevant hypothesis ) and as a suppressor of the perceptual surprise ( the attractor dynamics actively reduces the uncertainty about future sensory stimuli by predicting both the future expectation and precision of a categorical probability of hypothesis relevance; see Eq ( 9 ) ) . We believe that the probabilistic WCST provides a promising experimental paradigm for investigating complex behavioral models . However , one can probably improve on the current design using two changes . Firstly , in spite of the initial training , several subjects exhibit rather poor performance in the no-switch condition ( see Fig 2 ) . Ten out of twenty two subjects show poor performance in at least one experimental block of the no-switch condition . Importantly , we have included these subjects in our analysis , because the model comparison did not show any correlation between subjects’ performance and the best fitting behavioral model . Also note that a key strength of the proposed model is that it can explain this poor performance well , see for example Fig 10; insofar a potentially suboptimal performance does not pose a limitation to the proposed modelling approach . However , the obtained results may be even more compelling if subjects practiced the task until a stable performance is reached for both conditions . Secondly , as mentioned in the Methods section , the error rate ε was set to values that induced the most distinct behavioral responses between two experimental conditions , while rendering the switch condition informative enough to induce betting responses in subjects . However , these led to a partially imbalanced manipulation between conditions . Thus , a potential improvement would be to introduce a fractal design , such that both the error rate and the switch probability are incrementally increased . Such a fractal design would provide further insights into how each environmental parameter influences behavior and what effects , if any , each parameter might have on the model comparison . Similar to the experimental design , the analytical approach presented here may also be potentially improved upon . Firstly , as mentioned in the Methods section , the behavioral model proposed here is not the only possible formulation . Depending on how one defines the observation likelihood ( Eq ( 4 ) ) and the parametrization of the hypothesis probability ( Eq ( 5 ) ) , one can obtain different variants of the perceptual model . Although we have tested a couple of them ( one additional , alternative formulation is described in S1 Text ) , there is a large number of possible perceptual models . We anticipate that more studies are required to come to a general conclusion which of the models or model families is the most useful for describing behavioral data of studies similar to the one presented here . Secondly , the model comparison presented here relies solely on the Bayesian model selection that is useful for inferring which of the given models is most likely to generate the data . However , it cannot be directly used to answer the question whether a given model is a good predictor of behavior . To address this question one has to rely on cross-validation strategies , that is , on model testing [91] . Still , one important prior assumption of model testing is that the behavior can be described by parameters which are stable over blocks . We do not assume that this is the case for our experimental data as subjects were not over-trained which would motivate the assumption that subjects performed the task in some stable parameter regime . Thus , it is plausible that the experience in previous experimental blocks influences , at least slightly , the behavior in subsequent blocks . For this reason model testing may not be usefully applicable to our study . Nevertheless , for future studies changes to the training procedure may stabilize behavior across experimental blocks and would allow one to also apply model testing methods to predict behavior . Although the presented analysis has been applied to behavioral data only , it would be potentially useful and feasible to extend the behavioral analysis to the investigation of neuroimaging data . The inferred belief trajectories would be used as regressors [13] , and thus can provide insights into the functional aspects of specific brain areas involved in the decision making process during the ongoing task . We found strong evidence that an attention-like mechanism modulates the update of beliefs in human subjects who had to infer the relevance of various features in a dynamic and noisy environment . Effectively , this attentional focus facilitates the increase of expectations about the relevant feature and inhibits the expectations about irrelevant features . Subsequently , these modulated expectations affect update of beliefs . We expect that the same computational mechanism can be applied to modelling other complex tasks that impose high cognitive load on subjects , thus require the attentional focus strategies for decision making .
|
When making decisions in our everyday life ( e . g . where to eat ) we first have to identify a set of environmental features that are relevant for the decision ( e . g . the distance to the place , current time or the price ) . Although we are able to make such inferences almost effortlessly , this type of problems is computationally challenging , as we live in a complex environment that constantly changes and contains an immense number of features . Here we investigated the question of how the human brain solves this computational challenge . In particular , we designed a new experimental paradigm and derived novel behavioral models to test the hypothesis that attention modulates the formation of beliefs about the relevance of several environmental features . As each behavioral model accounted for a different hypothesis about the underlying computational mechanism we compared them in their ability to explain the measured behavior of human subjects performing the experimental task . The model comparison indicates that an attentional-focus mechanism is a key feature of behavioral models that accurately replicate subjects’ behavior . These findings suggest that the evolution of beliefs is modulated by a competitive attractor dynamics that forms prior expectation about future outcomes . Hence , the findings provide interesting and novel insights into the computational mechanisms underlying human behavior when making decisions in complex environments .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Modeling the Evolution of Beliefs Using an Attentional Focus Mechanism
|
NF-κB pathways are key signaling cascades of the Drosophila innate immune response . One of them , the Immune Deficiency ( IMD ) pathway , is under a very tight negative control . Although molecular brakes exist at each step of this signaling module from ligand availability to transcriptional regulation , it remains unknown whether repressors act in the same cells or tissues and if not , what is rationale behind this spatial specificity . We show here that the negative regulator of IMD pathway PGRP-LF is epressed in ectodermal derivatives . We provide evidence that , in the absence of any immune elicitor , PGRP-LF loss-of-function mutants , display a constitutive NF-κB/IMD activation specifically in ectodermal tissues leading to genitalia and tergite malformations . In agreement with previous data showing that proper development of these structures requires induction of apoptosis , we show that ectopic activation of NF-κB/IMD signaling leads to apoptosis inhibition in both genitalia and tergite primordia . We demonstrate that NF-κB/IMD signaling antagonizes apoptosis by up-regulating expression of the anti-apoptotic protein Diap1 . Altogether these results show that , in the complete absence of infection , the negative regulation of NF-κB/IMD pathway by PGRP-LF is crucial to ensure proper induction of apoptosis and consequently normal fly development . These results highlight that IMD pathway regulation is controlled independently in different tissues , probably reflecting the different roles of this signaling cascade in both developmental and immune processes .
In Drosophila , bacteria infection triggers NF-κB cascades ( called Toll and IMD ( Immune Deficiency ) ) leading to the production of immune effectors and regulators [1–3] [4] . This activation relies on the previous recognition of bacteria derived peptidoglycan ( PGN ) by host Peptidoglycan Recognition Protein ( PGRP ) family members . Recognition of Gram-positive PGN by circulating PGRP-SA triggers the maturation of the pro-Spätzle protein into an active ligand for the Toll membrane receptor [5] . The IMD pathway is triggered upon recognition of PGN by either transmembrane associated PGRP-LC or cytoplasmic PGRP-LE [5–10] . Receptor activation leads to the recruitment of the death-domain containing adapters IMD , FADD , and the caspase DREDD . Activated DREDD cleaves IMD , thus allowing its poly-ubiquitination that allows recruitment of the TAK1/TAB2 and the IRD5/Kenny kinase to the receptor complex [11] . These interactions ultimately lead to the nuclear translocation of the transcription factor Relish . In contrast to Toll signaling , IMD pathway activation after bacterial infection is transient and buffered by many repressors [12–14] . This tight control might reflect the essential role played by the IMD cascade in controlling antibacterial response in fly epithelia [4 , 15] . Indeed , the constant contact between bacteria and epithelia requires the presence of immune tolerance mechanisms through which the epithelium copes with the continuous input from microbiota derived immune-activating signals [16–19] . The homeobox transcription factor Caudal was one of the first proteins identified as an IMD pathway antagonist [20] . Through its occupation of some IMD target promoters , Caudal blocks IMD-dependent transcription . Negative regulation is also mediated through protein turnover by ubiquitous factors that regulate protein stability of identified IMD pathway components ( dUSP36 , CYLD , DNR-1 , Caspar ) [21–23] [24–26] . Some of IMD pathway regulation is also taking place at the level of the PGRP-LC receptor itself and of its PGN ligand . PGRP-LC transcription is under the control of the steroid hormone ecdysone [27] . The number of PGRP-LC molecules at the membrane , depends on intracellular ( Pirk ) and membrane associated ( nonaspanins TM9SF2 and TM9SF4 ) proteins that by sequestering PGRP-LC in the cytoplasm prevent its localisation at the membrane [16 , 28–31] . Another member of the PGRP family , PGRP-LF , antagonizes IMD pathway activation [32] . This transmembrane protein which has no intracytoplasmic tail but has two occluded PGRP domains is unable to bind PGN [32 , 33] . Plasma resonance data show that by interacting with PGRP-LC ectodomain , PGRP-LF prevents constant activation of the IMD pathway even in the absence of bacteria [34] . IMD pathway tuning is also mediated through the modulation of ligand availability , via a family of extracellular enzymes called amidases , which degrade PGN into non-stimulatory fragments [17 , 18 , 35–40] . While the inhibition provided by these regulators appears to be constitutive , the negative regulation brought by amidase and by PIRK is the result of a negative feedback loop . As a result , these factors additively regulate the amplitude of the IMD response . While the detrimental effects of runaway inflammation in mammals are well established , the situation is less clear with regards to Drosophila . The absence of ubiquitous negative regulators leads to a reduced lifespan , which however cannot be ascribed specifically to the constitutive activation of the IMD pathway , as these regulators act upon multiple targets . Modulation of amidase levels causes deregulation of NF-κB activity in the gut , resulting in commensal dysbiosis , stem cell hyper-proliferation , epithelial dysplasia and eventually reduced life span [18 , 40] . Importantly , these phenotypes can be partially rescued in germ-free conditions or by inactivating IMD pathway components demonstrating that they are direct consequences of IMD pathway stimulation by bacteria . Although IMD pathway is controlled at multiple levels along the cascade , the rationale behind this complex and multilayered regulation is not fully understood . It remains to be known whether this regulation is tissue specific and if yes , what are the consequences of IMD activation in each tissue . Our results show that PGRP-LF is expressed in ectodermal derivatives some of them with no known immune function . Phenotypic analysis of a newly generated PGRP-LF protein null allele demonstrates that its function is required for the formation of various epidermal derived adult structures . We showed that PGRP-LF's role in controlling developmental process is , like for the immune ones , mediated by a repression of the IMD pathway . Our results demonstrate that , in contrast to what has been shown in other tissues , IMD pathway activation in epidermal structures is blocking rather than triggering apoptosis . By preventing such IMD mediated anti-apoptotic signal , PGRP-LF is allowing normal development to take place . These results speak for the importance of a tissue specific regulation of NF-κB pathway in flies and for a strong link between immune and developmental processes .
To reveal the expression pattern of the PGRP-LF gene , we generated reporter lines ( later named PGRP-LFGal4 ) in which 1 . 4 Kb of genomic DNA 5’ of the PGRP-LF coding region ( corresponding to the PGRP-LC/PGRP-LF intergenic region , S1A Fig ) was cloned upstream of Gal4 coding sequences . PGRP-LFGal4 , UAS-GFP larvae showed fluorescence in most larval ectodermal derivatives such as the cuticle , the salivary glands , the fore and hindguts and to a lesser extend , the trachea ( Fig 1A and S2A Fig ) . mRNAs quantification confirmed the strong enrichment of PGRP-LF transcripts in ectodermal tissues ( especially the cuticle , the foregut and the hindgut ) and low levels in the fat body and in the midgut ( when compared to foregut and hindgut ) , in accordance with FlyAtlas data ( Fig 1B and S2B and S2C Fig ) . This was unexpected since these mesodermal derivatives are the main immune tissues in which bacteria infection is triggering IMD pathway activation . To elucidate the role of PGRP-LF in ectodermal tissues , we generated a novel PGRP-LF allele called PGRP-LFKO ( S1A Fig ) . Indeed , detailed molecular characterization demonstrated that the previously PGRP-LF200 allele is not molecularly null for PGRP-LF ( S1B Fig ) . In addition , it retains P-element sequences 5' of the PGRP-LF ORF that might interfere with neighboring loci that code for proteins putatively interacting with PGRP-LF ( PGRP-LC , PGRP-LA ) ( S1A Fig ) . The new PGRP-LFKO allele obtained by homologous recombination , completely abolished PGRP-LF mRNA expression without affecting the transcription of the neighboring UGP gene ( S1B and S3A Figs ) . The increased PGPR-LC transcription detected in PGRP-LFKO flies is secondary to IMD pathway activation since it was reduced in IMD pathway mutant backgrounds ( S3B Fig ) . Although previous results show that PGRP-LF negatively regulates the IMD pathway , the tissue specificity of PGRP-LF effects remain unknown [33] . We therefore compared the transcription levels of three IMD pathway target genes , AttacinD , Drosomycin and Diptericin , in wild-type and PGRP-LFKO mutant flies . PGRP-LFKO or PGRP-LFKO/Df ( 3L ) BSC113 flies showed a strong constitutive expression of AttacinD in most ectodermal larval derivatives ( Fig 2A and 2B ) . Although Diptericin mRNA levels were up-regulated in the trachea and the salivary glands of PGRP-LF mutants , they were not in the cuticle and in the hindgut , demonstrating that IMD dependent antimicrobial peptide ( AMP ) genes transcription is differently regulated in different tissues ( Fig 2C and S4A Fig ) . This is well illustrated with Drosomycin being only up-regulated in tracheal cells of PGRP-LFKO larvae ( S4A Fig ) . Although Diptericin is highly inducible by bacterial infection in fat body cells , a relatively mild increase of Diptericin mRNA was detected in this tissue when PGRP-LF was inactivated ( Fig 2C ) . By combining PGRP-LFKO mutants with AttacinD-cherry and PGRP-LFGal4 /UAS-GFP reporters , we could show that tissues expressing PGRP-LF are the ones that display constitutive AMP expression upon PGRP-LF inactivation ( S4B Fig ) . This tissue-autonomous expression indicated that AMP ectopic expression seen in PGRP-LF mutants was not due to global stress response of the larvae but rather to a direct consequence of IMD pathway activation in specific tissues . Epistatic experiments further showed that AMP ectopic expression observed in PGRP-LF mutants are due to PGRP-LC dependent IMD pathway activation . Indeed , AMP ectopic expression in PGRP-LF mutants was only moderately reduced by the functional inactivation of the intracytoplasmic receptor PGRP-LE and of the Toll signaling component dMyd88 . In contrast , it was completely suppressed when IMD pathway components were inactivated ( Fig 3A and S4C Fig ) . This is consistent with the fact that in tissues expressing PGRP-LF ( trachea , hindgut , epidermis… ) , IMD pathway activation has been shown to rely on the upstream PGRP-LC transmembrane sensor [33] . It should be noted that the constitutive AMP expression was not observed in single mutants for other IMD pathway negative regulators such as PGRP-LB or Pirk although synergistic effects were observed ( Fig 3B ) . The above results demonstrated that PGRP-LF acts to prevent constitutive IMD pathway activation in the absence of bacteria . To appreciate the reason of such a regulation , we compared wild-type and PGRP-LF mutant fitness in the absence of bacteria . When flies were grown in axenic conditions , PGRP-LF mutants succumbed earlier than their wild-type siblings ( Fig 4A ) . This premature death was largely suppressed by an absence of the caspase Dredd , demonstrating that the precocious lethality was mainly due to IMD constitutive activation . We then tested the ability of PGRP-LF mutant flies to mount an immune response to bacteria and to resist to them . While PGRP-LF adults showed , like PGRP-LF larvae , ectopic expression of AMP , their ability to trigger IMD pathway activation following Ecc septic injury was similar to that of wild-type controls ( Fig 3C ) . Consistently , PGRP-LF mutant flies survived Ecc septic injury as well as control flies ( Fig 4B ) . Interestingly , Ecc orally infected PGRP-LF mutants succumbed faster than controls with a kinetic close to that of PGRP-LB mutants ( Fig 4C ) . Remarkably , PGRP-LF adult showed increased systemic AMP production when compared to controls , although local gut AMP production was normal ( Fig 3D and S5 Fig ) . It is difficult to explain why PGRP-LF mutants show over-activation of IMD pathway in the fat body after Ecc oral infection but not after septic injury . The most likely reason is that the nature of the PGN that reaches PGRP-LC at the surface of the fat body cells is not the same whether it comes directly by pricking bacteria in the fly thorax or whether it translocates from the gut lumen . Indeed , PGN modifying enzymes , such as PGRP-SC and PGRP-LB , are differently expressed in the hemolymph and in the gut ( Fig 4C ) . In any case , the reduced lifespan of Ecc orally infected mutants was partially rescue by inactivating Dredd . These results demonstrate that by preventing IMD pathway activation in the absence of infection , PGRP-LF prevents precocious death . They also show that although PGRP-LF does not affect the ability of the fly to respond to septic infection , it is required to prevent death following Ecc infection . AMP levels indicated that precocious lethality is probably due to IMD pathway over activation in fat body cells and not in enterocytes themselves . The PGRP-LFKO allele is sub lethal with only some pupae ( 27% +/- 7 SD , n = 430 ) hatching as adults [33] . All adults and the dead pupae exhibited , with a range of severity , stereotypic malformations in the abdominal tergites with disruption in the joining of the cuticular plates along the dorsal midlines of the abdomen , exposing underlying soft tissue ( Fig 5A ) . In addition , all PGRP-LFKO male escapers displayed defects in male genitalia orientation ( Fig 5A and S6B Fig ) . Theses phenotypes were completely rescued by expressing the wild-type PGRP-LF cDNA with the PGRP-LFGal4 driver , demonstrating that they were indeed due to a lack of PGRP-LF protein in the ectodermal territories ( Fig 5A ) . To test whether these developmental phenotypes were , as for the AMP induction , consecutive to IMD pathway activation , we performed epistasis experiments . Inactivation of IMD , Dredd , Kenny , Relish but not the JNK mediator Hemipterous , completely suppressed abdominal tergites and genitalia orientation defects of PGRP-LF mutants ( Fig 5A and S6A Fig ) . Similar results were obtained when PGRP-LF was inactivated only in its expression domain via RNA interference ( Fig 5A ) . This suggests that uncontrolled activation of IMD pathway in cells that are fated to generate these structures is incompatible with proper developmental processes . We further tested this hypothesis by analyzing the consequences associated with a constitutive IMD pathway activation in cells that will give rise to tergites and genitalia . Ectopic expression of either IMD or PGRP-LC , whose overexpression is sufficient to activate IMD signaling in the absence of bacteria , in the PGRP-LF expression domain fully phenocopied PGRP-LF mutant in tergites and genitalia ( S7A Fig ) . Only one of the two adult phenotype was observed when Gal4 drivers specific to either tergites ( DdcGal4 ) or genitalia ( AbdBLDN-Gal4 ) was used ( S7A Fig ) . These lines specifically target larval epidermal cells ( DdcGal4 ) and the outer ring of cells of the A8 abdominal segment ( AbdBLDN-Gal4 ) , both being cells in which induction of apoptosis is essential for proper development . Altogether , these results demonstrate that PGRP-LF function is needed , in the absence of bacterial infection , to prevent constitutive activation of IMD pathway in ectoderm , which will otherwise provoke developmental defects . During metamorphosis , tergites and male genitalia development requires apoptosis [41–45] . Consistently , mutations in apoptotic genes such as Drice , or overexpression of the apoptosis inhibitors P35 or Diap1 in tergite and genitalia anlage , give rise to adult flies presenting cuticle midline and genitalia rotation defects [41] . To test if PGRP-LF mutant phenotypes are due to apoptosis inhibition in ectodermal anlage , we overexpressed P35 in PGRP-LF expression domain . PGRP-LFGal4 , UAS-P35 flies fully phenocopied PGRP-LF mutants . Although , such phenotypes were not observed with a weaker PGRP-LFGal4 driver , it occurred in flies heterozygous for the PGRP-LFKO allele ( S7B Fig ) . To further confirm these results , we compared the morphogenetic movements associated with the formation of these ectodermal structures in wild-type and PGRP-LF mutants . First , we investigated tergite patterning which requires replacement of larval abdominal epidermis with adult epithelium during pupariation . During this process , larval epidermal cells ( LECs ) undergo caspase-dependent apoptosis and are progressively replaced by histoblasts that expand as the LECs die . Inhibition of apoptosis in LECs by P35 overexpression or by Drice inactivation ( Fig 5B ) , leads to the persistence and accumulation of LECs along the dorsal midline of the pupal epidermis . A very similar LECs accumulation was observed in PGRP-LF mutant pupae but not in controls ( Fig 5B ) . To monitor apoptosis in these tissues , we took advantage of the anti-Dcp1 antibody that specifically recognizes the cleaved from of the effector caspase and therefore labels dying cells . Using such a tool , we detected a reduced number of LECs undergoing apoptosis in PGRP-LF mutant pupae , when compared to controls ( Fig 5C and 5D ) . The fact that PGRP-LF is expressed in LECs but absent from the histoblasts , suggests that it is required tissue autonomously in LECs to allow proper induction of apoptosis ( S7D Fig ) . We next examined male genitalia rotation using video recording of controls and PGRP-LF mutant pupae . During pupariation , the male genitalia rotates 360° clockwise and the acceleration and full completion of this process requires apoptosis of specific cells located in the AbdB expressing segment . We found that if genitalia of control flies rotate of 360° within 15 hours , those of PGRP-LF mutants only reach a 240° rotation during this period ( Fig 6A ) . Moreover , both the angle and the velocity of the genitalia rotation were reduced in PGRP-LF mutants compared to wild-type although they were not as affected as in a DriceΔ1 mutants ( Fig 6A and 6B ) . Hence , genetic data and live imaging data demonstrate that the lack of PGRP-LF protein affects developmental processes by preventing normal apoptosis to occur in these tissues . IMD pathway could block apoptosis by either repressing pro-apoptotic genes or by activating apoptosis inhibitors [46] . To distinguish between these possibilities , we compared expression levels of pro and anti-apoptotic genes in PGRP-LF mutants and in controls . First , using the Diap1-GFP4 . 3 and the PGRP-LFGal4 transgenes , we analyzed the expression pattern of Diap1 and PGRP-LF in wild type conditions . In both L3 and pupae , PGRP-LFGal4 was expressed in all LECs expected for one row of cells repeated along the antero-posterior axis ( S8A Fig ) . Remarkably , although Diap1-GFP4 . 3 expression was weak in most LECs , it was strongly expressed in these cells that do not express PGRP-LF , suggesting that PGRP-LF could repress Diap1-GFP4 . 3 expression . We next compared Diap1-GFP4 . 3 expression in wild-type and PGRP-LF mutant larvae and pupae . PGRP-LF inactivation was associated with an ectopic expression of Diap1-GFP4 . 3 in all LECs that became uniformly positive for Diap1-GFP4 . 3 , but also in other PGRP-LF expressing tissues such as the trachea and the hindgut ( Fig 7A and S9A and S9B Fig ) . q-RT-PCR quantification confirmed the increased expression of Diap1 mRNAs in PGRP-LF mutant larval tissues such as the cuticle , the hindgut and the trachea , when compared to controls ( Fig 7B ) . Consistently , ectopic activation of the IMD pathway in the larval epidermis and in the fat body was sufficient to induce ectopic expression of Diap1-GFP4 . 3 transgene in an autonomous fashion ( Fig 7C ) and this was confirmed by q-RT-PCR quantification ( Fig 7D ) . In contrast , neither loss-of function of PGRP-LF nor gain-of-function of IMD were able to activate pro-apoptotic Hid or Reaper transgenic reporter constructs in these tissues ( S10A–S10C Fig ) . We could also showed that Diap1 ectopic expression using either a ubiquitous Gal4 ( Act5CGal4 ) or PGRP-LFGal4 drivers was sufficient to mimic both cuticle and genitalia defects ( S7C Fig ) [41] . Altogether these results show that by preventing IMD pathway activation in ectodermal derivatives , PGRP-LF is allowing normal cell death to occur and pupal development to proceed .
We showed here that one of essential role of PGRP-LF is to prevent a bacteria independent constitutive activation of the NF-κB pathway , which otherwise perturbs tergite and genitalia formation during pupariation . We also confirmed previous data showing that a lack of IMD pathway repression by PGRP-LF is leading to AMP production in ectodermal derivatives [33] . Both PGRP-LF expression pattern and loss-of-function phenotype analyses showed that PGRP-LF is mainly acting in ectodermal cells . This is mostly evident in the intestinal tract that is formed during embryogenesis by associating domains of both mesodermal and ectodermal origins . Although IMD pathway is essential in regulating antibacterial response in the mesodermal derived midgut , PGRP-LF is only playing a minor role as an IMD regulator in this tissue . This contrasts with its importance in both neighboring ectodermal derivatives that are the fore and the hindgut . Loss of PGRP-LF function triggers in these tissues a massive AMP production . Interestingly also , is the fact that the effects of inactivating PGRP-LF , and hence of IMD pathway permanent activation , are not the same in all ectodermal derivatives . Whereas in trachea , epidermis or hind/foregut , it only leads to AMP constitutive production , it has profound and deleterious effects on tergites and genitalia . Of note , these are the two known structures whose proper morphogenesis has been shown via P35 overexpression to depend on caspase activity [41] . Removing PGRP-LF , blocks apoptosis and in turn interferes with developmental processes of these adult structures . It could be that apoptosis is also prevented in other ectodermal derivatives but that this has no impact on their development . Since mutations in caspases cause pleiotropic defects during development , it is obvious that PGRP-LF is only antagonizing a limited fraction of them , consistently with its restricted spatial pattern [47] . The fact that IMD pathway constitutive activation is blocking apoptosis is unexpected since previous results rather spoke for a pro-apoptotic function of IMD activation [48 , 49] . Ectopic activation of the IMD protein was shown to induce apoptosis in fat body cells while ectopic expression of the anti-apoptotic protein P35 was shown to prevent IMD induced AMP production [48] . In our hands , ectopic activation of IMD in imaginal discs , generates visible IMD pathway dependent phenotypes that were not suppressed by P35 overexpression ( S11 Fig and S1 Table ) . These results suggest that IMD pathway activation can have very different consequences depending on the cellular context . The rationale behind this could reflect the properties of the infected tissues . In proliferating epithelial tissues , such as the midgut or imaginal discs , dying cells induce their neighbors to divide to compensate for the lost space [50 , 51] . In such tissues , IMD pathway mediated cell death following infection should be without deleterious consequences . However , in cells that have left the mitotic cycle such compensatory proliferation cannot be a tissue repair system . Recent work has shown that these tissues are instead relying on compensatory cell hypertrophy and polyploidization for wound healing [52 , 53] . One can imagine that IMD triggered cell death in such tissues should be as much as possible prevented . PGRP-LF could play such a role . The results of this study reinforced the idea that in the absence of PGRP-LF , IMD pathway is constitutively active in the complete absence of bacterial product . This raises the question of the mechanisms of PGRP-LC mediated IMD activation . One possibility is that an endogenous ligand not derived from bacteria and probably endogenous is able to activate the IMD pathway . Before this putative elicitor is identified , this hypothesis would be difficult to prove . However , since PGRP-LC cleavage ( as for other receptor such as Notch ) has been shown to be sufficient for its activation and hence for IMD pathway triggering , endogenous proteases are good candidate for such PGRP-LC bacteria independent activators [54 , 55] . Consistently , clean wounding of the epidermis is sufficient to trigger AMP production in the absence of any bacteria [56] . Since embryonic development , and specially metamorphosis during which tergites and genitalia form , rely on extensive tissue remodeling including protease expression and release to eliminate dead tissues , this hypothesis is worth considering . Alternatively , PGRP-LC activation could take place in the complete absence of ligand . Crystal structure experiments have shown that the presence of PGRP-LF at the cell surface is reducing the probability of forming unwanted signaling PGRP-LC dimers in the absence of bacteria ligand hence contributing to the maintenance of a low IMD background level [34] . In this case , AMP expression in PGRP-LF deficient ectodermal tissues will be due to a PGRP-LC spontaneous dimerization sufficient to trigger signalling . Finally , PGRP-LF has been shown to compete with PGN derived ligand TCT for binding to PGRP-LC . By doing so , PGRP-LF could also increase the thresholds at which IMD pathway is activated by the presence of bacteria derived elicitor . Cells expressing PGRP-LF will require more PGN to activate the IMD pathway than cells devoid of it . It is interesting to note that PGRP-LF is expressed in foregut and hindgut epithelia that are not protected by a peritrophic membrane and are therefore in direct contact with bacteria . The presence of PGRP-LF at the membrane of such cells would insure that IMD pathway activation is only acting when a certain concentration of PGN elicitor is reached . Very transient and weak PGN level fluctuations will not be detected in these regions preventing responses to spurious stochastic events in the enterocytes . In contrast , neighboring midgut enterocytes are separated from bacteria by the peritrophic membrane that only allows diffusion of molecules such as nutriments or bacteria derived molecules . In the case of high bacteria loads , it is expected that high amount of PGN are released by bacteria , cross the peritrophic membrane and reach the midgut enterocytes . Detection of PGN by midgut enterocytes would therefore be the sign of a gut infection and should be followed by IMD pathway activation . The presence of PGRP-LF in such cells will prevent IMD activation . The same idea will hold true for the fat body cells in which PGRP-LF is playing a minor role . Fat body cells are sampling the circulating hemolymph to detect the presence of PGN that , in uninfected flies , should not be present in the body cavity . This sensing mechanism should be as sensitive as possible and not buffered by the presence of PGRP-LF at the fat body cell membranes . We propose here that the presence of PGRP-LF as an IMD pathway activation modulator is depending of cell types and on their role in the antibacterial response . The PGRP-LFKO mutant that we have generated is sublethal with few escapers showing tergites and genitalia defects but normal wings . We have demonstrated that these developmental defects are likely due to an apoptosis blockage . The previously characterized PGRP-LF200 allele did not present cuticle and genitalia defects but had wing notchings that were also observed by reducing PGRP-LF levels via RNAi in wing imaginal discs . These defects were associated with ectopic activation of the JNK pathway and increased apoptosis . To try to explain these discrepancies , we have characterized the PGRP-LF200 allele that was obtained by P element mobilization . Our results show that if the PGRP-LF200 allele has lost the white cDNA that serves as an eye marker for the P-element , it has kept some of the P element sequence 5’ to PGRP-LF coding region that might interfere with neighboring gene expression ( S1 Fig ) . Q-RT-PCR analysis indicates that PGRP-LF mRNA levels are globally not affected in these flies that do also not show ectopic Diptericin transcription . We hypothesized that wing notching in PGRP-LF200 flies are more likely due to an imaginal disc specific modulation of the relative PGRP-LF/PGRP-LC ratio than to a global loss of PGRP-LF function . Interestingly , a recent study identified some of the NF-κB pathway components , among which PGRP-LC , as proteins involved in a surveillance of cell fitness and cell competition in wing imaginal discs [57] . This might explain why elimination of PGRP-LF function in the entire wing imaginal disc ( such as in PGRP-LFKO ) or some domains only ( RNAi to PGRP-LF ) give rise to different phenotypes .
The following strains were used in this work: PGRP-LFGal4weak ( this work ) , PGRP-LFGal4strong ( this work ) , PGRP-LCGal4 ( this work ) , UAS-nlsGFP BL#4775 , UAS-myr-mRFP BL#7118 , DdcGal4 BL#7010 , AbdBLDNGal4 BL#55848 , FngGal4 BL#9891 , apmd544Gal4 BL#3041 , enGal4 , UAS-RFP BL#30557 , Act5CGal4 BL#25374 , UAS-mcd8CherryRFP BL#27392; AttacinD-Cherry ( this work ) , Diptericin-Cherry [58] , Drosomycin-GFP [59] , Diap1-GFP4 . 3 [60] , hid5’F-GFP , rprNP0520Gal4 DGRC#103634 , His2Av-mRFP BL#23651 , imdshadok [48] , PGRP-LE112 [8] , PGRP-LCΔE12 [61] , dMyd88c03881 [62] , DreddD55 [63] , Diap2c7c [64] , RelishE20 [65] , pirkEY0073 [31] , PGRP-LBKO [40] , dTak1 [66] , Dcp-1Prev1BL#63814 , DriceΔ1 [67] , UAS-P35 [48] , UAS-PGRP-LCa BL#30917 , UAS-p53 BL#8418 , UAS-Diap1 BL#63820 , UAS-PGRP-LF-IR ( this work ) , UAS-Dicer2 BL#24650 . Flies were grown at 25°C on a yeast/cornmeal medium . For 1l of food , 8 . 2g of agar ( VWR , cat . #20768 . 361 ) , 80g of cornmeal flour ( Westhove , Farigel maize H1 ) and 80g of yeast extract ( VWR , cat . #24979 . 413 ) were cooked for 10 min in boiling water; 5 . 2 g of Methylparaben sodium salt ( MERCK , cat . #106756 ) and 4 ml of 99% propionic acid ( CARLOERBA , cat . #409553 ) was added when the food had cooled down . For antibiotic treatment , standard medium was supplemented with Ampicillin , Kanamycin , Tetracyclin and Erythromycin at 50 μg/ml final concentrations . PGRP-LFKO line was generated by homologous recombination . The PGRP-LF gene was replaced by a mini-white gene . DNA flanking the 5’ and 3’ ends used were respectively , 2 , 914 bp and 3 , 060 bp for PGRP-LF locus . Sequences were cloned in the pW25 vector [68] . Larval , pupal or adult tissues were dissected in PBS , fixed for 20 min in 4% paraformaldehyde on ice and rinse 3 times in PBT ( PBS + 0 . 1% Triton X-100 ) . For antibody staining on pupal cases , cleaved anti-Dcp-1 antibody ( Cell Signaling #9578 ) was used at 1:200 . The tissues were mounted in Vectashield ( Vector Laboratories ) fluorescent mounting medium , with or without DAPI . Images were captured with either a Stereo Discovery V12 microscope or a LSM 780 Zeiss confocal microscope . Staged pupae ( 24 hours APF ) were washed in water and mounted on a glass slide using a drop of silicon grease ( Dow corning ) . The pupal case covering the caudal part of the abdomen was removed . A very wet filter paper was placed around the pupae to keep them hydrated . The pupae were covered with a cover glass in a small drop of water to avoid desiccation . High-vacuum silicone grease ( Dow Corning ) was also used to seal the chamber . In most cases , the animal survived the data acquisition and developed into an adult . Time-lapse images were captured using a Nikon Macroconfocal AZ100 . RNA from whole larvae or dissected organs ( n = 30 ) was extracted with RNeasy Mini Kit ( QIAGEN , cat . #74106 ) . Quantitative real-time PCR , TaqMan , and SYBR Green analysis were performed as previously described [36] . Primers information can be obtained upon request . The amount of mRNA detected was normalized to control rp49 mRNA values . Normalized data was used to quantify the relative levels of a given mRNA according to cycling threshold analysis ( ΔCt ) . The bacterial strain used was Erwinia carotovora carotovora 15 2141 ( Ecc ) cultured in Luria-Bertani medium at 30°C overnight . Bacterial cultures were centrifuged at 2500 g for 15 min at RT and resuspended in fresh Luria-Bertani medium . Cells were serially diluted in PBS and their concentration was determined by optical density ( OD ) measurement at 600 nm . For oral infection , flies were first incubated 2 hr at 29°C in empty vials and then placed in a fly vial with food . The food solution was obtained by mixing a pellet of an overnight culture of bacteria Ecc-15 ( OD = 200 ) with a solution of 5% sucrose ( 50/50 ) and added to a filter disk that completely covered the agar surface of the fly vial . Septic injuries were performed by pricking adult females with a thin needle contaminated with Ecc-15 . For oral infections , adult flies were infected every 2 days with a solution of Ecc ( OD = 200 ) 5% sucrose ( 50/50 ) . For septic injuries , adult females were pricked once with a thin needle contaminated with Ecc . At least two tubes of 20 flies were used for each survival assay and three replicates of this experiment were done . Survival was scored several times a day .
|
In multicellular organism such as mammals or insects , activation of innate immune responses occurs following detection of microbes by dedicated receptors called pattern recognition receptors . Such immune activation is taking place in immune competent tissue such as the skin , the digestive and respiratory epithelia and is under a tight negative control . Negative control is essential to finely adjust the duration and the intensity of the immune response to the level of infection . We found that the Drosophila innate immunity negative regulator PGRP-LF , is specifically expressed in non-immune tissues and plays an essential role during development , in absence of any infection . Lack of PGRP-LF function in these tissues inhibits apoptosis leading to incomplete genitalia rotation and tergite malformations . We show that such apoptosis inhibition results from the over expression of the negative regulator of apoptosis Diap1 specifically in PGRP-LF expressing cells . Our data highlight that proper negative regulation of immune signaling pathway in non-immune tissues is contributing to normal development and illustrate the growing evidence of the dual role of immune signaling pathway contribution to both immunity and in development processes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"skin",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"integumentary",
"system",
"cell",
"processes",
"developmental",
"biology",
"respiratory",
"system",
"pupae",
"epidermis",
"trachea",
"lipids",
"fats",
"gene",
"expression",
"life",
"cycles",
"biochemistry",
"anatomy",
"cell",
"biology",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"larvae",
"genital",
"anatomy"
] |
2017
|
Inhibition of a NF-κB/Diap1 Pathway by PGRP-LF Is Required for Proper Apoptosis during Drosophila Development
|
A vaccine against schistosomiasis would have a great impact in disease elimination . Sm29 and Sm22 . 6 are two parasite tegument proteins which represent promising antigens to compose a vaccine . These antigens have been associated with resistance to infection and reinfection in individuals living in endemic area for the disease and induced partial protection when evaluated in immunization trials using naïve mice . In this study we evaluated rSm29 and rSm22 . 6 ability to induce protection in Balb/c mice that had been previously infected with S . mansoni and further treated with Praziquantel . Our results demonstrate that three doses of the vaccine containing rSm29 were necessary to elicit significant protection ( 26%–48% ) . Immunization of mice with rSm29 induced a significant production of IL-2 , IFN-γ , IL-17 , IL-4; significant production of specific antibodies; increased percentage of CD4+ central memory cells in comparison with infected and treated saline group and increased percentage of CD4+ effector memory cells in comparison with naïve Balb/c mice immunized with rSm29 . On the other hand , although immunization with Sm22 . 6 induced a robust immune response , it failed to induce protection . Our results demonstrate that rSm29 retains its ability to induce protection in previously infected animals , reinforcing its potential as a vaccine candidate .
The development of a vaccine against schistosomiasis together with chemotherapy would have a great impact in the disease control and elimination . The ability of the parasite to evade the host immune system and its complex life cycle make the development of a vaccine against schistososmiasis a difficult task to achieve . But the presence of individuals naturally resistant to Schistosoma mansoni infection in endemic areas [1] , the evidence of acquired resistance by constant infection and treatments over the time [2] , and the high levels of protection induced by vaccination with irradiated cercariae suggest that developing a vaccine against the parasite is a feasible goal [3] . Many parasite antigens , especially the ones expressed on the parasite surface , have been studied as potential candidates for vaccine development [4 , 5] . Pre-clinical trials in the murine model represent an important step in anti-schistosomiasis vaccine development , and most of the antigens described as good candidates to be used in a vaccine formulation were evaluated in immunization protocols using naïve mice ( Sm14 , GST , Smp-80 , TSP-2 , Sm29 , Sm22 . 6 ) . Among the S . mansoni antigens , Sm29 and Sm22 . 6 are promising candidates . Sm22 . 6 or SmTAL-1 is a member of the Schistosoma mansoni Tegument-Allergen-Like ( TAL ) family [6] . Increased levels of IgE against Sm22 . 6 have been associated to resistance to reinfection in individuals living in endemic areas for schistosomiasis [6 , 7 , 8] . Also mice immunization with the recombinant form of Sm22 . 6 induced a significant decrease in parasite burden associated with increased levels of antibodies and mixed Type 1/Type 2 immune response [9] . Sm29 is a GPI-anchored parasite protein with unknown function [10] . High levels of IgG1 and IgG3 against the recombinant form of Sm29 were detected in individuals resistant to infection or reinfection [11] . Mice immunization with rSm29 elicited a significant antibody production and a Type 1 immune response , as well as a reduced parasite burden and pathology [12] . Besides the great results observed in pre-clinical trials in naïve mice using rSm29 and rSm22 . 6 as antigen , these antigens were never evaluated in pre-clinical trials using animals that had been previously exposed to parasite antigens . These evaluations are extremely important to determine antigen potential as a vaccine candidate , since the target population for the vaccine suffers several infections throughout life and is sensitized by parasite antigens in utero , which promotes a predominant Type 2 immunological profile [13 , 14 , 15] . In this context , we evaluated the ability of Sm22 . 6 and Sm29 recombinant proteins to induce protection with an immunization protocol using Balb/c mice previously infected with the LE strain of S . mansoni and treated with Praziquantel . We demonstrated that immunization of mice previously exposed to S . mansoni infection with rSm29 increased the levels of antibodies , IL-2 , IFN-γ , IL-17 and IL-4 production and the percentage of CD4+ central memory T cells , and elicited a protection level ranging from 26% to 48% , while immunization with rSm22 . 6 , despite inducing a robust immune response , failed to reduce worm burden .
Balb/c female mice aged 6–8 weeks were obtained from the Centro de Pesquisas René Rachou ( CPqRR ) —FIOCRUZ ( Fundação Oswaldo Cruz ) animal facility . Schistosoma mansoni cercariae ( Sambon , 1907 ) , LE strain , were maintained routinely on Biomphalaria glabrata ( Say , 1818 ) snails at CPqRR and were obtained by exposing infected snails to light for 1–2 hours to induce shedding . All the protocols involving animal use in this study were licensed by the Ethics Committee of Animal Use ( CEUA ) of FIOCRUZ , under license number LW12/12 . The recombinant Sm22 . 6 ( rSm22 . 6 ) and Sm29 ( rSm29 ) were produced and purified as previously described [9 , 11] . rSm22 . 6 and rSm29 concentrations were measured by BCA Protein Assay Kit ( Thermo Scientific Pierce , Rockford , IL , USA ) . Mice were infected through percutaneous exposure of shaved abdominal skin for 1h in water containing approximately 30 cercariae , as previously described [16] . After forty-five days , mice were treated with two oral doses of 800mg/Kg Praziquantel , with an interval of five days between doses to ensure that all parasites were killed . Fifteen days after treatment , mice were separated into four groups ( IT/rSm22 . 6; IT/Saline ( rSm22 . 6 ) ; IT/rSm29; IT/Saline ( rSm29 ) of ten animals each . Additionally Balb/c naïve mice were used in immunization protocols to evaluate the protection triggered by rSm22 . 6 and rSm29 in this mouse strain . Immunization protocol consisted in three doses of the vaccine injected subcutaneously in the nape of the neck with rSm22 . 6 ( 25μg/animal ) ; rSm29 ( 25μg/animal ) or saline plus Freund’s adjuvant . Mice received immunization doses in a fifteen-day interval regimen . In the first dose , mice were immunized with Complete Freund’s Adjuvant ( CFA ) and in the subsequent boosters , Incomplete Freund’s Adjuvant ( IFA ) was used . Thirty days after the last booster , animals were challenged through percutaneous infection with 100 cercariae ( IT/rSm29; IT/Saline ( rSm29 ) ; rSm29; Saline ( rSm29 ) groups ) or 50 cercariae ( IT/rSm22 . 6; IT/Saline ( rSm22 . 6 ) ; rSm22 . 6; Saline ( rSm22 . 6 ) groups ) . The number of cercariae used in challenge infections was similar to the number of those used in pre-clinical trials using rSm22 . 6 or rSm29 immunization in C57BL/6 naïve mice [9 , 12] ( Fig . 1 ) . Fifty days after challenge infection , adult worms were perfused by portal veins as described by Pellegrino and Siqueira [17] . Protection level was calculated comparing the number of worms recovered from the immunized group ( IT/rSm22 . 6 , IT/rSm29 , rSm22 . 6 or rSm29 ) with the number of worms recovered in the saline control groups ( IT/Saline ( rSm29 ) , IT/Saline ( rSm22 . 6 ) , Saline ( rSm22 . 6 ) or Saline ( rSm29 ) ) , using the formula below: Level Protection: Burden recovered from control group - Burden recovered from experimental group Burden recovered from control group ×100 Following perfusion , the intestine from infected , treated and vaccinated animals ( IT/rSm22 . 6 or IT/rSm29 ) and their control group ( inoculated with saline ) were removed , weighed and digested with 10% KOH overnight at room temperature . The eggs were obtained by centrifugation at 900 x g for 10 min and were resuspended in 1 mL of saline . The number of eggs present in intestine was determined using a light microscope Liver sections from IT/rSm29 immunized mice and from its control group ( IT/Saline ) were collected following perfusion . The liver sections removed from the central part of the lateral lobe were fixed in 10% buffered formaldehyde in phosphate buffered saline ( PBS ) . Histological sections were stained with Haematoxylin and Eosin ( HE ) . To determine granuloma area , approximately 100 granulomas from each group ( 10 granulomas/animal ) with a single well-defined egg and at exudative stage were randomly chosen at 10x objective lens through an AxioCam microcamera ( Carl Zeiss , Germany ) . Using the AxionVision 4 . 8 image analysis software ( Carl Zeiss MicroImaging GmbH , Germany ) , the total area of the granulomas were measured , and the results were expressed in square micrometers ( μm2 ) . Serum from mice of each of the different groups were obtained forty five days after infection , fifteen days after treatment and fifteen days after each immunization dose in individual basis . Sera were used to determine the production of IgG , IgG1 , IgG2a and IgE specific antibodies by ELISA . Briefly , MaxiSorp 96-well microtiter plates ( Nunc ) were coated with rSm22 . 6 or rSm29 at a concentration of 5μg/mL ( IgG , IgG1 and IgG2a ) , or 1 μg/mL ( IgE ) in carbonate-bicarbonate buffer , pH 9 . 6 , over night at 4°C . Then , the plates were blocked with 300μL/well phosphate-buffered saline , pH 7 . 2 , with 0 . 05% Tween-20 ( PBST ) plus 3% FBS ( fetal bovine serum , GIBCO , USA ) , for 2 hours at room temperature ( IgG , IgG1 and IgG2a ) , or with PBST plus 3% nonfat dry milk over night at 4°C ( IgE ) . One hundred microliters/well of pool of sera samples from each group was submitted to serial dilution beginning at 1:20 and ending in 1:1 . 310 . 720 , in duplicate , to determine antibody titers and to standardize the dilution of sera used in individual basis . Analyzes of antibody levels in individual basis were performed using one hundred microliters/well of each serum sample , diluted 1:1 . 000 ( IgG and IgG1—rSm29 tests ) , 1:100 ( IgG2a—rSm29 test ) , 1:600 ( IgG—rSm22 . 6 test ) , 1: 1 . 000 ( IgG1—rSm22 . 6 test ) , 1:400 ( IgG2a—rSm22 . 6 trial ) or 1:40 ( IgE—rSm22 . 6 and rSm29 test ) . Finally , the plates were incubated with peroxidase-conjugated anti-mouse IgG , IgG1 and IgG2a ( Southern Biotech , USA ) , diluted 1:10 . 000 , 1:10 . 000 and 1:8 . 000 , respectively , for 1h at room temperature ( RT ) . For IgE measurement , an anti-mouse biotin IgE ( BD Pharmingen ) diluted 1:250 was used and , after incubation ( 1h at RT ) , an avidin conjugated to peroxidase ( 1:250 ) was added for 30 minutes . Color reaction was developed by TMB incubation ( Microwell Peroxidase Substrate System ) and stopped with 5% sulfuric acid . The plates were read at 450 nm in an ELISA plate reader ( Bio-Rad , USA ) . Spleens from IT/Saline , IT/rSm29 or IT/rSm22 . 6 immunized animals were obtained 7 days after the last immunization dose ( Fig . 1 ) . As a control for the experiment , spleens from naïve Balb/c mice immunized with rSm22 . 6 or rSm29 or from non-immunized infected and treated mice were also obtained . Red cells were lysed with ACK , and spleen cells were washed twice with apyrogenic saline and adjusted to 1x106 cells/well . Cells were cultured in 5% CO2 at 37°C in the presence of medium , Concanavalin ( 5μg/mL ) , rSm29 or rSm22 . 6 ( 25μg/mL ) . Culture supernatants were collected 24 or 72 hours post stimulation for IL-4 ( 24h ) , IL-6 ( 24h ) , IL-10 ( 72h ) , IL-17 ( 72h ) , IFN-γ ( 72h ) and TNF-α ( 24h ) evaluation . Cytokine measurement was performed using Cytometic Bead Array Kit , anti-mouse CBA Th1/Th2/Th17 Kit ( BD Pharmingen , USA ) . CBA was performed according to the manufacturer’s protocol . Beads were acquired in FACScalibur flow cytometer ( BD , USA ) and data were analyzed using FCAP Array Software ( Becton Dickinson ) . For ex vivo analyses , spleen cells were adjusted to 5x106 cells/well and incubated for 15 min at 4°C with anti-mouse CD16/CD32 mAbs ( Fc-Block—BD-Pharmingen ) to block Fc gamma receptors . After a wash step with PBS ( Sigma ) , cells were incubated for 15 min at 4°C with antibodies against surface molecules: anti-CD4 ( BD-Pharmingen , clone GK1 . 5 ) , anti-CD8 ( BD-Pharmingen , clone 53–6 . 7 ) , anti-F4/80 ( eBioscience , clone BM8 ) and anti-IgG ( eBioscience , clone eBio299Arm ) conjugated to FITC; anti-CD69 ( eBioscience , clone H1 . 2F3 ) , anti-CD86 ( Accurate Chemical and Scientific Corporation , Westbury , NY , USA ) and anti-CD4 ( BD-Pharmingen , clone GK1 . 5 ) conjugated to PE; anti-CD27 ( eBioscience , clone LG . 7F9 ) and anti-CD25 ( BD-Pharmingen , clone 7D4 ) conjugated to biotin; anti-CD19 ( BD-Pharmingen , clone 1D3 ) and anti-CD127 ( BD-Pharmingen , clone A7R34 ) conjugated to PECy7; anti-CD44 ( BioLegend , clone IM7 ) conjugated to Pacific Blue and anti-CD62L ( BD-Pharmingen , clone MEL-14 ) conjugated to Alexa 700 . After that , cells were washed with PBS ( 0 . 15M ) , BSA ( 0 . 5% ) and NaN3 ( 2mM ) and incubated for 20 min with streptavidin-APC ( 1:200 ) at 4°C . Subsequently , cells were washed and fixed using 2% formaldehyde solution and were acquired using LSRFortessa ( Becton Dickinson , San Jose , CA ) . Data analysis was performed using FlowJO software ( TreeStar , Ashland ) . To assess the expression of Sm22 . 6 and Sm29 on parasite surface , skin-stage schistosomula were incubated with sera from Balb/c naïve mice immunized with three doses of rSm22 . 6 or rSm29 in Freunds’ adjuvant . Cercariae were mechanically transformed into skin-stage schistosomula as previously described [18] , with some modifications . Briefly , cercariae were incubated on ice for 30 min , centrifuged ( 1800g/3 min/4°C ) , and suspended in cold Glasgow medium ( Sigma-Aldrich , St . Louis , MO , USA ) plus 1% penicillin/streptomycin and 10% FBS . Cercariae tails were sheared off by vortexing for 2 min in high speed and were removed through several washing steps with Glasgow medium . The schistosomula were cultured for 3 hours at 37°C in medium . Later , the schistosomula were incubated with immune sera and FITC-conjugated anti-mouse IgG , as described [19] . Briefly , schistosomula were washed three times with DMEM medium by centrifugation at 1800g/5 min . Subsequently , the parasites were fixed with PBS plus 1% formaldehyde for 1 hour at 4°C . After this , schistosomula were washed three times with PBS and incubated in RPMI for 30 min , and in PBS + 1% BSA for additional 30 min . After another washing step , schistosomula were incubated with sera from Sm22 . 6 and Sm29 immunized mice diluted 1:50 in PBS overnight at 4°C . Afterwards , schistosomula were washed with PBS + 1% BSA ( 3x ) , incubated in RPMI for 30 min , washed with PBS ( 3x ) and incubated in PBS + 1% BSA for 30 min . Finally , schistosomula were incubated with an anti-IgG FITC-conjugated antibody for 2 hours in the dark , and antibodies binding to schistosomula surface were evaluated by fluorescence microscopy . The fluorescence intensity in the schistosomula tegument was measure using ImageJ software and the fluorescence analysis was determinate as described before [20] . Data normality was tested using the D’Agostino-Pearson omnibus test . Statistical analyses were performed using the Mann-Whitney nonparametric test for cytokine measurement and ex vivo analyses or Student’s t-test for parasitological analyses and antibodies measurement using the software package GraphPad Prism 5 . 0 ( Graph-Pad Software , San Diego , CA , USA ) .
The level of protection induced by rSm22 . 6 or rSm29 immunization in Balb/c mice , previously sensitized by S . mansoni infection and Praziquantel treatment , was evaluated 50 days after challenge infection with 50 ( IT/rSm22 . 6 trials ) or 100 ( IT/rSm29 trials ) S . mansoni cercariae . In order to establish the number of immunization doses required to induce a protective immunity , parasite burden after challenge infection was evaluated in mice immunized with one , two or three doses of vaccine formulations . No significant reduction in worm burden was observed in mice immunized with rSm22 . 6 ( IT/rSm22 . 6 ) compared to the control group regardless of the number of vaccination doses received ( Table 1 ) . In addition , IT/rSm22 . 6 group also showed no significant reduction in eggs trapped in intestine ( Table 1 ) . On the other hand , mice immunization with 3 doses of rSm29 ( IT/rSm29 group ) induced a significant reduction ( 26–48% ) in worm burden in comparison to saline control group ( Table 2 ) . A significant decrease in the numbers of eggs recovered from the intestine of the IT/rSm29 immunized group was observed in comparison with saline group ( Table 2 ) , reflecting the reduction in worm burden in this group . Despite this , the immunization with rSm29 in previously infected and treated mice failure to reduce granuloma area ( Fig . 2 ) . Interestingly , in naïve Balb/c mice neither immunization with rSm22 . 6 nor immunization with rSm29 induced reduction in parasite burden ( Table 3 ) . A significantly higher production of specific anti-rSm22 . 6 IgG , IgG1 , IgG2a and IgE was observed in IT/rSm22 . 6 immunized mice compared to control group ( IT/Saline ) in all of the time points evaluated and also compared to infected and infected/treated mice ( Fig . 3 ) . Furthermore , in the IT/rSm22 . 6 immunized group , the levels of specific IgG , IgG1 and IgG2a were significantly increased by the second and the third doses of the vaccine ( Fig . 3 ) . In IT/rSm29 immunized mice , significantly higher production of specific IgG and IgG1 in comparison to control group was observed in all of the time points evaluated ( Fig . 3 ) , while a significantly higher production of specific IgG2a and IgE in comparison with control group was only detected after the second immunization dose ( Fig . 3 ) . Moreover , in IT/rSm29 immunized group , a significant increase in IgG and IgG1 production was observed after each booster ( Fig . 3 ) , while a significant increase in specific anti-rSm29 IgG2a levels was observed after the third immunization dose in comparison with the first and the second doses of the vaccine ( Fig . 3 ) . The levels of anti-rSm29 IgE , in turn , increased only after the third immunization dose , compared to the first dose ( Fig . 3 ) . Also , a significant increase in anti-rSm29 IgG1 and IgE levels was observed in infected/treated mice compared to infected animals . Immunization with rSm29 significantly increased the levels of IgG , IgG1 , IgG2a and IgE in comparison with the levels detected in infected or infected/treated mice . The titers of specific Sm22 . 6 IgG did not differ between naïve immunized mice and infected/treated immunized mice ( Table 4 ) . But the titer of specific Sm22 . 6 IgG2a and IgG1 antibodies was lower in mice immunized after a previous infection whereas IgE titer was higher in these animals ( Table 4 ) . A higher titer of Sm22 . 6-specific IgG , IgG1 , IgG2a and IgE antibodies was observed in infected/treated immunized mice compared to infected/treated Saline inoculated animals ( Table 4 ) . In rSm29 immunized mice , the titers of Sm29-specific IgG2a did not differ between naïve immunized mice and infected/treated immunized mice ( Table 4 ) . A lower titer of specific IgE and IgG1 was observed in mice that had been previously infected and treated , while higher IgG titers were observed in these animals in comparison with the naïve immunized mice ( Table 4 ) . As in rSm22 . 6 immunized mice , a higher titer of IgG , IgG1 , IgG2a and IgE antibodies was observed in infected/treated immunized group compared to infected/treated Saline group ( Table 4 ) . To evaluate whether the antibodies produced in response to mice immunization with the recombinant form of Sm22 . 6 and Sm29 were able to recognize the native proteins in the parasite surface , an immunofluorescence staining assay was performed in newly-transformed schistosomula . Fig . 4A shows a representative picture of the results observed in immunofluorescence staining and demonstrates that the antibodies raised against the recombinant form of Sm22 . 6 or Sm29 recognize the native protein expressed on the parasite surface , since significant fluorescence measurement is detected in the presence of sera from rSm22 . 6 and rSm29 immunized mice compared to sera from saline inoculated animals ( Fig . 4B ) . Nonspecific recognition was not observed on parasites incubated with FITC conjugated anti-mouse IgG antibody ( Fig . 4B ) . To determine the cytokine profile induced by immunization , the levels of IL-2 , IFN-γ , TNF-α , IL-4 , IL-6 , IL-10 and IL-17 in the supernatant of spleen cells culture wer measured . In the group of mice immunized with three doses of rSm22 . 6 ( IT/Sm22 . 6 ) , rSm22 . 6 in vitro stimulation induced a significant production of IL-2 , IFN-γ , TNF-α , IL-4 , IL-10 and IL-17 in comparison to saline group ( Fig . 5 ) . Differences in the production of IL-2 , TNF-α and IL-4 were also observed between non-immunized mice that had been infected with S . mansoni and treated with Praziquantel and Sm22 . 6 immunized mice , with the higher production of those cytokines being observed in the immunized groups ( Fig . 5 ) . In the animals immunized with three doses of rSm29 ( IT/Sm29 ) in vitro re-stimulation induced significantly higher production of IL-2 , IFN- γ , IL-17 and IL-4 in comparison to saline control group and significantly higher IL-4 production in comparison to non immunized infected/treated animals ( Fig . 5 ) . The cytokine profile induced by rSm22 . 6 or rSm29 immunization in mice previously exposed to parasite antigen was similar to the one observed in naïve immunized mice , except for the IL-10 production , which was reduced in rSm29 immunized mice that had been previously infected and treated ( Fig . 5 ) . The expression of CD25 , CD69 and CD86 activation markers was evaluated on the surface of lymphocytes . A higher percentage of activated B cells ( CD19+CD86+ ) was observed in IT/Sm22 . 6 mice in comparison with IT/Saline group , while no difference was observed in TCD4+ and TCD8+ activation between IT/Saline group and IT/rSm22 . 6 or IT/rSm29 ( Fig . 6 ) . When compared to naïve Balb/c mice , immunization of infected/treated animals with rSm29 increase significantly the percentage of CD4+CD69+ and CD8+CD69+ cells , while immunization with rSm22 . 6 increase significantly the percentage of CD4+CD69+ cells ( Fig . 6 ) . Regarding memory cells , no differences in the percentage of CD8+ central memory cells ( CD44hiCD62hi CD127+ ) , T CD4+ or CD8+ effector memory cells ( CD44hiCD62loCD127+ ) and memory B cells ( CD19+CD27+ ) was observed between IT/rSm29 or IT/rSm22 . 6 and IT/Saline group ( Fig . 7 ) . A higher percentage of central memory TCD4+ lymphocytes were observed in the IT/rSm29 immunized group compared to control group ( Fig . 7 ) . Compared to naïve Balb/c mice , immunization with rSm29 in infected and treated animals induced significant higher percentage of CD4+ effector memory cells ( Fig . 7 ) . In contrast naïve Balb/c mice immunized with rSm22 . 6 presented a higher percentage of CD8+ central memory and memory B cells then IT/rSm22 . 6 group ( Fig . 7 ) . Furthermore , regarding monocytes activation status ( F4/80+CD86+ ) , no difference between immunized groups ( rSm22 . 6 or rSm29 ) and control group was observed ( S1 Fig . ) .
Currently , many efforts are focused on the schistosomiasis vaccine development , which will effectively contribute to disease control and eradication . Many vaccine candidates have been indentified and tested in pre-clinical trials with promising results . For instance , the recombinant proteins Sm22 . 6 or Sm29 induced significant worm burden reduction in C57BL/6 immunized mice [9 , 12 , 21] . In the case of schistosomiasis , the population most affected by the disease , and thus , the target to an anti-schistosomiasis vaccine , is represented by the residents of endemic areas , who have persistent contact with the parasite through constant infections/treatments throughout life responding differently to antigen stimulation if compared to an individual not sensitized previously by parasite antigens . Therefore it would be interesting , before moving forward to clinical trials , to evaluate promising vaccine candidates using experimental models that have already had previous contact with the parasite , approximating the vaccine research to the endemic area reality . Herein we performed an immunization protocol using Balb/c mice previously infected with 30 S . mansoni cercariae and treated with Praziquantel . Balb/c strain was the strain of choice in our studies since its Th2 genetic background resembles the immunological profile identified in individuals living in endemic areas for schistosomiasis [13 , 14 , 15 , 22] . Our results demonstrate that three doses of the vaccine containing rSm29 were necessary to elicit significant protection in mice previously exposed to parasite antigens , reducing the number of the worms recovered . This reduction reflected in significant decrease in the numbers of eggs recovered from the intestine . Although a reduction in parasite burden was observed in our study , immunization had no effect in liver pathology , thus this vaccine formulation would have an impact on disease transmission rather than in individual pathology . Immunization with Sm22 . 6 , on the other hand , failed to induce protection even after three doses of the vaccine . Since immunization of naïve Balb/c mice with rSm29 or rSm22 . 6 has never been evaluated , we also assessed the ability of both antigens to induce protection in this mice strain , interestingly neither rSm22 . 6 nor rSm29 was able to induce significant reduction in worm burden . Some studies have demonstrated the importance of antibodies in the protection induced by immunization . These studies demonstrate that antibodies are involved in parasite elimination [19 , 23 , 24] . Our results show a vigorous humoral immune response , with significant levels of IgG , IgG1 and IgG2a , and IgE in infected/treated Balb/c mice immunized with rSm22 . 6 or rSm29 plus Freund’s adjuvant . The humoral response induced by rSm22 . 6 immunization was observed after the first dose received . Our results corroborate with previous studies using rSm22 . 6 in vaccine formulations with Freund’s or Alum in naïve C57BL/6 mice , which showed a significant IgG production from the first immunization dose [9 , 21] . However , as in mice immunized with rSm22 . 6 + Alum [21] , we did not observe a significant reduction in parasite burden , suggesting that the IgG immunoglobulin induced by rSm22 . 6 immunization did not play a key role in worm elimination in these animals . Many studies have demonstrated the importance of immunoglobulin IgE in the parasite elimination and in the resistance to reinfection in individuals living in endemic areas [7 , 25 , 26 , 2 , 15 , 6] . An important mechanism involved in parasite elimination is mediated by IgE and eosinophils: in the presence of IgE , eosinophils can kill S . mansoni larvae through the antibody-dependent cytotoxicity ( ADCC ) process [25 , 27] . In this context , some studies demonstrate that individuals who present greater resistance to reinfection exhibit elevated IgE levels , associated with low levels of IgG4 against parasite antigens [2 , 6] . Indeed , many of these studies demonstrated that this resistance is mainly related to the response against the Sm22 . 6 ( SmTAL1 ) protein [28 , 15 , 29 , 30 , 31 , 32] . These studies revealed that , after treatment , individuals presented higher anti-Sm22 . 6 IgE levels , probably due to increased Sm22 . 6 exposure to the host immune system after parasite death . Indeed a higher titer of Sm22 . 6 specific IgE antibody was observed in infected and treated Sm22 . 6 immunized animals compared to rSm22 . 6 naïve immunized mice . However vaccine booster did not increase the levels of IgE against rSm22 . 6 in IT/rSm22 . 6 group . If IgE against Sm22 . 6 is a key factor associated to resistance , the absence of protection observed in Sm22 . 6 immunized mice can also be related to the lack of Fc-ɛ receptors expression in murine eosinophils [33] and the use of other models is necessary in order to confirm the inability of rSm22 . 6 to induce protection in an animal previously sensitized by parasite antigens . On the other hand , in Balb/c mice immunized with rSm29 , we observed that the production of IgG , IgG1 , IgG2a and IgE reaches the highest levels after the third immunization , which is consistent with the number of doses needed to confer protection against S . mansoni infection in animals previously exposed to parasite antigens . In individuals living in endemic areas for schistsosomiasis , the resistance to S . mansoni infection and reinfection is associated to increased production of IgG1 and IgG3 specific for Sm29 [11] . Regarding humoral profile the differences observed in antibody titers between rSm29 group , that did not develop a protective immunity and IT/rSm29 group , that developed a protective immunity , resides in an increased titer of IgG and decreased titer of IgG1 . These results suggest that other IgG isotypes may be associated with the protection induced by this vaccine formulation . Another concern regarding the antibodies produced after mice immunized with the recombinant proteins relied on their ability to recognize the native form of the protein on the parasite surface . In this context , the sera from Sm22 . 6 or Sm29 immunized mice recognized Sm22 . 6 and Sm29 expressed on the parasite surface , thus confirming the ability of the antibodies produced after immunization to bind the native form of the antigen . In mice immunized with rSm22 . 6 a significantly increased production of IL-2 , TNF-α , IFN-γ , IL-4 , IL-10 and IL-17 was observed , demonstrating that this formulation induced a mixed cytokine profile , similar to that observed in naïve Balb/c mice or naïve C57BL/6 immunized with the same protein [9] . Despite high levels of inflammatory cytokines , significant amounts of IL-10 were also produced . IL-10 is an important regulatory cytokine which regulates the immune response triggered by S . mansoni infection [34 , 35 , 36 , 37] . Significant amounts of IL-10 were also observed in naïve mice immunized with rSm22 . 6 plus Alum or without adjuvant , which also did not confer protection to challenge infection [21] . Increased production of IL-10 was also associated with absence of protection in mice immunized with schistosomula tegument in the absence of adjuvant [38] . If IL-10 production is responsible to the lack of protection observed in rSm22 . 6 immunized mice still need to be determined . In contrast to Cardoso and colleagues ( 2008 ) , that observed a significant production of IFN-γ , TNF-α and IL-10 cytokine in C57BL/6 naïve mice immunized with the recombinant protein , in our protocol Balb/c immunization with the rSm29 triggered a significant production of IL-2 , IFN-γ , IL-17 and IL-4 regardless of have been previous infected or not . Difference in cytokine profile between naïve Balb/c mice and infected/treated mice was only observed in IL-10 production that decreased in the infected/treated animals that presented a significant reduction in parasite burden . An important player on the protective immunity induced by vaccines is memory cells [39] . Our study demonstrate that IT/rSm29 group present a significant increase in the percentage of TCD4 central memory cells in comparison with IT/Saline group and also significant increase in the percentage of TCD4 effector cells in comparison with Balb/c naïve mice immunized with rSm29 . Since protection was observed only in the group of mice that had been previously infected with S . mansoni cercariae and treated with praziquantel , these cells might play an important role in parasite elimination . Although helminth infections have been associated with impaired protective vaccine-induced immunity against heterologous pathogens [40 , 41 , 42 , 43 , 44] , in our study a previous infection have contributed to vaccine-induced protection against schistososmiasis . Mice treatment before immunization schedule might have contributed to vaccination success either by enhancing the frequency of memory cells or decreasing IL-10 production in infected treated Sm29 immunized mice . Different type of immune response can be triggered by different vaccine formulations , but their ability to induce protection depends , besides other factors , on the role of the target antigen on pathogen development and survival . In the case of schistososmiasis , for instance , different types of immune mechanisms against different targets have been associated with protection . In vaccine formulations using Sm28 ( GST ) as antigen , the protective immunity is related to activation of a robust humoral response that blocks antigen enzymatic activity impacting on worm burden and female fecundity [45 , 46] . In the other hand , protective immunity induced by rSm14 is dependent on IFN-γ and TNF-α production [47] . In the case of Sm22 . 6 and Sm29 the knowledge about their functions is still extremely scarce . Lin and He ( 2006 ) have suggested that the Sm22 . 6 has an important role in the regulation of human coagulation [48] . For Sm29 , however , there is no information about its function . Functional analysis using RNA interference techniques ( RNAi ) can bring valuable information , allowing us to develop vaccine strategies that can effectively contribute to the parasite elimination [49] . In conclusion , ours results reinforces that the Sm29 is a promising vaccine candidate to compose an anti-schistosomiasis vaccine . Although Sm29 vaccination has failed to confer protection in Balb/c naïve mice , the fact that immunization of previously infected and treated animals trigger protection suggests that vaccination of individuals from endemic areas can be effective against S . mansoni infection , since it is a population sensitized by the parasite .
|
The development of a vaccine against schistosomiasis together with chemotherapy would have a great impact in the disease control and elimination . Sm29 and Sm22 . 6 are two promising antigens that have been associated with resistance to infection/reinfection in humans and also successfully induce protection in trials using C57BL/6 naïve mice . Despite the great results observed in C57BL/6 naïve mice , rSm29 and rSm22 . 6 ability to induce protection has never been assessed in mice previously exposed to the parasite antigens . In the case of schistosomiasis , this is an important assessment to be done , since the residents of endemic areas , the population mostly affected by the disease , are exposed to several infections through life . Here we evaluated these antigens in immunization trials using mice that had been submitted to a previous infection and treatment with Praziquantel . Both antigens induced a robust immune response triggering both cellular and humoral responses , but only rSm29 was able to induce a significant reduction on parasite burden and increased percentage of CD4+ memory cells . Our date reinforce Sm29 potential to compose an anti-schistosomiasis vaccine .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Sm29, but Not Sm22.6 Retains its Ability to Induce a Protective Immune Response in Mice Previously Exposed to a Schistosoma mansoni Infection
|
While it is widely held that an organism's genomic information should remain constant , several protein families are known to modify it . Members of the AID/APOBEC protein family can deaminate DNA . Similarly , members of the ADAR family can deaminate RNA . Characterizing the scope of these events is challenging . Here we use large genomic data sets , such as the two billion sequences in the NCBI Trace Archive , to look for clusters of mismatches of the same type , which are a hallmark of editing events caused by APOBEC3 and ADAR . We align 603 , 249 , 815 traces from the NCBI trace archive to their reference genomes . In clusters of mismatches of increasing size , at least one systematic sequencing error dominates the results ( G-to-A ) . It is still present in mismatches with 99% accuracy and only vanishes in mismatches at 99 . 99% accuracy or higher . The error appears to have entered into about 1% of the HapMap , possibly affecting other users that rely on this resource . Further investigation , using stringent quality thresholds , uncovers thousands of mismatch clusters with no apparent defects in their chromatograms . These traces provide the first reported candidates of endogenous DNA editing in human , further elucidating RNA editing in human and mouse and also revealing , for the first time , extensive RNA editing in Xenopus tropicalis . We show that the NCBI Trace Archive provides a valuable resource for the investigation of the phenomena of DNA and RNA editing , as well as setting the stage for a comprehensive mapping of editing events in large-scale genomic datasets .
With the exception of infrequent random somatic mutations , it is widely believed that the same genomic content should be fixed in an organism throughout its lifetime . This information will also serve as a template for exact RNA copies . Proteins that can modify genomic content , nevertheless , have been identified in humans and in many other organisms . RNA editing involves alteration of particular RNA nucleotides by specifically changing Adenosine ( A ) into Inosine ( I ) , which in turn is read as Guanosine ( G ) [1] . It is performed by the adenosine deaminase that acts on RNA ( ADAR ) family of deaminases [2]–[5] and this process has been implicated in several vital neurological functions [6] . A-to-I editing is known to target only RNA molecules [7] with numerous instances of editing events in the human transcriptome [8]–[12] . A different family of proteins , the AID/APOBEC family of deaminases , can edit both DNA and RNA nucleotides , specifically changing Cytosine ( C ) into Uracil ( U ) [13] . The first family member to be found and studied was the apolipoprotein B editing complex 1 ( APOBEC1 ) . This protein edits the apolipoprotein B ( ApoB ) RNA , which is involved in lipid transport [14] , [15] but APOBEC1 can also deaminate cytidine in DNA [16] . Additional members of the family were found to target DNA . Activation-induced deaminase ( AID ) was discovered to be vital for the antigen-driven diversification of immunoglobulin genes in the vertebrate adaptive immune system [17]–[19] and the APOBEC3s were shown to be involved in the restriction of retrovirus proliferation in primates [20] , [21] . For many years , the only known human endogenous target of the APOBEC protein family was the apoB RNA transcript . In this case , editing in position 6 , 666 by APOBEC1 leads to a stop codon and eventually results in two functionally distinct isoforms of apolipoprotein B ( ApoB ) [15] , [22] . This editing reaction is mediated by the APOBEC complementation factor ( ACF ) [23] , [24] which guides APOBEC1 to the target locus . Deamination of cytosines to uracils in DNA ( DNA editing ) by various APOBEC protein families is characterized , in many cases , by clusters of G-to-A mismatches between the reference genome and the edited sequence . These mismatches are the end product of deamination of “C” into “U” in the other DNA strand . Recently , it was found that APOBEC3G can serve as a potent inhibitor of a wide range of retroviruses , including endogenous retrotransposons . This protein introduces large numbers of C-to-U mutations in the minus-strand of the viral DNA , eventually leading to G-to-A mutations after plus-strand synthesis [25]–[29] . Also , it has been demonstrated that APOBEC3G is capable of editing the mouse IAP retrotransposon [30] . Little is known , however , about the frequency or localization of editing in vivo . Although editing of retrotransposons and their integration back into the genome is expected to be rare , very deep DNA sequencing can be used to identify these events . In this paper we report initial results of a novel bioinformatic approach for detection of endogenous RNA and DNA candidate sites in various organisms . We obtained 600 million sequence traces from the NCBI Trace archive . This data repository contains DNA sequence chromatograms ( traces ) from various large-scale capillary electrophoresis sequencing projects , base calls , and quality estimates . Next , we aligned these traces to their consensus reference genomes and searched for clusters of mismatches . Interestingly , we have found not only evidence of genuine RNA and DNA editing events but have also isolated a very common technical sequencing artifact that leads to such clusters .
Recently , DNA editing has been reported to be a powerful defense mechanism against the threat of genomic instability imposed by viruses and retrotransposons . However , the full magnitude of the phenomenon in vivo is not yet elucidated . We wanted to investigate whether our curated dataset of G-to-A mismatch clusters may actually include some examples of DNA editing . To test this assumption we looked at mismatch clusters in the mouse genome . We found that the total number of A-to-G and T-to-C mismatches was similar to the number of C-to-T and G-to-A mismatches ( 7 , 860 vs . 9 , 799 ) . However , in genomic regions of IAP ( intracisternal A-particle ) elements , for which a few members are still active , there was a significant dominance of the G-to-A / T-to-C mismatches ( 114 compared to 49 A-to-G / T-to-C ) ( p-value of 0 . 00018 , Fisher's Exact Test ) . This supports the idea that the origin of the mismatches is a result of editing by APOBEC after reverse transcription of the retrotransposons . An example of a DNA editing candidate , in a mouse retrotransposon , is given in Figure S1 . Active retrotransposons exist in human . For example , two edited HERVK elements have been recently discovered [34] . Thus , we applied our approach to human genomic sequences . Indeed we found evidence for DNA editing . We detected 247 events of G-to-A / C-to-T mismatch clusters versus 129 A-to-G / T-to-C events ( while overall in the genome the ratio is 91 , 120 to 79 , 401 respectively ) ( p-value of 0 . 0000017 , Fisher's Exact Test ) . One such candidate of editing by APOBEC in human retrotransposon HERVL-A1 is shown in Figure 3 . An additional example for a probable editing event in a human retrotransposon is present in Figure 4 where clusters of G-to-A mismatches are found in the most active SINE family in human , AluY . All of these mismatches have high sequencing quality ( Phred 40 or greater ) . Moreover , previously it was demonstrated that APOBEC3 can inhibit retrotransposition of Alu [35] . The actual number of edited traces in the trace archive is most probably much higher than we have found , for several reasons: More than half of all traces were rejected with our alignment parameters , at least partially due to the fact that DNA editing tends to lead to hyper-mutation in its target sequences [31] . Furthermore , we expect that a significant number of traces from retrotransposons , which are known targets for the APOBEC in their cDNA stage , are too redundant to align uniquely . Indeed , we found that in many cases the second best alignment of a putatively edited trace almost qualified for the 97% cut-off criteria , meaning that the trace was close to being rejected for having multiple possible genomic alignments . Thus , future work should find ways to curate the data in a less stringent manner so that editing , in traces with multiple hits to the genome or that do not meet our identity cut-offs , can still be detected . This would foster the development of a more complete picture of the occurrence of DNA editing in mammalian genomes . RNA editing is a general term for the modification of RNA after it is transcribed from DNA . The most common modification in mammals is A-to-I editing by the ADAR protein family . As I ( Inosine ) is read as a G ( Guanosine ) after sequencing , this editing type manifests itself as an A-to-G substitutions after cDNA sequencing and alignment to the original genomic locus . Recently it was found that the human genome harbors large numbers of editing events that are located in clusters , mainly in Alu repeats [9] , [10] , [11] , [12] . The origin of mismatch clusters in some of our traces , therefore , can be the result of ADAR activity . A fraction of the human , mouse and Xenopus tropicalis sequences obtained from the trace archive are labeled as derived from RNA , rather than DNA . In total , after passing the stringent alignment criteria , 250K , 513K and 454K traces , respectively , of those genomes have RNA origin , thus A-to-G or T-to-C mismatches in these traces could be the result of RNA editing . No over-representation ( 38% of the total MM clusters ) of A-to-G or T-to-C clusters appear in the RNA trace set ( Figure 5A ) , but as demonstrated above , the vast majority of mismatches are probably derived from a sequencing artifact . To overcome this issue we filtered those RNA traces and generated a higher quality , enriched set which required 3 consecutive mismatches of any quality and two mismatches separated by at least 100bp of phred 40 or greater . When we consider our higher quality , editing enriched set ( See Figure 5B ) , we find , in human , over-representation of mismatches that can be the result of RNA editing ( A-to-G and T-to-C ) , a total of 79% of the mismatch clusters are now of this type ( p-value 1 . 5e-119; Fisher's Exact Test . ) These observations suggest that RNA editing is the cause of the mismatches in the higher quality RNA sets . Further evidence that the higher quality set is indeed a result of RNA editing comes from two additional observations . First , a significant under-representation of “G” immediately upstream to the editing sites which is in agreement with the known sequence motif of the ADAR proteins [36] . In the enriched , higher quality set there was a G upstream of the mismatch in only 7 . 85% ( 265 out of 3 , 374 ) of the cases versus 30 . 3% ( 41 , 661 out of 137 , 313 ) in the non-enriched set ( p-value 1 . 9e-143 ) [36] , [37] ( See Figure 6 ) . Second , most known editing events in human are located in Alu repeats and indeed 72% of the mismatches in the higher quality set are located in Alu repeats while Alu represents only about 10% of human DNA ( p-value of 1 . 7e-110 ) . Detection of RNA editing from short EST sequences has proven to be challenging , due to their relatively low sequence quality [38] and indeed , almost all A-to-I sites found until now were detected from alignment of a small set ( <200 , 000 ) of full length RNAs [9] , [10] , [11] , [39] . In the present work we used the human EST data deposited in the trace archive ( currently including 2M ESTs which are mostly derived from poly-A mRNA ) and found thousands of potential editing sites . Only 156 sites out of the 3374 sites in the higher quality , enriched set overlap with the known set of about 20 , 000 editing sites reported by alignment of RNA to the genome ( total of 3 , 218 new sites ) . This suggests that ESTs , after accounting for sequence quality , can serve as a rich source for RNA editing site predictions . Of the organisms we studied , only human , mouse and Xenopus tropicalis had significant numbers of RNA traces . If we use our enriched , higher quality set as a proxy for the total number of editing events , our data shows that in mouse , editing occurs at an estimated rate of 1 mismatch per 100 , 000 unique , expressed base-pairs . In human , in agreement with previous publications [11] , [39] , [40] , our figures show ten-fold higher frequency . A striking picture emerges in Xenopus tropicalis . A closely related species , Xenopus laevis , is a principal model organism for the study of RNA editing as ADAR activity was first described in Xenopus laevis oocytes [41] and recently , research on hyper edited sequences in Xenopus laevis lead to the suggestion that editing can down-regulate gene expression in trans [42] . Only one endogenous hyper editing target is known in Xenopus - basic fibroblast growth factor ( bFGF ) [43] , [44] . Using our approach for detection of RNA editing we have observed significant over-representation of A-to-I derived mismatch in Xenopus tropicalis . In the enriched set 83% of the mismatch clusters are of the A-to-G and T-to-C type , while these types contribute only 39% of the mismatch clusters in the non-enriched set ( p-value≪e-200 ) ( Figure 5 ) . This strongly suggests that the mismatches in the enriched set are caused by RNA editing . The Xenopus tropicalis genome has not been completed yet and the annotation is still partial . Thus , we cannot determine if the editing sites are located in one type or a small number of genomic repetitive regions . Interestingly , we found that 10001 out of the total 18161 mismatches in our editing-enriched , higher quality set occur in clusters of ten sites or more , larger than the common clusters detected in human RNAs sequences which have a typical size of less than 6 mismatches . By further examining a few mismatch clusters , we found that they tend to occur in palindromic regions that can form tight double stranded RNA . These structures are known to be required for ADAR editing ( See Figure 7 ) . As in human , we observed the ADAR signature of low abundance of “G” upstream of editing sites ( 5 . 8% for the higher quality enriched set versus 24% in the non-enriched set ) ( Figure 6 , Tables S1 , S2 , S3 , S4 . ) . A full list with genomic coordinates of RNA editing sites in human and Xenopus is given in Datasets S1 , S2 .
The NCBI trace archive serves as a repository of raw data for the assembly of consensus genomes . Recently , it was utilized for a different purpose in the search for structural variation in the human genome [45] . Here , we show that it can also be used in the search for DNA and RNA editing . In the future , sequencing results deposited in the NCBI short-read archive might shed more light on these phenomena . Shorter reads , however , will pose a more challenging analysis problem . Recently , we did an initial analysis of Illumina's human resequencing reads and the SOLiD reads from the same individual . These reads are available at the NCBI short-read archive and are the basis for the first individual African consensus genome [46] , [47] . Given the importance of read-length and quality scores on the outcome of our current work , the current SOLiD and Illumina reads represent interesting trade-offs for the detection of editing . While Illumina's current read lengths are generally longer than SOLiD , the latter has much higher per-base quality . Adapting the techniques presented here to this new data presents an interesting opportunity for future research . The availability of computational resources for carrying out our analyses was essential to this project , as large computational effort was needed , six terabytes of disk for intermediate data and more than five “node years” of CPU time . With further computational effort , combining existing data in the trace archive with next generation sequencing data sets from multiple sequencing platforms and chemistries , it should be possible to greatly improve genomic databases and eliminate the sequencing errors reported here . By using well-calibrated quality scores and selecting traces with clusters of consecutive mismatches , we are able to investigate the scope of RNA editing sites in human and other genomes . The application of this technique in the search for editing events will make many large EST datasets more accessible for other organisms where quality scores are available . Currently , only a very small number of organisms , with large sets of full length RNA sequences , have been the subject of large-scale editing studies . Using quality scores , many additional genomes can be surveyed for editing with the opportunity for new discoveries in this emerging field . As a demonstration of the value of using quality data for ESTs , we are able to find a large number of candidate RNA editing events in Xenopus tropicalis . This discovery makes X . tropicalis the non human organism with the largest number of known editing sites so far . Since Xenopus is already an important model organism for the research of RNA editing , this new data-set could help foster new discoveries in this field . Despite the identification of thousands of newly discovered RNA editing sites in the current work , it is reasonable to believe that the actual number of editing sites is still significantly under-estimated . Support for this assertion comes from the stringency of our parameters: including length of alignment , percentage of identity and exclusion of insertions or deletions . These choices most likely limited the subset of EST data that we analyzed . Refinement of these criteria could lead to more comprehensive detection of RNA editing levels and , due to the breadth of EST data , even permit the comparison of editing levels in different tissues and disease conditions . In this work we also found evidence for recent or active events of DNA editing . While the true scope of these phenomena must be explored in future work , our approach , including the use of strict alignment criteria and quality scores , has proved effective at finding many intriguing examples . Using different parameters , mainly lower cutoffs and relaxation of the requirement for unique alignments , more DNA editing sites could be detected in the trace archive . Careful investigation , most likely combined with next-generation sequencing experiments , will help unravel the mechanisms of retroelement defenses in a variety of organisms . Moreover , DNA editing is known not to be limited to retrotransposons and can take place in other genomic loci . The most recognized example is the AID protein , which is a member of the AID/APOBEC protein family , and targets single stranded DNA in the immunoglobulin locus in B-cells . Similar approaches to the ones used here provide an exciting opportunity to survey how leakage of DNA editing events , outside retroelements , or immunoglobulins could cause many simultaneous mutations in the genome , a process that can eventually lead to cancer .
We obtained all traces for 10 organisms ( 600M traces in total ) , in FASTA format , at the NCBI Trace Archive [48] ( http://www . ncbi . nlm . nih . gov/Traces/home/ , May 2008 ) and aligned them with their reference genomes obtained from the UCSC Genome Browser [49] . We did not attempt to filter the initial set of traces by type which would have required the combination of FASTA format sequences with auxiliary information that provides the trace type . Instead we used strict placement criteria , described further below , to obtain the initial dataset summarized in Table 1 . We inspected chromatograms for individual traces using the tools provided at the trace archive . We further downloaded SCF raw binary data from the archive , by hand , and analyzed them using Phred version “0 . 071220 . b” [32] . This Phred version can generate an alternate base call for every position in the trace . This results in two sets of sequences for any given trace . By aligning the two sequences from the same trace separately , and looking for a large alignment with a single base-pair offset , we can identify the sequencing error from Figure 1 . This might be the basis for an automated test to eliminate this particular sequencing error . We augmented the above data by downloading auxiliary information and quality scores for a subset of about 20 . 7 million traces which were , potentially , enriched for editing events . We used runs of three consecutive mismatches of the same type as the enrichment criteria . The number of high quality traces for each editing type ( G-to-A , C-to-T , A-to-G , and T-to-C ) - is listed in Table 2 . For all organisms , except for mosquito and fly , there are more than ten times the number of examples from these four types than the next most frequent type . Furthermore , we extracted the lowest quality subset of these traces enriched for editing to be used for comparison purposes . The number of traces of each editing type from this set , G-to-A , C-to-T , A-to-G , T-to-C , as well as the most frequent or next most frequent type , is listed in Table S5 . For Mouse , Human , and Xenopus tropicalis these tables also provide ( in brackets ) the number of traces that likely originated from RNA . The complete set of mismatches found in these two sets of traces is available to the community as two files , “all . c2 . t100 . q40+ . bed . gz” ( 5 . 95MB ) and “all . c2 . t100 . q0-9 . bed . gz” ( 122MB ) , respectively . The first set is included on the journal's web-site while the second file is available , on request , from the authors . The files contain: the genomic coordinate of the mismatch , the mismatch type , the position on the trace , the quality of the mismatch , the length of the run in which the mismatch was found , the sequencing center , the trace id , the organism , and the likely origin of the trace , DNA or RNA . In order to be counted , each trace must have at least two mismatches with phred 40 or greater that are separated by 100bp or more . Only mismatches with phred scores of 40 or greater are included in the high quality set ( see Figures S2 , S3 , S4 for more data ) . In the lower quality set , at least two mismatches with phred less than 10 separated by 100bp or more are required . Only mismatches with phred scores of less than 10 are included in the low quality set . For sequence alignment , we used MegaBlast [50] version 2 . 2 . 13 from NCBI . The parameters used were: -W60 ( a 60bp seed was selected as a good compromise between computational efficiency and sensitivity , given our requirement of high identity to the reference ) , -s 400 -p 97 ( at least 400bp with 97% identity ) -F F ( no filtering ) -G25 -E10 ( these gap and extension penalties preclude insertions and/or deletions in matches ) . In addition , only unique alignments matching the above criteria were retained . These parameters were chosen for simplicity of subsequent analysis and to reduce the already onerous computational requirements . Two computational clusters were used to perform the analysis . These clusters were built to assist in deploying data intensive web services [51] . In total , the clusters use a variety of older and newer hardware and consist of 96 nodes w/ ( predominantly ) 4×1 . 8GHZ Opteron cores , 4–16GB of RAM per node , and 0–3750GB disk per node . The workflows to generate the initial analysis of the data are written in Perl . The human analysis consumed 347 node days and 530GB of space which was reduced to 22GB of compressed data after parsing the MegaBlast output and discarding redundant matches . A summary of the traces and space/time used by the computation can be found in Table 1 . The startling amount of intermediate space required by the mouse analysis , greater than 4 . 2 terabytes , suggests that many traces in mouse did not place uniquely and consumed large amounts of space , even with our strict chosen cut-offs and using gzip compression on the output of MegaBlast .
|
Most biomedical , genomic research begins with the painstaking assembly of a “reference genome” for the organism of interest . Implicit in this process is an assumption that genomic information is constant throughout an organism . There are enzymes , however , that can change , or “edit , ” genomic information so that variations from the reference can exist within a single organism . In this work , we analyze the raw data used to assemble the reference genomes of ten organisms to discover evidence for editing . We found candidates for DNA and RNA editing as well as a sequencing error that has become incorporated into commonly used genomic resources . Our analysis demonstrates the utility of raw genomic data for the discovery of some editing events and sets the stage for further analysis as sequencing costs continue to decrease exponentially .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics/comparative",
"genomics",
"genetics",
"and",
"genomics/genome",
"projects",
"computational",
"biology/genomics",
"computer",
"science/information",
"technology",
"genetics",
"and",
"genomics/bioinformatics"
] |
2010
|
A Survey of Genomic Traces Reveals a Common Sequencing Error, RNA Editing, and DNA Editing
|
Rift Valley fever ( RVF ) is a viral zoonosis that primarily affects animals resulting in considerable economic losses due to death and abortions among infected livestock . RVF also affects humans with clinical symptoms ranging from an influenza-like illness to a hemorrhagic fever . Over the past years , RVF virus ( RVFV ) has caused severe outbreaks in livestock and humans throughout Africa and regions of the world previously regarded as free of the virus . This situation prompts the need to evaluate the diagnostic capacity and performance of laboratories worldwide . Diagnostic methods for RVFV detection include virus isolation , antigen and antibody detection methods , and nucleic acid amplification techniques . Molecular methods such as reverse-transcriptase polymerase chain reaction and other newly developed techniques allow for a rapid and accurate detection of RVFV . This study aims to assess the efficiency and accurateness of RVFV molecular diagnostic methods used by expert laboratories worldwide . Thirty expert laboratories from 16 countries received a panel of 14 samples which included RVFV preparations representing several genetic lineages , a specificity control and negative controls . In this study we present the results of the first international external quality assessment ( EQA ) for the molecular diagnosis of RVF . Optimal results were reported by 64% of the analyses , 21% of the analyses achieved acceptable results and 15% of the results revealed that there is need for improvement . Evenly good performances were achieved by specific protocols which can therefore be recommended as an accurate molecular protocol for the diagnosis of RVF . Other protocols showed uneven performances revealing the need for improved optimization and standardization of these protocols .
Rift Valley fever ( RVF ) is a mosquito-borne viral zoonosis that primarily affects animals but also has the capacity to infect humans . An epizootic of RVF is usually first indicated by a wave of unexplained abortions as infected pregnant livestock abort virtually 100% of fetuses . The disease is less fatal to humans as most human infections are asymptomatic and when clinical symptoms appear they are in majority influenza-like . Nevertheless , some cases may develop a severe RVF disease with variable clinical signs . More severe cases occur in 2% of the RVF cases and fall into three categories: liver necrosis with hemorrhaging , retinitis with visual impairment and meningoencephalitis [1] , [2] . The causative agent of RVF , the RVF virus ( RVFV ) , is a negative-stranded RNA virus , a member of the genus Phlebovirus of the Bunyaviridae family . The number of identified viral lineages of RVFV has increased from 3 in an early analysis [3] to 7 in a 2007 study [4] , and in the most recent report 15 distinct genetic groups were reported [5] . Phylogenetic analysis shows that the virus emerged in the mid-19th century , but it was first identified in 1930 during an outbreak of abortions and deaths among sheep in the Rift Valley region of Kenya . In 1977–78 , several millions of people were infected and more than 600 died during a severe epidemic in Egypt [6] . Since then , the geographical distribution of the virus has widely spread and now includes most countries of the African continent as well as Madagascar and the Arabian Peninsula . During the past five years , outbreaks have been reported in Kenya [7] , Somalia , Tanzania [8] , Sudan [9] , Mayotte [10] , Madagascar [11] , Swaziland , South Africa and Mauritania [12] , [13] Another important concern is the increasing number of human fatalities during the most recent outbreaks [14] . The emergence or re-emergence of RVFV activity is periodic and associated with exceptionally heavy rainfalls which allow massive breeding of flood-water Aedes mosquitoes with the capacity for transovarial transmission [15] and other competent vectors such as Anopheles and Culex species [9] . These mosquitoes may initiate outbreaks among livestock , particularly breeds of cattle and sheep . The virus can be transmitted to humans by mosquito bite or by contact with infected tissues of domestic and wildlife ruminants . The sudden onset of large numbers of abortions and fatalities in RVFV affected livestock , resulting in the virus spread to humans can greatly strain public health and veterinary infrastructures . Unavailability of effective antiviral drugs and commercial vaccines for human or animal use outside endemic countries , including the US and Europe , and the recent spread of RVFV beyond its usual boundaries has resulted in increased international demand for qualified diagnostic tools for a rapid and accurate diagnosis of RVF . Diagnostic methods for RVFV detection include virus isolation [16] , antigen [17] , [18] and antibody detection methods [19]–[21] and nucleic acid amplification techniques . Isolation procedures are expensive , time-consuming and require high biocontainment facilities . Serological methods such as antigen or antibody-detection enzyme immunoassays ( EIA ) require several samples and often lack sensitivity . Therefore , considerable efforts have been made to develop molecular methods which allow a rapid , accessible and accurate detection of RVFV . The use of direct diagnostic methods such as molecular methods , can detect the disease during the acute phase of the infection thus allowing efficient patient management , avoiding nosocomial cases and providing rapid outbreak response . Highly sensitive nucleic acid detection methods have been developed including polymerase chain reaction ( PCR ) assays such as reverse-transcriptase PCR ( RT-PCR ) [22] , real-time RT-PCR ( qRT-PCR ) [23]–[25] and more recently real-time reverse-transcription loop-mediated isothermal amplification ( RT-LAMP ) [26] and recombinase polymerase amplification assays ( RPA ) [27] . The performance of the different techniques applied for molecular diagnosis of RVFV may vary between laboratories . External quality assessment ( EQA ) studies to assess the quality of RVFV molecular diagnostics have not been performed until now . The EQA study allows the participating laboratories to monitor the quality of current diagnosis , identify possible weaknesses of particular diagnostic methods and evaluate their capacity for surveillance activities . Therefore the first EQA study for the molecular diagnosis of RVFV was organized by the European Network for Diagnostics of ‘Imported’ Viral Diseases ( ENIVD ) ( http://www . enivd . org ) in 2012 . Using the results of this study , the ENIVD can also provide support and advice to all laboratories performing RVFV molecular diagnosis .
A total of 33 laboratories involved in diagnostics of RVF infections were invited to participate in this study . Invitees were selected from the register of ENIVD members , national/regional reference laboratories for RVF or vector-borne diseases as well as on the basis of their contributions to the literature relevant to this topic . The participation to the study was open and free of charge and included publication of the results in a comparative and anonymous manner . This EQA was coordinated by the ENIVD following comparable procedures used during previous studies performed by the network [28] , [29] . A proficiency test panel of 14 samples was prepared which included inactivated and stable RVFV preparations generated from Vero E6 cell culture supernatants of different RVFV genetic lineages and origin . Viral cell supernatants were inactivated by heating for 1 h at 60°C and gamma irradiation ( 25 kilogray ) to assure their non-infectivity . A serum sample spiked with Toscana virus , another phlebovirus , was included as a specificity control as well as two negative controls . The RVFV positive samples selected for this EQA panel are detailed in Table 1 . Two dilutions of sample Tambul/Egypt/1994 and 5 dilutions of sample F057/Kenya/2007 were obtained by serial 10-fold dilutions and included in the panel for sensitivity testing . All virus material used for the preparation of the EQA panel was obtained from cell culture and not from clinical samples of infected patients . Therefore , there is no requirement for any ethical statement in this study . All samples were diluted with fresh thawed human plasma previously confirmed as negative for RVFV . Aliquots of 100 µl were number-coded , freeze dried for 24 h ( Christ , AlphaI-5 , Hanau , Germany ) and stored at 4°C until dispatched . Before dispatching the panels , 3 different sets of EQA samples were tested and validated by 2 expert laboratories . For validation , the samples were resuspended in 100 µl of water and the RNA extracted using the QIAamp viral RNA minikit ( Qiagen , Hilden , Germany ) . The number of RVFV genome copies present in these samples was determined by qRT-PCR . Panel samples were shipped by regular post at ambient temperature . We requested participant laboratories to resuspend the samples in 100 µl of water and to analyze the material as serum samples for nucleic acid detection of RVFV following their routine protocols . The EQA panels were distributed to participants with documentation including full instructions and an evaluation form to fill in their results . Participants were also asked to report information on the adopted protocol , the type of RVFV strain and the number of genome copies in each sample when possible as well as any problems encountered concerning the shipment or the packaging of the samples . To guarantee anonymous participation , an individual numerical identification code was assigned to the results reported by each laboratory . This number was followed by a letter ( a , b , c ) when distinguishable data sets of results based on different methods were sent . The results were scored in reflection of analytical sensitivity and specificity as in previous EQA studies performed by the ENIVD [29] , [30] . We assigned one point for correct positive or negative result whereas false-negative/-positive results were not scored . Equivocal or borderline results were not counted as molecular diagnostic methods should always provide a clear positive or negative result . Results were classified as:
We obtained from the invitees a response rate of 91% representing a total of 30 participating laboratories from 16 different countries ( 10 European , 2 African , 3 Middle-Eastern/Asian countries and one American country ) : CODA-CERVA , Department of Virology , Epizootic Diseases Section , Uccle , Belgium; ANSES , Virology Unit , Laboratory of Lyon , France; CIRAD , Department BIOS «Control of exotic and emerging diseases» , Montpellier , France; IRBA-IMTSSA , Virology Unit , Le Pharo , Marseille , France; BNI , National Reference Centre for Tropical Infectious Diseases , Hamburg , Germany; Bundeswehr Institute of Microbiology , Munich , Germany; Institute for Novel and Emerging Infectious Diseases Friedrich-Loeffler-Institut , Germany; Robert Koch Institute , Berlin , Germany; Institute of Virology , Georg-August University , Gottingen , Germany; Central Virology Laboratory , Ministry of Health , Public Health Laboratories Sheba Medical Center , Israel; Army Medical and Veterinary Research Center , Rome , Italy; Department of Infectious , Parasitic and Immune-Mediated Diseases , Istituto Superiore di Sanità , Rome , Italy; Padiglione Baglivi National Institute for Infectious Diseases “L . Spallanzani” , Rome , Italy; Department of Histology , Microbiology and Medical Biotechnologies , University of Padova , Italy; Center for Vectors and Infectious Diseases Research , National Institute of Health , Aguas de Moura , Portugal; King Fahd Medical Research Center , King Abdulaziz University , Saudi Arabia; Arboviruses and viral hemorrhagic fever Unit , Institut Pasteur de Dakar , Senegal; Defense Medical & Environmental Research Institute , DSO National Laboratories , Singapore; Institute of Microbiology and Immunology , Faculty of Medicine , University of Ljubljana , Slovenia; Onderstepoort Veterinary Institute , South Africa; Deltamune ( Pty ) Ltd , Centurion , Gauteng , South Africa; Special Viral Pathogens Laboratory , National Institute for Communicable Diseases , South Africa; Laboratory of Arboviruses and Imported Viral Diseases , National Center for Microbiology , Instituto de Salud Carlos III , Spain; National Institute for Agricultural Research and Experimentation ( INIA ) , Madrid , Spain; Viral Diseases Unit , CReSA , Barcelona , Spain; Swedish Institute for Infectious Disease Control , Sweden; Virology group , Spiez Laboratory , Switzerland; Laboratory of Virology , University Hospitals of Geneva , Switzerland; WHO Collaborative Centre for Virus Reference and Research ( Arboviruses & VHFs ) , Health Protection Agency , United Kingdom; Viral Special Pathogens Branch , Infectious Diseases , CDC , Atlanta , United States of America . A total of 39 datasets were received including 5 double sets from laboratories using 2 methods ( lab #6 , 7 , 21 , 27 and 28 ) and 2 triple sets from lab #5 and #14 . Methods used by the same laboratory could differ from the type of technique , the protocol used for a specific technique or the type of instrument used for a specific protocol . Performances varied among laboratories and scores ranged from 7 to the maximum value of 14 . Optimal results were reported by 64% ( n = 25 ) of the analyses; 21% ( n = 8 ) of the analyses achieved acceptable results due to the inability to detect one positive sample , and 15% ( n = 6 ) revealed several false negative and/or one or more false positive results indicating that there is still need for improvement ( Table 2 and 3 ) .
RVF reference laboratories responded keenly to this EQA study ( 91% response rate ) , including laboratories situated in RVFV endemic countries such as South Africa and Saudi Arabia . Nonetheless , there is still a need to encourage more laboratories situated in RVF-endemic areas to participate in quality assurance programs . In fact , the increasing amplitude of this disease in Africa necessitates the rapid recognition of RVF outbreaks and implementation of effective control measures in order to prevent uncontrolled and wider spread of the virus . Most of the laboratories ( 93% , 28 out of 30 ) reported the use of qRT-PCR techniques allowing a rapid detection as well as quantification of the virus genome . This confirms that the use of qRT-PCR has remarkably expanded although it requires expensive equipment . All datasets obtained by qRT-PCR only were scored with 13 or 14 points indicating an evenly high performance of all qRT-PCR procedures performed by the different laboratories . Protocols from Drosten et al , 2002 [24] , Bird et al . 2007 [25] , Weidmann et al 2008 [31] as well as all in-house qRT-PCR protocols ( dataset #6b , #10 and #24 ) have demonstrated the capacity of providing optimal performances indicating a good specificity and sensitivity for these techniques . The sets of results obtained by applying the qRT-PCR protocols of Mweango et al . 2012 [35] , Garcia et al . 2001 [23] and Busquets et al . 2010 [32] did not achieve optimal performances ( scores 13 , 11 and 13 respectively ) but these techniques are not sufficiently represented to conclude on their overall performances . Information on the viral load of RVFV in human samples can be very useful to monitor the progress of clinical manifestations and to study the pathogenesis of RVFV . Interestingly , not all laboratories employing qRT-PCR techniques have reported quantified results and most of them ( 64% ) reported the results as cycle thereshold ( Ct ) values and not the number of genome copies . This indicates that most laboratories do not resort to RVFV standards while performing qRT-PCR although such standards would allow them to quantify viral genome in each sample without performing any additional assay . Accordingly to the results of this EQA as well as previous EQA studies , there is still room for improvement concerning viral load determination [29] , [30] . The most widely used technique after qRT-PCR was nested RT-PCR with 5 laboratories which referred to 2 different protocols [22] , [34] . Nested RT-PCR performances varied greatly compared to qRT-PCR with scores ranging from 7 to 13 thus never reaching optimal performances . The dataset #14c obtained a score of 13 with the protocol of Sall et al . 2002 [22] because it could not detect the highest dilution of the RVFV-Egypt/1994 strain indicating a slightly low sensitivity just as observed for some of the qRT-PCR methods . Nevertheless other datasets referring to nested RT-PCR ( #9 and #22 ) also reported false positive results indicating a lack of specificity of these procedures with both nested RT-PCR protocols [22] , . It is interesting to notice the appearance of newly developed techniques which are suitable for rapid field diagnostics such as RT-LAMP developed in 2009 [26] and RPA technology developed in 2012 [27] . No general conclusion can be achieved concerning the performances of these two techniques as they both have been performed by only one laboratory . However RPA has shown optimal results for this EQA demonstrating equivalent sensitivity and specificity to the qRT-PCR techniques ( dataset #27b ) . On the other hand , RT-LAMP results indicated difficulties in detecting RVFV genome in the less concentrated samples of the panel ( sample #4 , #13 and #14 ) . These results suggest some limitations in test sensitivity . However , very high test sensitivity is not essential for field diagnostics in an outbreak situation where most diagnosed patients are in the acute phase of the disease and are expected to present a high viremia . Three laboratories have provided different sets of results which referred to the same technique and protocol but using different instruments ( datasets #5b/c , #14a/b and #28 a/b ) . These datasets provided all optimal results by using two different instruments except for dataset #14 which reported a slightly lower sensitivity using the SmartCycler System from Cepheid ( #14b , 13 points ) compared to the 7500 Real-Time PCR System from Applied Biosystems ( #14a , 14 points ) . However , this difference cannot be attributed with certainty to the use of a different instrument as result variability can also arise from a lack of repeatability of the procedure . Only a few participants provided complete or partial information regarding strain typing ( 13% , 4 out of 30 ) . However , correct results without strain or genetic lineage specification are satisfactory in the context of laboratory diagnosis . Nonetheless , RVFV strain typing is relevant for surveillance activities in order to monitor which strains are circulating in RVFV-endemic areas and what type of clinical manifestations are associated with these strains . Comparing the results of this EQA panel to previous EQA studies [29] , [30] , [36] , we observe a higher concordance in terms of performance within laboratories using the same type of diagnostic method . In fact , all qRT-PCR techniques demonstrated an overall good performance with scores ranging from 13 to 14 . On the other hand , nested RT-PCR methods have shown a common need for improvement in terms of test sensitivity and/or specificity . Nevertheless , variations in performance between laboratories using the same method were noted . The reason for such variations is difficult to establish but can be minimized by standardizing procedures , including controls and testing conditions . In order to ensure optimal performances for RVFV molecular diagnosis in expert laboratories , we recommend conducting EQA studies on a regular basis . Future EQA studies should include a wide range of RVFV isolates with limiting concentrations to assess as precisely as possible the diagnostic performances of various molecular protocols in different reference laboratories .
|
Rift Valley fever ( RVF ) is a zoonotic viral disease posing an increasing threat to animals and humans worldwide . Recent severe outbreaks of the disease in animal and human populations in endemic regions and outside the disease's traditional geographic boundaries necessitate the need for evaluating the diagnostic performance of RVF expert laboratories . Molecular methods are increasingly used for a rapid and accurate detection of viral nucleic acid . In this study we present the results of the first international external quality assessment ( EQA ) for the molecular diagnosis of RVF . Such EQA studies allow participating laboratories to monitor the quality and identify possible weaknesses of current diagnostic methods . Participants to this RVF EQA were 30 expert laboratories from 16 different countries worldwide . The study demonstrated that optimal results could be achieved by the majority of laboratories . Specific protocols showed evenly good performances and can therefore be recommended to all expert laboratories . However , other methods showed uneven performances suggesting the need for improved optimization and standardization of these protocols .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"applied",
"microbiology",
"emerging",
"infectious",
"diseases",
"virology",
"biology",
"microbiology",
"viral",
"disease",
"diagnosis"
] |
2013
|
International External Quality Assessment of Molecular Detection of Rift Valley Fever Virus
|
Despite free of charge biomedical treatment , the cost burden of Buruli ulcer disease ( Bu ) hospitalisation in Central Cameroon accounts for 25% of households' yearly earnings , surpassing the threshold of 10% , which is generally considered catastrophic for the household economy , and calling into question the sustainability of current Bu programmes . The high non-medical costs and productivity loss for Bu patients and their households make household involvement in the healing process unsustainable . 63% of households cease providing social and financial support for patients as a coping strategy , resulting in the patient's isolation at the hospital . Social isolation itself was cited by in-patients as the principal cause for abandonment of biomedical treatment . These findings demonstrate that further research and investment in Bu are urgently needed to evaluate new intervention strategies that are socially acceptable and appropriate in the local context .
Buruli ulcer disease ( henceforth , Bu ) is the third most common mycobacterial disease in humans , after tuberculosis and leprosy [1]; nonetheless , despite the dramatic increase in incidence rates in West Africa during the last decade , it remains largely neglected [2] . Its causative agent , Mycobacterium ulcerans , is an environmental mycobacterium endemic to restricted foci throughout the tropics and directly related to stagnant or slow-flowing water [3] . Inoculation of Mycobacterium ulcerans into the subcutaneous tissues likely occurs through penetrating skin trauma , though the mode of transmission is not entirely clear . The agent produces a potent toxin known as mycolactone which destroys cells in the subcutis leading to the development of large skin ulcers [4] . Major advances have been made in the management of the disease with the introduction of rational antibiotic therapy [4] , however , successful clinical outcomes require that patients initiate treatment promptly and adhere to treatment regimens [5] , [6] , [7] . Furthermore , because Bu often leads to irreversible physical disabilities , the disease takes a significant toll on affected patients and their households [7] . In Cameroon , Bu was first documented in 1969 in 47 cases , all originating from the well-circumscribed rural area of Ayos and Akonolinga in Central Cameroon . The endemic region was later identified as an area stretching along the Nyong River and some of its tributaries for approximately 100 km in length and 10–30 km in width with a population of 98 , 500 people [3] . A study by Noeske and others [3] conducted in 2001 identified an overall prevalence rate of 0 . 44% constituting active and inactive Bu in the surveyed area . The highest prevalence of active cases found in a particular settlement was 8% . Disease prevalence was higher in villages closer to the Nyong river . Bu occurs principally among impoverished rural people with limited geographic and economic access to health facilities . Recently , a survey carried out by Um Boock in 2004 [8] detected new foci of Bu outside of the previously established endemic region , particularly in other areas of the Central Province and in the provinces of the East and Southwest . At the national level 930 cases were detected [8] . Biomedical treatment Bu in the endemic region is provided at the Ayos and Akonolinga Hospitals which have specialised Bu programmes sponsored by Aide aux Lépreux Emmaus Suisse and Médicins Sans Frontières respectively . Both offer free of charge medical treatment and supplementary aid . These services include free of charge medication and in-patient treatment; free meals served once or twice a day ( depending on the institution ) ; complementary accommodations for in-patients and their caretakers for the duration of their stays; extra schooling ( at Ayos Hospital ) ; and , the free provision of basic materials for everyday needs such as soap and bandages , sheets , etc . However , the provision of these materials is often irregular as stock-outs are common , directly affecting patients' hospitalisation costs . The main objective of this article is to evaluate the economic and social impact of hospital treatment for Bu disease on the patient and the household in a setting where medical costs for hospital treatment and supplementary aid for everyday needs are subsidised by the local health care system and/or through foreign aid .
The present study was conducted in the above-mentioned endemic region of Ayos and Akonolinga . It was an a focused ethnographic study using both qualitative and quantitative research methods . Field work was conducted in both community and clinical settings for a period of four months , three of which were spent at the Ayos and Akonolinga hospitals and one in the selected endemic communities of Eyess , Edou , Ebanda and Ngoulemakong , all belonging to the catchment areas of the respective hospitals . Considering the focalised character of Bu infection rates , [7] restricted local and geographical units were selected for analysis . During field work the following techniques were used: ( i ) Participant observation . Participant observation , or the observation of people's behaviour in its natural setting , is a fundamental and often neglected part of qualitative research . This technique consists of participating in everyday activities , working in the native language ( in this case the official language and lingua franca , French ) and observing events in their everyday context . Gaining patients' and household members' confidence through participant observation methods increases the validity of ethnographic data . In this study , participant observation was particularly important in establishing costs since patients and household members are unlikely to accurately recall all costs associated with this long-term illness . Participant observation also enabled us to contrast reported adherence to treatment , expenditures , frequency of visits to hospitalised Bu patients , and other behaviours with actual observed patterns . ( ii ) In-depth interviews . During field work , in-depth interviews were carried out with Bu patients and their households , extended family members , local traditional healers , community leaders and key informants ( such as teachers , school directors , priests , etc ) . One or more in depth interviews were carried out with Bu patients and their households . All interviews were conducted by the authors , personally , and were not mediated by hospital , public health or other biomedical staff for fear that their presence would guide responses ( e . g . downplay negative opinions of current health provisions ) . ( iii ) Focus group discussions . Focus group discussions were held with medical staff at the Ayos and at the Akonolinga Hospitals and with extended family groupings in the various communities . Treatment , costs and health seeking behaviour ( including perceived aetiology of the illness , traditional healing and delay ) were among the primarily topics during these sessions . Insights and established categories from an initial phase of qualitative research were used to assess preliminary data , and construct cost categories , which facilitated further systemisation and eliciting of costs from respondents , and were useful as secondary indicators to evaluate people's stated costs and economic coping strategies . After this stage , data were more systematically gathered and standardised through ( iv ) a quantifiable half-open structured questionnaire realised with all patients at Ayos and Akonolinga Hospitals and carried out in the local communities . At the Ayos and Akonolinga Hospitals , 79 clinically confirmed hospitalised Bu patients were included in the sample , representing all patients in treatment during the four month period of the study from November 2005 to February 2006 . The costs and cost burden presented in this article apply exclusively to the hospitalised patients . However , to gain a better understanding of Bu patients' and households' health seeking behaviour , the social and economic burden of the disease , and the relationship between local communities and the hospital setting , 73 patients and their households were further included for qualitative analysis at the community level , representing all active and inactive Bu cases living in the selected communities at the time of study . The costs accrued by the latter are outside the scope of this article . For the definition and categorisation of costs , the conceptual framework proposed by Russell [9] was employed to ensure clarity and precision and to allow for further theory building and comparability of the economic burden of the illness across settings ( Figure 1; Table 1 ) . In the present context , the household was operationally defined as a group of people who live together in a common residence , forming a unit of economic cooperation , and who are responsible for the socialisation of the children born of its members . The household was the preferred unit of analysis for assessing the economic costs and consequences of illness for two reasons . Firstly , local kinship relations are traditionally based both on patrilineal filiation ( kinship through the father's line ) and patrilocal residence patterns ( adoption of the father's place of residence in marriage ) . Communities are consequently divided into different settlements of patrilocally nucleated extended family clusters ( or patrikin groups ) , consisting of various related households that nevertheless function independently . Despite the proximity of patrikin , the household is the basic economic unit when coping with the illness costs of its members . Secondly , because decisions about treatment are based on negotiations within the household ( though not necessarily from an equal bargaining position ) , since the costs of illness reach beyond the sick to involve other household members who care for and accompany them to treatment and who ultimately are affected by the costs of illness which fall on the household budget and diminish the resources available for other household members [9] . Household subsistence strategies at the local level consist of a combination and alternation of revenues . Of primary importance is slash and burn agriculture , characterised by the simultaneous exploitation of several plots with a variety of intercropped products ( such as plantains , macabos , maniocs and peanuts ) that are harvested at different periods and can be more intensively exploited according to the necessities of general household spending . Earnings from slash and burn agriculture are often supplemented by earnings from cacao and/or café plantations ( a heritage of colonial times ) , which require different agricultural techniques and can only be harvested at specific intervals each year . Furthermore , the proximity of the Nyong and its tributaries provides communities with fish while the tropical rainforest offers possibilities for hunting . Both resources are used for consumption and sale . Occasional formal and salaried employment and migration to urban centres or abroad represent additional complementary subsistence strategies , although the latter signifies the separation of the household . Earnings were compiled through in-depth interviews with the household providers and was calculated based on the combined earnings of all household members for the period of one calendar year . For most cases , households were entirely dependent on subsistence farming . For slash and burn farming , activities were systemised in an agricultural calendar specifying products' harvest times or intervals , product prices and the subsequent estimation of earnings . Earnings from fishing were calculated according to an activity calendar , incorporating the amount of fish , fish prices and number of days worked . Hunting is a more irregular activity and meat is consumed more often then sold . Nevertheless , extra earnings of sold bush meat were also incorporated into calculations . In cases where one of the household members had formal employment , his/her yearly earnings were calculated according to the official salary . Usually this person further participates in the household's slash and burn agriculture , the revenues of which contribute to the general household earnings . The yearly revenues for informal urban and semi-urban employment ( i . e . undocumented , unofficial and irregular employment such as construction work , etc . ) were also included in calculations for the patient and/or other household members . The study was approved by the ethical committee of the Ministry of Health , Cameroon ( No . 0123/ARRO/MSP/DPSPL ) . For the field work , all interviewers were requested to follow the Code of Ethics of the American Anthropological Association ( AAA ) [10] . As proposed by the AAA , all interviewees were informed before the start of the interview about project goals , the topic and type of questions , their right to reject being interviewed , to interrupt the conversation at any time , and to withdraw any given information during or after the interview , and the intended use of results for scientific publications and reports to health authorities . Anonymity was guaranteed and confidentiality of interviewees was assured by assigning a number to each informant . The interviewers sought oral consent from all interviewees . Oral consent was preferred , since the interviewees were not put at any risk of being harmed in their safety or psychological well-being and the act of signing one's name when providing economic data was considered a potential reason for mistrust ( see also AAA which states that “It is often not appropriate to obtain consent through a signed form-for example ( … ) where the act of signing one's name converts a friendly discussion into a hostile circumstance” [11] .
During the study period , 34% of all patients registered at the Ayos and Akonolinga Hospitals since 2003 were classified as “healed” with a median hospitalisation time of 157 days ( range 47–645 days ) . 39% of hospital patients were female and 61% male . 56% were children and adolescents ( <20 ) while 44% were adults or elderly . Monthly household earnings of Bu hospital patients previous to their illness were calculated for all patients and households . The median value was €40 , 7 ( 26 . 400 FCFA ) . This was slightly higher than the median earnings for subsistence farming in the studied endemic communities , which was estimated during fieldwork at €33 , 8 ( 22 . 000 FCFA ) . These median values correspond with those of the World Bank [12] and UNDP [13] , which state that 51% of Cameroon's population lives below the $2/day mark while 17% earn less then $1/day . During in-patient treatment , the most common coping strategies included the use of savings ( 39% ) ; making claims from social networks ( 75% ) ; borrowing ( 53% ) ; reducing consumption of non-essentials ( 100% ) and essentials ( 69% ) ; patient informal employment ( 36% ) ; informal employment of the caretaker ( 43% ) ; the supplying of provisions from family relations in nearby villages ( 56% ) ; and , the social isolation of the patient ( 63% ) . The coping strategies employed were generally cost prevention strategies , similar to those used for other long-term and chronic illnesses , examples of which are found in studies of TB illness costs where households engage in either cost prevention strategies ( do not seek treatment or abandon treatment ) or asset strategies to mobilise substantial sums of money [9] , [14] , [15] . The median total costs for household members' involvement for a patient that was not isolated totalled €105 , 9 ( 68 . 848 FCFA ) while for an isolated patient the costs decreased dramatically to only €12 , 4 ( 8 . 051 , 87 FCFA ) . These numbers signify that for patients who were not isolated the costs for household involvement during the healing process was 8 , 6 times higher than for isolated patients . Subsequently 63% of households did avoid these costly direct expenses , leading to the social isolation of the in-patients . In the words of isolated patients:
Growing awareness of the connection between ill health and impoverishment has placed health at the centre of development agencies' poverty reduction targets and strategies and strengthened arguments for a substantial increase in health sector investment . Subsequent research , however , has shown the difficulty in alleviating the cost burden of illness for poor households due to the presence of hidden costs . Our research found that despite the availability of free of charge medical care , households of Bu patients in the study area faced an unbearable cost burden . The majority of households responded by withdrawing involvement in care , leaving patients socially isolated . This reality raises questions of how to increase accessibility to biomedical treatment and how to proceed with future programs . The median cost burden of Bu hospital treatment was estimated at 25% for costs directly covered both by the household and by the patient . In most illness studies mean direct costs are estimated between 2 , 5% and 7 , 0% of household earnings . However , many chronic illnesses , such as TB in Malawi , account for cost burdens between 8–20% of annual earnings; and in sub-Saharan Africa and Thailand , AIDS related treatment absorbed 50% or more of annual income [9] ( median values were not presented in the referenced articles ) . With its 25% cost burden , Bu surpasses the cost burden threshold of 10% , which has shown to be catastrophic for the household economy [9] , [16] , [17] and likely leads to further impoverishment . Households , therefore , recur to a variety of coping strategies to manage these expenses and indirect losses . A common coping strategy against the accumulation of unmanageable costs is manifest in the breaking of ties with the individuals most taxing on the household economy; in this case the patients . As stated in the results , of the total direct cost , only 43% are expenses directly related to the patient and his or her treatment , while 57% are due to household involvement during the illness period . However , the 57% of direct expenses can -and often is- avoided by households . This circumvention of costs was evident for 63% of in-patient children and 71% of the adults who had no caretaker present . These numbers signify that for patients who are not isolated the direct costs for their households during the healing process is 8 , 6 times higher than for isolated patients . Furthermore , qualitative research affirmed that the caretaker during hospital treatment is not a stable fixture; rather , a considerable percentage of caretakers leave during treatment , especially when they perceive the illness to no longer be ‘serious’ . Subsequently , 63% of in-patients were socially isolated at the hospital setting . Furthermore , initial qualitative data showed that , according to hospital patients , social isolation is the principal cause for abandonment of in-patient treatment . In endemic communities , it was further stated that fear of social isolation is one of the major reasons for postponing or avoiding hospital treatment and why traditional healing is often preferred since it is usually locally provided or provided in the vicinity of the patient's community . With respect to the cost burden equation , the weight of certain variables such as transportation costs , feeding costs and productivity loss of caretaker ( s ) are proportionate to the distance from the community to the treatment centre: ( 1 ) Transportation costs . While transportation costs are an additional cost for the patient , they represent a debilitating burden for caretakers since they continue to have social obligations to other household members ( i . e . other children ) , signifying repeated visits to the hospitalised patient . Accordingly , transportation costs account for 29% of median direct costs . ( 2 ) Feeding costs . The costs of food for the patient at the hospital setting can be minimised through the provision of agricultural products from the household's slash and burn cultivation . However , this coping strategy is only feasible in cases when regular visits are also feasible , such as when treatment is carried out in the general vicinity of the patient's community . This avoids extra expenses for paid meals for patients and caretakers ( feeding costs during hospitalisation represent 25% of median direct costs ) . ( 3 ) Productivity loss . Though the productivity loss of Bu patients could arguably be comparable in the hospital and community settings , that of the caregivers is greatly affected by the location of patient's treatment . When the patient receives treatment in the community or within the vicinity the caregiver can combine his/her daily economic activities with the patient's care without a serious impact on productive activities ( median indirect costs for caretakers represent one of the most taxing costs ) . The importance of the above-mentioned factors in relation to the economic and social impact of Buruli ulcer disease was also evident in patients' health seeking behaviour during field work . The fact that traditional healing minimised or largely avoided such costs was cited by respondents as a major reason why traditional treatment was often preferred to biomedical treatment . In this sense , decentralisation is expected to sidestep or drastically reduce the mentioned costs and to increase the sustainability of households' involvement during the patient's illness period , reducing patients' social isolation and create possibilities to increase adherence . From a socio-economic perspective , it can , therefore , be concluded that a decentralised system of treatment with minimal hospital stays could limit household impoverishment as the long term nature of the illness makes it impossible for a household to indefinitely sustain its involvement with the hospitalised patient . A multidisciplinary study to evaluate a decentralised system of care with minimal hospital stays is , therefore , essential .
|
The cost burden of free of charge Buruli ulcer disease ( Bu ) hospital treatment is not sustainable for a majority of patients and their families in Central Cameroon . The long term nature of Bu taxes the patients' and their families' resources often to a breaking point , consequently often leading to the abandonment of patients by the family . In the study area , 62% of families ceased providing social and financial support to the patient , which resulted in the patient's isolation at the hospital . Significantly , social isolation was cited by in-patients as the principal cause for abandonment of biomedical treatment . Paradoxically , this phenomenon was observed in settings where hospital in-patient treatment , room and board were provided free of charge for the patient and caretaker . These findings show that despite the significant reduction in costs for medical care , in its current form , hospital treatment for Buruli ulcer often remains financially and socially unsustainable for patients and their households , leading to the abandonment of biomedical treatment or altogether avoiding it . Further investment and research are urgently needed to evaluate new intervention strategies that are both socially and financially acceptable and appropriate in local settings .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"health",
"and",
"epidemiology/social",
"and",
"behavioral",
"determinants",
"of",
"health",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"public",
"health",
"and",
"epidemiology/health",
"services",
"research",
"and",
"economics",
"public",
"health",
"and",
"epidemiology/health",
"policy"
] |
2008
|
“It Is Me Who Endures but My Family That Suffers”: Social Isolation as a Consequence of the Household Cost Burden of Buruli Ulcer Free of Charge Hospital Treatment
|
Endometriosis is a gynecological disease defined by the extrauterine growth of endometrial-like cells that cause chronic pain and infertility . The disease is limited to primates that exhibit spontaneous decidualization , and diseased cells are characterized by significant defects in the steroid-dependent genetic pathways that typify this process . Altered DNA methylation may underlie these defects , but few regions with differential methylation have been implicated in the disease . We mapped genome-wide differences in DNA methylation between healthy human endometrial and endometriotic stromal cells and correlated this with gene expression using an interaction analysis strategy . We identified 42 , 248 differentially methylated CpGs in endometriosis compared to healthy cells . These extensive differences were not unidirectional , but were focused intragenically and at sites distal to classic CpG islands where methylation status was typically negatively correlated with gene expression . Significant differences in methylation were mapped to 403 genes , which included a disproportionally large number of transcription factors . Furthermore , many of these genes are implicated in the pathology of endometriosis and decidualization . Our results tremendously improve the scope and resolution of differential methylation affecting the HOX gene clusters , nuclear receptor genes , and intriguingly the GATA family of transcription factors . Functional analysis of the GATA family revealed that GATA2 regulates key genes necessary for the hormone-driven differentiation of healthy stromal cells , but is hypermethylated and repressed in endometriotic cells . GATA6 , which is hypomethylated and abundant in endometriotic cells , potently blocked hormone sensitivity , repressed GATA2 , and induced markers of endometriosis when expressed in healthy endometrial cells . The unique epigenetic fingerprint in endometriosis suggests DNA methylation is an integral component of the disease , and identifies a novel role for the GATA family as key regulators of uterine physiology–aberrant DNA methylation in endometriotic cells correlates with a shift in GATA isoform expression that facilitates progesterone resistance and disease progression .
Endometriosis is a painful and chronic gynecological disease that affects approximately 10% of reproductive-age women , causing infertility and development of adhesions due to the extrauterine growth of endometrium-like tissue [1] , [2] . The molecular cause of endometriosis is multifactorial , with disease occurrence and severity influenced by heritable components as well as environmental and lifestyle factors [3] , [4] . The cellular origin of endometriosis has been elusive , and several different models have been proposed to account for the many manifestations of the disease [5] . Of these , Sampson's model of retrograde menstruation remains most widely accepted , because the frequent occurrence of menstrual reflux explains the more common distribution of endometriosis to the ovaries and pelvic peritoneum , and because explanted endometrial tissue can give rise to endometriotic lesions [5]–[7] . Sampson's model was also remarkably intuitive , as we now observe endometriosis to be medically , historically , and evolutionarily linked to menstruation . Yet this model fails explain why only 10% of women develop endometriosis when most experience retrograde menstruation , nor can it explain instances of endometriosis that arise independently of menstruation [8] , [9] . Emera and Wagner provided clarity by proposing that menstruation is a mechanistic consequence of the evolution of hormone-induced spontaneous differentiation of the endometrium [10] . We would extend this model to the molecular level by suggesting that changes in the unique genetic regulatory networks controlling this hormonal trigger in menstruating primates permit the development of endometriosis . In most placental mammals , the differentiation ( decidualization ) of the endometrial stroma into the decidual cells of pregnancy is induced by the implanting blastocyst [11] , [12]; however , primates that menstruate initiate decidualization through an evolutionarily unique mechanism: the post-ovulatory rise in maternal progesterone [10] , [13] , [14] . Consequently , decidualization is triggered in women with every ovulatory cycle independent of pregnancy . Continued development and maintenance of the decidua is dependent on progesterone , and hormone withdrawal in the absence of a pregnancy provokes targeted apoptosis and eventual shedding of the superficial endometrium [15] . This physiological response is blunted in endometriosis . In contrast to healthy tissue , endometriotic tissues are progesterone-insensitive and resistant to apoptosis [16] , [17] , and many of the pathways utilized in progesterone-dependent decidualization are dysregulated in endometriotic lesions [18]–[20] . This suggests that alterations in the steroid-governed pathways unique to spontaneous decidualization underlie the pathogenesis of endometriosis . Intriguingly , the molecular characterization of diseased cells suggests that epigenetic defects strongly affect these pathways . DNA methylation serves as a critical regulator of gene expression , and global differences in DNA methylation affect multiple aspects of development and disease . Endometriotic cells express variable levels of the DNA methyltransferase enzymes ( DNMTs ) , which introduce and maintain DNA methylation on the C5 position of cytosine in CpG dinucleotides [21] . Abnormal DNA methylation in endometriosis affects the expression of several genes , including homeobox A10 ( HOXA10 ) , estrogen receptor beta ( ESR2 ) , steroidogenic factor 1 ( NR5A1 ) , and aromatase ( CYP19A1 ) , which alter steroid signaling and responsiveness , and are critically involved in development and decidualization [3] , [4] , [22]–[24] . While we and others have uncovered several individual genes with altered DNA methylation in endometriosis , the global profile of DNA methylation in endometriosis has not been characterized with the granularity necessary to detect many gene-specific methylation differences . We hypothesize that highly focused deviations in stromal cell DNA methylation , either inherited or acquired , affect key genes involved in spontaneous decidualization and contribute to the progression of endometriosis . These defects alter gene expression in stromal cells and their response to steroid hormone signaling during the menstrual cycle . Here , we used Illumina's high resolution HumanMethylation450 beadchips in conjunction with gene expression arrays to compare ovarian endometriotic stromal cells ( OSIS ) and healthy endometrial stromal cells ( EIUM ) treated with or without decidualizing stimuli .
After normalization , 45 , 429 probes from the gene expression array were retained for analysis . Using principal component analysis ( PCA ) , we examined the gene expression differences between 6 EIUM samples ( B , G , H , I , L , M ) and 6 OSIS samples ( O , P , Q , R , U , X ) both with ( + ) and without ( − ) in vitro decidualization ( IVD ) treatment . Nearly 66% of the variation across 45 , 429 probes was accounted for in the first two principal components ( 45 . 7% for PC1 , 19 . 9% for PC2 ) , and the samples clustered by both disease status and treatment when projected across these components ( Figure 1A ) . Hierarchical clustering of the same data , shown in the dendrogram in Figure 1B , separated the samples into two primary groups based on disease status . Within these groups , samples were further divided primarily by treatment , although EIUM I- and OSIS O+ were exceptions to this second level of clustering . Based on total signal from the array , one of the most abundant mRNAs in both cell types was smooth muscle actin ( ACTA2 ) . This was intriguing as we anticipated ACTA2 expression in OSIS , but not EIUM . We saw this more vividly at the protein level ( Figure 1C ) , where abundant ACTA2 fibers present in OSIS , both before and after IVD , coincided with some of the morphological differences seen in the diseased cells . ACTA2 protein was noticeably absent in EIUM , which typically appear as highly elongated , spindle-shaped cells that adopt a more rounded , epithelioid shape in response to IVD . OSIS are often elongated as well , but display a more irregular shape with a jagged or ruffled appearance along their edges . Moreover , OSIS cells were less rounded after IVD relative to EIUM . We compiled lists of differentially expressed genes for later comparison to the methylation array , and to confirm the phenotype of the cells . Comparing OSIS to EIUM ( Venn diagram Figure 1D ) , we identified 2 , 430 genes that were differentially expressed between the untreated groups , and 4 , 764 genes that were differentially expressed between the treated groups . Within each of these comparisons , the numbers of up- and downregulated genes were similar ( i . e . , in OSIS- vs . EIUM- , there were 1 , 143 upregulated genes and 1 , 287 downregulated genes ) . When the effect of treatment was examined in each population ( i . e . , EIUM+ vs . EIUM− and OSIS+ vs . OSIS−; Venn diagram Figure 1E ) , EIUM were observed to be more sensitive to IVD , with nearly 3 times as many genes differentially expressed in EIUM ( 4 , 549 ) compared with OSIS ( 1 , 517 ) . Consequently , EIUM showed a large number of unique genes that changed with treatment ( 3 , 695 ) compared to OSIS ( 663 ) . This suggests that the large differences in gene expression seen between normal and diseased cells are expanded further in response to IVD , when healthy stromal cells begin to decidualize but diseased cells showed a blunted response ( for full gene lists from these comparisons please see Table S1 ) . Using qPCR , we validated a panel of genes that were differentially expressed on the array and representative of gene targets known to be differentially expressed in endometriosis or in response to IVD ( right bar graph panels , Figures 1D , 1E ) . Consistent with previous reports , the expression of estrogen receptor alpha ( ESR1 ) , progesterone receptor ( PGR ) , matrix metalloproteinase-11 ( MMP11 ) , and retinaldehyde dehydrogenase 2 ( ALD1A2 ) was significantly lower in OSIS relative to EIUM; in contrast , the expression of estrogen receptor beta ( ESR2 ) , steroidogenic factor 1 ( NR5A1 ) , homeobox C6 ( HOXC6 ) , and aromatase ( CYP19A1 ) was significantly higher in OSIS relative to the EIUM ( p<0 . 05 , main effect two-way ANOVA ) [3] , [23] , [24] . Interestingly , both ESR2 and CYP19A1 showed a significant interaction between treatment and disease , with IVD increasing their expression in OSIS to a greater extent than in the other groups ( p<0 . 001 , Tukey's ) . The greatest differences in expression following IVD were seen for genes known to be induced during decidualization in vivo . The forkhead box protein O1 ( FOXO1 ) as well as the heart and neural crest derivatives-expressed protein 2 ( HAND2 ) are both essential mediators of the decidual response , and are increased in response to progesterone [22] , [25] . Additionally , the well-characterized decidual markers prolactin ( PRL ) and insulin-like growth factor-binding protein 1 ( IGFBP1 ) are strongly upregulated by IVD in EIUM but not OSIS [11] . There was a significant interaction across the groups for all 4 genes ( p<0 . 05 ) . In addition to confirming the results of the array , these differences function like a molecular signature highlighting established markers for both healthy and diseased cells and how they respond to IVD . After normalization , 470 , 540 probes from the methylation array were retained for analysis . PCA of the methylation variation for all 12 samples , with and without IVD treatment , is shown in Figure 2A . More than 64% of the variation across all probes could be accounted for by the first principal component , and samples strongly clustered along this component based on their disease status ( PC2 accounted for 6 . 7% of the variation ) . IVD treatment had little effect on the sample variation , and hierarchical clustering ( Figure 2B ) showed that subsequent branching was dictated by inter-sample variation . Density estimations derived from the full range of normalized β-values from each sample ( Figure 2C ) showed a similar bimodal frequency distribution of methylation . This indicates that global levels of methylation are comparable on somatic chromosomes , and that there is neither overall unidirectional shift in methylation nor substantial hemimethylation in either population . We then examined how average β-values for individual CpGs on the array differed between groups . Untreated OSIS compared with untreated EIUM revealed the largest difference in methylation , with 42 , 248 differentially methylated CpGs . To visualize the differences in methylation at discrete CpGs , scatter plots were generated using average β-values from each group for each of the 470 , 540 CpGs ( Figure 2D , β-values considered different shown in red , unchanged are in blue ) . For this comparison , there was a slightly higher number of probes showing greater methylation ( 24 , 208 CpGs , or approximately 57 . 3% ) in OSIS relative to EIUM . Consistent with the PCA , only 249 CpGs were differentially methylated as a consequence of IVD treatment in EIUM ( Figure 2E , red ) , and 244 of these showed a decrease in methylation . Only 2 CpGs were differentially methylated in OSIS with IVD treatment ( Figure 2F , red; full lists of differentially methylated CpGs are provided in Table S2 ) . The probe selection on the HumanMethylation450 beadchip is focused on genes and CpG islands ( CGI ) . Accordingly , the annotations for individual CpGs on the array are subdivided based on their positional context relative to both nearby transcripts and the closest CGI ( Figure 3A ) . Nearly 80% of the 470 , 540 CpGs examined on the array are linked to a transcript , mapping either near promoters ( a range that includes upstream proximal promoters , the 5′UTR of transcripts , and the first exon of the gene ) , within the gene body ( typically intronic ) , or to the 3′UTR of the transcript ( Figure 3B , top pie charts ) . The annotation of the array's probes based on proximity to CGIs maps nearly one third of the CpGs within predefined CGIs , while another third map to regions flanking the CGIs , termed “shores” and defined as within 4 kb of the nearest island ( as mentioned in the data analysis section , we merged Illumina's “shelf” and “shore” categories in to a single category , which we referred to as “shores” ) . CpGs unrelated to an island ( i . e . , more than 4 kb removed ) comprised the remaining third , and were termed “open sea . ” We examined the gene context of the 42 , 248 differentially methylated CpGs that were identified between OSIS and EIUM , and found a similar breakdown to the full array; however , CpGs mapping near promoters ( pink , purple , and blue ) were slightly underrepresented relative to CpGs found in the body ( orange ) of transcripts ( Figure 3B , middle ) . The CGI context showed a larger difference relative to the ratios built into the array . The majority of CpGs with differential methylation mapped to open seas ( light blue ) , while the fraction of CpGs in islands ( brown ) was reduced to one third of what we expected from the array . We also noticed that only 2 of the CpGs identified as different before and after IVD mapped to a CpG island; however , we did not follow this group of CpGs in subsequent analyses because so few were differentially methylated ( these are not represented graphically , see Table S2 ) . We then merged our results for the untreated OSIS and EIUM groups from the two arrays , and identified 1 , 402 differentially expressed mRNAs to which we could map 5 , 423 differentially methylated CpGs . We referred to these as matched CpGs ( Table S2 ) . Similar to above , matched CpGs occurred most frequently in the body of genes and distal to CGIs in open seas ( Figure 3B , bottom ) . However , the combined breakdown revealed that matched CpGs found near gene promoters were more often associated with shores or islands , fitting with the convention that CGIs are more frequently found in these regions . Matched CpGs in the body of a gene were more often observed in open sea . Inspection of the matched CpGs in gene bodies revealed that 88% were distributed within introns; however , these intronic regions did not appear to preferentially overlap with known enhancers ( data not shown ) . Because of the diverse distribution of these CpGs across different genomic contexts , we examined how changes in methylation were distributed across the matched genes based on CpG context . The heat maps in Figure 4 depict the direction of change in methylation between EIUM and OSIS side-by-side with the gene context and CGI context . Hypomethylated CpGs in OSIS are clustered near the top , whereas hypermethylated CpGs in OSIS are clustered near the bottom . The larger number of CpGs in gene bodies ( orange ) and in open seas relative to CGIs ( light blue ) is again apparent , but appears enriched near the bottom , suggesting they are more frequently hypermethylated . Additionally , a cluster of CpGs mapping to CGIs near promoters is observed more centrally , suggesting that the differences here are less pronounced . Methylation is conventionally thought to repress gene expression; thus , we examined whether hyper- or hypomethylation at each of the matched CpGs correlated positively or negatively with gene expression ( Table 1 ) . When comparing the 5 , 423 differentially methylated and matched CpGs in OSIS and EIUM , nearly 64% of the CpGs showed hypermethylation ( 3 , 485 ) . Interestingly , twice as many of the hypermethylated CpGs were negatively correlated with gene expression ( 2 , 363 vs . 1122 ) . The 1 , 938 CpGs that were hypomethylated were more evenly divided between the positively and negatively correlated groups . We next constructed a test of proportions to examine the distribution proportion of positive and negative correlation at each CpG context . When stratified by gene context ( Table 2 ) , the number of differentially methylated CpGs in the first exon of a gene were more frequently negatively correlated with gene expression ( p<0 . 00001 ) , whereas CpGs in the 3′UTR showed a relatively higher proportion that positively correlated with gene expression ( p = 0 . 01584 ) . When stratified by island context ( Table 3 ) , differentially methylated CpGs that mapped to CGIs were also more likely to be positively correlated with gene expression ( p<0 . 00001 ) We wanted to statistically infer the impact of methylation across a given gene on its expression . While several statistical approaches have been used to report average weighted changes in methylation across a given gene or region of chromatin , the large and punctuated variations in methylation we observed across the many different contexts suggested that variable differences in methylation might be a more useful model for correlating methylation with gene expression . To examine this , we used ANOVA to evaluate methylation variation across each matched gene . This approach was based on the prediction that statistically significant interactions across multiple gene or CGI contexts would identify genes more likely to have their expression affected by differential methylation . Consequently , cross-group interactions would preferentially identify genes where the methylation differences were both highly different in β-value and also widely distributed across unique contexts within the gene . At the same time , genes without multiple differentially methylated CpGs , or genes where methylation differences are clustered together , would be devalued . The ANOVA identified 403 genes ( Table S1 ) with a statistically significant interaction ( adjusted p<0 . 05 ) among their CpGs . This represented 2 , 978 of the differentially methylated CpGs ( Table S2 ) . One remarkable finding from this analysis was that it correctly identified the HOXA cluster , NR5A1 , and PGR—genes that are aberrantly methylated and differentially expressed in endometriosis—as highly significant . The 403 genes identified by ANOVA were uploaded to MetaCore and classified by protein function using ORA ( Table 4 ) . While many classes were present ( kinases , receptors , etc . ) , transcription factors were the only functional class to reach statistical significance ( p = 4 . 01×10−9 ) . To identify differentially methylated genes that may potentially alter the pathways involved in decidualization , we first identified the enriched GO processes among the differentially expressed genes observed when comparing IVD-treated OSIS and EIUM ( 4 , 764 transcripts shown in Figure 1C ) . We then determined which of these groups had significant overlap with the genes we identified by ANOVA interaction . The top processes identified through this intersection are shown in Figure 5 , and included several pathways important for the development and progression of endometriosis , such as organ development ( patterning ) , blood vessel development , neuronal development , and regulation of cell adhesion . From this we identified subsets of differentially methylated genes that encoded transcription factors enriched as hubs within these processes . We examined the methylation profiles at 11 regions that we predicted to be important in controlling the expression of transcription factors identified by ANOVA and in the enrichment analyses . Since these were identified by ANOVA , these genes were all both differentially methylated and differentially expressed on our arrays ( Tables S1 and S2 ) . The methylation plots for these regions were built by aligning the differentially methylated CpGs along diagrams for the associated genes ( Figures 6 and 7; orientation is relative to the “+” strand ) . Additionally , the average β-values for these CpGs from either EIUM or OSIS are plotted along the y-axis . The methylation profile for NR5A1 ( p = 4 . 11×10−39 , Figure 6A ) matches with those of previous reports for these cell types: relative to EIUM , the NR5A1 gene promoter in OSIS is hypomethylated but the gene body is hypermethylated [26] , [27] . Notably , we identified several regions of hypermethylation in the ESR1 gene ( p = 8 . 84×10−3 ) near its 3′ promoter as well as in regions flanking an intronic CGI ( Figure 6B ) . The ESR1 gene has been extensively studied in EIUM and OSIS , but differences in its methylation status have not been previously reported . HOXA10 and HOXA11 are known to be differentially methylated in endometriosis [28] , [29] . Our results extend these findings by identifying multiple regions across the HOXA cluster that are differentially methylated in OSIS relative to EIUM ( p-values are listed for the HOXA genes identified by ANOVA; Figure 6C ) . A unique pattern of differential methylation was seen across the central portion of the HOXC cluster on chromosome 12 , where both hypo- and hypermethylation occurred at CGI shores near the promoters of HOXC4 ( p = 2 . 16×10−22 ) , HOXC5 , and HOXC6 ( p = 1 . 80×10−7 , Figure 6D ) . The role of the HOXC genes in the endometrium is unknown . HOXC6 is regulated in part by two estrogen response elements in its promoter [30] , and is highly expressed in ovarian endometriotic tissue relative to eutopic tissue [23] . We also observed high levels of HOXC6 in OSIS relative to EIUM ( qPCR panel in Figure 1C ) , and the array reported higher levels of HOXC4 and HOXC8 as well . The T-box transcription factor 3 ( TBX3 ) is important for lineage decision and cell fate guidance during embryonic development , and the CGI overlapping its promoter and first exon is frequently differentially methylated in cancer [31] , [32] . We saw TBX3 expressed in EIUM but not in OSIS on our array , and noted that a CGI and much of this gene were uniformly methylated in OSIS ( p = 3 . 56×10−17 ) ( Figure 6E ) . In contrast , the zinc finger protein 423 ( ZNF423 ) was an example where hypermethylation throughout the gene body in OSIS was positively correlated with expression on our arrays ( Figure 7A ) . One of the most striking differences identified in this analysis was the high representation of GATA transcription factors and transcriptional coregulators of the GATA family . The zinc finger protein genes ZFPM1 and ZFPM2 , often referred to as friends of GATA , showed unique patterns of mixed methylation . In EIUM the 5′ region of ZFPM1 gene showed increased methylation while its 3′ region had reduced methylation . This pattern was reversed in OSIS where ZFPM1 had reduced methylation at its 5′ end relative to the 3′ ( Figure 7B ) . The ZFPM2 gene was largely hypomethylated intragenically in OSIS relative to EIUM , although the absolute differences in methylation varied across the gene ( Figure 7C ) . Both ZFPM1 and ZFPM2 were upregulated in OSIS relative to EIUM ) . Three GATA isoforms showed altered methylation . Multiple CpGs throughout the promoter and body of GATA2 showed higher methylation in OSIS relative to EIUM ( p = 6 . 02×10−10 , Figure 7D ) , while GATA4 and GATA6 had less methylation across ranges of intronic CpGs flanking CGIs ( 7E , p = 5 . 55×10−17 and 7F , p = 4 . 34×10−2 ) . Similar to the HOX genes , the individual methylation status of these GATA members was inversely correlated with gene expression , but because GATA family members are encoded on separate chromosomes , the observed methylation differences are not a shared occurrence . Because little is known regarding the role of GATA family members in the endometrium and endometriosis , we examined these genes more closely . Our microarray analysis demonstrated that GATA2 was more abundant in EIUM , whereas GATA4 , GATA6 , ZFPM1 , and ZFPM2 were more abundant in OSIS . To validate this , we examined mRNA and protein expression for all 5 genes in both cell types . GATA2 mRNA was 8 . 7-fold lower in OSIS than EIUM ( Figure 8A ) , whereas GATA4 and GATA6 were 1100-fold higher and 9 . 2-fold higher , respectively , in OSIS . Importantly , the qPCR results suggested that the abundance of GATA4 in EIUM and OSIS ( CT values from 35 to undetectable ) was very low relative to GATA2 and GATA6 ( CT values around 27 ) —thus although the fold induction of GATA4 was the largest , it was abundance was scarce relative to both the other isoforms . The data for the ZFPMs was less clear-cut . The mRNA levels of ZFPM1 and ZFPM2 were 1 . 8-fold and 2 . 4-fold higher ( respectively ) in OSIS than in EIUM ( Figure S1 ) . Only ZFPM2 was detectable at the protein level , where it was consistently expressed in OSIS , but more variably in EIUM ( ZFPM1 was undetectable using two different commercial antibodies ) . Immunoblots showed that GATA2 was highly abundant in EIUM , but scarcely detectable in OSIS ( Figure 8B ) . Immunofluorescence for GATA2 ( Figure 8C ) confirmed strong , uniform nuclear signal in EIUM , but OSIS cells stained with a weaker , more diffuse signal with a unique punctate appearance . GATA4 was detectable in both EIUM and OSIS by immunoblot , with slightly variable expression , but no signal was observed in either cell type by immunofluorescence . Immunoblots and immunofluorescence showed that GATA6 was robustly expressed and localized to the nuclei in all OSIS samples but barely detectable in EIUM . ( Notably , IVD had little effect on the expression of any of the GATA family members examined–compare figure 9A & 10A for EIUM , Figure 11A for OSIS . ) Given the striking differences in GATA2 and GATA6 expression , we proceeded to more closely examine their expression and function . We first validated the methylation data from the array using methylation specific PCR ( Figure S2 ) . Primers mapping to either exon 4 of GATA2 showed that this region was fully unmethylated in EIUM , and was predominantly methylated in OSIS . Primers targeting exon 2 of GATA6 showed that it was fully methylated in EIUM and fully unmethylated in OSIS . Given the stark differences in GATA expression in EIUM and OSIS , and that endogenous GATA2 was present only in EIUM , we wanted to examine how GATA2 affected gene expression and the decidual response in EIUM . We transiently transfected EIUM with gene-specific siRNAs ( xGATA2 , Figure 9 ) or scrambled siRNA controls ( xCONT ) , and the cells underwent IVD or control treatment for 6 days . GATA2 mRNA and protein were reduced in EIUM by more than 80% ( Figure 9A ) ; however , the loss of GATA2 did not affect the transcription of GATA4 or GATA6 ( Figure 9B ) . Similarly , the expression of the nuclear steroid hormone receptors were not significantly affected by the loss of GATA2 ( Figure 9C ) , nor were the other genes from our panel whose expression varied only as a function of disease ( NR5A1 , HOXC6 , CYP19A1 , and ALDH1A2 ( not shown ) . In contrast , when we examined the array of genes known to be differentially expressed in response to IVD ( shown in Figure 1 ) we found that silencing GATA2 significantly reduced the established markers of decidualization ( Figure 9C , bottom row ) . In the absence of GATA2 , the induction of HAND2 and PRL in response to IVD was reduced by 47% and 57% , respectively , and IGFBP1 induction was reduced by 88% ( p<0 . 05; Tukey's ) . The expression of FOXO1 was slightly reduced , but was not significant . This suggested that GATA2 expression may enhance decidual response . Using adenoviral vectors , we introduced either eGFP ( AdGFP ) or human GATA6 ( AdGATA6 ) into EIUM , treated the cells for 6 days with or without IVD , and then followed the effects of GATA6 on our panel of differentially expressed genes . The adenovirus significantly increased expression of the recombinant gene , and GATA6 expression was still abundant after the 6-day culture ( Figure 10A ) . As before , IVD did not affect GATA6 or the other GATA family members ( Figure 10B ) , but the overexpression of GATA6 resulted in a 2-fold reduction in GATA2 mRNA ( p<0 . 001 ) and a 4-fold increase in GATA4 mRNA ( p<0 . 001 ) . The effect of GATA6 overexpression on GATA2 protein levels was more pronounced ( Figure 10C ) , with expression decreased and redistributed in the nuclei to produce a punctate pattern similar to what we observed in OSIS in Figure 8C ( GATA2 pseudocolored from red to green due the use of GFP; GATA6 in red , Figure 10C ) . GATA6 overexpression also appeared to affect the cytoskeleton of EIUM , with a potent increase in ACTA2 expression that again resembled what we saw in OSIS ( see Figure 1E ) . The qPCR panel in EIUM cells overexpressing GATA6 revealed a pronounced shift in gene expression that mirrored the expression patterns we saw in OSIS ( compare Figure 10D and Figure 1 qPCR ) . All 4 of the nuclear receptors that we examined showed significant changes in expression . ESR1 mRNA was reduced by 1 . 6-fold when GATA6 was overexpressed without IVD , and by 3-fold after IVD ( both p<0 . 05 ) . The overexpression of GATA6 also reduced PGR transcript levels an average of 2 . 6-fold ( p<0 . 01 ) . In contrast , but consistent with the OSIS qPCR results in Figure 1 , the expression of ESR2 and NR5A1 were increased an average of 2 . 5-fold ( p<0 . 05 ) and 22 . 4-fold ( p<0 . 05 ) , respectively , and IVD did not significantly affect these genes . MMP11 and CYP19A1 mRNA levels were also strikingly altered by GATA6 overexpression . MMP11 was repressed by IVD ( by 2 . 7-fold ) , but overexpression of GATA6 further reduced its transcript levels by 30-fold ( p<0 . 01 ) , relative to untreated controls . As before , CYP19A1 was expressed at very low levels basally , but this was significantly increased after GATA6 overexpression ( 8 . 5-fold without IVD , 13 . 3-fold with IVD , p<0 . 01 ) . HOXC6 and ALDH1A2 did not change following either IVD or GATA6 overexpression . Finally , overexpression of GATA6 profoundly restricted the ability of EIUM to decidualize , with all four of the genes expected to increase with IVD ( FOXO1 , HAND2 , PRL , and IGFBP1 ) effectively blocked by GATA6 ( p<0 . 01 ) . The striking effects of altering GATA expression in EIUM led us to hypothesize that restoring the profile of GATA family members in OSIS to that seen in EIUM might recover the decidual response . Our experiments in EIUM showed that depleting GATA2 did not enhance GATA6 expression , and we anticipated that our experiments in OSIS would require simultaneous depletion of GATA6 and overexpression of GATA2 . To this end we used siRNAs to knockdown GATA6 in OSIS while simultaneously expressing GATA2 ( AdGATA2 ) via adenoviral transduction . The xCONT scrambled siRNA and eGFP expressing adenovirus were again used as controls . This was done both in the presence or absence of IVD treatment for 6 days ( n = 7 ) . Adenoviral-mediated expression of GATA2 increased mRNA and protein expression of GATA2 ( Figure 11A ) . After 6 days of culture , the GATA6 siRNAs reduced GATA6 mRNA by 74% , and protein levels by 51% . Remarkably , the expression of GATA2 also potently blocked GATA6 mRNA and protein , and the combined knockdown of GATA6 and overexpression of GATA2 reduced GATA6 mRNA by 95% and protein by 92% ( Figure 11A ) . Intriguingly , GATA4 mRNA expression in OSIS was reduced by GATA2 overexpression but not GATA6 knockdown ( Figure 11B ) . Exogenously expressed GATA2 protein was observed to be completely nuclear ( Figure 11C ) . We were surprised to observe that only 2 genes in our panel were significantly affected in response to manipulating GATA expression: ALDH1A2 and CYP19A1 ( Figure 11D ) . The nuclear hormone receptors were not affected by any treatments , although there was a trend for increased basal ESR2 after GATA2 overexpression . Likewise , neither silencing endogenous GATA6 nor expression of GATA2 improved the decidual response in OSIS . Quite opposite , GATA2 overexpression appeared to block the slight induction of HAND2 seen after IVD in OSIS , while PRL and IGFBP trended downward . As a result , IVD was the main effect for the expression of FOXO1 , HAND2 , PRL , and IGFBP1 in OSIS , the relatively heterogeneous expression demonstrated that subject-to-subject variation was greatly affecting these genes . The expression of ALDH1A2 was induced by GATA2 by 3-fold , but was not affected by IVD or GATA6 depletion . The expression of CYP19A1 revealed a more complex expression pattern in response to these treatments . As before , IVD strongly increased CYP19A1 expression . Silencing GATA6 decreased basal CYP19A1 by 87% , although this did not reach statistical significance . Moreover , silencing GATA6 nearly blocked IVD-stimulated CYP19A1 . Fitting with this , the expression of GATA2 , and concomitant loss of GATA6 , affected CYP19A1 expression in a pattern similar to GATA6 depletion .
We found that endometriotic cells possess a unique epigenetic fingerprint compared to healthy endometrial stromal cells . Moreover , we identified a large network of transcriptional regulators differentially methylated in endometriosis and linked to decidualization . This included a surprising number of GATA family members . With further examination of GATA2 and GATA6 , we found that GATA2 strongly regulates genes essential for decidualization , whereas GATA6 promotes an endometriotic phenotype . From this we suggest the possibility that an epigenetic switch controlling GATA isoform expression is important in the progression of endometriosis . Focused efforts and better techniques are rapidly improving our understanding of how DNA methylation affects cell differentiation and human disease . The converging evidence that endometriosis is linked evolutionarily to decidualization , that steroid-dependent pathways are dysregulated in the disease , and that DNMTs are differentially expressed and regulated by steroid hormones in both cell types led us and many others us to speculate that both decidualization and endometriosis could be affected by dynamic epigenetic cues arising from DNA methylation [21] , [33]–[38] . This appears to be the case in endometriosis , as the number of genes aberrantly methylated in endometriosis continues to grow , and the mutable and heritable nature of DNA methylation fits well as a mechanism to help to explain the enigmatic occurrence of endometriosis [3] , [4] , [39] , [40] . Our data suggests that this epigenetic component is a defining feature of endometriosis , and also helps to unify many of the diverse observations regarding its origin . While it is our opinion that Sampson's model can explain most instances of endometriosis , it is clear that lesions can arise through other mechanisms such the induction of either ectopic mesenchymal cells or stem cells , or similarly by müllerianosis [5] . We suggest that epigenetic defects in the unique genetic pathways of the primate underlie all these mechanisms , and that the spontaneous occurrence of endometriosis in primates can be traced to an epigenetic plasticity in mesodermal mesenchymal cells . Borghese et al . published the first global survey of DNA methylation in whole endometriotic tissues using MeDIP arrays to profile specific promoters [41] . Their work suggested that global patterns in methylation are similar between endometriosis , and that variation in methylation was more likely to occur at discreet loci across the genome . The gene-centric focus of our array compliments and expands these observations , and we observed genome-wide differences more frequently in the body of genes , and in the shore and open sea areas that flank CpG islands . Four of the loci they reported mapped to genes that we identified by ANOVA interaction: 5′-AMP-activated protein kinase subunit gamma-2 ( PRKAG2 ) , HOXD10 , zinc finger protein 22 ( ZNF22 ) , and the anoctamin 1 , calcium activated chloride channel ( ANO1 ) . In addition to having differentially methylated promoters , we saw methylation differences in the body and UTRs of these genes . These genes show a consistent correlation between DNA methylation and gene expression which is altered in endometriosis , although their function in the endometrium remains largely unknown . HOXD10 has been implicated in endometrial stromal cell proliferation , and our pathway analysis comparison identified it as well , suggesting that all of the HOX clusters may show aberrant methylation in endometriosis [42] . Notably , we saw very few changes in DNA methylation in response to IVD treatment . When changes did occur , they were almost exclusively instances of hypomethylation . Interestingly , one of these in EIUM mapped to IGFBP1 , which is massively upregulated in healthy cells both in vivo and in vitro in response to progesterone . It seems likely that the slightly increased number of CpGs that change in EIUM after IVD treatment reflects their increased sensitivity and dynamic response to steroid hormones , whereas the paucity of CpGs in OSIS that change with treatment underscore its more rigidly differentiated phenotype . Gao et al . recently used methylation-sensitive restriction fingerprinting to examine DNA methylation in the mouse endometrium during pregnancy and pseudopregnancy; they were able to identify several loci where differential methylation was induced during decidualization , although these did not correlate with our current findings [36] . It is important to recognize that the Gao and Borghese groups both used whole tissue as well as different platforms to interrogate DNA methylation , and these differences highlight important limitations when interpreting and comparing results . Our present study made extensive use of homogenous cultures of first passage primary stromal cell cultures , isolated several days in advance of starting the experiment . While the examination of whole tissue fragments taken fresh ex vivo is clearly relevant , and may ultimately be more informative , it is well established that large epigenetic differences occur between different cell and tissue types [43] . Since the Illumina beadchip can discern very subtle differences in methylation , and the signal from heterogeneous samples can deteriorate , our goal of examining methylation defects in the context of spontaneous decidualization required exceptionally pure sources of stromal cells [44] . We anticipate that this will provide a clear framework as we explore methylation in more complex samples , such as endometriotic cells from tissues other than endometrioma , and also the eutopic endometrium in women with endometriosis , where the populations of cells with altered DNA methylation will likely represent only a small fraction of the sample . A challenge in our work and to this field in general , is the difficulty in demonstrating and quantifying the influence of DNA methylation on gene expression . The lack of methods for experimentally manipulating CpG methylation in a site-specific manner prevents us from unambiguously interrogating the direct effects of these epigenetic marks . This problem is exacerbated by the complexity that arises when multiple methylation differences are observed across a region . Are some regions more influential than others ? What mechanisms govern how methylation will be correlated with gene transcription ? Most work addressing these questions has relied on correlating DNA methylation with respect to the genomic context in which it occurs . Early observations identifying CGIs near gene promoters led to the prediction that the methylation of these regions silenced downstream genes [45] , and this convention is important in cancer where hypermethylation of CGIs is frequently observed against a backdrop of global hypomethylation [46] . Growing evidence suggests this model to be over simplistic , in particular in non-cancerous cells , as CGIs more often escape methylation , while more isolated CpGs within the genome are more variably methylated , and better correlated with gene expression [47] , [48] . Moreover , these variable regions do not always demonstrate a negative correlation with gene expression , as we observed for genes such as NR5A1 , the HOXA cluster , and the ZFPMs , and further studies are needed to identify possible mechanisms at work . Often intronic enhancers or repressors are epigenetically regulated by methylation . This is the case for NR5A1 which is differentially expressed in the adrenal , the hypothalamus , and the gonad under the direction of multiple elements [27] , [49] , [50] . The intronic regions of NR5A1 are hypermethylated in the adrenal as well as in endometriotic stromal cells , which allows for higher levels of expression [26] . More recently , intragenic methylation has been show to alter tissue-specific transcription factor and methyl CpG binding protein MECP2 in order to alter tissue-specific splicing and gene expression [51] , [52] . The differential methylation we uncovered in endometriosis suggests that several different mechanisms are at work . Unlike cancer cells , there was significant hypo- and hypermethylation in endometriotic cells , and these were distributed in a variety of patterns . The majority of differential methylation was observed intragenically and at sites distal to CGIs . While the gene-centric bias in the array may explain the over-representation of intragenic CpGs , the array was similarly biased in favor of island CpGs , suggesting the increased incidence in differential methylation across shore and open sea regions is biologically relevant . Moreover , shore and open sea CpGs were much more likely to be negatively correlated with gene expression , particularly when they occurred near the TSS ( such as the TSS200 and 1stExon groups ) . Recent work from several groups suggests that intragenic methylation , in particular near the first exon , is important for coordinating tissue-specific nucleosome positioning and gene expression [53] , [54] . Although the well-spaced coverage of the 450K beadchip makes it possible to detect many of these differences , larger data sets are needed for the extrapolation of unique methylation patterns which correlate with gene expression[55] . We anticipate that our present data will continue to provide valuable insight into gene regulation in endometriosis as future studies decode the spatial and genomic context through which DNA methylation can affected transcription . Given the limitations we faced in correlating the effects of methylation on gene expression , we developed a novel interaction modeling pathway to better capitalize on the broad number of probe sets provided on the 450K beadchip . The rationale for this was to improve the fidelity with which we could identify the genes whose expression was associated with differential methylation . This model accurately identified genes either known or suspected to be affected by DNA methylation in endometriosis , such as ESR2 , NR5A1 , PGR and HOXA10 [3] , [4] , [56] . While the array we utilized extensively expands these findings by providing more detailed and quantified differences , our model focused on identifying regions with more extensive and significant deviations in methylation ( the HOXA cluster for example ) . Similarly , our model identified genes that are frequently affected by aberrant methylation in other diseases such as cancer , including the tumor suppressors deleted in cancer 1 ( DLC1 ) and transcription factor 21 ( TCF21 ) . The most exciting finding from this model was the large number of novel transcription factors , such as ESR1 and the HOXC cluster . Methylation of ESR1 frequently silences its expression in cancer , but our discovery of aberrant ESR1 methylation in endometriosis is novel [57] , [58] . Endometriotic cells typically have increased levels of ESR2 due to hypomethylation of its promoter , and the interplay between ESR1 and ESR2 results in the altered response of the diseased cells to estrogen [59] . The most surprising discovery in the list of differentially methylated genes was the GATA family of transcription factors , as the physiological role of this family in the uterus is largely unknown . In particular , the unique roles of GATA2 and GATA6 in apparent opposition to each other , is novel in the uterus ( Figure 12 ) . We expected GATA4 may also serve in conjunction with GATA6 , since GATA4 has been detected in rabbit endometrium [60] and is often seen to function alongside GATA6; however , our results suggest GATA4 is much less abundant in human EIUM and OSIS . Several studies have also demonstrated a role for GATA3 in the endometrium throughout the menstrual cycle and in women with endometriosis where it may play a role in modulating cytokine expression [61] , [62]; however its expression and methylation were not statistically different based on our arrays . More recently , GATA2 was identified in the mouse endometrium where it appears to coordinate PGR signaling in decidualizing stroma [63] , fitting well with our results . We identified GATA2 to be dominant in EIUM where it may be important for spontaneous decidualization , as either silencing GATA2 or overexpressing GATA6 can disrupt the effects of IVD . Notably , GATA2 knockdown in EIUM did not affect steroid hormone receptor expression or the expression of many of the genes on our panel , but disrupted the early targets of progesterone , such as HAND2 and IGFBP . Thus GATA2 may be important in amplifying signal transduction from maternal progesterone; however additional work is necessary to determine how GATA2 expression is affecting decidualization . Additionally , we saw that exogenous GATA2 in OSIS was able to promote the expression of ALDH1A2 . Retinoid synthesis is important for endometrial function , whereas reduced retinoid synthesis contributes to increased survival and decreased apoptosis in endometriosis [24] , [64] . Additionally GATA2 was able to antagonize GATA6 and concomitantly CYP19A1 in OSIS . While we do not know what predisposes GATA6 to be expressed in OSIS , is conceivable that GATA2 could help mitigate improper GATA6 expression . Thus in addition to enhancing progesterone sensitivity , GATA2 may be important in maintaining the unique differentiated state of the endometrial stroma . The role of GATA2 needs to be further studied with respect to decidualization , and in vivo models in mouse and non-human primates would be ideal for characterizing its function during the estrous cycle and menstrual cycle , respectively . The overexpression of GATA6 drives the gene expression profile in EIUM toward that seen in OSIS , which appear to only express GATA6 . Given this dramatic phenotype , it was remarkable that the depletion of GATA6 and exogenous GATA2 expression did not reverse the OSIS phenotype . This is likely a consequence of the more significant methylation defects that have accrued in the diseased cells , which render them unable to respond properly to steroid hormones . For example , even with GATA2 , the lack of PGR in OSIS cells would still prohibit effective decidualization . Based on its role in erythrocyte differentiation , GATA2 is thought to promote growth and stem-like properties of progenitor cells , and its transcriptional activity is opposed by GATA1 [65] . In this model , GATA2 binds and drives its own promoter , but the induction of GATA1 is able to supplant and inhibit GATA2 directly on the chromatin , creating a “GATA switch . ” It remains to be seen if a similar switch exists in EIUM and OSIS , as we did not examine the occupancy of GATA sites , nor have we demonstrated conclusively that methylation is responsible for maintaining repression of the GATAs in these cells . Likewise , it will be very exciting to determine if GATA isoform expression can affect DNA methylation patterns across the regions we have studied here . However , the idea of an epigenetic switch is provocative , especially for GATA6 , which demonstrates a strikingly inverted pattern of methylation based on β-values . From this perspective , it is also remarkable that GATA2 is expressed in the healthy endometrium; just like the erythrocyte progenitors , a substantial pool of endometrial stromal cells must be maintained between menstrual cycles , suggesting that GATA2 might also be important for ensuring a population of stromal cells is retained each cycle . GATA6 is crucial during the earliest stages of embryogenesis , and serves as a multifaceted differentiator in endoderm- and mesoderm-derived tissues [66]–[68] . Additionally , it directly enhances the ACTA2 promoter [69] , which helps explain the strong ACTA2 signals we observed in vitro , and likely explains some of the ACTA2 typically seen in vivo in endometriotic lesions [70] , [71] . Interestingly , ACTA2 shows a complex cyclic pattern of expression in different layers of the baboon and human endometrium , and it will be interesting to examine if GATA6 expression is coordinated in distinct zones of eutopic endometrium [72]–[74] . GATA6 is also a key regulator of many of the steroidogenic enzymes in the gonad , including aromatase [75] . We and others have shown that steroid signaling is altered in endometriotic cells , where loss of DNA methylation allows the expression of aromatase and other enzymes necessary to synthesize estrogen endogenously , locking them in a progesterone-resistant , estrogen-primed state that promotes their survival and growth [76] , [77] . Our observation that GATA6 increased NR5A1 and CYP19A1 in EIUM suggests that it may be instrumental in allowing endometriotic cells to become steroidogenic . We were very excited to notice that many of the same genes induced by GATA6 are normally repressed by methylation in OSIS . Could the different GATA transcription factors be involved in remodeling the methylation of the genes that drive endometriosis ? During knockdown experiments we noticed that siRNA-mediated loss of the endogenous GATA isoform did not result in increased expression of other isoforms . Similarly , and as mentioned above , the expression of NR5A1 and other nuclear receptors remained unchanged in OSIS after GATA6 knockdown and GATA2 expression , suggesting that their methylation pattern was fixed and/or was coordinated through other mechanisms . This fits with the mounting evidence suggesting that de novo DNA methylation does not initiate gene silencing , but instead occurs secondarily at genes that have already been transcriptionally silenced [78] . In such a scenario repressors and other factors that may in include GATA family members may serve to silence gene expression , and these loci would then subsequently methylated by recruited DNMTs . A possible example is GATA6 which has CGI in its proximal promoter that is not differentially methylated ( Figure 5K ) , and is uniformly unmethylated based on the data from our array . This region is densely populated by conserved transcription factor binding sites that coordinate its tissue-specific expression , and includes several conserved GATA binding sites [45] , [68] . It is plausible that tissue or even disease-specific factors able to bind this region can serve to coordinate the expression and methylation of the downsteam regions . The more important experiment will be to address the methylation status and expression of GATA6 and NR5A1 in the eutopic endometrium of women with endometriosis . This will be challenging , as the methylation defects may be rare occurrences in otherwise healthy stroma; however , our preliminary findings show increased GATA6 and NR5A1 , as well as their downstream targets CYP19A1 and the steroidogenic acute regulatory protein STAR ( [50] , [79]and unpublished observations ) . It will be very interesting to examine where and how GATA2 and GATA6 compete across the promoters of these genes , and if they can differentially affect DNMT recruitment . The significant differences in DNA methylation that we detected for genes such as GATA6 may also be suitable for use as a marker for endometriosis . A conclusive diagnosis of endometriosis can only be made histologically using surgically excised lesions [80]–[82] . Diagnosis is often further complicated as the pain and other symptoms of endometriosis are shared with many other conditions [2] . These problems often cause the definitive diagnosis of endometriosis to lag behind the advent of its symptoms [83] . The ability to use differential DNA methylation as a biomarker is rapidly evolving [84] , and would be immensely useful if developed as a sensitive and minimally invasive test for endometriosis . We anticipate that continued progress along this topic will reveal an epigenetic fingerprint for endometriosis that will shed insight both on the origin of the disease , and also open up new approaches for detecting the disease .
The acquisition of human tissue for this study was approved by the Northwestern Institutional Review Board for Human Research ( 1375–005 ) . Written , informed consent from each subject was obtained before surgery . Normal , eutopic endometrial tissue was obtained from subjects without endometriosis ( average age 42 . 6±5 . 1 years ) undergoing hysterectomy for benign conditions ( cervical dysplasia or uterine leiomyoma ) . Ectopic endometrium from the cyst walls of ovarian endometriomas was obtained immediately after surgery ( average age 41 . 3±4 . 6 years ) . All patients were premenopausal and underwent surgery during the proliferative phase of their menstrual cycle , having received no preoperative hormonal therapy . Endometriosis was confirmed for each sample by histological examination . Enzymes for tissue processing were obtained from Sigma ( St . Louis , MO ) . Cell culture media , trypsin , and supplements were from Gibco ( Life Technologies , Carlsbad , CA ) . Cell plastics were from TPP ( St . Louis , MO ) . Homogenous populations of primary stromal cells were isolated from eutopic endometrial tissue and from endometriotic tissue as previously described [85] , [86] . Briefly , stroma and glandular fragments were dissected from adjacent tissue , minced , and digested with collagenase and DNase at 37°C for 30 min . Samples were then treated with collagenase , DNase , pronase , and hyaluronidase at 37°C for an additional 30 min . Epithelial cells were eliminated by progressive filtration through sterile 70- and 20-µm sieves , and either human endometrial stromal cells ( EIUM ) or endometriotic stromal cells ( OSIS ) were dispensed into 100-mm dishes for adherent growth and maintained in DMEM/F12 supplemented with 10% fetal bovine serum , 100 IU/mL penicillin , 100 µg/mL streptomycin , and 2 . 5 µg/mL amphotericin B . Cells were maintained in grown in a humidified atmosphere with 5% CO2 at 37°C , and medium was replenished every 48 h . In vitro decidualization ( IVD ) regimens followed previous protocols [40] . Briefly , both normal and diseased stromal cells were grown to ∼75% confluency , then switched to phenol red-free DMEM/F12 media supplemented with 2% charcoal-dextran-stripped FBS and antibiotics as above . Controls were maintained in this reduced medium , while IVD treatment consisted of 1 µM medroxyprogesterone acetate ( MPA; Sigma ) , 35 nM 17β-estradiol ( E2; Sigma ) , and 0 . 05 mM 8-bromoadenosine 3′ , 5′-cyclic monophosphate ( BIOLOG Life Science Institute , Bremen Germany ) . Cells underwent 6-day IVD treatments following gene silencing or overexpression . All other IVD treatments were conducted for 10 days . For knockdown experiments , 400 , 000 EIUM/60-mm dish were transiently transfected with Silencer Select siRNAs from Ambion ( Life Technologies ) using Lipofectamine RNAiMAX ( Life Technologies ) . Individual siRNAs targeting GATA2 or GATA6 ( Table S3 ) were tested individually for specificity , and then optimized for transfection as a pair . On the day of transfection , EIUM were trypsinized and resuspended in antibiotic-free DMEM/F12 with 10% FBS , while siRNAs were prepared per the manufacturer's protocol for reverse transcription to yield a final concentration of 50 nM siRNA in 4 . 4 mL of medium/dish ( 0 . 22 nMol of total siRNA in complex with 15 µL of reagent ) . Cells were replaced with complete media 12 hours post-transfection , and IVD treatments were initiated 36 hours post-transfection , and continued for 6 days . For adenoviral transduction , 400 , 000 EIUM/60-mm dish ( or 30 , 000 per well in 24-well dishes ) were infected in complete medium , at an MOI of 20 with adenoviral particles ( Vector Biolabs , Philadelphia , PA ) carrying either enhanced GFP ( AdGFP ) , human GATA2 ( AdGATA2 ) , or human GATA6 ( AdGATA6 ) , each under direction of the CMV promoter . IVD treatments were initiated 24 hours post-infection ( at which point GFP expression was visually confirmed ) , and continued for 6 days . ( For combined transfection/transduction siRNAs were transfected first , transduction followed 12 hours as cells were transferred to complete media , and IVD treatments were started 24 hours after transduction ) . For staining and visualization , stromal cells were cultured on sterile 12-mm , #1 . 5 glass coverslips ( Thermo Fisher Scientific , Hampton , NH ) in 24-well dishes as described above . Cells were fixed in 1x PBS with 4% formaldehyde ( Electron Microscopy Sciences , Hatfield , PA ) , and then either stained with hematoxylin and eosin ( VWR ) , or for individual proteins using indirect immunofluorescence as described previously , but with minor modifications [87] . Briefly , after permeabilization and washing , fixed cells were washed twice with 1x TBS-T ( 20 mM Tris-HCl , 500 mM NaCl , pH 7 . 4 , 0 . 05% Tween-20 ) , and then blocked in TBS with 1% nonfat milk and 1% normal donkey serum ( Jackson Immunoresearch Inc . , West Grove , PA ) . Primary antibodies ( Table S3 ) were prepared in blocking solution ( 1∶50 dilution ) and incubated for 1 h . Secondary antibodies conjugated to Cy-3 ( Jackson Immunoresearch ) were prepared in blocking solution ( 1∶200 dilution ) , and incubated for 1 h . DNA was stained using DAPI , and the coverslips were then washed twice with TBS . Mounted coverslips were examined under brightfield or epifluorescence settings with a Zeiss Axiovert 200 using a 40x LDPlan-NEOFLUAR or a 63x Plan-APOCHROMATIC objective , and images were acquired using an Axiocam HRc . For microarray analyses , total genomic DNA and total RNA from each plate of cells were isolated using AllPrep DNA/RNA columns ( Qiagen , Valencia , CA ) . DNA quality was assessed by visualization following agarose gel electrophoresis . To selectively convert unmethylated cytosine to uracil , one microgram of genomic DNA was subjected to bisulfite treatment using EZ DNA Methylation kits ( Zymo , Orange , CA ) , and either frozen or then followed by isothermal amplification according to the manufacturer's protocol ( Illumina , San Diego , CA ) . The converted genomic DNA was then directly hybridized to Infinium HumanMethylation450 beadchips , and scanned using the Illumina iScan system . Total RNA quality was assessed using an Agilent Bioanalyzer 2100 , and 1 µg of high quality RNA ( RIN>9 ) from each subject was hybridized to HT-12v4 beadchips , and also scanned on the iScan system . Image data were processed in Genome Studio . The analysis of raw data was done with an in-house analysis pipeline described in the Data analysis subsection . For analysis of samples following transfection or transduction , total RNA was isolated using RNeasy columns ( Qiagen ) . Total cDNA was prepared with Q-script cDNA SuperMix ( Quanta Biosciences , Gaithersburg , MD ) [88] . Real-time PCR was performed as previously described , using Power SYBR green or Taqman Universal master mix , on an ABI 7900 ( Applied Biosystems , Foster City , CA , USA ) [89] . Relative gene expression was assessed using TATA-binding protein ( TBP ) as a reference gene . For endogenous expression before and after treatment , fold change was calibrated to the average ΔCT of untreated EIUM . For analysis of samples following transfection or transduction , average fold change was calculated after calibrating to each subject's untreated negative control . Primer information is provided in Table S3 . To validate the methylation status of GATA2 and GATA6 we employed methylation specific PCR using the following primers directed against differentially methylated regions of exon 4 of GATA2 , and exon 2 of GATA6 . Briefly , primers were designed to recognize either the methylated or unmethylated form of the sequence after the bisulphite converted sequences of the “-“ strand . Primer information is provided in Table S3 . Each MSP reaction utilized 25 ng of bisulfite converted DNA from either EIUM or OSIS as template . For controls , we used female genomic DNA that was either fully methylated ( M ) and bisulfite converted or fully demethylated ( U ) and bisulfite converted . Each reaction was carried out in a 20 µL volume containing 500 nM of each primer . Reactions proceeded using a hot start of 95C for 5 minutes , followed by 38 cycle , 3-step reaction using a melt at 95C for 1 min , annealing at 58C for 1 min , and an extension at 72C for 1 min . Amplification products were resolved on 3% agarose gel , and visualized with ethidium bromide under UV light . Whole cell lysates were prepared by washing cells with PBS , followed by lifting and homogenizing the cells in 120 µL RIPA buffer ( 50 mM Tris pH 7 . 6 , 150 mM NaCl , 0 . 1% SDS , 0 . 5% sodium deoxycholate , 1% NP-40 ) supplemented with protease inhibitor cocktail ( Sigma ) . Lysates were cleared by centrifugation at 14 , 000× g for 5 min . Equal amounts of protein ( 20 µg ) were resolved on NuPAGE Novex 4–12% bis-Tris Gels ( Life Technologies ) . Transfer and membrane blocking were performed as previously described [90] . Incubation with primary antibodies ( Table S3 ) was performed at 4°C in 2 . 5% nonfat milk overnight . The membranes were then washed and incubated with the appropriate HRP-conjugated secondary antibodies for 1 h . Detection was performed using Luminata Crescendo HRP substrate ( Millipore ) . We developed a pipeline in R/Bioconductor [91] to integrate the analysis of methylation and mRNA microarrays ( under submission ) . The mRNA data were preprocessed to eliminate probes associated to genes in the X and Y chromosomes . The raw probe intensities were then converted to expression values by firstly applying a variance-stabilizing transformation within each chip and , secondly , by performing a robust spline normalization between chips . The signal preprocessing was conducted in R using the lumi library [92] . Differentially expressed probes were identified between phenotypes ( 6 samples each ) using empirical Bayes correction of linear models provided by the limma library [93] , [94] . The p-values of probes were adjusted for multiple-hypotheses testing using the Benjamini-Hochberg algorithm . Probes with an adjusted p-value less than 0 . 05 were considered differentially expressed . Differentially expressed probes were then mapped to unique transcripts ( RefSeq IDs ) . If multiple probes mapped to the same RefSeq ID , then only the probe with the smallest adjusted p-value was kept . Probes that did not map to a known RefSeq ID were discarded . For methylation data , probes mapping to the X and Y chromosomes or to non-CpGs dinucleotides were excluded . Two-color data from the methylation array were normalized using quantile normalization . Probes were marked as “present” if their detectable probe ratio was greater than 0 . 01 . A probe was discarded if it was present in 4 or fewer samples of the 6 samples for each phenotype . The probe intensities were then converted to β-values . As it was the case for mRNA data , the methylation two-color data were preprocessed with the lumi library . A CpG probe was considered to be differentially methylated if the ( absolute ) difference between the average β-value for one phenotype ( 6 samples ) was greater than 0 . 15 with respect to the average β-value of the other phenotype ( 6 samples ) , i . e . , Δβ>0 . 15 . The average β-values in a phenotype were obtained from the probes marked as present in that phenotype . Differentially methylated CpG probes were linked to differentially expressed transcripts using the definitions in the chip manifests provided by the manufacturer . A Spearman rank correlation coefficient ( ρ ) was obtained between each differentially methylated CpG and its matched differentially expressed transcript , based on the β- values of the CpG and the expression values of the transcript across samples . The sign of ρ was used to classify CpGs as being positively or negatively correlated to the transcripts . For ρ>0 we had two cases: hypermethylation and over-expression , or hypomethylation and under-expression; Likewise , for ρ<0 we had hypermethylation and under-expression , or hypomethylation and over-expression . A test of proportion was conducted on each type of CpG location , stratified either by gene or CGI . The total proportion of negatively over positively correlated CpGs was compared against the proportions of negatively over positively correlated CpGs at each type of location . Two-tailed tests were conducted under the null hypothesis that the proportions at each location were equal to the total proportion . ANOVA interaction analysis was conducted for each differentially expressed transcript on the β-values of all the CpGs that were linked to the transcript . The interaction that was tested was between CpG status ( hypo- vs . hypermethylated ) and CpG location . As it was the case before , two different types of locations were used to determine the interaction with CpG status: location with respect to the transcript or with respect to the closest CGI . The location context for each CpG was taken from the manufacturer's manifest . Importantly , Illumina stratifies the regions flanking CGIs as either “shore” or “shelf” ( depending on proximity ) as well as being either north or south of the island depending on their orientation with respect to the starting coordinate for the chromosome . For simplicity , and because we only wished to stratify these regions relative to genes and CGIs , we merged the shelf and shores into a single category defined as a region 4 kb upstream or downstream of their most proximal CGI . Genes were considered to have a statistically significant interaction if the adjusted p-value for this test was less than 0 . 05 . MetaCore ( version 6 . 13 build 43450 , Thomson Reuters ) was used to perform enrichment analysis workflows on data matched to each target RefSeq ID . The “Enrichment by Protein Function” tool was used to perform over-representation analysis ( ORA ) based on protein class for the genes identified by ANOVA interaction analysis . To identify biological processes uniquely altered during IVD that were likely to be affected by differentially methylated genes , the “Compare Experiments Workflow” tool was used on the ANOVA interaction list and the list of differentially expressed genes in OSIS versus EIUM after IVD . Enrichment analysis was performed on an intersection of both groups' gene ontology processes , and then sorted and ranked where overlap was most similar . For qPCR data , differences across experimental/diseased groups treated with and without IVD were assessed by two-way ANOVA . When no significant interaction was present , a main effect was considered significant for a p-value of less than ( 0 . 05 ) . When an interaction was detected , multiple comparisons were made using Tukey's test . Differences in GATA isoform expression between EIUM and OSIS were made using t-tests .
|
Women develop endometriosis when endometrial tissue with altered sensitivity to ovarian hormones grows outside the uterus . The persistent survival of these cells results in chronic pelvic pain and infertility . Although the origin of the disease remains a mystery , it only occurs in women and menstruating primates , suggesting that the unique evolution behind primate uterine development and menstruation are linked to the disease . Epigenetic defects affecting the uterine physiological response to ovarian hormones are also involved in endometriosis , and several genes implicated in disease progression are differentially methylated . Here we compared DNA methylation with gene expression in endometriosis using large-scale arrays . By comparing healthy and diseased cells treated with or without hormones to mimic part of the menstrual cycle , we uncovered many differentially methylated genes with defective expression in endometriosis that also regulate the hormone-dependent aspects of menstruation . In addition to expanding our understanding of how methylation affects endometriosis many fold , this also led us to propose an epigenetic switch that permits GATA6 expression in endometriosis instead of GATA2 , and this switch promotes the aberrant expression of many of the genes seen in endometriosis . Our work provides novel unifying insight into the cause and development of endometriosis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"developmental",
"biology",
"menstrual",
"abnormalities",
"reproductive",
"system",
"obstetrics",
"and",
"gynecology",
"genetics",
"epigenetics",
"biology",
"dna",
"modification",
"anatomy",
"and",
"physiology",
"sexual",
"reproduction",
"cell",
"differentiation",
"reproductive",
"physiology"
] |
2014
|
Genome-Wide DNA Methylation Analysis Predicts an Epigenetic Switch for GATA Factor Expression in Endometriosis
|
The high environmental adaptability of bacteria is contingent upon their ability to sense changes in their surroundings . Bacterial pathogen entry into host poses an abrupt and dramatic environmental change , during which successful pathogens gauge multiple parameters that signal host localization . The facultative human pathogen Listeria monocytogenes flourishes in soil , water and food , and in ~50 different animals , and serves as a model for intracellular infection . L . monocytogenes identifies host entry by sensing both physical ( e . g . , temperature ) and chemical ( e . g . , metabolite concentrations ) factors . We report here that L-glutamine , an abundant nitrogen source in host serum and cells , serves as an environmental indicator and inducer of virulence gene expression . In contrast , ammonia , which is the most abundant nitrogen source in soil and water , fully supports growth , but fails to activate virulence gene transcription . We demonstrate that induction of virulence genes only occurs when the Listerial intracellular concentration of L-glutamine crosses a certain threshold , acting as an on/off switch: off when L-glutamine concentrations are below the threshold , and fully on when the threshold is crossed . To turn on the switch , L-glutamine must be present , and the L-glutamine high affinity ABC transporter , GlnPQ , must be active . Inactivation of GlnPQ led to complete arrest of L-glutamine uptake , reduced type I interferon response in infected macrophages , dramatic reduction in expression of virulence genes , and attenuated virulence in a mouse infection model . These results may explain observations made with other pathogens correlating nitrogen metabolism and virulence , and suggest that gauging of L-glutamine as a means of ascertaining host localization may be a general mechanism .
Listeria monocytogenes is a Gram-positive facultative intracellular bacterial pathogen and the causative agent of listeriosis in humans , a disease with deleterious impacts , such as increased risk of meningitis and miscarriage [1] . L . monocytogenes invades mammalian cells by expressing surface proteins named internalins , that bind host proteins to induce active bacterial uptake [2] . Upon entry , L . monocytogenes escapes the vacuole ( phagosome ) by producing a pore-forming hemolysin , listeriolysin O ( LLO , encoded by the hly gene ) and two additional phospholipases , PlcA and PlcB [3] , [4] . Once in the host cytosol , L . monocytogenes multiplies rapidly and expresses ActA , which recruits the host actin polymerization machinery to propel the bacteria within the cytosol and facilitate its spread from cell to cell [5] . Most of the known virulence factors involved in internalization , vacuolar escape and cell-to-cell spread are positively regulated by PrfA , the master virulence activator of L . monocytogenes [6] , [7] . Immediately upon infection , L . monocytogenes senses multiple host-derived signals that alert the bacteria of their intracellular localization . Temperature , iron , and the availability of specific metabolites control the transcription , translation , and activity of PrfA and consequently , the induction of virulence genes . For example , carbon sources that are encountered by L . monocytogenes in the soil ( e . g . , glucose and cellobiose ) , have been shown to repress PrfA activity [8] . In contrast , carbon sources encountered in the host ( e . g . , glucose-1-phosphate and glycerol ) are positive regulators of prfA [9]–[11] . The metabolism and virulence of L . monocytogenes are intimately linked via the combined regulation of PrfA and the global metabolism regulator CodY [12] , [13] . The latter , among other things , senses low concentrations of branched-chain amino acids ( BCAAs—isoleucine , leucine , and valine ) , as encountered within mammalian cells , and as a result affects transcription of prfA [14] . Recently , host derived glutathione was shown to serve as yet another intracellular signal that activates PrfA via allosteric binding [15] , [16] . Together , these and other signals reflect the metabolic conditions within the mammalian cell and assist the bacteria in apprising the required metabolic and virulence adaptations . While L . monocytogenes is well equipped to sense the mammalian niche , the host cell also employs sophisticated mechanisms to detect and respond to invading pathogens . For example , macrophage cells respond to L . monocytogenes infection by a robust activation of the type I interferon response , manifested by expression and secretion of the interferon β ( IFN-β ) and interleukin-6 ( IL-6 ) [17] , [18] . This response requires replication of L . monocytogenes within the host cell cytosol and secretion of c-di-AMP and other nucleic acids , demonstrating the existence of a specific sensing machinery for metabolically active bacteria [19] , [20] . These findings establish that both L . monocytogenes and its host have evolved techniques to sense the metabolic state of each other and respond accordingly . In efforts to identify L . monocytogenes novel genes that participate in host sensing and contribute to virulence , we have recently employed an L . monocytogenes transposon mutant library to infect bone marrow derived macrophages [21] . Using a reporter cell line featuring type I interferon-dependent luciferase expression [22] , we screened approximately 5 , 000 L . monocytogenes mutants for decreased induction of IFN-β response . Among the identified mutants were several well-established virulence determinants of L . monocytogenes ( e . g . , PrfA ) , as well as newly identified potential virulence factors [21] . One of these genes , LMRG_02270 , which encodes a polypeptide that is a fusion of a substrate-binding protein ( SBP ) and a transmembrane domain ( TMD ) of an ATP Binding Cassette ( ABC ) transporter , was chosen for further analysis . ABC transporters comprise a large super-family of proteins that couple the energy of ATP hydrolysis to the translocation of molecules across biological membranes against their concentration gradient [23–25] . ABC transporters have been shown to be involved in bacterial virulence and pathogenesis , as they allow the bacteria to acquire essential nutrients from the host [26–32] . In this work , we analyze the function of the L . monocytogenes ABC transporter GlnPQ , and show its importance in interpreting a newly identified signal that induces the expression of virulence genes: L-glutamine .
LMRG_02270 is the first gene of a two-gene operon ( Fig 1A ) , the second of which , LMRG_02271 , contains all of the canonical motifs of a nucleotide-binding domain ( NBD ) of an ABC transporter ( S1 Fig ) . Since both genes together are predicted to consist a functional ABC transport system , a mutant bearing deletions of both LMRG_02270 and LMRG_02271 ( LMRG_02270–1 double mutant ) was prepared and tested for induction of type I interferons response in infected bone marrow derived macrophages ( BMDMs ) . To this end , total mRNA from macrophages infected with either the LMRG_02270–1 mutant or wild-type ( WT ) L . monocytogenes was extracted and IFN-β and IL-6 transcription levels were quantified 6 hours post-infection , using real-time quantitative PCR ( RT-qPCR ) . Indeed , the LMRG_02270–1 mutant elicited a much weaker type I interferon response in comparison to WT bacteria ( Fig 1B and 1C ) , despite identical replication rates between the two bacterial populations ( Fig 1D ) . Collectively , these results demonstrate that LMRG_02770–1 contributes to the induction of the innate Type I interferon immune response to intracellularly growing L . monocytogenes bacteria . In ABC transporters that function as importers , the SBP is located extracellularly , where it recognizes the substrate with high affinity and delivers it to the transmembrane domain . A BLAST analysis of LMRG_02770 against the E . coli K-12 genome , revealed the highest homology to several ABC transporters ( importers ) of charged/polar amino acid , including those of glutamate , aspartate , glutamine , histidine , arginine and lysine . The transport specificity of an ABC importer is almost exclusively determined by the binding specificity of the SBP [33–35] . Therefore , to identify the substrate specificity of the LMRG_02270–1 transporter , a truncated , His-tagged version of the LMRG_02770 SBP domain ( containing only amino acids 29–254 , i . e . without the transmembrane domain ) was cloned , then overexpressed in E . coli and purified to near homogeneity ( S2 Fig ) . Isothermal titration calorimetry ( ITC ) was then applied to determine the binding specificity of the SBP to different amino acids . Of the tested amino acids ( L-isomers of Gln , Glu , Asn , Asp , His , Arg , Lys , Cys , Ser , Thr , and Tyr ) , detectable binding was only observed with L-glutamine , with a dissociation constant ( KD ) of 4 . 7 μM ( Fig 2A ) . To increase the fraction of protonated glutamate we repeated the experiments at pH 6 . 5 ( relative to the initial pH 8 ) , yet still detected no binding . The SBP was also highly stereo-specific as no interaction was detected with D-glutamine ( Fig 2B ) . Using the published X-ray structure of the Enterococcus faecalis L-glutamine SBP ( PDB 4G4P ) , we constructed a structural homology model of the SBP domain of LMRG_02270 . Similar to its homologues , the model of the SBP domain showed the characteristic positioning of the conserved amino acid residues that constitute the substrate-binding site in glutamine binding proteins ( S3 Fig ) . A number of these residues were conserved not only in glutamine binding proteins [36] , but also in SBPs of histidine and arginine transporters [37] . Of these , the highly conserved Arg residue ( R105 in LMRG_02270 ) is responsible for coordinating the backbone carboxy group of the amino acid . Thus , an arginine to alanine ( R105A ) mutation was designed , and indeed found to fully abolish L-glutamine binding ( Fig 2C ) . Collectively , these results suggest that LMRG_02270 and LMRG_02271 form a high-affinity L-glutamine-specific import system in L . monocytogenes . In accordance with the nomenclature used in other bacteria [36] , [38] , the gene encoding the SBP-TMD ( LMRG_02770 ) was designated glnP , and the gene encoding the ATPase ( LMRG_02271 ) was designated glnQ . Like most organisms , L . monocytogenes is unable to metabolize atmospheric nitrogen , and must acquire it in an organic form . Since L-glutamine was shown to be utilized by L . monocytogenes as a major nitrogen source [39] , we hypothesized that a ΔglnPQ mutant strain would show attenuated growth when supplied with glutamine as a sole nitrogen source . When grown in rich media ( brain-heart infusion ( BHI ) ) the growth of the ΔglnPQ and WT strains was very similar ( S4 Fig ) , indicating uncompromised overall fitness of this strain . In contrast , when the ΔglnPQ strain was grown on minimal defined medium ( MDM ) containing 1 mM L-glutamine as the sole nitrogen source , it displayed a 2-fold lower growth rate and a 2-times lower bacterial count in the stationary phase , relative to the WT bacteria ( Fig 3A ) . Even at very high L-glutamine concentrations , the ΔglnPQ mutant failed to grow to the same extent as the WT strain ( Figs 3B and S5 ) . On the other hand , a ΔglnPQ strain complemented with glnPQ genes delivered on the pPL2 plasmid ( ΔglnPQ-pGlnPQ ) , displayed the same growth profile as WT bacteria ( Fig 3B ) . To verify that L-glutamine utilization requires active transport by GlnPQ , the conserved Walker B glutamic acid ( E164 ) , that is essential for ATP hydrolysis and transport activity [33] , [34] , [40] , was mutated to alanine . The glnQ-E164A mutant failed to restore L-glutamine utilization , and its growth was indistinguishable from that of the ΔglnPQ strain ( Fig 3B ) . Taken together , these results indicate high dependency of L . monocytogenes on the GlnPQ transporter for acquisition of L-glutamine under limiting conditions . This dependency was further demonstrated by comparing the uptake rates of 3H-labeled L-glutamine by WT versus ΔglnPQ bacteria . L-glutamine uptake by WT bacteria reached saturation within 30 sec , which was shorter than the temporal resolution of our assay . This rapid plateau of accumulation was observed over a broad range ( 0 . 03–3 μM ) of L-glutamine concentrations ( Fig 4A ) . In contrast , the ΔglnPQ and glnQ-E164A mutant strains displayed practically zero uptake of L-glutamine ( Fig 4A ) , even at concentrations as high as 300 μM , whereas the GlnPQ-complemented strain displayed the same rapid L-glutamine uptake as the WT strain ( Fig 4A ) . These findings confirm that the rapid uptake of L-glutamine measured in the transport experiments is indeed mediated by an active , ATPase-dependent transport process , in which GlnPQ serves as the only high-affinity glutamine import system in L . monocytogenes . This conclusion is supported by a BLAST analysis of the genome of L . monocytogenes that failed to identify additional ( putative ) glutamine importers . Similarly , a bioinformatics study [41] revealed that unlike other Gram positive bacteria , L . monocytogenes harbors a single putative transporter in their nitrogen regulon , which is glnPQ . Next , we measured the concentration gradients ( Glnin:Glnout ) generated by GlnPQ over a broad range of external glutamine concentrations ( 0 . 03–1800 μM L-glutamine ) . Of note , the determination of concentration gradients assumes that L-glutamine is not further metabolized intracellularly . However , even within the short time scale of the transport experiments this is only an approximation . Therefore , the calculated concentration gradients are to be viewed only as estimates . We observed that GlnPQ generated the greatest concentration gradients ( Glnin:Glnout of ~ 60 ) when the external concentration of L-glutamine was < 1 μM . For example , when L-glutamine was added externally at 30 nM , its internal concentration reached ~ 2 μM ( Fig 4B ) . At higher external concentrations of L-glutamine the concentrative ability of GlnPQ gradually decreased: at 200 μM ( external ) L-glutamine the Glnin:Glnout ratio decreased to ~10 , leading to an internal concentration of ~ 2 mM , and a modest 2–3-fold concentration gradient was formed at 1 mM of externally added L-glutamine ( Fig 4B ) . In agreement with the strict binding specificity of the SBP domain ( Fig 2 ) , addition of a 300-fold excess of unlabeled L-Asp , L-Asn , L-Glu , or D-Gln did not affect the amount of 3H L-glutamine that was taken up . Only the addition of unlabeled L-glutamine significantly reduced the amount of accumulated radioactive label ( Fig 4C ) . These results substantiate the high specificity of GlnPQ for L-glutamine , with little to no cross-reactivity with other amino acids . To further study the role of GlnPQ in L . monocytogenes virulence , we first examined the transcription of the hly gene , encoding for Listeriolysin O toxin ( LLO ) , a well-established L . monocytogenes virulence factor , essential for phagosomal escape [3] . To this end , the pPL2 integrative plasmid harboring the hly promoter fused to a luminescent luxABCDE reporter system ( pPL2 hly-lux ) [42] , was conjugated to WT , ΔglnPQ and glnQ-E164A bacteria . Luminescence , indicative of hly transcription , was recorded during 30 hours of growth in MDM supplemented with L-glutamine ( 0 . 25 mM ) as a sole nitrogen source . Under these conditions , a high and sharp luminescence peak was observed in WT bacteria , reaching maximal levels in the middle of the exponential growth phase . In contrast , the clean deletion and the point mutation strains almost completely failed to express the lux genes , indicating no activation of the hly promotor ( Fig 5A ) . Luminescence was then quantified over a broad range of L-glutamine concentrations ( 0 . 02–2 mM ) . Relative to the ΔglnPQ and glnQ-E164A mutants , WT bacteria consistently presented a much higher activity of the hly promoter , reaching half its maximal activity at about 100–200 μM L-glutamine . At these external L-glutamine concentrations the ΔglnPQ and glnQ-E164A mutants displayed almost zero induction of hly ( Fig 5B ) . In complementary RT-qPCR experiments , we similarly observed a dramatic ( ~20-fold ) reduction of hly mRNA transcripts in ΔglnPQ versus WT bacteria ( Fig 5C ) . The ΔglnPQ mutant also exhibited a 5–10-fold reduction in transcription of other major Listerial virulence factors such as: plcA , plcB and actA ( Fig 5C ) . Collectively , these results demonstrate that an active GlnPQ transporter is important for transcription of L . monocytogenes virulence genes . One possible interpretation of the reduced transcription of the virulence genes in the ΔglnPQ strain is that nitrogen starvation restrains their expression . To test this , we sought to satisfy the metabolic need for nitrogen by supplying a nitrogen source other than L-glutamine . As semi-complex nitrogen sources we used a chemically defined media to which we added a tryptone or a peptone peptide digest . For growth in the presence of a defined nitrogen source we used a chemically defined media supplemented with ammonia , arginine , glutamate , or D-glutamine . As previously reported [39] , [43] , [44] , L . monocytogenes grew well when supplied with ammonia as the sole nitrogen source ( S6A Fig ) . However , despite the normal growth , luminescence driven from the hly promoter was very low in both WT and ΔglnPQ mutant , similar to those observed in the ΔglnPQ mutant grown in the presence of L-glutamine ( Fig 5D ) . Similar results were obtained using tryptone or a peptone peptide digest as the sole nitrogen source: WT and the ΔglnPQ bacteria strain grew equally well ( S6B Fig ) , yet in both cases the luminescence was very low ( S6C and S6D Fig ) . This indicates that the ΔglnPQ strain can still take up and utilize nitrogen from short peptides . Unlike short peptides or ammonia , glutamate and arginine proved to be poor nitrogen sources , and could not support growth even when supplied at high ( 10–20 mM ) concentrations . L . monocytogenes could also use D-glutamine as a nitrogen source , and in line with the strict stereo-specificity of GlnPQ ( Fig 2B ) , the ΔglnPQ mutant was not impaired in its ability to use D-glutamine ( S6E Fig ) . As observed with ammonia or short peptides for both the WT and the ΔglnPQ mutant hly transcription was not activated when the nitrogen source was D-glutamine ( Fig 5E ) . Collectively , these results show that nitrogen starvation per se is not the limiting factor for virulence machinery activation , and that L-glutamine is specifically required for the expression of the virulence genes . This conclusion is strongly supported by experiments where nitrogen was supplied in the form of a synthetic L-Gly-L-Gln dipetide . WT and ΔglnPQ bacteria efficiently utilized this di-peptide ( S6F Fig ) , and displayed comparable hly-associated luminescence , that was nearly as high as the luminescence observed in the presence of L-glutamine ( Fig 5F ) . The dependence of virulence genes transcription on the identity of the nitrogen source was also studied in RT-qPCR experiments . As shown , despite the normal growth ( S6A Fig ) , the transcription level of all examined virulence genes ( hly , plcA , plcB and actA ) was lower in the presence of ammonia relative to their transcription levels in the presence L-glutamine ( Fig 6A ) . In accordance , no protein activity of LLO ( encoded by the hly gene ) and PlcA was detected in culture supernatants of WT bacteria grown in the presence of ammonia in comparison to L-glutamine , as well as in supernatants of ΔglnPQ bacteria ( Fig 6B and 6C ) . Since activation of the virulence genes depends on the master virulence regulator PrfA , we examined whether L-glutamine affects its transcript or protein level . For this , WT and ΔglnPQ bacteria were grown in the presence of L-glutamine or ammonia and PrfA’s mRNA and protein levels were measured by RT-qPCR and Western blot analyses . As shown , a higher prfA transcription level was observed in WT bacteria grown with L-glutamine in comparison to ammonia and to ΔglnPQ bacteria ( S7A Fig ) . This increased transcription level was expected as under virulence conditions prfA is also transcribed together with plcA , which was shown to be induced upon L-glutamine uptake ( Fig 6A ) . Nevertheless , the Western blot analysis of PrfA protein demonstrated similar levels of PrfA under all conditions ( S7B Fig ) . While these findings indicate that L-glutamine does not affect PrfA transcription or translation , it is still possible that PrfA is regulated by L-glutamine at the protein level . Higher resolution inspection of the dependence of virulence gene transcription on L-glutamine levels revealed a non-linear and non-hyperbolic , on/off switch-type response ( S8 Fig ) . A Hill coefficient of nHill = 2 . 73 was calculated for the glutamine dependence , indicative of high signal amplification; very low hly induction is observed at low L-glutamine concentrations , and very high hly induction is achieved once a threshold concentration is crossed . These data suggest that the on/off transition occurs at external L-glutamine concentrations of ~100–200 μM . At these external concentrations the pumping activity of GlnPQ brings internal L-glutamine to ~3 mM ( Fig 4B , black curve , left Y-axis ) . For maximal growth , L . monocytogenes requires 2–4 mM external L-glutamine ( Fig 3B ) . However , saturation of hly transcription occurred at 10–20-fold lower concentrations ( S8 Fig ) . Therefore , full manifestation of the virulence gene induction occurs long before full satisfaction of the bacteria’s metabolic needs . To assess the role of GlnPQ during L . monocytogenes infection of BMDMs , macrophage cells were grown in glutamine-restricted medium ( no glutamine was added , and DMEM without glutamine was used ) , and then infected with WT , ΔglnPQ or glnQ-E164A bacteria . The transcription level of the virulence genes was assessed by measuring the transcription level of plcA , using a plcA-yfp transcriptional fusion ( 3 consecutive yfp genes were expressed from the integrative pPL2 plasmid under the regulation of the plcA promoter ) . We monitored YFP fluorescence , and observed reduced plcA expression in ΔglnPQ and glnQ-E164A bacteria in comparison to WT bacteria ( Fig 7A ) . Accordingly , ΔglnPQ and glnQ-E164A bacteria displayed lower infectivity than WT , but only when the macrophages were grown in L-glutamine-restricted medium , relative to when they were grown in L-glutamine-enriched ( 4 mM ) medium ( Compare Fig 7B , restricted , to Fig 1C , enriched ) . These results suggest that L-glutamine sensing is important during early stages of infection . Finally , to determine the contribution of GlnPQ to L . monocytogenes virulence in mice , young female C57BL/6 mice were intravenously injected with 4 × 104 cells of the ΔglnPQ mutant or WT strain , and bacterial counts in the spleens and livers of infected mice were analyzed at 72 h post infection . As shown in Fig 7C and 7D , relative to the WT , the ΔglnPQ mutant colonized the livers and spleens to a lesser extent , exhibiting a 30-fold ( liver ) and a 10-fold ( spleen ) decrease in recovered bacteria . Collectively , these results clearly demonstrate that GlnPQ is necessary to promote L . monocytogenes virulence .
When WT bacteria are supplied with ammonia and glutamate they synthesize L-glutamine ( via the glutamine synthase GlnA [39] ) . Similarly , L . monocytogenes successfully extracts glutamine from a complex mixture of peptides ( peptone or tryptone digest ) . However , in all of these cases , despite the intracellular production of L-glutamine virulence genes are not induced . We think that the distinction between externally supplied and internally produced L-glutamine is achieved via a threshold-based “switch mechanism” . Transcription of virulence genes only starts when the internal concentration of L-glutamine crosses a threshold , which according to our data is ~3 mM . In the absence of external L-glutamine the internally synthesized molecules are incorporated into proteins , utilized in other anabolic processes , or catabolically consumed , never reaching the concentrations that are needed to cross the activation threshold . The threshold can only be crossed via the concentrative action of GlnPQ in WT bacteria that are fed with L-glutamine , or when the bacteria are fed with an L-glutamine containing dipeptide . In retrospect , it makes perfect “biological sense” to use L-glutamine as an environmental signal . Gaseous nitrogen ( N2 ) is the most abundant molecule in the atmosphere ( 78% ) , but can only be exploited by a handful of organisms , primarily soil-residing , ‘nitrogen fixing’ bacteria that express the enzyme nitrogenase . In the soil , nitrogenase catalyzes the reduction of atmospheric nitrogen ( N2 ) to ammonia ( NH3 ) , which is the most abundant nitrogen source in the soil . Amino acids in general , and L-glutamine specifically , are scarce . In stark contrast , ammonia concentrations in the serum and cells of mammals are very low , while those of L-glutamine are very high ( 200–500 μM to several mM ) [50–53] , making L-glutamine a perfect signal for host localization . In this regard it is important to note that within the cytosolic niche L . monocytogenes most likely utilizes multiple nitrogen sources , for example ethanolamine and possibly glucosamine , but whether they also serve as signals for virulence gene activation is currently not known [39] , [54] , [55] . Of note , we were unable to test whether ethanolamine can serve as yet another signal , since L . monocytogenes bacteria could not grow in minimal medium with it as a sole nitrogen source unless glycerol was added , which is itself an inducer of virulence gene expression [10] . While we identified and characterized the L-glutamine transport system , we do not know how L-glutamine is sensed by the bacteria and how the signal is transduced . We found that L-glutamine does not activate prfA transcription or translation , though it is still possible it activates PrfA at the protein level . Notably , the crystal structure of PrfA shows a snugly bound L-glutamine , even though it was not included in the crystallization mixture [56] . Interestingly , the binding site of L-glutamine overlaps with the one proposed for GSH , which was recently shown to be an allosteric activator of PrfA [15] . In addition to PrfA , GlnR , the global nitrogen regulator common to many bacteria , and GlnA ( the glutamate synthase ) may serve as sensors/regulators . Aside from its enzymatic function , GlnA regulates the transcription of the glnRA operon and is also required for the regulatory functions of GlnR [57] . GlnA activity is directly controlled by L-glutamine [58] , [59] , and therefore L-glutamine also indirectly controls the regulatory functions of GlnR . Importantly , both genes have been correlated with virulence in other pathogens [52] , [60] , [61] . In this regard , in L . monocytogenes the global metabolic regulator CodY , that activates transcription of the global virulence regulator PrfA , has been shown to regulate also GlnR [13] . A correlation between nitrogen metabolism and virulence was observed also for other bacterial pathogens , including Staphylococcus aureus [62] , [63] , Streptococcus pneumonia [29] , [57] , [60] , Salmonella typhi [61] , Group B Streptococci [64] , and Mycobacterium tuberculosis [52] . However , it is unknown whether in these pathogens , virulence specifically depends on L-glutamine , or on nitrogen availability in general . The mechanism of L-glutamine sensing , and the generality of the mechanism across different pathogens , are open questions that remain to be addressed .
Experimental protocols were approved by the Tel Aviv University Animal Care and Use Committee ( 01-15-052 , 04-13-039 ) according to the Israel Welfare Law ( 1994 ) and the National Research Council guide ( Guide for the Care and Use of Laboratory Animals 2010 ) . Listeria monocytogenes 10403S was used as the WT strain and as the parental strain to generate allelic exchange mutant strains ( S1 Table ) . E . coli XL-1 Blue strain ( Stratagene ) was used for generation of vectors and E . coli SM-10 strain [65] was used for plasmid conjugation to L . monocytogenes . L . monocytogenes strains were grown in BHI ( Merck ) rich medium , or in minimal defined medium ( MDM ) [66] , [67] . For growth under limiting conditions , MDM without arginine was freshly prepared , supplemented with various concentrations of L-glutamine , NH4Cl , D-glutamine or L-Gly-L-Gln ( Sigma ) . All in-frame deletions and point mutation strains generated in this work were constructed using L . monocytogenes 10403S strain as the parental strain . Upstream and downstream regions of the selected gene were amplified by using Phusion DNA polymerase and cloned into the pKSV7oriT vector [68] . Cloned plasmids were sequenced and then conjugated to L . monocytogenes using the E . coli SM-10 strain . L . monocytogenes conjugants were selected on BHI–agar plates supplemented with chloramphenicol and streptomycin ( 10 μg/ml and 100 μg/ml , respectively ) , then grown at 41°C for two days on BHI–agar plates supplemented with chloramphenicol alone for plasmid integration into the bacterial chromosome by homologous recombination . For plasmid curing , bacteria were passed several times in fresh BHI without chloramphenicol at 30°C to allow plasmid excision by the generation of an in-frame deletion . Bacteria were then plated on BHI plates and chloramphenicol sensitive colonies were picked for validation of gene deletion by PCR using upstream and downstream primers . Complemented strains of deletion mutants were generated by introducing a copy of the deleted gene in trans under the control of its native promoter using the pPL2 integrative vector . For all infection experiments L . monocytogenes strains were grown overnight in BHI medium at 30°C without shaking . Bone Marrow Derived Macrophages ( BMDMs ) used for infection experiments were isolated from 6–8 weeks-old female C57/BL6 mice ( Harlan Laboratories , Israel ) as described previously [69] . BMDMs were cultured in Dulbecco’s Modified Eagle Medium ( DMEM ) - based media supplemented with 20% fetal bovine serum , 1 mM sodium pyruvate , 2 mM L-glutamine , 0 . 05 mM β- Mercaptoethanol , monocyte colony stimulating factor ( M-CSF ) ( L929-conditioned medium ) , and penicillin and streptomycin ( 5 μg ml−1 each ) . The cultures were incubated in a 37°C incubator with 5% CO2 . 1×105 macrophage cells per well in 100 μl of medium were seeded in a 96 well plate overnight . Before infection , macrophages cells were washed twice with phosphate buffer saline ( PBS ) , and fresh medium without antibiotics was added . Approximately 1 . 6×103 L . monocytogenes bacteria resuspended in PBS were used to infect each well . An hour post-infection , macrophage monolayers were washed twice with PBS and fresh medium supplemented with gentamicin ( 50 μg ml−1 ) was added to limit bacterial extracellular growth . At each time point , the media was aspirated from the wells and 100 μl of sterile water were added , followed by vigorous pipetting to release intracellular bacteria . Then serial dilutions of the lysate were plated on BHI-agar plates and Colony Forming Units ( CFU ) were counted after 24 h incubation at 37°C . The transcriptional levels of macrophage genes were analyzed using RT-qPCR . RNA from infected macrophages was extracted using TRIzol reagent according to standard protocols ( Biolab ) . 1 μg of RNA was reverse transcribed to cDNA using a QScript reverse transcription kit ( Quanta ) . RT-qPCR was performed on 10 ng of cDNA using SYBR green with the StepOnePlus RT-PCR system ( Applied Biosystems ) ( see S2 Table for RT-qPCR primers ) . The transcription levels of macrophage cytokine genes were normalized using glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) as a reference gene , and the uninfected cells as a reference sample . The gene encoding the SBP portion of GlnPQ from Listeria monocytogenes 10403S ( first 254 amino acids of LMRG_02770 ) , but without its N-terminal signal sequence ( first 28 amino acids ) , was synthesized and adjusted to the E . coli codon usage ( Genescript ) . The gene was cloned into pET-19b ( Novagen ) vector for expression with an N-terminal His-tag [70] , [71] . His-tagged GlnPQ SBP was overexpressed in E . coli BL21-Gold ( DE3 ) cells ( Stratagene ) cultured in Terrific Broth ( TB ) and induced at mid log phase by addition of 1 mM Isopropyl b-D-1-thiogalactopyranoside ( IPTG ) for 1 . 5 h at 37°C . Cells were harvested by centrifugation ( 13 , 600 × g , 20 min , 4°C ) and the pellet was stored at -80°C until use . For purification , cells were homogenized in 50 mM Tris-HCl pH 8 , 500 mM NaCl , complete EDTA-free protease inhibitor ( Roche ) , 30 mg ml-1 DNase ( Worthington ) , and 1 mM MgCl2 . The cells were then ruptured by three passages in an EmulsiFlex-C3 homogenizer ( Avestin ) , and the lysate centrifuged at 34 , 500 × g for 30 min at 4°C . The supernatant was loaded onto a nickel affinity column ( HisTrap HP , GE Healthcare ) on an AKTA Avant instrument . The protein was eluted using an imidazole gradient , and imidazole was eliminated from the sample by desalting ( HiPrep 26/10 , GE Healthcare ) . Protein purification was monitored by coomassie staining of SDS-PAGE and size exclusion chromatography ( Superdex 75 10/300 GL , GE Healthcare ) . The R105A point mutation was introduced to the GlnPQ SBP by the QuikChange Lightning site directed mutagenesis kit ( Agilent Technologies ) and confirmed by sequence analysis . The mutant protein was overexpressed and purified as described above for the wild type GlnPQ SBP . Calorimetric measurements were performed with Microcal iTC200 System ( GE Healthcare ) . Prior to measurement , the protein was dialyzed against three exchanges of 50 mM Tris-HCl pH 8 , 500 mM NaCl buffer . Amino acid stocks were prepared fresh in double-distilled water and diluted to working concentration using the buffer from the last protein dialysis exchange . 2 μL aliquots of 500 μM amino acid were added by a rotating syringe to the reaction well containing 200 μL of 50 μM WT or mutant GlnPQ SBP at 25°C [28] . Data fitting was performed with the Microcal analysis software . For luminescence assays , L . monocytogenes strains harboring the Phly-luciferase reporter system ( pPL2-Phlylux ) were used . Bacteria from overnight cultures grown in BHI medium were adjusted to OD600 of 0 . 05 in fresh BHI or MDM containing various concentrations of L- glutamine , D-glutamine , NH4Cl or tryptone digest , and 150 μL were transferred into a clear flat bottom white 96-well plate . The plate was incubated in an InfiniteM200 pro ( Tecan ) at 37°C for a period of 12–40 hours , during which luminescence and OD600 were measured . Bacteria from overnight cultures grown in BHI medium were adjusted to OD600 of 0 . 05 in fresh BHI and grown at 37°C to an OD600 of 0 . 4 . The cultures were harvested at 4°C , washed in ice-cold standard phosphate-buffer saline ( pH 7 . 4 ) supplemented with 1 . 6 mM MgSO4 , re-suspended to an OD600 of 15 and kept on ice until further analysis . Then , glucose ( 1% w/v ) was added to a 50 μl aliquot , and the cells were allowed to recover for 10 min at 37°C . The transport assay was initiated by the addition of 2 μl of a solution containing the desired concentration of non-labelled L-glutamine and 0 . 1 μCi of L-[2 , 3 , 4-3H] glutamine ( 60 Ci/mmol ) ( American Radiolabeled Chemicals , Inc . ) . For competition assays , the solution contained also the competing amino acid to a working concentration of 900 μM . After incubation at 37°C for the indicated times , 3 ml of ice-cold assay buffer were added and the sample was immediately filtered through a moistened 2 . 4-cm diameter glass microfiber filter ( GF/C; Whatman ) . The filter was then washed with an additional 3 ml of ice-cold assay buffer , dried and placed in a scintillation vial . 3 ml of Ultima Gold F scintillation fluid ( Perkin Elmer ) was added , and radioactivity was measured the following day on a β-counter . For all calculations of internal concentrations and concentration gradients we assumed that 100% of the accumulated label reports on glutamine , and that it has not been further metabolized during the transport experiment . The radioactive counts of several known concentrations of 3H-Gln were determined , including the stock solution that was later used in the uptake experiment . This was compared to the amount of radioactivity that was taken up by the cells . This amount was converted to concentrations assuming a cellular amount of 5·108 cells per 1mL of OD 1 and a cellular volume of 1 μm3 . For RNA extraction , bacteria from overnight cultures were adjusted to OD600 of 0 . 05 in 20 ml of fresh MDM containing 4 mM of L-glutamine or NH4Cl as the sole nitrogen source . The cultures were incubated with agitation at 37°C to an OD600 of 0 . 3–0 . 4 . Bacteria were harvested by centrifugation , washed with ice-cold PBS and stored frozen at -80°C until further analysis . Total nucleic acids were extracted by the RiboPure RNA Purification Kit ( Ambion ) . 1 μg of purified RNA was reverse transcribed to cDNA using the high-capacity cDNA reverse transcription kit ( Applied Biosystems ) . RT-qPCR was performed on 10 ng of cDNA using PowerUp SYBR green master mix ( Applied Biosystems ) and 500 nM forward and reverse primers , designed with the Primer3web software ( version 4 . 0 . 0 ) in the Rotor-Gene 6000 ( Qiagen ) system . The transcription level of each gene of interest was normalized to that of a reference gene , rpoD . Statistical analysis was performed using the Rotor-Gene Q series software . RT-qPCR primers are described in S2 Table . LLO hemolytic activity assay was performed as described previously [46] , [72] . Bacterial supernatants were treated with 5 mM DTT , serially diluted in PBS and incubated at 37°C with 0 . 5% of sheep blood red cells suspension ( NovaMed Ltd . ) in 35m M sodium phosphate buffer pH = 5 . 5 , 125 mM NaCl , 0 . 5 mg/ml BSA . Cells were then removed by centrifugation at 1000 g and hemolysis was estimated as optical density at 541 nm . PlcA PI-PLC activity assay was adopted from [73] , 0 . 05 gr phosphatidyl-inositol ( Sigma P6636 ) were mixed by sonication with 10 ml of 0 . 2% sodium deoxycholate , 1 mM CaCl2 , 114 mM ( NH4 ) SO4 and 40 mM Tris-HCl pH = 7 . 2 . 100μl of the assay solution was then mixed with 100 μl of bacterial supernatant and incubated in plate reader at 37C for 10 h , with continuous detection of turbidity at 510 nm . Lm bacteria were grown in MDM containing 4 mM of L-glutamine or ammonia as the sole nitrogen source at 37°C and harvested at a 0 . 3–0 . 4 OD . Then the bacteria were lysed by incubation with 5 units of mutanolysin ( M9901; Sigma ) for 1 hour at 37°C , followed by sonication . Cell debris were removed by centrifugation at 3000 g . Protein concentrations were quantified by modified Lowry method [74] . Samples with equal amounts of total proteins were separated on 12 . 5% SDS-PAGE , electro-blotted and probed either with rabbit anti-PrfA antibodies ( made in this study , Almog Diagnostics , dilution 1:2500 ) , followed by HRP-conjugated goat anti-rabbit IgG ( Jackson ImmunoResearch , USA ) , respectively . Anti-GroEL antibodies ( a kind gift from A . Azem , Tel Aviv University ) were used as a loading control at a dilution of 1:20 , 000 , followed by HRP-conjugated goat anti-rabbit IgG . Western blots were developed by enhanced chemiluminescence reaction ( ECL ) . WT and mutant strains of L . monocytogenes expressing three consecutive YFP proteins under the regulation of the plcA gene promoter in a pPL2 integrative plasmid were used to infect BMDMs on 20 mm slides . Upon infection , cells were cultured in BMDM medium with no added glutamine ( traces of glutamine may exist due to addition of SCF to the medium ) . Four hours post infection , cells were fixed with 4% v/v paraformaldehyde-PBS and permeabilized with 0 . 05% v/v Triton X-100 . DNA was stained with DAPI containing Vectashield mounting media ( Vector laboratories inc . ) . Images were taken using Zeiss LSM 510-META confocal microscope . L . monocytogenes bacteria were grown in BHI medium at 30°C overnight without shaking . Bacterial cultures were washed twice with PBS . BALB/c ( 6–8 weeks old ) female mice ( Harlan Laboratories ) were infected via tail vein injections with 2 . 7×104 bacteria in 200 μl PBS . Each group consisted of 5 mice for every mutant tested . Animals were observed daily for any signs of illnesses and were euthanized 72 h post-infection . Spleens and livers were harvested and homogenized in 0 . 2% v/v saponin . The numbers of viable bacteria in each organ were determined by plating serial dilutions of homogenates onto BHI agar plates .
|
To thrive in the human body and cause disease , bacterial pathogens rely on the expression of a battery of genes , termed virulence genes . However , to activate these genes , the bacterium must “know” that it is no longer free-roaming , and has invaded a host . Like all life forms , bacterial pathogens must acquire the four elemental building blocks ( carbon , nitrogen , phosphate , sulfur ) . The sources for these building blocks in the bacterial host are different from those present in the environment . It therefore makes perfect “biological sense” for bacteria to use such nutrients as localization beacons . We show for the first time , that the human pathogen Listeria monocytogenes surveys the available nitrogen sources before turning on its virulence genes . L-glutamine , an abundant nitrogen source in humans , greatly induces the expression of all major virulence genes . In contrast , a nitrogen source that is prevalent in soil and water , such as ammonia , fails to do so . To better treat infectious diseases , it is vital that we understand how bacteria switch to their virulent , disease-causing state . Our results provide a significant step in this direction .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"pathogens",
"immunology",
"microbiology",
"organic",
"compounds",
"electromagnetic",
"radiation",
"dna",
"transcription",
"acidic",
"amino",
"acids",
"amino",
"acids",
"bacterial",
"pathogens",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"chemistry",
"luminescence",
"physics",
"listeria",
"monocytogenes",
"biochemistry",
"ammonia",
"cell",
"biology",
"organic",
"chemistry",
"virulence",
"factors",
"genetics",
"glutamine",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"macrophages"
] |
2017
|
L-glutamine Induces Expression of Listeria monocytogenes Virulence Genes
|
LJ001 is a lipophilic thiazolidine derivative that inhibits the entry of numerous enveloped viruses at non-cytotoxic concentrations ( IC50≤0 . 5 µM ) , and was posited to exploit the physiological difference between static viral membranes and biogenic cellular membranes . We now report on the molecular mechanism that results in LJ001's specific inhibition of virus-cell fusion . The antiviral activity of LJ001 was light-dependent , required the presence of molecular oxygen , and was reversed by singlet oxygen ( 1O2 ) quenchers , qualifying LJ001 as a type II photosensitizer . Unsaturated phospholipids were the main target modified by LJ001-generated 1O2 . Hydroxylated fatty acid species were detected in model and viral membranes treated with LJ001 , but not its inactive molecular analog , LJ025 . 1O2-mediated allylic hydroxylation of unsaturated phospholipids leads to a trans-isomerization of the double bond and concurrent formation of a hydroxyl group in the middle of the hydrophobic lipid bilayer . LJ001-induced 1O2-mediated lipid oxidation negatively impacts on the biophysical properties of viral membranes ( membrane curvature and fluidity ) critical for productive virus-cell membrane fusion . LJ001 did not mediate any apparent damage on biogenic cellular membranes , likely due to multiple endogenous cytoprotection mechanisms against phospholipid hydroperoxides . Based on our understanding of LJ001's mechanism of action , we designed a new class of membrane-intercalating photosensitizers to overcome LJ001's limitations for use as an in vivo antiviral agent . Structure activity relationship ( SAR ) studies led to a novel class of compounds ( oxazolidine-2 , 4-dithiones ) with ( 1 ) 100-fold improved in vitro potency ( IC50<10 nM ) , ( 2 ) red-shifted absorption spectra ( for better tissue penetration ) , ( 3 ) increased quantum yield ( efficiency of 1O2 generation ) , and ( 4 ) 10–100-fold improved bioavailability . Candidate compounds in our new series moderately but significantly ( p≤0 . 01 ) delayed the time to death in a murine lethal challenge model of Rift Valley Fever Virus ( RVFV ) . The viral membrane may be a viable target for broad-spectrum antivirals that target virus-cell fusion .
Advances in antiviral therapeutics have allowed for effective management of specific viral infections , most notably human immunodeficiency virus ( HIV ) [1] . Yet , the one-bug-one-drug paradigm of drug discovery is insufficient to meet the looming threat of emerging and re-emerging viral pathogens that endangers global human and livestock health . This underscores the need for broad-spectrum antivirals that act on multiple viruses based on some commonality in their viral life cycle , rather than on specific viral proteins . Recently , a few broad-spectrum antivirals have been described that target enveloped virus entry [2] , [3] , [4] , [5] , [6] or RNA virus replication [7] , [8] , [9] , [10] . The former targets the viral membrane , or more precisely , the biophysical constraints of the virus-cell membrane fusion process , while the latter targets nucleic acid metabolic pathways . LJ001 is a membrane-binding compound with broad-spectrum antiviral activity in vitro . LJ001 acts on the virus , and not the cell , inhibiting enveloped virus infection at the level of entry [4] . LJ001 is non-cytotoxic at antiviral concentrations , yet had the remarkable property of inhibiting all enveloped viruses tested , including those of global biomedical and biosecurity importance such as HIV , hepatitis C virus ( HCV ) , Influenza , Ebola , henipaviruses , bunyaviruses , arenaviruses and poxviruses . LJ001 is also clearly not virolytic and does not act as a “detergent”: LJ001-treated virions remain intact and their viral envelopes functional , as LJ001-treated virions are still able to bind to their receptors . A panoply of assays showed that even though LJ001 was lipophilic , and could bind to both viral and cellular membranes , it inhibited virus-cell but not cell-cell fusion . This puzzling dichotomy was illuminated when studies with lipid biosynthesis inhibitors indicated that LJ001 was indeed cytotoxic when the ability of a cell to repair and turnover its membranes is compromised . Thus , we posited that the antiviral activity of LJ001 relies on exploiting the physiological difference between inert viral membranes and biogenic cellular membranes with reparative capabilities [4] . However , the molecular target of LJ001 remains to be defined , and a precise molecular mechanism that could explain the extraordinary breadth of LJ001's antiviral activity against lipid-enveloped viruses is lacking . This has limited consideration of the viral membrane as a plausible target for the development of broad-spectrum antivirals . Here , we identify the molecular target of LJ001 and present a strong body of evidence that supports a unifying hypothesis regarding its mechanism of action . Based on this mechanistic understanding , structure-activity relationship ( SAR ) optimization resulted in a new class of membrane-targeted broad-spectrum antivirals with markedly enhanced potencies and other relevant biophysical and pharmacokinetic properties that underscore the veracity of our mechanism of action ( MOA ) hypothesis . Finally , we validated our hypothesis in vivo by interrogating the efficacy of this new class of membrane-targeted antivirals against a virulent ( enveloped ) viral pathogen in a lethal challenge animal model .
To further define the molecular mechanism of LJ001's antiviral activity , we first investigated where LJ001 acts during the fusion cascade . A time-of-addition experiment , schematically shown in Figure S1 , indicated that LJ001 inhibited the HIV fusion cascade at a step subsequent to CD4-receptor binding and pre-hairpin intermediate ( PHI ) formation ( Figure 1A ) . Thus , the inhibitory half-life of LJ001 was longer than that of a CD4 blocking antibody ( Leu3A ) and T-20 , a heptad-repeat ( HR ) -derived peptide that targets the PHI and prevents six-helix bundle formation ( 6-HB ) [11] . LJ001 similarly inhibited Nipah virus ( another Class I fusion protein ) envelope mediated entry [12] , although in this case , the resolution of our assay couldn't distinguish between PHI and 6-HB formation ( Figure 1B ) . These results suggest LJ001 acts late in the fusion cascade , likely after PHI formation . LJ001 also acts late in the Class II fusion protein cascade , as we found that it did not affect homotrimer formation of the Semliki forest virus ( SFV ) E1 protein ( Figure 1C ) , even at concentrations that completely inhibited virus fusion ( Figure S2 ) . Class II E1 homotrimer formation is analogous to six-helix bundle ( 6-HB ) formation for Class I fusion proteins and marks a late step in the fusion cascade [13] , [14] . These data confirm that LJ001 inhibits both Class I and II fusion , highlight that LJ001 abrogates viral infectivity while maintaining the conformational integrity of the viral envelopes , and demonstrate that LJ001 inhibits fusion at a very late stage , likely just prior to virus-cell membrane merger . Lipid composition can affect the biophysical properties of viral membranes that impact the efficiency of virus-cell fusion . Insect cells are cholesterol auxotrophs and can be grown in the absence of sterols , and thus , SFV can be generated with or without cholesterol in viral membranes . The sensitivity of SFV to LJ001 did not differ significantly between viruses grown in the presence or absence of cholesterol ( Figure 2A ) , suggesting that cholesterol is not a membrane component essential for LJ001's antiviral activity . To determine if LJ001 affected the phospholipid composition of viral membranes , we treated influenza virus A ( A/PR/8/34 H1N1 ) with LJ001 or its inactive analog , LJ025 [4] , and analyzed the viral lipidome by mass spectrometry after liquid chromatography separation ( LC-MS ) . No difference was observed in the overall phospholipid composition of treated viruses ( Figure 2B ) . However , high-resolution LC-MS spectral analysis revealed that LJ001-treated viruses had up to 300-fold increase in the number of oxidized forms of unsaturated phospholipids , compared to LJ025-treated samples ( Figure 2C and Figure S3 ) . To rule out other virus-specific or virion-associated co-factors , we used liposomes with a defined phospholipid composition , and showed that LJ001 could mediate the specific and direct oxidation of linoleic acid ( 18∶2 ) ( Figure 2D ) , an unsaturated fatty acid present in viral and cellular membranes [15] , [16] , [17] . Reactive oxygen species such as singlet oxygen ( 1O2 ) are known to react readily with carbon-carbon double bonds ( alkenes ) present in the acyl chains of unsaturated phospholipids , and this process would generate the oxidized phospholipids described in Figures 2C–D . To evaluate the capacity of LJ001 to generate 1O2 , we added LJ001 to 9 , 10-dimethylanthracene ( DMA ) , a specific 1O2 trap , and quantified the oxidation of DMA by 1H-NMR ( Figure 3A and Figure S4 ) . LJ001 , but not LJ025 , exhibited 1O2-mediated oxidation of DMA , which was decreased by the antioxidant α-tocopherol ( α-toco ) and absent when molecular oxygen was replaced by argon ( Ar ) . Correspondingly , the ability of LJ001 to inhibit multiple viruses was abrogated not only by the addition of a lipophilic antioxidant ( α-toco ) or 1O2 quencher ( DMA ) , but also by a water-soluble 1O2 quencher ( NaN3 ) ( Figure 3B ) . Thus , we hypothesized that LJ001's antiviral activity is attributable to its properties as a type II photosensitizer [18] , [19] , a compound that generates highly reactive excited-state 1O2 by transferring energy of the excited sensitizer to ground-state ( triplet ) molecular oxygen ( 3O2 ) . Our hypothesis predicts that as a photosensitizer , LJ001's antiviral activity should also be dependent on light . Indeed , the antiviral activity of LJ001 was dependent on both its concentration and the time-of-exposure to white light . For example , doubling the time of light exposure achieved the same viral inhibitory effect at ten-fold lower concentrations ( Figure 3C , compare 50 and 500 nM curves ) . Importantly , LJ001's antiviral activity was absent when no visible light source was used ( Figure 3D ) . Since LJ001 membrane intercalation is dictated by its lipophilic properties and not the presence of light , this latter observation underscores our previous observations [4] that , at the active concentrations used , membrane insertion itself does not account for the antiviral activity of LJ001 . Finally , to provide independent confirmation of the type II photosensitizing properties of LJ001 , we subjected a solution of LJ001 in CD2Cl2 under ambient conditions to flash excitation , and observed the characteristic 1O2 emission in the near-infrared ( Figure S5 ) . We propose that after insertion into the viral membrane , light activation of LJ001 triggers the generation of 1O2 that oxidizes the unsaturated chains of fatty acids composing the phospholipids of the viral membrane . In further support of our model , we showed that LJ001 ( and LJ025 ) efficiently partitions into model lipid membranes mimicking the lipid packing density , fluidity , and composition of viral ( HIV-like ) or cell ( POPC ) membranes ( Figure 4A and Table S1 ) . Indeed , when lipid membranes were non-limiting ( >50-fold molar excess of lipid ) , over 85% of LJ001 or LJ025 were protected from the water-soluble quencher ( acrylamide ) , and thus , completely buried in the lipid bilayer ( Figure S6 ) . 1O2-mediated oxidation of unsaturated phospholipids proceeds by a “singlet oxygen ene” reaction , resulting in a cis-to-trans isomerization of a double bond in the unsaturated fatty acids and the presence of a polar group ( hydroperoxy- or hydroxy- ) in the hydrophobic core of the lipid bilayer ( Figure S7 , first and second panel ) . Cis-to-trans isomerization allows for closer packing of the fatty acid acyl chains in the lipid bilayer , which could result in a tighter positive curvature , while lipid oxidation results in clustering of the oxidized lipids into microdomains , reducing exposure of the polar groups to the hydrophobic acyl chains in the lipid bilayer core ( Figure S7 , third and fourth panel ) [20] . The latter effectively reduces membrane average fluidity ( and/or increases rigidity ) , as lipid species are now not as freely diffusible . Indeed , surface pressure and steady-state fluorescence anisotropy measurements indicated that LJ001 induced tighter lipid packing ( Figure 4B–C ) , and reduced membrane fluidity ( Figure 4D–E ) of various model lipid monolayers , significantly more than LJ025 . These effects were especially prominent using HIV membrane-like mixtures . Importantly , LJ001 did not show an effect on lipid packing when not exposed to light ( “dark” in Figure 4B–C ) , and neither compounds affected membrane fluidity when tested on biogenic cellular membranes ( primary peripheral blood mononuclear cells ( PBMC ) obtained from blood donors , Figure 4D–E ) . The former confirms that membrane insertion alone does not account for the change in membrane biophysical properties mediated by LJ001 , and the latter is consistent with our prior observations [4] that LJ001 damages inert viral membranes but not biogenic cellular membranes . In light of our elucidation that LJ001 acts as a lipophilic photosensitizer , the explanatory mechanism becomes clear: cells have multiple endogenous cytoprotection mechanisms against phospholipid hydroperoxides [21] that can overcome the oxidative damages done by LJ001 to cellular membrane lipids , whereas viral membranes have no such reparative capacity to guard against LJ001-mediated oxidative damage . In toto , these data indicate that LJ001 is a light-activated membrane-intercalating photosensitizer that catalyzes 1O2-mediated lipid oxidation of unsaturated phospholipids; this results in changes to the biophysical properties of the viral membrane that negatively impacts its ability to undergo virus-cell fusion . Having established that the broad-spectrum antiviral activity of LJ001 was due to its properties as a membrane-targeted photosensitizer , we sought to increase its antiviral potency by structure-activity relationship ( SAR ) experiments . LJ001 is a rhodanine derivative; rhodanines are derivatives of thiazolidines , such as the 5-membered ring on the left hand side of LJ001 ( Figure 5A ) . In order to maximize the absorption , and perhaps also shift the peak absorption ( λmax ) to longer tissue-penetrating wavelengths , we decided to investigate other ring systems analogous to the thiazolidine unit of the rhodanines . In particular we wanted to change the sulfur atom in the ring to a smaller atom , e . g . , nitrogen or oxygen to perhaps have better electronic overlap . While the imidazolidine ( nitrogen in the ring ) analogues had essentially no activity ( data not shown ) , we found that the oxazolidine analogues ( oxygen in the ring ) had superior activity . We therefore carried out a small SAR study of the 5- ( 5-arylfurfurylidene ) -2-thioxooxazolidin-4-one and the analogous 5- ( 5-arylfurfurylidene ) oxazolidine-2 , 4-dithiones ( see Text S1 ) that led us to an oxazolidine-2 , 4-dithione we named JL103 ( Figure 5A ) . Although it was still inactive against a non-enveloped virus ( Adenovirus serotype 5 , Ad5 ) , JL103 maintained the broad-spectrum activity of LJ001 against enveloped viruses—from all three classes of fusion proteins—with at least a 10-fold increase in potency ( Figure 5B and Figure S8 ) . JL103 was also mechanistically similar to LJ001 ( Figure S9 and Tables S1 and S2 ) : ( i ) it remained a membrane-targeted photosensitizer and its antiviral activity still required the presence of light , ( ii ) its antiviral activity could be reduced by antioxidants , and ( iii ) it acted on a similarly late stage of the HIV fusion cascade , but likely with a better efficiency than LJ001 at the same concentration . However , we noted a few differences that were mechanistically illuminative: the addition of a somewhat polar but uncharged substituent ( methoxy ) to the right-hand phenyl ring in JL103 decreased its partitioning into membranes ( Table S1 ) ; nevertheless , JL103's ability to generate 1O2 at a higher rate than LJ001 ( Figure S9 and Table S2 ) indicates that this increased quantum yield is the dominant factor that contributes to the enhanced antiviral potency of JL103 . Analysis of JL103's photophysical properties indicated that its absorption spectrum was red-shifted ( Figure 5C; λmax , LJ001 = 455 nm , JL103 = 515 nm ) , and that the total integrated absorption ( AUC ) within the optical spectrum ( λ = 400 to 750 nm ) was 1 . 53 times that of LJ001 ( Table S2 ) . Flash excitation of a solution of JL103 in CD2Cl2 under ambient conditions also resulted in the characteristic 1O2 emission in the near-infrared ( data not shown ) , confirming that JL103 is a bona fide 1O2 generator . However , compared to LJ001 , JL103 had improved 1O2 quantum yields ( QY ) at both 355 and 532 nm ( Table S2 ) . These results confirm that JL103 is more efficient in generating 1O2 than LJ001 [18] , [19] . Consequently , under the same conditions , JL103-treated liposomes had significantly more oxidized lipids than LJ001-treated liposomes ( Figure 5D ) , implicating the enhanced photosensitizing properties of JL103 in its increased antiviral potency . Of note , these photosensitizers have relatively small rates of 1O2 removal ( kT , Table S2 ) indicating that self-quenching of 1O2 by the photosensitizer-drug was not significantly limiting their antiviral function . Oxazolidine-2 , 4-dithiones ( e . g . JL103 ) are novel non-rhodanine compounds that are more potent inhibitors of virus-cell fusion than the rhodanine derivatives ( e . g . LJ001 ) we previously characterized as broad-spectrum antivirals [4] . Despite the increased potency and enhanced photosensitizing properties of JL103 , we thought it unlikely that JL103 ( λmax = 515 nm ) would exhibit antiviral activity neither in vivo nor in the common use of photosensitizers for whole blood or packed red blood cells ( RBC ) decontamination , known as Pathogen Reduction Technology ( PRT ) , as the hemoglobin present in molar excess would compete effectively for any incident photons with wavelengths <600 nm [19] . To confirm the competitive effect of hemoglobin , we tested the antiviral efficacy of JL103 in the presence of increasing amounts of human RBC . Indeed , the antiviral efficacy of JL103 was inversely proportional to the hematocrit ( Hct ) , and at physiological Hct ( ∼45% RBC v/v ) , the antiviral activity of JL103 was reduced by >50% ( Figure 6A ) . To rule out that this reduction in antiviral activity was not simply due to competition by the increasing amount of RBC membranes , we performed a second SAR study with the aim of developing new oxazolidine-2 , 4-dithiones with even more red-shifted absorption spectra . We hypothesized that compounds with equivalent 1O2 quantum yields , but with absorption spectra that extend beyond ∼600 nm , would maintain the potency of JL103 even at physiological hematocrits . The structures of the new JL compounds ( oxazolidine-2 , 4-dithiones ) are given in Figure S10 and their antiviral activity ( IC50 ) , cytotoxicity to primary PBMCs ( CC50 ) , and therapeutic indexes ( TI ) in Table S3 . We generated a series of active oxazolidine-2 , 4-dithiones by modulating the electron-donating nature of the substituents on the right-hand phenyl ring . Thus , JL108 ( 4-methoxy ) , JL109 ( 2 , 4-dimethoxy ) , JL122 ( 2 , 4 , 6-trimethoxy ) , and JL118 ( 4-dimethylamino ) were all as potent as JL103 , if not more , when tested against a representative panel of enveloped viruses ( Table S3 ) . Interestingly , these compounds exhibited increasingly red-shifted absorption spectra with λmax ranging from 530 ( JL108 ) to 550 ( JL109 ) , 545 ( JL122 ) , and 610 ( JL118 ) nm ( Figure S11 and Table S2 ) ( note: λmax for LJ001 and JL103 is 455 and 515 nm , respectively ) . All these compounds were also confirmed to be 1O2 generators with equivalent or greater quantum yields when compared to JL103 ( Table S2 ) . We chose to follow-up on JL118 and JL122 ( Figure 6B ) as they represent different classes of phenyl substituents ( dimethylamino versus methoxy ) , and were both at least as potent as JL103 in their antiviral activity , but had red-shifted absorption spectra beyond those of JL103 and hemoglobin ( Figure 6C ) . Indeed , in contrast to JL103 , and consistent with our hypothesis , JL118 and JL122 maintained their antiviral potency at physiological hematocrits ( Figure 6D ) . These results provide independent confirmation that the negative correlation seen in Figure 6A , between the antiviral activity of JL103 and Hct , was not simply due to the presence of extra RBC membranes , but indeed resulted from the hemoglobin competing for incident photons . JL118 and JL122 still insert into membranes , as indicated by their partitioning into membranes ( Table S1 ) , with Kp values between those of LJ001 and JL103 . As the addition of somewhat polar but uncharged substituents ( methoxy or dimethylamino ) to the phenyl ring may also improve the solubility and bioavailability of the compounds , we evaluated the pharmacokinetics of candidate compounds . Indeed , JL103 , JL118 and JL122 all exhibited >10-fold improvements in relevant pharmacokinetic ( PK ) parameters compared to LJ001 ( longer half-life , better AUC , improved bioavailability and lower clearance , see Figure 6E and Table S4 ) . Thus , we evaluated their potential antiviral activity in a stringent lethal challenge model of Rift valley fever virus ( RVFV ) , where the median lethal dose ( LD50 ) was ≤1 pfu ( plaque forming unit ) ( Figure S12 ) . In mice lethally challenged with 20×LD50 of RVFV , treatment with JL118 or JL122 resulted in a moderate but significant delay in time-to-death compared to untreated controls ( Figure 6F ) . As expected , treatment with JL103 had no significant effect on survival ( Figure S12 ) , indicating that the absorption spectrum of the compound plays a critical role in its antiviral activity in vivo . Furthermore , even at a higher challenge dose ( 50×LD50 ) , JL122 treatment still resulted in a significant delay in time-to-death when compared to JL103 treatment ( Figure 6G ) , suggesting that the red-shifted absorption spectra of JL122 and JL118 likely accounts for their improved antiviral activity in vivo compared to JL103 . Recall that JL103 , JL118 and JL122 all had similar PKs and in vitro IC50 values against diverse species of enveloped viruses ( Figure 6E and Tables S3 and S4 ) .
LJ001 was previously reported to be a small molecule broad-spectrum antiviral that targets entry of lipid-enveloped viruses [4] . Despite careful characterization of LJ001's antiviral properties , the molecular target and mechanistic basis for the broad-spectrum activity of LJ001 remained elusive . Here , we identify the unsaturated fatty acid chains of viral membrane phospholipids as the major targets of LJ001's antiviral activity . Furthermore , we not only confirmed that LJ001 insertion into membranes is necessary but not sufficient for its antiviral activity [4] , but also provided evidence for a unifying mechanistic hypothesis that accounts for the broad-spectrum antiviral activity of LJ001 against enveloped viruses . LJ001 acts as a membrane-targeted photosensitizer: the phospholipid modifications , resulting from the light-dependent LJ001-induced 1O2-mediated lipid oxidation , negatively impact on the fine-tuned biophysical properties of viral membranes critical for productive virus-cell membrane fusion ( e . g . by increasing membrane curvature and/or decreasing fluidity ) . Thus , the photosensitizing properties of LJ001 mediate its antiviral activity . Our proposed mechanism of action provides an explanatory basis for our observation that while LJ001 can clearly bind to both cellular and viral membranes , it is not cytotoxic to cells at antiviral concentrations unless the ability of the cell to repair its membranes is compromised [4] . This mechanism is consistent with our model that LJ001's antiviral activity exploits the inability of static viral membranes to repair LJ001-mediated damage , and also explains why this class of broad-spectrum antivirals affects virus-cell , but not cell-cell fusion [4] . Indeed , the effects of oxidized phospholipids on the biophysical properties of membranes ( Figure 4 ) are only apparent on viral membranes , and not on biogenic cellular membranes ( e . g . PBMCs ) , which are subject to repair , turnover , and translocation processes . These latter mechanisms have evolved to mitigate the negative effects posed by oxidized phospholipids [21] . Our mechanistic model for LJ001's mode of action was further confirmed by SAR experiments . We developed a new class of membrane-targeted broad-spectrum antivirals where , as hypothesized , the enhanced antiviral activity was correlated with improved 1O2 quantum yields , and more favorable photochemical and photophysical properties . These improvements overcame some of the limiting barriers that previously restricted the in vivo antiviral efficacy of this class of photosensitizers . Indeed , in proof-of-principle studies , we showed that JL118 and JL122 , from the new JL-series of membrane-targeted photosensitizing compounds , not only were more effective at inactivating HIV in the presence of a large excess of RBC ( i . e . hemoglobin ) , but also moderately , yet significantly , prolonged the time-to-death in a lethal challenge model of RVFV . Importantly , the demonstrated ex vivo and in vivo antiviral efficacy of JL118 and JL122 compared to JL103 provides functional validation of our SAR strategy , and is consistent with the panoply of in vitro assays that supports our model for the molecular mechanism that underlies the broad-spectrum antiviral activity of our novel series of membrane-targeted photosensitizers . Photosensitizers have been used clinically in many forms of photodynamic therapy . The majority of photosensitizers in clinical use focus on their ability to damage nucleic acids or proteins . There is also a large literature on membrane-targeted photosensitizers; many of them are porphyrin derivatives . Benzoporphyrin derivative monoacid ring A ( BPD-MA ) is a photosensitizer that has long been known to be a virucidal agent in vitro [22] . Remarkably , verteporfin , another BPD , was recently evaluated as an agent in extracorporeal photopheresis in HIV-infected patients , and shown to have a significant impact on viral load in a subset of patients that underwent an extended treatment course [23] , [24] . Due to logistical and practical considerations , photodynamic therapy to reduce viral pathogen load is unlikely to be an efficient application for chronic infections like HIV . However , our JL compounds with absorption spectra that are red-shifted beyond that of hemoglobin may warrant further evaluation of their use in PRT for transfusion medicine [25] . For example , whereas testing and PRT for blood products using photosensitizers are common in developed countries , they remain , as currently constituted , expensive and unaffordable in resource-poor countries , where blood-borne pathogens transmissions during transfusions is still present at unacceptable rates . Thus , the identification , development and testing of more affordable photosensitizers that can sustain greater variability in quality control processes are highly desirable . Incidentally , our experiments showing that JL118 and JL122 still maintained effective antiviral activity even at high hematocrits , and in the presence of just white ambient light , may provide proof-of-principle of this application . To our knowledge , despite the large literature on membrane-targeted photosensitizers and many claims as to their use as virucidal agents , no one has precisely identified the molecular mechanisms by which specific membrane-targeted photosensitizers inhibit virus-cell fusion [26] . In addition , the putative anti-viral activity of photosensitizers such as Hypericin and Rose Bengal , Hypocrellin A , Methylene Blue derivatives or Phthalocyanines , to name a few , has always been examined at concentrations at least 2 logs higher than what we have used for JL118 and JL122 , and their antiviral activity generally attributed to singlet oxygen's , or other ROS' , effects on proteins and/or nucleic acids [27] , [28] , [29] , [30] , [31] . Herein , we elucidated the molecular and biophysical mechanisms that underlie the antiviral activity of a well-known class of compounds: membrane-intercalating photosensitizers . In so doing , we generated a novel class of such compounds ( oxazolidine-2 , 4-dithione derivatives ) with effective nM IC50s , and showed that improving the relevant photophysical and photochemical properties can lead to increased antiviral efficacy . An exciting future prospect is to conjugate our lead compounds to lanthanide doped “upconversion” organic nanocrystals , which can absorb at deep tissue penetrating near infrared ( NIR ) wavelengths ( >900 nm ) and emit light at visible wavelengths [32] , [33] , [34] . The nitrogen on thiazolidine ring of LJ001 can tolerate many different substituents without loss of antiviral activity [4]; the nitrogen on the oxazolidine ring of JL118 and JL122 is likely suited for such conjugation purposes . Thus , an enhanced understanding of the precise molecular mechanism of action can guide the proper development of membrane-targeted photosensitizers as broad-spectrum antivirals . Taken together , this study suggests that targeting the physiological differences between virus and cell membranes represents a novel therapeutic antiviral strategy worthy of further investigation . Another class of membrane targeted broad-spectrum antivirals ( termed Rigid Amphipathic Fusion Inhibitors , RAFIs ) was described shortly after our original publication of LJ001 by St Vincent et al . [5] . The authors reasonably contend that the “inverted-cone” shape of RAFIs ( with respect to a larger hydrophilic headgroup ) impairs the positive-to-negative curvature transition that is critical for productive membrane fusion , a well-known property of other inverted cone-shaped molecules such as lysophospholipids [35] . However , it is also hard to attribute the nanomolar antiviral activity of RAFIs entirely to their lipid binding properties and changes to their molecular geometry , given the molar excess of cellular membranes in any viral-cell infection assay [36] , [37] . Although RAFIs are nucleoside derivatives with no chemical relation to LJ001 or the JL series of compounds , the hydrophobic group , perylene , present in effective RAFIs has a structure closely related to hypocrellin A , a well-known photosensitizer belonging to the family of quinones [36] , [38] . It will be of interest to determine if the potential photosensitizing properties of active RAFIs could contribute to their antiviral activity . In summary , thorough characterization of the mechanism of action and SAR optimization of LJ001 led to a new class of membrane-targeted photosensitizers ( oxazolidine-2 , 4-dithiones ) with increased potencies , 1O2 quantum yields , and red-shifted absorption spectra . Altogether , these improved properties resulted in membrane-targeted photosensitizers with encouraging in vivo antiviral efficacy against a lethal emerging pathogen . In light of our current study , the substantial literature on the in vivo use of photosensitizers [19] in the photodynamic therapy ( PDT ) of cancer should be re-examined for its applicability in the development of membrane-targeting broad-spectrum antivirals against lipid-enveloped viruses . Potentially , the most effective oxazolidine-2 , 4-dithiones could be evaluated as new candidate drugs in the photodynamic treatment of cancer .
All procedures and animal studies were in accordance with the National Research Council ( NRC ) Guide for the Care and Use of Laboratory Animals ( 1996 ) and/or approved by the Institutional Animal Care and Use Committee ( IACUC ) at the University of Texas Medical Branch ( UTMB ) and performed at the Robert E . Shope biosafety level 4 ( BSL-4 ) laboratory . Methodological details and PK results are further provided in Text S1 . The overall synthetic scheme for the JL series and structures of selected LJ and JL compounds are detailed in Text S1 and Figure S10 , respectively . The absorbance spectra of compounds were determined on a monochromator-based Tecan Infinite M-1000 PRO by continuous scanning ( λ = 250–800 nm ) in absorbance mode using 100 µM of compound in 100 µl DMSO . Viral strains used , determination of IC50 , and virus inhibition assays in the presence of red blood cells are indicated in Text S1 . All assays were performed at or above biosafety levels corresponding to the risk group of the agents and NIH requirements . Phospholipid species , liposome compositions , and fluorescent membrane probes are indicated in Text S1 . Partition and acrylamide quenching studies were carried out using a Varian Cary Eclipse fluorescence spectrophotometer . Excitation and emission wavelength of LJ001 and LJ025 used were described in [4] . Excitation and emission spectra were corrected for wavelength-dependent instrumental factors [39] , emission was also corrected for successive dilutions , light scattering [40] and simultaneous light absorption by quencher and fluorophore ( inner filter effect ) . Membrane partition studies were performed with LUV by successive additions of small amounts of lipid systems , including pure POPC and HIV membrane-like mixture ( POPC 5 . 3% , DPPC 3 . 5% , cholesterol 45 . 3% , SM 18 . 2% , POPE 19 . 3% and POPS 8 . 4%; mol % [15] ) , to 50 µM LJ001 or LJ025 solutions , with 10 min incubation between each addition . The partition coefficients ( Kp ) were calculated from the fit of the experimental data with [41] , [42]: ( 1 ) where IW and IL are the fluorescence intensities in aqueous solution and in lipid , respectively , γL the molar volume of the lipid [43] , and [L] the lipid concentration . Quenching of LJ001 or LJ025 by acrylamide [44] was studied in buffer and in the presence of POPC ( LUV ) as described elsewhere [44] , [45] and in Figure S6 . The changes of the surface pressure of lipid monolayers induced by LJ001 or LJ025 were measured in a Langmuir-Blodgett trough ST900 at constant temperature ( 25 . 0±0 . 5°C ) . The surface of an HEPES buffer solution contained in the Teflon trough was exhaustively cleaned by aspiration . Then , a chloroform solution of lipids was spread on this surface to reach surface pressures between 22 and 29 mN/m . At each chosen surface pressure , molecules solutions were injected in the subphase and the changes on the surface pressure were followed during time to reach a constant value . 3 mM LUV of POPC or HIV-like mixture prepared as described for partition assays were incubated with DPH or TMA-DPH to achieve a final probe concentration of 0 . 33 mol% ( relative to the lipid ) . Steady-state anisotropy 〈r〉 was calculated using: ( 4 ) where Ivv and Ivh represent the fluorescence intensities obtained with vertical excitation polarization and vertical or horizontal orientations of emission polarizers respectively . G = Ihv/Ihh is a correction factor accounting for the polarization bias in the detection system . DPH probe: excitation 350 nm , emission 452 nm . TMA-DPH probe: excitation 355 nm , emission 430 nm . Peripheral blood mononuclear cells ( PBMC ) obtained as described elsewhere [42] were incubated at 3×106 cells/ml in buffer with 2 . 5 µM of DPH or TMA-DPH , during 30 min , with gentle stirring . The <r> values obtained for control PBMC using DPH and TMA-DPH ( 0 . 302±0 . 016 and 0 . 317±0 . 055 , respectively ) are in a good agreement with reference values obtained in a previous works [46] . Fluorescently labeled PBMC were then incubated with LJ001 or LJ025 during 1 h , with gentle agitation , before the fluorescence anisotropy measurements , conducted as indicated above . 1O2 quantum yields ( QY ) and quenching rate constants were determined using a time-resolved set-up ( Nd∶YAG Minilase II , New Wave Research Inc . ) , excitation pulse duration 4–6 ns at 355 nm and 5–7 ns at 532 nm , and a liquid N2 cooled Ge photodetector ( Applied Detector Corporation Model 403 S ) . Details of the filters used have been described elsewhere [47] . Signals were digitized on a LeCroy 9350 CM 500 MHz oscilloscope and analyzed using Origin software . All experiments were carried out at ambient temperature and in air-saturated solutions . UV-visible spectra were recorded on a Cary 300 Bio Spectrophotometer ( Varian ) . Samples were prepared in deuterated methylene chloride ( CD2Cl2 ) with absorbances between 0 . 04–0 . 3 at 355 nm or 532 nm . The laser pulse energy was 1–2 . 5 mJ at 355 nm and 3–4 mJ at 532 nm . The absorbance of the reference sensitizer ( Rose Bengal , TPP and C60 ) and the series compounds were matched within 80% . The initial 1O2 intensity was extrapolated to t = 0 . Data points of the initial 0–5 µs were not used due to electronic interference signals from the detector . The quenching rates ( kT ) of 1O2 were measured by Stern–Volmer analysis using C60 as sensitizer at 355 nm in methylene chloride . Concentration of the samples used in the measurements ranged between 0 . 01–1 mM . Briefly , lipid oxidation in recombinant unilamellar liposomes ( 7∶3 molar ratio of phosphatidylcholine∶cholesterol , >60% linoleic acid ) untreated or treated with 10 µM compounds and light was determined on extracted lipids by LC-MS/MS analysis , as previously described [48] . The transitions monitored were mass-to-charge ratio ( m/z ) : m/z 295→194 . 8 for 13-HODE; 295→171 for 9-HODE; and 299→197 . 9 for 13-HODE-d4 . Methodological details are further provided in Text S1 . Viral lipidome analysis was performed on lipids extracted from Influenza A virus ( A/PR/8/34 H1N1 ) treated with 5 µM of LJ001 or the negative control LJ025 , exposed to light for 1 h as described [49] , [50] .
|
The threat of emerging and re-emerging viruses underscores the need to develop broad-spectrum antivirals . LJ001 is a non-cytotoxic , membrane-targeted , broad-spectrum antiviral previously reported to inhibit the entry of many lipid-enveloped viruses . Here , we delineate the molecular mechanism that underlies LJ001's antiviral activity . LJ001 generates singlet oxygen ( 1O2 ) in the membrane bilayer; 1O2-mediated lipid oxidation results in changes to the biophysical properties of the viral membrane that negatively impacts its ability to undergo virus-cell fusion . These changes are not apparent on LJ001-treated cellular membranes due to their repair by cellular lipid biosynthesis . Thus , we generated a new class of membrane-targeted broad-spectrum antivirals with improved photochemical , photophysical , and pharmacokinetic properties leading to encouraging in vivo efficacy against a lethal emerging pathogen . This study provides a mechanistic paradigm for the development of membrane-targeting broad-spectrum antivirals that target the biophysical process underlying virus-cell fusion and that exploit the difference between inert viral membranes and their biogenic cellular counterparts .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicinal",
"chemistry",
"virology",
"chemistry",
"chemical",
"reactions",
"biology",
"antivirals",
"microbiology",
"photochemical",
"reactions"
] |
2013
|
A Mechanistic Paradigm for Broad-Spectrum Antivirals that Target Virus-Cell Fusion
|
Leishmania uses the amino acid L-arginine as a substrate for arginase , enzyme that produces urea and ornithine , last precursor of polyamine pathway . This pathway is used by the parasite to replicate and it is essential to establish the infection in the mammalian host . L-arginine is not synthesized by the parasite , so its uptake occurs through the amino acid permease 3 ( AAP3 ) . AAP3 is codified by two copies genes ( 5 . 1 and 4 . 7 copies ) , organized in tandem in the parasite genome . One copy presents the expression regulated by L-arginine availability . RNA-seq data revealed 14 amino acid transporters differentially expressed in the comparison of La-WT vs . La-arg- promastigotes and axenic amastigotes . The 5 . 1 and 4 . 7 aap3 transcripts were down-regulated in La-WT promastigotes vs . axenic amastigotes , and in La-WT vs . La-arg- promastigotes . In contrast , transcripts of other transporters were up-regulated in the same comparisons . The amount of 5 . 1 and 4 . 7 aap3 mRNA of intracellular amastigotes was also determined in sample preparations from macrophages , obtained from BALB/c and C57BL/6 mice and the human THP-1 lineage infected with La-WT or La-arg- , revealing that the genetic host background is also important . We also determined the aap3 mRNA and AAP3 protein amounts of promastigotes and axenic amastigotes in different environmental growth conditions , varying pH , temperature and L-arginine availability . Interestingly , the increase of temperature increased the AAP3 level in plasma membrane and consequently the L-arginine uptake , independently of pH and L-arginine availability . In addition , we demonstrated that besides the plasma membrane localization , AAP3 was also localized in the glycosome of L . amazonensis promastigotes and axenic amastigotes . In this report , we described the differential transcriptional profiling of amino acids transporters from La-WT and La-arg- promastigotes and axenic amastigotes . We also showed the increased AAP3 levels under amino acid starvation or its decrease in L-arginine supplementation . The differential AAP3 expression was determined in the differentiation of promastigotes to amastigotes conditions , as well as the detection of AAP3 in the plasma membrane reflecting in the L-arginine uptake . Our data suggest that depending on the amino acid pool and arginase activity , Leishmania senses and could use an alternative route for the amino acid transport in response to stress signaling .
Leishmania is the causative agent of leishmaniasis , a complex disease characterized by cutaneous , mucocutaneous or visceral lesions [1–3] . It is currently endemic in 98 countries and territories around the world , with an annual incidence estimated at 1 million cases of cutaneous leishmaniasis and 300 , 000 cases of visceral leishmaniasis [4] . In its life cycle , the parasite alternates between the intestinal tract of the sand fly ( promastigote form ) and the phagolysosome compartment of mammalian host macrophages ( amastigote form ) . These environmental changes submit the parasite to undergo extensive modifications and can be considered a trigger for the gene expression regulations that lead to adaptation to the new milieu . Leishmania infection results in the specific activation of mammalian immune responses . Macrophages have a fundamental role in infection as the first line of defense . The production of nitric oxide ( NO ) , a potent molecule effective against pathogens , is one of the key defense mechanisms of mammalian phagocytes [5] . NO is produced by nitric oxide synthase 2 ( NOS2 ) using the amino acid L-arginine as substrate . Several studies have demonstrated that immune responses to infectious pathogens are strictly dependent on the expression of NOS2 [1 , 5 , 6] . Arginase is an immune-regulatory enzyme that can reduce the NO production by macrophages , limiting L-arginine availability to NOS2 and inducing resistance of some pathogens to host defense mechanisms [7–10] . In contrast , Leishmania also has arginase , which uses L-arginine to produce urea and ornithine as part of polyamine biosynthetic pathway , essential for the parasite replication and establishment of the infection [8 , 10–15] . The L-arginine uptake in macrophages is mediated by the cationic amino acid transporter family ( CAT ) , such as cationic amino acid transporter 2B ( CAT2B ) in Leishmania infection [16] . By contrast , Leishmania has a selective uptake of this amino acid by AAP3 [17 , 18] . L . amazonensis AAP3 is encoded by two gene copies ( 5 . 1 and 4 . 7 aap3 ) , arranged in tandem on the genome . The open reading frame ( ORF ) sequence shows 98% identity between the two copies and 93% identity to the AAP3 of L . donovani ( LdAAP3 ) [18] . Previous studies have demonstrated that L . donovani responded to amino acid starvation by increasing both mRNA and protein levels of the L-arginine transporter LdAAP3 [17 , 19] . Castilho-Martins et al . ( 2011 ) described an increase in the 5 . 1 aap3 mRNA half-life in L . amazonensis [18] . Leishmania also has mechanisms that sense both external and internal concentrations of L-arginine and can respond with an increase in amino acid uptake [17 , 18] . Therefore , L-arginine uptake control might be an important factor in the resistance of some pathogens to host defense mechanisms [14 , 20] . RNA-seq data analyses revealed 14 amino acid transporters differentially expressed in La-WT and La-arg- promastigotes and axenic amastigotes in the stationary growth phase . The differentiation from promastigotes to axenic amastigotes , independent of arginase activity , showed up-regulation of some these transporters , which may be involved in the Leishmania polyamine biosynthetic pathway . In fact , the 5 . 1 and 4 . 7 aap3 transcripts were down-regulated in axenic amastigotes when compared to the promastigote expresson . Up-regulation of other transporters was also identified , as the amino acid transporter aATP11 , suggesting that Leishmania senses the amino acid pool and regulates gene expression to use alternative route for parasite survival . In this work , we showed that changes in pH , temperature as well as L-arginine availability and the background from the host macrophage canmodulated the aap3 mRNA expression and the AAP3 protein amount . We also showed the AAP3 plasma membrane localization correlated with the arginine uptake in La-WT mid-logarithmic growth phase promastigotes . The change conditions aimed to simulate the environmental changes of the parasite in its life cycle from the sand fly to the mammalian host . Furthermore , we demonstrated that in addition to its plasma membrane localization , AAP3 was also localized partially in the glycosome of promastigotes and axenic amastigotes forms of the parasite , indicating that arginine uptake can be directed to this compartmentalized organelle , supplementing the polyamine production .
L . amazonensis ( MHOM/BR/1973/2269 ) wild type ( La-WT ) and L . amazonensis arginase knockout ( La-arg- ) [8] promastigotes were grown at 25°C in M199 medium supplemented with L-glutamine , 10% heat-inactivated fetal calf serum , 0 . 25% hemin , 12 mM NaHCO3 , 100 μM adenine , 40 mM HEPES , 50 U/mL penicillin and 50 μg/mL streptomycin , at pH 7 . 0 . La-WT and La-arg- axenic amastigotes were grown in M199 medium supplemented , as described above , at 34°C and pH 5 . 5 [21 , 22] . La-arg- cultures were grown in M199 supplemented as described above with hygromycin ( 30 μg/mL ) , puromycin ( 30 μg/mL ) and putrescine ( 50 μM ) addition . For in vitro macrophages infection , bone marrow-derived macrophages ( BMDMs ) from BALB/c or C57BL/6 mice were derived from the femurs and tibiae of females ( 6-8-weeks ) from the Animal Center of the Institute of Bioscience of the University of São Paulo . The femurs and tibiae were washed with cold PBS and the cells were collected at 500 x g for 10 min at 4°C . After lysis of erythrocytes , the cells were maintained in RPMI 1640 medium ( LGC Biotecnologia , São Paulo , SP , Brazil ) , supplemented with penicillin ( 100 U/ml ) ( Life Technologies , Carlsbad , CA , USA ) , streptomycin ( 100 μg/ml ) ( Life Technologies , Carlsbad , CA , USA ) , 5% heat-inactivated FBS ( Life Technologies , Carlsbad , CA , USA ) and 20% L9-29 supernatant . The cells were cultivated for 7 days at 34°C and 5% CO2 . After differentiation , cellular viability was evaluated with Trypan blue staining 1:1 , and cells were counted in Neubauer chamber . Approximately 2x105 BMDMs were incubated on sterile 8 wells glass chamber slide ( Lab-Teck Chamber Slide; Nunc , Naperville , IL , USA ) , overnight at 34°C and 5% CO2 to adhere to the coverslips . Non-adherent cells were removed by PBS washing . THP-1 human monocytic cell line was maintained in culture at the same conditions for BMDMs . Differentiation was performed plating 5×105 cells in 8 wells chamber slide with 30 ng/mL of phorbol 12-myristate 13-acetate ( PMA ) ( Sigma-Aldrich , St Louis , MO , USA ) diluted in RPMI 1640 medium for 72 h , followed by a 72-h resting phase with fresh RPMI 1640 medium before infection . The infection was performed with La-WT or La-arg- stationary growth phase promastigotes ( MOI 5:1 ) . After 4 h of infection , non-phagocytized parasites were washed with PBS and the cells were collected after 4 , 24 and 48 h . Non-infected macrophages maintained in culture in the same conditions were used as control . The infections were evaluated by determining the percentage of infected cells after counting 200 Panoptic-stained ( Laborclin , Parana , Brazil ) macrophages . The infection index was determined by multiplying the percentage of infected macrophages by the mean number of parasites per infected cell [23 , 24] . Statistical analyses were performed using non-parametric two-tailed Student t tests . Total RNA of BMDMs from BALB/c or C57BL/6 mice , and THP-1 derived macrophage infected with La-WT and La-arg- promastigotes; and La-WT promastigotes in different conditions of temperature , pH and amino acid starvation or L-arginine supplementation were isolated using TRIzol reagent ( Life Technologies , Carlsbad , CA , USA ) , according to the manufacturer’s instructions . RNA samples were treated with DNase I ( Thermo Scientific , Lithuania , EU ) and RNA concentration was determined using a spectrophotometer at A260/A280 ( Nanodrop ND1000 , Thermo Scientific , USA ) . For RNA-seq , total RNA from La-WT and La-arg- promastigotes and axenic amastigotes in the stationary growth phase were isolated using TRIzol reagent ( Life Technologies , Carlsbad , CA , USA ) , according to the manufacturer’s instructions . RNA samples were treated with DNase I ( Thermo Scientific , Lithuania , EU ) . Then , RNA concentration was determined using a spectrophotometer at A260/A280 ( Nanodrop ND1000 , Thermo Scientific , USA ) . In addition , RNA integrity was assessed using Agilent 2100 Bioanalyzer and Pico Agilent RNA 6000 kit ( Agilent Technologies , Santa Clara , CA , USA ) , according to the manufacturer’s instructions . Reverse transcription was performed using 2 μg of total RNA as a template , reverse transcriptase and random primers ( Revertaid H minus Reverse Transcriptase kit , Thermo-Scientific , Canada ) , according to the manufacturer’s instructions . Equal amounts of cDNA were assessed in triplicate in a total volume of 25 μL containing Maxima SYBR Green qPCR Master Mix ( Thermo Scientific , Lithuania , EU ) and the following primers ( 200 nM ) : AAP3_F ( 5 . 1 UTR ) 5´-GGTCCCCGATACACACATTC-3´ , AAP3_R ( 5 . 1 UTR ) 5´-GTCTCCCGTTTTGCAAGAGA-3´ , AAP3_F ( 4 . 7 UTR ) 5´-ACCATTGTGGGTTAGTTATACATCC-3´ , AAP3_R ( 4 . 7 UTR ) 5´-CAAGATCGC TAGCAGTGGAG-3´ , GAPDH_Leishmania_F 5´-TCAAGGTCGGTATCAACGGC-3´ and GAPDH_Leishmania_R 5´-TGCACCGTGTCGTACTTCAT-3´ . The mixture was incubated at 94°C for 5 min , followed by 40 cycles at 94°C for 30 s and 60°C for 30 s . A negative control in the absence of reverse transcriptase was included in RT-qPCR assays to detect DNA contamination in RNA samples . Reactions were carried out using an Exicycler 96 ( Bioneer , Daejeon , Korea ) . The copy number of the target genes ( aap3 5 . 1 and aap3 4 . 7 ) and reference gene ( gapdh ) were quantified in three biological replicate samples , considering the molar mass concentration , according to a standard curve generated from a ten-fold serial dilution of a quantified and linearized plasmid containing the target fragment for each quantification test . The normalized aap3/gapdh ratio of the absolute number of molecules of each target was used as the parameter to calculate the relative expression . Analyses were performed using Analysis Exicycler3 Software ( Bioneer , Daejeon , Korea ) . cDNA library preparations were performed using Stranded-specific TrueSeq RNA-seq Library Prep ( Illumina ) , according to the manufacturer´s instructions . Paired-end reads ( 125 bp ) were obtained using the Illumina HiSeq 2000 platform at the Norwegian Sequencing Centre at the University of Oslo . Trimmomatic was used to remove the Illumina adapter sequences [25] . The quality of the produced data was analyzed using FastQC by Phred quality score [26] . Reads with Phred quality scores lower than 20 were discarded . Reads were aligned to the L . mexicana ( MHOMGT2001U1103 ) genomic data obtained from TriTrypDB ( tritrypdb . org/tritrypdb/ ) using TopHat [27 , 28] . Thereafter , read mapping was performed for transcript assembly using Cufflinks [29] . After assembly , the abundance of transcripts was calculated as the Fragments Per Kilobase of transcript per Million mapped reads ( FPKM ) , which reflects the abundance of a transcript in the sample by normalization of the RNA length and the total read number [30] . Differentially expressed gene analysis was performed on four comparison pairs ( La-WT promastigotes vs . La-arg- promastigotes; La-WT axenic amastigotes vs . La-arg- axenic amastigotes , La-WT promastigotes vs . La-WT axenic amastigotes; La-arg- promastigotes vs . La-arg- axenic amastigotes ) [22] . Promastigotes in mid-logarithmic growth phase ( day 4 of culture ) were washed with Earl´s Salt Solution ( EBSS ) ( LGC Biotecnologia , São Paulo , SP , Brazil ) at pH 5 . 0 or pH 7 . 0 . Then , cells were starved of amino acids or supplemented with 400 μM L-arginine for 4 h at 25 or 34°C . The control parasites were those collected before starvation and/or L-arginine supplementation [18] . Arginine uptake assays were performed after amino acids starvation or L-arginine supplementation , as previously described [31 , 32] . Briefly , 5x107 promastigotes in the mid-logarithmic growth phase were washed twice with EBSS medium , resuspended in PBS and incubated at 25°C or 34°C for 3 min . Then , 3H-arginine ( 1mCi/43Ci/mmol ) ( GE Healthcare , UK ) was added . The uptake was stopped at different times by adding ice cold arginine . The parasites were washed twice with EBSS and the radioactivity was measured by liquid scintillation spectrometry Perkin-Elmer TRI-CARB 2910TR . The epitope ( ILYNFDPVNQP ) designed for a specific region of AAP3 through a high affinity MHC was synthesized and used to produce a rabbit anti-AAP3 polyclonal antibody by Proteimax Biotechnology ( São Paulo , SP , Brazil ) . Approximately 107 parasites in the different conditions were washed with PBS and then lysed with lysis buffer ( 100 mM Tris-HCl pH 7 . 5 , 2% Nonidet P40 , 1 mM PMSF and protease inhibitor cocktail ( Sigma-Aldrich , St Louis , MO , USA ) ) . Cells were disrupted by five freeze/thaw cycles in liquid nitrogen and 42°C , and then were cleared of cellular debris by centrifugation at 12 , 000 x g for 15 minutes at 4°C . Equal amounts of total protein ( 50 μg ) were solved using SDS-PAGE and then transferred to a nitrocellulose membrane ( LI-COR Bioscience , Lincoln , NE , USA ) using a Trans-Blot Semi-Dry apparatus ( Bio-Rad , USA ) . The membrane was incubated with Blocking Buffer ( LI-COR Bioscience , Lincoln , NE , USA ) and then with anti-AAP3 serum ( 1:500 dilution ) , overnight , at 4°C . After incubation with primary antibody , the membrane was incubated with goat anti-rabbit DyLight 680 conjugated antibody ( Thermo Scientific , IL , USA ) ( 1:10000 dilution ) for 1 h at room temperature . Anti-α-tubulin ( Sigma-Aldrich , St . Louis , MO , USA ) ( 1:5000 dilution ) was used to normalize the amount of protein in the blot . All steps were followed by washing 3 times with PBS . The membranes were scanned using an Odyssey CLx apparatus ( Li-COR , Lincoln , NE , USA ) in 700 channel using an Odyssey System . Odyssey Imaging CLx instrument was used at an intensity setting of 5 ( 700 nm ) . Approximately 106 promastigotes on days 3 , 5 , 7 and 9 of a growth curve , or in the mid-logarithmic growth phase after amino acid starvation or L-arginine supplementation in pH 7 . 0 or 5 . 0 maintained at 25°C or 34°C were washed with PBS and then fixed in 1% of paraformaldehyde ( 4°C , overnight ) . For the analysis of AAP3 on the external face of the plasma membrane , the cells were incubated with anti-AAP3 serum ( 1:500 dilution ) at 4°C with overnight shaking . Then , the cells were incubated with goat anti-rabbit FITC conjugated antibody ( Sigma-Aldrich , St . Louis , MO , USA ) ( 1:500 dilution ) at room temperature with shaking for 1 h . For the analysis of total AAP3 , the cells were permeabilized with 0 . 05% Tween-20 for 20 min at room temperature . Then they were incubated with anti-AAP3 and anti-rabbit FITC , as previously described . The cells were analyzed using FlowSight image flow cytometer ( Amnis-MerckMillipore , Darmstadt , Germany ) . 10 , 000 cells were acquired , sorted and analyzed using the gate based in gradient root mean square ( RMS ) . Single cells were analyzed using the gate based in the bright field channel and fluorescence intensity of AAP3 in channel 2 . Data were acquired using Inspire and analyzed using Ideas Software ( Amnis Corporation , Seattle , WA , USA ) . All analysis was performed at the Core Facility of the Centro de Aquisição de Imagens e Microscopia from Instituto de Biociências ( Caimi-IB ) at the University of São Paulo . Approximately 106 promastigotes La-WT , La-EGFP/SKL [33] or La-WT axenic amastigotes in the stationary growth phase were washed with PBS and adhered to coverslips treated with poly-L-lysine ( Sigma-Aldrich , St . Louis , MO , USA ) for 15 min . The cells were then fixed with 2% paraformaldehyde for 10 min at room temperature . The fixed cells were permeabilized and blocked with 0 . 1% Triton X-100 and 0 . 1% BSA in PBS for 1 h at room temperature . To analyze sub-cellular AAP3 localization , anti-AAP3 polyclonal antibody ( 1:500 dilution ) was visualized using an anti-rabbit secondary antibody conjugated to Alexa546 or Alexa 488 ( Life Technologies , Carlsbad , CA , USA ) ( 1:500 dilution ) . Anti-α-tubulin ( Life Technologies , Carlsbad , CA , USA ) ( 1:1000 dilution ) was visualized using an anti-mouse secondary antibody conjugated to Alexa594 ( Life Technologies , Carlsbad , CA , USA ) ( 1:500 dilution ) . Nuclear and kinetoplast DNA were labeled using DAPI . Each step was followed by washing 10 times with PBS . The coverslips were mounted in ProLong media ( Life Technologies , Carlsbad , CA , USA ) . All imaging was performed using confocal microscope ( Zeiss LSM 780 NLO ) at the Core Facility of the Centro de Facilidades para Pesquisa ( CEFAP ) at the University of São Paulo . The experimental protocols for the animals were approved by the Animal Care and Use Committee from the Institute of Bioscience of the University of São Paulo ( CEUA 233/2015 ) . This study was carried out in strict accordance with the recommendations in the guide and policies for the care and use of laboratory animals of the São Paulo State ( State Law 11 . 977 , de 25/08/2005 ) and Brazil government ( State Law 11 . 794 , de 08/10/2008 ) .
Transcriptomic profiling by RNA-seq was used to identify differential gene expression in La-WT and La-arg- promastigotes and axenic amastigotes . Sequencing data obtained are available on the NCBI BioProject under accession number PRJNA380128 and Sequence Read Archive ( SRA ) under accession number SRX2661998 and SRX2661999 [22] . More than one billion sequence reads were obtained by Illumina HiSeq2000 . Data were aligned to the L . mexicana genome ( MHOMGT2001U1103 ) , and 8253 transcripts , 180 hypothetical proteins and 443 novel transcripts were identified . Based on the DE genes analyzed , we identified 14 amino acid transporters differentially expressed in the comparisons La-WT and La-arg- promastigotes and axenic amastigotes . As shown in Tables 1 and 2 , we observed a down-regulation of both 5 . 1 and 4 . 7 aap3 in La-WT and La-arg- promastigotes and axenic amastigotes . In contrast , we observed up-regulation of other amino acid transporters . Then , to investigate this modulation , we analyzed the changes in environmental signals that could regulate this gene expression , such as pH , temperature and L-arginine availability , as intrinsic factors that can influence the differentiation of the parasite life cycle from the sand fly to the mammalian macrophage host . BMDMs from BALB/c or C57BL/6 mice or human lineage THP-1 derived macrophages were infected with La-WT or La-arg- ( MOI 5:1 ) promastigotes and the infection index was determined at 4 , 24 and 48 h post-infection . We did not observe differences in the infection index of BMDMs from BALB/c infected with La-WT or La-arg- after 4 and 24 h . A lower infection index was observed in BMDMs from BALB/c infected with La-arg- after 48 h compared to La-WT ( Fig 1A ) , corroborating with previous data and indicating the importance of arginase activity to stablish the infection [8] . The infection index of BMDM from C57BL/6 infected with La-WT or La-arg- only presented significant difference after 48 h of infection ( Fig 1B ) . In contrast , the infection index from THP-1 macrophages with La-WT was increased after 24 and 48 h of infection . And the infection index with La-arg- was lower in all time infections when compared to La-WT ( Fig 1C ) . In addition , we determined the 5 . 1 and 4 . 7 aap3 amount in the preparations from macrophages from BALB/c , C57BL/6 and THP-1 infected with La-WT or La-arg- . The 5 . 1 aap3 mRNA amount presented an increase during the time course of macrophages from BALB/c infected with La-WT ( Fig 1D ) . The absence of arginase activity did not change the 5 . 1 aap3 mRNA amount during the time course of infection with La-arg- , but it was lower when compared to La-WT infection ( Fig 1D ) . The 5 . 1 aap3 amount in macrophages from C57BL/6 mice infected with La-WT also did not change during the time course of infection . ( Fig 1E ) . And in these C57BL/6 macrophages , the absence of arginase activity showed lower expression when compared to La-WT infection after 24 and 48 h ( Fig 1E ) . The 5 . 1 aap3 amount in human THP-1 macrophage increased during the time course of infection with La-WT . Interestingly , in these THP-1 macrophages , the 5 . 1 aap3 amount was higher in La-arg- compared to La-WT after 4h of infection , and decreased after 24 and 48 h of infection ( Fig 1F ) . Furthermore , the 4 . 7 aap3 amount appeared up-regulated during the time course of infection in macrophages from BALB/c infected with La-WT ( Fig 1G ) . However , did not appear altered in macrophages from BALB/c or C57BL/6 or THP-1 with La-arg- infection ( Fig 1G , 1H and 1I , respectively ) . Interestingly , as observed for 5 . 1 aap3 , the 4 . 7 aap3 amount in THP-1 macrophages was higher in La-arg- compared to La-WT after 4h of infection ( Fig 1I ) . La-WT promastigotes in mid-logarithmic growth phase were starved of amino acids or supplemented with 400 μM L-arginine and incubated at 25°C or 34°C for 4 h . Then , total RNA was extracted and the 5 . 1 and 4 . 7 aap3 mRNA were quantified by RT-qPCR . Data were normalized to the gapdh transcripts in each condition . As shown in Fig 2A , in parasites maintained at 25°C and pH 7 . 0 , the starvation of L-arginine caused a significant increase in 5 . 1 aap3 transcript level compared with parasites that were supplemented with L-arginine . Interestingly , at 34°C , an increase in 5 . 1 aap3 transcripts was observed during starvation of L-arginine at pH 5 . 0 ( Fig 2A ) . By contrast , a significant decrease of the same transcript was observed at pH 7 . 0 during starvation as well asin parasites submitted to L-arginine supplementation . The decrease was also detected at pH 5 . 0 during L-arginine supplementation ( Fig 2A ) . Although no significant difference was observed in the amount of 4 . 7 aap3 transcripts in all conditions , a slight decrease in the mRNA was observed when the parasites were submitted to amino acid starvation at 25°C ( Fig 2B ) . The profiles at 25°C and 34°C were very similar , except at 34°C in pH 5 . 0 , when starvation slightly increased the mRNA level ( Fig 2B ) . Our data suggest that L-arginine availability and increased temperature regulates only 5 . 1 aap3 mRNA expression . To assess the amount of AAP3 present on the external face of the plasma membrane and the total AAP3 present in the whole cell , we performed a flow cytometry analysis of fluorescence labeled AAP3 antibody against non-permeabilized and permeabilized parasites . The fluorescence intensity values were normalized to the control ( parasite collected before amino acid starvation or L-arginine supplementation ) . Initially , the total amount of AAP3 protein was measured during the time course of the growth curve . We observed an increase of AAP3 protein on days 3 , 5 , 7 and 9 compared to day 2 ( S1 Fig ) . This data indicated that the increase of AAP3 in the plasma membrane was related from mid-logarithmic to late-stationary growth phase . Then , the total amount of AAP3 was measured during pH and temperature changes , and L-arginine availability . The total amount of AAP3 protein was increased in parasites kept at 25°C and pH 7 . 0 under amino acid starvation , compared to the control parasites , but not in the parasites at pH 5 . 0 ( Fig 3A ) . Interestingly , a decreased of total amount was observed in parasites at 25°C , pH 7 . 0 and supplemented with L-arginine when compared to parasites under amino acids starvation . No difference in the total AAP3 amount was observed in parasites at 34°C and pH 7 . 0 , compared to the control parasites . In contrast , an increase in the total amount was observed in parasites at pH 5 . 0 independent of amino acids starvation or L-arginine supplementation ( Fig 3A ) . Furthermore , the plasma membrane amount of AAP3 was measured and we observed that the parasites under amino acid starvation at 25°C in pH 7 . 0 presented increased AAP3 amount in the membrane compared to the control parasites ( Fig 3B ) . A decreased in the membrane amount was observed in parasites at 25°C , pH 7 . 0 , supplemented with L-arginine when compared to parasites under amino acids starvation . No significant difference was observed at pH 5 . 0 under amino acids starvation , compared to control parasites . However , an increased membrane AAP3 amount was observed at pH 5 . 0 under L-arginine supplementation when compared to the control parasites and to those at pH 7 . 0 . The AAP3 membrane amount was increased at 34°C , in both pH 7 . 0 and pH 5 . 0 , and in amino acid starvation or supplementation of L-arginine ( Fig 3B ) . The differences in the total and plasma membrane AAP3 amount reflected the increase in the protein expression and in the traffic of the transporter carrier to the membrane . The increase in total and plasmatic membrane AAP3 amount occurred in amino acid starvation at 25°C and pH 7 . 0 , and at 34°C and pH 5 . 0 , contrasting with the amount of total AAP3 and plasmatic membrane in L-arginine supplementation at 34°C ( Fig 3C ) . In addition , we performed Western blot analysis of cell lysates of La-WT promastigotes and axenic amastigotes during the time course of growth curve , and no difference was observed in the AAP3 protein level ( S2A Fig ) . Compared with the flow cytometry analysis results of total AAP3 protein ( Fig 3C ) , the protein level detected by Western blotting showed a similar profile in La-WT promastigotes after amino acid starvation or L-arginine supplementation at both 25°C and 34°C ( S2C and S2D Fig ) . To evaluate and correlate the results obtained for mRNA and protein levels , we analyzed L-arginine uptake in parasites submitted to the same pH and temperature changes , and L-arginine availability . The L-arginine uptake increased during the amino acid starvation at pH 7 . 0 ( Fig 4 ) , demonstrating a correlation with the increase exhibited by 5 . 1 aap3 mRNA in the same conditions ( Fig 2A ) . Changing the parasites to 34°C caused a 3-times increase in the rate of L-arginine uptake independent of pH , as well as amino acid starvation or L-arginine supplementation ( Fig 4 ) . Using the anti-rabbit antibody against the epitope AAP3 , confocal microscopy analysis was performed to localized that transporter in the parasites . Confocal images from La-WT promastigotes and axenic amastigotes in the stationary growth phase , and La-GPF/SKL promastigotes in the stationary growth phase confirmed the AAP3 localization in the plasmatic membrane as well as partially in the glycosome ( Fig 5 ) .
The importance of the amino acid L-arginine in Leishmania has been related to parasite replication as well as a requirement to establish the infection in the mammalian host [8 , 17 , 32 , 34][22 , 35] . In the course of the Leishmania life cycle , environmental changes may act as signals to regulate gene expression , which enables the parasite to adapt to the new conditions [7] . Stress signaling starts with the consumption of all available nutrients at the end of blood meal digestion in the insect's digestive tract . The starvation signal can cause the release of procyclic promastigotes from the insect mid-gut epithelia and its migration to the proboscis , promoting the differentiation of procyclic promastigotes into metacyclic promastigotes [36] . The deprivation of amino acids is an important signal for metacyclogenesis , providing the parasites with the infective capacity to establish the infection . The shift of temperature , from the sand fly ( 25°C ) to the mammalian body temperature ( 37°C ) , is the next shock and is also known to be an important signal for parasite differentiation . The heat-shock proteins are good examples of gene activation that allow parasite survival in rapid temperature changes [37–39] . The pH change is the following step upon the fusion of the phagosome to the lysosome in the formation of the phagolysosome [37 , 40 , 41] . In this context , we described here how the parasite was able to regulate the AAP3 expression in response to the L-arginine availability , arginase activity , pH and temperature modifications . AAP3 is an amino acid transporter described for L-arginine uptake in L . amazonensis and L . donovani [17 , 18] . This transporter is encoded by two copies of aap3 gene ( 5 . 1 and 4 . 7 ) . A possible explanation about the presence of these two copies can be related with the post-transcriptional gene regulation according to the environmental conditions , since the two open reading frames are similar [18] . The differentially gene expression based on the RNA-seq analysis in La-WT and La-arg- promastigotes and axenic amastigotes in the stationary growth phase revealed transcripts of amino acid transporters down- and up-regulated . The 5 . 1 and 4 . 7 aap3 appeared down-regulated in the comparisons of La-WT promastigotes vs . axenic amastigotes and La-WT vs . La-arg- promastigotes . Similar results were observed in RT-qPCR assays from intracellular La-WT or La-arg- amastigotes infecting macrophages with different genetic background ( BALB/c and C57BL/6 mice , and human THP-1 ) . Interestingly , in THP-1 macrophages , we observed an increased expression of both 5 . 1 and 4 . 7 aap3 after 4 h of infection with La-arg- when compared to La-WT . This differential behavior can be explained based on the distinct genetic background of the host that can influence in L-arginine accumulation . Then , in THP-1 and La-arg- infections , the absence of arginase activity can lead to a host transporter modulation . By its side , the parasite can respond to the L-arginine availability modulating the aap3 expression to establish the infection . Previous studies demonstrated increased aap3 mRNA expression during the time course of macrophage infection . Muxel et al . , 2017 showed that intracellular amastigotes increased the amount of 5 . 1 aap3 in the time course of macrophage infection with La-WT . On the other hand , the infection with La-arg- did not altered 5 . 1 aap3 levels . These data demonstrated the importance of L-arginine availability and arginase activity to regulate 5 . 1 aap3 mRNA expression and L-arginine transport to ensure the amastigote survival [7] . Additionally , when the L-arginine availability was lower in melatonin-treated infected macrophages , the levels of 5 . 1 aap3 and arginase mRNA of intracellular amastigotes were maintained on trial to keep the polyamine supply [42] . Goldman-Pikovich et al . , 2016 showed that L . donovani intracellular amastigotes presented higher levels of LdAAP3 . 2 mRNA than in axenic amastigotes kept in L-arginine-starvation condition [19] According to TriTryp database , we identified AAP3 orthologs in L . major ( LMJLV39_310014600 , LMJLV39_310014700 , LMJSD75_310014300 , LMJSD75_310014400 , LmjF . 31 . 0870 and LmjF . 31 . 0880 ) , L . gerbilli ( LGELEM452_310014200 ) , L . tropica ( LTRL590_310015200 and LTRL590_310015300 ) , L . turanica ( LTULEM423_310014200 ) , L . braziliensis ( LbrM . 31 . 1030; LBRM2903_000006400 and LBRM2903_310017700 ) , L . donovani ( LdBPK_310900 . 1 and LdBPK_310910 . 1 ) and L . infantum ( LinJ . 31 . 0900 and LinJ . 31 . 0910 ) . The AAP3 transcripts were annotated for L . donovani ( LdBPK_310900 . 1 . 1 ) [43] , L . infantum ( LinJ . 31 . 0910 ) [31] , L . major ( LmjF . 31 . 0870 ) [44] and L . mexicana ( LmxM . 30 . 0870 . 1 ) [45] ( TriTrypDB ) . These findings among the different Leishmania spp . show the importance of L-arginine metabolism and uptake indicating that it could contribute to the fine tuning of gene expression and consequently L-arginine uptake . Still , according to the RNA-seq data , we also identified other amino acid transporters differentially expressed in the following comparisons: La-WT promastigotes vs . axenic amastigotes , and La-WT vs . La-arg- promastigotes . The transcript aATP11 , an amino acid transporter , was previously described in association with amino acid starvation [19] , confirmed the gene expression regulation in stress conditions . Notably , some aATP11 members were up-regulated ( LmxM . 30 . 0330 , LmxM . 30 . 0571 and LmxM . 30 . 0570 ) and other was down-regulated ( LmxM . 30 . 0350 ) . Other studies have shown that amastigotes activate signals when internalized into the phagolysosome , which has been linked to the down-regulation of many surface nutrient transporters . The remodeling of amastigote central carbon metabolism also represents a programmed response to stress that cells undergo in the host macrophage [46 , 47] . The response to starvation or L-arginine supplementation in different conditions of pH and temperature revealed that 5 . 1 aap3 mRNA was down-regulated at 34°C and pH 7 . 0 in both amino acid starvation and L-arginine supplementation , and at pH 5 . 0 in L-arginine supplementation . These observations can indicate a gene expression regulation at 34°C , during amastigote differentiation . On the other hand , 5 . 1 aap3 mRNA was up-regulated at pH 5 . 0 in amino acid starvation , as previously described for L . amazonensis [18] and L . donovani [19] , suggesting up-regulation of the transporter in response to low L-arginine availability . The 4 . 7 aap3 mRNA did not show significant expression differences , indicating that probably only the 5 . 1 aap3 copy present the regulatory sequence for modulation . The data obtained after L-arginine starvation showed increased amino acid uptake corroborating previous findings [18] , indicating that Leishmania senses the concentration of this amino acid and regulates the expression of the transporter [17–19] . L-arginine starvation reduced the levels of arginine , ornithine and putrescine , but not spermidine , spermine and agmatine , in L . amazonensis promastigotes [48] . This condition allows the parasite to sense new signals , such as L-arginine availability in the phagolysosome environment , which is predominantly caused by the competition for L-arginine with the host cell [19] . The absence of L-arginine or polyamine could be surpassed by the polyamine transporter , described and characterized in L . major , with high affinity for putrescine and spermidine [49] , L . donovani [50 , 51] and L . mexicana [50] , and presenting an optimal transport function at pH 7 . 0–7 . 5 for promastigotes and pH 5 . 0 for amastigotes of L . mexicana [50] The increase of AAP3 protein levels in the plasmatic membrane reflected the increase of L-arginine uptake at 34°C , highlighting the participation of AAP3 and the importance of the uptake of the amino acid during promastigote to amastigote differentiation , as a response to the change in temperature ( 25 to 34°C ) [17–19] . And the increase amount of total AAP3 compared to the plasma membrane can suggest the directing of this transporter to other compartments . This hypothesis was confirmed with the cellular localization of AAP3 in promastigotes and axenic amastigotes . Previous studies already demonstrated the LdAAP3 localized in the flagella surface and in the glycosome [19] . As arginase enzyme was also localized in the glycosome [8] and as Szoor et al . ( 2010 ) reported that signaling in the nutrient-sensing pathway was targeted to this organelle in Trypanosoma brucei [52] , we hypothesized that the AAP3 could also be localized in the glycosome to supply polyamines biosynthesis . In this study , we demonstrated AAP3 localized in the plasma membrane as well as in the glycosome of La-WT promastigote and axenic amastigotes in the stationary growth phase , indicating that L-arginine uptake is directed to this organelle . The results presented in this communication indicated that as a strategy for controlling Leishmania infection could be focused in the inhibition of L-arginine flux into the glycosome of the parasite .
|
Leishmania alternates its life cycle between the invertebrate host , in which the promastigote forms reside at pH 7 . 0 and approximately 25°C; and the mammalian host , in which the amastigote forms reside at pH 5 . 0 and approximately 37°C . These environmental changes submit the parasite to dynamic undergo modifications in morphology , metabolism , cellular signaling and gene expression regulation to allow for a rapid adaptation to the new environmental conditions . Leishmania is auxotrophic for many amino acids , such as L-arginine . The L-arginine availability is important for the uptake control by the parasite as well as the intracellular amino acids pool maintenance . Leishmania arginase uses L-arginine to produce urea and ornithine , being the last one a precursor of polyamine biosynthetic pathway , which is used by the parasite to replicate and to establish the infection . L-arginine uptake in L . amazonensis is performed through AAP3 , which is encoded by two gene copies arranged in tandem on the genome ( 5 . 1 and 4 . 7 aap3 ) . In this report , we characterized the AAP3 function and expression regulation in parasites maintained in different pH and temperature conditions simulating both insect and mammalian micro-environment . We submitted the parasites to amino acid starvation , simulating mid-gut starvation , a signal for promastigote metacyclogenesis . Our results demonstrated that the changes in temperature and pH , in addition with amino acid starvation or L-arginine supplementation might represent important signals for aap3 expression regulation , mainly for the 5 . 1 aap3 copy . We also linked the regulation of 5 . 1 aap3 transcript to the genetic background from the host macrophage . In addition , we localized the AAP3 in the plasma membrane and in the glycosome of L . amazonensis promastigotes and axenic amastigotes , indicating that arginine uptake is directed to this organelle .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"microbiology",
"parasitic",
"diseases",
"protozoan",
"life",
"cycles",
"parasitic",
"protozoans",
"membrane",
"proteins",
"developmental",
"biology",
"protozoans",
"leishmania",
"promastigotes",
"cellular",
"structures",
"and",
"organelles",
"white",
"blood",
"cells",
"animal",
"cells",
"gene",
"expression",
"life",
"cycles",
"amastigotes",
"cell",
"membranes",
"eukaryota",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"protozoology",
"cellular",
"types",
"macrophages",
"organisms"
] |
2017
|
L-arginine availability and arginase activity: Characterization of amino acid permease 3 in Leishmania amazonensis
|
Many genes involved in the immune response of Anopheles gambiae , the main malaria vector in Africa , have been identified , but whether naturally occurring polymorphisms in these genes underlie variation in resistance to the human malaria parasite , Plasmodium falciparum , is currently unknown . Here we carried out a candidate gene association study to identify single nucleotide polymorphisms ( SNPs ) associated with natural resistance to P . falciparum . A . gambiae M form mosquitoes from Cameroon were experimentally challenged with three local wild P . falciparum isolates . Statistical associations were assessed between 157 SNPs selected from a set of 67 A . gambiae immune-related genes and the level of infection . Isolate-specific associations were accounted for by including the effect of the isolate in the analysis . Five SNPs were significantly associated to the infection phenotype , located within or upstream of AgMDL1 , CEC1 , Sp PPO activate , Sp SNAKElike , and TOLL6 . Low overall and local linkage disequilibrium indicated high specificity in the loci found . Association between infection phenotype and two SNPs was isolate-specific , providing the first evidence of vector genotype by parasite isolate interactions at the molecular level . Four SNPs were associated to either oocyst presence or load , indicating that the genetic basis of infection prevalence and intensity may differ . The validity of the approach was verified by confirming the functional role of Sp SNAKElike in gene silencing assays . These results strongly support the role of genetic variation within or near these five A . gambiae immune genes , in concert with other genes , in natural resistance to P . falciparum . They emphasize the need to distinguish between infection prevalence and intensity and to account for the genetic specificity of vector-parasite interactions in dissecting the genetic basis of Anopheles resistance to human malaria .
Human malaria is transmitted by female Anopheles mosquitoes , which vary in vector competence at both the species and individual level [1] . In Anopheles gambiae , the main malaria vector in Africa , it has been possible to select laboratory strains for their resistance or susceptibility to Plasmodium infection [2] , [3] , indicating that resistance has a genetic basis . This led to much effort being targeted towards understanding the genetic determinants of resistance with the hope of uncovering novel ways to reduce malaria transmission [4] . Although considerable progress has been made in model systems , the genetic basis of Anopheles resistance to Plasmodium remains to be understood in detail in epidemiologically meaningful vector-parasite species combinations . Resistance of natural populations of A . gambiae to Plasmodium falciparum , the deadliest human malaria parasite , is currently under scrutiny . The development of powerful genetic tools [5]–[7] in parallel with the sequencing of the A . gambiae genome [8] has substantially improved our knowledge of the molecular interactions between Anopheles and Plasmodium . It was shown that mosquito innate immunity plays a major role in controlling the level of infection by eliminating the majority of malaria parasites ( reviewed in [9] ) . The general scheme of the A . gambiae immune response has been deciphered: initially , pattern recognition receptors ( PRRs ) bind to pathogen-associated molecular patterns of the parasite that trigger signal transduction and modulation cascades; finally , effector molecules are activated to kill the parasites through a range of possible mechanisms [10] . The outcome of infection seems to depend on a fine balance between mosquito factors that act either positively or negatively on Plasmodium development [11]–[21] . Phenotypic variation in A . gambiae resistance to P . falciparum is likely influenced by naturally occurring polymorphism in genes that encode positive or negative modulators of the immune response . For instance , genetic variation at pathogen recognition and intracellular signaling loci may significantly contribute to phenotypic variation in immune competence [22] . If some mosquito immune variants are expected to perform better in controlling malaria infection , they are however not expected to reach fixation for at least two reasons . Firstly , even if not clearly documented in the A . gambiae - P . falciparum couple , the mosquito immune response is likely to be costly [23] , [24] , which may counteract the selection pressure exerted by the parasite and maintain the frequency of resistance at intermediate levels [25] . Secondly , interactions between A . gambiae and P . falciparum appear to be genotype-specific . Experiments using different A . gambiae families challenged with several field isolates of P . falciparum revealed significant mosquito genotype by parasite genotype ( G x G ) interactions , whereby the outcome of infection depends on the specific combination of mosquito and parasite genotypes [26] . Such G x G interactions can promote the maintenance of polymorphism through negative frequency-dependant selection [27] . Earlier studies exploring the genetic variation underlying Anopheles resistance to Plasmodium have mainly relied on Quantitative Trait Loci ( QTL ) mapping strategies . This has generally been conducted in model systems by exposing selected resistant/susceptible A . gambiae strains to rodent or simian Plasmodium species [28]–[30] . The most recent study identified loci associated to resistance in the chromosomal region containing TEP1 [6] , a gene that was previously shown to play a major role in the mosquito immune response [11] , [13] , [31] and in P . falciparum development [14] . However , mechanisms of anti-Plasmodium defense in mosquitoes initially uncovered in model systems do not always hold for the natural couple A . gambiae - P . falciparum [32] , [33] , likely because of the absence of a shared evolutionary history in artificial species combinations [34] , [35] . Studies of the natural A . gambiae - P . falciparum couple are still limited in number but have identified promising genetic markers of resistance . Associations were found between NOS and cecropin gene polymorphisms and infection status in wild mosquitoes from East Africa [36] , while a QTL mapping approach identified a Plasmodium Resistance Island ( PRI ) on chromosome 2L in mosquitoes from both East and West Africa [37] , [38] . Fine genetic dissection within the PRI highlighted a family of three genes ( APL1A , APL1B and APL1C ) with different roles in infection control . APL1A affects P . falciparum development and APL1C limits the development of the rodent parasite Plasmodium berghei [15] , [39] and is part of the important LRIM1/APL1C/TEP1 complex that is known to play a central role in controlling Plasmodium infection in A . gambiae [11] , [31] . More stringent tests have shown polymorphic sites within APL1C to account for much of the variation in immunity against P . berghei [39] . To date , studies on the natural resistance of A . gambiae to P . falciparum relied either on genome-wide QTL mapping strategies [37] , [38] , [40] or on genotype-phenotype associations using a very limited number of candidate genes [36] . Here , we conducted a large-scale candidate gene association study based on a set of 67 A . gambiae immune-related genes selected based on their role in innate immunity [10] or their functional implication in the response to Plasmodium [14] . We identified associations between gene polymorphisms and P . falciparum infection success in a semi-natural system consisting of a newly established mosquito colony experimentally exposed to field parasite isolates . We verified the validity of the approach by confirming the functional role of one of the genes found to be associated with infection phenotype .
Three separate experimental infections were carried out , each using a different gametocyte-positive blood isolate from carriers in Cameroon , hereafter named Isolate 1 , 2 and 3 . The number of mosquitoes included in the analysis ( that fed and survived until dissection 8 days after the infectious blood meal ) for Isolates 1 , 2 and 3 were 380 , 340 and 201 , respectively . The P . falciparum isolates contained 813 , 31 and 107 gametocytes/µl and infected 75% , 63% and 71% of the mosquitoes , with a mean number of oocysts per midgut of 11 . 5 , 5 . 3 and 15 . 2 , for Isolates 1 , 2 and 3 respectively . Thus , around one third of all mosquitoes remained uninfected after feeding on the same infectious blood . To evaluate P . falciparum genetic diversity in the blood isolates used for experimental infections , two parasite merozoite surface protein ( MSP ) alleles were genotyped . This analysis revealed that the three isolates used in this study contained distinct P . falciparum genotypes . MSP1 M had three alleles in Isolate 1 and one in Isolates 2 and 3 , and none were shared ( one null allele was found ) . One , four and two MSP2 FC alleles were identified in Isolates 1 , 2 and 3 , respectively , with Isolates 1 and 3 sharing one allele . Although the MSP alleles identified in the blood samples did not necessarily represent the gametocyte population ( the sexual stage infectious to the mosquito ) at the time of infection , the allelic pattern displayed by each parasite isolate indicated that they were genetically distinct . This confirmed that P . falciparum populations in sub-Saharan Africa are genetically diverse and that each human infection generally consists of multiple parasite strains during the period of highest transmission [41]–[44] . 157 single nucleotide polymorphisms ( SNPs ) located within and upstream 67 immune-related genes were successfully genotyped for mosquitoes fed on Isolate 1 with an extreme phenotype ( see Methods ) . Based on their statistical significance , 21 of the 157 SNPs were selected for genotyping of mosquitoes with an intermediate phenotype and six of these were further selected for genotyping of mosquitoes fed on Isolates 2 and 3 . Significant deviations from Hardy Weinberg Equilibrium ( HWE ) were detected in 18% of SNPs genotyped for mosquitoes with an extreme phenotype fed on Isolate 1 ( data not shown ) , in 38% of SNPs when all mosquitoes fed on Isolate 1 were pooled ( Table S1 ) , and in 39% of the SNPs genotyped across all three isolates ( Table S2 ) . Because deviations from HWE affected a relatively large proportion of the SNPs , nine genes representing the range of Fis values were selected and sequenced to verify that the sequence matched the assigned genotype . Each gene contained between one and three of the selected SNPs with a total of 20 SNPs sequenced from the original DNA stock before whole genome amplification . Nucleotide identity ranged between 85 and 100% for each SNP with an average of 95% , which ruled out the possibility that the significant HWE deviations observed were due to technical errors . A possible biological explanation is that the mosquitoes used for each infection derived from a relatively small number of parental pairs , resulting in some degree of population structure in the offspring . Of ten neutral microsatellite loci that were genotyped , however , none were found to be in HW disequilibrium ( Table S3 ) , indicating that any genetic stratification must have been minor . It is worth mentioning that although HW disequilibrium may have affected the power of the study ( i . e . the ability to detect genotype-phenotype associations ) , it did not affect the significance of the results ( i . e . the probability of falsely rejecting the null hypothesis that a SNP is not associated with the phenotype ) . Overall , linkage disequilibrium ( LD ) between SNPs was low . A cut-off of r2 = 0 . 8 is commonly used to exclude redundant SNPs in association studies because a non-causative SNP in LD with a causative SNP will generally be found associated to the phenotype if r2≥0 . 8 [45] . Between 0 . 6 and 2 . 1% of SNP pairs were above this cut-off depending on the chromosome arm ( Table 1 ) . Of the 21 SNP pairs over this threshold , 67% were located less than 500 nucleotides apart , 29% were between 0 . 5 and 5 . 5 kb from each other and one SNP pair located 5 . 2 Mb apart showed long-range LD . When considering only the mosquitoes with an extreme phenotype fed on Isolate 1 ( n = 192 females ) , significant association between the genotype and the level of infection was found in 21 of the 157 SNPs examined . After inclusion of intermediate phenotypes fed on the same isolate ( n = 380 females ) , six SNPs remained significantly associated to infection phenotype . These SNPs are named as follows: AgMDL1-40910564 , CEC1-12441661 , CLIPB4-34473971 , SpPPOact-58805968 , SpSNAKElike-40693950 , TOLL6-41490803 ( associated gene name-genomic position ) . The statistical significance of the associations was assessed through a False Discovery Rate ( FDR ) procedure to correct for multiple testing . If all null hypotheses ( SNP genotypes are not associated to infection phenotype ) were true , the FDR procedure would find zero significant tests in 95% of replicate studies and one significant test in 5% of replicate studies . The robustness of the six significant genotype-phenotype associations was evaluated across different parasite genotypes by repeating the experimental infections using three different genetically distinct P . falciparum isolates ( Table 2 ) . The parasite isolate used for infection generally had a significant effect on both components of the infection phenotype , prevalence ( proportion of infected mosquitoes ) and intensity ( number of oocysts in infected mosquitoes ) . This isolate effect encompasses inherent experimental variation due to the day of infection as well as the genetic identity of the parasite isolate ingested . Overall tests across all isolates and both phenotypes confirmed genotype-phenotype associations showing significance for all SNPs with the exception of CLIPB4-34473971 , which was marginally non-significant . Breaking down the analysis of the infection phenotype into tests of a total SNP effect within each component ( prevalence and intensity ) , a SNP additive effect and an Isolate x SNP interaction effect ( Table 2 ) provided additional information on the lack of significance of CLIPB4-34473971 . Although none of the effects were significant , both interaction P-values were low , suggesting that the lack of overall effect may have resulted from the absence of genotype-phenotype association with Isolates 2 and 3 . AgMDL1-40910564 genotype was associated to infection intensity , but not prevalence . The genotype-phenotype pattern was similar across the three parasite isolates as confirmed by the significant SNP additive effect and non-significant SNP x Isolate interaction ( Table 2 ) , suggesting that the effect of this SNP may be independent of parasite genotype . Heterozygotes were significantly more susceptible to P . falciparum infection than both homozygote genotypes ( Figure 1 ) . AgMDL1-40910564 is located within the coding region of AgMDL1 , which encodes a PRR [14] . The two alleles observed in the population ( A and G ) correspond to a synonymous substitution , suggesting that the causative SNP ( s ) are likely to be distinct . As mentioned previously , LD is generally low in the A . gambiae colony used in this study . The nearest known gene to AgMDL1 is >1 kb away , making it likely that the causative SNP ( s ) are located within the same gene and lead to an amino acid modification altering parasite recognition . AgMDL1 is thought to initiate an immune response upon recognition of a parasite , similar to its vertebrate homolog [14] . The gene is up-regulated 1 . 7 fold in response to P . falciparum infection and its silencing facilitates P . falciparum oocyst development but has little effect on P . berghei . Although prevalence and intensity are confounded in the analysis , AgMDL1 appears to affect both components of the infection phenotype [14] . The identification of the association between AgMDL1-40910564 with infection intensity in addition to the effect of AgMDL1 silencing previously observed highlights the major role of this gene in controlling P . falciparum infection and the interest of deciphering its function . It is surprising that in the present study AgMDL1-40910564 heterozygotes showed higher infection levels across three different isolates . One potential explanation is that when the parasites contained in the blood meal are genetically diverse ( as in this study ) , different allelic forms of AgMDL1 allow recognition of different Plasmodium genotypes , resulting in reduction of within-host competition between parasite strains and increased infection level in heterozygotes [46] , [47] . The hypothesis of a parasite genotype-specific function of AgMDL1 contrasts with the previous observation that the effect of AgMDL1-40910564 did not depend on parasite isolate , but this interpretation may be complicated by complex interactions between co-infecting parasite strains [48] and requires further investigation . CEC1-12441661 and SpPPOact-58805968 genotypes were associated to infection prevalence , but not intensity . A similar pattern was observed across isolates as confirmed by the significant SNP additive effect , suggesting that the effect of these SNPs did not depend on parasite genotype ( Table 2 ) . Heterozygotes were significantly more resistant to P . falciparum infection than homozygotes ( Figure 2 ) . Higher resistance in heterozygotes is consistent with the idea that genetic heterozygosity enhances host immunity to infectious agents [49] , [50] . CEC1-12441661 is about 500 bp upstream from the coding region of Cecropin 1 on one side and of Cecropin 3 on the other side . Thus , this SNP could be causative or linked to causative SNP ( s ) in the regulatory regions affecting the expression of Cecropin 1 or Cecropin 3 , or linked to non-synonymous causative SNP ( s ) in the coding regions of either gene . Although CEC1-12441661 was not in LD with any of the other SNPs genotyped on the chromosome , it is worth noting that two SNPs located on either side in Cecropin 2 and Cecropin 3 showed an r2 of 0 . 92 at a distance of 4 . 1 kb indicating that LD in this region can be high ( Table 2 ) . Cecropin 1 and Cecropin 3 are anti-microbial peptides for which different allelic variants could confer enhanced efficacy against a mixed-genotype Plasmodium infection . Consistently , allelic variants of Cecropin 1 have previously been associated to natural P . falciparum infection [36] . SpPPOact-58805968 is located 1 . 5 kb upstream of the coding region of Sp PPO activate but also in the coding region of the gene AGAP004639 causing a synonymous mutation . This SNP could be causative by affecting regulation of Sp PPO activate expression or linked to causative non-synonymous SNP ( s ) in either of the genes . Sp PPO activate is part of a serine protease cascade up-regulated in response to P . falciparum infection [14] and thus represents a strong candidate gene whose polymorphism may underlie variation in resistance . AGAP004639 is an ortholog of genes encoding clip domain serine proteases in Aedes aegypti and Culex quinquefasciatus involved in signal modulation and amplification following non-self recognition [10] . To our knowledge it has not already been implicated in the mosquito response against Plasmodium but is also a promising candidate . Causative SNPs within either gene will likely interact indirectly with the parasite altering the amount or type of effector molecule produced . SpSNAKElike-40693950 genotype was significantly associated to infection intensity and marginally to prevalence . In both cases , associations were parasite isolate-specific as indicated by the significant Isolate x SNP interaction ( Table 2 ) . For infection prevalence , AA homozygotes showed greater resistance than AG heterozygotes against Isolate 1 , but the opposite trend was observed with Isolates 2 and 3 ( Figure 3A ) . For intensity , AA homozygotes showed greater susceptibility than AG heterozygotes against Isolate 1 , and to a lesser extent Isolate 2 . The opposite trend was observed with Isolate 3 , although the effect was relatively modest ( Figure 3B ) . The non-significant total SNP effect on prevalence indicates that this association was weaker than for intensity ( Table 2 ) . The significant Isolate x SNP interactions suggest that the effect of this SNP , or linked causative SNP ( s ) , on the outcome of infection depends on the parasite genotype . Although the SNP effect is highly significant for Isolate 1 , it is not significant with either Isolates 2 or 3 alone , as shown by separate Kruskal-Wallis tests for each isolate ( data not shown ) . The strong association between SpSNAKElike-40693950 genotype and infection phenotype for Isolate 1 but not for the two other isolates is consistent with the potential implication of this SNP , or linked causative SNP ( s ) , in specific G x G interactions with the parasite . The SNP is located in the coding region of Sp SNAKElike , which is involved in a serine protease cascade and is up-regulated following P . falciparum infection [14] . It causes a synonymous mutation and is therefore likely to be linked to causative SNP ( s ) . Although the nearest gene is>1 . 2 kb away , this particular SNP is in high LD with another SNP located 5 . 2 kb away , pointing to a larger region that may contain the causative SNP ( s ) . The causative SNP may act indirectly with the parasite affecting the downstream immune signal . Isolate-specific association was also observed between TOLL6-41490803 genotype and infection prevalence , as indicated by the significant Isolate x SNP interaction ( Table 2 ) . For Isolate 1 , heterozygotes were more susceptible to infection than both homozygote genotypes ( Figure 4 ) , a similar scenario to AgMDL1-40910564 . Isolates 2 and 3 showed a similar trend , although the effects were not statistically significant when analyzed independently by a Kruskal-Wallis test ( data not shown ) . The significant Isolate x SNP interaction suggests that this SNP , or linked causative SNP ( s ) , may be responsible for G x G interactions with the parasite . The SNP is located within TOLL6 , which encodes a toll-like receptor involved in immune signal transduction [10] , [51] causing a non-synonymous mutation . This could therefore be causative or be linked to causative SNP ( s ) probably located within the same gene ( the nearest known gene is>0 . 5 Mb away ) . This SNP may act indirectly with the parasite affecting the downstream immune signal . Statistical association between a genetic marker and infection phenotype does not provide conclusive evidence that the gene where the genetic marker is located plays a functional role in controlling infection . We verified the validity of our approach by testing the functional role of Sp SNAKElike in P . falciparum infection through gene silencing assays using five new P . falciparum isolates ( named 4–8 hereafter ) . Sp SNAKElike was selected due to the level of significance of SpSNAKElike-40693950 to both components of infection phenotype ( intensity and prevalence ) , together with the longer range LD exerted by this SNP decreasing the specificity of the association . Overall , RNAi knockdown of Sp SNAKElike resulted in increased susceptibility to P . falciparum infection ( Figure 5 ) . In a combined analysis of prevalence and intensity , mosquitoes depleted for Sp SNAKElike expression harbored significantly more oocysts per midgut than control mosquitoes ( overall P-value = 0 . 0027 ) . Further analysis showed that Sp SNAKElike silencing had a main effect on infection intensity ( P = 0 . 01 ) but not prevalence ( P = 0 . 08 ) , which was consistent with the results of the association study . The amplitude of the effect varied with the isolate . As SpSNAKElike-40693950 is in high LD with another SNP 5 . 2 kb away , the region potentially containing the causative SNP ( s ) in the association study is relatively large . Increased oocyst numbers upon Sp SNAKElike knockdown confirm that this gene is an antagonist to P . falciparum development and makes it likely that the causative SNP ( s ) include SpSNAKElike-40693950 and/or closely linked SNP ( s ) within the same gene . This gene was selected for inclusion in the study based on its up-regulation in response to ingestion of a P . falciparum infected blood meal [14] . Sp SNAKElike in Drosophila is up-regulated in response to Gram-positive bacteria and fungi and predicted to activate the TOLL immune response pathway [52] . It is therefore likely that Sp SNAKElike plays a role in immune signaling in A . gambiae .
This study is , to our knowledge , the first one that examined associations between natural polymorphisms in a large number of immune-related genes in A . gambiae and P . falciparum infection success . Out of 67 initial candidates , five immune genes had polymorphisms that strongly associated with infection phenotype through our screening scheme . Genetic variation in these genes ( or closely linked loci ) thus contributes to phenotypic variability of A . gambiae resistance to P . falciparum in natural populations . Although our approach bears some limitations ( discussed below ) , it provides several new insights into our understanding of the genetic basis of A . gambiae resistance to P . falciparum in natural populations . Our results show that genes underlying variation in infection intensity may be partly distinct from those controlling variation in infection prevalence . Importantly , some of the associations found were isolate-specific , indicating that the alleles underlying A . gambiae resistance or susceptibility to P . falciparum can differ across parasite genotypes . We validated our approach by confirming the functional role of a gene that contained a SNP significantly associated to infection phenotype . Earlier studies have successfully used a genome-wide QTL mapping strategy in field-derived isofemale mosquito families to detect associations between microsatellite markers and P . falciparum infection phenotype [37] , [38] , [40] . Here we tested associations between SNPs identified in a large set of candidate immune genes and infection phenotype in a newly established mosquito colony . Although our method likely excluded potentially important genes because of the initial selection of candidates , it has the major advantage of the minimal LD harbored by a randomly mating colony compared to isofemale pedigrees for which LD spans over genomic regions of several Mb . Whereas associations found using isofemale pedigrees necessitate to gradually refine genetic mapping to identify the causative polymorphisms , our strategy likely identified SNPs that are either causative or closely linked to the causative SNPs [53] . The Plasmodium Resistance Island 1 ( PRI1 ) previously uncovered in natural populations of A . gambiae [37] includes several immunity genes but the causative polymorphic sites remain to be identified . In the present study , 12 SNPs were genotyped in the PRI1 , within and upstream of APL1 and APL2 , but none were associated to infection phenotype . This emphasizes the need for complementary approaches to unravel the complex genetic basis of mosquito resistance to P . falciparum in natural populations . Polymorphisms in the five genes identified in the present study most likely act in concert with genetic variation in many other genes to drive phenotypic variability [54] . In other words , the five candidate genes that we identified probably represent only a glimpse into the complex genetic basis of A . gambiae resistance to P . falciparum in natural populations . Several aspects of our approach that were designed to increase our power of detection also limited its scope . Firstly , our study was initiated with a set of candidate genes known to be implicated in the mosquito immune response . Most of these genes have been identified in functional studies based on their up- or down-regulation upon Plasmodium infection [14] . That a gene is functionally involved in anti-Plasmodium immunity does not necessarily imply that its polymorphism underlies phenotypic variation in resistance . Conversely , other genes that are not regulated upon infection may contribute to phenotypic variation in resistance . Thus , by using a selected set of genes based on the available data we inevitably excluded an unknown number of potential candidates . Secondly , the first two steps of our procedure excluded SNPs that did not show a significant association to infection intensity by a single P . falciparum isolate . As a result , genetic polymorphisms acting on prevalence and/or involved in isolate-specific resistance may have been missed . Ideally , equal numbers of mosquitoes from different infections should be genotyped before selecting SNPs for more stringent analyses . Such a strategy would be less likely to exclude SNPs only effective against a subset of isolates . By decomposing the infection phenotype into infection prevalence and infection intensity , we found that four of the five SNPs significantly associated to infection phenotype were associated to only one but not both components . This indicates that different mechanisms , involving different gene pathways , may control the different steps of the infection process . We hypothesize that the pathways triggered to prevent infection differ to some extent from those involved in minimizing infection intensity . Differences in the genetic basis underlying Plasmodium infection prevalence and infection intensity in Anopheles mosquitoes have been previously observed [39] , [55] . For two of the five significant SNPs ( AgMDL1-40910564 and TOLL6-41490803 ) heterozygous mosquitoes had an increased parasite load , which was unexpected as heterozygosity is generally expected to increase resistance [49] , [50] . When previously observed [56]–[58] , it has been hypothesized that both alleles alone may have beneficial effects in homozygotes if reduced protein production of each of the two allelic variants in heterozygotes leads to reduced fitness [59] . Both under-dominance ( homozygote advantage ) and over-dominance ( heterozygote advantage ) are suggested in the associated SNPs , both of which were shown to have important consequences for the maintenance of polymorphism in immunity genes [60]–[62] . Our results suggest that evolutionary forces maintain polymorphism in the A . gambiae immune system , although so far purifying selection was identified as the most common form of selection [63]–[66] . By explicitly accounting for parasite variation in the analysis , we identified SNPs that were associated with infection phenotype in an isolate-specific manner . Two of the five significant SNPs ( SpSNAKElike-40693950 and TOLL6-41490803 ) had an effect that strongly depended on the parasite isolate . By replacing the autologous serum with naïve serum , we standardized the infectious blood meal that was offered to the mosquitoes . We did not control , however , potential differences in the nutritive quality of the blood meal , which may have affected parasite infection success . Nonetheless , isolate-specific associations detected between SNPs and infection phenotype are consistent with earlier evidence that P . falciparum infection in A . gambiae is governed by G x G interactions [26] , [67] . The present findings go one step further by providing evidence for specific mosquito genotype by parasite isolate interactions ( an approximation of G x G interactions ) at the molecular level . That only two SNPs showed significant SNP x Isolate interactions indicates that among A . gambiae genes underlying resistance/susceptibility to P . falciparum , only some may be responsible for G x G interactions whereas others would provide a generalist effect against all parasite genotypes . For example , it was observed that APL1A had a similar effect across multiple P . falciparum genotypes [39] , [55] . By measuring significant SNP x Isolate interactions for SpSNAKElike-40693950 and TOLL6-41490803 , we may have found the two first genes governing specific compatibility between A . gambiae and P . falciparum . The suggestion of mosquito genotype by parasite isolate interactions at the SNP level has important implications for current efforts towards identification of Anopheles genes underlying resistance to P . falciparum [67] . Genetic dissection of mosquito resistance to Plasmodium is usually conducted either in model systems or in semi-natural systems that do not account for the genetic variation among parasites ( e . g . [6] , [11] , [14] , [37] , [39] , [40] , [68] ) . Data obtained from infections using different parasite isolates are generally pooled or the effect of genetic polymorphism determined by computing statistical significance across experiments . The present study reveals the importance of considering the variation observed between infections with different parasite isolates in functional genetic or association studies . A significant portion of the observed variation in infection phenotype seems to result from specific interactions between mosquito and parasite genotypes . A total SNP effect was found for two of the three SNPs involved in significant Isolate x SNP interaction effects , but when each of the three isolates was considered separately the SNP effect was only significant for Isolate 1 . Previous studies may therefore have reported significant genotype-phenotype associations based on a total effect and presumed to be parasite isolate-independent that may have in fact resulted from an interaction effect . This highlights the need to integrate the effect of the parasite genome in interaction with the mosquito genome if we are to fully understand the genetic architecture of mosquito resistance to malaria parasites [69] . LD reduces the specificity of associations and is highly influenced by chromosomal inversions [70] . The mosquito strain used in the current study is fixed for the Forest ( standard ) pattern of inversions [71] , so that genomic regions with significantly increased LD due to chromosomal inversions are not expected . In future studies using mosquito strains with polymorphic inversion patterns , specific SNP locations and LD relationships will have to be carefully examined to interpret the results . Detecting associations between infection phenotype and individual SNPs supports the potential functional importance of the genes in which they lie , whether the association is isolate-specific or not . Most of the genes in which we uncovered significant SNPs have been relatively well characterized for their role in the immune response of A . gambiae or closely linked species [10] , [14] , [36] , [51] . The exact role of TOLL6 is still unknown , but it may play a role in the TOLL pathway , which has been mainly implicated in the A . gambiae response to P . berghei and not P . falciparum [39] . A recent study , however , suggests a role for the two major immune signaling pathways , TOLL and IMD , with the IMD pathway controlling infection prevalence and the TOLL pathway controlling infection intensity [19] . The present study supports a role for both pathways although TOLL6-41490803 showed association to infection prevalence , suggesting that some aspects of these major immune pathways in the mosquito remain to be discovered . Little is known , however , about the functional role of Sp SNAKElike , which contains the SNP with the most significant association in this study , besides its up-regulation in A . gambiae in response to ingestion of a P . falciparum-infected blood meal [14] . We used RNAi gene knockdown to show that Sp SNAKElike plays an important role in controlling infection with a major effect on infection intensity . This result is consistent with the hypothesis that ‘SNAKE like’ genes are responsible for activating the TOLL immune pathway in Drosophila [52] , which in A . gambiae plays a major role in controlling P . falciparum infection intensity [19] . The phenotypic effect of Sp SNAKElike silencing supports the conclusion that the causative SNP ( s ) in the association study are located within this gene rather than further away , as could have been suggested by the LD data . The variation observed in the gene knockdown effect across isolates could result from mosquito genotype by parasite isolate interactions in agreement with the association analysis . Such isolate-dependent variation in the functional effect of gene silencing is expected if the function of the gene depends on a specific interaction between its own polymorphism and that of one or several parasite genes . In particular , when the gene variant ( s ) present in a mosquito genotype does not allow effective recognition or interaction with a certain parasite genotype , gene silencing will lead to little or no phenotypic difference for this isolate . To conclude , we used a multi-step association procedure that provided strong support for the role of genetic variation within or near five candidate immune genes in natural resistance of A . gambiae to P . falciparum . The relevance of our approach was validated functionally for the candidate genes identified in the association analysis . Although our approach was not exhaustive , this information will be useful for future allele-specific functional characterization of the corresponding genes or their immediate genomic region . In addition , our association analysis at the SNP level provided important information for further dissection of the genetic basis of natural A . gambiae resistance to P . falciparum . Four of the five significant SNPs were associated to either the probability of infection or the parasite load , but not both , indicating that genetic variation underlying infection prevalence likely differs from that underlying infection intensity . The effect of two SNPs on infection phenotype was isolate-specific , suggesting that G x G interactions in this system likely occur at the gene level . This information will be useful to identify molecular targets for strategies aimed at interrupting malaria transmission during parasite development in the vector .
Ethical approval was obtained from the Cameroonian National Ethics Committee . All human volunteers were enrolled after written informed consent from the participant and/or their legal guardians . Mosquitoes and blood donors came from the vicinity of Yaoundé , a rainforest area in Cameroon , where the intensity of malaria transmission is relatively constant throughout the year , but slightly higher during the rainy seasons [72] . An A . gambiae s . s . colony was established in January 2006 from larvae collected in Ngousso , a suburb of Yaoundé and reared at the OCEAC insectary under standard conditions ( 12 h day/night cycle , 28 +/− 2°C , 85 +/− 5% humidity , adults maintained on 8% sucrose ) . The mosquito colony , named “Ngousso” , was used within 6 months from its establishment to limit the loss of polymorphism due to maintenance under laboratory conditions . The colony is of the M molecular form and Forest chromosomal form . Potential genetic structure in the mosquito colony was tested to confirm that the adult mosquitoes were freely interbreeding . Ten neutral microsatellite markers distributed throughout the genome were genotyped for 100 randomly selected mosquitoes from the colony and potential deviations from HWE measured in Genepop [73] and corrected for multiple testing using the Bonferroni procedure . The loci used were AG3H119 , AG3H242 , AG3H249 , AG3H555 , AG3H577 , AG3H59 , AG3H746 , AG3H812 , AG3H817 and AG3H93 [74] . P . falciparum gametocyte carriers were selected by examining thick blood smears from school children aged between five and eleven , who lived and attended school in Mfou , a small town located 30 km from Yaoundé . Malaria positive individuals were treated according to national recommendations . Up to 8 ml of venous blood was taken from selected carriers with at least 20 gametocytes/µl of blood ( estimated based on an average of 8000 white blood cells/µl ) . In order to limit the potential effect of human transmission blocking immunity [75] , the blood was first centrifuged at 2000 rpm at 37°C for three minutes and the serum changed to European naive AB serum with 0 . 225 UI heparin/ml ( to prevent clotting ) . 500 µl of reconstituted blood was added to membrane feeders maintained at 37°C by water jackets . Two to three day-old female mosquitoes were allowed to feed for up to 30 minutes through a Parafilm membrane . Un/partially fed mosquitoes were removed and fully fed mosquitoes maintained under standard conditions on an 8% sucrose diet . At day eight post infection , midguts were dissected in 0 . 4% mercurochrome stain and the infection load of each individual female was determined by counting oocysts under a light microscope and carcasses kept for genotyping . This procedure was repeated three times , each experimental infection using a different gametocyte carrier , referred to as Isolate 1 , 2 and 3 . DNA was extracted from an aliquot of the blood used in each infection using DNAzol ( Medical Research Centre ) and parasites were genotyped at the MSP1 M and MSP2 FC loci by measuring nested PCR fragment lengths as previously described [76] . Here , we used fluorescently labeled reverse primers and detected sizes on an Applied Biosystems 3130xl Sequencer . Names of the 67 genes included in this study are either from VectorBase ( http://www . vectorbase . org/ ) or published literature ( see Table S4 for corresponding VectorBase gene IDs ) . Genes were selected to represent the range of immune families previously characterized [10] with emphasis put on those implicated in the A . gambiae response to Plasmodium ( e . g . [14] ) and were amplified using previously published [37] , [63] , [64] and newly designed primers . Sequences were obtained from VectorBase and primers designed with Primer 3 [77] ( Table S4 ) to amplify approximately 700 bp upstream and/or within coding regions of each selected gene . These regions were amplified by PCR using 25 µl reaction mixes as previously described [63] for eight mosquitoes from the Ngousso colony . PCR products were sequenced using the Big Dye Terminator v3 . 1 Sequencing Kit ( Applied Biosystems ) , and run on an Applied Biosystems 3130xl Sequencer . Sequences were verified using SeqScape ( Applied Biosystems ) and aligned in Mega v3 . 1 [78] . 157 SNPs found more than once in the eight mosquitoes were selected . Genotyping single base pair extension primers were designed directly upstream of each SNP using Oligo Explorer and Oligo Analyser ( http://www . cmbn . no/tonjum/biotools-free-software . html ) . GACT repeat tails were added to genotyping primers to allow pooling of up to ten per reaction , and migration distances tested using the SNaPshot Primer Focus kit ( Applied Biosystems ) . For genotyping , DNA was isolated from the remaining mosquito carcasses after midgut dissection as previously described [79] and the Genomiphi kit applied ( Amersham ) for whole genome amplification in an unbiased manner [80] . The SNaPshot method ( Applied Biosystems ) was used to genotype mosquitoes before samples were run on an Applied Biosystems 3130xl Sequencer and results analysed using GeneMapper v4 . 1 software ( Applied Biosystems ) . For each genotyped SNP , deviations from HWE were determined as described above . The procedure consisted of three steps . Firstly , 192 females fed on Isolate 1 with an extreme phenotype [81] , of which 96 had 0–1 oocysts/midgut and 96 had 14+ oocysts/midgut , were genotyped for all 157 SNPs . The Kruskal-Wallis test was applied in R v . 2 . 10 . 0 [82] to detect significant genotype-phenotype associations . Specifically , this test compares oocyst counts between the three possible genotypic categories ( heterozygotes and both types of homozygote ) . Secondly , significant SNPs from the first step of the analysis were genotyped in all of the remaining females ( n = 185 ) fed on Isolate 1 that had intermediate phenotypes ( 1–14 oocysts/midgut ) . Genotyping data from all individuals fed on Isolate 1 were combined into a full data set for the significant SNPs . These were reanalyzed using the Kruskal-Wallis test . An FDR analysis [83] , [84] was used on the final P-values of per-SNP Kruskal-Wallis tests , with the original FDR procedure [85] applied at the 5% level . We therefore increased the sample size for the SNPs that were most notable in the 192 initial individuals before applying the FDR analysis . Although this procedure increases the detection power for SNPs with strong effects , it is conservative because it tends to decrease significance when there is no true genotype-phenotype association . The basic genotype-phenotype analysis for SNP filtering required up to this point was based on univariate , non-parametric tests . Thirdly , significant SNPs following the first two steps of the analysis ( based on genotype-phenotype associations from a single P . falciparum isolate ) were genotyped for all mosquitoes fed on Isolates 2 and 3 , giving a data set including these SNPs across all three parasite isolates . Infection phenotype was decomposed into prevalence ( proportion of mosquitoes with at least one oocyst ) and intensity ( number of oocysts in individuals with at least one oocyst ) . Prevalence was analyzed by binomial logistic models and intensity in infected individuals by linear models . In the latter analysis , oocyst numbers were square-root transformed to achieve normality of the residuals . For each individual SNP , both components of the infection phenotype were analyzed as a function of the mosquito genotype , the parasite isolate , and their interaction . Effects were tested by standard analysis of variance . To avoid multiple testing issues , each SNP was analyzed in a stepwise manner , starting with an overall test by Fisher's combination of probabilities method [86] , combining the P-values of the intensity and prevalence analyses , and further analyzing significant results in terms of their components ( prevalence or intensity ) . Likewise , in each case , the complete model ( SNP and Isolate effects with interaction ) was compared to the model with Isolate effect alone to obtain a single test of the total SNP effect ( resulting from both additive and interaction effects ) . A total SNP effect means that the oocyst distribution ( prevalence and/or intensity ) differs depending on the SNP genotype . The total SNP effect was further analyzed by stepwise deletion of effects . When an Isolate x SNP interaction was statistically significant ( P<0 . 05 ) , the model with interaction was retained . An interaction effect means that the SNP effect on the oocyst distribution ( prevalence and/or intensity ) differs depending on the parasite isolate . When no significant interaction was detected , a SNP additive effect , i . e . measuring the same trend across isolates , was tested in a model with Isolate effect . All analyses were performed with the functions lm , glm and anova in the R software [82] . For graphical representation , when the SNP x Isolate interaction was not statistically significant , phenotypic values were corrected for the main effect of the isolate so that visual differences could be directly attributed to the genotypes . For infection prevalence the proportion was standardized by isolate , whereas for infection intensity we plotted the residuals of a one-way analysis of oocyst numbers as a function of isolate . When the SNP x Isolate interaction was statistically significant , no correction was made and the raw data was plotted separately for each isolate within the same graph . LD between SNP pairs was measured for each chromosome arm as r2 in Haploview [87] . A cut-off of r2 = 0 . 8 [45] was used to identify SNP pairs in high LD and estimate the accuracy , in terms of genetic distance , of the associations found . The gene Sp SNAKElike was functionally tested for its role against P . falciparum . Double stranded RNA ( dsRNA ) was produced using the T7 Megascript Kit ( Ambion ) as described previously [7] , [68] . cDNA was obtained using Sp SNAKElike primers ( SNL Inner F: 5′-TTGCACTGGTGAAGCTCAAG-3′; SNL Inner R: 5′-CCTTCGTGTAGATCGCCTTC-3′ ) with A . gambiae Ngousso RNA as a template . Double stranded LacZ RNA was used as a control . For both genes , Sp SNAKElike and LacZ , 200 ng of dsRNA was injected into the thorax of 1-2 day-old mosquitoes anesthetized with CO2 , using a nano-injector ( Nanoject II , Drummond Scientific ) [7] . Up to 100 dsLacZ and dsSp SNAKElike injected mosquitoes were experimentally challenged with an infectious blood meal four days post injection as described above and oocysts counted eight days later . A total of ten feedings with P . falciparum gametocyte positive blood isolates were performed . Only experiments with at least 20 live mosquitoes eight days after feeding in each treatment were included in the analysis . Data from the gene knockdown assays were analyzed with non-parametric tests because no transformation of oocyst counts yielded normally-distributed residuals after linear modeling . Infection phenotype was analyzed in two steps . Firstly , total oocyst counts were compared with separate Wilcoxon Mann-Whitney tests for each isolate and P-values were combined using Fisher's meta-analysis approach [86] . Secondly , infection phenotype was decomposed into prevalence and intensity ( as described above ) , which were compared for each isolate with an exact chi-square test for contingency tables and a Wilcoxon Mann-Whitney test , respectively . P-values were combined for each isolate using Fisher's meta-analysis approach . All performed in R [82] . Gene knockdown success was confirmed by semi-quantitative PCR from mosquitoes collected four days after dsRNA injection and prior to feeding . Total RNA was extracted from 15 mosquitoes using Trizol reagent ( Invitrogen ) and cDNA was synthesized using the SuperScript III Reverse Transciptase Kit and an oligo ( dT ) 20 primer ( Invitrogen ) . The A . gambiae S7 ribosomal gene was used to normalize the amount of RNA between knockdown and control mosquitoes . Semi-quantitative PCRs were conducted using the primers SNL Outer F 5′-ACGCTAATACCGCTCACGAT-3′ and SNL Outer R 5′-CCCCACACGTTGTCCTCTAT-3′ for Sp SNAKElike and S7 F 5′-AGGCGATCATCATCTACGTGC-3′ and S7 R 5′-GTAGCTGCTGCAAACTTCGG-3′ for S7 .
|
Anopheles gambiae is the main malaria vector in Africa , transmitting the parasite when it blood feeds on human hosts . The parasite undergoes several developmental stages in the mosquito to complete its life cycle , during which time it is confronted by the mosquito's immune system . The resistance of mosquitoes to malaria infection is highly variable in wild populations and is known to be under strong genetic control , but to date the specific genes responsible for this variation remain to be identified . The present study uncovers variations in A . gambiae immune genes that are associated with natural resistance to Plasmodium falciparum , the deadliest human malaria parasite . The association of some mosquito genetic loci with the level of infection depended on the P . falciparum isolate , suggesting that resistance is determined by interactions between the genome of the mosquito and that of the parasite . This finding highlights the need to account for the natural genetic diversity of malaria parasites in future research on vector-parasite interactions . The loci uncovered in this study are potential targets for developing novel malaria control strategies based on natural mosquito resistance mechanisms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/genetics",
"of",
"the",
"immune",
"system",
"genetics",
"and",
"genomics/complex",
"traits",
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"immunology/genetics",
"of",
"the",
"immune",
"system",
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2010
|
Polymorphisms in Anopheles gambiae Immune Genes Associated with Natural Resistance to Plasmodium falciparum
|
Quorum sensing ( QS ) is a bacterial cell-cell communication process that relies on the production and detection of extracellular signal molecules called autoinducers . QS allows bacteria to perform collective activities . Vibrio cholerae , a pathogen that causes an acute disease , uses QS to repress virulence factor production and biofilm formation . Thus , molecules that activate QS in V . cholerae have the potential to control pathogenicity in this globally important bacterium . Using a whole-cell high-throughput screen , we identified eleven molecules that activate V . cholerae QS: eight molecules are receptor agonists and three molecules are antagonists of LuxO , the central NtrC-type response regulator that controls the global V . cholerae QS cascade . The LuxO inhibitors act by an uncompetitive mechanism by binding to the pre-formed LuxO-ATP complex to inhibit ATP hydrolysis . Genetic analyses suggest that the inhibitors bind in close proximity to the Walker B motif . The inhibitors display broad-spectrum capability in activation of QS in Vibrio species that employ LuxO . To the best of our knowledge , these are the first molecules identified that inhibit the ATPase activity of a NtrC-type response regulator . Our discovery supports the idea that exploiting pro-QS molecules is a promising strategy for the development of novel anti-infectives .
Quorum sensing ( QS ) is a process of bacterial cell-cell communication that relies on the production , release , detection , and response to extracellular signaling molecules called autoinducers . QS allows groups of bacteria to synchronously alter behavior in response to changes in the population density and species composition of the vicinal community . QS controls collective behaviors including bioluminescence , sporulation , virulence factor production , and biofilm formation ( Reviewed in [1] , [2] ) . Impairing virulence factor production or function has gained increasing attention as a method to control bacterial pathogenicity . The advantage of anti-virulence strategies over traditional antibiotics is presumed to be reduced pressure on bacteria to develop resistance [3]–[5] . Because QS controls virulence in many clinically relevant pathogens , disrupting QS is viewed as a promising possibility for this type of novel therapeutic development [6]–[8] . Many pathogenic Gram-negative bacteria use acylhomoserine lactones ( HSLs ) as QS autoinducers , which are detected by either cytoplasmic LuxR-type or membrane-bound LuxN-type receptors [9] . To date , efforts to interfere with HSL QS in Gram-negative bacteria have yielded several potent antagonists [10]–[15] . While these strategies are exciting , some globally important Gram-negative pathogens do not use HSLs as autoinducers . Thus , additional strategies that target non-HSL based QS systems are required . Here , we describe the identification and characterization of a set of small-molecule inhibitors that act on the non-HSL QS system of Vibrio cholerae by targeting two independent steps in the signal transduction pathway . V . cholerae is the etiological agent of the disease cholera and its annual global burden is estimated to be several million cases [16] . V . cholerae produces and detects two QS autoinducer molecules called CAI-1 and AI-2 . CAI-1 ( ( S ) -3-hydroxytridecan-4-one ) is produced by the CqsA synthase [17] , [18] and AI-2 ( ( 2S , 4S ) -2-methyl-2 , 3 , 3 , 4-tetrahydroxytetrahydrofuran borate ) is produced by the LuxS synthase [19] , [20] . Detection of CAI-1 and AI-2 occurs through transmembrane receptors CqsS and LuxPQ , respectively [21] , [22] . CqsS and LuxPQ are two-component proteins that possess both kinase and phosphatase activities ( Figure 1 shows the CqsA/CqsS system ) . At low cell density ( LCD ) , when the receptors are devoid of their respective ligands , their kinase activities predominate , resulting in the phosphorylation of the response regulator LuxO . LuxO∼P is the transcriptional activator of four genes encoding small regulatory RNAs ( sRNAs ) , Qrr1-4 [23] . The Qrr sRNAs target the mRNAs encoding the quorum-sensing master transcriptional regulators AphA and HapR . At LCD , facilitated by the RNA chaperone Hfq , Qrr1-4 stabilize and destabilize the aphA and hapR mRNA transcripts , respectively [23] . Therefore , AphA protein is made while HapR protein is not ( Figure 1 ) . When autoinducer concentration increases above the threshold required for detection ( which occurs at high cell density ( HCD ) ) , binding of the autoinducers to their cognate receptors switches the receptors from kinases to phosphatases ( Figure 1 ) . Phosphate flow through the signal transduction pathway is reversed , resulting in dephosphorylation and inactivation of LuxO . Therefore , at HCD , qrr1-4 are not transcribed , resulting in cessation of translation of aphA and derepression of translation of hapR . This QS circuitry ensures maximal AphA production at LCD and maximal HapR production at HCD . AphA and HapR each control the transcription of hundreds of downstream target genes [24] , [25] . Hence , reciprocal gradients of AphA and HapR establish the QS LCD and HCD gene expression programs , respectively ( Figure 1 ) . In pathogens that cause persistent infections , QS commonly activates virulence factor production at HCD . However , in V . cholerae , which causes an acute disease , HapR production at HCD represses genes important for biofilm formation and virulence factor production [22] , [26]–[30] . This peculiar pattern of virulence gene regulation can be understood in terms of the disease caused by V . cholerae [31] . Following successful V . cholerae infection , the ensuing diarrhea washes huge numbers of bacteria from the human intestine into the environment . Thus , expression of genes for virulence and biofilm formation at LCD promotes infection , while repression of these genes by autoinducers at HCD promotes dissemination . Thus , molecules that activate QS have the potential to repress virulence in V . cholerae . Moreover , QS plays an essential role in virulence in other pathogenic vibrios including Vibrio parahaemolyticus , Vibrio alginolyticus , and Vibrio vulnificus [32]–[35] . The components of the QS circuits in these species are similar to those of V . cholerae . Therefore , QS-activating molecules identified for V . cholerae could be broadly useful for controlling diseases caused by other vibrios . Here , we report the identification of a set of small molecules that activate the QS system of V . cholerae . We classify the QS-activating molecules as either QS receptor agonists or LuxO inhibitors . Because we have already reported analyses of QS receptor agonists , we focus here on the LuxO inhibitors . At LCD , LuxO∼P activates production of the Qrr sRNAs , which repress HapR; inhibitors of LuxO thus activate QS due to derepression of HapR . LuxO belongs to the NtrC protein family , σ54-binding transcriptional activators that rely on ATP hydrolysis to promote open complex formation [36] . The LuxO inhibitors identified here function uncompetitively to perturb LuxO ATPase activity . Genetic analysis of LuxO mutants that are insensitive to the inhibitors suggests that the inhibitors interact with a region adjacent to the ATP binding pocket . Finally , using a set of phenotypic assays , we show that the inhibitors broadly activate different vibrio QS circuits and , in turn , repress virulence factor production and reduce cytotoxicity . Because LuxO is conserved among vibrio QS circuits , the molecules we characterize here are capable of inhibiting HSL-based and non-HSL-based vibrio QS systems . Numerous NtrC-type proteins homologous to LuxO act in two-component signaling systems and their roles in controlling nitrogen metabolism , virulence , motility , and other important processes have been extensively studied ( Reviewed in [37] ) . To the best of our knowledge , there exists no previous report of a chemical probe that modulates the activity of a NtrC-family response regulator .
We are interested in identifying small molecules that activate QS in V . cholerae , in order to induce the HCD state and thus repress virulence factor production . To do this , we developed a whole-cell high-throughput screen that relies on QS-dependent induction of bioluminescence ( lux ) in V . cholerae [22] . We exploited V . cholerae mutants genetically locked into the LCD state and carrying the lux operon from V . harveyi to screen for molecules that induce light production , indicating that they activate QS responses . We performed the screen on two different LCD mutants . The first mutant lacks the two autoinducer synthases , CqsA and LuxS . Therefore , both CqsS and LuxPQ QS receptors function as kinases and constitutively phosphorylate LuxO , resulting in transcription of the Qrr regulatory RNAs , and repression of translation of HapR ( see INTRODUCTION ) . In the absence of HapR , there is no transcription of the heterologous lux operon , and thus , this strain is dark . The second strain carries the luxOD47E allele . This luxO mutation mimics LuxO∼P , rendering LuxO constitutively active [23] , [38] . Therefore , HapR is repressed and the strain is dark . We anticipated identifying two classes of molecules that could induce light production: Class 1 ) Molecules that induce bioluminescence in the double synthase mutant but not in the luxOD47E mutant . These compounds are predicted to be QS receptor agonists; and Class 2 ) Molecules that induce bioluminescence in both the double synthase mutant and the luxOD47E mutant . Class 2 compounds likely target QS components that lie downstream of the receptors . We screened 90 , 000 molecules and identified eight Class 1 compounds and three Class 2 compounds ( Figures 2A and 2B ) . The EC50 of Class 1 compounds are comparable to that of CAI-1 and generally lower than those of Class 2 compounds ( Figure 2C ) . These differences support the idea that the two classes of molecules potentiate QS responses by distinct mechanisms . None of the compounds affected cell growth ( Figure S1 ) . To determine which QS component each compound acts on , we first tested the eight Class 1 compounds against V . cholerae mutants that lack only the CqsS receptor or only the LuxPQ receptor . All eight Class 1 compounds induced light production in the ΔluxPQ strain but not the ΔcqsS strain; hence , these eight molecules function as CqsS agonists ( Figure S2 ) . Interestingly , none has structural homology to the native CAI-1 autoinducer [17] , [18] , [39] , [40] ( Figure 2A ) . The Class 1 molecules are currently being characterized and are not discussed further here . The three Class 2 compounds that activate QS in both of the LCD screening strains likely act downstream of the QS receptors . These three compounds are structurally homologous ( Figure 2A ) ; therefore , they may function by an identical mechanism . Here , we focused on the compound displaying the highest potency ( i . e . , compound 11 , Figures 2A and 2C ) . Class 2 compounds could potentially target one or more of the V . cholerae QS cytoplasmic components that function downstream of the receptors: LuxO , σ54 , Hfq , and/or Qrr1-4 . We reasoned that if these compounds interfere with LuxO or σ54 , transcription of qrr1-4 would decrease in the presence of the inhibitors . By contrast , if the compounds target Hfq or act directly on Qrr1-4 , they should not affect qrr1-4 transcription . GFP production from a qrr4-gfp transcriptional fusion decreased ∼3-fold when the luxOD47E strain was treated with compound 11 ( Figure 2D ) . This result suggests that compound 11 targets either LuxO or σ54 . If the target of compound 11 is σ54 , transcription of other σ54-dependent genes should be affected when V . cholerae is treated with the compound . We examined transcription of the σ54-dependent gene vpsR [41] and found that it did not change significantly in the presence of compound 11 ( data not shown ) . These results suggest that compound 11 targets LuxO . The three identified Class 2 compounds share a 5-thio-6-azauracil core and only their side chains vary ( Figure 2A ) . In addition , several 5-thio-6-azauracil analogs with other modifications on their side chains displayed weak or no activity in the screen . Therefore , differences in the hydrocarbon side chains must be responsible for the corresponding differences in potency with compounds harboring branched side chains displaying greater potency ( i . e . , compound 11 , Figure 2C ) . To explore the relationship between structure and activity , we synthesized a focused library of compounds bearing the conserved 5-thio-6-azauracil core , and we altered the branching in the side chains . We measured activities using bioluminescence in the V . cholerae luxOD47E mutant . Several of the side chain modifications decreased potency ( as shown by an increase in EC50 , Figure 3 ) . However , increasing steric bulk by incorporation of a tert-butyl carbinol side chain led to a 3-fold enhancement in potency ( i . e . , compound 12 , Figure 3 ) . Thus , the activity of the 5-thio-6-azauracil compounds within this series is highly sensitive to the structural features of the alkyl side chain . In the focused group of molecules we investigated , a bulky , hydrophobic terminal t-butyl moiety is optimal . NtrC-type response regulators including LuxO possess three biochemical activities: phosphoryl-group accepting activity , DNA-binding activity , and ATP hydrolyzing activity [36] . We investigated which of these activities is inhibited by compounds 11 and 12 . First , using whole-cell bioluminescence assays , we found that both compounds activate QS in V . cholerae strains expressing either wild type LuxO or LuxO D47E ( Figures 2B and 3 ) . Wild type LuxO is activated by phosphorylation via the QS cascade , and the LuxO D47E variant , which mimics LuxO∼P , while not phosphorylated is constitutively active [22] , [23] , [26] , [38] . Because both wild type LuxO and LuxO D47E are vulnerable to inhibition , it cannot be the ability of LuxO to participate in phosphorylation or dephosphorylation that is impaired by compounds 11 and 12 . LuxO , as a NtrC-type response regulator , binds to σ54-dependent promoters to activate transcription . Compounds 11 and 12 could prevent LuxO from binding to DNA , and in so doing , prevent qrr transcription . To investigate this possibility , we used electrophoretic-mobility-shift and fluorescence anisotropy assays to probe the LuxO interaction with qrr promoter DNA . Even in the presence of a high concentration ( 200 µM ) of the inhibitors , no significant change in LuxO D47E binding to qrr4 promoter DNA occurred as judged by mobility shift ( Figure 4A ) . Quantitative fluorescence anisotropy assays revealed that , in the presence and absence of the LuxO inhibitors , LuxO D47E interacts with the qrr4 promoter DNA with an identical binding constant ( ∼300 nM ) ( Figure 4B ) . Thus , binding to DNA is not altered by the inhibitors . Finally , we examined whether compounds 11 and 12 affect LuxO ATPase activity . To do this , we used a coupled-enzyme assay [42] to assess the rate of ATP hydrolysis by LuxO in the presence and absence of the compounds . Both compounds inhibit ATP hydrolysis in a dose-dependent manner ( Figures 5A–C ) . Using traditional Michaelis-Menton enzyme kinetic analyses , we found that both compounds decrease the Km and the Vmax of the LuxO ATPase reaction ( Figures 5B and 5C ) . The Lineweaver-Burk plots of curves derived from control reactions and from inhibitor-containing reactions display parallel slopes ( Km/Vmax ) , indicating that compounds 11 and 12 function as uncompetitive inhibitors ( Figures 5B and 5C ) , suggesting they bind to the pre-formed LuxO-ATP complex to inhibit ATP hydrolysis . Indeed , inhibition of LuxO ATPase by the analogs we identified or synthesized ( as represented by % inhibition ) is correlated with their potency ( EC50 ) in inducing QS in the luxOD47E mutant ( Figure 5D ) . We conclude that the LuxO inhibitors discovered here activate QS in V . cholerae by specifically inhibiting the ATPase activity of LuxO . Presumably , in the presence of the inhibitors , LuxO is incapable of participating in open complex formation at the qrr promoters , which prevents transcription of the Qrr sRNAs . In turn , translation of HapR is derepressed and the QS response occurs prematurely . Compounds 11 and 12 likely bind to LuxO at an allosteric site that negatively regulates ATP hydrolysis activity . To determine where compounds 11 and 12 bind , we screened for LuxO mutants refractory to inhibition . To do this , we engineered random mutations into the cloned luxOD47E gene and introduced the mutant library into a V . cholerae ΔluxO strain carrying the lux operon . We screened for clones that conferred a dark phenotype in the presence of compound 12 , hypothesizing that such mutants harbor alterations in the inhibitor binding-site . Four such mutants were identified ( Figure 6A ) . These LuxO D47E variants all possess an active ATPase and are functional , as judged by their ability to repress light production in the absence of inhibitor ( Figure 6A ) . Sequencing revealed that the four LuxO D47E mutants carry I211F , L215F , L242F , or V294L alterations , implicating these residues as important for binding of the inhibitors . We mapped these four alterations onto the existing crystal structure of ATP-bound Aquifex aeolicus NtrC1 ( PDB:3M0E ) [43] , which has high sequence homology to LuxO ( Figure 6B ) . The four residues we identified in the screen map to three regions that abut the Walker B motif ( D245 , E246 , L247 , and C248 in LuxO ) ( Figure 6B ) . In other NtrC-type proteins , mutations in this region have been shown to prevent ATP hydrolysis ( See DISCUSSION ) . These four luxO mutations were introduced into wild type LuxO and the resulting mutants are similarly resistant to inhibition ( Figure S3 ) . Thus , binding of compounds 11 and 12 to this region may induce a conformational change in the nearby ATP-binding pocket that inhibits ATP hydrolysis . As mentioned , LuxO is a conserved member of vibrio QS circuits . We therefore wondered if , similar to what we found in V . cholerae , compounds 11 and 12 could activate QS in other Vibrio species . To test this idea , we exploited two well-characterized phenotypes controlled by QS: light production in V . harveyi and colony opacity in Vibrio parahaemolyticus [44]–[46] . In V . harveyi , light production is induced by QS and a V . harveyi luxOD47E mutant is dark . Treatment of V . harveyi luxOD47E with compounds 11 and 12 induced light production 10 , 000-fold , indicating that these compounds are indeed active in V . harveyi ( Figure 7A ) . In V . parahaemolyticus , the HapR ortholog , OpaR , controls colony opacity . OpaR production is repressed at LCD by LuxO∼P via the V . parahaemolyticus Qrr sRNAs . V . parahaemolyticus mutants that produce low and high levels of OpaR form translucent and opaque colonies , respectively [32] , [46] . Thus , V . parahaemolyticus is naturally translucent at LCD and opaque at HCD . McCarter et al [32] recently identified a constitutively active LuxO mutant ( LM4476 , luxO* ) in V . parahaemolyticus that confers a constitutive translucent colony morphology ( Figure 7B , left ) . By contrast , an isogenic V . parahaemolyticus ΔluxO strain ( LM9688 ) forms opaque colonies ( Figure 7B , left ) . When the luxO* mutant is plated on medium containing compound 11 or compound 12 , the colonies switch from translucent to opaque , a phenotype indistinguishable from the ΔluxO mutant ( Figure 7B , right ) . These results suggest that compounds 11 and 12 inhibit V . parahaemolyticus LuxO from repressing the OpaR-dependent QS program . We conclude that the LuxO inhibitors identified in this study are broadly capable of activating QS in Vibrio species that employ LuxO as the central QS regulator . In pathogenic vibrios , HapR and its homologs ( e . g . , V . parahaemolyticus OpaR and V . vulnificus SmcR ) function as repressors of virulence factor production at HCD [32]–[34] . For example , in V . cholerae , the genes encoding the key V . cholerae virulence factors , the CTX toxin and the Toxin Co-regulated Pilus ( TCP ) , are targets of HapR repression at HCD [17] , [27] , [30] . V . parahaemolyticus uses Type Three Secretion Systems ( TTSS ) for pathogenesis , and at HCD , OpaR represses the expression of one of the TTSS operons ( TTSS-1 ) [32] , [47] . Thus , luxO mutants that constitutively produce HapR ( V . cholerae ) or OpaR ( V . parahaemolyticus ) are attenuated in virulence [22] , [30] , [32] . The previous section shows that our LuxO inhibitors are active in multiple vibrios . To test whether the inhibitors can disrupt the QS-controlled virulence outputs of pathogenic vibrios , we assayed their effects on TcpA production in V . cholerae and production and secretion of VopD , a TTSS-1 effector protein , in V . parahaemolyticus . Western blot analysis showed that , in a V . cholerae luxOD47E strain , HapR and TcpA levels increased and decreased , respectively , in the presence of compound 12 ( Figure 8A ) . Likewise , exposing the V . parahaemolyticus luxO* mutant to compound 12 resulted in decreased production and secretion of VopD ( Figure 8B ) . To begin to explore whether repression of these in vitro virulence phenotype translates to repression of the in vivo phenotype , we exploited an established V . parahaemolyticus cytotoxicity assay [48] to investigate whether pathogenicity could be inhibited by treatment with the LuxO inhibitors . We infected cultured HeLa cells with the untreated or compound 12-treated V . parahaemolyticus luxO* mutant and assayed HeLa cell lysis by measuring lactate dehydrogenase released from the host cytoplasm . Consistent with a previous report [32] , the V . parahaemolyticus luxO* mutant is more cytotoxic to HeLa cells than the isogenic ΔluxO mutant ( Figure 8C ) . At 2 to 3 hours post-infection , HeLa cell lysis was significantly lower in samples infected with the luxO* mutant treated with compound 12 than in samples infected with the luxO* mutant that had not been treated ( average cytotoxicity is ∼30% and ∼100% for treated and untreated , respectively , p<0 . 01 ) . At that time point , the cytotoxic capability of the Compound 12-treated luxO* mutant is slightly higher than that of the isogenic ΔluxO mutant ( Figure 8C ) . At 4-hour post-infection , the compound 12-treated luxO* mutant was equally toxic ( ∼100% ) as the untreated the luxO* mutant , while the ΔluxO mutant caused only ∼60% HeLa cells lysis . This residual cytotoxicity is consistent with earlier results showing that the ΔluxO mutant is not completely impaired for cytotoxicity [32] . Thus , the level of in vitro inhibition of TTSS-1 ( Figure 8B ) is a good indicator of the ex vivo inhibition of cytotoxicity ( Figure 8C ) . The increase in cytotoxicity in Compound 12-treated V . parahaemolyticus that occurred at late time points could be due to incomplete inhibition of LuxO , uptake , or degradation of the compound by the HeLa cells . Nonetheless , the progression of V . parahaemolyticus killing of mammalian cells is impaired by compound 12 , consistent with the notion that virulence factor production can be controlled by small molecule inhibitors of LuxO .
As part of a continuing effort to identify molecules that modulate QS in bacteria , we have identified two classes of molecules that activate QS in V . cholerae . These newly identified molecules serve two important purposes . First , they can be used as novel chemical probes to study QS signal transduction mechanisms . Second , from a practical standpoint , because QS represses virulence factor production in many pathogenic Vibrio species , molecules that activate QS , which decreases virulence , have the potential to be developed into anti-virulence agents to combat infectious diseases caused by pathogenic vibrios . The first class of molecules identified here acts on the V . cholerae CqsS receptor . These molecules , surprisingly , do not resemble the native CAI-1 family of ligands ( Figure 2A ) . Previous studies revealed that CqsS receptors from different vibrios possess distinct ligand detection specificities . The V . cholerae receptor is promiscuous in detecting a range of CAI-1-type molecules , while the V . harveyi receptor is relatively stringent [39] . Interestingly , none of the Class 1 molecules identified here activates QS in V . harveyi , lending support to the idea that CqsS receptors , although sharing extensive homology , possess different overall stringencies for ligands . We altered a single specificity-determining residue in the V . cholerae CqsS receptor ( Cys 170 ) to the corresponding amino acid ( Phe ) in the V . harveyi receptor . This alteration is sufficient to increase stringency in detection of CAI-1 type molecules [39] , [49] , however , it did not abolish detection of the Class 1 molecules ( Figure S4 ) . Identification of CqsS receptor mutants with altered selectivity to the Class 1 molecules will provide additional insight into the molecular basis of ligand-CqsS interactions . The second class of molecules identified , and the focus of this work , act on LuxO , the central QS regulator that controls transcription of the four Qrr sRNA genes . LuxO , which is a member of the NtrC family of two-component response regulators , possesses an N-terminal regulatory receiver domain , a central ATPase domain ( AAA+ type ) , and a C-terminal DNA-binding domain . Three inhibitors have previously been identified that target non-NtrC type response regulators , AlgR1 of Pseudomonas aeruginosa [50] , WalR in low-GC Gram-positive bacteria [51] , and DevR in Mycobacterium tuberculosis [52] . The molecules function by perturbing phosphorylation ( AlgR1 and WalR ) and DNA binding ( DevR ) . Our LuxO inhibitors , by contrast , function by an uncompetitive mechanism , presumably by binding to the pre-formed LuxO-ATP complex to prevent ATP hydrolysis . Thus , multiple families of response regulator can be selectively inhibited using small molecules . Furthermore , all three known response regulator activities; phosphorylation , DNA binding , and ATPase , are potential targets for inhibition . Analyses of LuxO inhibitor-resistant mutants suggest that our inhibitors bind to a region close to the predicted Walker B motif . Additional support for this idea comes from studies of other NtrC-type proteins , which show that mutations that affect ATP hydrolysis but do not interfere with ATP binding also map to the Walker B motif and to amino acid residues preceding the conserved GAFTGA domain [43] , [53] , [54] . Indeed , one of the LuxO inhibitor-resistant mutations identified here ( L242F ) lies immediately upstream of the predicted Walker B motif , while both the I211F and L215F mutations map to the helix containing the GAFTGA domain . In addition , the residue identified in the final inhibitor-resistant mutant , V294L , is predicted to sit facing the putative catalytic arginine ( R306 ) . The GAFTGA domain is important for interaction with the σ54-RNAP holoenzyme [55] . Thus , it was possible that the mutations we isolated in this region ( I211F and L215F ) suppress inhibition by compounds 11 and 12 by stabilizing the LuxO-σ54-RNAP interaction without affecting inhibitor binding . If this were the case , the ATPase activity of the purified LuxO D47E/I211F and D47E/L215F variants would be inhibited by these compounds . However , we purified LuxO D47E/I211F protein and found that the ATPase activity is not inhibited ( Figure S5 ) . This result is consistent with the idea that these mutations abolish inhibitor binding and , in so doing , prevent ATP hydrolysis . High sequence conservation in the ATPase domain exists between different NtrC-type response regulator family members . Thus , we were interested to test whether the LuxO inhibitors could inhibit other NtrC-type response regulators . Compounds 11 and 12 only modestly inhibit ( ∼10% ) the ATPase activity of purified E . coli NtrC at 250 µM ( a concentration at which >80% of the LuxO ATPase activity is inhibited , Figure S6 ) . This finding is surprising because the key residues ( I211 , L215 , L242 , and V294 ) that , when mutated , confer resistance to the inhibitors in LuxO are all present in E . coli NtrC . Thus , NtrC must possess additional structural features that render it resistant to inhibition . Structural comparisons between these two related RRs , coupled with identification of inhibitor-sensitive NtrC mutants , should allow us to understand the basis of the differences in inhibitor sensitivity . Two-component signaling ( TCS ) proteins are widely distributed in bacteria . In addition to their global importance in microbial physiology , the absence of TCSs in mammalian cells makes them attractive drug targets in pathogenic bacteria [56] , [57] . Even though significant effort has been devoted to identifying novel TCS inhibitors , to date , none has been developed into a new class of anti-infective . Problems such as undesirable properties associated with lead molecules have been encountered [56] , [57] . In particular , inhibitors that generally target the conserved hydrophobic kinase domains of TCS histidine kinases suffer from drawbacks such as low cell permeability , poor selectivity , and unfavorable non-specific off-target effects ( e . g . membrane damaging ) [58]–[60] . By contrast , approaches to target the sensory domains of histidine kinases have yielded a handful of promising TCS inhibitors . For instance , LED209 , an antagonist of the QseC histidine kinase , which regulates motility and pathogenicity in enterohaemorrhagic E . coli , reduces virulence in several pathogens both in vitro and in vivo [61] . In addition , in Staphylococcus aureus , inhibitory Agr peptide analogs antagonize the AgrC histidine kinase receptors and block abscess formation in an experimental murine model [62] . Targeting response regulators as a broad-spectrum anti-infective strategy has been considered challenging because response regulator functions , such as phosphorylation and DNA binding , are thought to be specific . In spite of this , a handful of molecules that inhibit particular response regulator functions have been reported [50]–[52] . For example , as mentioned , Walrycins , molecules that inhibit the phosphorylation of the essential WalR response regulator , are active in suppressing growth in multiple Gram-positive bacteria [51] . In the context of our work , the ATPase domain is highly conserved between all members of the NtrC response regulator family . Therefore , molecules that specifically target the ATPase domain of a single response regulator in this family ( e . g . , LuxO ) could potentially be developed into general inhibitors of NtrC-family of proteins . Because NtrC-type proteins control virulence , nitrogen metabolism , motility , and other vital processes in bacteria [37] , targeting the ATPase domain offers an additional route for anti-TCS drug development . The LuxO inhibitors identified here possess certain favorable drug-like characteristics: potent inhibition , water-solubility , good stability , and cell-permeability . The molecules also display low host-cell cytotoxicity ( undetectable cytotoxicity at 500 µM ) . These broadly-active LuxO inhibitors are not broad-spectrum NtrC-type inhibitors . Microarray analyses reveal that fewer than 40% of genes affected by the inhibitors are non-LuxO targets ( data not shown ) . Nonetheless , our LuxO inhibitors could be used as preliminary scaffolds for building a general NtrC-type RRs inhibitor . Future improvements to these molecules will be focused on the structure-activity relationships of the thio-azauracil core , combined with simultaneously screening for molecules that inhibit LuxO and other NtrC type response regulators . Although NtrC is not affected by the inhibitors discovered here , multiple LuxO response regulators from different Vibrio species are targeted by our inhibitors . Vibrio species detect a wide array of autoinducers ( HSLs , CAI-1 , and AI-2 ) , thus , molecules that interrupt QS in Vibrio species by targeting the cognate receptors/synthases are likely to be autoinducer-specific and will have a limited spectrum . By contrast , because LuxO is nearly identical in all Vibrio species , our inhibitors can broadly activate vibrio QS irrespective of what type of autoinducer is detected . More importantly , we showed here that treatment of V . cholerae and V . parahaemolyticus with the LuxO inhibitors reduces virulence factor production and impedes cytotoxicity . Thus , our LuxO inhibitors , upon refinement , can at a minimum be used broadly to control virulence factor production in a variety of Vibrio species that use QS to repress pathogenesis . The central ATPase module of the NtrC-type RR is classified as AAA+ type [63] . This module is present in multiple domains of life . For example , AAA+ ATPases are important in functions including protein unfolding and degradation ( ClpXP , FtsH , and p97 ) , organelle function and maintenance ( PEX1 and VPS4 ) , replication and recombination ( RuvBL1 and helicases ) , and intracellular transport ( Dyneins ) . Some eukaryotic AAA+ ATPases have been proposed to be drug targets [64] . Therefore , it will be particularly fascinating to investigate whether the thio-azauracil core discovered here can be developed into an inhibitor of AAA+ ATPases across different domains . Antagonizing QS in bacteria represents a promising new approach that is an alternative to traditional antibiotics [8] , [12] , [14] , [15] , [61] , [65] . Likewise , using pro-QS agents to treat acute infections , in which bacteria use QS to repress virulence , should be further explored . Using the native CAI-1 ligand , we previously showed that V . cholerae virulence factor production is repressed in vitro [17] . In the same vein , we show here that our synthetic pro-QS molecules reduce virulence by inhibiting LuxO . March et al reported that pretreatment with commensal E . coli over-producing the V . cholerae autoinducer CAI-1 increases the survival rate of mice following V . cholerae infection [66] , which further supports the idea of QS potentiators as drugs . Use of CAI-1 , LuxO inhibitors , or other QS-activating molecules as prophylactics could conceivably prevent V . cholerae or other pathogenic vibrios from initiating the LCD virulence gene expression program that is required for colonization . In this scenario , inhibiting the launch of virulence factors would provide sufficient time for the host immune system to eliminate the pathogen . In contrast to traditional antibiotics that target essential bacterial processes , growth is not affected by interfering with QS , so development of resistance could potentially be minimized [8] , [14] .
All V . cholerae strains are derivatives of wild type C6706str [67] . All V . harveyi strains are derivatives of wild type V . harveyi BB120 [68] . V . parahaemolyticus strains were generously provided by Dr . Linda McCarter . Escherichia coli S17-1 pir , DH5α , and Top10 were used for cloning . The relevant genotypes of all plasmids and strains are provided in Supporting Table S1 . Unless specified , E . coli and V . cholerae were grown in LB medium at 37°C and 30°C with shaking , respectively . V . harveyi and V . parahaemolyticus were grown in LM medium at 30°C with shaking . Colony opacity of V . parahaemolyticus was monitored on LM with 2% agar . Unless specified , antibiotic concentrations are as follows: ampicillin , gentamicin , and kanamycin , 100 mg/L; chloramphenicol and tetracycline , 10 mg/L; streptomycin , 5 g/L; polymyxin B , 50 U/L . The 90 , 000 molecule library was supplied by the High-Throughput Screening Resource Center of the Rockefeller University . The V . cholerae strains BH1578 ( ΔcqsA ΔluxS pBB1 ) and BH1651 ( luxOD47E pBB1 ) were grown overnight in LB medium with tetracycline and diluted 25-fold . The diluted cultures were dispensed into 384-well microtiter plates containing screening molecules that were previously added to each well . The final concentration of each compound was ∼20 µM . Light production was measured on an Envison Multilabel Reader after 6-hour incubation at 30°C without shaking . Compounds that induced light production >100-fold were reordered from suppliers and tested . Overnight cultures of reporter strains were grown in LM medium ( for V . harveyi ) or LB with tetracycline ( for V . cholerae carrying pBB1 ) and diluted 20-fold with sterile medium . Bioluminescence and OD600 were measured in an Envison Multilabel Reader following 4-hour incubation at 30°C with shaking . Synthetic molecules were dissolved in DMSO and supplied at varying concentrations to the reporter strains . DMSO was used as the negative control . The open reading frame encoding V . cholerae LuxO D47E was amplified by PCR and cloned into plasmid pET28B that had been previously digested with NdeI and BamHI . The resulting plasmid was transformed into E . coli BL21 Gold ( DE3 ) resulting in strain WN133 . Strain WN133 was grown in LB with kanamycin at 30°C with shaking until the OD600 of the culture reached ∼1 . 0 . IPTG was added at a final concentration of 200 µM , and the culture was incubated for an additional 4 hours at 30°C with shaking . Cells were harvested by centrifugation , suspended in lysis buffer ( 20 mM Sodium phosphate buffer pH 7 . 4 , 0 . 5 M NaCl , 10% glycerol , and 5 mM imidazole ) , and lysed using a Cell Cracker . Soluble materials were loaded onto a HiTrap chelating column charged with nickel , the column was washed extensively with lysis buffer , and His6-tagged V . cholerae LuxO D47E enzyme was eluted using a linear gradient of increasing concentration of imidazole dissolved in lysis buffer . Fractions containing LuxO D47E were pooled and concentrated with an Amicon Untra-15 filter . Protein was snap-frozen in liquid nitrogen and stored at −80°C . Protein concentrations were determined by UV absorbance at 280 nm . E . coli NtrC and other LuxO D47E variants were purified using the same method . A modified coupled-enzyme assay was used to measure the rate of ATP hydrolysis by LuxO D47E [42] . Briefly , ADP released from ATP by LuxO D47E is reacted with phosphoenolpyruvate ( PEP ) to form pyruvate using pyruvate kinase ( PK ) . Pyruvate is reacted with NADH to form NAD and lactate using lactate dehydrogenase ( LDH ) . The rate of NAD production is followed at 340 nm using a spectrophotometer . ATP hydrolysis rates were inferred from the absorbance change observed ( εNADH , 340−εNAD , 340 = 6220 M−1 cm−1 for NADH ) [42] . The rates of ATP hydrolysis by LuxO D47E were measured in reactions containing 100 mM Sodium phosphate buffer pH 7 . 4 , 5 mM MgCl2 , 0 . 2 mM NADH , 1 mM PEP , 5–20 units of PK/LDH mix ( Sigma ) , and 10 µM LuxO D47E . ATP and inhibitors were added to the reactions at indicated concentrations . The rate of ATP hydrolysis was monitored for 5 minutes . Data were fitted using Graphpad Prism to obtain the kinetic parameters . Percent ATPase inhibition was calculated using the following formula: Electrophoretic mobility shift assays to study LuxO and Qrr promoter DNA interactions were performed as described in [69] . Fluorescence anisotropy assays using LuxO D47E were modified from [70] . The luxOD47E allele was removed from plasmids harbored in WN133 with the enzymes XbaI and BamHI and ligated into pEVS143 [71] that had been previously digested with AvrII and BamHI . The luxOD47E reading frame of the resulting plasmid ( WN2029 ) was randomly mutated using the GeneMorph II Random Mutagenesis Kit . The resulting mutagenized luxOD47E plasmid library was introduced into a V . cholerae ΔluxO strain by conjugation . Individual colonies from this V . cholerae luxOD47E mutant pool were arrayed into 96-well plates containing LB medium with 100 µM compound 12 . The V . cholerae ΔluxO strain harboring non-mutated luxOD47E was grown in the absence of compound 12 to provide the reference for background light production . Following overnight static incubation at 30°C , clones that produced light comparable to the background were selected and re-tested in the presence and absence of compounds 11 and 12 . DNA sequencing was used to determine the alterations in luxOD47E for inhibitor-resistant mutants . Site-directed mutageneses were performed with the QuikChange II XL Site-Directed Mutagenesis Kit to uncouple multiple mutations . Overnight cultures of the V . cholerae luxOD47E strain were diluted 1000-fold in AKI medium containing the indicated concentrations of compound 12 . The cultures were statically incubated at 37°C for 4 hours and subsequently shaken for 4 more hours at 37°C . Cells were collected by centrifugation , TcpA from different samples was analyzed by Western blot as previously described [17] . Overnight cultures of the V . parahaemolyticus luxO* strain ( LM4476 ) were washed and diluted 50-fold in LM medium with 10 mM MgCl2 and 10 mM sodium oxalate in the presence of the indicated concentrations of compound 12 . The cultures were grown for 4 hours with shaking at 37°C . Viable cell count showed that all cultures contained ∼1×109 CFU/mL after incubation . Cells were collected by centrifugation , and the secreted and cytoplasmic VopD from different samples were analyzed by Western blot as previously described [47] . Cytotoxicity assays were modified as previously described [48] . HeLa cells ( 2×104 cells/well ) were cultured for 48 hours at 37°C and 5% CO2 in a 96-well plate containing DMEM with 10% fetal bovine serum prior to infection . V . parahaemolyticus strains were grown as described above for VopD analysis and used in the infection assays . Immediately prior to V . parahaemolyticus infection , DMSO or compound 12 ( 500 µM ) was added to the HeLa . Serially diluted bacteria were added to HeLa cells at multiplicity of infection of 10 . Lactate dehydrogenase release from HeLa cells was assayed between 1–4 hours after infection using the CytoTox 96 nonradioactive cytotoxicity kit ( Promega ) . All chemical syntheses and analytical methods are provided in the Supporting Text S1 .
|
The disease cholera , caused by the pathogenic bacterium Vibrio cholerae , is a major health concern in developing regions . In order to be virulent , V . cholerae must precisely control the timing of production of virulence factors . To do this , V . cholerae uses a cell-cell communication process called quorum sensing to regulate pathogenicity . In the current work , we identify and characterize new classes of small molecules that interfere with quorum-sensing-control of virulence in multiple Vibrio species . The molecules target the key quorum-sensing regulator LuxO . These molecules have the potential to be developed into new anti-infectives to combat infectious diseases of global importance .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"medicinal",
"chemistry",
"chemistry",
"biology",
"microbiology"
] |
2012
|
Broad Spectrum Pro-Quorum-Sensing Molecules as Inhibitors of Virulence in Vibrios
|
Caulobacter crescentus undergoes an asymmetric cell division controlled by a genetic circuit that cycles in space and time . We provide a universal strategy for defining the coding potential of bacterial genomes by applying ribosome profiling , RNA-seq , global 5′-RACE , and liquid chromatography coupled with tandem mass spectrometry ( LC-MS ) data to the 4-megabase C . crescentus genome . We mapped transcript units at single base-pair resolution using RNA-seq together with global 5′-RACE . Additionally , using ribosome profiling and LC-MS , we mapped translation start sites and coding regions with near complete coverage . We found most start codons lacked corresponding Shine-Dalgarno sites although ribosomes were observed to pause at internal Shine-Dalgarno sites within the coding DNA sequence ( CDS ) . These data suggest a more prevalent use of the Shine-Dalgarno sequence for ribosome pausing rather than translation initiation in C . crescentus . Overall 19% of the transcribed and translated genomic elements were newly identified or significantly improved by this approach , providing a valuable genomic resource to elucidate the complete C . crescentus genetic circuitry that controls asymmetric cell division .
The C . crescentus genome encodes instructions to perform asymmetric cell division using a genetic circuit that integrates transcriptional control from differential chromosome methylation , activation of transcription factors by phosphosignaling pathways , specific proteolysis events , and the subcellular localization of regulatory proteins [1] . Multiple cell cycle events are coordinated with the replication and segregation of the chromosome once and only once per cell cycle [2] . While the C . crescentus genome was sequenced 13 years ago [3] , our understanding of the transcribed and translated elements in the genome is far from complete . Tiling arrays have previously been used to map 27 ncRNAs and 769 transcription start sites ( TSSs ) in the C . crescentus genome [4] , [5] . Now , using RNA sequencing one can identify transcript architectures at single base-pair resolution and with genome-wide coverage [6] . Recently , global identification of 5′ PPP sites of transcription initiation in the genome using a modified global RACE approach enabled mapping of 2726 TSSs in the C . crescentus genome ( Zhou et al . [unpublished data] ) . Liquid chromatography-mass spectrometry ( LC-MS ) based proteomics methods have identified peptides in 66% of annotated coding DNA sequences ( CDSs ) [7] , but poor peptide coverage severely limits mapping of entire CDSs . However , with ribosome profiling , which maps translating ribosomes [8] , [9] , we have successfully mapped the C . crescentus CDSs genome-wide . We report application of a multi-omic approach utilizing RNA-seq , global 5′-RACE , LC-MS proteomics , and ribosome profiling data sets to identify the RNA transcripts and CDSs in the C . crescentus genome at high resolution . We identified transcription units at single nucleotide resolution , 5′ and 3′ UTRs , and the position of all translated CDSs at near single codon resolution in the C . crescentus genome . Integration of these datasets allowed the identification of 375 leaderless mRNAs , 94 new small open reading frames , and 106 new noncoding RNAs . Additionally , we mapped 3235 CDSs in the C . crescentus genome transcribed from 2201 mRNA transcripts . Our integrated analysis also identifies a plethora of genetic regulatory elements , significantly increasing the knowledge of regulatory complexity encoded by the C . crescentus genome . With the identification of the genomic transcription and translation elements , a systems map of the genetic network that controls asymmetric cell division is within reach . Analysis of the translation initiation sites shows that a majority ( 75 . 4% ) initiate without a Shine-Dalgarno sequence . A majority of Shine-Dalgarno sites are encoded within the CDSs and , as with E . coli and B . subtilis , these Shine-Dalgarno sites correlate with pauses in translation elongation [10]–[12] . This suggests that C . crescentus uses the Shine-Dalgarno site more commonly for ribosome pausing rather than translation initiation . As suggested from a multitude of predicted bacterial genome annotations [13]–[16] , our genomic map provides further experimental evidence that the Shine-Dalgarno-based translation initiation model is not applicable to all bacteria .
We integrated multiple C . crescentus genomic datasets to map global gene expression features at base-pair resolution ( Figure 1 ) . We used a genomic RACE dataset that mapped 2726 TSSs in minimal defined medium allowing promoter and 5′ end RNA identification ( Zhou et al . [unpublished data] NCBI GEO accession number GSE57366 ) . Additionally , we used RNA-seq data derived from base-hydrolyzed RNA fragments from complex and minimal defined medium to find both the stable 5′ end of the transcript and the length of the transcript onto which we mapped the individual CDSs . To identify translated CDSs , we used both genome coverage of trypsin-digested peptides ( identified for 2559 annotated CDSs ) in minimal medium during log growth and starvation [7] and ribosome profiling data . Ribosome profiling data were collected from mid-log phase C . crescentus NA1000 cultures grown in complex ( peptone-yeast extract; PYE ) and minimal defined ( M2 glucose; M2G ) medium . Translation was arrested with 100 µg/mL chloramphenicol , polysomes were digested with micrococcal nuclease , and ribosome-protected mRNA fragments were purified on a sucrose gradient and prepared for high throughput sequencing ( Figure S1 ) [8] , [17] . Although the extent of peptide coverage within the CDS was not consistent due to the non-uniform distribution of trypsin cut sites , the ribosome profiling data allowed us to map the expressed CDSs in the genome with high coverage and resolution . The 5′ and 3′ UTRs of the transcript can thus be identified . With this approach we have now identified the global transcript and CDS architecture of the C . crescentus genome under the specified growth conditions . Our updated version of the C . crescentus genome annotation can be downloaded here ( Dataset S1 ) , and has been incorporated in the NCBI NA1000 annotation ( accession CP001340 ) . We identified differentially expressed genes by comparing RNA-seq levels between M2G and PYE medium , and these results agree well with previous microarray measurements ( Dataset S1 ) [18] . In addition , we find that the ribosome profiling levels correlate with the relative amount of protein present in the cell , validating that the ribosome profiling assay is measuring protein production ( Fig . S15 ) . The ribosome profiling data also revealed additional changes in translation between growth conditions ( Dataset S1 ) . We found 39 genes that are differentially translated with a >2-fold change in translation efficiency ( as defined in the Methods ) between M2G and PYE medium ( Dataset S1 ) . The largest class of differentially translated genes includes eight genes involved in amino acid catabolism . These genes are repressed in M2G , likely due to the absence of amino acids in the medium . We mapped the CDSs in the C . crescentus genome using both LC-MS peptide coverage [7] and ribosome profiling . We initially used the LC-MS peptide coverage and the specificity of trypsin protease to map start codons . Since trypsin cuts proteins after Arg or Lys residues , we identified translation start sites as N-terminal sites not preceded by Arg or Lys codons . To avoid false signals from peptides generated from protein degradation we searched for peptides >20 amino acids , thereby omitting products from the major protease ClpP [19] . A majority of the remaining peptides mapped to ATG , GTG , or TTG start codons or the next codon that would result from cleavage of fMet . In this manner , we identified 621 start codons out of the 3818 annotated CDSs in the NA1000 ( CP001340 ) genome . The remainder could not be identified due to the poor intra-CDS coverage of peptides . Since the ribosome profiling read density matched the 621 verified start codons remarkably well ( Figure S2 ) , we used the ribosome profiling data to map all start codons . Importantly , ribosome profiling relies on sequencing the protected mRNA fragment from actively translating ribosomes; thus , the ribosome profiling results can be used to globally map start codons at near complete coverage . Using the density of ribosomes along CDSs , we searched for start codons in the predicted annotation ( CP001340 ) by looking for a continual density of ribosomes from the stop codon to the furthest upstream in-frame start codon . If a peptide was found in the LC-MS data , we refined the search for the start codon from the most N-terminal codon of the peptide to the furthest upstream in-frame start codon covered by ribosome footprints . Additionally , we found many LC-MS peptides and ribosomes positioned outside of annotated CDSs either within intergenic regions or on the opposite strand of hypothetical CDSs . We manually curated these regions to identify the boundaries of the corresponding CDS . Using this multipronged approach we mapped 3235 CDSs in the C . crescentus genome . The average increase in the density of ribosomes at the start codon ( Figure S2 ) aided the detection of start codons and , despite heterogeneity in mRNA footprint sizes , allowed us to identify start codons at near single codon resolution . While 74 . 3% of the start codons identified were ATG , many CDSs initiate with GTG ( 14 . 5% ) , TTG ( 10 . 3% ) , and a few with CTG ( 0 . 7% ) ( Dataset S1 ) . We also observed a small number of CDSs that begin with other potential near-cognate start codons ( 0 . 25% ) , including one double mismatch GTC codon verified by LC-MS ( Dataset S1 ) . In total , we corrected the start codons of 12 . 8% of annotated CDSs ( or 15 . 7% of those that were mapped ) , including many that were previously reported to be misannotated or involved in cell cycle regulation including gcrA , chpT , sciP , sidA , divJ , parB , and ftsA ( Dataset S1 ) [20]–[22] . We verified that the ftsA start codon is 18 codons upstream using western blots ( Figure 2A , Figure S11 ) and found that overexpression from a high-copy plasmid containing the correct start codon yielded a strong cell division phenotype while that of the previously annotated form lacking the N-terminal 18 amino acids ( [23]–[25] ) causes a less severe phenotype ( Figure 2A ) even after 24 hours of overexpression ( Figure S16 ) , suggesting these 18 N-terminal amino acids are likely functional . In general , predicted start codons are further upstream than our experimentally determined start codons due to the biases of start codon prediction algorithms to pick longer CDSs . However , we identified 69 CDSs with start codons further upstream than the original annotation . We also identified 94 previously unidentified CDSs , most of which encode small proteins of less than 100 amino acids . Some of these small CDSs appear to be leader peptides , such as the small CDS positioned in front of the trpS gene ( Figure 2B ) [26] . It is likely that some of these small leader CDSs have a regulatory role in the expression of their downstream genes [26] . Additionally , we found that 62% of small CDSs are not encoded in the same direction as the downstream genes , indicating that they are not leader peptides and instead likely encode functional proteins ( Figure 2C ) . As tracking the ribosome profiling footprint density allowed us to globally map CDSs in C . crescentus , we analyzed the E . coli and B . subtilis ribosome profiling datasets [10] alone and identified 53 and 70 putative changes to the CDSs in each respective genome ( Dataset S6 ) . Finally , we observed cases where a single mRNA has multiple start codons that initiate different isoforms of the protein ( Figure 2D ) [9] . We identified 75 alternative start codons in the C . crescentus genome by searching for internal peptides with N-terminal residues mapping to non-typsin digested ATG sites ( Dataset S1 ) . Despite a conserved 3′ end of the rRNA anti-Shine-Dalgarno ( aSD ) sequence ( CCUCC ) in all bacteria , only 24 . 6% ( 957 ) of C . crescentus CDSs contain a Shine-Dalgarno ( SD ) sequence in the translation initiation site as determined by the predicted ΔG° of annealing between the aSD with the mRNA ( Figure 3A , Figure S3A ) [16] . While the C . crescentus genome is GC-rich ( 67 . 17% ) , the random chance of finding a SD sequence in a translation initiation region is 19 . 2% , suggesting only slight enrichment of SD sites . The C . crescentus translation initiation site motif contains little or no consensus information other than the start codon ( Figure 3B ) . Globally , the predicted RNA stability at the translation initiation site revealed it to be less stable than other regions of the mRNA ( Figure S3B ) , consistent with the model that an unstructured region at the translation initiation site is required to translate mRNAs without a SD sequence at the initiation site [27] . On average we observe a peak of ribosome density at the start codon and a peak , albeit smaller , at the stop codon , suggesting that initiation and termination may be slow steps in C . crescentus translation ( Figure S4A ) . However , as the ribosomes were arrested with chloramphenicol , which blocks elongation but not initiation of translation , the enrichment observed at the start codon may not accurately reflect the natural abundance of initiating ribosomes . The ribosome occupancy along genes has peaks along the coding sequence caused by pausing of elongating ribosomes ( Figure 3C ) . As observed in E . coli and B . subtilis [10] , many of the internal pauses in translation elongation appear not to be driven by codon usage ( Figure S4B ) , but instead correlate with internal SD sites in the mRNA coding sequence that base-pair with the 3′ end of the rRNA , stalling ribosome movement ( Figure 3D ) [10] . The aSD binding strength for the SD sequences correlates with the ribosome occupancy , suggesting that the annealing of the rRNA to the mRNA slows translocation of elongating ribosomes ( Figure 3E , Figure S14 ) . These results support the hypothesis that internal SD sites provide a conserved pausing mechanism for bacterial ribosomes even in a genome that has high GC content where SD sequences are more abundant . In accordance with a more prevalent role of the SD in elongation , we see poor correlation with the translation efficiency of mRNAs and the aSD binding strength of their SD sequence at the start codon ( Figure S5 ) . To identify the RNA transcript units we used a global RACE dataset that maps 5′ PPP-sites of transcription initiation ( Zhou et al . [unpublished data] ) together with RNA-seq density measured here . We found good overlap of the TSSs between the datasets . When the RNA-seq density is centered at the TSSs identified by 5′ global RACE , we observed an increase in RNA-seq read density at the same 5′ nucleotide ( Figure 4A ) . By comparing the RNA-seq data to the TSSs we were able to map the length of the major form of the transcriptional unit and in some cases where an internal TSS exists , allowing us to identify potential isoforms of transcripts . The transcripts mapped by our RNA-seq approach agree well with published northern blots ( Dataset S7 ) . In addition , by comparing the transcript unit with the mapped CDSs , we were also able to determine which RNAs encode proteins under the growth conditions tested . In the RNA-seq data sets we found 74% of reads map within CDSs , 21% of reads map to intergenic regions , and 5% map antisense to CDSs . In total , 96 . 2% of the genome was transcribed among the conditions tested . Together , these data now provide a comprehensive map and functional classification ( coding or noncoding ) of the expressed RNAs in the C . crescentus genome with single base-pair resolution . The global distribution of mRNA leader lengths in C . crescentus ( Figure 4B ) shows that 57% of 5′ UTRs are between 15 and 60 nt with some spanning >100 nt . Surprisingly , we observed 375 leaderless mRNAs ( 9 . 6% of the cell's CDSs ) ( Figure 4C ) . The 5′ nucleotide is the first base of the start codon in a leaderless mRNA that is able to initiate translation on bacterial , archaeal , and eukaryotic ribosomes , suggesting it is an ancient mechanism of translation initiation [28] . Leaderless mRNAs have been found to be rare in most bacteria [29]–[31] and previously only two leaderless mRNAs were identified in C . crescentus: dnaX and hemE [32] . The presence of many leaderless mRNAs in C . crescentus and 171 in S . meliloti [33] suggests translation of leaderless mRNAs may occur more commonly in the alpha-proteobacteria than previously anticipated . In contrast to leaderless mRNAs , we identified 286 mRNAs that have long 5′ UTRs >100 nt ( Figure 4D ) , which may play regulatory roles in translation . For example , dnaA mRNA encodes a 155 nt 5′ UTR that contributes to the repression of translation , suggesting the 5′ UTR helps regulate the level of the protein needed for proper cell cycle regulation [34] . Additionally , four 5′ UTRs appear to encode conserved riboswitches that are capable of regulating the expression of downstream genes upon direct metabolite binding to the RNA [35] ( Table S1 ) . We also observed genes for which the mRNA is transcribed from an internal site driving translation of an alternative translation initiation site . For example , CCNA_00832 has an internal TSS which is translated from a start codon in the +2 reading frame , resulting in a distinct protein compared to the lowly expressed full length mRNA isoform ( Figure 4E ) . In addition , we find that the cell division gene ftsW has two mRNA isoforms which result in translation of two different length proteins in the same reading frame , with the smaller form ( ftsWs ) being more highly translated ( Figure 4F ) . No PPP site was identified for the small form of the mRNA; however , it does contain a good sigma 70 site [36] 35 nt upstream of the internal 5′ end . Additionally , when ftsW was inserted into a low copy plasmid lacking the promoter for the full length ftsW CDS , we observed accumulation of FtsWs protein ( Figure S11 ) suggesting it is transcribed from an internal transcript . Overexpression of ftsWs causes a marked motility and cell division phenotype in a low-agar swarming plate assay as well as an increase in cell length when grown in liquid culture ( Figure 4F , S6 ) . Despite the small size of this 35 amino acid isoform , FtsWs can localize to sites of constriction ( Figure 4F ) suggesting it may play a role in cell division . Altogether , we observe that alternative transcripts can drive alternative translation products increasing the diversity of proteins encoded in the genome . We observed 133 non-coding RNAs ( ncRNAs ) , adding 106 new ncRNAs to the 27 previously identified using tiling arrays [4] ( not including conserved ncRNAs such as tRNAs , rRNAs , RNaseP , 6S RNA , 4 . 5S RNA , and tmRNA ) . Most of the ncRNAs are expressed from intergenic regions ( Figure 5A ) and ribosome profiling data showed that these regions are not translated . Some ncRNAs are transcribed from TSSs in the 3′ end of a CDS ( Figure 5B ) , which , similar to Salmonella , allows the 3′ UTR regions to act as a reservoir for ncRNAs [37] . RNA-seq data showed widespread antisense RNA transcribed throughout the C . crescentus genome accounting for 5% of non-tRNA/rRNA reads . Global RACE 5′ PPP mapping revealed that antisense TSSs are found within 15% of CDSs ( Zhou et al . [unpublished data] ) . We observed that the 3′ UTR of an mRNA can extend into the coding regions of downstream genes oriented in the opposite direction forming a long antisense RNA with respect to the mRNAs of these downstream genes ( Figure 5C ) . We found overlaps extending over up to three genes . For example , the 3′ UTR of CCNA_03120 , a gene predicted to encode a protein involved in chemotaxis , extends into the coding regions of an operon containing genes CCNA_03121 , CCNA_03122 ( putative integral membrane protein ) , and CCNA_03123 ( metal regulated homo-dimeric repressor ) . The operon has been traditionally defined as a single co-transcribed unit that yields a single polycistronic mRNA . Using our CDS and RNA maps , we were able to identify operons as mRNAs with >1 CDS ( Figure 6A ) . We observe 863 operons in the C . crescentus genome encoding 65% of all CDSs in the genome . We found that 55% of operons contain 2 CDSs ( Figure S7 ) ; however , a few operons are quite large with up to 29 CDSs in a single operon . Examples of these include the type IV pilus operon ( 12 CDSs ) [38] , one of the ribosomal protein operons ( 24 CDSs ) , and the NADH dehydrogenase operon ( 29 CDSs ) , the largest C . crescentus operon . The distribution of operon sizes for C . crescentus is similar to that for M . pneumonia [39] , H . pylori [31] , and E . coli [40] . In many operons , such as those of ribosomal proteins ( Figure 6A ) , the expression level of each CDS is similar yielding the proper stoichiometry of ribosomal proteins of one per ribosome . However , we find that many C . crescentus operons do not have equal expression of the encoded CDSs at the RNA and translation levels ( Figure S8 ) . Different levels of expression of RNA for contiguous CDSs within a single operon can be caused by a multitude of factors . The most well characterized mechanism is transcriptional polarity , driven by translation rate , transcription elongation factors , and/or termination factors to cause the 3′ end genes to have reduced levels of expression ( Figure 6B ) . Additionally , operons can be regulated by transcriptional attenuators that down-regulate transcription of the trailing genes ( Figure 6C ) . As shown originally in the E . coli trp operon [26] , the ilvBN operon leader has tandem Ile and Val codons which , upon conditions of low tRNAIle and tRNAVal aminoacylation , cause ribosome pausing at these codons , blocking a rho-independent terminator hairpin from forming and allowing RNA-polymerase to elongate through the ilvBN operon [41] . Uneven expression can also cause the 3′ end CDSs to be expressed higher than 5′ end CDSs . We observed that 349 operons contained an alternative TSS ( Figure 6D ) that could potentially drive higher expression of downstream CDSs . Expression from these internal TSSs was observed to be dynamically regulated during the cell cycle ( Zhou et al . [unpublished data] ) . We also observed operons that appear to have 3′ end genes whose mRNAs are more stable ( Figure 6E ) . In these cases , the operon has only a single TSS and contains a downstream 5′ P site , indicative of an RNaseE cut site . Since the last CDS has a higher mRNA level , it is likely that the 5′ end of the transcript is less stable . Altogether , 64% of the operons appear to have a >2 fold change in RNA level among different CDSs suggesting that most operons are regulated co- and post-transcriptionally to ensure appropriate RNA levels of each encoded CDS . Operons appear to be highly regulated by having both multiple TSSs and different transcription termination sites . We therefore calculated the total number of TSSs per operon and found that C . crescentus operons have an average of 1 . 3 TSSs per operon driving multiple mRNA isoforms . Additionally , the number of operons that have successive ≥5-fold drops in RNA level between encoded CDSs is 125 , suggesting that polarity of operons also drives many isoforms . In total , we estimate that C . crescentus operons have an average of 1 . 5 isoforms per operon generated either from alternative TSSs or polarity and 0 . 5 cis-encoded regulatory features including antisense RNAs , riboswitches , and transcription attenuators . The high number of isoforms and regulatory features suggests that operons can be highly regulated at the transcription and RNA levels . Together with the 75 CDSs that can be initiated internally to drive different protein isoforms , this suggests that the C . crescentus genome contains significant regulatory complexity .
We used multiple datasets from ribosome profiling , RNA-seq , 5′ global RACE , and LC-MS [7] to identify and quantify the transcribed and translated elements of the C . crescentus genome with high resolution and near complete coverage ( Table 1 ) . Ribosome profiling provides a way to map CDSs that greatly surpasses LC-MS in coverage . We found misannotation of the start codons of many important genes involved in the C . crescentus cell cycle ( Dataset S1 ) , including the essential cell division gene ftsA ( Fig . 2A , Fig . S11 ) , and found that the truncation of the N-terminal 18 amino acids absent in the previously annotated start codon affects the function of FtsA . We found two cases where previously predicted ncRNAs [4] are , in fact , translated ( Dataset S1 ) . Additionally , ribosome profiling identified 94 previously unknown CDSs , a majority of which are <50 amino acids . In total we observe 94 small CDSs of <50 amino acids in the genome . The role of these small proteins in C . crescentus is largely unexplored; however , small proteins have been reported to have important functions in B . subtilis , E . coli , and eukaryotes [42] , [43] . A recent identification of a small protein in C . crescentus that can delay cell division upon DNA damage suggests this class of proteins indeed can perform important cellular functions in C . crescentus [20] . We discovered 106 new ncRNAs in the C . crescentus genome that are expressed during normal growth . However , most of the identified C . crescentus ncRNAs are not conserved in other genomes outside of the Caulobacteraceae with only a few present in other alpha-proteobacteria . The function of only one of these ncRNAs in C . crescentus has been characterized , crfA , which was shown to be involved in the response to carbon starvation [4] , [44] . In other bacteria , small ncRNAs have a variety of functions , but most commonly they are involved in annealing to mRNAs with complementary sequences and regulating translation or mRNA stability [45] . Most ncRNAs identified in bacteria function through the RNA chaperone Hfq [45] , [46] . Hfq is thought to both stabilize the ncRNA and facilitate annealing between the ncRNA and the target mRNAs . In C . crescentus the ncRNA substrates of Hfq have not been identified; however , Hfq was found to be non-disruptable in a high-throughput transposon mutagenesis screen [22] suggesting an important role for ncRNA regulation . Additionally , 14 of the C . crescentus ncRNAs are cell cycle regulated ( Zhou et al . [unpublished data] ) , suggesting these ncRNAs may play a role in cell cycle progression . With our CDS mapping approach we identified upstream leader peptides and alternative start codons ( Figure 2A , B , D ) . While translation of upstream leader peptides can often regulate expression of the downstream gene , it is possible that these CDSs may also produce functional proteins . Alternative start codon selection in eukaryotes has been shown in some cases to control subcellular localization and to cause functional switches in proteins by translating forms lacking functional domains [47]–[49] . The cell division gene ftsW is made in a full length and short form ( Figure 4F ) , both of which can localize to the site of constriction at the midcell ( Figure 4F ) [25] . Overexpression of ftsWs , the short form of ftsW , gave rise to a motility and cell division defect in the swarmer plate assay , leading to a modest elongation of the cells ( Figure 4F , S6 ) . As the mRNA for the full length ftsW is activated in the late predivisional cell , it will be important to measure the cell cycle-regulated translation of both the ftsW long and ftsWs short forms to understand their roles in regulating cell division . The vast amount of regulatory RNA elements identified by this approach suggests that there's an unexplored level of cell cycle gene expression control that remains to be investigated . Indeed , as seen in other bacteria , the examination of RNA levels in operons suggests that most operons are not consistent with the classical model of one polycistronic transcriptional unit , suggesting that regulation of operons is more complex [31] , [39] , [50] . In support of this we estimate that on average , each mRNA and operon contains 2 . 0 cis-encoded regulatory features ( alternative TSS , antisense RNA , internal TSS , internal start codon , lower 3′ RNA density ) suggesting combinatorial regulation . Altogether , in the C . crescentus genome we identified ncRNAs , leaderless mRNAs , alternative translation initiation sites , small upstream CDSs , antisense RNAs , alternative transcription initiation sites , transcriptional polarity of operons , and differential RNA stability of operons . These elements are spread throughout the genome and suggest that co/post-transcriptional regulation is likely an important mechanism for cell cycle regulation of gene expression . In support of this , many antisense RNAs and ncRNAs are differentially activated at specific stages of the cell cycle ( Zhou et al . [unpublished data] ) . An important goal will be to understand how the RNA regulatory elements affect cell cycle stage-specific translation and mRNA stability to identify their role in the genetic circuitry that drives the cell cycle . Bacterial translation start site selection is thought to occur by the 30S ribosome subunit binding to the SD site on the mRNA [51] , [52] , spaced approximately 5 nt away from the start codon [53] . While the kinetic events of translation initiation on SD led mRNAs have been well studied [54] , initiation on leaderless mRNAs and non-SD containing mRNAs are less well understood . Recent reports suggest that non-SD led mRNAs have an unstructured region at the start codon [27] , which was also seen in C . crescentus ( Figure S3B ) . Additionally , non-SD led mRNAs may contain motifs that bind to sites on the rRNA outside the aSD region [55] , [56]; however , we do not see abundant motifs that can explain initiation ( Figure 3B ) . We observe that around the start codon the predicted mRNA folding stability is lowest , suggesting that having an unstructured region may be vital for non-SD mRNA binding to ribosomes ( Figure S3B ) [27] . Leaderless mRNAs are initiated by preassembled 70S/80S ribosomes and can be initiated by ribosomes from all three domains of life [28] , [57] , [58] . We find that leaderless mRNAs have no specific motif for translation initiation , but instead have an unstructured region that is shifted from the start codon further towards the 3′ end of the translation initiation site likely ensuring the AUG is accessible to bind initiator tRNA in the mRNA channel of the ribosome ( Figure S3B ) . The C . crescentus genome appears to contain the second highest relative number of leaderless mRNAs of any bacterium characterized to date with 375 in a 4 . 0 mb genome , only behind Mycobacterium tuberculosis with 505 in a 4 . 4 mb genome [59] . Interestingly , in C . crescentus leaderless mRNAs are translated with similar efficiency to mRNAs containing a leader ( Figure S10 ) suggesting C . crescentus translation is adapted to use leaderless mRNAs as substrates during normal growth and not with a stress induced mechanism as in M . tuberculosis or E . coli [59] , [60] . Analysis of sequenced bacterial genomes shows an abundance of non-SD led mRNAs across bacteria , suggesting that the SD dominated mechanism , which is abundant in E . coli and B . subtilis ( 66 . 9% and 94 . 3% of CDSs use a SD sequence , respectively ( Figure S9 [16] ) ) , is not abundantly used in other bacterial species [13]–[16] , [27] , [61] . Furthermore , bioinformatics predictions have estimated that the fraction of genes with start codons preceded by SD sites is only 54 . 3% across bacteria [15] . In C . crescentus only 24 . 6% of all start codons are preceded by a SD sequence , providing direct evidence that SD mediated translation initiation is not the major mechanism . Interestingly , C . crescentus ribosomes do not initiate on bacteriophage Ms2 or T4 mRNAs and E . coli ribosomes do not initiate on Caulobacter phage Cb5 mRNA , suggesting the translation machinery of these bacteria have different specificities for translation initiation sites despite a similar aSD sequence of the rRNA [62] , [63] . Overall , this suggests that the low level of SD sites in C . crescentus translation initiation sites ( 24 . 6% ) may be due to an adaptation of C . crescentus translation machinery to initiate on non-SD led mRNAs . In support of this , we observe equivalent translation efficiency of leaderless , non-SD led , and SD led mRNAs ( Figure S10 ) . Thus , C . crescentus provides a useful model system to investigate the molecular mechanisms of translation initiation on both non-SD and leaderless mRNAs . Using our experimentally determined CDS features we found that C . crescentus uses SD sites primarily for ribosome pausing within the CDSs instead of for translation initiation . We did not observe that C . crescentus ribosomes preferentially paused at rare codons ( Figure S4B ) , similar to E . coli and B . subtilis when cultured in conditions with sufficient nutrients , but instead at internal SD sites within the mRNA ( Figure 3D , E ) [10] . Upon starvation of E . coli or B . subtilis cells for serine , pausing is observed at serine codons [10] , [12] suggesting that depleting aminoacyl-tRNA levels can cause significant codon dependent pausing [26] , [64] . In C . crescentus , as in E . coli [10] , SD sites are selected against in the CDS ( Figure S13 ) , presumably due to their strong ability to pause ribosomes . Indeed , the presence of internal SD sites within CDSs has been shown to cause long pauses in a single molecule ribosome translocation assays [65] . Additionally , ribosome pausing at internal SD sites has also been shown to be an important element for ribosome frame shifting [66] , [67] and likely affects other cotranslational processes such as protein folding [68] . The aSD site in the ribosome is conserved across all known bacteria ( Figure S12 ) [14] , even in those lacking abundant SD sites at start codons . As C . crescentus has evolved to have a larger apparent role of the SD for pausing than initiation , perhaps the strong conservation of the aSD site is due in part to its role in programmed ribosome pausing .
C . crescentus strain NA1000 was grown in M2G or PYE overnight in 5 mL , transferred to 25 mL and grown overnight , then diluted into 500 mL and grown to an OD600 of 0 . 5 . Cells were treated with 100 µg/mL of chloramphenicol for 2 minutes then harvested by centrifugation and flash frozen in liquid nitrogen . Cells were subjected to mixer milling ( 6 cycles for 3 min at 15 Hz ) while frozen in liquid nitrogen . A small amount of the lysate was saved for RNA-seq and the rest was used for ribosome profiling . Ribosome profiling was performed as in [10] , [17] . To prepare the RNA-seq libraries , total RNA was extracted from the frozen cell pellet by hot acid-phenol extraction and RNA integrity was verified on the bioanylzer ( Agilent ) . rRNA was removed by MICROBExpress gram negative rRNA removal kit ( Ambion ) . The resulting RNA was base hydrolyzed at 95°C in alkaline hydrolysis buffer ( 50 mM sodium carbonate pH 9 . 2 , 1 mM EDTA ) for 23 minutes and size selected between 20 and 45 nt on a denaturing PAGE gel ( 10% acrylamide 1× TBE/7M Urea ) . Library prep was performed as in [10] , [17] for both RNA-seq fragments and ribosome footprints . DNA libraries were sequenced on the Illumina Hiseq 2000 or Genome Analyzer platforms . Ribosome profiling reads were mapped to the NA1000 genome sequence ( CP001340 ) using bowtie 0 . 12 . 8 [69] and center weighted as in [10] . RNA-seq reads were mapped to the 5′ nucleotide to find the 5′ ends or to the full read sequence for mapping transcripts . Data for two ribosome profiling and two RNA-seq datasets ( one set for both M2G and PYE ) were deposited into the gene expression omnibus ( accession number GSE54883 ) . Ribosome profiling read density ( Datasets S2 & S3 ) and the LC-MS derived tryptic peptides were both mapped to the NA1000 genome sequence ( CP001340 ) . Using the predicted CDS architecture in the annotation file downloaded from genbank ( accession number CP001340 ) we found tryptic peptides in 66% of the CDSs . Tryptic peptides were directly used to map start codons if the N-terminal codon ( or the previous codon in the case of formyl-Met processing ) mapped to regions where the previous codon was not an Arg or Lys codon . Since the coverage of the tryptic peptides at the start codon is poor we used the ribosome profiling read density to map the remainder of start codons . We defined start codons as the most upstream ATG , GTG , CTG , or TTG codon with >1/20 the ribosome profiling read density . If no 1st position mismatches were found we searched for single position mismatches in the 2nd and 3rd positions . If no single position ATG mismatches were found , we used the resulting codon only if they matched the beginning of ribosome density and contained a LC-MS tryptic peptide not preceded by an Arg or Lys codon . Each potential start codon which fit this criterion was manually annotated to ensure accuracy . If we identified two adjacent potential start codons we selected the most upstream start codon . To identify new CDSs we searched for intergenic regions of significant ribosome density . We considered a region a CDS if the ribosome density strictly mapped between start codons to stop codons . We also checked for CDSs that had greater antisense than sense ribosome footprints and manually corrected genes predicted on the wrong strand . We deleted hypothetical CDSs that significantly overlapped other CDSs encoded on the opposite strand or that significantly overlapped tRNA genes . To map CDSs on leaderless mRNAs we found that the center-weighted ribosome footprints often began 12–18 nt after the start codon as the ribosome footprints were shorter . We therefore identified either 5′ PPP ends or the 5′ end of the RNA-seq read density for each potential start codon . If tryptic peptides matched the 5′ end we annotated it as a leaderless mRNA . Alternatively if no tryptic peptide was found , we mapped leaderless mRNAs if the 5′ end matched the first nucleotide of the start codon and the center-weighted ribosome footprints mapped to the 5′ end of the mRNA . We verified this signature on leaderless mRNAs dnaX and hemE [32] . The 5′ end was mapped based on the increased peak intensity of the RNA-seq data at the 5′ nucleotide resulting from partial shearing of the RNA [70] ( from Dataset S4 & S5 ) and/or presence of a 5′ PPP site . 3′ ends were mapped based on an increased 3′ end peak intensity before a drop in RNA level if present , or estimated based on the drop in RNA reads . Non-coding RNAs were identified by examining intergenic or antisense stretches of RNA-seq density . We considered an RNA non-coding if no CDSs were detected within the transcript boundaries . 5′ UTR length distribution was calculated using mapped 5′ RNA ends identified within 300 nt upstream of the start codons or within the last 30% of the upstream CDS , whichever is the shorter distance . 5′UTRs longer than 300 nt were curated manually . To identify known riboswitch elements we searched the 5′ UTR sequences in the Rfam database . Using the predicted NA1000 operon predictions [71] we appended new CDSs to operons using the following criteria: 1 ) CDSs that overlap or were less than 40 nt away with the upstream operon or CDS were annotated as either part of the previous operon or as a new operon if overlapping with an upstream single CDS . 2 ) CDSs less than 260 nt from an upstream CDS were manually inspected and annotated . To use the new CDS map to refine operon predictions we split predicted operons at sites between individual CDSs if they met the following criteria: 1 ) Intergenic region between CDSs must be >40 nt , 2 ) Reads per nucleotide must be >20 , and 3 ) a >10 fold difference in RNA-seq read density between the CDS and intergenic region must be observed . SD sites were calculated using the Free2Bind package [16] . To identify SD affinity for a translation initiation site , we calculated the annealing affinity of 5′-CACCUCCU-3′ sequence of the rRNA with a 1 nt sliding window from −100 to +100 nt of the translation initiation site . Presence of a SD motif was determined if the lowest predicted ΔG° of annealing between the rRNA and mRNA was less than −4 . 4 kcal/mol [13] in a window between −20 and −5 nt upstream of the translation initiation site [53] . To estimate the background SD affinity encoded by a random sequence of nucleotides at the GC% of the C . crescentus genome we calculated the SD affinity on 10 , 000 randomized sequences . 19 . 2% of random sequences contained our criteria for a SD motif . Global ribosome pausing analysis was performed as in [10] on genes with average read coverage >10 reads per codon in M2G medium . The average normalized cross-correlation function of sequence elements relative to pause sites was calculated on genes greater than 160 nt long and >10 reads per codon . Predicted ΔG° of RNA structures were calculated using the RNAFold program in the Vienna RNA package [72] as in [27] . The minimum free energy was calculated in a 50 nt sliding window moving in 1 nt increments from 100 nt before to 100 nt after the start codon . Levels of gene expression were calculated using the reads per kilobase per million mapped reads ( R . P . K . M . ) [73] between samples . Ribosome profiling data were corrected for initiating and terminating ribosomes by removing the first 10 codons and the last 5 codons from the R . P . K . M . calculation . After removing genes with less than 30 reads in a given sample , genes were classified as differentially translated between M2G and PYE if they had a greater than 2-fold change in the translation efficiency = . Images were collected as described in [74] on M2G 1 . 5% agarose pads using a Leica DM6000B microscope . For image analysis MicrobeTracker software [75] was used to determine cell outlines and measure the cell length . Cells were grown to mid-log phase , normalized to OD600 0 . 3 , and spotted on PYE/0 . 3%-Bacto-Agar/0 . 3%-xylose/kanamycin plates . Cells were grown for 2–4 days in a humid 28°C chamber , and imaged on a gel imager . Colony size was calculated using imageJ . Whole cell lysates were generated by growing 1 mL cultures to mid-log , resuspending the cells in 1× Laemmli sample buffer , and boiling at 95°C . Lysates were run on TRIS-Gly SDS-PAGE gels ( Bio-Rad 4–15% or 10% acrylamide ) and transferred to PVDF membranes ( Millipore ) . Immunoblotting was performed using anti-GFP ( Roche ) or anti-FtsA sera followed by detection using chemiluminescent substrate ( PerkinElmer ) . Band intensity was calculated using ImageJ . A list of all strains can be found in Table 2 . To generate all replicating plasmid containing strains , C . crescentus NA1000 was transformed with the following plasmids and selected using standard procedures on PYE plates supplemented with antibiotics . All plasmids were sequence verified . To generate pBXSPA small CDS overexpression plasmids , an SPA tag was inserted into pBXMCS-2 between the EcoRI and SacI sites . Then , ftsWs , sidA , and CCNA_03934 were inserted between NdeI and EcoRI . To generate pftsWs , ftsW was inserted into pRVYFPC-6 between HindIII and KpnI , removing any promoter for ftsW and blocking full length expression with two 5′ transcription terminators . To generate YFP integrating strains , 500 bp of DNA upstream of the stop codon was cloned in frame with YFP in pYFPC-4 using Gibson assembly . Resulting integrating plasmids were transformed into NA1000 and selected on PYE gentamycin . To generate pBX-ftsA , ftsA was inserted into pBXMCS-2 [76] between the NdeI and PstI sites . To generate ftsA::ftsAΔC PxylX-ftsA , ftsA1–375 was inserted into pXMCS-2 [76] between the NdeI and PstI sites . The resulting plasmid ( pXMCS-2 ftsAΔC ) was subsequently transformed into NA1000 resulting in a single integration event at the ftsA locus that simultaneously truncated the native ftsA gene while introducing Pxyl-ftsA . Transformants were selected on PYE kanamycin and xylose . pBX- ftsAΔN1-18 was a gift from Erin Goley .
|
Caulobacter crescentus is a model system for studying asymmetric cell division , a fundamental process that , through differential gene expression in the two daughter cells , enables the generation of cells with different fates . To explore how the genome directs and maintains asymmetry upon cell division , we performed a coordinated analysis of multiple genomic and proteomic datasets to identify the RNA and protein coding features in the C . crescentus genome . Our integrated analysis identifies many new genetic regulatory elements , adding significant regulatory complexity to the C . crescentus genome . Surprisingly , 75 . 4% of protein coding genes lack a canonical translation initiation sequence motif ( the Shine-Dalgarno site ) which hybridizes to the 3′ end of the ribosomal RNA allowing translation initiation . We find Shine-Dalgarno sites primarily inside of genes where they cause translating ribosomes to pause , possibly allowing nascent proteins to correctly fold . With our detailed map of genomic transcription and translation elements , a systems view of the genetic network that controls asymmetric cell division is within reach .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome",
"expression",
"analysis",
"caulobacter",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"microbiology",
"genomic",
"databases",
"prokaryotic",
"models",
"model",
"organisms",
"genome",
"analysis",
"bacteria",
"research",
"and",
"analysis",
"methods",
"genome",
"complexity",
"caulobacter",
"crescentus",
"systems",
"biology",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"gene",
"regulatory",
"networks",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology",
"organisms"
] |
2014
|
The Coding and Noncoding Architecture of the Caulobacter crescentus Genome
|
Critical transitions are sudden , often irreversible , changes that can occur in a large variety of complex systems; signals that warn of critical transitions are therefore highly desirable . We propose a new method for early warning signals that integrates multiple sources of information and data about the system through the framework of a generalized model . We demonstrate our proposed approach through several examples , including a previously published fisheries model . We regard our method as complementary to existing early warning signals , taking an approach of intermediate complexity between model-free approaches and fully parameterized simulations . One potential advantage of our approach is that , under appropriate conditions , it may reduce the amount of time series data required for a robust early warning signal .
Critical transitions are sudden , long-term changes in complex systems that occur when a threshold is crossed [1] . Many systems are known to be at risk of such transitions , including systems in ecology [2] , climate research [3] , economics [4] , sociology [5] and human physiology [6] . Examples of critical transitions in ecology include shifts in food web composition in shallow lakes [7] , degradation of coral reefs [8] , degradation of managed rangelands [9] , and desertification [10] . Warning signals for impending critical transitions are highly desirable , because it is often difficult to revert a system to the previous state once a critical transition has occurred [2] , [11] . If an accurate mathematical model of the system is available then critical transitions can be predicted straight-forwardly , either by numerical simulation or by direct computation of the dynamical thresholds . For real world complex systems , however , sufficiently accurate models are in general not available , and predictions based on models of limited accuracy face substantial difficulties [12] . Recent research has therefore focused on model-free approaches that extract warning signals from observed time series [13] . Two of the most widely used approaches are increasing variance [14] and autocorrelation [15] , both of which are caused by critical slowing down [16] . Other approaches consider warning signals based on skewness [17] , flickering [18] and spatial correlation [19] . One strategy for improving the quality of an early warning signal , which to our knowledge has not been explored , is to utilize other information that may be available . This other information may take the form of other time series data , for example in ecological applications birth rates as well as population sizes , or additional knowledge about the system , such as that the top-predator mortality is likely to be linear . This highlights the need for intermediate approaches , which can efficiently incorporate available information without requiring a fully specified mathematical model . In the present Letter , we propose an approach for the prediction of critical transitions based on the framework of generalized modeling [20] , [21] . The approach allows available information to be used , subject to certain limitations on the quality and availability of the information . Our results indicate that in the cases considered here , the approach can reduce the amount of time series data required or increase the quality of the prediction . We demonstrate the applicability of the proposed approach by considering a simple one-population model , a previously studied fisheries model and a tri-trophic food chain .
Suppose that a system has been identified as being at risk of a critical transition . Even if very little specific information is available , the dynamics can generally still be captured by a so-called generalized model [20] . Such a model captures the structure of the system , without restricting it to specific functional forms . To formulate a generalized model we first identify important system variables ( say , abundance or biomass of the populations in the system ) and processes ( for example , birth , death , or predation ) . As a first step , the generalized model can then be sketched in graphical form , such as in Fig . 1 below . This graphical representation is sometimes called a causal loop diagram [22] . To obtain a mathematical representation of the model we write a dynamical equation for each variable . In these equations we represent the processes by general functions . For instance we can model a single population subject to gains and losses by an ordinary differential equationor as a discrete-time mapNote , that in contrast to conventional models , we do not attempt to describe the processes G and L by specific functional forms . Instead , we use unspecified functions and as formal placeholders for the ( unknown ) relationships realized in the real system . We assume that before the critical transition , the system can be considered in equilibrium . We emphasize that this does not require the system to be completely static or closed in a thermodynamic sense , but that , on the chosen macroscopic level of description , the system can be considered at rest . For example , the system may be undergoing stochastic fluctuations of a fixed distribution around a stable fixed point . Additionally , the system is subject to a slowly changing external parameter that puts it at risk of undergoing a critical transition . The system is therefore at equilibrium only on a certain timescale . In the following we refer to this timescale as the fast timescale , while the dynamics of the whole system , including the slow change of the external parameter , takes place on the slow timescale . Using the definitions above critical transitions can be linked to instabilities ( bifurcations ) of the fast subsystem [23] . For detecting these instabilities we construct the Jacobian matrix , a local linearization of the system around the steady state [24] . A system of ordinary differential equations ( ODEs ) is dynamically stable if all eigenvalues of the Jacobian have negative real parts , whereas a discrete time map is stable if all eigenvalues have an absolute value less than one . Critical transitions are thus signified by a change in the external parameter causing at least one of the eigenvalues to cross the imaginary axis ( ODE ) or a unit circle around the origin ( map ) . To warn of impending critical transitions we monitor the eigenvalues of the Jacobian of the fast subsystem , which usually change slowly in time . A warning is raised if at least one of the eigenvalues shows a clear trend toward the stability boundary ( for ODEs , zero real part; for maps , absolute value of one ) . The Jacobian itself can be computed directly from the generalized model , but will contain unknown terms reflecting our ignorance of the precise functional forms in the model . Previous publications [20] have shown that these unknowns can be treated as well-defined parameters with clear ecological interpretations . In the present applications we estimate the unknowns in the Jacobian matrix from short segments of time series data . Thereby , a pseudo-continuous monitoring of the eigenvalues of the fast subsystem is possible . The generalized model that is constructed should reflect existing knowledge about the structure of the system . It should contain terms that represent relevant and observable processes ( or relevant processes whose magnitudes can be deduced from other processes , as we will see below ) . The generalized model should also have a structure that permits bifurcations that are relevant for the system; if not , the generalized model cannot be used to anticipate those bifurcations . We note that with given time series data estimating the generalized model parameters is simpler than estimating the entries of the Jacobian matrix directly , because the generalized model already incorporates structural information about the system . Further , many of the parameters in the generalized model may already be available in a given application , because they refer to well-studied properties of the species , such as natural life expectancy or metabolic rate .
Allee effects , that is , positive relationships between per-capita growth rate and population size , are postulated in many populations and have been conclusively demonstrated in some [25] . A population with an Allee effect can suddenly transition from a stable , non-zero population size to unconditional extinction [26] . We supposed that an early warning signal was desired for a population in which a slowly increasing death rate ( for example the spread of a new disease , the appearance of a new predator , or habitat destruction ) was pushing the population towards a critical transition associated with an Allee effect . We assumed that regular observations of the population size and birth rate were available . Accordingly , we constructed the generalized model ( 1 ) where and are the birth and death rates of the population , respectively , and represents the external factor pushing the system towards the critical transition . We refer to the population and birth rate observations as and , taken at times , . ( Observations of the death rate would also be acceptable in place of the birth rate . ) From the generalized model of Eq . ( 1 ) , we constructed the Jacobian ( in this 1-D system , also the eigenvalue ) of the system ( 2 ) near its steady state , where the prime denotes the derivative with respect to . To calculate the changing values of the eigenvalue as the external parameter changes , we need to estimate the gradients and from our time series observations of and . We calculated as follows . Since the birth rate and the population have been directly observed , could therefore be computed immediately , where we use the notation . ( These one-sided derivative estimators involve a loss in accuracy but allow the eigenvalues to be estimated at the most recent observation time , which is important when attempting to predict an imminent transition . ) A discretization of Eq . ( 1 ) gives . We cannot calculate in the same way as , because also depends on . Instead , we make one additional assumption: That the mortality is linear in ( although the coefficient of this linearity may change with ) . Then we can estimate . ( Suppose . Then . ) Finally , the eigenvalue . To test the early warning signal , we simulated a simple model ( given in the Supporting Information as Text S1 ) of an Allee effect with additive noise . A critical transition occurred , causing subsequent extinction of the population ( Fig . 2 ) . The challenge we addressed is predicting the critical transition from a limited number ( here , fifteen ) of observations of population size and birth rate . We emphasize that we did not utilize any information on the functional forms of processes employed in the simulation , so that the prediction is based solely on the 15 observations and the assumed structural information ( that is , one population subject to gains and losses ) . By estimating the parameters of the generalized model as described above , we determined the eigenvalues of the Jacobian as a function of time ( Fig . 2b ) . A clear increase in the eigenvalue is detectable well before the critical transition , giving ample warning of the impending collapse . Due to a phenomenon called bifurcation delay [23] , the population size did not start to change rapidly until well after ( ) the bifurcation point ( ) . As previously observed by Biggs et al . [27] , management action to reverse the change in bifurcation parameter may successfully avert the critical transition even after the fast subsystem's bifurcation has occurred , if still within the range of the bifurcation delay . In the case of Fig . 2b , the eigenvalue trend is directed more towards the last possible time that successful management action can be taken than towards the time of the actual bifurcation . Our second case study focuses on an example from fisheries . Increased harvesting of piscivores can induce a shift from the high-piscivore low-planktivore regime to a low-piscivore high-planktivore regime [28] . Many fisheries are suspected to have undergone such transitions [29] , [30] . Based on the causal loop diagram ( Fig . 1 ) , we formulated a discrete-time generalized model , describing the piscivore and planktivore populations at the end of each year , in the spirit of stock-assessment modeling ( see Text S1 ) . Thereby detailed modeling of the intra-annual dynamics was avoided . To test the warning signal , we generated time series data with a detailed fishery model by Biggs et al . [27] , which was developed from a series of whole-lake experiments [31] . We describe this model more fully in Text S1 , but note here that the model differs significantly from our generalized model by a ) accounting for the intra-annual dynamics and b ) containing an additional state variable denoting the juvenile piscivore population . These discrepancies were intentionally included to reflect the limited information that would be available for the formulation of the generalized model in practice . In simulations the detailed model showed a transition to a low-piscivore high-planktivore regime as the harvesting rate was increased ( Fig . 3a ) . From this simulation , we recorded the simulated piscivore and planktivore density and piscivore catch at the end of each year . Because the simulated data was very noisy we estimated the Jacobian's eigenvalues after smoothing the recorded data ( see Text S1 ) . In addition to the time series data , the information on the natural adult piscivore mortality and reproduction rate and the planktivore influx from refugia were required ( see Text S1 ) . This type of information can be reasonably well estimated for most systems . We confirmed that our predictions ( reported below ) are not sensitive to the specific values used . Indeed , a simple approach for estimating these parameters is to recognize that the initial state , before the critical transition , is stable . In a number of test trials we confirmed that any reasonable combination of parameters used that corresponded to an initially stable steady state provided an early warning signal comparable to the results reported below . An estimate of the Jacobian eigenvalues for the fisheries example is shown in Fig . 3b . As the system approaches the critical transition we observe that an eigenvalue approaches one , which signifies a critical transition for discrete time systems . This result is compared to the variance early warning signal computed by Biggs et al . [27] , which uses a much more densely sampled time series including intra-annual dynamics . The comparison shows that the approach proposed here produces a signal of similar quality ( although possibly too early ) , while requiring significantly less time series data . Further , comparison with a variance signal using the same amount of time-series data as the generalized model shows that the generalized model-based signal is a much clearer early warning signal in this case . In particular , the variance signal only rises during or after the transition . For our final example we consider a tri-trophic food chain . In ecological theory food chains play a role both as a prominent motif appearing in complex food webs and as coarse-grained models . Using generalized models , a general Jacobian for a continuous-time model of the tri-trophic food chain can be derived ( see Text S1 and Gross et al . [32] ) . We generated example time series data using a set of three ordinary differential equations that modeled a producer biomass , , predator biomass , , and top predator biomass , , as described in Text S1 . We included additive noise terms in the equations , and if any biomass decreased to zero we suppressed the noise term so that the corresponding population remained extinct . We simulated these equations while increasing the mortality rate of the top predator . The resulting time series , for the chosen combination of parameters , show a slowly changing steady-state followed by a sudden transition to large oscillations , and a sudden collapse of all three populations ( Fig . 4 ) . To provide an early warning signal for the transition we recorded time series of the three biomasses and the top-predator's death rate , and estimated the parameters of the generalized model from smoothed segments of these time series . Even for the smoothed data we find that one of the eigenvalues is very noisy and sometimes positive . We believe that the presence of this eigenvalue reflects the response of the prey to fluctuations on the higher trophic levels and therefore exclude this value from our analysis . As is increased toward the onset of oscillations , two eigenvalues show a clear increase toward zero real part ( Fig . 4 ) . The two eigenvalues approach zero as a complex conjugate eigenvalue pair , which is indicative of the system undergoing a Hopf bifurcation [24] , which in turn generally implies a transition from stationary to oscillatory dynamics . The early warning signal constructed here , consisting of the approach of this eigenvalue pair towards the imaginary axis , warned of the transition to an oscillatory state significantly before the transition occurred . These large oscillations combined with stochastic fluctuations then led rapidly to extinction . Supercritical Hopf bifurcations , to which class the bifurcation in the present system belongs , are by themselves not critical transitions . The detection of Hopf bifurcations is nevertheless of interest . First , subcritical Hopf bifurcations are indeed true critical transitions . Second , even supercritical Hopf bifurcations have long been associated in ecology with rapid destabilization and extinction of populations [33] , a chain of events that we characterize as a critical transition and that we observed to occur in the present system . We also note that although to linear order sub- and super-critical Hopf bifurcations cannot be distinguished , generalized modeling can be extended to higher orders where these cases can be identified [34] .
In this Letter we proposed an approach for anticipating critical transitions before they occur . In particular we showed that generalized modeling of the system can facilitate the incorporation of the structural information that is in general available . We demonstrated the proposed approach in a series of three case studies . The first example showed that in simple systems even very few time points can be sufficient for clean prediction of the critical transition . The second example posed a hard challenge , where test data was generated by a model that differed considerably from the generalized model . Yet even in this case the generalized model significantly reduced the amount of data needed for predicting the transition . The third and final example demonstrated the ability of the proposed approach to distinguish between different types of critical transitions ( in this case , through the presence of a complex conjugate pair of eigenvalues approaching the imaginary axis ) . In all case studies we found that the proposed approach can robustly warn of critical transitions in the presence of noise . We believe that the performance of the approach under noisy conditions can be further improved by subsequent refinements . Such refinements could include combination with dynamic linear modeling [14] , utilization of a parameter transformation ( to ‘scale’ and ‘elasticity’ parameters ) previously proposed for generalized models [20] , or the use of optimized sampling procedures . Two important rules for constructing the generalized model are as follows . First , there must be sufficiently few unobservable processes ( represented by placeholder functions in the generalized model ) that their magnitudes can be inferred from balancing the observable processes . For example , in the Allee effect study , the unobserved process was estimated by . Second , where a process is a function of variables ( although in the cases studied here was never larger than 2 ) , our method at present requires assumptions or other knowledge about the dependence of the process on of those variables . This requirement could be relaxed in future work , although probably at the cost of requiring more time series data . An advantage of the proposed approach is that it generally becomes more reliable closer to a critical transition , where rates of change of state variables and other observables are generally larger , which may lead to better sampling , although noise will also increase close to the transition due to critical slowing down . In such situations statistical methods such as variance may become more difficult to estimate as the time series becomes less stationary , for example since detrending becomes more difficult . On the other hand , the model-free statistical approaches may be more useful where little knowledge is available about the system , or where trends in the means of observed quantities are strongly obscured by noise . In these respects , the proposed approach provides a tool complementary to established statistical methods , each method with its own domain of utility . One limitation shared by both our and the statistical early warning methods is that large noise can bias the estimation of the respective warning signals . In our case , an asymmetric distribution of fluctuations can bias the estimation of the underlying steady state . That our approach effectively involves derivatives of time series can increase the sensitivity to high observation noise or otherwise poor-quality data . Another important assumption in our present treatment ( that is also shared by the statistical approaches ) was that the dynamics of the fast subsystem could , at least at some level of description , be considered as stationary . Let us emphasize that this is not a strong assumption because even systems that are primarily non-stationary , such as the fisheries example , can be modeled as stationary if a suitable generalized modeling framework is chosen . Furthermore , ongoing efforts aim at extending the framework of generalized modeling to non-stationary dynamics , which may lead to a further relaxation of that assumption in the future [35] . A thorough statistical analysis of the generalized modeling and the statistics-based approaches is another topic for future work . Such a study would help to quantify under exactly what conditions the generalized modeling approach can operate effectively and offer advantages compared to statistics-based approaches . In summary , we used generalized models to efficiently incorporate available information about a system without requiring detailed knowledge about that system . Our intermediate-complexity method provides an early warning signal approach complementary to existing statistics-based methods . In the cases studied here , our method could provide early warning signals with significantly less time series data than statistical approaches . Thereby the proposed approach can , under suitable conditions and with good quality data , contribute to the warning of critical transitions from a realistic sampling effort .
|
Fisheries , coral reefs , productive farmland , planetary climate , neural activity in the brain , and financial markets are all complex systems that can be susceptible to sudden changes leading to drastic re-organization or collapse . A variety of signals based on analysis of time-series data have been proposed that would provide warning of these so-called critical transitions . We propose a new method for calculating early warning signals that is complementary to existing approaches . The key step is to incorporate other available information about the system through the framework of a so-called generalized model . Our new approach may help to anticipate future catastrophic regime shifts in nature and society , allowing humankind to avert or to mitigate the consequences of the impending change .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"mathematics",
"ecology",
"biology",
"nonlinear",
"dynamics"
] |
2012
|
Early Warning Signals for Critical Transitions: A Generalized Modeling Approach
|
Bdelloid rotifers are microinvertebrates with unique characteristics: they have survived tens of millions of years without sexual reproduction; they withstand extreme desiccation by undergoing anhydrobiosis; and they tolerate very high levels of ionizing radiation . Recent evidence suggests that subtelomeric regions of the bdelloid genome contain sequences originating from other organisms by horizontal gene transfer ( HGT ) , of which some are known to be transcribed . However , the extent to which foreign gene expression plays a role in bdelloid physiology is unknown . We address this in the first large scale analysis of the transcriptome of the bdelloid Adineta ricciae: cDNA libraries from hydrated and desiccated bdelloids were subjected to massively parallel sequencing and assembled transcripts compared against the UniProtKB database by blastx to identify their putative products . Of ∼29 , 000 matched transcripts , ∼10% were inferred from blastx matches to be horizontally acquired , mainly from eubacteria but also from fungi , protists , and algae . After allowing for possible sources of error , the rate of HGT is at least 8%–9% , a level significantly higher than other invertebrates . We verified their foreign nature by phylogenetic analysis and by demonstrating linkage of foreign genes with metazoan genes in the bdelloid genome . Approximately 80% of horizontally acquired genes expressed in bdelloids code for enzymes , and these represent 39% of enzymes in identified pathways . Many enzymes encoded by foreign genes enhance biochemistry in bdelloids compared to other metazoans , for example , by potentiating toxin degradation or generation of antioxidants and key metabolites . They also supplement , and occasionally potentially replace , existing metazoan functions . Bdelloid rotifers therefore express horizontally acquired genes on a scale unprecedented in animals , and foreign genes make a profound contribution to their metabolism . This represents a potential mechanism for ancient asexuals to adapt rapidly to changing environments and thereby persist over long evolutionary time periods in the absence of sex .
Bdelloid rotifers ( Rotifera , Bdelloidea ) are abundant , ubiquitous microinvertebrates that inhabit aqueous habitats [1] . They possess an extraordinary and unique combination of characteristics among the Metazoa: they have survived for tens of millions of years without sexual reproduction , while speciating similarly to sexual organisms; they can withstand extreme desiccation by undergoing anhydrobiosis; and they display other properties usually associated with extremophiles such as tolerance of high levels of ionizing radiation [2]–[7] . In addition , the bdelloids Adineta vaga and Philodina roseola contain foreign DNA sequences in at least some subtelomeric chromosomal regions of their genomes , and these probably derive from horizontal gene transfer ( HGT ) [8] . Three of these genes were shown to be transcribed , and Boschetti et al . [9] showed that in a related bdelloid species , A . ricciae , four different foreign genes , out of a set of 36 identifiable foreign and native sequences sampled , were expressed . Of these , three were upregulated by evaporative water loss and were therefore part of the desiccation stress response . This suggests that horizontal gene transfer ( HGT ) might contribute significantly to the remarkable biology of the bdelloid rotifer . However , the proportion of the bdelloid genome harbouring foreign sequences , how many of these sequences are expressed , and their contributions to bdelloid physiology , are completely unknown . To address these issues , we present the first global analysis of the transcriptome of a bdelloid rotifer , A . ricciae , which shows that horizontally acquired genes are expressed on a scale unprecedented in animals and that they make a profound contribution to bdelloid metabolism . We suggest this is highly significant in the context of the extremophile nature of bdelloids and their long term evolutionary persistence without sex , which theory suggests should limit their ability to adapt to changing environments [10]–[13] .
To capture expression of genes active during the hydrated and dehydrated states , cDNA was prepared and pooled from a laboratory strain of A . ricciae under both conditions , then partially normalised to reduce coverage of abundant transcripts . Paired-end , massively parallel sequencing was performed on cDNA fragments of mean size 200 bp using the Illumina platform; 19 . 5 million 76-base reads were assembled to give an initial library of 61 , 219 transcript contigs of size range 118–3674 bp . Of these , 28 , 922 contigs gave at least one significant blast hit ( E-value≤10−5 ) when compared to the UniProtKB database , allowing the identification of their likely product , and these were used for further analysis . Those transcripts originating from horizontally acquired genes were identified by assigning each contig an HGT index , hU , defined as the difference between the “bitscore” ( i . e . score in bits ) of the best non-metazoan match and the bitscore of the best metazoan match in the database . The subscript , U here , signifies the database used , UniProtKB; S for Swiss-Prot is used where appropriate below . A positive hU value for a given transcript means that its translation gives a better alignment to a non-metazoan protein than to a metazoan protein , and vice versa for a negative hU value . For comparison with other invertebrates , we carried out the same analysis with transcript datasets from the monogonont rotifer Brachionus plicatilis ( a distinct class within phylum Rotifera , that has both sexual and asexual life phases , and is not considered anhydrobiotic , but can form desiccation-tolerant resting eggs ) , the nematode Caenorhabditis elegans and the fly Drosophila melanogaster . Although for hU>0 , a non-metazoan origin is indicated , there will be some uncertainty where non-metazoan bitscores are close to those of metazoans . Therefore , a threshold signifying foreign origin needs to be set at some value higher than zero . Figure 1A shows that the bdelloid contains many more foreign transcripts than other invertebrates , regardless of where a threshold might be set , and therefore other species can be used as a reference for ‘background’ levels of HGT in invertebrates . We calculate R ( hU ) , the relative proportion of transcripts with HGT index value greater than a given value of hU , where R = ( the percentage of transcripts from species 1 with HGT index≥hU ) ÷ ( the percentage of transcripts from species 2 with HGT index≥hU ) . In comparisons between A . ricciae and other invertebrate species , we notice that , for hU≤0 , R is relatively constant since both metazoan and non-metazoan sequences are included . However , as the hU = 0 threshold is passed , R increases with hU as metazoan sequences are excluded , and the greater proportion of foreign sequences in the bdelloid transcriptome becomes apparent . R then plateaus around hU = 25–30 and is approximately constant up to hU∼100 ( Figure 1B ) . This suggests that , as the threshold of hU = 30 is exceeded , the proportion of sequences judged to be foreign decreases , but at a similar rate in both the bdelloid and the comparator species , i . e . the ratio between species remains constant , indicating that increasing stringency above hU = 30 only results in loss of truly foreign genes from the count , and does not give a better test of “foreignness” . Figure 1B also shows that there is approximately 5-fold more HGT in A . ricciae than in either B . plicatilis or C . elegans , since R≈5 for hU≥30 . For the comparison of A . ricciae and D . melanogaster , the ratio is appreciably higher at R≈16 ( data not shown ) , in line with the apparently very low levels of HGT in the fly ( Figure 1A ) . A comparison of B . plicatilis with C . elegans ( Figure 1B ) does not show the inflection between hU = 0 and hU = 30 , consistent with these species having a similar proportion of foreign sequences in their transcriptomes . We used linear models to test whether differences in hU were significant among taxa . Results confirmed that A . ricciae had both a significantly higher mean hU score and a significantly higher probability per gene of hU>30 than the other taxa , even when controlling for differences in contig length between the assemblies ( all comparisons , p<0 . 001 , details in legend to Figure S1 ) . Of the identified bdelloid contigs , 9 . 7% ( 2 , 792/28 , 922 ) were shown to have hU≥30 and so were considered to be of foreign origin ( Figure 1A , Table S1 ) . In B . plicatilis , 1 . 8% ( 171/9 , 685 ) of transcripts have hU≥30 , while in C . elegans and D . melanogaster this figure is 1 . 8% ( 206/11 , 168 ) and 0 . 6% ( 105/18 , 368 ) , respectively ( Figure 1A ) . This demonstrates that , independent of the dataset dimensions , the level of expressed HGT in bdelloid rotifers is far greater than in other invertebrates tested . Phylogenetics was used to validate the foreign origins of putative horizontally acquired sequences [14] and this can be performed meaningfully where contigs with hU≥30 have a significant ( E-value≤10−5 ) blast match to at least one metazoan sequence , allowing a phylogenetic tree to be constructed . However , two-thirds ( 1 , 884/2 , 792; 67% ) of sequences with hU≥30 do not give a significant metazoan match , which strongly supports a foreign origin . For the remaining ( 908/2 , 792 ) contigs , phylogenetic trees were built in PhyML from amino-acids sequences using a JTT model [15] . Each contig was assigned to a particular group according to the aLRT support for each metazoan or non-metazoan taxon as follows: group 1 contains sequences that are monophyletic with Metazoa ( or where there were only metazoan hits from the blast analysis ) ; group 2 contains sequences for which monophyly with Metazoa cannot be strongly rejected; group 3 contains cases where there are too few sequences to define a meaningful clade; group 4 contains cases where monophyly with Metazoa can be strongly rejected; group 5 contains transcripts which are monophyletic with another single ( non-metazoan ) taxon . Analysis of these data showed that 98% of A . ricciae transcripts with hU≥30 and at least one significant metazoan hit fall into groups 4 and 5 with high node support ( summarised in Table 1; Table S1; Figure S2 ) and therefore are supported as truly non-metazoan . For example , an acetyl-CoA synthetase ( Enzyme Commission [EC] number 6 . 2 . 1 . 1 ) does not cluster with metazoan sequences for this enzyme , instead grouping within the eubacterial clade with high support ( aLRT support = 0 . 86 ) ( Figure 2A; Figure S2C ) . More than half of foreign transcripts appeared prokaryotic ( 59% eubacterial , 1% archaeal ) ; the remainder were eukaryotic in origin: 23% fungal , 6% from algae or plants , and 11% from other eukaryotic taxa ( largely protists ) . A similar analysis can be performed for other invertebrates . For example , there are 206 transcripts from C . elegans with hU≥30 of which 108 give significant blast matches only with non-metazoan sequences . For the remaining transcripts , phylogenetic analysis shows that 92% ( 90/98 ) fail to cluster with metazoan examples ( summarised in Table 1; Table S2; Figure S3 ) . Therefore , 96% ( 198/206 ) of these C . elegans transcripts were verified as foreign by the phylogenetics . Although there are no comprehensive studies in the literature , the frequency of HGT we detect in C . elegans is higher than inferred in an earlier study [16] . One possible confounding factor might be that the phylogenetic placement of individual C . elegans sequences is impaired by filtering out other nematode sequences ( see Materials and Methods ) . To check this , we repeated the evaluation including the top blast hits from nematodes , i . e . homologous and paralogous examples ( Table S2; Figure S3 ) . From the phylogenetics , we found that 93% ( 91/98 ) of C . elegans sequences did not cluster with the metazoa and therefore 97% ( 199/206 ) of the total set of transcripts with hU≥30 are likely to be foreign . This shows that the vast majority still lack a close non-nematode metazoan match when additional nematode sequences are included in the analysis . We interpret this finding as evidence of HGT in an ancestor of nematode species in the sample . However , as our aim here is not to evaluate levels of HGT in other metazoa beyond providing a baseline for comparison with bdelloids , these analyses are meant to illustrate that the results are robust to variations in the method , such as which sequences are included for evaluation . To confirm that foreign transcripts originated from the bdelloid genome and were not due to contaminating or commensal organisms , several corresponding genomic regions were PCR-amplified and Sanger sequenced , and this showed that foreign genes were linked to a gene of metazoan origin or to another foreign gene from a different taxon ( Figure 2B ) . In some cases ( asterisks in Figure 2B ) , the foreign transcript was close to a gene previously described in a bdelloid rotifer . The sequences were also aligned to an early draft of the A . ricciae genome , where 91% of foreign transcripts aligned for at least 50% of their length , compared to 90% of all transcripts and for metazoan transcripts only ( Figure 2C; data not shown ) . Furthermore , 81% of foreign transcripts were aligned to the same genomic contig as metazoan transcripts or foreign transcripts of a different phylogenetic group , which rules out an origin from contamination for this set ( examples given in Table S3 correspond to some foreign sequences in Figure 3 ) . This proportion is likely to rise as genome assemblies improve , but even if 10–20% of foreign genes cannot be shown to be part of the bdelloid genome , and thus represent contamination , this would only reduce the foreign component of the transcriptome to 8–9% , rather than 10% , which is still remarkably high . Where HGT has been observed between prokaryotes , operational genes encoding , for example , enzymes , predominate over informational genes concerned with transcription and translation [17] , [18] . If a similar situation pertains in bdelloids , we would expect to find many foreign genes that encode enzymes , which largely fall into the operational category [18] . Bdelloid transcripts with biochemical functions were identified by alignment to proteins with EC numbers in the Swiss-Prot database . This database was used as the quality of annotation is higher than UniProtKB and the smaller number of proteins should reduce the false positive rate ( although it will also increase the number of false negatives ) . Of the 26 , 001 transcript contigs with matches in the Swiss-Prot database , 2 , 947 ( 11 . 3% ) had hS≥30 and were categorised as foreign , i . e . a similar proportion to the previous analysis using UniProtKB ( Figure S4 ) . Approximately 50% ( 13 , 059/26 , 001 ) of contigs ( irrespective of their hS value ) had a match with an assigned EC number . These were then tagged as either metazoan ( hS≤0 ) , indeterminate ( 0<hS<30 ) or foreign ( hS≥30 ) . Therefore , of the foreign transcripts , 79% ( 2 , 341/2 , 947 ) were annotated with an EC number , showing that a large majority are concerned with metabolism . In fact , when the functions of those without an EC number were analysed , a further 93 sequences could be identified as enzymes that lacked EC numbers due to poor annotation . This increases the proportion of foreign transcripts encoding metabolic functions to 83% ( Figure S4 ) . Transcript contigs ( in all categories ) with assigned EC numbers were mapped onto the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) reference pathways ( denoted ‘K’ plus a number in the following ) . In total , 839 EC numbers assigned to the rotifer transcriptome were matched to 152 metabolic pathways ( Table 2 and Table S4 ) . Of the 839 EC numbers , 23% ( 191 ) were only assigned to foreign transcripts , with a further 16% ( 138 ) being assigned to both foreign and metazoan transcripts . This made a total of 39% of identified enzyme activities with a contribution from foreign transcripts , suggesting that HGT has the potential to radically diversify bdelloid biochemistry . Many pathways containing foreign transcript products specify metabolism found only in micro-organisms and unknown in metazoans ( Figure 3 , Figure S5 , Table 2 , Table S4 ) . Several of these are concerned with degradation of toxic compounds , and we give three examples here: 1 ) breakdown of phenylacetonitrile ( benzyl cyanide ) is initiated by the products of two genes derived from bacteria ( EC 4 . 2 . 1 . 84 or EC 3 . 5 . 5 . 1; K00643; Figure 3A , Figure S5A ) , and other nitrile compounds , such as benzonitrile , can also be metabolised similarly ( K00627 , Figure S5B ) ; 2 ) the organochloride pesticide , 1 , 3-dichloropropene , is degraded in five steps to the central metabolite , acetaldehyde , and the first of these is exclusively specified by the foreign-encoded enzyme haloalkane dehalogenase ( EC 3 . 8 . 1 . 5; K00625; Figure 3B , Figure S5C ) ; 3 ) branches of the degradation pathways for benzoate ( K00362 ) and bisphenol ( K00363 ) are also represented by foreign gene products ( Figure S5D , S5E ) . Not all steps in these pathways are present in our transcriptome sample . This is partially due to the use of the Swiss-Prot database to assign EC numbers; performing the same analysis using the UniProtKB database adds steps to many pathways . However , there might also be incomplete capture of transcripts during cDNA cloning and sequencing , or bdelloids might only partially metabolise certain compounds . If the latter is correct , such partial metabolism might still be sufficient for detoxification or metabolite utilisation in other pathways . HGT is also implicated in improved resource acquisition , e . g . two-step pathways to convert the ubiquitous natural phosphonates 2-aminoethylphosphonate ( AEP ) and 3-phosphonopyruvate into useable metabolites are enabled by foreign transcripts encoding 2-aminoethylphosphonate-pyruvate transaminase ( EC 2 . 6 . 1 . 37 ) or phosphonopyruvate decarboxylase ( EC 4 . 1 . 1 . 82 ) and phosphonoacetaldehyde hydrolase ( EC 3 . 11 . 1 . 1 ) ( K00440; Figure 3C , Figure S5F ) . Furthermore , several foreign transcripts are implicated in utilisation of a range of polysaccharides not normally directly available to metazoans , e . g . cellulose ( K00500; Figure 3D , Figure S5G ) and polygalacturonate ( K00040; Figure 3E , Figure S5H ) breakdown; α-N-arabinofuranosidase ( EC 3 . 2 . 1 . 55; K00520 , Figure S5I ) , glucan endo-1 , 3-β-glucosidase ( EC 3 . 2 . 1 . 39; K00500 , Figure S5G ) and fructan β-fructosidase ( EC 3 . 2 . 1 . 80; K00051 , Figure S5J ) are also encoded . Cellulase activity has been described in other invertebrates but , where this does occur , it seems to result from HGT ( e . g . ref . [19] ) . Other pathways novel to metazoans but represented in the bdelloid transcriptome are biosynthetic , some of which are associated with robustness . These include production of the powerful antioxidant , trypanothione , normally only produced by parasitic protozoa , which is specified by two foreign transcripts: a glutathionylspermidine synthetase ( EC 6 . 3 . 1 . 8 ) , and a trypanothione synthase ( EC 6 . 3 . 1 . 9; K00480; Figure 3F , Figure S5K ) . Such antioxidants could play a role in desiccation tolerance , where protection of repair systems against oxidative stress is thought to be crucial [20]–[22] . Foreign gene products can also add extensions or linking steps to existing metazoan metabolism in A . ricciae . Valine and isoleucine are essential amino acids in animals and must normally be accumulated from the diet . However , foreign transcripts encode ketol-acid reductoisomerase ( EC 1 . 1 . 1 . 86 ) and dihydroxy-acid dehydratase ( EC 4 . 2 . 1 . 9 ) , allowing completion of biosynthetic routes to these amino acids from pyruvate ( K00290; Figure 3G , Figure S5L ) . A . ricciae also encodes a fungal form of pyruvate decarboxylase ( EC 4 . 1 . 1 . 1; K00010 ) , allowing an additional end step to glycolysis for the regeneration of NAD+ from NADH under anaerobic conditions with the production of ethanol ( Figure 3H , Figure S5M ) ; animals usually only form lactate from pyruvate under these conditions . A further intriguing possibility highlighted by the transcriptome analysis is that the bdelloid can fix carbon from CO2 , using eubacterial forms of phosphoenolpyruvate synthase ( EC 2 . 7 . 9 . 2 ) and phosphoenolpyruvate carboxylase ( EC 4 . 1 . 1 . 31; K00720; Figure 3I , Figure S5N ) , by a route used in plants and bacteria , but not in fungi or animals . Where it is meaningful to do so , i . e . where there are significant metazoan blast matches , phylogenetic trees are shown in Figure S2G–S2M for example transcript contigs representing foreign-encoded activities in Figure 3 . In a few cases , where we would expect to find a metazoan sequence , this is absent from the transcriptome and the activity is instead represented by a foreign counterpart . For instance , a fungal form of stearoyl-CoA delta-9 desaturase ( EC 1 . 14 . 19 . 1; K01040; Figure S5O ) , an essential enzyme for the synthesis of monounsaturated fatty acids , is present , but no metazoan equivalent was discovered in the transcriptome . To control for the possibility that relevant metazoan genes had not been expressed in study samples , we searched a draft A . ricciae genome sequence , but failed to find them , although the gene encoding the foreign transcript was present . While the inability to detect a particular sequence is not proof of its absence , this suggests that the metazoan form of stearoyl-CoA delta-9 desaturase has been lost in the bdelloid , perhaps following a detrimental mutation , and that a foreign gene has been co-opted in this role . Other examples of a foreign sequence potentially replacing a metazoan counterpart include nicotinate-nucleotide diphosphorylase ( EC 2 . 4 . 2 . 19; K00760 ) , which catalyses a step in NAD+ biosynthesis , and the antioxidant peptide-methionine ( S ) -S-oxide reductase ( EC 1 . 8 . 4 . 11 ) .
In recent years , there has been increasing interest in HGT , but most investigations have been performed in prokaryotes or in unicellular eukaryotes . In these organisms , HGT is considered a main driver of innovation , often associated with speciation [23] , [24] . In multicellular eukaryotes , there has been less emphasis on HGT , partly because it is thought to occur on a much smaller scale [14] , [25] , and partly because there are fewer well-annotated genome sequences available . Since de novo whole genome assembly is still a significant challenge for complex organisms , particularly for the bdelloid rotifer with its unusual genome characterised by degenerate tetraploidy , divergence of formerly allelic sequences , and gene conversion between gene copies [7] , [26] , [27] , we chose to assess HGT primarily at the transcriptome level . This study represents the first global analysis of the expressed genes in a bdelloid rotifer , A . ricciae , and the contribution of horizontally acquired sequences to its transcriptome . The results reveal a remarkable degree of HGT in the bdelloid , with approximately 10% of identifiable , transcribed sequences deriving mainly from prokaryotes , but also from fungi , plants and algae , and protists . The method for assessing HGT in the bdelloid transcriptome is novel , but follows principles currently recognised as the most rigorous , where sequence matching is coupled with phylogenetics [14] . There have been relatively few such global analyses among the Metazoa that test for expression of horizontally acquired sequences , one example being in Hydra magnipapillata , where seventy-one “gene models” apparently derive from bacteria , 70% of which were shown to be transcribed [28] . For the bdelloid work , we introduced the HGT index , h , which is calculated as the difference in bitscores between best non-metazoan and best metazoan matches in blast alignments , to give a measure of the “foreignness” of any sequence . We preferred the HGT index to the alien index ( AI ) , developed previously for assessing foreign sequences in bdelloid subtelomeric regions [8] and also used in the Hydra study [28] , because h is calculated from bitscores and is therefore not influenced by the sizes of the databases used to perform the blast screen . In contrast , if E-values are used , as for the AI , the score changes with database size . Additionally , an arbitrary constant must be included in the AI formula so that the index does not become infinite with identical matches to database sequences; this adjustment is unnecessary with the HGT index . Although Figure 1A showed that , whatever value of h is chosen , there is a greater proportion of foreign sequences in the bdelloid than in other invertebrates , it is useful to adopt a threshold value to signify a foreign sequence . In principle , any sequence with h>0 is more likely be foreign , but there will be uncertainty at values close to zero where non-metazoan and metazoan sequences have similar degrees of divergence from the test sequence . One technique for identifying a reliable threshold value of h is to normalise the proportion of foreign sequences against the “background” levels found in other invertebrates . The greater proportion of horizontally acquired sequences in the bdelloid then becomes apparent above the minimum threshold level of h required to confidently identify their foreign nature , as shown in Figure 1B . This was validated by phylogenetics , where possible ( i . e . where matching metazoan counterparts exist ) , which showed that the vast majority of bdelloid transcript contigs with hU≥30 did not cluster with metazoan sequences . There are other technical considerations in any assessment of HGT . For example , we classified sequences as either metazoan or non-metazoan , and therefore any HGT from other animals ( including other bdelloids ) into the A . ricciae genome would be missed . Of course , there is no reason to believe that bdelloids are unable to acquire genes from other metazoans , or indeed from other rotifers; in fact , this might be more efficient than acquisition from microorganisms , since fewer changes to metazoan genes should be required before they become expression competent . Therefore , our approach is likely to give a minimum estimate of the extent of HGT in the bdelloid . Another factor that might influence this estimate is the approximately half of transcript contigs that showed no match with known sequences and therefore had to be excluded from further analysis . If all these sequences originate from vertical transmission into the bdelloid lineage , then this would reduce the estimate of HGT . However , there is no a priori reason to assume this: the proportion of foreign sequences in this non-matched set could be higher , lower or about the same as in the matched set . How the matched and non-matched sequence sets are defined could also potentially influence the proportion defined as HGT . We used 10−5 as a maximum value for a significant match when blast screening the transcript contigs against the databases and this gave 28 , 922 contigs in the matched set . If 10−10 or 10−15 is used as a cut-off value , the number of matched contigs decreases to 22 , 719 and 17865 , respectively , but the fraction scored as foreign ( i . e . with hU≥30 ) remains high , at 11 . 5% and 11 . 7% of matched sequences , respectively . Which database is used for blast matching also does not seem to be a major factor since both UniProtKB and Swiss-Prot gave similar proportions of foreign transcripts at 9 . 7% and 11 . 3% , respectively . A final technical consideration might be to ask whether the HGT resulting from the endosymbiosis of the mitochondrial precursor affects our results . Endosymbiosis was a primordial event for eukaryotes , with acquisition of mitochondrial precursors taking place in the earliest eukaryotic cells , perhaps two billion years ago [29] . The horizontal gene transfer we describe is very unlikely to have occurred before the divergence of bdelloids from monogonont rotifers ( or B . plicatilis would share similarly high levels of foreign genes ) , and therefore probably took place at most 100 , more likely 65–80 , million years ago [30] . If horizontal gene transfer has continued throughout bdelloid evolution , many events will be more recent . Consequently , most , perhaps all , gene flux from mitochondrial precursor to nucleus would have occurred before bdelloids arose . Thus , we would not expect significant differences in numbers of nuclear mitochondrial genes between bdelloids and the other major class of rotifers , the monogononts , as exemplified by B . plicatilis in our study . To test what proportion of foreign genes apparently derive from mitochondrial nuclear genes , we blast aligned sequences of 1 , 098 known nuclear mitochondrial genes from MitoCarta ( www . broadinstitute . org/pubs/MitoCarta ) against our transcripts . Using a cut-off of 10−5 , only 0 . 7% of transcripts of foreign origin ( hU≥30 ) matched mitochondrial nuclear genes , whereas 2 . 9% of those of metazoan origin ( hU≤0 ) gave matches . If we adjust the blast cut-off to 10−10 and 10−15 , these proportions are approximately the same: 0 . 7% vs . 3 . 3% , and 0 . 8% vs . 3 . 6% , respectively . This shows that transcripts for nuclear mitochondrial genes are less likely to be found in the foreign sequence set than among metazoan transcripts and therefore will not cause an overestimate of HGT . The complexity of foreign gene expression observed in the bdelloid rotifer A . ricciae is comparable to that in prokaryotes [31] and is far greater than in other animals where relatively few genes are involved [14] , [25] , [28] . For example , while in Hydra perhaps 50 foreign genes are active [28] , in Drosophila ananassae , which has acquired most of the genome of its endosymbiont , Wolbachia , by HGT , only 28 genes are transcribed; the model fly , D . melanogaster , has not acquired the Wolbachia genome [32] , [33] . In pea aphids , red body colour results from the expression of carotenoid genes acquired and diversified from fungal counterparts [34] , [35] . In the sea slug , Elysia chlorotica , HGT and expression of the algal psbO gene allows photosynthesis in plastids also acquired from the alga [36] . However , there is a need for more animal studies at the whole transcriptome level . It is surprising , for example , that there are no comprehensive global studies of HGT in C . elegans in the literature [37] , as our analysis suggests there are approximately 200 foreign transcripts in the model nematode . The software pipeline developed for this study has the potential to be used more widely where expression data are available to gain a more complete picture of HGT in metazoans . Nevertheless , the scale of HGT in the bdelloid seems to be unusual among animals and it would be interesting to address the importance of asexuality and desiccation tolerance in this phenomenon . For example , transcriptome data from the nematode Panagrolaimus superbus , which is anhydrobiotic , but gonochoristic ( i . e . reproduces only sexually ) , has recently been published [38] . The authors highlighted one foreign sequence in the P . superbus transcriptome , but did not perform a global analysis for HGT . If this nematode contains low numbers of foreign sequences , it would rule out that desiccation tolerance per se , without asexuality , is associated with extensive HGT . Another characteristic of HGT in A . ricciae is that the source organisms are extremely diverse and include examples that are unlikely to be symbionts or even in the bdelloid's immediate habitat , such as the trypanosome relative from which trypanothione biosynthesis genes derive . Therefore , bdelloids are likely able to readily scavenge and incorporate DNA from the environment , and desiccation , which could lead to both leakiness in cell membranes and double-strand breaks in rotifer chromosomes , might facilitate this . HGT in the bdelloid has the potential to radically extend and complement metazoan biochemistry , since approximately 80% of foreign sequences are concerned with enzyme activity , much of which is novel in animals . This supports the complexity hypothesis , which states that genes whose products are involved in relatively few protein-protein interactions , such as those specifying enzymes , are more likely to be horizontally transferred than those with a higher degree of connectivity , like transcription factor genes [17] , [39] , [40] . Thus , although the complexity hypothesis was developed to explain observations in prokaryotes , it also seems to apply to the large scale HGT observed in the bdelloid . It would be interesting to investigate in the bdelloid a more recent suggestion from a study in prokaryotes that highly expressed genes are less likely to be horizontally transferred between organisms [41] . Technically , this might be difficult to achieve , as we estimate there are at least 533 source organisms that have contributed to the bdelloid genome by HGT , but we will explore this in future work . The novel biochemistry implicated includes the ability to degrade toxins , and indeed to exploit them and a range of otherwise unmetabolisable organic molecules as food sources , and to use novel biosynthetic pathways to generate protective molecules , for example antioxidants , or nutrients that are rare in the environment . This is expected to increase bdelloid stress tolerance and competitiveness , and could be important for anhydrobiosis . Bdelloids do not produce trehalose or other non-reducing disaccharides [42]–[44] and have unusual LEA proteins [26] , [44] , [45] , and therefore mechanisms associated with desiccation tolerance in other anhydrobiotes do not apply . Recently , the bdelloid A . vaga was shown to have high antioxidant capacity; this reduces protein oxidation , which is thought to be a major problem caused by desiccation and the dry state [22] . Antioxidants in bdelloids have not been characterised , but it will be of interest to determine how far HGT plays a role; this is currently under investigation . It is also tempting to speculate that HGT facilitates long-term persistence in the absence of sex: asexuals are unable to bring together novel gene combinations arising within a population since they lack conventional genetic exchange mechanisms; equally , asexuals cannot eliminate detrimental mutations readily [10] , [11] . Uptake and expression of genes from other organisms is a means of diversifying functional capacity , particularly biochemical capacity , and the potential to replace defective genes with foreign counterparts could protect against loss of function through mutation .
The bdelloid rotifer Adineta ricciae was supplied by Claudia Ricci , University of Milan . A clone culture was split into four populations: one was kept hydrated and the other three were dehydrated for 24 , 48 and 72 h , as described previously [9] . RNA was extracted separately from each bdelloid population with TRI reagent ( Sigma ) according to manufacturer's instructions . RNA purity and concentration were measured with a NanoDrop spectrophotometer . Oligo ( dT ) -primed cDNA from all four sets was prepared with a Clontech/Takara SMART PCR cDNA Synthesis Kit and an Advantage 2 PCR Enzyme System using Invitrogen SuperScript III Reverse Transcriptase . 1 µg cDNA from each preparation was pooled and the resulting mixed cDNA library was normalized with Evrogen Trimmer cDNA normalization kit , according to manufacturer's instructions . About 8 µg of both the normalized and non-normalised cDNA library ( each made of the mixed of hydrated and desiccated rotifers ) were pooled and a paired-end sequencing library with insert size 200 bp was prepared . Massively parallel Illumina sequencing was performed , resulting in 19 . 5 million 76-base reads . These were assembled with the CLC-bio ( www . clcbio . com ) assembler , using a k-mer size of 22 , no minimum contig length and all other options at the default settings . The resulting assembly used 9 , 048 , 520 of the reads ( 46 . 4% ) for a total length of 27 , 227 , 333 bp giving an average coverage of 25 . 3 times . This produced a library of 61 , 219 transcript contigs of size range 118–3674 bp , with median size 341 bp , and mean size 445 bp ( standard deviation 295 bp ) . Transcript contigs have accession numbers HE687510 to HE716431 . Analysis of the bdelloid transcriptome was performed using R ( The R Project for Statistical Computing , http://www . r-project . org/ ) complemented with NCBI-Blast 2 . 2 . 23+–2 . 2 . 25+ ( Basic Local Alignment Search Tool ) [46] , ClustalW2 ( EMBL-EBI ) and PhyML 3 . 0 [47] . Blastx was used to compare the complete set of 61 , 219 bdelloid transcripts against taxa-specific subsets of UniProtKB , labelled as Metazoa , Plantae , Fungi , Eubacteria , Archea and “Other Eukaryotes” ( Eukaryotes which are neither Metazoa nor Plants nor Fungi ) . The taxa-specific subsets only included sequences from complete proteomes ( keyword: KW-0181 ) in order to reduce the search space and to avoid bias towards specific types of proteins that have been sequenced in many organisms . E-value and bitscores were collected for the best five hits of each transcript against each taxon , and 32 , 297 sequences that did not have any match with at least one taxon with an E-value≤10−5 were excluded from further analysis . The alien index [8] and the HGT index ( hU ) were calculated for each of the remaining 28 , 922 sequences . The HGT index ( hU ) is calculated as the difference between the highest non-metazoan and the highest metazoan bitscore . Bitscores , being independent of the search space , do not depend on the size of the database used to calculate the blast score , reducing the incorrect determination of sequences . Setting the hU threshold value is explained in the text . Similar analyses were performed for the C . elegans ( WormBase release WS226; www . wormbase . org ) , D . melanogaster ( FlyBase release r5 . 37 ) and B . plicatilis transcriptomes . In the first two of these cases , proteins from the phylum containing the test organism ( i . e . Nematoda/Arthropoda ) were excluded from the Metazoan database , as is common practice [8] , [28] . For both A . ricciae and B . plicatilis this exclusion was not necessary as there are currently no complete proteomes available for the phylum Rotifera . For B . plicatilis , ESTs with accession numbers FM897377–FM945301 [48] were first assembled with CAP3 [49] into 16024 contigs , which became 9685 contigs after filtering for a blastx E-value≤10−5 . To confirm the non-metazoan origin of the sequences with hU≥30 and at least one significant metazoan hit , each transcript meeting these conditions was translated and aligned using ClustalW2 to the output ( the best hits for each of the five taxa ) of the previous blastx analysis . Each alignment was then trimmed to exclude regions where only one of the sequences was present , and phylogenetic trees were built in PhyML from amino-acids sequences using a JTT model [15]; branch support was calculated with the aLRT ( approximate Likelihood-Ratio Test ) method . The transcripts were then assigned to one of five groups according to the aLRT support for each metazoan or non-metazoan taxon: group 1 contains sequences that are monophyletic with Metazoa ( or where there were only metazoan hits from the blast analysis ) ; group 2 contains sequences for which monophyly with Metazoa cannot be strongly rejected; group 3 contains cases where there are too few sequences to define a meaningful clade; group 4 contains cases were monophyly with Metazoa can be strongly rejected; group 5 contains transcripts which are monophyletic with another single ( non-metazoan ) taxon . Analysis of these data showed that 98% of the sequences with at least one significant metazoan hit and hU≥30 are truly non-metazoan as they fall into groups 4 and 5 ( Table 1; Table S1 ) . To compare the bdelloid transcriptome to those of other species , the same analysis was performed on the published transcriptomes from the monogonont rotifer B . plicatilis , the nematode C . elegans and the fly D . melanogaster , calculating the percentage of sequences above threshold for a given value of hU as shown in Figure 1A . To confirm the presence of foreign genes in the bdelloid genome and to assess the possibility of contamination from food , symbionts , parasites and other organisms , we manually sequenced the genomic DNA around some genes of interest . A number of assembled transcript fragments , chosen at random from a subset of foreign sequences encoding biochemical functions that have never been reported in metazoans , were blast screened against a ( partial ) genome assembly of A . ricciae and the longest genomic DNA contig for each transcript was identified . This was then compared using blastx to the NCBI non-redundant database to find other genes on the same genomic DNA fragment , and primers were designed around these regions . Genomic DNA was extracted from an A . ricciae population derived from the original , and 11 individual regions , were PCR amplified using Finnzymes Phusion High Fidelity Taq polymerase , adding an A overhang after PCR with Advantage 2 PCR Enzyme System . The resulting PCR product was cloned into pCR 2 . 1 TOPO TA ( Invitrogen ) , inserted into competent E . coli ( New England BioLabs ) and white-blue colony screening was performed . Ten positive colonies for each PCR product were chosen , and plasmid DNA was purified and restriction digested to check for insert size . One clone for each genomic DNA region was sequenced via primer walks using a standard dideoxy method at the University of Cambridge Department of Biochemistry Sequencing Facility . Of the 11 attempted , eight are shown as Figure 2B . For the remaining three examples , one amplification worked , but sequencing could not be completed since the insert was long and unstable in E . coli: although the sequence of the middle of this fragment could not be determined , we confirmed that one end contained a metazoan gene and the other contained two genes of bacterial origin . Another amplification was not of the target region , and one amplification failed altogether . The successfully amplified and sequenced genomic DNA regions were then manually aligned in Geneious ( www . geneious . com ) with the relevant transcripts from the library , then blastx aligned against the non-redundant NCBI database and annotated . Each annotated gene was considered metazoan or not-metazoan according to its best hits in the published database . Figure 2B represents eight genomic regions with the annotated genes colour-coded according to the tree in Figure 2A ( metazoa , black; eubacteria , blue; fungi , pink; protists , grey ) . In two cases , two foreign genes are present on the same genome fragment , but they derive from different taxa . Occasionally , a gene next to a foreign representative has been identified previously in a bdelloid rotifer species , and is annotated with an asterisk in the figure . Figure 2B shows the shortest region including one foreign gene and one metazoan ( or another non-metazoan from a different taxon ) gene , but in a few examples the actual sequenced region was longer than shown . Accession numbers for these eight genomic regions are HE662868 to HE662875 . Transcripts were aligned to the draft genome using blastn . To determine the total length of alignment of a transcript all matches for that transcript fulfilling the following criteria were used: 1 ) E-value≤10−3; 2 ) non-overlapping with any previous matches; 3 ) longer than 40 bp ( a minimum exon length constraint ) ; and 4 ) on the same genomic contig as a previous match OR within 1000 bp of the start/end of a genomic contig when a previous match was also within 1000 bp of the start/end of a genomic contig ( a maximum intron length constraint ) . The total aligning length ( sum of the length of the matches that fulfill these conditions ) was then divided by the length of the transcript and this percentage plotted as a histogram for all transcripts . Transcripts were considered to be “on” a genomic contig if they had a match on it fulfilling the above criteria . For each genomic contig with a foreign transcript on it the number of metazoan transcripts and foreign transcripts with a different origin was calculated . The 28 , 922 sequences with at least one match with E-value≤10−5 were blastx aligned to the whole of Swiss-Prot ( 532 , 146 proteins at time of analysis ) and the results were filtered to give only transcripts with at least one match to a protein that was annotated with an EC number . An HGT index for these transcripts was then calculated as before ( but denoted hS , to show that the comparison was done with Swiss-Prot rather than with UniProtKB , cf . hU ) . Based on hS , the transcripts were then subdivided into three groups: horizontally transferred ( hS≥30 ) , indeterminate ( 0<hS<30 ) and metazoan ( hS≤0 ) . For horizontally transferred and metazoan transcripts the EC numbers of their first match were collated and input into the KEGG website ( http://www . genome . jp/kegg/tool/map_pathway1 . html ) to determine which KEGG pathways they occurred in . EC numbers were also assigned a colour: green ( EC number is only annotated to matches of transcripts with metazoan origin ) , red ( EC number is only annotated to matches of horizontally transferred transcripts ) , grey ( EC number is only annotated to matches of transcripts with indeterminate origin ) , orange ( green plus red ( and possibly grey ) ) , pink ( red plus grey ) , light green ( green plus grey ) . These were input into the KEGG website ( http://www . genome . jp/kegg/tool/map_pathway2 . html ) to produce the coloured pathway diagrams shown in Figure 3 and Figure S5 . The results were then extracted from the KEGG website and a hypergeometric test performed to calculate which KEGG pathways were enriched for horizontal transfer as compared to the total number of unique EC numbers found for all transcripts and the total number of unique EC numbers found for horizontally transferred transcripts . Benjamini-Hochberg multiple testing correction was performed to reduce the false positive rate ( Table S4 ) . The workflow is shown in Figure S6 .
|
Bdelloid rotifers are tiny invertebrates with unusual characteristics: they withstand stresses , such as desiccation and high levels of ionising radiation , that kill other animals , and they have survived over millions of years without sexual reproduction , which contradicts theories on the evolutionary advantages of sex . In this study , we investigate another bizarre feature of bdelloids , namely their ability to acquire genes from other organisms in a process known as horizontal gene transfer ( HGT ) . We show that HGT happens on an unprecedented scale in bdelloids: approximately 10% of active genes are “foreign , ” mostly originating from bacteria and other simple organisms like fungi and algae , but now functioning as bdelloid genes . About 80% of foreign genes code for enzymes , and these make a major contribution to bdelloid biochemistry: 39% of enzyme activities have a foreign contribution , and in 23% of cases the activity in question is uniquely specified by a foreign gene . This indicates biochemistry , such as toxin degradation and antioxidant generation , that is unknown in other animals and that is expected to improve the “robustness” of the bdelloid . It also represents a possible mechanism for survival without sex , by diversification of functional capacity and even replacement of defective genes by foreign counterparts .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genomics",
"functional",
"genomics",
"genetics",
"biology",
"evolutionary",
"biology",
"evolutionary",
"genetics",
"genetics",
"and",
"genomics"
] |
2012
|
Biochemical Diversification through Foreign Gene Expression in Bdelloid Rotifers
|
Syphilis is a chronic disease caused by the bacterium Treponema pallidum subsp . pallidum . Treponema pallidum disseminates widely throughout the host and extravasates from the vasculature , a process that is at least partially dependent upon the ability of T . pallidum to interact with host extracellular matrix ( ECM ) components . Defining the molecular basis for the interaction between T . pallidum and the host is complicated by the intractability of T . pallidum to in vitro culturing and genetic manipulation . Correspondingly , few T . pallidum proteins have been identified that interact directly with host components . Of these , Tp0751 ( also known as pallilysin ) displays a propensity to interact with the ECM , although the underlying mechanism of these interactions remains unknown . Towards establishing the molecular mechanism of Tp0751-host ECM attachment , we first determined the crystal structure of Tp0751 to a resolution of 2 . 15 Å using selenomethionine phasing . Structural analysis revealed an eight-stranded beta-barrel with a profile of short conserved regions consistent with a non-canonical lipocalin fold . Using a library of native and scrambled peptides representing the full Tp0751 sequence , we next identified a subset of peptides that showed statistically significant and dose-dependent interactions with the ECM components fibrinogen , fibronectin , collagen I , and collagen IV . Intriguingly , each ECM-interacting peptide mapped to the lipocalin domain . To assess the potential of these ECM-coordinating peptides to inhibit adhesion of bacteria to host cells , we engineered an adherence-deficient strain of the spirochete Borrelia burgdorferi to heterologously express Tp0751 . This engineered strain displayed Tp0751 on its surface and exhibited a Tp0751-dependent gain-of-function in adhering to human umbilical vein endothelial cells that was inhibited in the presence of one of the ECM-interacting peptides ( p10 ) . Overall , these data provide the first structural insight into the mechanisms of Tp0751-host interactions , which are dependent on the protein’s lipocalin fold .
Syphilis is a chronic , multistage disease caused by Treponema pallidum subsp . pallidum , with a global burden of 36 million cases and 11 million new infections per year [1] . Syphilis remains prevalent in resource-poor settings and the incidence rate is rising in Europe and Britain , the United States , Canada and China [2–7] . Congenital syphilis is the most common infection associated with fetal loss or stillbirth in low-income populations , with approximately 1 . 4 million pregnant women infected with active syphilis per year [8–10] . Symptomatic syphilis infections increase HIV transmission and acquisition 2- to 5-fold , and modeling studies predict that eradication of syphilis would have a significant impact on HIV prevention [11 , 12] . Elimination of syphilis as a risk factor for HIV can be achieved only through prevention of new syphilis cases , since the highest risk for transmitting and acquiring HIV coincides with early syphilis and frequently precedes diagnosis . The continuing high rates and global public health threat of syphilis , despite the effectiveness of penicillin treatment , highlights the need for enhanced understanding of the mechanisms of T . pallidum pathogenesis . The highly invasive nature of T . pallidum is reflected in its ability to cross the placental barrier ( congenital syphilis ) , to invade the central nervous system ( neurosyphilis ) , to cause a widespread rash ( characteristic of secondary syphilis ) , and to invade immunologically privileged sites such as the eye ( ocular syphilis ) [13 , 14] . Animal studies suggest dissemination via the bloodstream and lymphatics begins within hours of infection [15 , 16] , and early involvement of the liver and kidneys in patients implies that systemic dissemination is also an early event in humans [17 , 18] . Although the invasive capability of T . pallidum is crucial to the pathogenesis of this microorganism , the molecular mechanisms underlying dissemination are incompletely understood . This is due , in part , to the fact that only a limited number of T . pallidum proteins have been identified that could be directly involved in molecular interactions with the host . Our understanding of the mechanisms underlying T . pallidum pathogenesis , and of dissemination in particular , is also limited by the inability to genetically modify this pathogen , and the associated challenges of studying the roles of candidate virulence factors in pathogenesis . Heterologous expression of candidate T . pallidum virulence factors in other spirochetes , including Treponema phagedenis [19] and more recently the Lyme disease spirochete Borrelia burgdorferi [20] , is a crucial strategy for investigating the biological function of these factors . One protein suggested to play a role in T . pallidum dissemination within the host is Tp0751 ( also referred to as pallilysin ) . This protein is a primary target of opsonic antibodies , and thus is predicted to be surface-exposed on T . pallidum , and it binds and degrades host components encountered by T . pallidum during dissemination [21–25] . In particular , Tp0751 has a propensity to bind host molecules that are in close proximity to the vasculature , including ECM components found within the sub-endothelial matrix ( laminin ) , and associated with the glycocalyx on the apical surface of endothelial cells ( fibronectin and fibrinogen ) . Moreover , previous investigations using synthetic peptides identified the regions of Tp0751 that directly interact with laminin [22] . Although the functional studies performed to date indicate that Tp0751 interacts with multiple host components , the molecular basis for these interactions are unknown . A complicating factor is that the amino acid sequence of Tp0751 offers little insight into the molecular architecture and associated functions of the protein . Towards establishing the molecular mechanisms by which Tp0751 engages host components , we first determined the three dimensional structure of the mature region of Tp0751 encompassing residues Ser78 to Pro237 . Intriguingly , structural analysis revealed a beta-barrel fold displaying key hallmarks of a non-canonical lipocalin domain . Using a synthetic peptide library , we identified several peptides with the capacity to coordinate host ECM components and identified one of these peptides as possessing the ability to reduce adhesion of an engineered Tp0751-expressing B . burgdorferi strain to endothelial cells . Collectively , these data provide the first structural and mechanistic insight into Tp0751 interactions with host components .
Analysis of the Tp0751 sequence C-terminal to the signal peptide cleavage site ( Cys24 to Pro237 ) suggested a prolonged region of disorder extending to Pro98 , followed by a set of defined secondary structure elements encompassing Val99 to His228 ( Fig 1A ) . However , the lack of any significant sequence identity with known domains or structurally characterized proteins offered little insight into the architecture , and therefore function , of the C-terminal region of Tp0751 . To address this knowledge gap , we first generated constructs for recombinant protein production that extended from Ser78 ( putative thrombin cleavage site ) or Val99 ( beginning of region of predicted secondary structure ) to the C-terminus ( Tp0751_78 and Tp0751_99 , respectively; Fig 1A ) . We also mutated Glu199 to Ala to stabilize the protein for crystallization studies [23] . Recombinant proteins produced in E . coli were purified and showed expected elution patterns using size exclusion chromatography , with the similar elution profiles of Tp0751_78 WT and Tp0751_78 E199A ( Tp0751_78A ) indicating that the point mutation did not alter protein folding ( Fig 1B ) . We next crystallized and determined the three dimensional structure of Tp0751_78A using selenium single wavelength anomalous dispersion ( SAD ) phasing to a resolution of 2 . 15 Å ( Table 1 ) . Structural analysis revealed that the C-terminal domain of Tp0751_78A adopts a compact eight-stranded antiparallel beta-barrel with +1 topology , capped by a short 310-like helix and a longer N-terminal helix ( Fig 1C ) . The first ordered residue in the Tp0751_78A structure was Gln96 indicating the N-terminal residues Ser78 to Thr95 were either disordered or proteolyzed in the crystal . To investigate these two possibilities , crystals of Tp0751_78A , and the related Tp0751_99A construct for comparison , were isolated , washed and analyzed by SDS-PAGE . Analysis clearly revealed an intact N-terminal extension present in the Tp0751_78A crystals ( Fig 1D ) indicating that the lack of electron density for the N-terminal residues in the structure reflects disorder and not proteolysis . This confirms that the region of Tp0751 from Gln96 to Ala229 represents the core structural domain of the protein . In addition , an extended N-terminus on the Tp0751_78A construct is consistent with its faster elution from the size exclusion column compared to the globular protein standards ( Fig 1B ) . Comparison of the Tp0751_78A structure against the database of known structures using the DALI server [26] revealed a striking similarity to lipocalin domains , specifically nitrophorins . Tp0751_78A achieved a Z-score of 8 . 8 ( Z < 2 is spurious ) with the top hit Nitrophorin 4 ( PDB ID 1KOI ) , corresponding to a root mean square deviation of 3 . 0 Å over 184 aligned positions ( Fig 2A ) . Although Tp0751_78A and Nitrophorin 4 share only 6% sequence identity , low sequence identity is a common feature of lipocalins [27 , 28] . Lipocalins , along with fatty acid-binding proteins and avidins , are members of the calycin superfamily , which is defined by the distinct features of a central beta-barrel and a key structural signature consisting of three short conserved regions ( SCR1 , SCR2 , and SCR3 ) [27] . The designation of the Tp0751 structural domain as a lipocalin within the calycin superfamily is confirmed by the presence of eight antiparallel beta-strands with +1 topology that comprise the beta-barrel , combined with the elliptical shape of the barrel cross-section and readily distinguishable open and closed barrel ends ( Fig 2A ) . While Tp0751_78A lacks certain features common to lipocalins , such as N- and C-terminal regions that are pinned to the outside of the beta-barrel by disulfide bonds , these are not requirements for classification as a lipocalin ( Fig 2A , right ) . Ultimately , the classification of Tp0751 as an outlier lipocalin domain , as opposed to a kernel lipocalin , is based on the observation that is does not contain all three SCRs . This is also the basis for Nitrophorins being classified as outlier lipocalins [27] . The SCR1 in Tp0751 localizes to the 310 helix and Strand A , centered on a GxW motif ( Gly125 and Trp127 in Tp0751; Fig 2B ) [27] . Although SCR3 is distal in amino acid sequence , residing on Strand H , the key basic residue in this region ( Arg226 in Tp0751 ) stacks on top of the SCR1 Trp and forms a hydrogen bond to the backbone carbonyl of the preceding coil ( Fig 2B ) [27] . However , similar to other bacterial lipocalins [27] , the SCR2 , which is localized to the Strand F-Loop 6-Strand G region , is not conserved in Tp0751 . Notably , this region was predicted previously [24] to incorporate a metal binding motif , however , the structure of Tp0751_78A reported here reveals no clear mechanism for metal coordination . Lipocalins that are closely structurally related tend to have similar functions [28] . However , it is highly unlikely that Tp0751 shares a similar function with nitrophorins , as the residues required for heme coordination and transport in nitrophorins are not conserved in Tp0751 . More broadly , a common feature of lipocalins is the presence of a hydrophobic ligand-binding pocket within the beta-barrel [27 , 29] . This pocket and surrounding loops of the open end ( L1 , L3 , L5 , and L7 ) serve as a cup to coordinate an extensive variety of hydrophobic ligands for transport , catalytic , or sequestration purposes . In conventional lipocalins , Loop 1 serves as a lid for the cup-like binding site . Although Loop 1 of Tp0751_78A is exposed and partially disordered in the structure ( Fig 2C ) , there is no discernable hydrophobic pocket , primarily because Loop 7 caps the hydrophobic core with a stacked Trp-Met pair ( Fig 2C , right ) . While displacement of Loop 7 could expose a potential ligand-binding site , several other polar residues also cap the hydrophobic core ( Fig 2C , right ) , suggesting that the open end of the Tp0751 lipocalin domain lacks a hydrophobic binding pocket . To dissect the contributions of individual substructures in mediating attachment to host ECM components , we took advantage of a Tp0751 peptide library [22] . Of the 13 native Tp0751 peptides tested ( p1-p13 ) in our initial binding screen , p4 , p6 , and p11 displayed statistically significant binding to fibrinogen ( p≤0 . 0004 ) , fibronectin ( p≤0 . 0001 ) , collagen I ( p≤0 . 0113 ) , and collagen IV ( p≤0 . 0002 ) ( Fig 3A ) . Peptide p10 also showed significant binding to fibrinogen ( p<0 . 0001 ) , fibronectin ( p<0 . 0001 ) , and collagen IV ( p≤0 . 0006 ) , but not to collagen I ( p≤0 . 0653 ) . The overlapping peptides 3 , 5 and 7 exhibited little to no binding , consistent with a key role for the central four amino acids unique to p4 ( PVQT ) and p6 ( LWIQ ) in mediating interactions . Notably , scrambled versions of p4 ( p4scr ) and p6 ( p6scr ) showed no binding , yet a scrambled p10 ( p10scr1 ) showed enhanced binding relative to p10 ( Fig 3A ) . To further investigate the strength and specificity of these interactions , we performed dose-dependent binding assays . First we measured the apparent dissociation constants ( Kd ) of Tp0751_78A ( 2 . 0 ± 0 . 4 μM ) and Tp0751_78 WT ( 6 . 1 ± 1 . 7 μM ) to fibrinogen ( S1 Fig ) . The tighter binding observed with Tp0751_78A is likely due to its greater stability as observed previously [23] . Next we showed that the Tp0751 peptides p4 , p6 , p10 , and p11 exhibited dose-dependent binding to both fibrinogen and fibronectin with apparent dissociation constants ranging from 1 . 6 to 10 . 5 μM ( see Fig 3B legend for individual apparent Kd values ) . To further probe the previously observed enhanced binding with p10scr1 in the single point assay , we measured dose dependent binding and calculated corresponding apparent Kd values of 1 . 5 ± 0 . 2 μM and 1 . 3 ± 0 . 1 μM to fibrinogen and fibronectin , respectively ( Fig 3B ) . Analysis of p10 sequence revealed an arginine triplet framed by hydrophobic residues ( Fig 3C ) that remained largely intact in p10scr1 ( RxRxxR ) . To test the contribution of this Arg rich region , we generated a second scrambled version of p10 ( p10scr2 ) in which the arginine residues are more spatially separated . Intriguingly , p10scr2 exhibited weaker binding to fibrinogen ( apparent Kd = 11 ± 1 . 3 μM ) and fibronectin ( apparent Kd = 7 . 9 ± 0 . 7 μM ) compared to p10scr1 and p10 ( Fig 3B ) . Collectively , these data reveal a key , yet highly contextual role for the arginine motif in coordinating host proteins . We next mapped each of the ECM binding peptides onto the Tp0751 structure . Notably , each peptide that exhibited binding is contained within the lipocalin domain , highlighting the importance of this ordered region for interfacing with the host ( Fig 3C ) . Peptide 4 ( yellow ) maps to the N-terminal helix while p6 ( purple ) maps to the 310-like helix and a short strand ( Strand A labeled in Fig 1C ) and peptides p10 ( blue/cyan ) and p11 ( cyan/green ) map to Strands E/F and F/G , respectively . We next sought to investigate the capacity of wild-type Tp0751 to attach to endothelial cells in the biologically relevant context of a live spirochete . Due to the technical limitations associated with direct experimentation with T . pallidum , we conducted these studies by engineering the model spirochete , B . burgdorferi , to heterologously express Tp0751 as a surface-localized protein ( strain Bb-Tp0751 ) . To enable these investigations , we cloned the tp0751 open reading frame , including the Tp0751 lipoprotein localization signal sequence , in fusion with sequences encoding a C-terminal 3X-FLAG tag under the control of a constitutive B . burgdorferi promoter ( PflaB ) and inserted this construct into a cp32-based shuttle vector ( Fig 4A ) . A construct for constitutive expression of C-terminally FLAG-tagged BBK32 ( a B . burgdorferi adhesin [30 , 31] ) was also generated as a positive control . The resulting constructs were transformed into a non-infectious , BBK32-negative , adhesion-attenuated , GFP-expressing , B31-A-derived B . burgdorferi strain [30 , 31] . Positive clones were screened by PCR with adhesin-encoding inserts sequence verified , and antibiotic selected ( S2 Fig ) . Digital PCR also verified tp0751 transcripts were produced in the Bb-Tp0751-transformed strain ( S3 Fig ) . To determine if Tp0751 was expressed and surface-localized on the heterologous expression strain , we measured expression of the C-terminal FLAG tag of fusion proteins and a non-surface-localized control protein ( periplasmic flagellin B , FlaB ) by fluorescence-activated cell sorting ( FACS ) analysis using antibodies to the FLAG tag and FlaB ( Fig 4B–4D ) . FACS analysis was performed with equal numbers of methanol-permeabilized and mock saline-treated bacteria to distinguish between surface-localized and intracellular proteins . A representative histogram of mock saline-treated strains probed with anti-FLAG antibodies is shown in Fig 4B . Approximately 50% of bacteria transformed with plasmids encoding FLAG-tagged BBK32 and Tp0751 were FLAG-positive , with no significant differences among strains ( p = 0 . 94 ) , whereas 0% of parent strain bacteria were positive for the FLAG epitope ( Fig 4B ) . This indicated that detection of surface-localized proteins via their FLAG tag was specific . Quantification of FLAG tag and FlaB expression levels in methanol-permeabilized and mock-treated bacteria showed that FLAG fusion proteins were exclusively localized to bacterial surfaces ( Fig 4C: p≥0 . 66 mock vs permeabilized ) and that periplasmic FlaB was not readily detected without permeabilization ( Fig 4D ) . Thus , FACS analysis was performed under conditions that did not damage bacterial outer membranes and expose intracellular proteins . The BBK32- and Tp0751-expressing strains showed similar relative abundance of FLAG-tagged proteins and there was no difference in expression of FlaB ( Fig 4C and 4D: p≥0 . 7 for all comparisons ) . Therefore , Tp0751 was expressed and surface-localized on B . burgdorferi as efficiently as the B . burgdorferi lipoprotein BBK32 . Although a constitutive B . burgdorferi promoter was used to drive expression of FLAG fusion proteins and FACS analysis showed that 50% of bacteria expressed these proteins ( Fig 4B ) , we did not detect expression of Tp0751 by either immunoblotting or immunofluorescence . Although the reason for this discrepancy is unknown , we hypothesize that the capacity of FACS to facilitate analysis at the single-cell level , combined with the exquisite sensitivity of FACS [32] , accounts for the divergent results obtained between FACS and the other immunological techniques . With the knowledge that lipocalin-derived Tp0751-specific peptides bind host components , we next investigated the possibility that the Tp0751 ECM-binding peptides could modulate adhesion of our Tp0751-expressing B . burgdorferi ( strain Bb-Tp0751 ) to host cells ( Fig 5A ) . We compared the ability of Bb-Tp0751 to adhere to HUVECs ( Human Umbilical Vein Endothelial Cells ) relative to the non-adherent , non-transformed B . burgdorferi strain ( Parent ) , alone or individually preincubated with the Tp0751 synthetic peptides . While peptides p4 , p6 , and p11 did not alter adhesion of Bb-Tp0751 , incubation with peptide p10 significantly reduced binding of Bb-Tp0751 to HUVEC monolayers ( p≤0 . 005 ) compared to the binding levels exhibited following preincubation with the scrambled peptide , p10scr1 ( Fig 5B ) . A negative control peptide ( p8 ) that failed to bind the ECM components tested in the current study ( Fig 3A ) had no effect on HUVEC adherence by either B . burgdorferi strain . Furthermore , preincubation of HUVEC cells with increasing concentrations of p10 ( 0 nM– 545 nM ) resulted in dose-dependent and statistically significant lower levels of binding of Bb-Tp0751 to HUVEC monolayers compared to the levels of binding when preincubated with p8 ( ≥ 5 . 45 nM p10; p≤0 . 005 ) . These results show that p10 is uniquely capable of inhibiting binding of Tp0751-expressing B . burgdorferi to HUVEC monolayers .
The ability to disseminate throughout the host and extravasate from the vasculature is central to the pathogenesis of the spirochete bacterium T . pallidum . As with other invasive pathogens , these processes in T . pallidum are undoubtedly governed by surface displayed proteins that enable adhesion to host components . The identification and characterization of T . pallidum adhesins , however , has proven particularly challenging due to the intractability of the pathogen to in vitro culturing , genetic manipulation and the use of conventional experimental methodologies . As a result , very few T . pallidum adhesins ( Tp0136 [33 , 34] , Tp0155 [35] , Tp0483 [35] and Tp0751 [23 , 24] ) have been identified . Of these , we have focused on Tp0751 due to its role in modulating host ECM interactions and because its targeting by opsonic antibodies in intact T . pallidum [24] indicates that it is surface-localized in this pathogen . The structural , biochemical and functional analyses described herein reveal an intriguing profile for Tp0751 that relies on a compact , eight-stranded beta-barrel classified as a non-canonical or outlier lipocalin fold ( Fig 1 ) . The overall architecture of lipocalins supports diverse functional roles in both prokaryotes and eukaryotes , where the latter are localized to the extracellular milieu and implicated in binding cell surface receptors , regulation of cell homeostasis , and modulation of immune and inflammatory responses [36] . Eukaryotic lipocalins are also involved in modulating host cell signaling pathways that regulate cell motility and cell differentiation , and in this way have been reported to promote tumor metastasis [37–39] . In contrast , many of the earliest characterized prokaryotic lipocalins are localized to the inner leaflet of the outer membrane in Gram negative bacteria [40] . They rely on a central cavity to coordinate hydrophobic ligands , and function in outer membrane biogenesis and repair [40] and membrane adhesion [41] . More recently , however , an expanded functional repertoire for prokaryotic lipocalins has been recognized . These include the secreted lipocalin-containing protein Hp1286 from Helicobacter pylori that is involved in bacterial colonization and persistence in the stomach [41] , and the recent report of lipocalin YxeF in the Gram-positive bacterium Bacillus subtilis [42] . Most importantly , the lipocalin domain of the factor H binding protein ( fHbp ) from Neisseria meningitides has been shown to be surface-localized and lack a hydrophobic pocket [43–45] . Intriguingly , our structural investigations of Tp0751 also reveal the absence of a defined hydrophobic binding pocket ( Fig 2 ) . This feature , combined with the prior observation that Tp0751 binds host ECM components [23 , 24] and our observation reported here that Tp0751 is surface-localized when heterologously-expressed in the related spirochete B . burgdorferi , is consistent with a functional role for Tp0751 that centers on host-pathogen protein-protein interactions . To further dissect the capacity of Tp0751 to coordinate host partners , we employed a peptide library that spanned the entire post-signal peptide Tp0751 sequence and showed that host ECM component binding was entirely contained within the lipocalin domain ( Fig 3 ) . Moreover , and in similar fashion to N . meningitides fHbp [43–45] , much of the binding surface is localized to one face of the lipocalin beta-barrel and formed by peptides p10 and p11 ( Fig 3 ) . Intriguingly , the 10 amino acid overlapping region ( Fig 3—cyan ) between these peptides harbors an arginine triplet framed by hydrophobic residues . Consistent with the observation of an influential role for basic residues in host component attachment , it has been shown that the FbsA adhesin from Streptococcus agalactiae binds fibrinogen via a 16-amino acid motif containing RRxR/K and xxR/Kxx sequences [46] . Studies using site-directed mutagenesis and synthetic peptides have also shown positively charged residues to be important in mediating binding of different MSCRAMMs ( Microbial Surface Components Recognizing Adhesive Matrix Molecules ) to fibrinogen and fibronectin [47–49] . Furthermore , both arginine and lysine residues have been shown to comprise part of the fibronectin binding motif in BBK32 from B . burgdorferi [50] . While peptides 4 and 6 do not harbor a basic region , they also show significant binding to host ECM components . Thus , these data reveal a spatially and chemically diverse strategy employed by Tp0751 to coordinate host ECM partners that likely supports adhesion of T . pallidum to the vasculature and has the potential to serve as an immune-masking strategy [51] . To investigate the role of Tp0751 in mediating binding to the host endothelium , we engineered a B . burgdorferi strain to heterologously express and display Tp0751 on its surface ( Fig 4 ) . The expression of the tp0751 gene within B . burgdorferi was verified at both the transcript and protein levels using digital PCR and FACS analyses , respectively ( Figs 4 and S3 ) . The detection of Tp0751 expression in intact Borrelia , combined with the detection of FlaB solely in permeabilized Borrelia , verifies expression of Tp0751 on the borrelial surface , thus placing Tp0751 in a location that is directly exposed to the host environment . Thus , our studies clearly show that the native Tp0751 lipoprotein localization peptide is sufficient for surface localization on B . burgdorferi . Notably , this observation suggests the existence of a conserved lipoprotein export system among diverse spirochetes and provides further evidence to suggest that Tp0751 , ( and potentially other lipoproteins ) , is a surface-localized lipoprotein in T . pallidum . With this powerful new model system , we then showed that p10 was uniquely capable of competitively inhibiting binding of the Tp0751-expressing B . burgdorferi to HUVEC monolayers ( Fig 5 ) in a dose-dependent manner . Based on the important role of p10 we further investigated the structure of this peptide in the context of the lipocalin fold . Specifically , the 24mer sequence can be broken into four distinct sub-structures: L4 ( RK ) , Strand E ( TVSFLTRN ) , L5 ( TAISS ) , and Strand F ( IRRRLEVTF ) . In the context of the lipocalin fold , nearly all the hydrophobic residues of p10 are buried in the protein core , while the six basic residues are clustered predominately in two basic patches separated at the closed end ( RK on L4 ) and near the open end ( RRR on Strands E and F ) of the lipocalin . However , the interaction with HUVECs appears to be not solely dependent on the arginine-rich region found in p10 , since both p11 and the scrambled p10 peptides contained this positively charged region yet did not significantly inhibit binding . Thus , it appears that the capacity of Tp0751 to engage molecular partners on endothelial cells is mediated by a different , more narrowly defined mechanism , compared with those used in ECM coordination . The lack of a central hydrophobic pocket in Tp0751 suggests a protein-protein interaction function mediated by the extended surfaces of the compressed beta-barrel . Dissecting Tp0751 using peptide libraries and host cell component binding assays ultimately validated this hypothesis . To place these data in the context of T . pallidum dissemination and invasion , we propose a three stage model , with two stages for binding to host components , followed by a final stage of extravasation ( Fig 6 ) . In the first stage , the surface-exposed Tp0751 leverages its promiscuous binding surface ( p4 , 6 , 10 and 11 ) to engage host ECM components as an effective mechanism of slowing bacterial movement through the vasculature . In the second stage , once the bacterium has slowed , Tp0751 engages specific host endothelial receptors using a more narrowly focussed molecular interaction strategy mediated solely by the region represented by p10 . Such a multi-stage binding mechanism has been demonstrated to occur between B . burgdorferi and endothelial cells under conditions of flow , with BBK32 mediating initial interactions with ECM components followed by more specific interactions with endothelial surfaces [30] . We propose that a similar binding mechanism plays a key role in treponemal virulence , and that disrupting these interactions has the potential to limit infectivity of this invasive pathogen . Thus Tp0751 , much like the N . meningitides fHBP surface-displayed lipocalin , which is one of three protein antigens included in the highly successful 4CMenB vaccine [52 , 53] , may be an ideal candidate for targeted vaccine development . Collectively , our findings provide the first detailed structural and mechanistic insight into the molecular cross talk between T . pallidum and host cells and offer intriguing potential for developing Tp0751-based measures to control dissemination and , ultimately , pathogenesis of the syphilis spirochete .
Constructs encoding N-terminally truncated forms of Tp0751 ( Ser78 to Pro237 , Tp0751_78; Val99 to Pro237 , Tp0751_99 ) with an E199A mutation as previously described [24] were cloned into a modified pET28a vector with a TEV protease cleavable N-terminal hexa-histidine tag . Constructs were produced in E . coli BL21 and purified from the soluble fraction by Ni-affinity chromatography . Tp0751_78A and Tp0751_99A were cleaved with TEV protease , purified further by size exclusion and cation exchange chromatography , and exchanged into a final buffer of 20 mM HEPES pH 8 . 0 , 150 mM NaCl , 1% glycerol , and 5 μM zinc chloride . Selenomethionine ( SeMet ) -labelled Tp0751_78A protein was produced using E . coli 834 cells in SeMet media ( Molecular Dimensions ) . The culture was grown at 37°C to an A600 of 1 . 2 , cooled to 16°C and then induced at an A600 of 1 . 8 with 0 . 4 mM isopropyl 1-thio-β-d-galactopyranoside . After 18 h of growth at 16°C , the cells were harvested by centrifugation , and the SeMet-labeled protein was purified using the same protocol as for the native protein . Crystals of Tp0751_78A were originally identified in the SaltRx screen ( Hampton Research ) using sitting drops at 18°C . The final , refined drops consisted of 0 . 8 μL Tp0751_78A at 15 mg/mL with 0 . 8 μL of reservoir solution ( 0 . 1 M sodium acetate pH 5 . 4 , 1 . 6 M sodium formate ) and were equilibrated against 120 μL of reservoir solution . Crystals appeared within 5 days and grew to a final size within 3 weeks . For crystallization of SeMet-derivatized protein , 1 . 0 μL of Tp0751_78A-SeMet at 15 mg/mL was mixed with 1 . 0 μL of reservoir solution ( 0 . 1 M sodium acetate pH 3 . 6 , 1 . 6 M sodium formate ) and equilibrated against 120 μL of reservoir solution . Crystals were cryoprotected in paratone and flash cooled in liquid nitrogen . Diffraction data were collected on beamline 08ID-1 at Canadian Light Source using a wavelength of 0 . 984 Å for the native data , or an optimized wavelength of 0 . 9794 Å for the f” selenium edge . Diffraction data were processed to 2 . 15 Å ( Tp0751_78A ) or 2 . 80 Å ( Tp0751_78A-SeMet ) resolution using Imosflm [54] and Scala [55] or Aimless [56] in the CCP4 suite of programs [57] . The structure of Tp0751_78A-SeMet was solved by Selenium single wavelength anomalous dispersion . A total of 6 high confidence Se sites were identified using Phenix . hyss , and enabled building and registering of approximately 70% of the backbone using Phenix . autosol followed by Phenix . autobuild [58] . Native data were twinned; while Phenix . xtriage [58] and Aimless [56] identified P622 as the lattice point group , Zanuda in CCP4 [57] identified P321 as the most likely space group . The Tp0751_78A native structure was solved by molecular replacement using a single Tp0751_78A chain from the Se-phased model in Phaser [59] , which identified P3121 as the optimal space group . COOT [60] was used for model building and selection of solvent atoms , and the model was refined in Phenix . refine [61] using a twin fraction of 0 . 43 and the merohedral twin operator -H , -K , L . The structure of Tp0751_78A has 9 copies in the asymmetric unit; each chain overlays on chain A with an rmsd of 0 . 30 to 0 . 60 Å over 115 to 127 aligned Cα positions . Chain A was the most completely modeled with the lowest thermal motion parameters and was used for all analyses and figures . Complete structural validation was performed with Molprobity [62] , including analysis of the Ramachandran plots , which showed greater than 93% of residues in the most favored conformations . Five percent of reflections were set aside for calculation of Rfree . Data collection and refinement statistics are presented in Table 1 . The atomic coordinates and structure factors for Tp0751_78A have been deposited in the Protein Data Bank under the following PDB ID: 5JK2 . Thirteen overlapping 24-mer peptides ( p1-p13 ) that spanned the Tp0751 sequence from T46-P237 were synthesized as described previously [22] . Each peptide shared 10 overlapping amino acids with neighboring upstream and downstream peptides . Scrambled versions of peptides 4 ( p4scr ) , 6 ( p6scr ) , 10 ( p10scr1 ) , and a second scrambled version of peptide 10 ( p10scr2—VAREFNKSRTILFRTSVTRLSTRI ) were prepared as described previously [22] . All synthetic peptides contained N-terminal hexa-histidine tags to allow for detection . Plasminogen-depleted human fibrinogen ( Calbiochem ) was purchased from VWR International . Laminin isolated from Engelbreth-Holm-Swarm murine sarcoma basement membrane and fibronectin isolated from human plasma were purchased from Sigma-Aldrich Canada Ltd . ( Oakville , ON ) . Human collagen types I and IV ( Rockland Immunochemicals , Inc . ) were purchased from VWR International . To test for binding of synthetic Tp0751 peptides p1-p13 , p1scr , p6scr , p10scr1 , and p10scr2 to the host proteins fibrinogen , fibronectin , laminin , collagen type I , and collagen type IV , initial binding assays were performed as described previously [21 , 22] . For dose-dependent binding assays , recombinant proteins were serially diluted 1:2 from either 10 μM to 0 . 156 μm ( Tp0751_78A used for crystallization and Tp0751_78 WT ) or 40 μm to 0 . 625 μm ( peptides p4 , p6 , p10 , p11 , p10scr1 , and p10scr2 ) . Average absorbance readings ( 600 nm ) from three wells are presented with bars indicating standard error and the results are representative of two independent experiments . Plates were read at 600 nm with a BioTek enzyme-linked immunosorbent assay plate reader ( Fisher Scientific , Ottawa , ON ) . All statistical analyses were performed using the Student’s two-tailed t-test . Non-linear regression curves were fitted and apparent dissociation constants ( Kd ) were calculated using GraphPad Prism data analysis software ( San Diego , CA ) . All bacterial strain details are provided in S1 Table . Construction and characterization of GFP-expressing the non-infectious B31-A-derived parent ( GCB706 ) strain was previously reported [63] . Construction and characterization of the GCB706-derived BBK32-3XFLAG- ( TMB103 ) and Tp0751-3XFLAG-expressing ( designated TMB49 or Bb-Tp0751 ) strains are described below and in S2 Table . B . burgdorferi was cultivated as reported [63] in Barbour-Stoenner-Kelly-II ( BSK-II ) medium supplemented with 100 μg/ml gentamicin and/or 200 μg/ml kanamycin for plasmid selection . All primers and templates used for construct cloning , E . coli construct names and strain numbers , and B . burgdorferi strain names are described in S1–S3 Tables . Constructs for expression of C-terminally 3XFLAG-tagged BBK32 ( pTM259 ) and Tp0751 ( pCC_3–1 ) were assembled by overlap extension PCR , cloned into pJET1 . 2 by blunt ligation , sequenced using pJet1 . 2 forward and reverse primers ( Fermentas , Burlington , ON ) , excised and cloned into XhoI/NotI sites of cp32-derived shuttle vector pCE320 as described [64] , followed by sequencing with primers B1723 and B1724 . E . coli strain TOP10 ( Life Technologies , Burlington , ON ) was used for cloning Tp0751 expression cassettes into pJET1 . 2 . E . coli DH5α was used for all other cloning steps and preparation of plasmid for B . burgdorferi transformations . Plasmids were prepared using Qiagen maxiprep kits ( Qiagen , Toronto , ON ) . Electrocompetent GCB706 B . burgdorferi were prepared and transformed with 50 μg pTM259 or pCC_3–1 as described [30] , followed by cloning by limiting dilution and selection in kanamycin- and gentamicin-supplemented medium . Clones were PCR-screened using primers for kanamycin and gentamicin-resistance cassettes ( B70 , B71 , B348 , B349 ) , as well as BBK32 and Tp0751 expression cassettes ( B1723 , B1724 ) . Screening results for Tp0751-expressing B . burgdorferi clones are shown in S2 Fig . BBK32- and Tp0751-expressing B . burgdorferi clones used in subsequent experiments were TMB103 and TMB49 . The sequence of the Tp0751 expression cassette in TMB49 was confirmed by sequencing of DNA isolated from this strain . No notable differences in bacterial morphology , motility or length were observed for any of the B . burgdorferi strains generated ( S5 Table ) , suggesting that the presence of the tp0751 coding sequence did not markedly affect B . burgdorferi viability . The Bb-Tp0751 strain was grown to logarithmic phase ( 5 x 107 cells/ml ) in 15 ml of BSK-II . Cells were spun down at 4 , 000 x g/15 min at 4°C , and frozen at -20°C to enhance cell lysis . Genomic DNA was isolated using PureLink Genomic DNA Mini Kit ( Invitrogen , Burlington , ON , CA ) according to manufacturer’s protocol for DNA isolation from Gram-negative bacteria with the following modifications . Thawed cells were resuspended in 180 μl of Genomic digestion buffer supplemented with 20μl of proteinase K , and incubated for 2 hours at 55°C with slow shaking . Genomic DNA was eluted using 50 μl of PCR grade water pre-warmed to 55°C . One microliter of isolated Bb-Tp0751 gDNA was used as a template for PCR . The reaction mixture contained 1 μL of 10 mM dNTPs ( ThermoFisher Scientific , Toronto , ON ) , 1 μL of 25 μM F_Xho_Pfla_short , 1μL of 25μM Bb-751_int_R , 10 μL 5x Phusion HF buffer , 0 . 5 μL Phusion polymerase ( both New England Biolabs ( NEB ) , Whitby , ON ) , 1 . 5 μl of 100% DMSO and PCR grade water in 50 μL volume . PCR conditions were set as follows: 98°C/3min , 35 cycles ( 98°C/15 s , 58°C/30 s and 72°C/1 min ) and 72°C/7 min . PCR products were purified using Qiaquick kit ( QIAgen , Montréal , QC ) according to manufacturer’s instructions , and eluted using 30 μL of provided elution buffer ( pre-warmed to 55°C ) . Amplicons were Sanger sequenced using amplification primers at The Center for Applied Genomics at The Hospital for Sick Children ( Toronto , ON ) . Each sequencing reaction contained 50–100 ng of PCR product and 7 pmol of the amplification primer ( F_Xho_Pfla_short or Bb-751_int_R ) . Biological triplicates and duplicates of the Bb-Tp0751 and the parent strain GCB706 , respectively , were grown to logarithmic phase ( 5 x 107 cells/ml ) in 15 ml of BSK-II . Cells were harvested at 4 , 000 x g/15 min at 4°C , resuspended in 1 ml of Trizol reagent ( Life Technologies ) , and placed at -80°C until processing . RNA was isolated according to the manufacturer’s instructions under RNase free conditions , and eluted using 30 μl of Ultrapure DNase/RNase-free water ( Life Technologies ) . Average RNA yield was 805 ng/μl and 465 ng/μl for Bb-Tp0751 and the parent strain , respectively . Ten micrograms of RNA were treated with DNase using the Turbo DNA-free kit ( Life Technologies ) according to the manufacturers’ instructions . Briefly , the reaction was set up in a total volume of 50 μL with 1 μl of DNase . Samples were incubated at 37°C for 30 min , an additional 1 μl of DNase was added , and the incubation was continued for another 30 min at 37°C . DNase was inactivated using reagents provided in the kit . Integrity of RNA after DNase treatment was verified on an agarose gel . After DNase treatment , average RNA concentrations were 127 . 7 ng/μl ( A260/280: 2 . 03–2 . 08 ) and 161 . 7 ng/μl ( A260/280: 2 . 02–2 . 05 ) for the Bb-Tp0751 and the parent strain , respectively , as measured using a Nanodrop . Five microliters of isolated RNA were used as a template for reverse transcription PCR ( RT-PCR ) , which was performed using the iSCRIPT cDNA synthesis kit ( BioRad , Mississauga , ON ) according to the manufacturer’s protocol . The Bb-Tp0751 RNA samples were pooled , and 5 μl were used as a template for no reverse transcriptase control ( NRT ) reaction . Three microliters of the RT-PCR sample was used as a template for digital PCR ( 10 μl of QX200 ddPCR EvaGreen Supermix ( Biorad ) , 0 . 5 μl of 10 mM qTp0751FI , 0 . 5 μl of 10 mM qTp0751RI and PCR grade water in a 20 μl volume ) . Droplets were generated on the QX200 Droplet Generator ( BioRad ) and transferred to a 96-well plate for amplification . PCR conditions were set as follows with ramp rate 2°C/s: 95°C/5 min , 40 cycles ( 95°C/30 s . 58°C/1 min 30 s ) , 4°C/5 min , 90°C/5 min , and 4°C/5 min . Fluorescence measurements were performed using QX200 Droplet Reader ( BioRad ) . Results were normalized to 100 ng of mRNA used for RT-PCR ( S4 Table ) . Flow cytometry to measure expression and surface-localization of 3 XFLAG-tagged adhesins was performed as described with 1 x 108 B . burgdorferi , using 1:250 dilutions of mouse anti-Flag-M2 , mouse anti-FlaB and Alexa635-conjugated goat anti-mouse antibodies ( Sigma , Invitrogen ) [65] . After fixation in formalin , FACS analysis of equal numbers of bacteria was performed within 2 days of sample preparation using a Becton Dickinson FACSCalibur flow cytometer equipped with a 15 mW 488 nm argon laser , standard three-color filters , and CELLQuest Software ( BD Bioscience , Franklin Lakes , NJ ) . Mean fluorescence intensities ( MFI ) of spirochetes obtained by analysis using FlowJo software ( Three star Inc , Ashland , OR ) were used to calculate relative levels of protein production in different strains , which were all normalized to values for the FLAG-tagged BBK32-expressing strain . Spirochetes with MFIs of less than 10 were considered negative for FLAG-tagged proteins . Human umbilical vein endothelial cells ( HUVECs ) pooled from multiple donors and purchased from Lonza ( Allendale , NJ ) were cultured in endothelial growth medium-2 ( EGM-2 ) ( Lonza ) at 37°C in an atmosphere of 5% CO2 , as per manufacturer’s instructions . B . burgdorferi strains were cultivated at 36°C in an atmosphere of 1 . 5% CO2 in BSK-II medium [66] prepared in-house with appropriate antibiotics ( 100 μg/mL gentamycin , 200 μg/mL kanamycin ) . Stationary adhesion assays were adapted from Szczepanski and colleagues [67] and Cameron and colleagues [19] . HUVECs ( passage 2–3 ) were seeded in 4-well chamber slides ( Nalge Nunc International , Rochester , NY ) , coated with 500 μg/mL phenol red-free matrigel ( Corning , Tewksbury , MA ) and grown 20 h at 37°C in 5% CO2 to form confluent monolayers . Synthetic Tp0751 peptides ( p4 , p6 , p10 , p11 , p8 , p4scr , p6scr , and p10scr1 ) were diluted to 500 μg/ml in HEPES buffered saline solution ( Lonza ) and HUVECs were pre-incubated with 0 . 1 μg of Tp0751 peptide per well for 3 h at 37°C in 5% CO2 . Peptide solutions were removed from HUVECs prior to the addition of bacteria . Borrelia burgdorferi strains were cultured in biological triplicate , and two days prior to the experiment , B . burgdorferi cultures were passaged to 6x105 cells/ml and grown for 48 h to reach a concentration of 2x107 cells/ml . Borrelia burgdorferi cultures were then centrifuged and resuspended in a 3:1 mixture of BSK-II:EGM-2 and 1 . 4x107 cells of each biological replicate were added to HUVECs in duplicate wells . Chamber slides were incubated for 12 h at 36°C in 1 . 5% CO2 , washed three times with warm HEPES buffered saline to remove non-adherent bacteria , and fixed in buffered 10% formalin ( Fischer Scientific , Ottawa , ON ) . To determine the effect of increasing concentrations of p10 and a negative control peptide ( p8 ) on adherence of Parent and Bb-Tp0751 to HUVECs , cell binding assays were performed as above with cells preincubated with increasing concentrations of peptide ( 0 nM , 0 . 54 nM , 5 . 45 nM , 54 . 5 nM , and 545 nM ) . Quantitation of B . burgdorferi adhesion to HUVECs was performed by counting GFP-expressing bacteria in ten fields of view ( FOV ) from duplicate wells for each biological replicate under 400X magnification on a Nikon 80i fluorescence microscope ( Meridian Instrument Company , Inc . , Kent , WA ) fitted with a monochrome digital camera , a dark-field condenser , and fluorescein filter ( Excelitas Technologies , Mississauga , ON ) . Statistical analyses were performed using the Student’s two-tailed t-test .
|
The Treponema pallidum protein , Tp0751 , possesses adhesive properties and has been previously reported to mediate attachment to the host extracellular matrix components laminin , fibronectin , and fibrinogen . Herein we demonstrate that Tp0751 adopts an eight-stranded beta barrel-containing lipocalin structure , and using a peptide library approach we show that the extracellular matrix component adhesive functionality of Tp0751 is localized to the lipocalin domain . Further , using a heterologous expression system we demonstrate that Tp0751 mediates attachment to endothelial cells , and that this interaction is specifically inhibited by a peptide derived from the Tp0751 lipocalin domain . Through these studies we have delineated the regions of the Tp0751 protein that mediate interaction with host extracellular matrix components and endothelial cells . These findings enhance our understanding of the role of this protein in treponemal dissemination via the bloodstream and provide defined regions of the Tp0751 protein that can be targeted to disrupt the treponemal-host interaction .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biotechnology",
"synthetic",
"biotechnology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"fibrinogen",
"engineering",
"and",
"technology",
"pathogens",
"synthetic",
"biology",
"microbiology",
"peptide",
"libraries",
"molecular",
"biology",
"techniques",
"glycoproteins",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"borrelia",
"burgdorferi",
"artificial",
"gene",
"amplification",
"and",
"extension",
"medical",
"microbiology",
"peptide",
"mapping",
"microbial",
"pathogens",
"proteomics",
"borrelia",
"pathogenesis",
"molecular",
"biology",
"biochemistry",
"treponema",
"pallidum",
"peptides",
"host-pathogen",
"interactions",
"polymerase",
"chain",
"reaction",
"synthetic",
"peptides",
"biology",
"and",
"life",
"sciences",
"glycobiology",
"organisms"
] |
2016
|
The Structure of Treponema pallidum Tp0751 (Pallilysin) Reveals a Non-canonical Lipocalin Fold That Mediates Adhesion to Extracellular Matrix Components and Interactions with Host Cells
|
Spike timing dependent plasticity ( STDP ) has been observed experimentally in vitro and is a widely studied neural algorithm for synaptic modification . While the functional role of STDP has been investigated extensively , the effect of rhythms on the precise timing of STDP has not been characterized as well . We use a simplified biophysical model of a cortical network that generates pyramidal interneuronal gamma rhythms ( PING ) . Plasticity via STDP is investigated at the excitatory pyramidal cell synapse from a gamma frequency ( 30–90 Hz ) input independent of the network gamma rhythm . The input may represent a corticocortical or an information-specific thalamocortical connection . This synapse is mediated by N-methyl-D-aspartate receptor mediated ( NMDAR ) currents . For distinct network and input frequencies , the model shows robust frequency regimes of potentiation and depression , providing a mechanism by which responses to certain inputs can potentiate while responses to other inputs depress . For potentiating regimes , the model suggests an optimal amount and duration of plasticity that can occur , which depends on the time course for the decay of the postsynaptic NMDAR current . Prolonging the duration of the input beyond this optimal time results in depression . Inserting pauses in the input can increase the total potentiation . The optimal pause length corresponds to the decay time of the NMDAR current . Thus , STDP in this model provides a mechanism for potentiation and depression depending on input frequency and suggests that the slow NMDAR current decay helps to regulate the optimal amplitude and duration of the plasticity . The optimal pause length is comparable to the time scale of the negative phase of a modulatory theta rhythm , which may pause gamma rhythm spiking . Our pause results may suggest a novel role for this theta rhythm in plasticity . Finally , we discuss our results in the context of auditory thalamocortical plasticity .
In many systems , synaptic plasticity is believed to depend on the spike timing of the pre- and postsynaptic cells [1] , [2] . This spike timing dependent plasticity ( STDP ) has been observed in vivo [3] and has been studied in many in vitro preparations in the context of learning and sensory processing [1] , [2] , [4] . Though there have been many modeling papers on STDP and learning [5]–[8] , the effect of rhythms on STDP has not been explored extensively , despite evidence that rhythmic activity is physiologically relevant . For example , a significant gamma rhythm ( 30–90 Hz ) is present in cortical areas during attention to learning-related tasks [9] and is associated with the specificity of learning in auditory areas [10] , [11] and with plasticity in vitro [12] . Spatially distinct gamma rhythms around 35 Hz are observed in granular Layer IV ( LIV ) and supragranular Layer II/III ( LII/III ) auditory cortical layers of monkeys that are awake but not involved in a task [13] . In this paper , we are concerned with an input that is periodic in the gamma frequency range , directed to an oscillatory network that produces an independent gamma frequency . The input is considered as an information encoding input , while the network is considered as a cortical network model in which gamma rhythms are modulated by non-specific drive ( See Discussion ) . With temporal structure in the input and in the cortical network itself , the effects of even a simple STDP rule become much less transparent . If the input did not perturb the target output , the dynamically changing phase lags would be determined completely by the periods of the input and cortical network oscillation . When this is not true , as in the present work , the phase lags can change in a way that seems random . However , statistically there is a bias , and when the STDP effects of the different lags are averaged over some time window , cumulative changes in potentiation are seen . The frequency of the input influences the output network , and the phase difference between an input spike and a nearby output spike depends on the dynamics of the model system . While N-methyl-D-aspartate receptor mediated ( NMDAR ) currents are known to be necessary in mediating many synapses affected by STDP [1] , [2] , [14] , the precise effect on spike timing is less clear . We show that the slower decay kinetics of NMDAR currents in the excitatory cortical cell model affect the cycle of firing and increase the amplitude and time course of potentiation at that synapse . The plastic changes can be either potentiating or depressing , depending on the frequencies of the input and that of the receiving network . Indeed , we show that there are interspersed bands of input frequency over which there is potentiation or depression . Furthermore , a persistent stimulus can first lead to potentiation and then switch to depression . We explore the robustness of this system in detail . Finally , we discuss a possible implication for our model results in the context of gamma rhythms nested in theta rhythms , and we discuss how the gamma frequency input in the model may represent an information specific corticocortical or thalamocortical input to a primary sensory cortical network with an independent gamma frequency .
The simplest case of a single gamma frequency input to the cortical network was considered initially . The model showed that broad bands of potentiation and depression existed for different frequency regimes that were periodic in the gamma frequency range ( Fig . 2 ) . Beta frequency ( 20–25 Hz ) inputs exhibited bands analogous to the low gamma frequency region with the frequency ratio less than 1 . For a constant applied drive to the E cell model , which set each , both potentiation and depression were possible , depending on the specific . This suggested that the STDP mechanism naturally favored potentiating the responses of certain network frequencies to particular input frequencies . The inputs to the oscillator represented only two of several currents that affect the voltage dynamics of the network , and the resultant period of the network oscillation does not generally follow that of the gamma frequency input . With broad bands of potentiation , a range of input potentiated the E cell synapse being driven independently by . Narrower or absent bands of potentiation would have signified that a very precise input frequency would be required to potentiate the glutamatergic E cell synapse . This qualitative behavior also occurred with a constant NMDAR current having no decay ( Data not shown ) , indicating that the global behavior of the model was more dependent on the frequency ratio than on the dynamics of the NMDAR current model , which played a more crucial role in regulating the specifics of the potentiation ( See below ) . Unlike the rhythmic gamma frequency inputs , Poisson distributed stochastic inputs did not lead to broad bands of potentiation ( Fig . 3 ) . However , when directed at the gamma generating network , theta frequency ( 4–12 Hz ) input also did not exhibit these bands ( Data not shown ) , though separate simulations , in which a pause was inserted into the input spiking ( See below ) , may be thought of in terms of theta frequency modulation ( See Discussion ) . In potentiating frequency regimes , the rise of potentiation was generally followed by the subsequent decline . Therefore , each frequency regime had a maximal level of potentiation and a finite time over which potentiation lasted . Of the 483 potentiating frequency regimes shown in Fig . 2 , 445 displayed this transition into depression within the 500 ms run . All 483 frequency regimes displayed the transition within 2000 ms . Potentiation occurred in the model when input spikes preceded E cell spikes and was strongest when the spike time difference was much less than the time constant for potentiation ( Fig . 4A ) . Strong depression occurred when the E cell spiked before the presynaptic input ( Fig . 4B ) . In a typical example of potentiation ( Fig . 5 ) , spike rasters confirmed positive spike pairs , in which the presynaptic input spike preceded the postsynaptic cortical E cell spike ( Fig . 5A ) . The increase in was also transiently observed ( Fig . 5B ) . The model NMDAR current decayed more slowly than the model AMPAR current , demonstrated by the open probability kinetics ( Fig . 5C and 5D ) . The slow decay of the NMDAR current dynamics had a prolonged effect on the cycle of E cell spike times , which affected the plasticity . Though a constant instead of decaying activation of the NMDAR current model did not qualitatively change the frequency map , NMDAR currents in the model were a natural candidate for mediating the transition from potentiation to depression that occurred over time scales around 100 ms . Increases in increased the average amplitude and duration of potentiation for a potentiating frequency regime . In the common frequency regimes that showed potentiation for a control value of 80 ms and an increased of 130 ms , there were increases in the amplitude of potentiation with the longer ( Fig . 6 ) , with a mean increase of 54% . Additionally , the time over which potentiation occurred was on average 2 . 9 times longer with the increased . Decreases in to 50 ms resulted in frequency regimes that continued to show potentiation compared to the control , but on average the amount of potentiation was 22% lower than the control ( Data not shown ) . The decay of the NMDAR current can account for the amplitude increase by prolonging E cell spiking activity . When an input spike immediately preceded an E cell spike , the STDP rule increased the AMPAR conductance , and the event-related NMDAR current was initiated . This raised conductance caused the next cycle of the E cell to fire faster , changing the rhythm of the network oscillator . However , as the NMDAR current decayed , the network rhythm was slowly pulled out of a potentiating frequency regime . Due to the prolonged availability of slowly decaying NMDAR currents , a potentiating network rhythm could conceivably be prolonged by allowing for another positive spike timing event to occur . However , once the NMDAR current was extinguished completely , and the spiking rhythm fell out of a potentiating regime , only a strong positive spike timing event could have prolonged or re-initiated the potentiation . The long time course of the model NMDAR current may be an advantage of the system by allowing multiple gamma period spike timing events to occur during the NMDAR current decay . This sets a natural window for greater potentiation to occur , while also limiting the possibility for unlimited spiking pairing to occur after the initial potentiating event . While transient potentiation was observed in nearly all simulations run with the standard of 80 ms , monotonically increasing potentiation was observed in some simulations with a constant NMDAR current that exhibited no decay ( Fig . 7 ) . The lack of NMDAR current decay dynamics eliminated the gradual change of the E cell spiking rate that is typically present . Since the rate of E cell spiking did not change dramatically without the NMDAR current decay , positive spike pairs were observed on each cycle of E cell spiking , contributing to a monotonic rise in potentiation . Temporarily suspending presynaptic input prolonged potentiation in the time period after the pause . In all frequency regimes in which potentiation was observed without pauses , greater potentiation was observed after insertion of certain pause conditions , which were identified by start time and duration . Almost any length of pause resulted in post-pause potentiation , as long as spiking resumed before the NMDAR current was completely extinguished ( approx . 80–130 ms ) . The model suggests that there was an optimal pause duration that resulted in maximal post-pause potentiation , which was dependent upon the specifics of the frequency regime . For all frequency regimes of potentiation , several positive spike pairs prolonged the potentiation . However , the increased activation drove the E cell spikes to precede the input spikes , and the potentiation turned to depression , as previously described . Further potentiation was observed ( Fig . 8B ) compared to the unpaused condition ( Fig . 8A ) when the pause in input spiking was 125 ms , less than . The AMPAR conductance was constant for the duration of the pause since no plasticity can occur , but once the input spiking was resumed , further potentiation occurred . When the input spiking was paused , starting at a time far after the peak of potentiation , depression persisted when the spiking resumed . This occurred because the NMDAR current had decayed completely , and no other positive spike pairs were observed ( Fig . 8C ) . In simulations with multiple inputs , the maximum contribution from each individual input was scaled inversely to the number of inputs . In this condition , coherence of inputs was necessary for strong potentiation . However , the simulations also showed that when inputs were slightly perturbed , potentiation was greater in some frequency regimes compared to fully coherent cases . For three inputs in a potentiating frequency regime , high coherence inputs ( C = 0 . 99 , Fig . 9A ) exhibited greater potentiation than lower coherence inputs ( C = 0 . 10 , Fig . 9B ) . The applied current , and therefore , was identical for both cases . Because the inputs in the low coherence case arrived at drastically different times with respect to the E cell spikes , their individual effects were not strong enough to potentiate the E cell appreciably , as in the high coherence case . However , the difference between completely coherent inputs ( C = 1 , Fig . 10A ) and slightly incoherent inputs ( C = 0 . 99 , Fig . 10B ) exhibited the opposite behavior . For a frequency regime that showed potentiation when the inputs were completely coherent , slight decreases in coherence increased the potentiation . This was due to the increased robustness of the input spiking: with completely coherent inputs , as soon as the E cell spike preceded all of the input spikes , the entire system tended toward depression , as all of the spike pairs contributed to depression . However , less coherent inputs allowed for some spike pairs to contribute to potentiation while others contributed to depression . This small perturbation or noise in the input spiking may have allowed the potentiation to persist , since it did not require all of the inputs to spike before the E cell . Greater decreases in coherence , however , still led to depression , since far fewer positive spike pairs contributed strongly to an overall effect of potentiation . Changes in from 1 . 2 ms to 2 . 1 ms resulted in higher maximal potentiation and prolonged potentiation in some frequency regimes ( Fig . 11 ) . This result persisted when was modified to affect only AMPAR currents , only NMDAR currents , or both simultaneously . However , no gross qualitative changes were seen in the frequency map due to this change ( Data not shown ) . In contrast , more dramatic changes to from 1 . 2 ms to 5 . 0 ms disturbed the frequency map , and the bands of potentiation and depression became less distinct ( Fig . 12 ) . At certain , from 40–45 Hz , very little depression was seen irrespective of , which suggested poor discrimination to the input frequency . The longer enabled more activation of the AMPAR and NMDAR currents in the model , since pre- and postsynaptic spikes could be paired within a longer interval . Persistent activation of glutamatergic conductances allowed for a greater contribution to the E cell's firing . In contrast , a shorter elicited fewer AMPAR and NMDAR currents , which enabled a faster frequency regime transition from potentiation to depression . To our knowledge , it is not known if glutamate is ever available for 5 . 0 ms endogenously , though bath application of glutamatergic agonists like kainate are common in slice preparations . The output of the model depended heavily upon the STDP rule given by Eqn . ( 3 ) . Both the temporal and amplitude asymmetries of the STDP rule were necessary for the frequency map structure seen in Fig . 2 . Specifically , temporal asymmetry , in which was greater than , was required for bands of depression . There have been reports of temporal symmetry in some preparations [2] , [17] , though others appear to have observed temporally asymmetric windows [1] , [7] , [18] . The model results remain valid as long as is greater than . With temporally symmetrical windows , in which was equal to at 20 ms , every frequency regime became potentiating , irrespective of and ( Fig . 13 ) . Further reductions of and to 10 ms resulted in identical behavior ( Data not shown ) . As expected , when was less than , all frequency regimes potentiated as well , since the amount of time between spikes favored potentiation ( Data not shown ) . Amplitude asymmetry , in which the ratio was greater than 1 , was necessary for observing potentiation . When was 1 or slightly less than 1 , favoring greater depression for the same , the frequency map no longer showed any bands of potentiation ( Fig . 14 ) . With this amplitude asymmetry , the maximal change in potentiation was greater than the maximal change in depression for the same between pre- and postsynaptic spike times . This amplitude asymmetry was consistent with several experimental reports [1] , [4] , [17] , [18] . Other models have also employed similar temporal and amplitude asymmetries [6] , [19] .
The time course and amplitude of potentiation for a single frequency regime in the model increased with greater . Longer decays are associated with earlier developmental stages [20] , [21] , at which time greater plasticity is exhibited [22] . Conversely , greater expression postnatally of the NMDAR subunit NR2A results in faster current decay [23] . The model results suggest that plasticity at synapses exhibiting shorter NMDAR current decays would be less likely to continue to potentiate with prolonged thalamocortical inputs . The time scale over which the potentiation switched to depression in most frequency regimes ( Fig . 5 ) was comparable to the decay time of the NMDAR current , motivating our focus on this current . However , the current dynamics also exhibit a time scale around 80 ms ( Eqn . 1 ) , but simulations that changed this time constant did not result in qualitative changes in either the frequency map or the average amplitude of potentiation ( data not shown ) . However , it is thought that the -mediated component of NMDAR currents may be important in STDP [24] , [25] . Other computational work addresses questions of how is used to induce plasticity [25]–[28] . Here , we focus on the implications of an STDP rule for plasticity rather than the underlying -dependent mechanisms . For simplicity , we have not considered possible changes of that are generated by NMDAR currents . While different gamma frequency inputs in the model led to broad bands of potentiation and depression ( Fig . 2 ) , theta frequency inputs did not . Yet gamma rhythms are often found nested within lower frequency theta rhythms [13] , [29] , [30] , and functionally , theta frequency modulation of gamma frequencies has been implicated in short term memory performance [31] . Our network model is specifically a gamma generating model and does not explicitly take into account theta frequency modulation . As such , the lack of broad bands of potentiation by theta frequency spiking input is not necessarily surprising , since the input spike events did not occur frequently enough to sustain potentiation in a predictable manner . However , it may be possible to consider the effect of theta rhythm modulation of the network gamma within the context of our model . Our simulations showed that pauses can help to increase potentiation in certain cases . In our simulations , the optimal pause lengths ( 80–130 ms ) approximately corresponded with the period of a theta frequency oscillation ( 80–250 ms , 4–12 Hz ) . The reduced excitability that marks the negative phase of the theta rhythm , approximately half of the theta rhythm period , may have a similar effect as the pauses in the input spiking . Future work could account for a theta frequency envelope that modulates the excitability of the gamma rhythm in the network model , which might result in further potentiation during the positive phase of the theta rhythm . In our simulations , a gamma frequency input was necessary for the presence of robust bands of potentiation in the frequency map , suggesting the involvement of a gamma frequency component in the input . Many aspects of our model parallel certain features of the auditory systems of some animals . The cortical network in the model most accurately depicts a supragranular gamma rhythm in primary auditory cortex ( A1 ) . However , there is uncertainty regarding the source of the input gamma rhythm . While it is known that there are tone frequency specific lemniscal inputs to A1 from the ventral portion of the medial geniculate complex of the thalamus ( MGv ) [32] , the connections between granular and supragranular layers within A1 are less clear . As mentioned previously , at least one report has observed gamma rhythms that appear to be anatomically localized , with a supragranular gamma rhythm and a granular gamma rhythm [13] . Because the gamma frequency input in the model is simply a spike time list , and the learning rule is dependent only on spike timing , the input can represent either a granular input to the supragranular layer or a thalamocortical input directly to the supragranular layer . The granular gamma rhythm may be endogenous or be driven by activity of thalamic origin , suggested by anatomical evidence of a projection from the medial geniculate to Layers III/IV ( LIII/IV ) in A1 [33] , [34] . It has been suggested that the MGv may convey temporally encoded information to the cortex [35] , but it is not known whether or not there is a gamma frequency component in cells projecting from MGv during auditory tasks . In the visual system , there is evidence from the lateral geniculate nucleus of the thalamus suggesting that some thalamic cells generate gamma frequency oscillations that are coherent with corresponding cortical oscillations [36]–[38] . Further investigation will be necessary to understand the specifics of the thalamocortical and interlaminar anatomy and the roles of these possibly distinct gamma rhythms . Certain predictions of our model may be testable in the thalamocortical auditory system of rats . In one circuit of thalamocortical auditory plasticity [32 , Review] , the specific thalamocortical input that carries tone frequency information ascends via at least two distinct pathways from the medial geniculate nucleus ( MG ) . The lemniscal pathway terminates in LIII/IV of A1 . The input to the cortical network in the model could represent this pathway . In some species , the non-lemniscal pathway terminates in both Layer I/II and LIII/IV of cortex from the medial portion of MG ( MGm ) [39] , [40] and is thought to carry non-specific information [41] , such as nociceptive information from a shock during a classical conditioning paradigm [32 , Review] . Additionally , there are some suggestions that this pathway may carry both specific and non-specific information [32] , [42] . In our model , the non-lemniscal pathway would be represented simply by the non-specific cortical drive ( ) . Each pyramidal cell in A1 responds selectively to a small range of tone frequencies , which peaks around the so-called best frequency ( BF ) [43] . In rats , this BF can shift to a new trained frequency ( TF ) in a classical conditioning paradigm [42 , Review] , which presumably requires some sort of synaptic modification . In these experiments , the neural response to the new TF is potentiated while the response to the old BF is slightly depressed . The model results show that STDP may be able to account for this dual plasticity between the cortical gamma rhythm and the input gamma rhythm . The variable input frequency in the model represents the encoded TF and BF frequencies , which result in potentiation and depression , respectively , and is summarized in Fig . 2 . The region highlighted in Fig . 2 shows that potentiation occurs most robustly when the gamma frequency of the encoded TF input is higher than the natural cortical frequency . We regard this as a prediction of our model . Within the context of the auditory circuit , the model results suggest that long tone shock pair presentations in a fear conditioning paradigm can result in decreased efficacy of potentiation , especially if the presence of a slower rhythm to effectively pause the gamma rhythm is attenuated [44] . In many fear conditioning experiments , the tone and shock presentations on each trial co-terminate temporally [45]–[47] . Co-activation is represented in our model by simultaneously driving the cortical oscillator during the thalamocortical input . Potentiated frequency regimes all show potentiation followed by eventual depression ( Fig . 5 ) . This observation is related to the NMDAR current decay time and the availability of glutamate to the synapse . Applied to auditory fear conditioning , the model suggests that longer NMDAR current decays at the thalamocortical E cell synapse increase the maximal potentiation and prolong potentiation . Prolonging the tone-shock pair beyond 1–2 , or 80–160 ms , would result in diminished potentiation or even depression . The model result that pauses can prolong potentiation may suggest that pulsed tone-shock presentations could result in faster , more robust potentiation . While pulsed tones have been employed in a few classical conditioning studies [48] , [49] , to our knowledge the effect of pulsed tones on the neural response , BF shift , or the efficacy of learning have not been systematically explored . Both the start time and the duration of the pause is significant in prolonging potentiation in the model ( Fig . 8 ) . The model results suggest that pauses effective in prolonging potentiation would start during the rise of potentiation and would persist for approximately the length of 1–2 . The intrinsic NMDAR current dynamics may need to be taken into consideration when choosing appropriate lengths for pauses . Further experimental work that assesses learning in an auditory fear conditioning experiment using pulsed tone shock pairs is needed to verify this result . Our results suggest that the interaction of different gamma frequency rhythms can lead to potentiation or depression , depending on the frequency ratio . The amplitude and temporal profile of this potentiation is dependent on the NMDAR current dynamics , a result relevant to both models and experimental paradigms .
The frequency-specific thalamocortical input is modeled as a list of discrete spike time events with interspike intervals ( ISIs ) corresponding to rhythms in the gamma frequency range ( 11–24 ms ) . Results are reported without noise in the input ISI , unless otherwise noted . In noisy simulations , Gaussian random noise is applied with a mean equal to a predetermined ISI and a standard deviation of about the ISI . In most noisy simulations , is 0 . 1 . Multiple inputs are also considered in certain simulations . Multiple inputs are modeled by distinct spike timing lists corresponding to each input , each with a specified average ISI and noise measure . Coherence between the inputs is defined as . The glutamate concentration external to the cortical E and I cells is based on the timing of the most recent presynaptic event . The amount of glutamate ( Eqn . 1 ) decays exponentially from each synapse with a time constant [50] , unless otherwise noted . ( 1 ) The single compartment E and I cells are Hodgkin-Huxley type cells that are coupled via synaptic currents and closely follow the formulation used in other work to represent a single auditory frequency channel [51] . Parameters for spiking currents follow previous models of a gamma rhythm [52] . The units for potentials are mV . All units for current are . The units for conductance are . Opening rates ( ) are expressed in , and closing rates ( ) are in . The specific membrane capacitance is . All alphabetical superscripts refer to particular gating variables or channel identifiers . The dynamics for the voltage of the E cell ( ) are given by:for , , , -activated , a synaptic GABAA-like I to E , AMPAR , NMDAR , leak , and applied currents , respectively . The currents allow the E cell to exhibit spike frequency adaptation ( Fig . 1B ) , as observed experimentally in neocortical pyramidal cells [3] , [53] , [54] . The strength of represents cholinergic or other neuromodulation [55] , [56] to the Layer I apical dendrites of the pyramidal cells . In the absence of any plasticity , determines the natural firing frequency of the PING oscillation . The voltage of the I cell ( ) is given similarly to that of the E cell by:for , , a synaptic AMPAR-like E to I , AMPAR , leak , and applied currents . Because the I cell is driven by the E cell , is 0 . Alone , the I cell is a fast spiking cell that exhibits no spike frequency adaptation [54] ( Fig . 1C ) . The currents for both the E and I cell are given by:for constants and . The currents are given by:for constants and . The current is given by:for constants and . The -dependent current is given by: ( 2 ) for constants , , and . The synaptic currents that couple the E and I cells are given by:for constants and , andfor constant . The leak current is given by:for and . The model includes glutamatergic AMPAR and NMDAR currents , which are both activated only within of the most recent thalamocortical spike time . For both the E and I cells , the AMPAR current is given by:for the open probability , the number of presynaptic ( thalamocortical ) inputs , and the AMPAR current reversal potential . The opening and closing rate constants are , , , and . The total AMPAR conductance for is a scalar value and is fixed for the I cell at . varies based on the spike timing rule . Each individual synapse begins with for a fixed initial value of the total conductance and the number of inputs . The total is therefore independent of in this model . For the E cell , the NMDAR current is given by [57]: ( 3 ) The constants are and . Unless otherwise noted , the closing rate is , corresponding to an NMDAR current decay time constant of 80 ms . In the absence of any presynaptic input , the spiking behavior of the mutually coupled E and I cells for a given is shown in Fig . 1D . At the thalamocortical E cell synapse , the model utilizes a classical additive STDP rule that takes into account the contributions of every pair of pre- and postsynaptic spikes [5] , [16] . Here , the presynaptic spike times are the thalamocortical input spike times , and the postsynaptic spike times are calculated when the E cell's voltage crosses an arbitrarily chosen +40 mV threshold . At each detection of an E cell spike , the model calculates the total contribution to the synaptic modification , which updates after detection of each postsynaptic spike and immediately affects the next cycle of firing . The fractional synaptic modification for each individual presynaptic ( thalamocortical ) spike time and each postsynaptic ( cortical E cell ) spike time is calculated according to Eqn . ( 3 ) for [5] . ( 4 ) Unless otherwise noted , the potentiation decays with a time constant of , and depression decays with a time constant of [5] . For a given positive spike pair , the maximal amplitude for modification is . For a given negative spike pair , the maximal amplitude for depression is . The total synaptic modification over all spike pairs is given by:The AMPAR conductance for each input is given by the initial conductance plus the synaptic modification:Additionally , at each synapse , the total maximal conductance was restricted to , to ensure a maximal equal to , whose value was 0 . 25 . Two primary parameters are tuned in each of the simulations: the current applied to the E cell ( ) and the average interspike interval ( ISI ) of the thalamocortical input . Changing the changes the natural frequency of the cortical oscillator . A range of 6–18 elicits a cortical oscillation within the gamma frequency range of interest ( 30–90 Hz ) . Changing the ISI from 11–33 ms changes the thalamocortical input gamma frequency from 30 . 3 Hz to 90 . 9 Hz . The closing rate for NMDAR currents is equivalent to and is varied in some simulations . Other simulations investigate the effects of changing glutamate time constants for AMPAR and NMDAR currents , independently , motivated by evidence concerning the time courses of activation and decay [50] , [58] . This model is based on several biophysical constraints and , though simplified , is not meant for mathematical analysis . Further work may elucidate some aspects of the results , including the potential resonance between the inputs and the network . Potentiation and depression are defined in this paper by AMPAR conductance increases and decreases , respectively . The total potentiation is defined by the area underneath the curve within a width about the peak conductance value , as shown in Eqn . ( 4 and 5 ) . Depression is defined when is less than zero . ( 5 ) ( 6 ) Each frequency regime is run for 500 ms of simulation time , in which both the thalamocortical input and cortical drive are activated . The simulations and subsequent analysis are run using MATLAB 7 . 6 . 0 . 324 ( The MathWorks , Natick , MA ) . The differential equations are integrated using MATLAB's built in ode15s stiff solver of variable order and time step .
|
Rhythms are well studied phenomena in many animal species . Brain rhythms in the gamma frequency range ( 30–90 Hz ) are thought to play a role in attention and memory . In this paper , we are interested in how cortical gamma rhythms interact with information specific inputs that also have a significant gamma frequency component . The results from our computational model show that plasticity associated with learning depends on the specific frequencies of the input and cortical gamma rhythms . The results show a mechanism by which both increases and decreases in the strength of the input connection can occur , depending on the specific frequency of the input . A current mediated by NMDA receptors may be responsible for the temporal course of the plasticity seen in these brain regions . We discuss the implications of our results for conditioning paradigms applied to auditory learning .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"neuroscience",
"neuroscience/theoretical",
"neuroscience",
"neuroscience/sensory",
"systems"
] |
2009
|
Cortical Gamma Rhythms Modulate NMDAR-Mediated Spike Timing Dependent Plasticity in a Biophysical Model
|
EHBP-1 ( Ehbp1 ) is a conserved regulator of endocytic recycling , acting as an effector of small GTPases including RAB-10 ( Rab10 ) . Here we present evidence that EHBP-1 associates with tubular endosomal phosphatidylinositol-4 , 5-bisphosphate [PI ( 4 , 5 ) P2] enriched membranes through an N-terminal C2-like ( NT-C2 ) domain , and define residues within the NT-C2 domain that mediate membrane interaction . Furthermore , our results indicate that the EHBP-1 central calponin homology ( CH ) domain binds to actin microfilaments in a reaction that is stimulated by RAB-10 ( GTP ) . Loss of any aspect of this RAB-10/EHBP-1 system in the C . elegans intestinal epithelium leads to retention of basolateral recycling cargo in endosomes that have lost their normal tubular endosomal network ( TEN ) organization . We propose a mechanism whereby RAB-10 promotes the ability of endosome-bound EHBP-1 to also bind to the actin cytoskeleton , thereby promoting endosomal tubulation .
Transmembrane proteins enter cells via several endocytic pathways including clathrin-dependent endocytosis ( CDE ) and a variety of less well understood clathrin-independent endocytosis ( CIE ) mechanisms [1–3] . After internalization some receptors will be recycled back to the plasma membrane via the endocytic recycling compartment ( ERC ) [4 , 5] . Recycling endosome transport is known to be essential for diverse biological processes , including cell migration , cytokinesis , and synaptic plasticity [5] . In the C . elegans intestine the small GTPase RAB-10 resides on a subset of basolateral endosomes where it regulates basolateral cargo recycling upstream of RME-1/EHD , a membrane remodeling protein with Dynamin-like features [6–9] . While the cargo-specificity of RME-1 is broad , RAB-10 appears more specific , with especially potent effects on the recycling of transmembrane proteins internalized by CIE , such as the model CIE cargo hTAC ( the alpha-chain of the human IL2 receptor ) [6 , 10] . Rab10 function in mammalian cells appears highly conserved , where Rab10 is highly enriched on the membranes of the common recycling endosomes and regulates basolateral recycling in polarized epithelial cells [11] . Likewise , in mammalian adipocytes , Rab10 functions in the insulin-stimulated recycling of glucose transporter GLUT4 [12] . The calponin homology ( CH ) domain protein Ehbp1 has also been reported to function in GLUT4 recycling in adipocytes , associated with the RME-1 homologs EHD1 and EHD2 [13 , 14] . In our previous work we determined that C . elegans EHBP-1 binds to the GTP-loaded conformation of RAB-10 through its C-terminal domain ( a predicted coiled-coil ) and functions with RAB-10 in the intestinal basolateral recycling of hTAC , and in the neuronal recycling of AMPA-type glutamate receptor GLR-1 [10 , 15] . EHBP-1 labels an extensive network of tubular endosomes in the intestine where it colocalizes with recycling cargo , and is also found on connected punctate endosomal membranes where it colocalizes with RAB-10 . Loss of EHBP-1 produces phenotypes that strongly resemble those produced upon loss of RAB-10 . These include RAB-10-specific phenotypes in polarized cells such as the intestinal epithelium , including accumulation of enlarged basolateral endosomes filled with fluid-phase markers and hTAC , and the abnormal accumulation of endosomal GLR-1 in interneurons [10 , 15] . ehbp-1 mutants or RNAi also produce phenotypes in non-polarized cells very similar to simultaneous loss of RAB-10 and its closest paralog RAB-8 , including variable larval arrest , and fully penetrant adult sterility due to a failure in germline membrane transport and oocyte growth [15] . In Drosophila dEHBP1 has also been reported to act with Rab11 [16 , 17] . Our previous studies found that a truncated form of EHBP-1 lacking the RAB-10 interaction domain remained membrane associated , raising the question of how EHBP-1 associates with endosomal membranes [15] . Although not apparent in simple homology searches , a purely computational study using sequence profile searches with profile–profile comparison and fold recognition methods classified the EHBP-1 N-terminus as a putative C2-like domain ( NT-C2 ) that could potentially mediate direct membrane binding [18] . It has been shown that endosomal recruitment of some conserved recycling regulators depends on the regulatory lipid phosphatidylinositol-4 , 5-bisphosphate [PI ( 4 , 5 ) P2] [9] . PI ( 4 , 5 ) P2 is enriched at the plasma membrane and recycling endosomes , and membrane bending proteins associated with recycling function such as RME-1/EHD and AMPH-1/Amphiphysin/BIN1 have been shown to associate with membrane structures enriched in PI ( 4 , 5 ) P2 [9 , 19 , 20] . In fact , we have previously shown that the PI ( 4 , 5 ) P2 level in basolateral recycling endosomes is modulated by RAB-10 , in part through its effector CNT-1 , an ARF-6 GAP [20] . Other reports also indicate a requirement for phosphatidylinositol-4-phosphate ( PI4P ) in recycling endosome function [21] . These findings imply that EHBP-1 could be targeted to recycling endosomes via PI ( 4 , 5 ) P2 and/or PI ( 4 ) P binding . In addition to its N-terminal C2-like and C-terminal RAB-10-binding domains , EHBP-1 harbors a central CH domain . CH domains in different proteins are known to bind to the cytoskeleton , but vary in their specificity , with some binding to the microtubule cytoskeleton and others binding to actin filaments [22] . Requirements for the microtubule and actin cytoskeletons are well established in the endosomal system [23–26] . The actin cytoskeleton also plays essential roles along the endocytic pathway . First identified in studies of endocytosis in yeast , Arp2/3-dependent nucleation of actin at endocytic sites has been observed in many organisms , including mammals , and is thought to contribute to membrane fission [27–31] . Furthermore , certain forms of endocytic recycling are also actin-dependent . For instance , actin depolymerization results in the retention of TAC in tubular recycling endosomes together with Arf6 , suggesting the necessity of actin function in Arf6-mediated recycling transport [32 , 33] . Retromer/WASH mediated local actin polymerization on endosomes has alternately been reported to enhance the fission of tubular cargo carriers from endosomes , or to stabilize tubular extensions for cargo loading prior to their release by fission [34 , 35] . Cargo carriers in the endosomal system are often tubular in nature , and their tubular shape has been proposed to help sort membrane intrinsic components away from lumenal content [4 , 36] . The endocytic recycling compartment in mammals is composed of a dense collection of endosomal tubules and vesicles [4] . In the C . elegans intestine the basolateral recycling compartment enriched in CIE cargo , EHBP-1 , and RME-1 , is highly tubular in nature and appears to have many interconnections [8 , 15 , 37] . The entire network collapses to vesicles upon loss of RAB-10 or EHBP-1 , suggesting that they contribute to the formation and/or maintenance of such tubular endosomes [6 , 15] . To further dissect the function of RAB-10 effector EHBP-1 , we studied individual domains of EHBP-1 in vitro and in vivo , and characterized EHBP-1 regulation by RAB-10 . Here we demonstrate that the NT-C2 and CH domains are both indispensible for proper EHBP-1 function . We found that the EHBP-1 NT-C2 domain has an intrinsic ability to associate with endosomal membranes . RNAi-mediated knockdown of phosphoinositide kinases , colocalization assays with PI ( 4 , 5 ) P2 biosensor PH ( PLCδ ) -GFP , and liposome co-sedimentation assays revealed that the EHBP-1 NT-C2 domain preferentially associates with PI ( 4 , 5 ) P2 enriched endosomes via predicted patches of basic residues within the NT-C2 domain . Our biochemical studies indicate that the EHBP-1 CH domain preferentially binds to actin filaments and not microtubules , and EHBP-1 colocalizes with endosomal actin in vivo . Remarkably we find that the interaction of the EHBP-1 C-terminal domain with RAB-10 ( GTP ) enhances the actin filament affinity of EHBP-1 via its central CH domain . Our data demonstrates that RAB-10 regulates EHBP-1 actin binding and suggests that RAB-10 and EHBP-1 function together with actin to create and/or maintain endosomal tubulation .
In ehbp-1 ( tm2523 ) deletion mutants , or upon RNAi of ehbp-1 , the tubular endosomal network is disrupted and large vacuoles accumulate near the basolateral membranes of the intestinal cells , a phenotype very similar to rab-10 mutants . Such vacuoles are grossly enlarged early endosomes that can be labeled by fluid-phase endocytosis markers taken up from the basolateral surface ( Fig 1B , S1A and S1A' and S1E Fig ) [15] . This vacuole phenotype can be fully rescued by intestine-specific expression of tagged forms of full-length EHBP-1 ( Fig 1C and 1C' , S1B–S1B'' and S1E Fig ) . EHBP-1 contains three distinct protein domains , including an N-terminal C2-like domain ( NT-C2 ) , central CH ( Calponin Homology ) domain , and C-terminal predicted coiled-coil ( CC ) domain ( Fig 1A ) [15 , 18] . Our previous studies showed that the predicted CC domain of EHBP-1 binds to RAB-10 ( GTP ) , and EHBP-1 missing the CC domain does not rescue the ehbp-1 mutant intestinal vacuole phenotype [15] . However , removal of the RAB-10-binding CC-domain does not cause redistribution of EHBP-1 to the cytoplasm . Rather , EHBP-1 lacking the CC-domain remains associated with misshapen endosomal membranes and acts as a dominant negative , impairing recycling [15] . These results indicated that while the CC-domain is important for function , EHBP-1 must have a mechanism for membrane association independent of the RAB-10 binding domain . Bioinformatics analysis suggests that the N-terminus of EHBP-1 contains a C2-like domain termed the NT-C2 . Since many C2 domains bind directly to membrane lipids , the newly proposed NT-C2 domain is an excellent candidate to mediate such EHBP-1 membrane binding [18] . To better understand the functional significance of the predicted EHBP-1 NT-C2 domain , we analyzed the ability of GFP-tagged EHBP-1 missing the predicted NT-C2 domain ( EHBP-1 ( ΔNT-C2 ) -GFP ) to rescue the intestinal vacuolation of ehbp-1 ( tm2523 ) mutants . Intestinally expressed EHBP-1 ( ΔNT-C2 ) -GFP failed to rescue ( Fig 1D and 1D' , S1C–S1C'' and S1E Fig ) , and was much more diffusive than intact EHBP-1 , consistent with a function for the NT-C2 domain in recruiting EHBP-1 to membranes . Furthermore , intestinal over-expression of the CH-CC fragment ( EHBP-1 ( ΔNT-C2 ) ) in an otherwise wild-type background disrupted recycling cargo hTAC-GFP tubular endosomal localization and caused hTAC-GFP accumulation on enlarged endosomes and vacuoles ( S2E–S2F' and S2H and S2I Fig ) . In the absence of the NT-C2 domain , the only remaining localized EHBP-1 signal was restricted to small punctate structures ( S2A Fig ) , unlike full length EHBP-1-GFP , which localizes strongly to abundant tubular endosomes in the basolateral cortex , as well as apparently attached endosomal puncta ( Fig 1C' ) . Notably , the residual EHBP-1 ( ΔNT-C2 ) -GFP labeled puncta were lost upon removal of RAB-10 ( S2B Fig ) . Taken together these results suggest that the NT-C2 domain of EHBP-1 is important for EHBP-1 function and the recruitment to tubular endosomal membranes , with a contribution by the RAB-10-binding CC-domain in recruitment to punctate endosomes . Next we asked if the NT-C2 domain of EHBP-1 is sufficient to direct GFP to endosomal structures . When expressed in the C . elegans intestine , EHBP-1-NT-C2 ( aa1-223 ) -GFP localized to tubular and punctate endosomes , very similar to full length EHBP-1-GFP ( Fig 2A and 2B and Fig 2F ) , co-localizing with PI ( 4 , 5 ) P2 biosensor PH ( PLCδ ) -GFP on basolateral tubular and punctate membrane structures ( Fig 3A–3A" ) [15] . Like full length EHBP-1 , EHBP-1-NT-C2 ( aa1-223 ) -GFP also overlapped with ARF-6-RFP and RFP-RAB-10 on basolateral puncta ( S3A–S3B'' Fig ) . EHBP-1-NT-C2 ( aa1-223 ) -GFP showed little colocalization with RFP-2xFYVE , a marker for early endosome enriched lipid PI ( 3 ) P ( S4A–S4B'' Fig ) . Together our results indicate that the EHBP-1 NT-C2 domain is sufficient to direct EHBP-1 to endosomes , probably via direct membrane binding . Since tubular recycling endosomes are enriched in PI ( 4 , 5 ) P2 and the EHBP-1 NT-C2 domain displayed a similar subcellular distribution to a PI ( 4 , 5 ) P2 biosensor , we tested whether PI ( 4 , 5 ) P2 or other phosphoinositides are important for NT-C2 membrane recruitment , using RNAi-based knockdown of PI-kinases involved in phosphatidylinositol metabolism ( Fig 3B–3I ) . Loss of a key EHBP-1 binding lipid is expected to result in diffusion of NT-C2-GFP protein in vivo . Depletion of PI-directed PI4-kinase PIFK-1 produced a nearly 3-fold decrease in EHBP-1 ( NT-C2 ) -GFP labeling of endosomal puncta and tubules ( Fig 3F and Fig 3I ) . RNAi knockdown of PI ( 5 ) P-directed PI4-kinase PPK-2 produced a more modest ( ~35% ) decrease in EHBP-1 ( NT-C2 ) -GFP endosome association ( Fig 3E and Fig 3I ) . PI ( 4 , 5 ) P2 levels are also regulated by PI5-kinases . Consistent with PI ( 4 , 5 ) P2 being a key lipid in EHBP-1 membrane recruitment , we observed a strong decrease in EHBP-1 ( NT-C2 ) -GFP labeling of endosomal puncta and tubules after RNAi-mediated depletion of PI ( 4 ) P-directed PI5-kinase PPK-1 , with a ~98% decrease in average intensity ( Fig 3G and Fig 3I ) . Knockdown of PI ( 3 ) P-directed PI5-kinase PPK-3 did not affect EHBP-1 ( NT-C2 ) -GFP distribution ( Fig 3H and Fig 3I ) . In addition we did not find any effects on EHBP-1 ( NT-C2 ) -GFP localization after RNAi of PI3-kinases including PI ( 4 , 5 ) P2-directed PI3-kinase AGE-1 and PI-directed PI3-kinase VPS-34 ( Fig 3C and 3D and Fig 3I ) . Taken together , these results suggested that EHBP-1 membrane recruitment requires PI ( 4 , 5 ) P2 in endosomes , whose level is mainly regulated by PI5-kinase PPK-1 and PI4-kinase PIFK-1 , and to a lesser extent PI4-kinase PPK-2 . Indeed , similar to the ehbp-1 ( tm2523 ) mutant phenotype , hTAC-GFP lost its tubular endosomal localization and accumulated in punctate structures upon the knockdown of PPK-1 , suggesting that endosomal PI ( 4 , 5 ) P2 is required for the recycling function of EHBP-1 ( S2G–S2G' Fig ) . PI ( 4 , 5 ) P2 levels on endosomes is regulated by GTPase ARF-6 , presumably through activation of PI5-kinase PPK-1 [20] . Therefore , EHBP-1 recruitment to endosomes may be dependent on ARF-6 . Consistent with the action of ARF-6 in PPK-1 regulation , we found that EHBP-1 ( NT-C2 ) -GFP basolateral tubular and puncta labeling decreased in arf-6 mutants ( ~60% ) ( Fig 3J–3L ) . To further define the endosomal phosphoinositide association preference of EHBP-1 , we tested the interaction of purified recombinant GST-NT-C2 with liposomes in sedimentation assays . We found that GST-NT-C2 preferentially pelleted with liposomes containing 5% PI ( 4 , 5 ) P2 , with much less co-sedimentation with liposomes containing PI or PI ( 4 ) P ( Fig 3M and 3N and S4C Fig ) . To determine residues that may contribute to membrane association , we analyzed evolutionary sequence conservation within the NT-C2 domain , and mapped conserved positively charged residues onto a homology model that we constructed . Our model suggests that the N-terminal 160 amino acids of EHBP-1 folds into a globular domain consisting of seven β-strands and an α-helical segment between strand-5 and strand-6 . A patch of basic residues at the extreme N-terminus of the fold prior to strand-1 is predicted in this model to contribute to the formation of a concavity in the β sheet , on the “upper surface” comprised of a constellation of basic and hydrophobic residues ( Fig 4H ) . We performed alanine substitution in three areas of the NT-C2 domain in the context of the NT-C2-GFP intestinal expression construct ( Fig 4G ) . Compared with wild type , modification of the four arginines within the patch prior to strand-1 , predicted to line the concavity , results in loss of association with tubular endosomal membranes and diffusion of the mutated NT-C2-GFP within the cytosol ( Fig 4A and 4B and Fig 4I ) . Mutation of pairs of arginines within this sequence reduced but did not eliminate membrane association , suggesting that all four arginines contribute to NT-C2 domain membrane binding ( Fig 4E and 4F and Fig 4I ) . Furthermore , we assayed the vacuole phenotype in ehbp-1 ( tm2523 ) mutants animals expressing EHBP-1 ( RRLRR6AALAA ) -GFP . In contrast to WT EHBP-1 ( S1B–S1B'' Fig ) , the number and size of vacuoles were not rescued by EHBP-1 ( RRLRR6AALAA ) -GFP , indicating that the RRLRR motif is critical for EHBP-1 recycling function ( S4D and S4E Fig ) . Mutation of nearby lysines , predicted to face away from the cleft , had no effect on the association of NT-C2-GFP with tubular endosomes ( Fig 4D and Fig 4I ) . Another patch of basic residues ( HRRRK of strand-2 ) on the predicted surface of the NT-C2 domain also appears to contribute to NT-C2 membrane association , as mutation of this sequence also produced a diffusive localization ( Fig 4C and Fig 4I ) . Collectively , our results suggest that EHBP-1 NT-C2 domain associates with PI ( 4 , 5 ) P2 enriched endosomal membranes through two patches of basic amino acids . To determine if the CH domain is important for the subcellular localization of EHBP-1 to endosomes we expressed GFP-tagged EHBP-1 lacking the CH domain ( EHBP-1 ( ΔCH ) -GFP ) in the intestinal epithelia . The CH domain did not appear to be a major determinant of localization , since EHBP-1 ( ΔCH ) -GFP was enriched on tubular and punctate membranes in a pattern indistinguishable from intact EHBP-1-GFP ( S2C Fig ) . This localization was also RAB-10-dependent like full-length EHBP-1 , displaying significant intracellular accumulation on puncta and vacuoles in rab-10 ( ok1494 ) mutant animals ( S2D Fig ) . Nevertheless , intestinally expressed EHBP-1 ( ΔCH ) -GFP failed to rescue the ehbp-1 ( tm2523 ) mutant vacuole phenotype ( Fig 1E–1E' , S1D–S1D'' and S1E Fig ) , indicating that the CH domain is also indispensible for EHBP-1 function . Unlike the NT-C2 domain above , which conferred robust localization to endosomes on its own , expression of the EHBP-1 CH-domain ( aa260-510 ) , fused to GFP , localized relatively diffusely in the intestinal cells . Sparse puncta were visible above background . These puncta partially overlapped with ARF-6 and RAB-10 , indicating that they represent very weak recruitment to endosomes ( Fig 2C and Fig 2F and S3C–S3D'' Fig ) . By contrast , the C-terminal RAB-10-binding domain , expressed as EHBP-1-CC ( 510-901aa ) -GFP , lacked tubular localization but retained visible localization on punctate endosomes labeled by ARF-6-RFP and RFP-RAB-10 ( Fig 2D and 2D' and Fig 2F and S3E–S3F'' Fig ) . RAB-10 was required for this punctate recruitment , since the punctate labeling of EHBP-1-CC-GFP was lost in rab-10 ( ok1494 ) mutant animals ( Fig 2E and 2E' and Fig 2F ) . This is distinct from the full length EHBP-1 protein that remains membrane associated in a rab-10 mutant background , presumably through the NT-C2 domain [15] . CH domains have the potential to interact with cytoskeletal elements [22] . Structural studies on the kinetochore attached Ndc80 complex indicated that a CH-domain pair is involved in microtubule binding [38] . Mammalian EHBP1 was shown to colocalize with the cortical actin cytoskeleton in COS-1 cells , and overexpression of HA-EHBP1 induced cortical actin rearrangement [13] . Homology analysis suggests that the CH domain of C . elegans EHBP-1 most closely resembles the CH domain of β spectrin and the second CH domain of utrophin [39 , 40] . To determine if the EHBP-1 CH-domain interacts with actin microfilaments or microtubules , we assayed for interaction in vitro using co-sedimentation assays ( Fig 5A–5F ) . First , we validated our co-sedimentation assays using human Utrophin actin binding CH-domains ( aa1-261 ) and detected significantly elevated co-sedimentation of this Utrophin fragment with filamentous actin ( S7C Fig ) . Similarly , we found that a purified fusion of GST to the EHBP-1 CH-domain increased its sedimentation by more than 5-fold in the presence of actin microfilaments , indicating that the EHBP-1 CH domain binds to polymerized actin ( Fig 5A–5C ) . By contrast , addition of microtubules to the reaction failed to enhance GST-CH sedimentation ( Fig 5D–5F ) . These results suggested that the EHBP-1 CH-domain functions to link EHBP-1 to polymerized actin . Previous work in C . elegans indicated that microtubules are required for the structure of tubular endosomes in the basolateral intestine [37] . Because we found binding of EHBP-1 to actin microfilaments in vitro , we asked whether actin polymerization is also important for the structure of these endosomes . Thus we injected the actin depolymerizing drug latrunculin B ( LatB ) into the worm pseudocoelom ( body cavity ) and assayed for effects on the EHBP-1-GFP labeled tubular endosomal meshwork . Indeed , LatB treatment greatly disrupted the EHBP-1-GFP pattern , converting many of the tubules to puncta ( Fig 5G–5H and Fig 5J and 5K ) . Similar treatment with the microtubule-depolymerizing drug nocodazole ( Noc ) affected the EHBP-1-GFP meshwork in a different manner . The tubular network was still observed , but in a dotted line pattern ( Fig 5I–5K ) . We also assayed the distribution of PH ( PLCδ ) -GFP labeled basolateral endosomal tubules after LatB and Noc treatments and observed similar results ( S5A–S5D Fig ) . These results indicate that formation or maintenance of basolateral tubular endosomes labeled by EHBP-1 requires both actin and microtubule cytoskeletal elements , although EHBP-1 itself is probably actin-specific in its interactions . To further test the functional involvement of actin and microtubules in EHBP-1 mediated recycling we assayed Lat B and Noc treatments for effects on the well-defined recycling CIE cargo marker hTAC-GFP [6 , 15] . In our previous work we showed that loss of EHBP-1 specifically impaired hTAC-GFP recycling [15] . Our analysis indicates that hTAC-GFP accumulates intracellularly in intestinal epithelial cells after depolymerization of either actin or microtubules ( S6A–S6D Fig ) . If EHBP-1 links endosomal membranes to actin microfilaments then we would expect to find colocalization of EHBP-1-GFP with F-actin marker Lifeact-RFP . Indeed we found that many punctate regions of endosomes labeled by EHBP-1-GFP in intestinal cells were positive for Lifeact-RFP ( Fig 5L–5L" ) . This is consistent with the localization of actin to endosomes [19] . Likewise , some tubular overlap was observed between EHBP-1-RFP and EMTB-GFP , a marker for microtubules ( S6E–S6E" Fig ) , indicating that EHBP-1 labeled tubular endosomes orient co-linearly with cortical microtubules . We also analyzed the importance of the EHBP-1 CH domain for EHBP-1 colocalization with actin . We found that CH-GFP , lacking the other domains of EHBP-1 , colocalized with Lifeact-RFP on intestinal endosomes ( S6F–S6F" Fig ) . To determine whether the EHBP-1 localization with actin is dependent on its CH domain in vivo , we assayed for colocalization of EHBP-1 missing the CH domain with Lifeact-RFP in wild-type and in a rab-10 ( ok1494 ) mutant background . EHBP-1 ( ΔCH ) -GFP retained some overlap with Lifeact-RFP ( Fig 5M–5M" and Fig 5O ) . However , in rab-10 ( ok1494 ) mutant animals , EHBP-1 ( ΔCH ) -GFP fusion protein overlap with Lifeact RFP decreased ~83% , with most remaining GFP positive structures offset from Lifeact-RFP puncta ( Fig 5N–5N" and Fig 5O ) . Thus we interpret the EHBP-1 ( ΔCH ) -GFP colocalization with Lifeact-RFP to be mediated via interaction with endogenous RAB-10 and not direct interaction with actin . Our data suggested that EHBP-1 CH-domain associates with endosomal actin microfilaments in vitro and in vivo . To determine whether the RAB-10-binding CC-domain influences the actin affinity of the EHBP-1 CH-domain , we assayed for effects of RAB-10 on the ability of a GST-CH-CC fusion protein to co-sediment with F-actin . Without addition of RAB-10 to the reaction , an EHBP-1 fragment containing the CH and CC domains displayed a similar level of interaction with F-actin as the CH domain alone ( Fig 5A and S7A and S7B Fig ) . However , we detected elevated co-sedimentation of CH-CC with F-actin in the presence of active HA-RAB-10 ( Q68L ) , suggesting that RAB-10 ( GTP ) interaction with the EHBP-1 CC-domain enhances the ability of EHBP-1 to bind to actin filaments ( Fig 6A and Fig 6C and 6D ) . In contrast , addition of HA-RAB-10 ( Q68L ) did not enhance the ability of an EHBP-1 fragment lacking the RAB-10 binding CC-domain ( C2-CH ) to co-sediment with actin ( S7D and S7E Fig ) . As expected , we did not detect interaction between a GST control protein and F-actin ( Fig 6B and Fig 6C ) . Consistent with the in vitro data , we found that the EHBP-1 ( CH-CC ) -GFP colocalized well with Lifeact-RFP in the C . elegans basolateral intestine ( S6G–S6G" Fig ) . Importantly , the augmented physical interaction between EHBP-1 and F-actin in the presence of RAB-10 was further confirmed in vivo by co-immunoprecipitation experiments between GFP-tagged EHBP-1 and endogenous actin in whole-worm lysates . EHBP-1-GFP was immunoprecipitated from worm lysates using an anti-GFP antibody and the precipitants were probed with an anti-actin antibody on western blots . The amount of actin co-immunoprecipitating with EHBP-1-GFP was strongly reduced in rab-10 ( ok1494 ) mutants , suggesting that RAB-10 promotes the interaction of EHBP-1 with F-actin ( Fig 6E ) . Loss of EHBP-1 disrupts the tubular endosomal network as visualized by hTAC-GFP ( Fig 6F and 6G ) or ARF-6-RFP ( S8A and S8B Fig ) . Using the integrity of the hTAC-GFP labeled network as an assay , we sought to test the functionality of versions of EHBP-1 containing different combinations of domains . Importantly we found that overexpression of an EHBP-1 fragment including the membrane associating NT-C2 and F-actin binding CH domains can partially rescue the steady state tubular pattern of recycling cargo marker hTAC-GFP ( Fig 6F and 6H and Fig 6K ) . This was in sharp contrast to the effects of expressing a CH-CC fragment or C2-CC fragment , neither of which could restore hTAC-GFP tubularity ( Fig 6I and 6J and Fig 6K ) . This difference in rescuing ability was even more apparent in time-lapse imaging . Normally the hTAC-GFP labeled endosomal network in the basolateral intestine is highly dynamic , with frequent movement of puncta and tubules ( Fig 6L and S1 Video ) . In ehbp-1 mutant animals the hTAC-GFP labeled endosomes are devoid of movement , appearing almost completely static ( Fig 6L and S2 Video ) . This could be significantly rescued in an ehbp-1 mutant expressing the C2-CH fragment , but not upon expression of CH-CC or C2-CC fragments ( Fig 6L and S3–S5 Videos ) . Compared with ~18 tubule movement events ( per unit area ) /180 sec in wild-type animals , and ~1 event/180 sec in ehbp-1 ( tm2523 ) mutant animals , C2-CH expression animals presented moderate dynamics with ~7 events/180 sec ( Fig 6L ) . These results are consistent with an important role for EHBP-1 in linking the endosomal membrane to the actin cytoskeleton , and the EHBP-1 CC domain-RAB-10 interaction acting as an enhancer for EHBP-1 CH domain-actin filaments binding during endocytic recycling , regulating membrane tubule formation and function .
Our studies in C . elegans have demonstrated a requirement for EHBP-1 in basolateral recycling of CIE cargo in intestinal epithelia and postsynaptic recycling of AMPA receptors in interneurons , functioning with the small GTPase RAB-10 [10 , 15] . EHBP-1 is enriched in the intestinal cells on basolateral tubular and punctate endosomes , and loss of EHBP-1 results in reduced levels of interacting protein RAB-10 on endosomal membranes [15] . Loss of RAB-10 or EHBP-1 also completely disrupts the tubular character of these endosomes [15] . Our new data suggests that this loss of tubular character , which is closely linked with recycling endosome function , is due to a loss of EHBP-1-dependent linkage between endosomal membranes and F-actin . The EHBP-1 N-terminal domain was predicted by bioinformatics to adopt a C2 domain-like fold ( termed NT-C2 ) that might allow it to bind to membrane phosphatidylinositols , while the central CH domain suggested an interaction with the cytoskeleton [18 , 22] . In this study , we demonstrated the pivotal roles of EHBP-1 NT-C2 domain and CH domain in EHBP-1-mediated recycling regulation . Using in vitro and in vivo assays , we showed that the NT-C2 domain association with endosomal membranes requires two groups of basic residues predicted to form surface patches that could interact with phosphoinositides . We also found that the CH-domain associates with actin filaments but not microtubules , and that F-actin is important for developing the tubular character of these EHBP-1 associated endosomes . Remarkably , we found that the interaction of the EHBP-1 CC domain with RAB-10 ( GTP ) enhanced the CH domain affinity for actin filaments . Thus our studies suggest that RAB-10 promotes bridging of recycling endosomes and actin filaments via EHBP-1 to create or maintain endosomal tubulation . Recent phylogenetic analysis and structural modeling predicted an NT-C2 domain in the Ehbp1/EHBP-1 extreme N-terminus , providing a potential interface for EHBP-1 membrane lipid binding [18] . Studies focusing on the well known Ca2+-dependent C2 domain of PKC and the Ca2+-independent C2 domain of PI3K proposed that C2 domain lipid binding capacity involves two structural segments including a calcium binding pocket-like structure and a β-sandwich surface respectively [41 , 42] . The negatively charged acidic residues in the pocket can coordinate Ca2+ and lipid binding [41 , 43] . The clustered positively charged basic residues ( H , R and K ) within the β-sandwich regions of Ca2+-independent C2 domains participate in the interaction with negatively charged lipids [18] . However , bioinformatics predictions of the membrane binding mode of NT-C2 family proteins suggested that the NT-C2 extreme N-terminus , prior to strand-1 , contains a patch of basic residues on the surface , contributing to lipid binding in parallel with the β-sandwich concave surface [18] . Our data revealed two regions of basic residues that appear to contribute to EHBP-1 NT-C2 membrane association . In our structural model these two basic regions appear to be located on opposite sides of the domain . Further structural dissection of the NT-C2 will be required to determine the true arrangement . As reported in previous studies , PI ( 4 , 5 ) P2 and PI4P are both enriched in recycling endosomes and are important for recycling transport [20 , 21] . Our experimental results clearly indicated that the EHBP-1 NT-C2 domain is required for association with tubular endosomes and interacts with PI ( 4 , 5 ) P2 . Within the limits of our assays , we did not observe obvious changes in the tubular endosomal network upon knockdown of PI3 kinases known to be important for early endosome function , such as type III PI3-kinase VPS-34 or type I PI3-kinase AGE-1 , suggesting that they mainly affect other aspects of endosome function [44–55] . Phosphatidylserine ( PS ) is also known to be quite important for recruitment of many peripheral membrane proteins necessary for membrane traffic , including endocytic recycling , and phosphatidic acid ( PA ) has been implicated in recycling tubule formation in mammalian cells [56–61] . It will be important to test for roles of PS and PA in EHBP-1 function in the future . C2 domains display a wide range of lipid selectivity , with preference for anionic PS and phosphatidylinositol-phosphates ( PIPs ) [41] . Unlike lipid binding PH domains [62] , C2 lipid targeting often involves two recognition components , such as two lipids or a lipid/protein combination . For instance the protein kinase C ( PKC ) C2 domain uses its basic surface residues to bind plasma membrane PS and PI ( 4 , 5 ) P2 [63] , while cytosolic phospholipase A2 ( cPLA2 ) binds to the neutral lipid phosphatidylcholine ( PC ) and the anionic lipid ceramide-1-phosphate ( C1P ) through C2 domain Ca2+ site charged hydrophobic side chains and a basic cluster [64] . Synaptotagmin utilizes two C2 domains to bridge the vesicular and plasma membranes , with the C2A domain binding vesicular PS and SNARE , while the C2B domain binds plasma membrane PI ( 4 , 5 ) P2 and SNARE [65 , 66] . The molecular basis for phosphoinositide-binding specificity of C2 and C2-like domains has been explored in recent years . Structural analysis of the PKCα C2 domain showed that PI ( 4 , 5 ) P2 binds to the concave surface of β3 and β4 strands . Intriguingly , aromatic residues Tyr195 ( strand 2 ) and Trp245 ( strand 5 ) interact directly with the inositol ring phosphate moieties of PI ( 4 , 5 ) P2 . Loss of Tyr195 and Trp245 abrogated PI ( 4 , 5 ) P2 recognition and plasma membrane association of PKCα [67] . Phylogenetic analysis showed that Tyr195 and Trp245 are conserved among different C2 domains except in the DOCK-C2 and NT-C2 families [18] . However , Trp71 of EHBP-1/Ehbp1 NT-C2 strand 4 is highly conserved among NT-C2 family members . One plausible possibility is that the Trp71 residue participates , at least in part , in PI ( 4 , 5 ) P2 binding specificity . Further functional analysis will be required to test this model . Filamentous actin has long been known to be particularly important for the Arf6-mediated recycling of CIE cargo such as TAC and MHCI [5 , 32] . Colocalization assays and the presence of predicted actin-binding domains have indicated that NT-C2 proteins are involved in actin binding [13 , 18] . For instance Ehbp1 colocalizes with cortical actin filaments in cultured mammalian adipocytes and in Drosophila pII cells of the external mechanosensory organs [13] . Since certain CH domains have extensively documented actin binding and bundling functions , we hypothesized that EHBP-1 would link to the actin cytoskeleton via its CH domain [68–70] . Accordingly , our work strongly suggests that EHBP-1 promotes endosomal tubulation by linking PI ( 4 , 5 ) P2 enriched endosomal membranes to F-actin . Microtubules are also critical players in many different intracellular trafficking processes . Although some CH domains bind to microtubules , C . elegans tubular endosomes aligned along microtubules , and hTAC recycling in the intestine is impaired upon microtubule disruption , no interaction of the EHBP-1 CH-domain with microtubules was detected in our assays . Thus we infer that while EHBP-1 is microfilament-specific , microtubules play an important role in C . elegans CIE cargo basolateral recycling , collaborating with the microfilament cytoskeleton to shape the endosomal network [37] . EHBP-1 may promote endosome tubulation by transducing force from growing actin filaments to endosomal membranes . Alternatively , EHBP-1 may anchor endosomal membranes to the actin cytoskeleton while other forces , such as pulling by microtubule motors , acts to deform the membranes . CH-domain based actin binding structures , such as those found in alpha-actinin and spectrin , often present as a tandem arrangement of two CH domains ( CH1-CH2 ) [22 , 71] . The CH1-CH2 dimer takes on a juxtaposed conformation , with weak F-actin affinity until the dimer adopts an open conformation [72 , 73] . Utrophin and dystrophin atomic structural models suggest a theme of tandem CH-domains , with one CH domain apposed to the other CH , within the same molecule or provided by two different molecules [74 , 75] . In the current study our experiments indicated that the EHBP-1 CH domain mediates the interaction with F-actin , and suggested that the RAB-10 interaction with the EHBP-1 CC-domain somehow potentiates the CH domain-F-actin interaction . Since we did not detect a difference in F-actin binding of the EHBP-1 CH only versus CH-CC fragments , and we also did not detect binding of the CH domain to the CC domain , we do not favor an auto-inhibition model for RAB-10 mediated activation of EHBP-1 actin binding activity ( S8C Fig ) . Rather , since RAB-10 interacts with a predicted coiled-coil domain in EHBP-1 , RAB-10 binding may potentiate EHBP-1 multimerization , producing a multivalent presentation of apposed CH domains from the dimerized EHBP-1 molecules . Further analysis will be required to test this model ( S8D Fig ) . We also note that in mammalian adipocytes Rab10 and Ehbp1 are key regulators of insulin stimulated GLUT4 recycling , but their relationship has not been tested [12 , 14 , 76] . Future investigation of the Ehbp1-mediated bridging of endosomal membranes and the actin cytoskeleton in human adipocytes could prove fruitful .
All C . elegans strains were derived originally from the wild-type Bristol strain N2 . Worm cultures , genetic crosses , and other C . elegans husbandry were performed according to standard protocols [77] . Strains expressing transgenes were grown at 20°C . A complete list of strains used in this study can be found in S1 Table . RNAi was performed using the feeding method [78] . Feeding constructs were either from the Ahringer library [79] or prepared by PCR from EST clones provided by Dr Yuji Kohara ( National Institute of Genetic , Japan ) followed by subcloning into the RNAi vector L4440 [78] . For most experiments , synchronized L1 or L3 stage animals were treated for 48–72 h and were scored as adults . The following antibodies were used in this study: rabbit anti-actin polyclonal antibody ( sc-1616-R ) ( Santa Cruz Biotechnologies , Dallas , TX ) , rabbit anti-HA monoclonal antibody ( C29F4 ) ( Cell Signaling Technology , Beverly , MA ) , rabbit anti-GST monoclonal antibody ( 91G1 ) ( Cell Signaling Technology , Beverly , MA ) and rabbit anti-GFP polyclonal antibody-Chip Grade ( ab290 ) ( Abcam , Cambridge , UK ) . N-terminally hemagglutinin ( HA ) -tagged proteins , 2xHA only and RAB-10 ( Q68L ) were synthesized in vitro using the TNT-coupled transcription-translation system ( Promega , Madison , WI ) using DNA templates pcDNA3 . 1-2xHA-Gtwy and pcDNA3 . 1-2xHA-RAB-10 ( Q68L ) ( 1μg/each 50μl reaction ) , respectively . The reaction cocktail was incubated at 30°C for 90 min . Control glutathione S-transferase ( GST ) , GST-EHBP-1 ( 260-510aa ) , GST-EHBP-1 ( 1-510aa ) , GST-EHBP-1 ( 260-901aa ) and GST-hUtrophin actin binding domain ( 1-261aa ) fusion proteins were expressed in the ArcticExpress strain of Escherichia coli ( Stratagene , La Jolla , CA ) . Bacterial pellets were lysed in Lysis solution ( 50 mM HEPES pH 7 . 5 , 400 mM NaCl , 1 mM DTT , 1 mM PMSF or Complete Protease Inhibitor Cocktail Tablets ( Roche , Indianapolis , IN ) ) . Extracts were cleared by centrifugation , and supernatants were incubated with glutathione-Sepharose 4B beads ( Amersham Pharmacia , Piscataway , NJ ) at 4°C overnight . For GST pull down , beads were washed six times with cold STET buffer ( 10 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 1 mM EDTA , 0 . 1% Tween-20 ) . In vitro synthesized HA-tagged protein ( 15 μl TNT mix diluted in 500μl STET ) was added to the beads and allowed to bind at 4°C overnight . After six additional washes in STET , the proteins were eluted by boiling in 30μl 2xSDS-PAGE sample buffer . Eluted proteins were separated on SDS-PAGE ( 12% polyacrylamide ) , blotted to PVDF , and probed with anti-HA ( C29F4 ) and anti-GST ( 91G1 ) antibodies . For protein purification , beads were then washed six times with cold PBS . The bound proteins were eluted with 50 mM Tris-HCL pH 8 . 0 , and 20 mM reduced L-glutathione . Eluted GST fusion peptides were then exchanged into 20 mM HEPES-KOH pH 7 . 5 , 5 mM Mgcl2 , 1 mM EGTA , 1 mM DTT . All GST fusion proteins are centrifuged at 150 , 000x g for 1 h at 4°C prior to use in the co-sedimentation assays at the indicated molar concentrations . 3ug GST-EHBP-1 ( aa1-223 ) or GST was mixed with 10ul 1mM Control PolyPIPosomes , PI PolyPIPosomes , PI4P PolyPIPosomes , PI ( 4 , 5 ) P2 PolyPIPosomes , respectively ( Echelon Biosciences , Salt Lake City , UT ) and rotated for 15 min at room temperature in 1 ml liposome binding buffer ( 20 mM HEPES pH 7 . 5 , 150 mM NaCl , 1 mM MgCl2 ) . The mixture was centrifuged at 90 , 000xg for 15 min , collecting the supernatant . The liposome pellet was resuspended in 1 ml liposome binding buffer and centrifuged at 90 , 000xg for 15 min to wash off unspecific bound proteins , this step was repeated three times . The pellet and 20ul supernatant samples were resolved by SDS-PAGE , and GST fusion proteins were detected by western analysis using anti-GST antibody . Actin co-sedimentation assays were performed using an Actin-Binding Protein Biochem Kit: Non-Muscle Actin ( BK013 ) ( Cytoskeleton , Denver , CO ) , essentially as described by the manufacturer . Supplied α-actinin was used as a positive control . Briefly , protein preparations were incubated with 40ul freshly polymerized non-muscle actin ( 21 μM F-actin ) or F-actin buffer alone . In order to test whether RAB-10 ( Q68L ) enhances GST-EHBP-1 ( 260-901aa ) binding to F-actin , in vitro synthesized HA-RAB-10 ( Q68L ) and HA-only were added to the mixture . After incubation for 30 minutes at room temperature , samples were centrifuged at 150 , 000x g for 1 . 5 h at 24°C to pellet F-actin and the co-sedimenting proteins . Supernatants were collected on ice , and pellets were resuspended on ice for 10 min . SDS-PAGE sample buffer was added to both supernatant and pellet fractions , and the entire fractions were then resolved by SDS-PAGE gel and processed for western blot or stained with coomassie blue . GST-EHBP-1 ( 260-510aa ) , GST-EHBP-1 ( 260-901aa ) , GST-EHBP-1 ( 1-510aa ) and GST-hUtrophin ( 1-261aa ) co-sedimentations with F-actin were quantified by densitometry using FluorChem FC3 version 3 . 4 . 0 ( ProteinSimple , San Jose , CA ) . The microtubule-binding assays were performed using the Microtubule Binding Protein Spin-Down Assay Kit ( BK029 ) ( Cytoskeleton , Denver , CO ) . Microtubules were polymerized in cushion buffer ( 80 mM PIPES pH 7 . 0 , 1 mM MgCl2 , 1 mM EGTA , 60% glycerol ) for 20 min at 35°C and stabilized with taxol . GST-EHBP-1 ( 260-510aa ) and the control proteins were mixed separately with microtubules ( 50μl final volume ) , incubated at room temperature for 30 min and centrifuged at 100 , 000x g for 40 min at room temperature on top of a 100 μl of cushion buffer supplemented with taxol . All supernatants and pellets were analyzed by SDS-PAGE as described above . Worms ( 9cm plates x 10 ) were collected and washed with M9 buffer . The worm pellet was lysed by French Press in ice-cold lysis buffer ( 25 mM Tris-HCl pH 7 . 5 , 100 mM NaCl , 1 mM EDTA , 0 . 5% NP-40 , 1 mM PMSF , 1 mM Na3VO4 , 1 μg/ml Pepstatin-A and 10 mM NaF ) containing protease-inhibitor cocktail ( Sigma , St . Louis , MO ) . The lysates were incubated at 4°C for 30 min and centrifugated at 13 , 000xg for 30min . Then , supernatant was incubated with 80μl Protein A+G Agarose ( Beyotime , Shanghai , China ) for 1h at 4°C to pre-clear non-specific bead-protein interactions . 2μl anti-GFP antibody ( ab290 ) was added into pre-cleared supernatant and incubated at 4°C overnight , followed by incubation with 80μl Protein A+G Agarose ( Beyotime , Shanghai , China ) at 4°C for 4 hours . Precipitates were washed five times with lysis buffer and subjected to immunoblotting using anti-actin and anti-GFP polyclonal antibodies . rab-10 ( Q68L ) cDNA clones were transferred into an in-house modified vector pcDNA3 . 1 ( + ) ( Invitrogen , Carlsbad , CA ) with 2xHA epitope tag and Gateway cassette ( Invitrogen , Carlsbad , CA ) for in vitro transcription/translation experiments . For actin binding experiments an equivalent ehbp-1 ( 260-510aa ) , ehbp-1 ( 1-510aa ) , ehbp-1 ( 260-901aa ) and hUtrophin ( 1-261aa ) PCR product was introduced in frame into vector pGEX-2T ( GE Healthcare Life Sciences , Piscataway , NJ ) modified with a Gateway cassette . To construct GFP or RFP fusion transgenes for expression specifically in the worm intestine , a previously described vha-6 promoter-driven vector modified with a Gateway cassette inserted just upstream of the GFP or red fluorescent protein ( tagRFP-T ) coding region was used . The sequences of C . elegans ehbp-1 ( cDNA ) , ehbp-1 ( aa1-223 ) , ehbp-1 ( aa1-223 ) ( RRLRR6AALAA ) , ehbp-1 ( aa1-223 ) ( RR6AA ) , ehbp-1 ( aa1-223 ) ( RR9AA ) , ehbp-1 ( aa1-223 ) ( KK13AA ) , ehbp-1 ( aa1-223 ) ( HRRRK46AAAAA ) , ehbp-1 ( 260-510aa ) , ehbp-1 ( 260-901aa ) and ehbp-1 ( 1-259aa , 511-901aa ) lacking a stop codon were cloned individually into entry vector pDONR221 by PCR and BP reaction , and then transferred into intestinal expression vectors by Gateway recombination cloning LR reaction to generate C-terminal fusions [6] . Integrated transgenic lines for all these plasmids were obtained by microinjection or microparticle bombardment . Nocodazole ( 50 μg/mL , M1404 ) ( Sigma , St . Louis , MO ) or Latrunculin B ( 10 μM , sc-203318 ) ( Santa Cruz Biotechnologies , Dallas , TX ) was injected into the pseudocoelom of young adult worms 2 h before imaging . Drugs were diluted in DMSO and used at a final concentration of 1% DMSO in egg buffer [118 mM NaCl , 48 mM KCl , 2 mM MgCl2 , 2 mM CaCl2 , and 25 mM HEPES ( pH 7 . 3 ) ] . Live worms were mounted on 2% agarose pads with 10 mM levamisole . Multi-wavelength fluorescence images were obtained using an FLUOVIEW FV1000 microscope ( Olympus , Tokyo , Japan ) and captured using FV10-ASW Ver . 3 . 1 software . Images taken in the DAPI channel were used to identify broad-spectrum intestinal autofluorescence caused by lipofuscin-positive lysosome-like organelles . Fluorescence images were obtained using an FV1000-IX81 confocal laser scanning microscope ( Olympus , Tokyo , Japan ) equipped with a 60×N . A . 1 . 2 oil-immersion objective . Z series of optical sections were acquired using a 0 . 5μm step size . Dynamic fluorescence imaging was performed on a spinning-disk confocal imaging system ( CSU-X1 ) ( Yokogawa , Tokyo , Japan ) equipped with an EM CCD camera ( iXon DU897K ) ( ANDOR , Belfast , UK ) ) and oil-immersion objectives ( 60×N . A . 1 . 45 ) . A 50 mW solid state lasers ( 491 nm ) coupled to an acoustic-optical tunable filter ( AOTF ) were used to excite GFP . hTAC-GFP labeled endosomes dynamic images were obtained over 180–240 sec with an exposure every 1 sec . To compare the subcellular distribution of GFP-tagged proteins , fluorescence data from GFP channel were analyzed by Metamorph software version 7 . 8 . 0 . 0 ( Universal Imaging , West Chester , PA ) . The “Integrated Morphometry Analysis” function of Metamorph was used to detect the fluorescent structures that are significantly brighter than the background and to measure total puncta number ( referred as “structure count” ) and total fluorescence area ( referred as “total area” ) within unit regions . From total 6 animals of each genotype , “structure count” and “total area” were sampled in three different unit regions of each intestine defined by a 100 x 100 ( pixel2 ) box positioned at random ( n = 18 each ) . In most cases , “total area” was used to compare tubularity , as the normal endosomal tubule network covers much more area than when the network collapses into puncta . Another parameter “structure count” was also sometimes used to assay this aspect , where the structure count increases as the network breaks down into puncta . GFP and RFP-tagged proteins colocalization analysis were performed using “Measure colocalization” App of Metamorph software . After thresholding , the percentage of GFP fluorescence area ( area A ) overlapping with RFP fluorescence area ( area B ) in eighteen intestinal unit regions ( 3 regions per animal ) was analyzed for each genotype . Most GFP/RFP colocalization experiments were performed on L3 and L4 larvae expressing GFP and RFP markers . To establish a quantitative index for the vacuole phenotype , total vacuole number and size were quantified in three intestinal cells of 6 ehbp-1 ( tm2523 ) mutant or rescue animals ( n = 18 per genotype ) using endosome visualization marker ARF-6-RFP . Vacuoles were classified into 3 different size groups: small vacuole ( diameter <5 μm ) , medium vacuole ( diameter 5–10 μm ) and large vacuole ( diameter >10 μm ) .
|
Endosomes are intracellular organelles that sort protein and lipid components integral to the membrane , as well as more loosely associated lumenal content , for delivery to distinct intracellular destinations . Endosomes associated with recycling cargo back to the plasma membrane are often tubular in morphology , and this morphology is thought to be essential for recycling function . Our previous work identified a particularly dramatic network of endosomal tubules involved in membrane protein recycling in the basolateral intestinal epithelial cells of C . elegans . Our subsequent genetic analysis of basolateral recycling in this system identified a number of key regulators of these endosomes , including the small GTPase RAB-10 and its effector EHBP-1 . Our new work presented here shows that EHBP-1 promotes endosomal tubulation by linking the membrane lipid PI ( 4 , 5 ) P2 to the actin cytoskeleton , and that the linkage of EHBP-1 to actin is enhanced by the interaction of EHBP-1 with RAB-10 . This work has broad implications for how endosomal tubulation occurs in all cells , and has specific implications for the role of EHBP-1 in related processes such as insulin-stimulated recycling of glucose transporters in human adipocytes , a process intimately linked to type II diabetes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"vesicles",
"actin",
"filaments",
"vacuoles",
"microtubules",
"cellular",
"structures",
"and",
"organelles",
"cytoskeleton",
"digestive",
"system",
"endosomes",
"contractile",
"proteins",
"actins",
"proteins",
"cell",
"membranes",
"gastrointestinal",
"tract",
"biochemistry",
"cytoskeletal",
"proteins",
"cell",
"biology",
"anatomy",
"protein",
"domains",
"biology",
"and",
"life",
"sciences"
] |
2016
|
RAB-10 Promotes EHBP-1 Bridging of Filamentous Actin and Tubular Recycling Endosomes
|
Echinococcus multilocularis is the source of alveolar echinococcosis , a potentially fatal zoonotic disease . This investigation assessed the presence of E . multilocularis infection in definitive hosts in the Chenaran region of Razavi Khorasan Province , northeastern Iran . Fecal samples from 77 domestic and stray dogs and 14 wild carnivores were examined using the flotation/sieving method followed by multiplex PCR of mitochondrial genes . The intestinal scraping technique ( IST ) and the sedimentation and counting technique ( SCT ) revealed adult Echinococcus in the intestines of five of 10 jackals and of the single wolf examined . Three jackals were infected only with E . multilocularis but two , and the wolf , were infected with both E . multilocularis and E . granulosus . Multiplex PCR revealed E . multilocularis , E . granulosus , and Taenia spp . in 19 , 24 , and 28 fecal samples , respectively . Echinococcus multilocularis infection was detected in the feces of all wild carnivores sampled including nine jackals , three foxes , one wolf , one hyena , and five dogs ( 6 . 5% ) . Echinococcus granulosus was found in the fecal samples of 16 . 9% of dogs , 66 . 7% of jackals , and all of the foxes , the wolf , and the hyena . The feces of 16 ( 21 . 8% ) dogs , 7 of 9 ( 77 . 8% ) jackals , and all three foxes , one wolf and one hyena were infected with Taenia spp . The prevalence of E . multilocularis in wild carnivores of rural areas of the Chenaran region is high , indicating that the life cycle is being maintained in northeastern Iran with the red fox , jackal , wolf , hyena , and dog as definitive hosts .
Echinococcus multilocularis is the agent of alveolar echinococcosis , a potentially fatal zoonotic disease [1] , [2] . The life cycle of E . multilocularis is sylvatic; adult worms are found in wild carnivores , principally foxes , and in the raccoon dog , wolf , coyote , and jackal , while their metacestodes develop in small mammals , predominantly rodents such as Cricetidae , Arvicolidae , and Muridae [3] , [4] , [5] , [6] , [7] . In some rural areas domestic dogs and sometimes cats can be definitive hosts after acquiring the infection from wild rodents , and thus become a major zoonotic risk for infecting humans [3] , [8] , [9] . Humans can serve as an aberrant intermediate host for E . multilocularis , with transmission occurring through direct contact with the definitive host or by ingestion of contaminated water , vegetables , or other foods [10] . Human alveolar echinococcosis is a lethal zoonotic disease caused by infection with the multivesiculated metacestode of E . multilocularis [11] , [12] . The geographical distribution of the parasite is restricted to the northern hemisphere . The cestode has been reported in areas of central Europe , the Near East , Russia , central Asian republics , northern Japan , and Alaska [11] , [13] , [14] , [15] . In the Middle East , cystic echinococcosis is prevalent in most countries , although a low prevalence of alveolar echinococcosis is reported in Iran , Iraq , and Tunisia [16] . In Asia , canine infections have been recorded in dogs in China , Kazakhstan and Kyrgyzstan [17] , [18] , [19] . Echinococcus multilocularis in canids was reported in northwestern Iran for the first time in 1971 [20] , [21] . Further investigation in 1992 found its infection in 22 . 9% of red foxes ( Vulpes vulpes ) and 16% of jackals ( Canis aureus ) [22] . The latest research in 2009 reported no evidence of E . multilocularis infection in canids of the Moghan Plain in northwest Iran [23] . The majority of previous research focused only on the northwestern part of the country [20] , [21] , [22] , [23] . Alveolar echinococcosis , based on histopathological and clinical data , was first reported in a village in Chenaran County of Razavi Khorasan Province in 2007 [24] . The disease was subsequently confirmed by molecular evaluation , and a second case reported ( E . Razmjou , unpublished data ) . Razavi Khorasan Province is located in northeastern Iran , near the border with Turkmenistan , where E . multilocularis is endemic . A pilot study revealed that suitable hosts such as foxes , jackals , dogs , wolves , and rodents are frequently present near villages in the mountains of Chenaran region . The presence in Razavi Khorasan of suitable conditions for completing the life cycle of this parasite , such as presence of definitive and intermediate hosts in mountainous areas and proximity to other countries where the parasite is endemic , led to the present study to assess the prevalence of E . multilocularis infection in carnivores , and to identify natural definitive hosts of this life-threatening parasite in the Chenaran region of Razavi Khorasan Province , Iran .
Animals were shot under license from the Iran Environment Protection Organization , solely for the purpose of investigating the presence of Echinococcus multilocularis in wild carnivores . The Protocol of this investigation was reviewed and approved by the Ethics Committee of Tehran University of Medical Sciences . The study area , the Chenaran region , covers approximately 2400 km2 in northeastern Iran , 55 kilometers northwest of Mashhad ( Figure 1 ) ( 36°4′N , 59°7′E ) . It lies between the Binalood Heights and the Hezar Masjed Mountains . Chenaran city is surrounded by rural areas , mainly consisting of human habitations , gardens , farms , and moorland . The region has cold and snowy winters and mild summers . The average annual temperature is 13 . 4°C with variable rainfall; mean annual precipitation of 212 . 6 mm . It is rich in wildlife , including carnivores and small rodents appropriate for supporting the life cycle of E . multilocularis . From November 2009 to January 2010 , fecal samples from 77domestic and stray dogs from 17 villages and the entire gut of three foxes , ten jackals , and one wolf ( Canis lupus pallipes ) , either shot or killed accidentally , were collected . In addition , during October and November 2010 , the intestines of one fox and one hyena ( Hyena hyena ) killed on roads were added to our samples . A standard form including place and date of killing was completed . Fecal samples were collected from the rectum of each wild carnivore . The intestine and fecal samples were placed in labeled ziploc bags , stored at −80°C for at least seven days [25] to reduce the risk of laboratory infection by inactivating any Echinococcus oncospheres and other infective materials , and subsequently stored at −20°C until further examination . The intestinal scraping technique was performed as described by Deplazes and Eckert [25] . The intestine was opened full length and after removal of undigested food and visible parasites from the proximal , middle , and posterior parts of the small and large intestine , 15 deep mucosal scrapings were taken using microscope slides . Material adhering to the slides was transferred to plastic Petri dishes and examined stereomicroscopically at 120× magnification . Echinococcus worms were isolated and stored in 85% ethanol for molecular examination and in 10% formalin for morphological diagnosis . The sedimentation and counting technique was done as previously described [26] . The intestine was cut into 10 cm pieces and each was placed in a flask containing one liter of 0 . 9% saline . After vigorous shaking for a few seconds , the pieces of intestine were pressed firmly between the fingers using gloves and with care to avoid contamination with eggs to remove attached worms . The supernatant was decanted and the procedure was repeated several times with physiological saline . The sediment was placed in plastic Petri dishes and examined stereomicroscopically at 120× magnification . All isolated worms , including Echinococcus , were stored in 85% ethanol and 10% formalin for molecular and microscopic identification , respectively . Fecal samples were submitted to flotation with zinc chloride for isolating parasite eggs . Each sample ( 4–5 g ) was stirred into 50 ml distilled water until completely dispersed . The suspension was passed through four layers of gauze and large particles removed . The suspension was transferred into a 50 ml Falcon tube and centrifuged at 1000×g for 5 min . For isolating eggs , zinc chloride solution ( specific gravity 1 . 45 g ml−1 ) was added to sediment up to a final volume of 12 ml and , after complete mixing , centrifuged at 1000×g for 30 min [27] . The supernatant was passed through sequential sieves on 50 ml falcon tubes with metal and polystyrene screens of mesh sizes 37 and 20 µm , respectively [27] . The sieves were inverted and washed thoroughly with distilled water containing 0 . 2% Tween 20 . After adding phosphate-buffered saline ( PBS; pH 7 . 2 ) to a final volume of 50 ml , suspensions were centrifuged at 1000×g for 30 min , the supernatant fraction was aspirated , and sediment ( approximately 400 µl ) was transferred to 1 . 5 ml tubes and stored at −20°C until further examination . Detection of Echinococcus spp . was based on morphological characteristics . Adult worms of E . multilocularis and E . granulosus were differentiated using morphological characteristics including size , length of gravid proglottids , shape of the uterus , number of eggs per proglottid , and position of genital pore after acetic acid alum carmine staining and mounting in Canada balsam . DNA of adult worms and taeniid eggs was extracted using the QIAamp DNA Mini kit ( Qiagen , Germany ) according to the protocol of Verweij et al . [28] with slight modifications . Briefly , one Echinococcus worm was removed from 85% ethanol and washed in sterile PBS buffer three times . The worm was then placed in 200 µl of PBS buffer and , after 10 min boiling at 100°C , an equal volume of ATL buffer plus 10% proteinase K was added and completely mixed and incubated two hours at 55°C in heat block . DNA extraction was continued according to manufacturer's instructions with the minor modification of increasing incubation time to five minutes to increase the yield of DNA in the final step . DNA was stored at −20°C until analysis . Before submitting the eggs from fecal samples to the DNA extraction procedure described for adult worms , they were subjected to seven freeze/thaw cycles , using liquid nitrogen and boiling water , to disrupt the egg wall . Then , 200 µl of the sample was heated at 100°C for 10 min as in Verweij's procedure [28] . The concentration of extracted DNA was measured spectrophotometrically by Biophotometer ( Biophotometer Plus , Eppendorf , Germany ) . Multiplex PCR of adult worms and eggs was performed as described [29] . The mitochondrial multiplex reaction was designed to amplify a 395 bp fragment of NADH dehydrogenase subunit 1 ( nad1 ) of E . multilocularis and 117 bp and 267 bp of a small subunit of ribosomal RNA ( rrnS ) of E . granulosus and other Taenia spp . , respectively . Primers , conditions , and parameters for PCR were as previously described [29] . All samples were tested in 25 µl amplification reaction mixtures with 12 . 5 µl of the master mix ( QIAGEN Multiplex PCR , Germany ) , 2 . 5 µl of primers ( 2 µM of primers Cest1 , Cest2 , Cest3 , Cest4 and 16 µM of primer Cest5 in H2O ) , 8 µl H2O , and 2 µl of template DNA . Initially , multiplex PCR was confirmed with standard DNA of E . multilocularis , E . granulosus , Taenia multiceps , and T . hydatigena provided by Professor Deplazes of the Institute of Parasitology of Zurich , Switzerland . Additionally , for all PCR reactions one negative control without DNA and one positive control with standard DNA were included to confirm the results of multiplex PCR . Finally , 10 µl of the PCR products were loaded on 2% ( W/V ) agarose gels , and stained with ethidium bromide to visualize by electrophoresis . The results of multiplex PCR were confirmed by single PCR using the primer pair Cest1/Cest2 and Cest4/Cest5 for E . multilocularis and E . granulosus , respectively [29] , and EM-H15/EM-H17 [9] for E . multilocularis . All E . multilocularis PCR-positive samples were confirmed by sequencing of a 395 bp amplified fragment . PCR products were excised from agarose gels and purified using the QIAquick Gel Extraction Kit ( QIAgen , Germany ) , according to the manufacturer's instructions . Products were sequenced in both directions using the Cest1/Cest2 primers by MilleGen Company ( France ) . Sequences were read by CHROMAS ( Technelysium Pty Ltd . , Queensland , Australia ) and aligned using the DNASIS MAX ( version 2 . 09; Hitachi , Yokohama , Japan ) software program .
The entire small intestine of 16 wild carnivores comprising 10 jackals , four foxes , one hyena , and one wolf were examined by both IST and SCT . These techniques found 6 of 16 ( 37 . 5%; 95% CI: 18 . 5%–61 . 4% ) wild canids to be infected with Echinococcus spp . The worms were isolated from five of 10 ( 50% , 95% CI: 23 . 7–76 . 3% ) jackals and one wolf , while the remaining jackals , the foxes , and the hyena tested negative . The intensity of infection was classified as low ( 1–100 ) , medium ( 101–1000 ) , or high ( >1000 ) worm burden [30] . All positive jackals showed a high Echinococcus worm burden , while the wolf had a low burden . We differentiated all Echinococcus worms by microscopic examination ( Figure 2 ) . Among six Echinococcus positive samples , three jackals ( 30% ) had a single species infection with E . multilocularis but two jackals ( 20% ) and the wolf were infected with both E . multilocularis and E . granulosus . To detect eggs , fecal samples of dogs were investigated by direct microscopic examination and the flotation method . Eggs were observed in 13 of 77 ( 16 . 9% , 95% CI: 10 . 1–26 . 8% ) dog fecal samples . The result of multiplex PCR by amplification of 395 bp fragment of nad1 indicated that 19 of the carnivores were infected with E . multilocularis . The 117 bp fragment of rrnS identified E . granulosus in 24 , and the 267 bp fragment found Taenia spp . in 28 of the fecal samples . Echinococcus multilocularis infection was detected in the feces of all wild carnivores ( 100%; 95% CI: 78 . 5–100% ) including nine jackals , three foxes , one wolf and one hyena , and five dogs ( 6 . 5%; 95% CI: 2 . 8–14 . 3% ) . Echinococcus granulosus was found in the fecal sample of 16 . 9% ( 95% CI: 10 . 1–26 . 8% ) of dogs , 66 . 7% ( 95% CI: 35 . 4–88 . 0% ) of jackals , and all of the foxes , the wolf , and the hyena . The feces of 16 of 77 ( 21 . 8%; 95% CI: 13 . 2–31 . 1% ) dogs , 7 of 9 ( 77 . 8%; 95% CI: 45 . 3–94 . 0% ) jackals , and all three foxes , one wolf and one hyena were infected with Taenia spp . ( Table 1 ) ( Figures 3 , 4 ) . Among 26 PCR positive dog samples , a single DNA fragment amplified in 18 samples indicated two E . multilocularis ( 2 . 6%; 95% CI: 0 . 7–9 . 0% ) , six E . granulosus ( 7 . 8%; 95% CI: 3 . 6–16 . 0% ) , and ten Taenia spp . ( 13%; 95% CI: 7 . 2–22 . 3% ) infected dogs . Two species-specific fragments in seven cases showed five dogs ( 6 . 5%; 95% CI: 2 . 8–14 . 3% ) to be co-infected with E . granulosus and Taenia spp . , and two E . multilocularis infected dogs were also infected with Taenia spp . ( 1 . 3%; 95% CI: 0 . 2–7 . 0% ) or E . granulosus ( 1 . 3%; 95% CI: 0 . 2–7 . 0% ) . Three amplicons revealed one dog ( 1 . 3%; 95% CI: 0 . 2–7 . 0% ) infected simultaneously with Taenia spp . and two species of Echinococcus . Of 14 E . multilocularis infections in wild carnivorous , 11 showed triple infections including six jackals ( 66 . 7% ) , three foxes ( 100% ) , the hyena and the wolf . One jackal was co-infected with Taenia spp . , and two jackals were infected with E . multilocularis ( Table 2; Figure 4 ) . The prevalence of E . multilocularis infection was high in wild carnivores ( 100%; 95% CI: 78 . 5–100% ) , whereas the rate of infection in domestic and stray dogs was low ( 6 . 5%; 95% CI: 2 . 8–14 . 3% ) . The results of multiplex PCR were confirmed with single PCR . Sequencing results of amplicons obtained from worms and eggs were identified as E . multilocularis . The nucleotide sequences of the amplified nad1 were equivalent to positions 7645 to 8040 of the published E . multilocularis mitochondrion complete genome ( accession no . AB018440 ) . Sequences were aligned using DNASIS MAX ( version 2 . 09; Hitachi , Yokohama , Japan ) with the published reference sequences . Analysis revealed 100% identity between our isolates and the corresponding published reference sequences for E . multilocularis . The nucleotide sequence data reported in this paper will appear in the DDBJ/EMBL/GenBank nucleotide sequence databases with the accession numbers AB617846–AB617855 and AB621793–AB621801 .
Iran is located in the Middle East and central Eurasia . It is bordered on the north by Armenia , Azerbaijan , Turkmenistan and Caspian Sea . Afghanistan and Pakistan are Iran's direct neighbors in the east . The country borders the Persian Gulf and the Gulf of Oman in the south , Iraq in the west , and Turkey to the north-west ( Figure 1 ) . In the Middle East , echinococcosis is one of the most important zoonotic diseases [16] . Cystic echinococcosis ( CE ) and alveolar echinococcosis ( AE ) have been reported in the Mediterranean region , but CE is more prevalent [31] , [32] . Although Iran is an endemic area for echinococcosis , most studies have been carried out only on E . granulosus . Mobedi and Sadighian in 1971 reported E . multilocularis for the first time in three of 30 red foxes tested from northwest Ardebil Province [20] , [21] . A study of E . multilocularis in carnivores of that area was followed in 1992 by Zariffard and Massoud [22] showing comparable results , infection in 22 . 9% ( 16/70 ) of red foxes and 16% ( 4/25 ) of the jackals . In a study in 2009 Zare-Bidaki et al . did not observe E . multilocularis in the investigated canids [23] . With the exception of these three studies , E . multilocularis infection has not been investigated outside of the northwest part of the country . Although there is limited information about AE in Iran , Torgerson et al . [10] suggested that Iran , since it is bordered by highly endemic countries , is an endemic area for E . multilocularis , and that the estimated annual incidence of eleven cases of AE are likely underreported . Molecular confirmation of some AE reports from inhabitants of a village in Chenaran County of northern Razavi Khorasan Province [24] ( E . Razmjou , unpublished data ) , which neighbors a hyperendemic country , Turkmenistan , led to our investigation of the establishment of E . multilocularis' life cycle in this area . Assessment of the occurrence of E . multilocularis in definitive hosts showed that this cestode has a high prevalence in the wild carnivores of the Chenaran area in northeastern Iran ( 100%; 95% CI: 78 . 5–100% ) . In comparison to the prevalence of infection in foxes in Belgium ( 24 . 55% ) [33] , Switzerland ( 47%–67% ) [26] , Ukraine ( 36% ) [34] , Kyrgyzstan ( 64% ) [35] , and Japan ( 49% ) [36] , it is assumed that Razavi Khorasan Province is hyperendemic for this tapeworm . The significantly lower prevalence of E . multilocularis infection in dogs ( 6 . 5% ) than in other carnivores ( 100% ) may be due to decreased ingestion of metacestode infected intermediate hosts through controlled and limited diet of domestic dogs . As the multiplex PCR is subject to amplification of DNA from taeniid eggs , the results could also be associated with overlooking some positive cases due to prepatent infections , the intermittent egg excretion during the patent period [29] , or degradation of taeniid eggs in the environment and unsuitable conditions such as solar radiation [37] . The rate of infected dogs in our study ( 6 . 5% ) was less than reported in China ( 12% ) [38] and Kyrgyzstan ( 18% ) [19] . On the other hand , the infection rate observed in our study was considerably greater than prevalence reported in Germany ( 0 . 24% ) [4] and Lithuania ( 0 . 8 ) [39] . The low rates found there may have been a consequence of conducting PCR only on positive taeniid egg samples by the sieving/flotation method . Our experiment showed Taenia eggs in only 16 . 9% of the 77 dog feces by sieving/flotation and microscopic examination . This increased to 33 . 9% with multiplex PCR on the flotation material . It assumed that results might be related to difficulty in detecting small numbers of eggs by microscopic examination , while DNA of a single taeniid egg from a sieved fecal sample can be amplified by the multiplex PCR [29] . While the prevalence of infection in dogs is low , the large population of domestic and stray dogs in the villages that are in close contact with inhabitants must be considered a potential source and risk factor for transmission of E . multilocularis to humans . The results of multiplex PCR in the current study showed that most of the wild ( 78 . 6% ) and some domestic ( 2 . 6% ) canids were co-infected with E . granulosus and E . multilocularis . Previous studies have reported a high rate of CE in livestock and humans in this province [32] . Consequently , Razavi Khorasan should be considered an endemic area for both E . granulosus and E . multilocularis infection . Razavi Khorasan Province is a tourist area , and many travelers are at the risk of exposure to these zoonotic diseases . The province has extensive agriculture and export of fruits to other parts of the country . For these reasons and because of close contact of humans with infected domestic dogs and other definitive hosts that forage for food on farms and gardens of this region , it is important to initiate intensive health initiatives . In conclusion , our study confirms the presence of the life cycle of E . multilocularis in the Chenaran region of northeastern Iran . In this cycle the red fox , jackal , wolf , hyena , and dog play the role of the definitive host , and efforts are underway to elucidate the intermediate host of this parasite . Since the current survey is the first evidence of existence of E . multilocularis in domestic and wild animals in northeastern Iran , further studies should be conducted to investigate the presence of E . multilocularis in other parts of the country .
|
Echinococcus multilocularis causes alveolar echinococcosis , a serious zoonotic disease present in many areas of the world . The parasite is maintained in nature through a life cycle in which adult worms in the intestine of carnivores transmit infection to small mammals , predominantly rodents , via eggs in the feces . Humans may accidentally ingest eggs of E . multilocularis through contact with the definitive host or by direct ingestion of contaminated water or foods , causing development of a multivesicular cyst in the viscera , especially liver and lung . We found adult E . multilocularis in the intestine and/or eggs in feces of all wild carnivores examined and in some stray and domestic dogs in villages of Chenaran region , northeastern Iran . The life cycle of E . multilocularis is being maintained in this area by wild carnivores , and the local population and visitors are at risk of infection with alveolar echinococcosis . Intensive health initiatives for control of the parasite and diagnosis of this potentially fatal disease in humans , in this area of Iran , are needed .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"biology",
"microbiology",
"zoology",
"parasitology",
"helminthology"
] |
2011
|
Detection of Echinococcus multilocularis in Carnivores in Razavi Khorasan Province, Iran Using Mitochondrial DNA
|
There is a strong need for computational frameworks that integrate different biological processes and data-types to unravel cellular regulation . Current efforts to reconstruct transcriptional regulatory networks ( TRNs ) focus primarily on proximal data such as gene co-expression and transcription factor ( TF ) binding . While such approaches enable rapid reconstruction of TRNs , the overwhelming combinatorics of possible networks limits identification of mechanistic regulatory interactions . Utilizing growth phenotypes and systems-level constraints to inform regulatory network reconstruction is an unmet challenge . We present our approach Gene Expression and Metabolism Integrated for Network Inference ( GEMINI ) that links a compendium of candidate regulatory interactions with the metabolic network to predict their systems-level effect on growth phenotypes . We then compare predictions with experimental phenotype data to select phenotype-consistent regulatory interactions . GEMINI makes use of the observation that only a small fraction of regulatory network states are compatible with a viable metabolic network , and outputs a regulatory network that is simultaneously consistent with the input genome-scale metabolic network model , gene expression data , and TF knockout phenotypes . GEMINI preferentially recalls gold-standard interactions ( p-value = 10−172 ) , significantly better than using gene expression alone . We applied GEMINI to create an integrated metabolic-regulatory network model for Saccharomyces cerevisiae involving 25 , 000 regulatory interactions controlling 1597 metabolic reactions . The model quantitatively predicts TF knockout phenotypes in new conditions ( p-value = 10−14 ) and revealed potential condition-specific regulatory mechanisms . Our results suggest that a metabolic constraint-based approach can be successfully used to help reconstruct TRNs from high-throughput data , and highlights the potential of using a biochemically-detailed mechanistic framework to integrate and reconcile inconsistencies across different data-types . The algorithm and associated data are available at https://sourceforge . net/projects/gemini-data/
The inference of transcriptional regulatory networks ( TRNs ) from high-throughput data is a central challenge in systems biology . TRN models provide a mechanistic framework for describing interactions between transcription factors and their target genes . Cellular phenotypes are influenced by the differential activity of these networks , and reconstructing the regulatory network enables one to understand the underlying molecular processes that cause phenotypic changes and better predict the response of a cell to an external perturbation . Current network inference algorithms enable rapid reconstruction of TRNs by utilizing high-throughput data such as protein-DNA binding , DNA sequence or gene expression [1]–[12] . However , the overwhelming number of possible regulatory interactions between thousands of genes and transcriptional regulators in a cell—combined with the complex and dynamic nature of these interactions—limits the success of these inference approaches [1] , [13] , [14] . Recent analyses in Saccharomyces cerevisiae ( baker's yeast ) have shown that even though there are a multitude of predicted interactions , very few have a functional effect on the pathway activity or the metabolic flux distributions [13] , [15] . Furthermore , a large-scale comparative study of expression-based network inference algorithms found poor performance in yeast [1] . One reason for this is that a connection to a growth or metabolic phenotype is missing during the inference process , making it difficult to assess the plausibility of the predicted interactions in a systems context . Connecting TRN inference to the phenotype data can lead to a more seamless connection between genomic measurements and phenotype . We hypothesized that integrating regulatory interactions with metabolic networks would make it possible to more directly connect the regulatory interactions with their downstream phenotype , and thus allow us to use a broader range of data for network curation . Genome-scale models of metabolic networks have been constructed using growth phenotype data for a wide range of organisms , and these models accurately predict the response of the cell to environmental and genetic perturbations [16]–[19] . These models explicitly represent the mechanistic relationships between genes , proteins , and the chemical inter-conversion of metabolites within a biological system . The success of this integration would then allow the utilization of large-scale phenotypic data , which are commonly used to curate metabolic networks [16] , [20] , [21] , to also refine regulatory interactions . To enable the concurrent analysis of transcriptional regulation and metabolism , we recently developed the Probabilistic Regulation of Metabolism ( PROM ) approach for integrating biochemical networks with TRNs in an automated fashion [22] . We used PROM to demonstrate that phenotypic states can be predicted from the combined TRN and metabolic network models . PROM takes in a genome-scale metabolic network model , a regulatory network structure consisting of TFs and their targets , and gene expression data across different conditions , as inputs to predict the phenotypic outcome of transcriptional perturbations . PROM solves the forward problem of combining disparate networks to predict phenotype ( e . g . , flux and growth rates ) . In the work described herein , we iteratively use PROM to aid in solving the more challenging inverse problem [23]—guiding TRN structure prediction using the metabolic network and the emergent phenotype measurements . In doing so , our new method serves as a tool to refine the inferred TRN and improve the predictive power of the integrated network models . This new approach , Gene Expression and Metabolism Integrated for Network Inference ( GEMINI ) , discerns functional regulatory interactions in high-throughput data by taking advantage of PROM , the growing amount of information in phenotype databases , and the observation by Barrett et al [24] that only a fraction of functional regulatory network states are compatible with a viable metabolic network . GEMINI produces a regulatory network state that is simultaneously consistent with observed gene knockout phenotypes , gene expression data , and the corresponding metabolic network state . While there have been approaches to model the constraints imposed by regulation and signaling networks on metabolism [18] , [22] , [25] , [26] or to readjust manually curated regulatory rules based on metabolism [18] , [27] , no method thus far have utilized metabolic constraints to refine high-throughput interaction data as GEMINI does . Here we describe the GEMINI approach and then test it by building a genome-scale integrated model for yeast . We compare the refined network model across various high-throughput data sets , and demonstrate that GEMINI effectively recalls known mechanistic interactions . We then iteratively expand and refine the integrated model using published genome-wide chromatin immunoprecipitation , TF knockout gene expression and binding-site-motif data sets , and show the ability of our integrated metabolic and regulatory network model to predict growth phenotypes of transcription factor knockout strains in new conditions . We also use GEMINI to identify potential condition-specific interactions and post-transcriptional regulatory mechanisms in S . cerevisae .
GEMINI takes in a draft regulatory network and integrates it with the corresponding metabolic network and gene expression data using PROM . PROM uses conditional probabilities , viz . the probability of a given gene being ON or OFF when the regulating transcription factor is ON or OFF , to represent gene states and gene–transcription factor interactions . The ON/OFF state of the TFs is then used to determine the likelihood of an ON/OFF state of the target genes based on the probabilities estimated from microarray data . PROM then utilizes the Gene-Protein-Reaction ( GPR ) relationships present in the metabolic network models to connect the regulatory targets to the corresponding metabolic reactions . The GPRs take into account the presence of isozymes or multi-gene/multi-subunit complexes that may be involved in catalyzing each metabolic reaction . The probabilities are then used to constrain the fluxes through the metabolic network ( detailed below ) , and an optimal state of the network that satisfies topological and transcriptional constraints is determined . Using this integrated metabolic-regulatory network , PROM can simulate metabolic phenotypes under different conditions using Flux Balance Analysis ( FBA ) [28] . FBA identifies the optimal state of the metabolic network that would allow the system to achieve a particular objective , typically the maximization of an organism's growth rate or biomass production . Mathematically , FBA is framed as a linear programming problem: ( 1 ) ( 2 ) ( 3 ) where i is the set of metabolites , j the set of reactions in the network , Sij is the stoichiometric matrix , cj designates the objective function ( the cellular growth rate in this case ) and vj is the flux through reaction j . PROM finds a flux distribution that satisfies these physico-chemical constraints plus additional constraints to account for the transcriptional regulation [22]: ( 4 ) subject to constraints ( 5 ) ( 6 ) ( 7 ) where lb' and ub' are constraints based on transcriptional regulation and are estimated based on the probabilities . Vmax and Vmin are the systemic maximum and minimum fluxes through a reaction and are determined using Flux Variability Analysis ( FVA ) [29] . α and β represent the deviation from those constraints ( determined by the algorithm for each reaction ) , and κ represents the penalty for such deviations . The higher the value of κ , the greater the transcriptional regulation constraint is on the system . The value of κ is determined in a data-driven manner ( See Methods ) . Once the initial PROM model is built , GEMINI then performs in silico knockouts of each TF in the integrated model and compares the predictions with experimental observations . GEMINI identifies and removes interactions that do not lead to the measured growth phenotype , while retaining the phenotype-consistent interactions . This is achieved by comparing the flux state predicted by PROM for the TF knockout ( v1 ) with the closest flux state that represents the measured growth phenotype ( v2 ) . The flux state v2 is obtained by forcing the model to match the observed phenotype , while still attempting to satisfy as many of the transcriptional constraints as possible . Mathematically , we solve the same constraints as above with the additional constraint that the predicted growth phenotype matches the observed phenotype ( See Methods ) . Unlike mass balance or thermodynamic constraints that cannot be violated , PROM imposes “soft” constraints on the system due to transcriptional regulation , thereby enabling us to force the model to match the measured phenotype . This procedure results in a flux solution that is geometrically closest to the flux state v1 , based on absolute distance , while still satisfying the observed growth phenotype . We then compare the new flux state v2 with the original flux state v1 , and prioritized reactions regulated by the perturbed TF based on their magnitude of change . Interactions regulating these reactions were removed consecutively and PROM is run on each new network to predict the growth phenotype . This process is repeated until the inconsistency is resolved ( Figure 1 ) . We demonstrate the GEMINI approach using the model organism Saccharomyces cerevisiae . Because of the availability of a large amount of data about regulatory interactions , a vast amount of gene expression and phenotype data , and the existence of a well-curated genome-scale metabolic model for yeast , this organism makes an ideal test case for GEMINI . Most importantly , highly accurate inference of regulatory interactions has been a major challenge in yeast as it is a more complex system than bacterial model organisms such as Escherichia coli [1] , [30] . To apply our approach to yeast , we downloaded transcriptional regulatory interactions from the Yeastract database [31] , which were compiled from various literature sources . These Yeastract interactions have a high-confidence subset ( direct/gold-standard interactions ) for which strong experimental evidence ( supporting the interaction of the TF with the promoter of the specified target gene ) is available [31] . This gold-standard subset is commonly used as a benchmark for validating inference algorithms [1] . This dataset allowed us to test our hypothesis that metabolic phenotype-consistency can be used as a criterion for improving the identification of functional regulatory interactions . The effectiveness of GEMINI was evaluated by measuring its ability to differentiate between the validated direct interactions and the remaining low-confidence interactions ( putative/potential interactions ) , which were inferred using motif search algorithms [32] . It should be noted that the gold-standard interactions are not necessarily perfect and may contain false-positive interactions [1]; similarly , the low-confidence interactions could be either false-positives or true interactions that have not been validated yet . However , on average , the gold-standard interactions have stronger supporting evidence from ChIP-binding or directed mutagenesis—giving them a higher probability of being true than the lower confidence set . According to our hypothesis , gold-standard interactions are more likely to be consistent with phenotype data than the potential interactions . With an unlabeled list of Yeastract interactions as input to GEMINI , what we aimed to test in the refined output network was enrichment for the gold-standard interactions over the potential interactions . The initial TRN , formed by compiling the Yeastract interactions , was integrated with the yeast metabolic network [33] ( composed of 1597 reactions and 901 genes ) and gene expression data [34] ( consisting of 904 expression arrays in 435 conditions ) using PROM ( See Methods ) . 14% of all the interactions in the Yeastract database involved interactions with metabolic genes and the integrated model contains 31 , 075 interactions between 179 TFs and 863 metabolic genes . GEMINI performed in silico knockouts of each TF in the model and compared the predictions ( i . e . , lethal or viable ) to data from growth viability assays in glucose minimal media [35] . Running GEMINI on this network eliminated over 9 , 000 phenotype-inconsistent interactions and results in a final network containing 22 , 059 phenotype-consistent regulatory interactions . In comparison to the original YEASTRACT network , we found the final integrated network built using GEMINI to be highly enriched ( p-value = 10−172 , hyper-geometric test ) for validated gold-standard interactions; this result suggests that GEMINI preferentially removed low-confidence interactions ( Figure 2 ) . These results were robust to the chosen growth conditions – glucose , galactose , glycerol and ethanol minimal media all led to significant enrichment of gold-standard interactions ( Table 1 ) . We also observed the same effect when we did the same analysis using a different metabolic network model ( iMM904; See Methods ) , regulatory networks from different sources ( binding , motif-based and expression-based inference; see section below ) , different subsets of the Yeastract TRN ( Figure S4 ) and using different metrics to prioritize interactions ( Figure S9 ) . To determine whether a similar accuracy could have been obtained using expression data alone ( i . e . , without adding constraints based on the phenotypic outcomes predicted by the metabolic network ) , we compared our GEMINI results to a more commonly used approach for curating TRNs—sorting predicted interactions based on the correlated expression of the TFs and their putative target genes . Specifically , we measured the Mutual Information ( MI ) and Pearson's correlation among all of the interactions in our original YEASTRACT network . To ensure comparison was not biased towards GEMINI , we tuned the size of the network using MI and correlation over all possible values ( over-fitting to the best outcome that could be achieved for MI or correlation for any cutoff ) . The maximum enrichment obtained by MI and correlation ( even when overfit ) was lower than that obtained using GEMINI ( the lowest p-value measured over all possible network sizes for MI was 10−6 and for correlation was 10−3; Figure S1 shows the enrichment obtained over the entire range of thresholds for both MI and correlation ) . The high enrichment obtained by GEMINI strongly supports our hypothesis that additional phenotype data and integration with the biochemical details represented through the metabolic network can be used as an effective constraint to refine high-throughput interaction data . To gain further insight into the types of interactions recalled by the different methods , we examined another subset of interactions having “indirect evidence”—interactions inferred based on changes in the mRNA or protein expression of a target gene after perturbing its putative regulator [31] ( Figure 2 ) . MI and correlation performed significantly better at recalling indirect interactions than direct interactions ( p-value of 10−19 and 10−4 for the best cutoffs of MI and correlation , respectively ) ; this is not surprising since the indirect relationships are defined by gene expression changes . However , GEMINI still outperformed these methods in recalling indirect interactions ( p-value of 10−104 ) for any network size ( Figure 2c and Figure S2 ) . Therefore , GEMINI seems to more effectively distinguish both evidence-based direct and indirect interactions from a background of lower-confidence inferred interactions . Furthermore , no significant difference in the distributions of the MI scores was observed between the interactions retained and removed by GEMINI based on the Kolmogorov-Smirnoff test , showing that the phenotype data and integration with the metabolic network provides significant independent information ( Figure S3 ) . The biological relevance of the interactions retained by GEMINI is also supported by the enrichment for biological processes relevant to the set of target genes for each regulator . As compared to regulons ( target genes for each regulator ) in the original network , regulons in the refined network were found to be more specific , on average , to a given metabolic pathway ( p-value<0 . 01; Methods ) . The number of enrichments for specific metabolic pathways increased from 165 to 184 despite the removal of over 9000 interactions , suggesting that the phenotype-consistent regulons identified by GEMINI are associated with a more coherent set of molecular and metabolic functions , and most TFs tend to regulate distinct cellular processes as has been observed previously [3] , [8] , [36] . Through this process of refinement , we identified new statistical associations between TF and specific metabolic pathways ( Table S1 ) . More interestingly , GEMINI removed an association between the TFs , Msn4 and Gis1 , and the TCA cycle . The availability of flux measurements for the knockout strains of these two TFs enabled us to validate this prediction . Comparison with C13 flux data [13] showed that the knockout of these TFs did not in fact affect the flux through the TCA cycle . Comparison with TF knockout expression data from a recent study [15] also supported the functional significance of the phenotype-consistent interactions . This expression set was not part of the microarray compendium used for running GEMINI and allowed us to assess the predictive ability of the phenotype-consistent interactions . For 152 transcription factors in our network , we obtained a list of genes that were differentially expressed after the TF was knocked out ( FDR<0 . 05; [37] ) . We compared this list with the list of predicted target genes in the original Yeastract network and the refined network . We found that the targets of TFs in the refined network were more likely to be differentially expressed than those in the original network when their corresponding TF was knocked out ( p-value = 10−9; Methods ) . While we had selected interactions based on their consistency with phenotype , their ability to match expression changes in new conditions provided additional support for GEMINI . The phenotype-consistent interactions also had higher TF-DNA binding affinity than the original network ( p-value = 0 . 01; t-test; Methods ) , as measured from protein binding microarray ( PBM ) data [38] . These results also provide additional evidence supporting the validity of the potential interactions that were predicted to be phenotype-consistent by GEMINI . This suggests that GEMINI is effective at identifying functional interactions and is consistent with various heterogeneous data . One interesting observation from our results is that GEMINI can differentiate interactions from different sources based on their effect on the predicted phenotype . We next checked to see if we can use this to evaluate newly inferred interactions in the context of available known interactions . We can subsequently reconcile inconsistencies that arise from these interactions with metabolic phenotypes . To simulate such a scenario , we added new interactions onto the refined Yeastract network model and refined the expanded network model using GEMINI . We chose three commonly used data types: We found that for both the motif and CLR network , we could refine the network further and significantly enrich once again for direct and indirect interactions ( enrichment p-value compared to the original inferred network ( direct , indirect ) = ( 10−44 , 10−73 ) and ( 10−13 , 10−31 ) for motif and CLR , respectively; Table 2 ) . A wide variety of reverse engineering algorithms have been developed recently to infer potential regulatory interactions from sequence , gene expression data [1] , [4] , [8] or through integration of various data types [2] , [8] . These algorithms rely on correlated patterns of expression or the occurrence of a sequence motif in the upstream region of the target gene [5] . The enrichment for gold-standard interactions suggests that GEMINI could be integrated with these network inference and reverse engineering approaches to improve the identification of functional regulatory interactions . This result is consistent with the observation that an integrative network inference approach combining heterogeneous omics data could lead to more predictive TRN models [2] . While inference approaches like CLR allow for predicting potentially new TF-gene interactions , GEMINI is a refinement algorithm and it is not an alternative to these de novo inference approaches , but may be used in conjunction with such approaches to enhance their prediction by combining orthogonal data types . Overall , this result provides additional validation that GEMINI works across multiple data sets from different sources . In contrast to the inferred interactions , very few interactions ( ∼66 ) from the validated interaction data set ( Network III ) were removed by GEMINI . This interaction set is similar to the gold-standard set in the Yeastract database and was thus retained in the network . While these interactions were consistent with the simple lethal/non-lethal constraint we used in glucose minimal media , we predicted that by adding more constraints , we could narrow down the solution space further , and remove more phenotype-inconsistent interactions . With this aim , we employed PROM to quantitatively predict the growth rate ( as opposed to just lethal/non-lethal outcomes ) . Doing so allowed us to partition the non-lethal predictions into two categories: suboptimal and optimal ( Methods ) . Using this strategy for the 118 TFs in our network for which experimental measurements of this kind were available for comparison [13] , we were able to eliminate 4874 more interactions , while still improving the enrichment for the validated interaction set ( p-value of 10−27; Table 2; Figure 3 ) . Importantly , we observed that the refined network had a greater consistency with growth phenotype data in new conditions than the original network . Thus , by learning only on glucose minimal medium , the network model had greater correlation with growth rate measurements in galactose minimal medium ( correlation of 0 . 47 , p-value = 10−7 vs . a correlation of 0 . 2 , p-value = 0 . 04 for the original unrefined Yeastract model ) and in urea minimal medium ( correlation of 0 . 62 , p-value = 10−14 vs . a correlation of 0 . 22 , p-value = 0 . 02; data from Fendt et al . [13] ) . This is not unexpected because we were removing inconsistencies in one condition , which may have produced the same discrepancy in the other conditions as well . Nevertheless , the result suggests that GEMINI improves the overall predictive ability of the integrated regulatory-metabolic network model under new environmental conditions ( Figure 3 ) . We were also able to expand our integrated network model from 22 , 059 to 25 , 000 interactions through the addition of this validated interaction set .
By applying the GEMINI approach to our yeast model , we identified phenotype inconsistencies for 80 TF knockout predictions . The majority of the inconsistencies ( 85% ) were of the type NGG ( No Growth – Growth ) , for which the model predicts lethality ( or suboptimality ) , while the actual phenotype was non-lethal ( or optimal ) . Because this scenario was the most commonly identified inconsistency type , we concentrated on reconciling this set alone . Also , this case is more tractable to resolve than the opposite case ( GNG ) , which involves adding interactions from a very large multi-optimal solution space . Further , a TF knockout may be lethal or suboptimal due to a non-metabolic reason , meaning that even an optimal metabolic model would not be expected to resolve all GNG inconsistencies; in contrast , if a knockout is non-lethal and the model predicts it to be lethal , then that implies there is an inconsistency with the integrated model . GEMINI integrates two different network models ( metabolic and regulatory ) and inconsistencies could arise due to either network . In this work , we assumed that the metabolic network , being better curated and having a biochemical basis , could be used to identify inconsistencies in the regulatory network . Additional evidence from the distribution of inconsistencies also supports our assumption ( Figures S4 and Figure S10; discussed below ) . Furthermore , NGG inconsistencies arising due to the metabolic model were circumvented by using a metabolic model in which the GrowMatch algorithm [21] was run to resolve the NGG inconsistencies ( Zommorodi and Maranas model [33] ) . To test the sensitivity of our approach to the metabolic model used , we repeated our analysis with an older version of the metabolic model ( iMM904 [41] ) , which has a lower predictive accuracy than the Zommorodi and Maranas model . We found that even with iMM904 , GEMINI was able to strongly enrich for direct interactions ( p-value = 10−104 ) , but not as strongly as when using the more predictive model by Zommorodi and Maranas . This suggests that as the predictive ability of the metabolic models improves , we should be able to refine these interactions further . In theory , a trivial solution for resolving NGG inconsistencies is to remove all of the interactions for the respective TF . However , interestingly , GEMINI resolved all 80 NGG inconsistencies without reverting to the trivial solution . Furthermore , the elimination of phenotype inconsistent interactions by GEMINI based on one condition might lead to inconsistent predictions in a different condition . We found that this was the case for a small fraction ( 4% ) of the interactions that were phenotype-inconsistent in glucose minimal media , but were predicted to be consistent with growth phenotype data in galactose minimal media . Analyzing inconsistencies over different set of conditions would help us avoid over fitting the model to the growth phenotype data . Further analysis across conditions would help uncover interactions that are condition-specific and post-transcriptionally regulated ( discussed below ) . In the present analysis , we used the predicted growth rate as the only phenotype to constrain the regulatory network . If the interactions regulating biomass-related metabolic reactions were enriched for potential interactions , this would lead to an apparent enrichment for direct gold standard interactions on running GEMINI as an artifact . We tested this by evaluating the metabolic genes for which their knockout affected the maximum growth rate of the model . No difference was observed in the number of gold-standard interactions regulating this set of genes versus the rest ( both the sets had the same fraction ( 14% ) of gold-standard interactions; Methods ) . A similar distribution of gold-standard interactions was also found for interactions regulating dead-end reactions that do not contribute to the biomass and the rest of the metabolic network . Hence , there were no apparent underlying biases in the metabolic network architecture that led to the enrichment of gold-standard interactions . We predicted that the effectiveness of GEMINI would also depend on the scale of the regulatory network model used . GEMINI evaluates interactions in the context of other interactions in the network and so its effectiveness will depend on the size and degree of completeness of the entire network . To test this , we ran GEMINI using different fractions of the entire TRN and looked at the enrichment for gold-standard interactions . As expected , we found that GEMINI's effectiveness to refine the network increased with the size of the input network . To control for size bias on the enrichment p-value , we also looked at the fraction of gold-standard interactions in the initial and final refined network and observed the same effect ( Figure S4 ) . GEMINI utilizes the mechanistic information in biochemical networks to refine high-throughput interaction data . We next sought to determine which parts of the yeast transcriptional regulatory network were prone to inconsistencies across different growth conditions ( Table 1 ) . We analyzed the distribution of inconsistencies among the 41 TFs that led to inconsistencies using the qualitative phenotype data across the four carbon sources . The distribution was approximately exponential suggesting that a few TFs led to most of the inconsistencies ( Figure 4 ) . By identifying key regions that lead to the most inconsistencies , we can prioritize experiments to refine the regulatory network . Further , it highlights regions that are prone to inconsistent predictions while analyzing integrated network models . The top three TFs with most inconsistencies were Ash1 , Fkh1 and Fkh2; Ash1 encodes a transcription factor that is involved in mating type switching and while the genes Fkh1/2 are involved in cell cycle regulation . Interestingly , all these TFs have important roles outside metabolism suggesting that the interactions with metabolic enzymes might be false positives due to sequence-based inference . In contrast to the regulatory network , analysis of the distribution of inconsistencies across the metabolic network did not reveal any strong trend towards specific metabolic pathways . The distribution was linear rather than exponential across the metabolic genes ( Figure 4 ) . This suggests that relative to the regulatory network there were no specific genes in the metabolic network that were much more prone to inconsistencies . This is consistent with our previous observation that no underlying biases in the metabolic network architecture led to the enrichment of gold-standard interactions . Among the metabolic pathways highlighted in Table S1 , the pentose phosphate ( PP ) pathway had the most number of inconsistencies . Being a well-studied pathway in yeast and other organisms , it's more likely that the inconsistency arose due to the regulatory interactions rather than due to the PP pathway . Among the carbon sources , galactose led to the least enrichment for both validated gold standard interactions and indirect interactions . Both glucose and galactose enter central metabolism at the level of glucose-6-P , but they lead to primarily fermentative or respiro-fermentative metabolism , respectively [13] , [42] This suggests that we have perhaps incomplete knowledge about the regulatory network changes that happen during growth in galactose , though extensively studied [13] , [43] , [44] , and despite being similar at the metabolic level to glucose . GEMINI also performed poorly on rich media , which is primarily due to the limitations in the representation of the media constituents in rich media within a constraint-based modeling framework . Analysis of phenotype-consistent interactions inferred using GEMINI under different environmental conditions ( Table 1 ) revealed potential post-transcriptional regulatory mechanisms . Although there was considerable overlap between the phenotype-consistent interactions predicted from different minimal media conditions , we identified 1170 interactions that were phenotype-inconsistent in only one condition , but were retained in all the other three conditions ( Table S2 ) . The fraction of direct and indirect interactions among the 1170 interactions was quite similar to those interactions that were retained in all conditions . We predicted that these interactions might be true interactions that are conditionally-inactive , and the phenotype inconsistency might have arose due to post transcriptional regulatory mechanisms inactivating these interactions in these conditions . While the static information from the gene regulatory network and gene expression predicted the interactions to be active , combining this information with phenotypic data resulted in identifying post-transcriptional regulatory mechanisms that may have turned off these interactions . Glucose repression is one of the most well-studied processes in yeast and we focused on a subset ( 408 ) of these 1170 interactions that were predicted to be inactive only in glucose minimal media . The top 3 TFs with most interactions in this list—Rph1 , Hsf1 and Adr1 , were all activated during glucose starvation and are regulated via signaling and phosphorylation [45] [46] [47] . For example , the TF Hsf1 is constitutively phosphorylated , but under glucose starvation , it becomes hyper-phosphorylated and adopts an activated conformation resulting in the transcription of target genes [45] . The other TFs are activated through similar mechanisms in the absence of glucose . This is consistent with our prediction that the interactions that lead to inconsistencies only in glucose media were true interactions that are conditionally-inactive in the presence of glucose . Thus , we can potentially infer interactions that are not transcriptionally mediated through this approach . The condition-specific predictions also agreed well with a list of manually curated TF-environment interactions from the regulatory network model of Herrgard et al . [19] for 6 of the 7 predicted glucose-repressed TFs that were present in both the models . This strategy shows the utility of looking across multiple conditions to identify discrepancies in the data , which might be due to additional biological regulation . This also highlights the importance of incorporating signaling networks as they become available into these integrated network models . Given the large amount of data required to run GEMINI , we are currently restricted to a few well-studied systems with adequate expression , knockout phenotype and network data . However , with the development of automated methods for reconstructing metabolic networks [48] , GEMINI could be used as part of a network inference pipeline to identify functional regulatory interactions that are inferred from omics data , and reconcile the interactions with metabolic phenotypes for a large number of sequenced organisms . Another limiting factor in this study was the phenotype data used for analysis . The use of gene deletion growth phenotype data in the current study might restrict GEMINI's application only for microbes for which such a knock-out library exists and has been measured in great enough detail across different conditions . This approach might not be feasible for use in higher organisms like humans and mice . Yet , in theory , phenotype data other than that from growth assays such as metabolite uptake or secretion could be used to limit the space of possible functional states of the TRN and could be applied to higher organisms . The regulatory network model used in this study , despite being genome-scale and much more comprehensive than the current integrated model for yeast [19] , does not comprise the entire TRN . We have focused only on a subset ( 14% ) of the TRN that regulate metabolism . Nevertheless , this subset of the TRN is very well studied and has important applications in metabolic engineering and synthetic biology . Further , the scale of the TRN is primarily limited by the size of the biochemical model with which it interfaces . Although we have restricted our analysis to interactions involving metabolic reactions in the present work , the GEMINI approach is generally applicable to other cellular network types [25] , [49] , such as signaling networks , as they become available . By integrating other network types , one might account for additional regulatory-phenotype relationships and thus improve predictions even further . Regulatory network inference is a significant challenge today as the system is underdetermined and often results in multiple models that could explain the same data with equal efficacy . Thus , it is important to incorporate diverse heterogeneous data types like expression , binding and growth phenotype to constrain the solution space . GEMINI exploits this principle to refine high-throughput regulatory interaction data and identifies interactions that are consistent with various data types . Importantly , this is the first such approach that ties the inference of a transcriptional regulatory network from high-throughput data with a biochemically detailed metabolic network . We believe this to be an important first step towards mechanistically refining a network model of one type ( gene regulatory ) using data from another network type ( metabolic ) . Further , our approach highlights the potential of using a biochemically-detailed mechanistic framework to interpret high-throughput data and identify and reconcile inconsistencies across different data types . We find that the data types that are more consistent with each other also have greater evidence supporting their existence . While there are still several challenges ahead for regulatory network inference , the methods presented here lay the foundation for the rapid refinement of omics data using a mechanistic framework , which will advance the study of metabolic regulation and lead to better predictive models of the cell .
Using PROM , we predicted the growth outcome of knocking out each TF in the network under a specific condition . By comparing our simulations with experimental growth viability data , we identified and reconciled inconsistent predictions . TF knockouts were predicted to be lethal if the respective maximal growth rate prediction of the mutated organism was less than 5% of the wild-type growth rate [22] , [50] . Any knockout that resulted in a growth-rate lower than 95% of the wild-type was considered suboptimal , as has been used previously in other analyses [22] , [26] . These results were robust to the choice of the growth thresholds ( Figure S5 ) . While we used the values commonly used in the literature , tuning these thresholds indicated that higher enrichments could be achieved by varying this parameter . However , we recommend using the default values to avoid over-fitting . The closest flux state that represents the measured growth phenotype ( v2 ) was obtained by solving the same optimization problem for PROM with the additional constraint that the predicted model growth rate matches the observed growth phenotype: ( 8 ) subject to constraints ( 9 ) ( 10 ) Additional constraint – ( 11 ) or ( 12 ) f is the predicted growth rate by the model , and 0 . 05 and 0 . 95 are the growth rate thresholds for determining viability and suboptimality as mentioned above . The solution obtained by solving this above problem gives flux state v2 . The entire steps in GEMINI are described in the pseudo code below: For each TF { ELSE REPEAT from 3e till it matches actual phenotype } The flux solutions in FBA have multiple possible states , while the growth rate or the objective function is unique . Since we relied only on the growth rate and the transcriptionally constrained reactions ( part of the objective function in PROM ) as the metric to refine the network , the final network structure was identical across different runs of GEMINI . To further investigate how alternate optimal solutions alter the effectiveness of GEMINI we generated new flux solutions by introducing small changes to the growth threshold ( step 3b in pseudo code and equation 11 ) . We compared five different networks across different combinations generated by changing the growth threshold . This generated new flux solutions with approximately the same growth rates . We found that the networks were 99 . 9% similar across these small perturbations that led to alternate flux solutions ( Table S3 ) . We can infer that the same subset of phenotype-inconsistent interactions is removed across various growth thresholds and flux optima . The use of regulatory-constrained reactions in the objective function in PROM ensures that there is no variability between runs and we get the same solution each time while running GEMINI . Furthermore as mentioned earlier , strong enrichment for validated interactions were obtained over a wide range of these growth thresholds ( Figure S5 ) . The above analysis of inferring regulatory networks across alternate metabolic flux solutions also resolves the possibility of multiple alternate optima with respect to the regulatory network . We found that the same core set of interactions was removed across different runs . In addition , we also compared network generated using much larger changes in growth rate threshold used for inferring the flux state v2 . We once again found that while the refined network sizes changed across different thresholds , they were >95% similar to each other among the interactions that were retained . These results indicate that that there is a strong global optimal state for the regulatory network and by perturbing the model and constraints we still converge close to the global optima . In terms of network refinement , all these results suggest that there is a core set of regulatory interactions that are removed across different constraints and conditions ( Table S3 ) . While its still certainly possible that there are multiple other optimal flux and regulatory network solutions , the use of regulatory-constrained reactions in the objective function in PROM ensures that there is no variability between runs and we get the same solution each time while running GEMINI . The value of κ , which determines the strength of the transcriptional regulatory constraint , was determined in a data-driven manner by tuning across a range of values . We set κ to be the lowest value above which there was no increase in the number of interactions removed ( Figure S6 ) . We obtained a κ value of 10 for all of our simulations based on this strategy . The results were robust to the value of κ chosen this way for a wide range of values above 10 ( Figure S7 ) . We have used a metabolic network-based approach for prioritizing regulatory interactions for pruning . One can envision other approaches and metrics to prioritize these interactions . As an alternative metric , we sorted interactions based on probabilities instead of predicted flux difference ( see step 3 in the pseudo-code ) . While this seems to be a straightforward metric , this ignores the system-level effect of these interactions on the biochemical network for prioritizing the interactions . Using this approach on the yeastract data , we obtained an enrichment of 10−20 for direct interactions . Note that even though we only use transcriptomic data to prioritize interactions , this approach yields higher enrichment than MI or correlation . This is because we prune interactions till the predicted systems-level growth phenotype matches the experimental measurement; thus the systems level constraint is unchanged while only transcriptomic data is used for prioritizing interactions . As a second alternative approach for prioritizing interactions , instead of sorting interactions based on the flux difference between the predicted ( v1 ) and expected ( v2 ) flux state , we assigned the reactions into two groups – the first group of reactions change significantly based on a z-score threshold between v1 and v2 and the rest that did not change significantly . Interactions that regulate these reactions were then pruned randomly from the first group and then from the second group . The rationale being that this strategy doesn't rely significantly on the absolute difference between reactions and allows for alternate flux solutions . We once again found strong enrichment for gold standard interactions through this approach across different thresholds ( p-value = 10−143 ) . This method is further discussed in Figure S9 . The strong enrichments using different metrics and thresholds suggest that the systems level constraints are relatively more important than the order in which the different inconsistencies are solved . Both expression randomization and phenotype swapping removed the enrichment for gold-standard interactions ( p-value = 1 ) . We also performed bootstrapping of expression data to determine sensitivity to the gene expression data used . This was done by running GEMINI using random subsets comprising 80% of the expression data . We found strong enrichment in all of the runs ( p-value<1E-90; Figure S8 ) , suggesting that the data were sufficiently powered for this analysis . All parameters were left at the default value as recommended for running PROM ( binarization threshold – 0 . 33 i . e . the 33rd percentile of gene expression data ( Figure S10 ) ; lethal/non-lethal growth threshold – 0 . 05 ( Figure S5 ) ) . The parameter Kappa is determined in a data driven manner by the GEMINI algorithm as mentioned above . We found that much higher enrichment could be achieved by changing the binarization threshold ( upto 10−220; Figure S10 ) ; nevertheless , we recommend using the default parameter values while running GEMINI to avoid over fitting . For the analysis to identify potential biases in the network architecture , we identified genes affecting maximal growth rate by doing a systematic single gene deletion of all the metabolic genes in the model in glucose minimal media . We identified interactions that regulate this set of genes and compared it with the rest of the interactions in the network . We found the fraction of gold standard interactions to be the same in both sets of interactions . Dead end reactions used for this analysis were identified using the removeDeadends algorithm in the COBRA toolbox in MATLAB . We used the reconstructed yeast metabolic network by Zommorodi and Maranas because it had the highest predictive ability among the available yeast models [33] . In our simulations , the carbon source and oxygen uptake were constrained to 10 mmol/h/gDW and 2 mmol/h/gDW , respectively . Ammonia , phosphate , and sulfate were assumed to be non-limiting . Trace amounts of essential nutrients that were present in the experimental minimal media formulation ( 4-aminobenzoate , biotin , inositol , nicotinate , panthothenate , and thiamin ) were also supplied in the simulations . Flux variability analysis for PROM was performed using the FastFVA algorithm [51] . Robust multi-array averaged ( RMA ) -normalized gene expression data consisting of 904 arrays in 435 conditions were obtained from the Many-Microbes Microarray Database [34] . This microarray compendium was chosen with the aim of maximizing the number of conditions under which gene expression is measured , while reducing array platform-induced variations [22] , [52] . All the regulatory interaction data were obtained from the supplementary material of the respective publications [37] or from the author's website [38] , [40] and from the YEASTRACT database [31] . Among these interactions , only those involving metabolic genes , and those that had corresponding expression data in the Many Microbes Database were retained . Growth phenotype data for yeast TF knockout strains grown in glucose , galactose , glycerol and ethanol minimal media were obtained from Kuepfer et al [35] . These data provided a list of lethal/non-lethal predictions under different conditions . Quantitative growth data were obtained from Fendt et al [13] in glucose , galactose and urea minimal media . TFs with missing values in the Kuepfer et al or Fendt et al phenotype data were not refined using GEMINI . Metabolic pathway enrichment analysis was done by overlapping genes in each regulon with genes in each pathway ( like TCA cycle or glutamate metabolism ) as defined in the metabolic network model . The p-value for overlap between the regulons and pathway genes was calculated using the hyper-geometric test . In the analysis to determine the functional significance of the interactions , the differentially expressed genes ( FDR<0 . 05 ) were obtained from Reimand et al . [37] based on their analysis of a comprehensive TF knockout experiment by Hu et al . [15] . For the comparison with PBM data [38] , we compared the distribution of interaction ranks for the original and refined network . We used a t-test to test the hypothesis that the mean rank for the refined network was lower than the mean rank of interaction for the original network ( p-value = 0 . 001 ) . The sequence motif data were obtained from the supplement of MacIsaac et al . [39] . A TRN model was inferred using the algorithm CLR [4] with default parameters ( number of bins = 10 and spline degree = 3 ) , and using the expression data from the Many Microbes Database . Predicted interactions with z-scores greater than two ( mutual information greater than two standard deviations above than background ) were chosen in the final network . Interactions involving metabolic genes were then identified and used for the analysis . For the TF knockout data described previously , the top 100 genes with the lowest p-value ( below a FDR threshold of 0 . 05 ) were considered to be targets for each TF . This was done to limit the size of the TRN . Table 2 gives the network sizes and the number of interactions retained in each case . Mutual Information between interactions was measured using the algorithm ARACNE [7] with default parameters . The p-value for overlaps and enrichments with different interaction sets was calculated using the hyper-geometric test . We calculated a p-value for each comparison by summing over probabilities for all values of overlap> = L , the length of the overlap . The obtained p-values were rounded off to the closest power of 10 for clarity . All the simulations and statistical analyses were performed in MATLAB . The COBRA toolbox [53] was used to load and optimize the metabolic model . The optimization problem was solved using the GNU Linear Programming Kit ( GLPK ) solver . The GEMINI algorithm along with a faster version of PROM , and the integrated metabolic-regulatory network models for yeast , are available for download at https://sourceforge . net/projects/gemini-data/ .
|
Cellular networks , such as metabolic and transcriptional regulatory networks ( TRNs ) , do not operate independently but work together in unison to determine cellular phenotypes . Further , the phenotype and architecture of one network constrains the topology of other networks . Hence , it is critical to study network components and interactions in the context of the entire cell . Typically , efforts to reconstruct TRNs focus only on immediately proximal data such as gene co-expression and transcription factor ( TF ) -binding . Herein , we take a different strategy by linking candidate TRNs with the metabolic network to predict systems-level responses such as growth phenotypes of TF knockout strains , and compare predictions with experimental phenotype data to select amongst the candidate TRNs . Our approach goes beyond traditional data integration approaches for network inference and refinement by using a predictive network model ( metabolism ) to refine another network model ( regulation ) – thus providing an alternative avenue to this area of research . Understanding how the networks function together in a cell will pave the way for synthetic biology and has a wide-range of applications in biotechnology , drug discovery and diagnostics . Further we demonstrate how metabolic models can integrate and reconcile inconsistencies across different data-types .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2013
|
Metabolic Constraint-Based Refinement of Transcriptional Regulatory Networks
|
Syphilis , a sexually transmitted disease caused by spirochetal bacterium Treponema pallidum , can progress to affect the central nervous system , causing neurosyphilis . Accumulating evidence suggest that regulatory T cells ( Tregs ) may play an important role in the pathogenesis of syphilis . However , little is known about Treg response in neurosyphilis . We analyzed Treg frequencies and Transforming Growth Factor-β ( TGF-β ) levels in the blood and CSF of 431 syphilis patients without neurological involvement , 100 neurosyphilis patients and 100 healthy donors . Suppressive function of Tregs in peripheral blood was also assessed . Among syphilis patients without neurological involvement , we found that secondary and serofast patients had increased Treg percentages , suppressive function and TGF-β levels in peripheral blood compared to healthy donors . Serum Rapid Plasma Reagin ( RPR ) titers were positively correlated with Treg numbers in these patients . Compared to these syphilis patients without neurological involvement , neurosyphilis patients had higher Treg frequency in peripheral blood . In the central nervous system , neurosyphilis patients had higher numbers of leukocytes in CSF compared to syphilis patients without neurological involvement . CD4+ T cells were the predominant cell type in the inflammatory infiltrates in CSF of neurosyphilis patients . Interestingly , among these neurosyphilis patients , a significant decrease in CSF CD4+ CD25high Treg percentage and number was observed in symptomatic neurosyphilis patients compared to those of asymptomatic neurosyphilis patients , which may be associated with low CSF TGF-β levels . Our findings suggest that Tregs might play an important role in both bacterial persistence and neurologic compromise in the pathogenesis of syphilis .
China has experienced an expanding epidemic of syphilis infection in the last 10 years [1] , [2] . In 2011 , the national incidence rate was 32 . 04 per 100 , 000 population and 429 , 677 new cases were reported [3] . This sexually transmitted disease has reemerged as a significant public health issue in China due to its serious , irreversible sequelae [4] and its strong association with HIV infection [5] . The rapid rise in syphilis rates in China highlight the importance of understanding of the pathogenesis of syphilis and its complications . The spirochetal bacterium , Treponema pallidum ( T . pallidum ) , is the etiologic agent of syphilis [6] , [7] . After T . pallidum infection , mammalian hosts mount robust humoral and cellular immune responses aimed at spirochetal clearance [8] , [9] , [10] , [11] . However , T . pallidum has the ability to escape the host immune response and establish persistent infection . There are several strategies used by the spirochete to resist host immune effector mechanisms including poor antigenicity [12] , [13] , antigenic variation of membrane proteins [14] , [15] , [16] , and impaired antibody-mediated opsonization [17] . Interestingly , several studies have demonstrated that T . pallidum may also actively harness host immune suppression mechanisms to facilitate persistence and dissemination [18] , [19] . A recent study has demonstrated that T . pallidum antigen TpF1 could promote development of regulatory T cells ( Tregs ) in the patients with secondary syphilis [18] . Tregs represent a unique population of CD4+ T cells with potent immune suppressive activity [20] , [21] . This regulatory CD4+ T cell population is classically defined by high expression of CD25 ( IL-2 receptor α-chain ) [22] . The forkhead family transcription factor Foxp3 , the most definitive signature , is critical for Treg development and function [23] . Emerging evidence from human patients and animal models has demonstrated that Tregs contribute to impaired immune responses and chronic infection with diverse organisms [24] , including mycobacterium tuberculosis [25] , helicobacter pylori [26] , hepatitis B virus [27] , [28] , HIV [29] , and plasmodium falciparum [30] . The enhanced Treg response in early syphilis patients may down-regulate immune effector function to allow survival of T . pallidum within the host . T . pallidum infection can infect many organs , including central nervous system ( CNS ) . This form of syphilis is termed neurosyphilis . Neurosyphilis may affect the meninges or brain or spinal cord parenchyma and may be asymptomatic or symptomatic [4] , [31] . Meningeal neurosyphilis usually appears during the first few years of T . pallidum infection . Patients with meningeal neurosyphilis may be manifested by meningitis ( headache , stiff neck , and cranial nerve abnormalities ) or meningovasculitis ( focal CNS ischemia or stroke ) . Parenchymal neurosyphilis , presenting as general paresis and tabes dorsalis , occur in the later course of the disease , often decades after the primary infection [4] , [32] . The mechanisms underlying the development of symptomatic neurosyphilis in some patients are largely unknown . Previous studies have extensively characterized immune cell infiltrates of early syphilis lesions [8] , [9] , [10] and indicated that the clinical manifestations of early syphilis result from collateral tissue damage caused by host immunity to T . pallidum [6] , [33] . However , little is known about the immune response in neurosyphilis patients . In the present study , we performed a comparative analysis of Tregs in peripheral blood and cerebrospinal fluid ( CSF ) from neurosyphilis patients and syphilis patients without neurological involvement . We found that symptomatic neurosyphilis patients had lower Treg frequencies and numbers in CSF compared to asymptomatic neurosyphilis patients , indicating that an immunopathological mechanism might be present in the onset of neurological symptoms .
This study was performed at the Shanghai Skin Disease Hospital between June 2009 and Jan 2012 . The hospital is located in central Shanghai , where the syphilis prevalence is highest in China [1] . The Sexually Transmitted Diseases ( STD ) center in this hospital is the major STD clinic in Shanghai , which provides screening , diagnosis and treatment for most sexually transmitted diseases , including syphilis . As one of the biggest STD centers in China , more than 300 patients are served in this clinic per day . Although most of our patients are walk-in , some are referred to our clinic by their doctors at other hospitals across the country . This study was approved by the Ethics Committee of the Shanghai Skin Disease Hospital . Written informed consent was obtained from all participants . Syphilis was determined based on medical history , physical , neurological and psychiatric symptoms and signs , and the results of nontreponemal and treponemal serological tests . The excluded criteria include HIV; prior syphilis or syphilis treatment ( except in the serofast syphilis group ) ; history of systemic inflammatory , autoimmune disease , other underlying acute or chronic disease , were receiving anti-inflammatory medications , were immunocompromised , or use of antibiotics or immunosuppressive medications in the last four weeks . Peripheral blood was collected from all healthy donors and syphilis patients . Lumbar punctures were encouraged to be performed if i ) patients had neurological or psychiatric signs or symptoms , ii ) patients whose serum RPR≥1∶32 , regardless of stage or presentation , iii ) patients whose serofast state was more than 2 years and who are anxious regarding their serofast state . 100 healthy donors , who visited Shanghai Skin Disease Hospital voluntarily for STD prevention and a medical check-up , were recruited to the study . All healthy control subjects were negative for HIV and serological tests for syphilis . Primary syphilis: i ) Chancres or ulcers; and/or ii ) detection of spirochetes in a dark-field microscopy examination; and iii ) positive RPR confirmed by Treponema pallidum particle agglutination assay ( TPPA ) ; and iv ) absence of other causes of genital ulcers , including herpes simplex virus ( HSV ) infections . Secondary syphilis: i ) positive RPR confirmed by TPPA; and ii ) skin or mucocutaneous lesions; Latent syphilis: i ) positive RPR confirmed by TPPA; and ii ) without skin or mucocutaneous lesions or any symptoms of syphilis; Serofast syphilis: i ) previously treated syphilis of any stage; ii ) an appropriate 4-fold decline in serum RPR titer at 6 months after treatment ( Benzathine penicillin 2 . 4 MU/qw im for 2 or 3 weeks or procaine penicillin 0 . 8 MU/day im for 15 days in most cases , if patient allergic to penicillin ceftriaxone 250 mg/day im for 10 days would be as an alternative ) ; iii ) persistently reactive serum RPR two or more years after treatment; iv ) no evidence of reinfection . The clinical and laboratory characteristics of 71 patients with primary syphilis , 136 patients with secondary syphilis , 127 patients with latent syphilis , and 97 patients with serofast syphilis were shown in Table 1 . All neurosyphilis patients have positive serum RPR and TPPA tests . The diagnosis of confirmed neurosyphilis also includes reactive CSF-VDRL ( Venereal Disease Research Laboratory ) and CSF-TPPA tests in the absence of substantial contamination of CSF with blood . Presumptive neurosyphilis was defined as nonreactive CSF-VDRL but reactive CSF-TPPA with either or both of the following: i ) CSF protein concentration >45 mg/dL or CSF white blood cell ( WBC ) count ≥8/µL in the absence of other known causes for these abnormalities; ii ) neurological or psychiatric manifestations consistent with neurosyphilis without other known causes for these abnormalities . Fourteen patients with presumptive neurosyphilis were also included in the study and the data of these patients were combined with those of confirmed neurosyphilis patients for analysis . In the case of presumptive neurosyphilis , the patient has a nonreactive CSF-VDRL test plus a reactive CSF-TPPA along with either or both of the following: ( i ) elevated CSF proteins ( normal: 15–45 mg/dL ) or elevated CSF white blood cell ( WBC ) count ( normal: <8/µL ) in the absence of other known causes of the abnormalities; ( ii ) clinical neurological or psychiatric manifestations without other known causes of these clinical abnormalities . Neurosyphilis is categorized as asymptomatic , meningeal ( meningitis and meningovasculitis ) and parenchymal ( general paresis and tabes dorsalis ) . Asymptomatic neurosyphilis is defined by the presence of CSF abnormalities consistent with neurosyphilis and the absence of neurological and psychiatric signs or symptoms . Meningitis is diagnosed by CSF abnormalities and headache , stiff neck , nausea , or cranial neuropathies . Meningovasculitis is defined by clinical features of meningitis and stoke with or without neuroradiological confirmation . General paresis is characterized by personality changes , dementia and psychiatric symptoms including mania or psychosis . Tabes dorsalis is characterized by sensory loss , ataxia , lancinating pains , and bowel and bladder dysfunction . All patients diagnosed with neurosyphilis should have no other known causes for these clinical abnormalities . The features of 100 neurosyphilis patients are shown in Table 2 . These patients are mutually exclusive of those in Table 1 . Peripheral blood mononuclear cells ( PBMC ) were isolated from whole blood from syphilis , neurosyphilis patients and healthy donors via density centrifugation over Lymphoprep ( Axis-Shield ) . CSF was centrifuged and stained immediately at 4°C after spinal tap . The volume was 5 mL . Multicolor fluorescence activated cell sorting ( FACS ) analysis was performed using the following antibodies: PE- , FITC- , PerCP , or PE-Cy5-conjugated antibodies against human CD45 ( Biolegend ) , CD3 ( Biolegend ) , CD4 ( Biolegend ) , CD25 ( Biolegend ) . For Foxp3 staining , cells were stained using One Step Staining Human Treg Flow Kit ( Biolegend ) according to the manufacturer's protocols . Cells were assessed with FACScalibur ( Becton Dickinson ) or Epics XL ( Beckman Coulter ) cytometers as previously described [34] . For CSF samples , acquisition of ≥5 , 000 events for gated CD45+ cells was performed . The CSF Treg number was defined as the total number of CSF cells multiplied by the percentage of Tregs identified by flow cytometry . Data were analyzed using FlowJo software ( Tree Star ) . Treg suppression assay was performed as described [35] , [36] . Briefly , PBMC were used for CD4+ CD25+ and CD4+ CD25− T cell isolation using a Regulatory T Cell Isolation Kit according to the manufacturer's instruction ( Miltenyi Biotec ) . Purity of the cell fractions as determined by flow cytometry was >90% . Purified CD4+ CD25− T responder cells ( 5×104 cells/well ) were incubated in RPMI 1640 medium with 10% FBS in 96-well U-bottom plates precoated with anti-CD3 antibody ( 1 µg/mL; eBioscience ) . To assess suppressive ability , purified autologous CD4+ CD25+ T cells were added , at a CD25+/CD25− ratio of 1∶1 , 1∶2 , 1∶4 , or 1∶8 . All cells were cultured in a final volume of 200 µl in the presence of 2×104 irradiated allogeneic PBMC/well . After 4 days of culture , [3H] thymidine ( Amersham ) was added for an additional 18 h to each well . [3H] thymidine incorporation was measured using a liquid scintillation counter . Percent inhibition of proliferation was determined as ( 1- [3H] thymidine incorporation of CD25+ and CD25− T cells coculture/[3H] thymidine incorporation of CD25− T cells alone ) ×100 . Serum and CSF TGF-β1 levels were determined using Human TGF-β1 ELISA kit from eBioscience . We performed statistical analysis using GraphPad Prism version 5 . 01 ( GraphPad Software ) . All datasets were first assessed for deviation from a normal distribution using the D'Agostino-Pearson omnibus normality test . Non-normally distributed variables were compared between groups using the nonparametric Kruskal–Wallis test followed by Dunn's multiple comparison tests . If the variables were approximately normally distributed , differences between experimental groups were analyzed using one-way ANOVA followed by Bonferroni test for the selected pairs . Pearson correlation analysis was used to determine the relationship between the frequency of CD4+ CD25high Treg and other parameters . A value of P<0 . 05 was considered significant .
Human Tregs were identified as CD4+CD25high or CD4+Foxp3+ T cells [20] , [21] . The representative gating strategy for CD4+ CD25high and CD4+ Foxp3+ T cells are depicted in Figure 1A . The majority of Foxp3+ T cells co-expressed high levels of CD25 ( Figure 1A ) . The baseline frequency of CD25high Tregs among CD4+ T cells in PBMCs from healthy individuals was 2 . 7%±0 . 1% ( Figure 1B ) . A comparison between syphilis patients and healthy individuals revealed a 1 . 3-fold increase in mean frequency of CD4+ CD25high T cells in primary syphilis patients ( 3 . 6%±0 . 2% , p<0 . 01 ) , 1 . 7-fold increase in secondary syphilis patients ( 4 . 5%±0 . 2% , p<0 . 001 ) , 1 . 5-fold increase in early latent syphilis patients ( 4 . 1%±0 . 2% , p<0 . 001 ) , and 1 . 7-fold increase in serofast syphilis patients ( 4 . 7%±0 . 3% , p<0 . 001 ) ( Figure 1B ) . Consistently with CD25 expression , the highest percentage of Foxp3+ Tregs among CD4+ T cells were observed in patients with secondary syphilis ( 4 . 3%±0 . 4% , p<0 . 001 ) and serofast syphilis ( 4 . 3%±0 . 3% , p<0 . 001 ) patients , followed by latent syphilis ( 3 . 9%±0 . 4% , p<0 . 01 ) and primary syphilis patients ( 3 . 6%±0 . 4% , p<0 . 05 ) , which were all significantly higher than healthy donors ( 2 . 3%±0 . 1% ) ( Figure 1B ) . We next investigate the suppressive function of Tregs from syphilis patients on T cell proliferation . CD4+ CD25+ suppressor T cells were cocultured with autologous CD4+ CD25− T responder cells at different ratios ( suppressor/responder ratios: 1∶1 , 1∶2 , 1∶4 , and 1∶8 ) . We found that blood CD4+ CD25+ Tregs isolated from secondary syphilis ( 84 . 0%±1 . 4% , P<0 . 05 ) and serofast syphilis ( 84 . 3%±3 . 0% , P<0 . 01 ) but not primary syphilis ( 74 . 5%±1 . 1% , P>0 . 05 ) and latent syphilis ( 73 . 8%±1 . 1% , P>0 . 05 ) patients exhibited significantly higher suppressive activity than healthy controls ( 66 . 3%±1 . 1% ) at a 1∶1 ( suppressor: responder ) ratio ( Figure 1C ) . Significant increases in suppressive effect of CD4+ CD25+ Tregs were also observed at ratios of 1∶2 and 1∶4 in secondary and serofast syphilis patients compared with healthy donors ( Figure 1C ) . These data indicated that CD4+ CD25+ Tregs derived from secondary and serofast syphilis patients display enhanced suppressive function . Since Transforming Growth Factor-β ( TGF-β ) is critical to Treg differentiation and suppressive function [37] , [38] , [39] , we determined whether higher Treg frequency and function in syphilis patients were associated with serum TGF-β levels . It was shown that serum concentrations of TGF-β were significantly increased in patients with secondary ( 5 . 4±0 . 7 ng/ml , P<0 . 001 ) and , to a lesser extent , in primary syphilis patients ( 4 . 4±0 . 9 ng/ml , P<0 . 05 ) , latent patients ( 4 . 6±0 . 5 ng/ml , P<0 . 01 ) and serofast patients ( 4 . 4±0 . 6 ng/ml , P<0 . 01 ) compared with healthy controls ( 1 . 1±0 . 2 ng/ml ) ( Figure 1D ) . There was a positive correlation between the percentage of circulating CD4+ CD25high Tregs and serum TGF-β levels in these syphilis patients ( r = 0 . 20 , P<0 . 05 , Figure 1E ) . Nontreponemal test antibody titers usually correlate with disease activity [40] . We thus assessed whether serum RPR titers were associated with circulating Treg percentage in these syphilis patients . Pearson correlation analysis showed that there was a positive correlation between the percentage of circulating CD4+ CD25high Tregs and serum RPR titer in secondary syphilis ( r = 0 . 27 , P<0 . 01 , Figure 2B ) , latent syphilis ( r = 0 . 27 , P<0 . 05 , Figure 2C ) and serofast ( r = 0 . 44 , P<0 . 01 , Figure 2D ) syphilis patients , but no correlation in primary syphilis patients ( r = 0 . 10 , P = 0 . 44 , Figure 2A ) . If untreated or treated improperly , some syphilis patients will progress to neurosyphilis . To investigate whether Tregs are associated with the progression of neurosyphilis , we analyzed Treg numbers in the peripheral blood of 49 asymptomatic and 41 symptomatic neurosyphilis patients . As shown in Figure 3A and 3B , syphilis patients with neurological involvement ( including both asymptomatic and symptomatic syphilis patients ) had higher percentage of CD4+ CD25high Tregs ( 4 . 7%±0 . 2% , P<0 . 001 ) and CD4+ Foxp3+ Tregs ( 5 . 0%±0 . 4% , P<0 . 001 ) in peripheral blood compared with healthy individuals ( 2 . 7%±0 . 1% and 2 . 4%±0 . 1% , respectively ) . Compared to syphilis patients without neurological involvement ( including primary , secondary , latent and serofast syphilis patients ) , there was a slight but not significant increase in CD4+ CD25high Treg frequency in peripheral blood of neurosyphilis patients ( P = 0 . 06 ) ( Figure 3A ) , but the percentage of CD4+ Foxp3+ Treg were significantly higher ( P<0 . 05 ) ( Figure 3B ) . Among syphilis individuals with neurological involvement , there was no significant difference in CD4+ CD25high Treg frequency ( P>0 . 05 ) ( Figure 3C ) and CD4+ Foxp3+ Treg frequency ( P>0 . 05 ) ( Figure 3D ) in peripheral blood among asymptomatic , meningeal , and parenchymal neurosyphilis patients . CSF mononuclear pleocytosis is one of diagnostic criteria for neurosyphilis [41] , [42] . As expected , higher numbers of leukocytes were observed in asymptomatic ( 14±3 cells/µL ) , meningeal ( 35±15 cells/µL ) and parenchymal ( 16±4 cells/µL ) neurosyphilis patients compared to those from syphilis patients without neurological involvement ( 4±1 cells/µL ) ( P<0 . 001 , P<0 . 001 , and P<0 . 001 , respectively ) ( Table 3 ) . Among the CSF leukocytes , higher percentage of CD4+ T cells were found in patients with asymptomatic ( 41 . 8%±2 . 3% , P<0 . 01 ) and parenchymal ( 46 . 4%±2 . 1% , P<0 . 001 ) neurosyphilis compared with syphilis patients without neurological involvement ( 29 . 7%±2 . 4% ) ( Table 3 ) . There was no significant difference in CSF CD4+ T cell frequency ( P>0 . 05 ) among different types of neurosyphilis patients ( Table 3 ) . The average percentage of CD25high Tregs in the CD4 compartment was 22 . 0%±1 . 0% for the patients without neurological involvement and did not differ from those with asymptomatic neurosyphilis ( 20 . 0%±1 . 1% , P>0 . 05 ) . Both meningeal ( 12 . 5%±1 . 4% ) and parenchymal ( 12 . 0%±1 . 2% ) neurosyphilis patients showed pronounced decreases in CD4+CD25high Treg percentage compared to syphilis patients without neurological involvement ( P<0 . 05 , P<0 . 001 , respectively ) and asymptomatic neurosyphilis patients ( P<0 . 05 , P<0 . 001 , respectively ) ( Table 3 ) . Due to preferential accumulation of CD4+ T cells in the CSF of neurosyphilis patients , both asymptomatic and symptomatic neurosyphilis patients have higher numbers of CD4+CD25high Tregs than syphilis patients without neurological involvement . Interestingly , lower number of Tregs was observed in meningeal ( 0 . 9±0 . 3 cells/µL ) and parenchymal ( 0 . 5±0 . 1 cells/µL ) neurosyphilis patients than asymptomatic neurosyphilis patients ( 1 . 2±0 . 2 cells/µL ) . In addition , meningeal ( 3 . 4±0 . 9 ng/ml ) and parenchymal ( 2 . 8±0 . 5 ng/ml ) neurosyphilis patients had significantly lower CSF TGF-β levels than asymptomatic neurosyphilis ( 10 . 7±2 . 0 ng/ml ) and syphilis patients without neurological involvement ( 8 . 2±1 . 7 ng/ml ) , indicating that decreased CD4+ CD25high Treg frequencies in CSF of symptomatic neurosyphilis patients may be associated with low CSF TGF-β concentration .
Syphilis is a multistage chronic disease , which can cause damage to diverse tissues and organs . An influx of immune cells to skin lesions of early syphilis patients not only mediates bacterial clearance but also lead to tissue damage and clinical symptoms [9] , [10] , [43] , [44] . Our prior study has shown that immune cells can also infiltrate into the CSF of syphilis patients [45] . However , this study was limited because of a small number of patients ( n = 32 ) , selected patient populations ( latent syphilis and neurosyphilis ) and lack of characterization of neurosyphilis patients [45] . In the present study , a total of 431 syphilis patients without neurological involvement ( including 20 latent syphilis patients in the previous report ) and 100 neurosyphilis patients ( including 12 patients in the previous report ) were included . This larger number of syphilis patients enables further stratification according to stage and symptoms . Interestingly , we observed an accumulation of CD4+ T cells in the CSF of both asymptomatic and symptomatic neurosyphilis patients , which were consistent with several previous reports showing that CD4+ T cells were the primary responders to T . pallidum in syphilis lesions [8] , [9] , [46] . CD4+ T cells can be divided into a variety of effector subsets , including classical Th1 cells and Th2 cells , the more recently defined Th17 cells , follicular helper T cells , and regulatory T cells [47] . Though we did not elucidate the precise identity of the CD4+ T cell subset , we observed a decreased frequency of CD4+ CD25high Tregs in the CSF of symptomatic neurosyphilis patients compared with those of non-neurosyphilis and asymptomatic neurosyphilis patients . Given the important role of Tregs in controlling immune-mediated tissue damage , our results suggest that the CNS damage in neurosyphilis patients may be due to an uncontrolled host immune response . A local decrease in Tregs may facilitate CNS injury in neurosyphilis patients . A similar scenario has been observed in other CNS disorders [48] , [49] . T . pallidum can establish persistent infection by promoting Treg response in early stage of syphilis . In marked contrast to reduced local Treg response in symptomatic neurosyphilis , we found that Treg numbers in circulation of neurosyphilis patients were even higher than early syphilis patients without neurological involvement . This finding suggests that suppression of the systemic immune response against T . pallidum may favor neurological progression . Consistent with this notion , studies have found that HIV-positive people infected with T . pallidum are more likely to develop neurosyphilis , even during the early stages of infection [5] , [50] . The mechanisms underlying Treg differences among syphilis patients are poorly understood . Given that TGF-β was implicated in modulating Treg differentiation and activity [37] , [38]; we investigated whether the frequency and functional status of Tregs were associated with this cytokine . We confirmed that serum from the patients with secondary and serofast syphilis did express significantly higher levels of TGF-β than those of healthy control subjects , which may be related to the increased frequency and enhanced function of Tregs in these patients . Lower TGF-β levels were observed in CSF of symptomatic neurosyphilis patients than asymptomatic neurosyphilis patients , which may be associated with a decrease in CSF Treg numbers . We propose a model to summarize the role of T cell subsets in the pathogenesis of syphilis in Figure 4 . T . pallidum penetrates through abraded skin where antigen presenting cells ( APC ) , such as dendritic cells ( DC ) , process the bacteria and then migrate to the subcutaneous lymph nodes . These activated APC [51] may present T . pallidum-derived antigens to naïve T cells and induce production of Th1 [9] and Treg [18] , which enter the peripheral blood and circulate widely throughout the body . T . pallidum has the ability to preferentially enhance the generation of Tregs through TGF-β [18] , which may impair Th1 function to favor bacterial persistence in the circulation and skin . Antigenic variation and poor antigenicity also enable T . pallidum to evade cell mediated immune response [13] , [15] , [16] . However , a defective accumulation of Tregs in the CNS ( Table 3 ) may fail to suppress T cell-mediated inflammation and tissue damage in the meninges and parenchyma of brain and spinal cord , resulting in neurological symptoms and signs . In our study cohort , there are differences in sex distribution among syphilis patients of different stages: 78 . 0% neurosyphilis patients ( 78/100 ) were male , while only 35 . 1% serofast syphilis patients ( 34/97 ) were male . However , there was no significant difference in blood and CSF CD4+ CD25high Tregs between males and females in each group ( data not shown ) , which indicating that the Treg differences between stages were not due to gender preference . Serofast status represents a clinical challenge for treatment of syphilis . There is no universally accepted definition of “serofast” . The definition of “serofast” in this manuscript is “having had an appropriate 4-fold titer decline after treatment , but not reverting to seronegative” . Although these syphilis patients meet criteria for being adequately treated , we and others have shown that such “serofast” patients can progress to neurosyphilis [52] , [53] , suggesting that they still harbor T . pallidum . The immune status of serofast patients is unclear . A recent study reported that HIV-infected patients are at increased risk for serofast state after treatment [54] . Our results showed that these patients had enhanced circulating Treg numbers and suppressive function , also suggesting serofast status may be associated with a systemic immune suppression . There are several limitations in the analysis of Treg activity in this study . First , future studies should examine Foxp3 expression and define the functional status of CD4+ CD25high Tregs in CSF in neurosyphilis patients . We were not able to conduct such studies because of the limited availability of CSF T cells . In addition , studies of Treg loss-of-function and gain-of-function are needed to further explore their role in syphilis , but these experiments have been hampered by inherent difficulty in conducting immunologic studies of syphilis in experimental animal models [9] . In conclusion , our findings demonstrate for the first time that neurological progression in syphilis patients is associated with increased circulating Tregs and CSF CD4+ T cells and reduced local Treg response is implicated in the development of symptoms in neurosyphilis patients .
|
Syphilis , caused by the bacterium Treponema pallidum , can progress to affect the central nervous system ( CNS ) and cause damage in the brain and spinal cord , which is called neurosyphilis . While many affected neurosyphilis patients may not have any symptoms , some of the patients will develop severe symptoms that can be life-threatening . Regulatory T cells ( Tregs ) are a subpopulation of CD4+ T cells functioning in suppression of immune-mediated bacterial clearance and tissue damage . In this study , we conduct a comparative analysis of regulatory T cells ( Tregs ) in the blood and cerebrospinal fluid ( CSF ) of syphilis patients without neurological abnormalities , and neurosyphilis patients with or without symptoms . Our results show that neurosyphilis patients had higher Treg percentage in blood than syphilis patients without neurological involvement , suggesting that neurological progression in syphilis patients is associated with an increase in blood Treg numbers . Strikingly , a decrease in Treg percentage and numbers in CSF of symptomatic neurosyphilis patients was observed compared to asymptomatic neurosyphilis patients . These results may implicate reduced CNS Treg response as a factor underlying the development of symptoms in some neurosyphilis patients . Our findings highlight a discordant Treg response in blood and CSF in symptomatic neurosyphilis patients and further underscore the fascinating complexity of immune response in syphilis .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2013
|
Regulatory T Cells in Peripheral Blood and Cerebrospinal Fluid of Syphilis Patients with and without Neurological Involvement
|
Many genes are recruited to the nuclear periphery upon transcriptional activation . The mechanism and functional significance of this recruitment is unclear . We find that recruitment of the yeast INO1 and GAL1 genes to the nuclear periphery is rapid and independent of transcription . Surprisingly , these genes remain at the periphery for generations after they are repressed . Localization at the nuclear periphery serves as a form of memory of recent transcriptional activation , promoting reactivation . Previously expressed GAL1 at the nuclear periphery is activated much more rapidly than long-term repressed GAL1 in the nucleoplasm , even after six generations of repression . Localization of INO1 at the nuclear periphery is necessary and sufficient to promote more rapid activation . This form of transcriptional memory is chromatin based; the histone variant H2A . Z is incorporated into nucleosomes within the recently repressed INO1 promoter and is specifically required for rapid reactivation of both INO1 and GAL1 . Furthermore , H2A . Z is required to retain INO1 at the nuclear periphery after repression . Therefore , H2A . Z-mediated localization of recently repressed genes at the nuclear periphery represents an epigenetic state that confers memory of transcriptional activation and promotes reactivation .
The subnuclear localization of DNA has important roles in regulating transcription [1 , 2] . In particular , localization of chromatin near the nuclear periphery has well-documented effects on transcription . Heterochromatin and developmentally repressed genes localize at the nuclear periphery in metazoan cells , and peripheral localization promotes silencing of telomeres and the mating type loci in yeast [1 , 3–5] . Conversely , recent studies have shown that certain genes are conditionally recruited to the nuclear periphery when transcriptionally activated in both yeast and mice [6–13] . The yeast genes INO1 and GAL1 distribute randomly within the nucleoplasm under repressing conditions , but become co-localized with the nuclear periphery upon activation [6 , 7] . Live-cell four-dimensional imaging experiments reveal that recruitment is associated with both a change in the subnuclear distribution of genes and a reduction in their mobility , resulting in constrained movement near the nuclear envelope [9 , 14] . Chromatin immunoprecipitation experiments suggest that these and many other transcriptionally active genes physically interact with components of the nuclear pore complex ( NPC ) and associated factors [7] . The mechanism and functional significance of peripheral localization is unclear . Interaction of GAL1 with the nucleoporin Nup2 requires the Gal4 activator , but does not require the SAGA histone acetylase complex , and is not affected by inactivation of RNA polymerase II [13] . Thus , the association with the NPC and , presumably , recruitment of these genes to the nuclear periphery are regulated upstream of TBP binding and transcription . Furthermore , artificial tethering at the nuclear periphery promotes transcriptional activation of the INO1 gene [6] and the HXK1 gene [8] , and is sufficient to activate certain reporter genes [11] . Thus , recruitment to the nuclear periphery appears to have a functional role in promoting transcriptional activation . In contrast , recruitment of genes to the nuclear periphery has also been suggested to reflect coupling between transcription and mRNA export . Chromatin immunoprecipitation studies suggest that the interaction of mating pheromone–induced genes with the NPC is mediated by the mRNA [12] . Likewise , recruitment of HXK1 and GAL1 to the nuclear periphery is affected by sequences in the 3′ UTR [8 , 15] , and recruitment of GAL1 requires proteins that have been implicated in mRNA export [9] . These results raise the possibility that recruitment of genes to the nuclear periphery might simply be the product of physical interactions between nascent transcripts , the mRNA export machinery , and the NPC . Using a quantitative chromatin localization assay [6] , we find that the transcriptional activation of both INO1 and GAL1 genes in yeast is biphasic , with the mRNA levels increasing dramatically after gene recruitment is complete . RNA polymerase II activity was not required for peripheral recruitment of INO1 . Furthermore , when cells were shifted from activating to repressing conditions , INO1 and GAL1 remained localized at the nuclear periphery for generations . We find that localization at the periphery defines a distinct , heritable state that marks recently repressed genes and promotes reactivation . The reactivation of GAL1 was more rapid in cells that had previously activated the gene , even after six generations of repression . The rate of activation of INO1 was accelerated when the gene was artificially tethered to the nuclear envelope and was delayed in a mutant blocked for gene recruitment . Epigenetic mechanisms of transcriptional memory are employed extensively during metazoan development to stably propagate transcriptional states [16] . Such memory can be mediated by DNA methylation [17] , by histone H3 acetylation and methylation [18 , 19] or by incorporation of variant histone H3 . 3 [20] . We find that the histone variant H2A . Z was specifically required for reactivation of recently repressed INO1 and GAL1 , but had no role in the activation of the long-term repressed states of these genes . H2A . Z associated with nucleosomes in the promoter of the recently repressed INO1 gene , but not in the promoter of either activated or long-term repressed INO1 . Finally , we find that H2A . Z is essential for retention of recently repressed INO1 at the nuclear periphery . These results identify a new form of chromatin-based transcriptional memory and highlight an important role for H2A . Z in regulating subnuclear localization to mark recently repressed genes and promote their reactivation .
To determine whether gene recruitment to the nuclear periphery requires transcription , we used a chromatin localization assay [6] . This is a quantitative assay for localization of genes at the nuclear periphery based on a system developed by Belmont , Murray , and colleagues [21 , 22] . An array of 128 lac repressor–binding sites is targeted for integration to a location in the yeast genome by homologous recombination . The array can then be localized as a green fluorescent protein ( GFP ) -labeled spot in cells expressing the lac repressor tagged with GFP ( Lac I-GFP ) . Cells within a population are individually analyzed by confocal microscopy and scored as either peripheral , if the Lac I-GFP co-localizes with the nuclear envelope ( marked by the endoplasmic reticulum/nuclear envelope membrane protein Sec63-myc ) , or nucleoplasmic , if the Lac I-GFP does not co-localize with the nuclear envelope [6] ( Figure 1A ) . The URA3 gene , which distributes randomly within the nucleus , co-localizes with Sec63-myc in 27%–30% of cells [6] ( Figure 1A ) . This represents the baseline for this assay ( indicated with a hatched blue line in all relevant figures in this work; [6] ) . When the INO1 gene is artificially tethered to the nuclear envelope , we observe peripheral localization in 81% ± 7% of cells [6] . Therefore , the dynamic range of the chromatin localization assay is between 25% and 80% . For this reason , data from chromatin localization experiments were plotted on an axis from 20%–80% . The repressed INO1 gene distributes randomly , co-localizing with the nuclear envelope in 31% ± 1% of cells in the population ( Figure 1A , +inositol; [6] ) . The activated INO1 gene is recruited to the nuclear periphery , co-localizing with the nuclear envelope in 60% ± 5% of cells in the population ( Figure 1A , −inositol; [6] ) . We used the chromatin localization assay to compare the change in the peripheral localization of INO1 with the change in transcription after shifting cells from repressing to activating conditions . We quantified the levels of INO1 mRNA relative to ACT1 mRNA by reverse transcriptase real-time quantitative PCR ( RT Q-PCR ) . After shifting cells into medium lacking inositol , INO1 mRNA levels increased slowly for the first 2 . 5 h ( Figure 1B , left panel ) . The mRNA levels then increased more rapidly over the next several hours and reached steady state after 5–6 h ( unpublished data ) . Recruitment of INO1 to the nuclear periphery was rapid . The fraction of cells in which INO1 localized to the nuclear periphery increased approximately 10% in the first 5 min after shifting cells to the activating condition and was complete after 60 min . Therefore , INO1 recruitment to the periphery occurred prior to the rapid accumulation of mRNA . However , plotting the data on a logarithmic scale revealed that there was a substantial fold increase in the concentration of the mRNA during this time , consistent with the possibility that mRNA production might lead to recruitment ( Figure 1B , right panel ) . We conclude that ( 1 ) INO1 was activated quickly , resulting in an approximately 50-fold increase in the mRNA level over the first 45 min , ( 2 ) recruitment of INO1 to the nuclear periphery correlated with this early increase , and ( 3 ) the maximal rate of INO1 mRNA accumulation occurred after relocalization was complete . We next adapted the chromatin localization assay to compare the localization and transcriptional activation of the GAL1 gene , which is repressed in cells grown in glucose and expressed in cells grown in galactose . We integrated the lac repressor–binding site array downstream of the GAL1 gene and quantified its co-localization with the nuclear envelope as in Figure 1A . Repressed GAL1 localized at the nuclear periphery in 35% ± 1% ( five replicates of 30–50 cells ) of cells , and activated GAL1 localized at the nuclear periphery in 70% ± 2 . 5% ( three replicates of 30–50 cells ) of cells ( unpublished data ) . When cells were shifted from glucose to galactose , GAL1 mRNA levels increased slowly for the first 60 min and then more rapidly , reaching steady state after approximately 2 h ( Figure 1C ) . Like INO1 , GAL1 was recruited to the nuclear periphery rapidly , increasing approximately 15% in the first 5 min after shifting cells to galactose medium ( Figure 1C ) . Peripheral localization increased to 56% ± 2% after 60 min ( Figure 1C ) and continued to increase to 70% over the next 2 h ( Figure S1 ) . Therefore , like INO1 , the rate of accumulation of GAL1 mRNA was fastest after recruitment to the nuclear periphery . We next tested how localization to the nuclear periphery changed after repressing transcription ( Figure 2 ) . Both GAL1 and INO1 are repressed rapidly [23 , 24] . After addition of inositol to cells expressing INO1 , the mRNA levels decreased quickly , with no lag phase , and returned to the fully repressed level within 30 min ( Figure 2A ) . Likewise , in cells shifted from galactose to glucose , the GAL1 mRNA levels dropped rapidly , with no lag phase ( Figure 2C ) . However , both INO1 and GAL1 remained localized at the nuclear periphery for more than 2 h after repressing transcription ( Figure 2B and 2D ) . This persistent localization at the nuclear periphery suggested that these genes are actively retained . The rapid relocalization of both genes upon shifting cells to activating conditions ( Figure 1 ) indicates that they are capable of rapidly changing their distribution . Furthermore , the diffusion coefficient of repressed GAL1 is approximately 0 . 18 μm2/min [9] . This mobility would predict that , in the absence of an active mechanism of retention , GAL1 should assume a random distribution within minutes of shifting the cells from activating to repressing conditions . To rule out the possibility that the localization of INO1 to the nuclear periphery after repressing transcription was due to very low levels of transcription , we analyzed the localization of INO1 after inactivating a temperature-sensitive version of RNA polymerase II . RNA polymerase II–mediated transcription is blocked within 5 min after shifting rpb1–1 mutant cells to the non-permissive temperature ( Figure S2 and [25] ) . We grew rpb1–1 cells in the absence of inositol to activate INO1 expression , and then shifted the cells to the non-permissive temperature and quantified the localization of INO1 to the nuclear periphery over time . After 30 min at the non-permissive temperature , the INO1 gene remained localized to the nuclear periphery in 60% ± 3% of the cells , despite a 5-fold decrease in INO1 mRNA levels ( Figure 2E and Figure S2A ) . Therefore , ongoing transcription is not required to maintain INO1 at the nuclear periphery . To test if transcription is required to establish INO1 recruitment , we inactivated RNA polymerase II for 15 min before shifting cells into the activating condition . This treatment completely blocked INO1 activation , resulting in an approximately 420-fold difference in the levels of INO1 mRNA ( Figure S2B ) . In the absence of RNA polymerase II function , the INO1 gene was still recruited rapidly to the nuclear periphery ( Figure 2F ) . These results indicate that transcription is not required for either the establishment or maintenance of gene recruitment to the nuclear periphery . This conclusion is consistent with studies of the interaction of the nucleoporin Nup2 with the GAL1 promoter [13] . We next tested if the lingering localization of INO1 and GAL1 at the nuclear periphery was inherited . We quantified the peripheral localization of both INO1 and GAL1 in cells that had repressed transcription through several cell divisions . Cells were maintained in logarithmic growth by continual dilution , and their doubling time was approximately 110 min . The INO1 gene remained localized at the nuclear periphery in more than 50% of the cells after 6 h of repression and then returned to a random distribution after 12 h ( Figure 3A ) . Therefore , localization of INO1 at the nuclear periphery was maintained through at least three to four cell divisions . The retention of the GAL1 gene was even more stable , remaining localized at the nuclear periphery in more than 60% of cells after 12 h of repression ( Figure 3A ) . This suggests that GAL1 is maintained at the nuclear periphery indefinitely in logarithmically growing cells . Consistent with this indefinite switch , we find that GAL1 remained localized at the nuclear periphery for greater than 120 h , or approximately 65 generations ( Figure S3 ) . Therefore , the localization of INO1 and GAL1 at the nuclear periphery is stably maintained after repressing transcription and is inherited by subsequent generations , suggesting that it represents an epigenetic state . Our data suggest that there are two different forms of repressed INO1 and GAL1 . Whereas INO1 and GAL1 that have been repressed for many generations distribute randomly within the nucleus , recently repressed INO1 and GAL1 localize at the nuclear periphery . Therefore , peripheral localization distinguishes between recently repressed and long-term repressed states . This raised the possibility that localization might function as an epigenetic marker to allow cells to “remember” recent transcription of these genes , potentially affecting their rate of reactivation . To test this idea , we compared the rate of transcriptional activation of long-term repressed and short-term repressed GAL1 . The rate of reactivation of GAL1 in cells in which the gene had been repressed for 12 h ( six to seven generations ) was much more rapid than in cells grown continuously in glucose ( Figure 3B ) . Thus , in a culture in which only approximately 1% of the cells have previously experienced galactose , the reactivation of the GAL1 gene is enhanced . We next compared the rate of activation of long-term repressed INO1 to the rate of reactivation of short-term repressed INO1 after 3 h of repression ( ∼1 . 5 generations ) . In contrast to GAL1 , the reactivation of the INO1 gene after 3 h of repression was delayed compared with activation of the long-term repressed gene ( Figure 4A ) . However , this rate of reactivation was clearly enhanced by the localization at the nuclear periphery . Nup2 , a component of the nuclear pore complex that physically associates with transcriptionally active genes such as GAL1 [7 , 13] , is required for recruitment of both INO1 and GAL1 to the nuclear periphery ( Figure 4B ) . Mutants lacking Nup2 exhibited a delay in the reactivation of INO1 ( Figure 4C ) , suggesting that recruitment to the nuclear periphery promotes more rapid reactivation . To determine if recruitment to the nuclear periphery is sufficient to promote activation , we compared the activation of INO1 that was artificially tethered to the nuclear envelope to untethered INO1 . The lac repressor array was integrated beside INO1 in strains expressing either the wild-type Lac I-GFP ( untethered INO1 ) or a modified version possessing an FFAT motif to target the protein to the nuclear envelope ( tethered INO1; [6 , 26] ) . Expressing this form of the lac repressor results in efficient targeting of the INO1 gene to the nuclear envelope [6] . Tethering the INO1 gene to the nuclear envelope had no effect on steady-state levels of INO1 mRNA under activating or repressing conditions ( Figure S4 ) . However , tethered INO1 was activated more rapidly than untethered INO1 ( Figure 4D ) . Therefore , localization at the nuclear periphery enhances the rate of reactivation of INO1 . If so , then why is the rate of reactivation of recently repressed INO1 slower than the rate of activation of long-term repressed INO1 ? This difference is likely due to differences in the physiology of cells grown continuously in the presence of inositol and cells to which inositol has recently been added . Activation of INO1 is regulated by the concentration of phosphatidic acid , a lipid precursor of phosphatidylinositol [27] . Phosphatidic acid consumption is stimulated by both exogenous inositol and the action of the Ino1 enzyme [27] . After repressing INO1 transcription , the Ino1 enzyme in the cells will continue to produce inositol , driving a higher flux through the pathway and depleting phosphatidic acid . We think this may explain the longer lag phase in the reactivation experiment , which represents the time required for phosphatidic acid to accumulate to levels that activate transcription . This feedback , combined with the shorter duration of the memory phenomenon for the INO1 gene , complicates a direct comparison between the rate of activation between the short- and long-term repressed states of INO1 . To explore the molecular nature of the difference between short-term and long-term repressed INO1 , we asked if remaining at the nuclear periphery after repression affects the chromatin state of the gene . Because nucleosome remodeling is important for both INO1 activation and repression [28–32] , we compared the positioning of nucleosomes within the short-term repressed and long-term repressed INO1 promoter . Permeablized cells were treated with micrococcal nuclease for various times to digest unprotected DNA ( Materials and Methods ) . As an internal control for nucleosome protection , we used a known , well-positioned nucleosome within the GAL1 promoter ( GAL NB; Figure 5A; [33] ) and an adjacent , non-nucleosomal sequence ( GAL I; Figure 5A ) . Using Q-PCR to define the concentration of these two sequences in our digestion , we observed protection of the nucleosomal sequence relative to the non-nucleosomal sequence ( Figure 5A , left panel ) . Furthermore , after 15 min and 30 min of digestion , we observed the production of clear mononucleosome and dinucleosome bands , indicating that nucleosomes were providing protection from the nuclease and that linker DNA had been digested ( Figure 5A , right panel , arrows ) . Previous studies have established that relative nucleosomal protection is observable over a large range of digestion and with or without gel purification of mononucleosomes [34] . Therefore , we used Q-PCR and a set of 27 different primer pairs to amplify overlapping 80–100 base pair fragments from the INO1 promoter ( Table S1 ) . The concentration of each of the templates for these 27 primer pairs was quantified after 30 min of digestion . The protection of each template was calculated relative to the GAL NB sequence . Using this method , we identified one well-positioned nucleosome within the INO1 promoter and one at the junction between the promoter and the transcript ( Figure 5B ) . Comparison between short-term and long-term repressed INO1 revealed no significant change in the positioning of these nucleosomes . However , we did observe a decrease in the relative protection provided by these two nucleosomes in the short-term repressed state ( Figure 5C ) . This difference resulted in a 2-fold decrease in the protection at these two sites relative to the GAL NB site . This difference may reflect either an increase in the fraction of cells in the population in which these nucleosomes are absent , or a change in the stability of these nucleosomes in the lysates subjected to nuclease digestion . The positioning of the pair of nucleosomes present in the INO1 promoter suggested that they might contain the histone H2A variant H2A . Z . H2A . Z is incorporated into pairs of nucleosomes that are typically found in the promoters of repressed genes , and incorporation of H2A . Z has been proposed to promote more rapid activation [35–38] . However , genome-wide chromatin immunoprecipitation experiments did not demonstrate a strong association of H2A . Z with the long-term repressed INO1 promoter [37] . Yeast H2A . Z is encoded by the non-essential HTZ1 gene [39] . To test if H2A . Z is important for transcriptional memory , we compared the rates of reactivation of recently repressed INO1 and GAL1 in wild-type and htz1Δ mutant cells ( Figure 6 ) . Loss of H2A . Z led to a strong delay in the rate of reactivation of both short-term repressed INO1 and short-term repressed GAL1 ( Figure 6A and 6E ) . Surprisingly , loss of H2A . Z had no effect on the rate of activation of long-term repressed INO1 or GAL1 ( Figure 6B and 6F ) . These results suggest that H2A . Z plays an important and specific role in the reactivation of these genes . H2A . Z is exchanged for H2A within intact nucleosomes by the SWR1 ATPase complex [40–42] . To test if SWR1 plays a role in the H2A . Z-dependent reactivation of INO1 , we next tested the effect of loss of SWR1 on INO1 activation and reactivation . We find that swr1Δ mutant strains were also defective for reactivation of recently repressed INO1 ( Figure 6C ) , and had little effect on the activation of long-term repressed INO1 ( Figure 6D ) . To examine the deposition of H2A . Z nucleosomes in the INO1 promoter , we used chromatin immunoprecipitation with antiserum against Htz1 . Consistent with previous work , immunoprecipitation of H2A . Z from either long-term repressed cells or the activated cells gave low recovery of the INO1 promoter ( Figure 7A ) . In contrast , immunoprecipitation of H2A . Z from recently repressed cells gave a clear enrichment for the INO1 promoter ( Figure 7A ) , suggesting that H2A . Z is specifically incorporated into promoter nucleosomes in the recently repressed state . We next tested if H2A . Z had any role in the localization of the INO1 gene . Loss of H2A . Z had no effect on recruitment of activated INO1 to the nuclear periphery ( Figure 7B ) . This is not surprising since the histone variant generally associates with repressed genes ( Figure 7A; [35–38] ) . However , cells lacking H2A . Z were unable to retain INO1 at the nuclear periphery after repressing transcription ( Figure 7C ) . Therefore , H2A . Z nucleosomes in the recently repressed INO1 promoter function both to retain recently repressed INO1 at the nuclear periphery and to promote optimal reactivation .
Our results show that the recruitment of genes to the nuclear periphery is a rapid , active process that is independent of transcription . The most robust transcription of the GAL1 and INO1 genes occurred after these genes had fully relocalized to the nuclear periphery , suggesting that recruitment to this subnuclear environment allows optimal expression of these genes . Furthermore , both genes remained at the periphery for generations after repressing transcription , suggesting that cells can inherit localization information . Retention of the INO1 gene and optimal reactivation of both INO1 and GAL1 required the histone variant H2A . Z , which associated with nucleosomes within the recently repressed INO1 promoter . Thus , cells have both molecular and cellular sources of memory of past transcriptional activation , and they are able to pass on this information to their progeny . This type of memory is mediated by local changes in chromatin structure that mark recently repressed genes to alter their transcriptional potential and localization , and perhaps to provide a mechanism for inheritance . What is the functional significance of this epigenetic memory ? In the case of the GAL1 gene , the recently repressed state is reactivated much more rapidly than the long-term repressed state , which presumably confers an adaptive advantage upon cells that have previously grown in galactose . We do not see this for INO1 , perhaps because physiological differences between recently repressed and long-term repressed cells complicates the comparison of the rate of INO1 activation and reactivation . However , we can conclude that ( 1 ) there are two distinct states of repressed INO1 and GAL1 , distinguishable by their localization , their transcriptional histories , and the molecular requirements for activation , ( 2 ) localization of INO1 at the nuclear periphery is necessary and sufficient to promote more rapid activation , and ( 3 ) incorporation of H2A . Z is the molecular mechanism of transcriptional memory , retaining INO1 at the nuclear periphery and promoting reactivation of both INO1 and GAL1 . Histone variant H2A . Z is enriched in pairs of nucleosomes within the promoters of repressed genes [35–38] . The histone appears to play an important role in the loss of nucleosomes from promoters upon their activation [37] . This observation , coupled with the fact that H2A . Z nucleosomes are less tightly bound to DNA than H2A nucleosomes , suggests that H2A . Z nucleosomes promote activation by being more easily removed [37] . We find that H2A . Z deposition and function can depend on the transcriptional history of the promoter into which it is incorporated . H2A . Z is required for rapid reactivation of short-term repressed INO1 and GAL1 and for retention of recently repressed INO1 at the nuclear periphery . It is possible that these results represent an indirect effect of loss of H2A . Z . However , we think H2A . Z most likely plays a direct role in promoting the reactivation of INO1 and GAL1 because ( 1 ) loss of H2A . Z ( and SWR1 ) affects reactivation of recently repressed INO1 and GAL1 , but not the activation of long-term repressed INO1 and GAL , 1 and ( 2 ) H2A . Z physically associates with the recently repressed INO1 promoter . Therefore , we have identified a new and novel role for this histone variant: H2A . Z can serve as a molecular identifier of recently repressed genes to promote their retention at the nuclear periphery and their rapid reactivation . Our current model for the mechanism of gene recruitment and transcriptional memory is shown in Figure 8 . In response to signals that regulate transcriptional activation , genes physically interact with the nuclear pore complex via the mobile nucleoporin Nup2 . Recruitment to the nuclear periphery allows access to the optimal subnuclear environment for transcription and , potentially , for mRNA export . After transcription is repressed , previous transcriptional activation of genes such as INO1 and GAL1 is remembered through retention in this optimal environment . Localization at the nuclear periphery is epigenetically inherited and requires incorporation of histone variant H2A . Z . Finally , the reactivation of INO1 and GAL1 is optimized by both localization at the periphery and through more rapid loss of H2A . Z nucleosomes [37] . What is the role of DNA localization in promoting transcriptional memory ? Our data suggest two possible models for how peripheral localization affects H2A . Z-mediated transcriptional memory . In the first model , H2A . Z incorporation into promoter nucleosomes is promoted by Nup2-mediated gene recruitment to the nuclear periphery , and functions to promote reactivation by altering the rate of nucleosome loss or local histone modifications . This model is consistent with several observations in the literature . Tethering of Nup2 to DNA promotes “boundary activity , ” insulating euchromatin from the spread of heterochromatin [43 , 44] . Intriguingly , one of the most dramatic phenotypes of mutants lacking either Nup2 or H2A . Z is the spread of silenced heterochromatin [43 , 45] . Thus , it is possible that tethering genes to the nuclear periphery through Nup2 leads to the incorporation of H2A . Z nucleosomes , which functions as a boundary . Furthermore , it is possible that boundary elements normally associate with the NPC . We find that H2A . Z is involved in both the activation of recently repressed genes and their retention at the nuclear periphery . Thus , a second model for the importance of H2A . Z is that H2A . Z nucleosomes promote reactivation of recently repressed genes by retaining them in the optimal environment for transcriptional activation . These models are not mutually exclusive , and we favor the possibility that H2A . Z incorporation is promoted by localization and that , once incorporated , H2A . Z affects localization . Transcriptional memory is employed extensively during development in multi-cellular organisms . In Drosophila , Hox gene expression throughout development is determined early in embryogenesis by transcriptional regulators that control segmentation [46] . The initial expression states defined by the segmentation genes are maintained by the action of either polycomb group proteins ( generally repressive ) or trithorax group proteins ( generally activating ) through a number of chromatin-based mechanisms such as nucleosome positioning and histone modification [47] . Similarly , the variant histone H3 . 3 is incorporated selectively into transcriptionally active parts of the genome , which may promote the epigenetic maintenance of an activated state [20 , 48] . Like these forms of transcriptional memory , the transcriptional memory described here is mediated by chromatin-based changes that mark recently repressed genes and distinguish them from long-term repressed genes . However , unlike these forms of memory , which serve to maintain a previously established transcriptional state , the transcriptional memory described here serves an informational role , revealing previous transcriptional activity and altering the transcriptional potential of previously expressed genes . Previous work has hinted that transcriptional activity of GAL1 can alter the degree of methylation of histone H3 , marking the chromatin for hours after repressing transcription [19] . However , in this case , the mark was lost after cell division . Our data suggest that the past experiences of microbial organisms can affect their cellular organization and their physiology for many generations . The efficiency of inheritance of the memory state was different for the two genes we examined , suggesting that there are different timing mechanisms for each . In the case of the GAL1 gene , after exposure to galactose , logarithmically growing cells appeared to undergo an indefinite switch to the recently repressed state . It will be fascinating to determine if there are conditions or stimuli that can reset the GAL1 gene to the long-term repressed state . In contrast , the transcriptional memory of INO1 activation was relatively short lived . The previous transcriptional state of INO1 is imprinted in its chromatin and its subnuclear localization for 6 h or more ( two to three cell doublings ) , but this information is eventually lost . Why do cells optimize reactivation of genes ? We speculate that rapid reactivation of certain genes confers an adaptive , and therefore an evolutionary , advantage to cells . This might be particularly important in the case of stress-responsive genes such as INO1 or genes involved in metabolizing non-glucose hexose sugars . Also , epigenetic mechanisms may be useful in allowing cells to alter their transcriptional output rapidly under highly variable environmental conditions or under physiological circumstances in which they rapidly undergo reversible changes in physiology [49] . It will be interesting to see if this mechanism is also operative in metazoan organisms , perhaps to establish epigenetically “primed” states for dynamically regulated genes in response to transient physiological or environmental cues .
Unless stated otherwise , chemicals were from Sigma ( St . Louis , Missouri , United States ) , oligonucleotides were from Operon ( Huntsville , Alabama , United States ) , restriction enzymes were from New England Biolabs ( Ipswitch , Massachusetts , United States ) , yeast media components were from Q-Biogene ( Irvine , California , United States ) , antibodies against GFP and myc were from Invitrogen/Molecular Probes ( Carlsbad , California , United States ) , and antiserum against Htz1 was from Abcam ( Cambridge , Massachusetts , United States ) . Yeast strains used in this study are listed in Table 1 . Except for BY4741 , BY4741 htz1 , Δ and BY4741 swr1Δ [50] , all strains were constructed from CRY1 ( ade2–1 can1–100 his3–11 , 15 leu2–3 , 112 trp1–1 ura3–1 MATa ) [51] . Strain JBY451 is the product of a cross between JBY376 ( ade2–1 can1–100 his3–11 , 15 leu2–3 , 112 trp1–1 ura3–1 INO1:LacO128:URA3 HIS3:LacI-GFP MATa ) and BY4742 nup2Δ mutant strain ( his3Δ1 leu2Δ0 lys2Δ0 ura3Δ0 nup2Δ::Kan̂r MATα ) from the genome-wide null mutant collection [50] . Random spores JBY451-r1 and JBY451-r7 were selected . For JBY451-r7 , the identity of the ura3 allele was confirmed to be ura3–1 by transforming JBY451-r7 with StuI-digested pRS306 [52] . Strain JBY462 was created by transforming JBY451-r1 with pRS304-Sec63-myc digested with NheI . Strain JBY467 was created by transforming JBY451-r7 with p6LacO128GAL1 and pRS304-Sec63-myc . Finally , strains JBY451-r1 , JBY451-r7 , JBY462 , and JBY467 were confirmed to be nup2Δ by PCR from genomic DNA . Strain JBY461 is the product of a cross between JBY397 [6] and JCY218 [53] . Random spores were selected that were Ura+ Trp+ His+ and temperature sensitive for growth ( JBY461-r2 ) . These were then visually scored for expression of Lac I-GFP . Strain DBY50 is the product of a cross between DBY49 ( htz1Δ::His5+ ade2–1 can1–100 his3–11 , 15 leu2–3 , 112 trp1–1 ura3–1 MATα ) and JBY397 . The resulting diploid was sporulated , and tetrads were dissected to generate DBY50 . Plasmids p6LacO128 [6] , p6LacO128-INO1 [6] , pAFS144 [21] , and pAFS144-FFAT [6] have been described . To create the plasmid p6LacO128-GAL1 to mark the GAL1 gene with the lac repressor–binding site array , the 3′ end of the GAL1 gene , and downstream sequences were amplified by PCR using the following primers ( 5′ to 3′ ) : GAL1up , GTTCAAACCGCAGTTGAAGG and GAL1down , CCGAAAGATCTTCTCTATGGGG . The resulting PCR product was cloned into the TOPO4 vector ( Invitrogen ) . This was then cloned into p6LacO128 as a SpeI fragment . The plasmid was integrated downstream of GAL1 by digestion with NruI . Plasmid pRS304-Sec63-myc was created by amplifying SEC63-13myc from JBY397 genomic DNA using the following primers ( 5′ to 3′ ) : SEC63up: GTATTTCGGAGAGGGGGC; Pringledown: ACTATACCTGAGAAAGCAACCTGACCTACA . The resulting PCR product was TOPO cloned into pCR2 . 1 ( Invitrogen ) . The insert was then cloned into pRS304 [52] as a NotI-KpnI fragment . The plasmid was digested with NheI to target integration at SEC63 . Unless noted otherwise , all experiments were performed on cells grown in synthetic complete medium at 30 °C . For experiments involving INO1 , cells were grown in medium lacking inositol or supplemented with 100 μM myo-inositol . For experiments involving GAL1 , cells were grown in media with either 2% glucose or 2% galactose . RNA was prepared as described [54] . A total of 2–4 μg of DNAse-treated total RNA was reverse transcribed using 5 μM Oligo dT and 20 units of Superscript III reverse transcriptase ( Invitrogen ) at 42 °C for 1 h . The reaction was diluted 5-fold , and 1/20th was used for Q-PCR . The sequences of the primers used for real-time PCR were ( 5′ to 3′ ) : INO1CDS F , TAGTTACCGACAAGTGCACGTACAA; INO1CDS R TAGTCTTGAACAGTGGGCGTTACAT; ACT1CDS F , GGTTATTGATAACGGTTCTGGTATG; ACT1CDS R , ATGATACCTTGGTGTCTTGGTCTAC; GAL1CDS F , GTTCGATTTGCCGTTGGACGG; GAL1CDS R , GGCAAACCTTTCCGGTGCAAG . The relative concentration of cDNA templates for both the target gene ( INO1 or GAL1 ) and the control gene ( ACT1 ) were calculated for each sample using standard curves for each primer set that were defined by linear regression analysis of Ct values using a series of 5-fold dilutions of yeast genomic DNA covering a 3 , 125-fold range . Long-term repressed cells were harvested at an optical density ( OD600 ) of 0 . 8–1 . 0 from 1 l of SDC + inositol . Short-term repressed cells were grown in 1 l of SDC − inositol to an OD600 of 0 . 7 , and inositol was added to 100 μM . After 1 h of repression , cells were harvested by filtration . Cell permeablization and micrococcal nuclease digestion were performed as described , except that DNA was not size selected [55] . Q-PCR analysis on digested DNA was performed using the oligonucleotides listed in Table S1 . To map the protected sequences onto the INO1 promoter , we used the experimentally determined transcriptional start site and initiation codon [56 , 57] . Chromatin immunoprecipitation experiments were performed using anti-Htz1 antiserum ( Abcam ) as described [37] , with the following modifications: 2 μg of anti-Htz1 were used to immunoprecipitate Htz1 from 4 . 8 mg of chromatin , and immunoprecipitates were recovered using Protein G-dynabeads ( Invitrogen ) . Immunoprecipitated DNA was recovered and analyzed by Q-PCR as described [6] . Recovered INO1 promoter was expressed relative to recovered ACT1 coding sequence .
|
Eukaryotic cells control the spatial arrangement of chromosomes; the localization of genes can both reflect and contribute to their transcriptional state . A number of genes in the simple eukaryote brewer's yeast are “recruited” to the nuclear periphery through interactions with the nuclear pore complex when they are expressed . The functional significance of peripheral recruitment is unclear . Here , we show that recruited genes are actively retained at the periphery for generations after transcription is repressed . This suggests that localization at the nuclear periphery represents a novel inherited state that might allow simple eukaryotic organisms to “remember” previous transcriptional activation . This type of memory allows for more robust reactivation of genes , suggesting that it is adaptive . Finally , both retention at the nuclear periphery and rapid reactivation require a variant form of histone H2A . Adaptive memory is distinct from other types of transcriptional memory . In developmental memory , transcriptional states established by transcriptional regulators early in embryogenesis are propagated long after these regulators have disappeared . Adaptive memory does not propagate a state , but represents a novel state that serves as a source of information . In this way , it resembles a rudimentary form of cellular learning that allows cells to benefit from recent experience .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology",
"yeast",
"and",
"fungi",
"eukaryotes",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] |
2007
|
H2A.Z-Mediated Localization of Genes at the Nuclear Periphery Confers Epigenetic Memory of Previous Transcriptional State
|
The abrupt origin and rapid diversification of the flowering plants during the Cretaceous has long been considered an “abominable mystery . ” While the cause of their high diversity has been attributed largely to coevolution with pollinators and herbivores , their ability to outcompete the previously dominant ferns and gymnosperms has been the subject of many hypotheses . Common among these is that the angiosperms alone developed leaves with smaller , more numerous stomata and more highly branching venation networks that enable higher rates of transpiration , photosynthesis , and growth . Yet , how angiosperms pack their leaves with smaller , more abundant stomata and more veins is unknown but linked—we show—to simple biophysical constraints on cell size . Only angiosperm lineages underwent rapid genome downsizing during the early Cretaceous period , which facilitated the reductions in cell size necessary to pack more veins and stomata into their leaves , effectively bringing actual primary productivity closer to its maximum potential . Thus , the angiosperms' heightened competitive abilities are due in no small part to genome downsizing .
The flowering plants are highly competitive in almost every terrestrial ecosystem , and their rapid rise during the early Cretaceous period irrevocably altered terrestrial primary productivity and global climate [1–3] . Terrestrial primary productivity is ultimately determined by the photosynthetic capacity of leaves . The primary enzyme in photosynthesis , rubisco , functions poorly when CO2 is limiting , which requires leaf intercellular CO2 concentrations ( ci ) to be maintained within a narrow range [4] through adjustments in leaf surface conductance to CO2 and water vapor . This surface conductance is one of the greatest biophysical limitations on photosynthetic rates across all terrestrial plants [5 , 6] . In order for CO2 to diffuse from the atmosphere into the leaf , the wet internal surfaces of leaves must be exposed to the dry ambient atmosphere , which can cause leaf desiccation and prevent further CO2 uptake . As a consequence , increasing leaf surface conductance to CO2 also requires increasing rates of leaf water transport in order to avoid desiccation [7] . Both theory and empirical data suggest that among all major clades of terrestrial plants , the upper limit of leaf surface conductance to CO2 and water vapor is tightly coupled to the biophysical limitations of cell size [8–11] . Cellular allometry , in particular the scaling of genome size , nuclear volume , and cell size , represents a direct physical constraint on the number of cells that can occupy a given space and , as a result , on the distance between cell types and tissues [12–14] . Because leaves with many small stomata and a high density of veins can maintain higher rates of gas exchange than leaves with fewer , larger stomata and larger , less numerous veins [15] , variation in cell size can drive large changes in potential carbon gain . Without reducing cell size , increasing stomatal and vein densities would displace other important tissues , such as photosynthetic mesophyll cells [16] . Therefore , the densities of stomata on the leaf surface and of veins inside the leaf are inversely related to the sizes of guard cells and the primary xylem elements comprising them . While numerous environmental and physiological factors can influence the final sizes of somatic eukaryotic cells , the minimum size of meristematic cells and the rate of their production are strongly constrained by nuclear volume , more commonly measured as genome size [17–19] . Among land plants , the bulk DNA content of cells varies by three orders of magnitude , with the angiosperms exhibiting both the largest range in genome size and the smallest absolute genome sizes [20] . Whole-genome duplications and subsequent genomic rearrangements , including genome downsizing , are thought to have directly contributed to the unparalleled diversity in anatomical , morphological , and physiological traits of the angiosperms [12 , 21–28] . We extend this prior work and test the hypothesis that genome size variation is responsible not only for gene diversification but also directly limits minimum cell size and , thus , is the underlying variable constraining stomatal size and density and leaf vein density ( Dv ) . Due to the strong influence of cell size on maximum potential carbon gain , the allometric scaling of genome size and cell size is predicted to directly influence primary productivity across all major clades of terrestrial plants [12 , 13 , 27 , 29] .
To determine whether genome downsizing among the angiosperms drove the anatomical and physiological innovations that resulted in their ecological dominance , we compiled data for genome size , cell size ( guard cell length; lg ) , stomatal density ( Ds ) , and Dv for almost 400 species of ferns , gymnosperms , and angiosperms . Consistent with prior studies and with our predictions , genome size varied substantially among major clades ( Fig 1 ) and was a strong predictor of anatomical traits across the major groups of terrestrial plants even after accounting for phylogenetic relatedness of species ( Fig 2 , Table 1 ) . Species with smaller genomes have smaller , more numerous stomata and higher leaf vein densities . Genome size explained between 31% and 54% of interspecific variation in lg , Ds , and Dv across the major groups of terrestrial plants , and both phylogenetic and non-phylogenetic analyses showed that a single relationship predicted each of these traits from genome size across all species ( Table 1 ) . In both phylogenetic and non-phylogenetic analyses there were strong , significant correlations between anatomical traits both among the major clades and within the angiosperms , highlighting the coordinated evolution of these traits throughout the history of seed plants ( S1 Table ) . Because genome size directly affects minimum cell size , variation in genome size has numerous consequences for the structure and organization of cells and tissues in leaves , which directly influence rates of leaf water loss ( transpiration ) and photosynthesis . Physical resistance to diffusion across leaf surfaces is ultimately determined by the sizes of epidermal cells , and the maximum diffusive conductance of CO2 and water vapor is higher in leaves with more numerous , smaller stomata [8 , 10 , 11] . While the effects of cell size on leaf epidermal properties have been well characterized , the effects of cell size on the efficiency of liquid water supply through the leaf are , perhaps , less obvious . Because the highest hydraulic resistance in the leaf occurs in the path between the veins and the sites of evaporation , shortening this path length by increasing Dv reduces the resistance outside the xylem and increases leaf hydraulic conductance [7 , 30] . Given a constant leaf volume , increasing Dv without displacing photosynthetic mesophyll cells requires reductions in vein and conduit sizes that can only be accomplished by decreasing cell size [16 , 31] . However , smaller conduits have higher hydraulic resistances . To overcome hydraulic limitations associated with reductions in conduit size , other innovations in xylem anatomy that reduce hydraulic resistance have been hypothesized to facilitate narrower xylem conduits and high Dv . In particular , the development of low resistance end walls between adjacent cells is thought to have given angiosperms a hydraulic advantage as conduit diameters decreased . Only in angiosperm lineages with very high Dv do primary xylem have simple perforation plates , which have lower resistance to water flow than scalariform perforation plates [16] . Similarly , the low resistance of gymnosperm torus-margo pits compared to angiosperm pits can result in higher xylem-specific hydraulic conductivity for small diameter conduits [32] . In both cases , while smaller conduits have higher resistance , this potential cost has been offset by other innovations that reduce hydraulic resistance at the scale of the whole xylem network . We examined the consequences of variation in genome size on terrestrial primary productivity by calculating maximum stomatal conductance ( gs , max ) and operational stomatal conductance ( gs , op ) using theoretical and empirical models that directly relate leaf anatomy to gas exchange ( see Materials and Methods ) . Genome size was a highly significant predictor of both gs , max and gs , op , whether or not phylogenetic relatedness of species was incorporated ( Fig 3 , Table 1 ) . Scaling relationships that accounted for phylogenetic relatedness of all species in our dataset were as significant as non-phylogenetic analyses and had similar slopes . Thus , a single relationship between genome size and stomatal conductance exists among all land plants . We tested assumptions about how vein positioning in the leaf influences gs , op by modeling stomatal conductance for leaves of varying thickness and found that regardless of leaf thickness ( 70 , 100 , 130 μm ) , the slopes of the relationships between genome size and gs , op were significantly steeper than the slope of the relationship between genome size and gs , max ( all p < 0 . 001 ) . Thus , across all species , shrinking the genome brings gs , op closer to gs , max ( Fig 3 , Table 1 ) , which facilitates faster rates of growth . The timing of these physiological innovations further corroborates their role in promoting angiosperm domination of terrestrial ecosystems . Unlike other major clades of terrestrial plants , genome sizes , Dv , Ds , and lg of the angiosperms expanded into new regions of trait space during the Cretaceous period ( Fig 4 ) , increasing rates of leaf level carbon assimilation and ushering in an era of greater terrestrial primary productivity [12 , 15 , 27] . To determine how the upper or lower limits of trait values changed through time , linear and nonlinear curves were fit through the upper or lower 10% of trait values during the period of rapid angiosperm diversification ( 165–60 Ma ) . For the angiosperms , extreme values of genome size and anatomical traits were fit by a logarithmic curve better than by a linear relationship ( genome size change in Akaike Information Criterion ( ΔAIC ) = 31 . 8; Dv ΔAIC = 6 . 6; lg ΔAIC = 16 . 3; Ds ΔAIC = 5 . 7 ) , indicating that Cretaceous angiosperms pushed the frontiers of genome size , cell size , and vein and stomatal densities . In contrast to the angiosperms , fern and gymnosperm lineages exhibited no such sudden change in any trait during the Cretaceous period ( Fig 4 ) . Reconstructions of Dv matched well with fossil data , but the limited available data for lg and Ds among Cretaceous angiosperms precludes a comparable analysis ( S1 Fig ) . These results suggest that the ability to develop leaves with high vein and stomatal densities derives not exclusively from common developmental programs underlying these traits nor from genetic correlations ( i . e . , linkage between genes controlling both traits ) , but—even more fundamentally—from biophysical scaling constraints that limit minimum cell size [8 , 29] . Together with analyses of trait evolution , the scaling relationships between genome size and gas exchange rates suggest that rapid genome downsizing among the angiosperms during the Cretaceous period facilitated increased rates of photosynthesis and biomass accumulation ( Fig 2 , Fig 3 and Fig 4 ) . Importantly , while genome downsizing has been critical to increasing leaf gas exchange rates among the angiosperms , it was not a key innovation that occurred only at the root of the angiosperm phylogeny . Rather , the angiosperms exhibit a wide range of genome sizes , and coordinated changes in genome size and physiological traits have occurred repeatedly throughout their evolutionary history ( Table 1 , S1 Table ) . While whole-genome duplications have been particularly important in promoting diversification among the angiosperms [21] , larger genomes increase minimum cell size and depress maximum potential gas exchange , thereby reducing competitive ability in productive habitats . Our results suggest that reductions in minimum cell size through genome downsizing can recover leaf gas exchange capacity subsequent to genome duplications and diversification events . If heightened competitive ability among the angiosperms drove their ecological dominance , then innovations that reduced minimum cell size were critical to this transformative process [29] . Although genome size limits minimum cell size [19 , 25] , final cell size can vary widely as cells grow and differentiate . After cell division and during cell expansion , various factors influence how large a cell becomes . Intracellular turgor pressure overcomes the mechanical rigidity of the cell wall to enlarge cellular boundaries . The magnitude of turgor pressure is itself controlled by water availability around the cell and by the osmotic potential inside the cell . Final cell size is influenced , therefore , by both biotic and abiotic factors that affect pressure gradients in and around the cell . By reducing nuclear volume and the lower limit of cell size [19 , 25] , genome downsizing expands the range of final cell size that is possible . Species that can vary cell size across a wider range can more finely tune their leaf anatomy to match environmental constraints on leaf gas exchange . Indeed , Dv , Ds , lg , and gs are more variable among species with small genomes ( Fig 2 , Fig 3 and Fig 4 ) , and the variance in these traits unexplained by genome size is likely due to environmental variation . Thus , genome size may predict ecological breadth insofar as species with small genomes can exhibit greater plasticity in final cell size and can inhabit a wider range of environmental conditions , although more analyses of within- and between-species variation in genome size are needed to clarify this [33 , 34] . Interestingly , only the angiosperms occupy this region of trait space , and the angiosperms tend to be more productive than either the ferns or the gymnosperms across a broad range of environmental conditions . Therefore , rapid genome downsizing by the angiosperms during the Cretaceous period likely explains not only their greater potential and realized primary productivity ( Fig 3 and Fig 4 ) but also why they were able to expand into and create new ecological habitats , fundamentally altering the global biosphere and atmosphere [3] . Prevailing theories have suggested that the global dominance of angiosperms occurred due to higher maximum photosynthetic capacity and growth , despite Cretaceous declines of atmospheric CO2 that would have otherwise depressed rates of photosynthesis [3 , 12 , 15 , 35] . In habitats that can support high rates of primary productivity , maximum rates of gas exchange and growth are generally greater for angiosperms than for gymnosperms and ferns and are due , we show , to reductions in genome and cell sizes that occurred after the appearance of early angiosperms . Smaller genomes and cells increased leaf surface conductance to CO2 and enabled higher potential and realized primary productivity . Furthermore , because genome downsizing lowers the limit of minimum cell size , final cell size can vary much more widely , which may facilitate a closer coupling of anatomy and physiology to environmental conditions [36] . Therefore , genome downsizing among the angiosperms allowed them to outcompete other plants in almost every terrestrial ecosystem .
Published data for lg , Ds , and Dv were compiled from the literature ( S1 Data ) . Genome size data for each species were taken from the Plant DNA C-values database ( release 6 . 0 , December 2012 ) , managed by the Royal Botanic Gardens , Kew [37] . In total , our dataset comprised 393 species of vascular plants , of which 289 were angiosperms , 53 were gymnosperms , and 51 were ferns . The dataset comprised here represents 0 . 1% of the estimated angiosperm species diversity . Of the 416 families and 64 orders of extant plants currently accepted by the Angiosperm Phylogeny Group IV , the 289 species in our dataset represented 102 families and 43 orders . Among angiosperm clades , the species diversity in our dataset is positively correlated with the number of known species in those clades ( S2 Fig ) . The Plant DNA C-values database currently contains data for over 7 , 000 angiosperms , and our sample of 289 for which there were anatomical traits had genome sizes highly representative of all angiosperms in the database with no significant differences between the mean genome sizes of the two datasets ( S3 Fig ) . For the 289 angiosperms in the dataset , there were Dv data for 165 , guard cell size data for 184 , and Ds data for 184 . Similarly , there were Dv data for 23 gymnosperms and for 10 ferns , there were lg data for 20 gymnosperms and for 38 ferns , and there were Ds data for 37 gymnosperms and 26 ferns . Fossil data for Dv [38 , 39] , lg [8 , 27 , 40 , 41] , and Ds [8 , 40] were compiled from published sources ( S1 Data ) . For each species , we calculated gs , max and gs , op . gs , max is defined by the dimensions of stomatal pores and their abundance , and represents the biophysical upper limit of gas diffusion through the leaf epidermis . Anatomical measurements of guard cells were used to calculate gs , max as [8 , 9]: gs , max=DsamaxdH2Omvdp+π2amax/π ( 1 ) where dH2O is the diffusivity of water in air ( 0 . 0000249 m2s−1 ) , mv is the molar volume of air normalized to 25°C ( 0 . 0224 m3mol−1 ) , Ds is stomatal density ( mm−2 ) , amax is maximum stomatal pore size , and dp is the depth of the stomatal pore . The amax term can be approximated as: π ( lp/2 ) 2 , where lp is stomatal pore length with lp being approximated as lg/2 , where lg is guard cell length . For studies that only reported lp , we calculated lg as 2∙lp [8 , 42] . dp is assumed to be equal to guard cell width ( W ) . If W was not reported dp was estimated as 0 . 36∙lg [11] . gs , op , by contrast , more accurately defines the stomatal conductance leaves attained under natural conditions when limitations in leaf hydraulic supply constrain stomatal conductance . We used an empirical model of gs , op that directly relates Dv to stomatal conductance during periods of steady state transpiration ( E ) [7] as: E=gs , opv=KleafΔΨ ( 2 ) Kleaf=12 , 670dm-1 . 27 ( 3 ) where: dm=π/2 ( dx2+dy2 ) 1/2 ( 4 ) dx=650/Dv ( 5 ) gs , op= ( KleafΔΨ ) /v . ( 6 ) Kleaf is leaf hydraulic conductance ( mmol m−2s−1MPa−1 ) , dm is the post vein distance to stomata ( μm ) , dx is the maximum horizontal distance from vein to the stomata ( μm ) , dy is the distance from vein to the epidermis ( μm ) , ΔΨ is the water potential difference between stem and leaf ( set to 0 . 33 MPa [43] ) , and v is vapor pressure deficit set to 2 kPa . Variation in v would affect the intercept but not the slope of gs , op . In order to test the influence of variation in leaf thickness on gs , op , we used three values of dy ( 70 , 100 , and 130 μm ) . The steady state equations presented above can be related directly to photosynthesis as: Aop=E1 . 6v ( ca ( 1−cica ) ) = ( KLeafΔΨ ) 1 . 6v ( ca ( 1−cica ) ) =gs , op1 . 6 ( ca ( 1−cica ) ) ( 7 ) where Aop is operational photosynthetic capacity ( μmol m−2s−1 ) , ca is the molar concentration of CO2 in the atmosphere , ci is the molar concentration of CO2 in the air spaces inside the leaf , and 1 . 6 accounts for the difference in diffusivity between H2O and CO2 in air . To determine the temporal patterns of trait evolution , we generated a phylogeny from the list of taxa ( S1 Data ) using Phylomatic ( v . 3 ) and its stored family-level supertree ( v . R20120829 ) . To date nodes in the supertree , we compiled node ages from recent , fossil-calibrated estimates of crown group ages . Node ages were taken from Magallón et al . [44] for angiosperms , Lu et al . [45] for gymnosperms , and Testo and Sundue [46] for ferns . The age of all seed plants was taken as 330 million years [47] . Because there is some uncertainty in the maximum age of the ancestor of all angiosperms , we took the angiosperm crown age used by Brodribb and Field [12] to make our results directly comparable to theirs . We tested this assumed angiosperm age by using different ages for the crown group angiosperms ranging from 130 Ma to 180 Ma , and the results were not qualitatively different . Of the 254 internal nodes in our tree , 82 of them had ages . These ages were assigned to nodes and branch lengths between these dated nodes evenly spaced using the function “bladj” in the software Phylocom ( v . 4 . 2 [47] ) . Polytomies were resolved by randomly bifurcating and adding 5 million years to each of these new branches and subtracting an equivalent amount from the descending branches so that the tree remained ultrametric . For all subsequent analyses of character evolution , this method for randomly resolving polytomies was repeated 100 times to account for phylogenetic uncertainty . For ancestral state reconstructions , the ages and character estimates at each node were averaged across the 100 randomly resolved trees . Ancestral state reconstructions were calculated using the residual maximum likelihood method , implemented in the function “ace” from the R package ape [48] . To determine when changes in traits pushed the frontiers of trait values , the upper ( Dv and Ds ) and lower ( genome size and lg ) limits of traits were estimated by first extracting the upper or lower ten percent of reconstructed trait values in sequential 5 million-year windows and then attempting to fit curves to these values . This method is similar to a previous analysis of Dv evolution through time [38] , which is included here for comparison . We compared three types of curve fits: a linear fit that lacked slope ( equivalent to the mean of the reconstructed trait values ) , a linear fit that included both a slope and an intercept , and a nonlinear curve of the form trait = a + b / ( 1 + e^[− ( time + c ) / d] ) . Curves were fit to reconstructed trait values for each clade between 165 and 60 Ma , which corresponds to the time period encompassing the major diversification and expansion of the angiosperms , and the best fit was chosen based on AIC scores with a difference in AIC of 5 taken to indicate significant differences in fits . Phylogenetic generalized least squares regression was used to determine whether traits underwent correlated evolution . A regression was performed for each pairwise combination of traits for only species with data for both traits . Phylogenetic regressions used a Brownian motion correlation structure from the R package ape [49] . We acknowledge the potential for high uncertainty in ancestral state character reconstructions when working with a small subset of species relative to the broader species pool [50 , 51] . In an effort to minimize uncertainty , we sampled basal angiosperms as much as possible and performed two additional analyses that suggest our dataset is robust to incomplete sampling . First , we performed a bootstrapping analysis in which we randomly sampled species from our entire genome size dataset ( 35% , 52% , and 78% of angiosperm species ) , reconstructed genome size , and fit curves to the lower limit of reconstructed genome sizes , as before . This procedure was replicated 100 times at each level of sampling diversity . This analysis revealed that using only 35% of the angiosperms in our dataset still produced estimates of minimum genome size that are consistent with the entire dataset ( S4 Fig ) . Second , the species diversity of 20 named nodes in our dataset is strongly correlated with the actual extant species diversity of those clades ( S2 Fig ) . Additionally , our sample of genome size variation does not differ significantly from the genome size variation among approximately 7 , 000 measured species ( S3 Fig ) . Furthermore , our analysis of vein density evolution based on 151 angiosperm species is almost identical to the previous analysis by Brodribb and Feild [12] , which relied on 504 angiosperm species ( Fig 4 ) , and both of these modeled limits of vein density agree strongly with fossil data [38] . Overall , these analyses strongly suggest that the trait values represented in our taxon sampling is robust , given the incredible extant diversity of angiosperms and the data currently available . Scaling relationships between genome size and Dv , lg , gs , max , and gs , op were calculated from log-transformed data and analyzed using the function “sma” in the R package smatr [52] . Analyses were performed for the entire dataset and also for individual clades . Slope tests were used to determine whether the scaling relationship between genome size and gs , max was significantly different from the relationship between genome size and gs , op and whether the scaling relationships between genome size and gs , op and gs , max differed among clades . To account for the non-independence of sampling related species , phylogenetic standard major axis regressions were performed on all species using the function “phyl . RMA” in the R package phytools .
|
The angiosperms , commonly referred to as the flowering plants , are the dominant plants in most terrestrial ecosystems , but how they came to be so successful is considered one of the most profound mysteries in evolutionary biology . Prevailing hypotheses have suggested that the angiosperms rose to dominance through an increase in their maximum potential photosynthesis and whole-plant carbon gain , allowing them to outcompete the ferns and gymnosperms that had previously dominated terrestrial ecosystems . Using a combination of anatomy , cytology , and modelling of liquid water transport and CO2 exchange between leaves and the atmosphere , we now provide strong evidence that the success and rapid spread of flowering plants around the world was the result of genome downsizing . Smaller genomes permit the construction of smaller cells that allow for greater CO2 uptake and photosynthetic carbon gain . Genome downsizing occurred only among the angiosperms , and we propose that it was a necessary prerequisite for rapid growth rates among land plants .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] |
[
"biotechnology",
"taxonomy",
"plant",
"anatomy",
"stem",
"anatomy",
"ferns",
"stomata",
"phylogenetics",
"plant",
"science",
"data",
"management",
"plant",
"genomics",
"plants",
"flowering",
"plants",
"leaf",
"veins",
"computer",
"and",
"information",
"sciences",
"plant",
"genetics",
"leaves",
"short",
"reports",
"evolutionary",
"systematics",
"eukaryota",
"genetics",
"biology",
"and",
"life",
"sciences",
"gymnosperms",
"genomics",
"evolutionary",
"biology",
"plant",
"biotechnology",
"organisms"
] |
2018
|
Genome downsizing, physiological novelty, and the global dominance of flowering plants
|
Exposure to Plasmodium falciparum is associated with circulating “atypical” memory B cells ( atMBCs ) , which appear similar to dysfunctional B cells found in HIV-infected individuals . Functional analysis of atMBCs has been limited , with one report suggesting these cells are not dysfunctional but produce protective antibodies . To better understand the function of malaria-associated atMBCs , we performed global transcriptome analysis of these cells , obtained from individuals living in an area of high malaria endemicity in Uganda . Comparison of gene expression data suggested down-modulation of B cell receptor signaling and apoptosis in atMBCs compared to classical MBCs . Additionally , in contrast to previous reports , we found upregulation of Fc receptor-like 5 ( FCRL5 ) , but not FCRL4 , on atMBCs . Atypical MBCs were poor spontaneous producers of antibody ex vivo , and higher surface expression of FCRL5 defined a distinct subset of atMBCs compromised in its ability to produce antibody upon stimulation . Moreover , higher levels of P . falciparum exposure were associated with increased frequencies of FCRL5+ atMBCs . Together , our findings suggest that FCLR5+ identifies a functionally distinct , and perhaps dysfunctional , subset of MBCs in individuals exposed to P . falciparum .
Naturally acquired immunity is vital in reducing morbidity and mortality from Plasmodium falciparum malaria in endemic areas , where some individuals receive hundreds of infectious mosquito bites per year . Humoral responses to P . falciparum may be a critical component of this immunity , and P . falciparum-specific memory B cells ( MBCs ) are likely important in the development and maintenance of an effective response [1–3] . Unfortunately , protection from symptomatic disease takes many years to develop , during which time children living in endemic areas experience multiple episodes of symptomatic malaria , resulting in over half a million deaths annually [4–8] . One possible explanation for the slow and incomplete development of immunity to malaria is that chronic exposure to P . falciparum alters the immune response in ways that interfere with the development of protective B cell responses [9] . In particular , P . falciparum exposure has been associated with higher frequencies of circulating CD21-CD27- “atypical” memory B cells ( atMBCs ) [10–17] . These cells are distinct in their surface phenotype , and possibly function , from CD21+CD27+ classical memory B cells ( MBCs ) , which are capable of undergoing a recall response that includes proliferation and differentiation into antibody-secreting cells . The surface phenotype of atMBCs exhibits commonalities with a subset of dysfunctional B cells found in viremic HIV patients . These cells express inhibitory receptors , such as FCRL4 and SIGLEC6 , that block their ability to undergo recall in response to mitogenic stimuli [18–20] . In addition to malaria and HIV , nonclassical MBC phenotypes have been identified in the context of other chronic diseases such as common variable immunodeficiency ( CVID ) , systemic lupus erythematosus ( SLE ) , and HCV [21–26] , and they bear similarities to B cells found in the tonsils of healthy individuals [27 , 28] . This has led to the notion that atMBCs might represent a functionally inhibited state that results from chronic antigen exposure [11 , 12] , in analogy to the induction of exhaustion in T cells [29 , 30] . Malaria-associated atMBCs were originally reported in individuals living in Mali [11] , and their association with increasing exposure to P . falciparum has been corroborated in several studies using distinct cohorts from different geographical locations [10–17] . Although this association is increasingly well established , there are limited available data on the function of atMBCs in the context of malaria [11] . A recent study of atMBCs concluded that they are capable of producing P . falciparum-specific antibodies found in the serum [31] , suggesting that these cells are not dysfunctional but rather may play an important role in host protection . However , this study did not define atMBCs with markers to specifically exclude antibody-producing plasmablasts , which may confound findings of antibody production . Importantly , the conclusion of antibody production was also based on indirect evidence correlating circulating antibody fragments with atMBC-encoded repertoires , which does not exclude the alternative possibility that circulating antibodies were produced by other B cell subsets . Thus , whether atMBCs are capable of producing antibody remains unclear . A more global investigation of the functional programs expressed in malaria-associated atMBCs would help to define their role in immunity . To this end , we performed microarray-based transcriptome analysis of highly purified atMBCs from Ugandan children . Using paired comparisons to classical MBC transcriptomic profiles from the same individuals , we present a detailed examination of the functional programming of these cells . We demonstrate that atMBCs express FCRL5 , but not FCRL4 as reported in other studies , and that expression of FCRL5 is associated with a poor capacity for antibody production . Our findings provide unique insights into the functional programming of these nonclassical MBCs and the nature of B cells in immunity to malaria .
A number of studies have established an association between higher frequencies of atMBCs and increasing exposure to P . falciparum [10–17] , but the functional programming of these cells remains poorly characterized . Consistent with prior reports , we found that the frequencies of circulating atMBCs in individuals from our cohort living in a high P . falciparum transmission region in Uganda were higher than in malaria-naïve controls , and increased with age ( S1 Fig ) . To better understand differences between atMBCs and classical MBCs , we performed microarray-based whole transcriptome comparisons of atMBCs to classical MBCs within asymptomatic parasitemic individuals living in areas of intense P . falciparum transmission . Sort-purified class-switched atMBCs ( CD3-CD14-CD19+CD10-CD27-CD21-IgD-IgG+ ) and classical MBCs ( CD3-CD14-CD19+CD10-CD27+CD21+IgD-IgG+ ) were processed for whole human transcriptome microarray analysis using previously described methods [32 , 33] . Differential gene expression analysis demonstrated that atMBCs express a transcriptional repertoire distinct from that of classical MBCs . Using a false discovery rate of 3% and a 1 . 5-fold change threshold , we identified 2226 differentially expressed probes representing 1479 unique genes ( S1 Table ) . Approximately 60% of these genes were more highly expressed in atMBCs than classical MBCs . Functional enrichment analysis demonstrated significant differences in categories related to multiple B cell functions ( Fig 1 ) . For example , atMBCs exhibited lower expression of genes associated with co-stimulation of BCR signaling , such as CD79b , CD70 , CD24 , and CD44 . This was accompanied by higher expression of regulators of BCR signaling ( LILRB2 , ITGAX ) , Fc receptor family inhibitory receptors ( FCRLA , FCRL3 , FCRL5 ) , and genes known to promote B cell anergy and exhaustion ( SIGLEC6 , PDCD1 , LGALS1 ) . Together , the differences in regulation of these genes are suggestive of cell-intrinsic down-modulation of BCR signaling in atMBCs . Genes involved in apoptosis , particularly those related to p53 signaling , were expressed at lower levels in atMBCs than classical MBCs . For example , HIPK2 , a pro-apoptotic protein that phosphorylates p53 in response to DNA damage [34–36] , exhibited lower expression in atMBCs . Other pro-apoptotic genes with lower relative expression in atMBCs included TP53INP1 , which promotes cell cycle arrest and apoptosis [37]; TNFSF10 ( TRAIL ) , a gene target in the p53 cell death pathway [38]; PERP , a mediator of p53-dependent apoptosis [39]; and TNFRSF25 ( Death Receptor 3 ) , which functions similarly to CD95 ( Fas ) , with over-expression leading to NF-κB induction and apoptosis [40] . We concomitantly detected higher expression of TNFRSF1B and IL21R , both of which can promote B cell survival [41–44] . Together , suppressed expression of these pro-apoptotic factors could promote the survival of atMBCs , suggesting one mechanism by which they might accumulate with increasing parasite exposure . To better understand the relationship of atMBCs to nonclassical memory B cell subsets found in other disease contexts , we collated data from diverse studies characterizing the mRNA and protein expression levels of signature genes in these cells [11 , 18 , 20–28 , 31 , 45 , 46] ( Fig 1 ) . The direction of gene expression in malaria-associated atMBCs relative to classical MBCs corresponded well with gene expression patterns of other nonclassical memory B cell subsets; specifically , 88% of the changes occurred in the same direction , with the highest proportion of overlap occurring with CD27-CD21- cells in HIV ( 89% , 21 of 23 genes ) and CD21lo cells in CVID ( 97% , 32 of 33 genes ) . Functional overlap extended to most categories , with the notable exception of apoptosis . Together , these data suggest that in addition to similarity in surface phenotypes , atMBCs may exhibit functional similarity to nonclassical memory B cells associated with other chronic diseases . Notably , we detected a decrease in expression of CXCR3 in atMBCs , despite reports that this marker is increased on malaria-associated atMBCs and similar cells in the tonsil and in individuals with HIV , SLE , and CVID [11 , 20–22 , 27 , 45] . We did not detect a relative increase in expression of FAS ( CD95 ) , though this has been reported for cells in the tonsil and in individuals with HIV , SLE , and CVID [21–23 , 28 , 45] . Other genes previously described to be differentially expressed in similar B cells from other contexts , but not detected in our microarray analysis , included LAIR1 , CXCR4 , and the genes encoding caspase-1 ( CASP1 ) and caspase-9 ( CASP9 ) , which further distinguishes malaria-associated atMBCs with reports from HIV , SLE , and CVID [18 , 22 , 23 , 25 , 26 , 28] . Thus , although there are abundant commonalities between malaria-associated atMBCs and cells of similar surface phenotype associated with other diseases , there are also unique aspects that differentiate malaria-associated atMBCs from other exhausted and nonclassical memory B cell subsets ( S2 Table ) . A key functional phenotype of exhausted MBCs found in HIV-viremic individuals is their decreased ability to differentiate into antibody-secreting cells [19 , 20] , leading early reports to propose that malaria-associated atMBCs might be similarly dysfunctional [11] . Consistent with this , we observed that atMBCs expressed higher levels of SIGLEC6 and BCL6 , which negatively regulate B cell proliferation and differentiation [19 , 47 , 48] . Similarly , PDCD1 , which encodes the signaling regulator PD-1 [49] , was more highly expressed in atMBCs than in classical MBCs . Surprisingly , we also observed that atMBCs express higher levels of PRDM1 ( the gene encoding BLIMP-1 ) , a regulator of plasmablast differentiation which acts in opposition to BCL6 . This raised the possibility that plasmablasts comprised a subset of these CD21-CD27-IgG+ cells , a phenotype previously used to define atMBCs [11 , 31] . To test this hypothesis , we examined spontaneous antibody production from CD20+ and CD20- subsets in the absence of stimulation , which is a property of antibody-secreting cells such as CD20- plasmablasts . We found that among CD20+ atMBCs ( CD19+IgG+CD10-CD27-CD21-CD20+ ) , only 1 . 6% of cells spontaneously secreted IgG ex vivo ( Fig 2A ) . In contrast , 18% of cells with a similar surface phenotype but lacking expression of CD20 spontaneously secreted IgG . These CD19+IgG+CD10-CD27-CD21-CD20- cells also expressed high levels of CD38 , which is consistent with the surface phenotype of plasmablasts/plasma cells ( Fig 2B ) . We found that on average , 2 . 6% of the cells within the CD19+IgG+CD10-CD27-CD21- gate were CD20- and CD38hi . Therefore , to distinguish atMBC from this minor population of likely plasmablasts , we incorporated CD20 and CD38 into all analyses below , defining atMBCs as CD19+ CD20+ CD21- CD27- CD38int/lo IgG+ . Among the genes we identified as relatively enriched in atMBCs , only 6 ( 0 . 5% ) were identified as being enriched in plasmablasts in a previous study [50] . Thus , the likely inclusion of a small number of plasmablasts along with atMBCs was unlikely to have significantly affected our microarray results . The surface phenotype of atMBCs is most commonly defined by the absence of expression of CD21 and CD27 . In accord with protein levels , transcripts of both CR2 ( the gene encoding CD21 ) and CD27 were significantly lower in atMBCs than classical MBCs , indicating that down-regulation of the expression of these markers occurs , at least in part , at the level of transcription . Previous studies also described differential expression of protein levels of CD85j , CD11c , CXCR5 , CD24 , CD84 , and CD319 [11 , 13 , 31] , which we corroborated at the transcript level as differential expression of LILRB1 , ITGAX , CXCR5 , CD24 , CD84 , and SLAMF7 , respectively ( Fig 1 ) . Notably , these markers represent high confidence signatures , given that they have been identified as markers of atMBCs at both the mRNA and protein levels in studies of distinct cohorts performed by different laboratories . In addition to the above , we detected significantly increased expression of LILRB2 ( CD85d ) , TNFRSF1B ( CD120b ) , and IL21R ( CD360 ) in atMBCs relative to classical MBCs . Expression of LILRB2 and TNFRSF1B was previously reported to be increased in exhausted B cells during HIV infection [18] , and LILRB2 and its encoded protein , CD85d , were expressed in CD21lo B cells from patients with combined variable immunodeficiency ( CVID ) [23] . We corroborated the expression of CD85d , CD120b , and CD360 at the protein level on samples from our highly P . falciparum-exposed individuals by surface staining of atMBCs ( Fig 3A ) . As in previous studies , we also found CD11c protein to be significantly increased on the surface of atMBCs relative to classical MBCs . The Ig-beta chain of the BCR , encoded by CD79B , is required for proper trafficking of the BCR; diminished expression of CD79B in atMBCs would be predicted to result in lower levels of surface-localized BCR . As previously reported by others [31 , 51] , we found surface IgG levels to be significantly lower on atMBCs than classical MBCs from the same individuals ( Fig 3A ) , consistent with down-modulation of surface BCR . FCRL4 protein was previously reported to be expressed on malaria-associated atMBCs [13 , 31] , and elevated gene and/or protein expression has been reported for HIV-associated exhausted MBCs [18 , 20] , tonsillar B cells [27 , 28] , and nonclassical memory B cells associated with CVID and hepatitis C infection [22 , 26 , 46] . Surprisingly , we did not detect significantly increased expression of FCRL4 by atMBCs in our microarray analysis . Quantitative RT-PCR analysis of FCRL3 , FCRL4 , and FCRL5 corroborated the microarray data , demonstrating that FCRL3 and FCRL5 , but not FCRL4 , transcripts were present at higher levels in atMBCs than classical MBCs ( S2 Fig ) . FCRL3 and FCRL5 share 28–60% extracellular amino acid sequence identity with FCRL4 [52] , suggesting that antibodies used to detect surface-localized FCRL4 in other studies might have cross-reacted with other FCRL family members . To test this possibility , we assessed the specificity of various anti-FCRL antibodies using cell lines constitutively expressing FCRL4 or FCRL5 [53] . Consistent with the original study that produced these antibodies [53] , the anti-FCRL4 antibody clone 1A3 and the anti-FCRL5 antibody clone 7D11 bound specifically to the expected cell lines ( Fig 3B ) . In contrast , the widely used anti-FCRL4 antibody clone 2A6 [27] , which was employed in previous malaria-associated atMBCs studies [11 , 31] , bound strongly to both FCRL4- and FCRL5-expressing cell lines . Thus , the 2A6 antibody binds to both FCRL4 and FCRL5 , whereas the 1A3 and 7D11 antibodies are specific for FCRL4 and FCRL5 , respectively . Having determined the specificity of these antibodies , we measured the surface expression of FCRL4 and FCRL5 on MBCs from 8–10 year old children and adults from our high exposure Ugandan cohort; all selected subjects were smear positive for P . falciparum but lacked fever . Consistent with previous reports [11 , 31] , the nonspecific 2A6 clone labeled atMBCs more strongly than classical MBCs ( Fig 3C ) . Similar results were seen with the anti-FCRL5 antibody 7D11 . In contrast , the anti-FCRL4 antibody 1A3 failed to exhibit binding above an isotype control background to either atMBCs or classical MBCs . Given that these protein level data are consistent with our microarray and qRT-PCR observations that FCRL5 , but not FCRL4 , is more highly expressed by malaria-associated atMBCs than classical MBCs , it is likely that FCRL5 is the actual target recognized on these cells by previous malaria studies that used clone 2A6 [11 , 31] . FCRL5 expression followed a heterogeneous distribution on atMBCs ( Fig 4A ) . The proportion of atMBCs that were FCRL5+ between individuals was variable ( mean 53% , range 18–74% ) , but was consistently higher than the proportion of classical MBCs that were FCRL5+ in the same individual ( mean 23% , range 10–52%; p < 0 . 001 ) ( Fig 4B ) . Given the non-uniformity in FCRL5 expression on atMBCs , we considered the possibility that FCRL5+ atMBCs might represent a distinct subset from FCRL5- atMBCs . To assess this possibility , we compared the surface phenotypes of FCRL5- and FCRL5+ atMBCs and classical MBCs . Compared to FCRL5- atMBCs , the FCRL5+ subset expressed significantly higher levels of FCRL3 , CD19 , and CD20 , but lower levels of CD21 , with no significant difference in either CD27 or IgG expression ( Fig 4C ) . Similar trends were observed for classical MBCs , with the exception that CD21 was unchanged . These findings are consistent with FCRL5- and FCRL5+ cells being distinct , though perhaps developmentally and/or functionally related , subsets of memory B cells . A number of studies have reported that the frequency of atMBCs increases with age and P . falciparum exposure [11–13 , 15 , 16] . If exposure induces phenotypic changes in atMBCs consistent with reduced responsiveness , we predicted that the FCRL5+ subpopulation of atMBCs would similarly increase with P . falciparum transmission intensity . To test this hypothesis , we compared expression of FCRL5 on atMBCs from study participants living in Nagongera , Uganda , where transmission is very high , to those from Walukuba , a periurban area of Uganda where transmission is ~30 fold lower [54 , 55] . Subjects living in the area of higher malaria transmission had a significantly higher proportion of FCRL5+ atMBCs than subjects living in the area with lower transmission ( Fig 5; mean difference of 25% , p = 0 . 004 by Wilcoxon rank-sum test and in multivariate regression including age ) . Having shown that CD20+ atMBCs are poor spontaneous producers of antibody ex vivo , we evaluated their capacity to differentiate into antibody-secreting cells following stimulation ( i . e . , recall ) . Given the heterogeneity in FCRL5 expression in atMBCs and the potential inhibitory role of this surface receptor [56] , we also evaluated whether this surface marker distinguished subsets with different capacities to undergo recall . FCRL5- and FCRL5+ subsets of atMBCs and classical MBCs were isolated by flow cytometry and stimulated for 4 d in vitro with an activating anti-BCR antibody and CpG to induce a recall response . Following stimulation , FCRL5- classical MBCs exhibited robust production of antibody as expected , with a mean of 6 . 3% of these cells capable of secreting IgG ( Fig 6 ) . In comparison , FCRL5- atMBCs exhibited reduced capacity to produce antibody ( 3 . 4% IgG-secreting cells ) , though this difference did not reach statistical significance . More strikingly , FCRL5 expression defined strongly inhibited subsets of both classical and atMBCs , with only 1 . 1% of FCRL5+ classical MBCs and 0 . 2% of FCRL5+ atMBCs capable of a recall response . Of note , FCRL5- atMBCs produced a higher proportion of IgG-secreting cells than FCRL5+ classical MBCs . Thus , expression of FCRL5 , more so than the traditional subset-defining markers , strongly delineates functionally distinct groups of memory B cells and is correlated with inhibition of antibody production .
We have performed a detailed molecular characterization of malaria-associated atMBCs , beginning with an unbiased transcriptome-wide comparison with classical MBCs and leading to functional characterization of atMBC subsets defined by differential expression of FCRL5 . We show that in comparison to classical MBCs , atMBCs obtained from individuals living in an area of intense malaria transmission in Uganda have a distinct transcriptional program , with down-modulated BCR signaling that may contribute to reduced function , and changed apoptosis programs which may contribute to accumulation . This analysis reveals new surface markers that identify atMBCs , particularly FCRL5 , which we show is a specific correlate of poor recall capacity . Based on the surprising finding that FCRL5 , but not FCRL4 , was enriched for expression in atMBCs , we confirmed that an anti-FCRL4 antibody used in many prior studies cross-reacts with FCRL5 . The extent to which these molecules have been confused in the literature is unclear , and it is certainly possible that FCRL4 is expressed by some analogous B cell subsets given that increased gene expression and functional studies of FCRL4 perturbation have been reported [19 , 22 , 28 , 46 , 57] . A re-examination of antibody specificity is warranted to determine if in some studies , the functional consequences of FCRL5 expression might have been missed and/or ascribed to FCRL4 as a result of non-specific recognition or perturbation . Interestingly , some evidence suggests that FCRL5 is a receptor for IgG , which circulates at high levels during malaria [58 , 59] . Thus , FCRL5 expression on B cells could participate in a feedback mechanism for IgG homeostasis during hypergammaglobulinemia , thereby impacting memory B cell responses . In accord with studies of an analogous subset in HIV-viremic individuals [19 , 20] , we find that atMBCs in malaria-exposed individuals are comparatively ineffective at producing antibody ex vivo , either spontaneously or following re-stimulation . These findings contrast with those of a recent study which concluded that atMBCs actively produce protective antibodies in vivo [31] . However , the authors reached this conclusion based on the indirect observations that transcripts of secretory IgG , along with membrane IgG , were detected in atMBCs , and that BCR sequences from some atMBCs matched those of serum IgG fragments in a single subject . In light of our findings , two alternative possibilities that could explain the detection of secretory transcripts are that: a ) these transcripts were derived from a minority population of CD20- plasmablasts and not atMBCs; or b ) transcripts were derived from CD20+ atMBCs but these cells were not actively producing antibody , possibly due to transient or permanent arrest of differentiation by inhibitory molecules such as FCRL5 . To additionally explain the detection of overlapping repertoires in serum IgG and atMBCs [31] , we suggest the possibilities that: a ) atMBCs do not themselves produce antibody , but at some frequency can eventually differentiate into antibody secreting cells; or b ) atMBCs do not produce antibody nor do they differentiate into antibody secreting cells , but antibody secreting cells and atMBCs share antibody repertoires [60] due to derivation from a common progenitor such as classical MBCs . In any case , CD20+ atMBCs have a relative decrease in the capacity to secrete antibody in response to stimulation versus classical MBCs . However , this difference is modest compared with the marked decrease seen in the FCRL5+ subsets of either MBC population . Based on the magnitude of the effect , the traditional subset markers that distinguish atMBCs from classical MBCs ( CD21 and CD27 ) are less effective than FCRL5 in defining a functionally distinct subset . This raises the question of how best to consider the relationships of the various sub-populations , and suggests the possibility that up-regulation of FCRL5 expression precedes down-regulation of CD21 and/or CD27 , in a progression through which MBCs adopt a state of reduced antibody production . This model is also in accord with a very recent report that BCR variable region sequences in atMBCs are largely indistinguishable from those found in classical MBCs [60] . Further experimentation to better define the relationships between these subsets is urgently needed , as this will have an important influence on our thinking about the ontogeny and function of these populations . Consistent with down-modulation of B cell functions , increasing evidence suggests that higher levels of exposure to P . falciparum induce immunoregulatory processes that dampen infection-associated immune activation [10 , 32 , 61] . This development of immunological tolerance might underlie the decreasing severity of malaria disease with increasing exposure and age , but may come at the expense of inhibiting sterilizing immunity . Similar to upregulation of expression of immunoregulatory receptors on γδ T cells [32] , we show here that FCRL5 expression on B cells is associated with higher levels of exposure to P . falciparum . In turn , FCRL5 is associated with poor antibody production , suggesting that upregulation of this receptor may be a mechanism of cell-intrinsic immunoregulation . We note , however , that atMBCs are associated with increasing age and exposure to P . falciparum , the same factors which are associated with acquired immunity [11–13 , 15 , 16 , 62] . Acquired immunity allows individuals , like those studied here , to remain asymptomatic while parasitemic , not to receive antimalarial therapy , and therefore to remain infected . It is possible that immune activation associated with this state of asymptomatic parasitemia in part drives the accumulation of atMBCs and affects aspects of their phenotype , such as the expression of FCRL5 . Frequencies of atMBC appear to decrease following elimination of P . falciparum exposure [10 , 16] , but further studies will be required to assess the dynamics of atMBC frequency and phenotype in response to acute and chronic P . falciparum infection . It remains to be determined whether atMBCs are truly dysfunctional , with immunity being acquired despite their accumulation; play an immunoregulatory role , aiding in the development of tolerance to P . falciparum infection; or have an as yet undefined role in anti-parasite immunity , e . g . , antigen presentation . Further functional studies will also be required to elucidate the roles of FCRL5 and other similarly expressed immunoregulatory molecules in this process . Given the similarities between atMBCs and similar B cell subsets found in other contexts of chronic antigen exposure , such as HIV infection , HCV infection , SLE , and CVID [11 , 18–26 , 31 , 45 , 46] , it may be that these cells are not so “atypical” at all . These subsets all share a similar biomarker phenotype ( CD19+CD21lo/-CD27- ) and are all hypothesized or demonstrated to have refractory responses to B cell mitogens . In addition to functional and biomarker similarities , we found that many of their gene expression signatures were also shared , including similarities in expression of immunoregulatory receptors , proteins involved in migration , and BCR co-stimulatory transcripts , which were down-regulated . However , key differences from other studies were also observed , especially with regard to B cell trafficking and survival [11 , 19 , 21–23 , 27 , 45] . It is possible that the differential expression of these markers is rooted in ontogeny; however , these markers could also reflect contextual differences , such as those driven by tissue localization , kinetics , or differences in the antigenic and/or inflammatory environment . Further studies will be needed to better define the relationships of these populations to one another through detailed functional and global transcriptomic analyses . In summary , comparison of the gene expression of malaria-associated atMBCs vs . classical MBCs highlights key differences in these subsets and provides a foundation for comparison with analogous subsets seen in other conditions of chronic antigen exposure . High expression of FCRL5 defines distinct subsets of MBCs and appears to be a key marker of functional deficiency , at least with respect to the ability to secrete antibody in response to stimulation . Further studies of the function of these cells will be required to define their relevance to disease and immunity .
Ethical approval was obtained from the Makerere University School of Medicine Research and Ethics Committee , the Uganda National Council for Science and Technology , the London School of Hygiene & Tropical Medicine Ethics Committee , and the University of California , San Francisco Committee on Human Research . All adult study participants provided written informed consent , and a parent or guardian of all child participants provided written informed consent on their behalf . Samples were obtained from participants enrolled in cohort studies as part of the East African International Center for Excellence in Malaria Research in Uganda . These cohorts of children aged 6 months—10 years of age and their adult primary caregivers were followed for all their health care needs in dedicated study clinics as previously described [55] . Samples for the majority of experiments came from parasitemic , but non-symptomatic , children 8–10 years old and adult caregivers from the Nagongera cohort in Tororo District , where malaria transmission is very high ( annual entomological inoculation rate ~ 310 infectious bites per person per year ) [54 , 55] . To compare phenotypes in different malaria transmission settings , samples were also analyzed from the Walukuba cohort in Jinja District where transmission is lower ( annual entomological inoculation rate ~ 2 . 8 infectious bites per person per year ) . Older children and adults were selected since individuals in these age ranges have previously been shown to have the highest frequencies of atMBCs [11 , 13] . Subjects without fever were selected to avoid transient effects on B cell function associated with inflammation from symptomatic malaria or other acute illness . Subjects with documented parasitemia by microscopy were selected to keep subjects as similar to each other as possible; asymptomatic parasitemia is common in older children and adults in high transmission settings and those without documented parasitemia may or may not have had submicroscopic parasitemia . For transcriptomic analysis of atMBCs , samples were selected from children aged 8–10 . Approximately ten million cryopreserved PBMCs from each child were stained with antibodies specific for CD3 ( clone UCHT1 ) , CD14 ( clone M5E2 ) , CD19 ( clone HIB19 ) , CD10 ( clone HI10a ) , CD38 ( clone HIT2 ) , CD27 ( clone O323 ) , CD21 ( clone B-ly4 ) , IgG ( clone G18-145 ) ( all BioLegend ) ; and IgD ( clone IA6-2 ) ( BD Biosciences ) ( see S3 Fig for gating strategy ) . Classical and atMBCs were processed for microarray analysis as previously described [32 , 33] . In brief , cell subsets were isolated to >99 . 8% purity using two successive rounds of purity-optimized sorting on a FACSAria , with 5 , 000 total cells on the second round sorted directly into 100 μl RNAqueous Micro lysis buffer . RNA was isolated with the RNAqueous Micro kit ( Life Technologies ) , and was amplified in two rounds with the Amino Allyl MessageAmp II kit ( Life Technologies ) . Amplified RNA was covalently labeled with Cy3 and hybridized to SurePrint G3 Unrestricted GE 8x60K human V2 gene expression microarrays ( Agilent Technologies ) . Microarrays were scanned on an Agilent microarray scanner at 3 μm resolution into a 20-bit TIFF , and raw intensities were extracted with Agilent Feature Extraction . Raw intensities were log2-transformed and quantile-normalized using the R package limma [63] . Probes not expressed above background ( normalized intensity of 128 ) in either sample group were removed from the data set . Significantly differentially expressed genes were identified using Significance Analysis for Microarrays in a paired comparison using a false discovery rate of 3% and 1 . 5-fold change threshold [64] , and expression values were median centered across samples for visualization as heat maps . Functional enrichment analysis was performed using DAVID [65] , using a Benjamini-corrected p value of 0 . 05 to determine significance . All microarray data are available in the NCBI Gene Expression Omnibus under accession number GSE64493 . We evaluated data from 14 studies that reported transcriptional or protein differences in nonclassical MBCs compared to controls in the contexts of malaria , HIV , CVID , SLE , and HCV , as well as data for tonsillar B cells , where FCRL4+ B cells were first described [11 , 19 , 21–28 , 31 , 45] . Studies for comparison were identified by searching PubMed for “FCRL4 B cells” or “FCRL5 B cells” [11 , 18 , 20 , 22 , 23 , 31] . We then searched on the diseases identified from “FCRL4 B cells” and “FCRL5 B cells” search terms to include studies with relevant transcript and protein information in B cells that did not specifically identify FCRL4 or FCRL5 [21 , 25 , 45] . Nonclassical MBCs were defined differently between studies as follows: CD21lo/-CD27- in the context of P . falciparum exposure , HIV , CVID , and HCV cirrhosis [11 , 20 , 22 , 23 , 31 , 46]; HIV-specific CD21lo/-CD27- [18]; IgD-CD27- in SLE [21 , 24]; CD21loCD27+ in HCV with mixed cryoglobulinemia [26]; FCRL4+ ( CD21lo/-CD27- ) in the tonsil [27 , 28]; and bulk B cells from subjects with HIV or SLE [25 , 45] . Transcriptional or protein differences in nonclassical MBCs were measured in comparison to either classical MBCs ( CD21+CD27+ ) [11 , 31 , 46] , HIV-specific classical MBCs [18] , activated/classical MBCs ( CD27+CD21+/- ) [20 , 21] , CD27+IgD+/- B cells [24] , FCRL4- B cells in the tonsil [27 , 28] , CD21loCD27+ cells in HCV without mixed cryoglobulinemia [26] , or bulk B cells from healthy donors [22 , 25 , 45] . Our comparison includes genes and proteins that were determined to be significantly differentially regulated in at least one of the studies above as well as in our own analysis , and reports the direction of the change . Reverse transcription was performed on 600 ng aminoallyl-incorporated amplified RNA using SuperScript III Reverse Transcriptase ( Life Technologies ) and poly dT20V oligonucleotide primer in a 20 μl reaction . Samples were incubated with primer for 10 min at 70°C prior to addition of RT to allow primers to anneal . After addition of RT , tubes were incubated for 10 min at 25°C , then 50 min at 42°C , then 15 min at 70°C to inactivate RT , following previously published methods [66] . One μl RNase H was added and samples were incubated at 37°C for 20 min to degrade the input RNA . Samples were diluted 1:5 in nuclease-free water and 5 μl of diluted sample was used in a 25 μl quantitative PCR reaction using PerfecTa 2x qPCR Mix ( Quanta ) . Primers used in the qPCR reactions were biased toward the 3' end of mRNA transcripts , and annealed no further than 500 bp upstream from the polyA tail , to account for product shortening during amplification . Specific sequences used in this study were ACTB-F: 5’-AGTTCACAATGTGGCCGAGGA-3’; ACTB-R: 5’-TGTGTGGACTTGGGAGAGGA-3’; FCRL3-F: 5’-GAGGGCCCTCAGCTCCTA-3’; FCRL3-R: 5’-AAAGGGAAACAAAATATTTGGAGCA-3’; FCRL4-F: 5’-AAAACTTAAGTACCAACTCTCCAAA-3’; FCRL4-R: 5’-AATAAAACCTCTCTGCAAGGAGT-3’; FCRL5-F: 5’-AGAACAAACTCCACCCTAATGTG-3’; and FCRL5-R: 5’-CCAAGAAGAGCCATTTTTCAGTTTG-3’ . FCRL transcript levels were normalized to levels of actin mRNA . Samples were selected from children and adults over 8 years old , unless specifically noted otherwise . All had concurrent asymptomatic parasitemia as identified by microscopy ( blood smear positive , in the absence of fever ) . B cell subsets were defined using the antibodies described above , with the addition of CD20 ( clone B9E9 ) ( Beckman Coulter ) , as follows , unless otherwise specified: atypical MBCs ( CD19+CD20+CD21-CD27-IgG+IgD- ) , classical MBCs ( CD19+CD20+CD21+CD27+IgG+IgD- ) , transitional B cell ( CD19+CD20+CD10+ ) , and plasmablast/plasma cell ( CD19+CD20-CD38++CD27+/- ) . For some experiments , we also stained B cells to detect expression of CD120b ( clone hTNFR-M1 ) ( BD Biosciences ) ; CD85d ( clone 42D1 ) and CD360 ( clone 17A12 ) ( BioLegend ) ; FCRL3 ( clone 546828 ) ( R&D Systems ) ; FCRL4 and FCRL5 ( clone 2A6 ) ( generously provided by M . Cooper ) ; FCRL5 ( clone 7D11 ) and FCRL4 ( clone 1A3 ) ( generously provided by A . Polson and Genentech Inc . ) . Isotype controls included mouse IgG1 ( clone MOPC-21 ) ( Tonbo Biosciences ) , and IgG2a ( clone MOPC-173 ) and IgG2b ( clone MPC-11 ) ( BioLegend ) . Detection of mAb clones 2A6 and 1A3 was performed with rat anti-mouse IgG2a PE ( clone RMG2a-62 ) ( BioLegend ) , and 7D11 was detected with polyclonal goat anti-mouse IgG2b PE ( Life Technologies ) . Confirmation of the FCRL specificities of mAb 2A6 , 7D11 , and 1A3 was performed using cell lines expressing recombinant FCRL genes ( generously provided by A . Polson and Genentech Inc . ) [53] . Cell lines were cultured as previously described [53] , stained with mAb 2A6 , 7D11 , or 1A3 in the presence of Fc Block ( eBioscience ) , and stained with the secondary antibodies described above . Human B cell FCRL staining was similar , except that cells were first stained with mAb 2A6 , 7D11 , 1A3 , IgG2a isotype control , or mouse IgG2b isotype control in the presence of Fc block , followed by secondary antibody staining and subsequent staining for lineage markers . PBMCs were stained as above and flow cytometrically sorted into the following subsets: atMBCs ( CD19+CD20+CD21-CD27-IgG+IgD- ) , classical MBCs ( CD19+CD20+CD21+CD27+IgG+IgD- ) , transitional ( CD19+CD20+CD10+ ) , and plasmablast/plasma cell ( CD19+CD20-CD38++ ) . To measure spontaneous antibody secretion , sorted cells were placed in 200 μl of RPMI media supplemented with 5% FBS for 18 h in ELISpot plates ( Millipore ) that had been coated overnight at 4°C with 10 μg/ml of goat anti-human IgG ( Life Technologies ) . IgG-secreting cells were detected using alkaline phosphatase-conjugated goat anti-human IgG ( Life Technologies ) and a blue alkaline phosphatase substrate kit ( Vector Laboratories ) . Spots were enumerated using an AID ELISpot Reader and software ( AID ELISpots ) . To measure antibody production after stimulation , B cells were sorted based on FCRL5 expression in the atMBC and classical MBC subsets , and autologous CD3+ T cells were added at a 20:1 T cell to B cell ratio . Sorted cells were then stimulated with 2 . 5 μg/ml of CpG ODN 2006 ( InvivoGen ) and 2 . 5 μg/ml of F ( ab’ ) 2 goat anti-human IgG H+L chain ( Jackson ImmunoResearch ) for 4 d . After 4 d , cells were washed and incubated for 12 h on an ELISpot plate coated with goat anti-human IgG ( Life Technologies ) . ELISpot plates were developed and enumerated as described above . Statistics for microarray analysis are described above . All other comparisons between groups utilized nonparametric Wilcoxon rank-sum or signed-rank tests for unpaired and paired comparisons , respectively . Comparisons of the percentage of atMBCs expressing FCRL5 between Nagongera and Walukuba were also evaluated using multivariate linear regression to account for potential confounding by age . A p-value of < 0 . 05 was considered significant .
|
A subset of “atypical” memory B cells found in individuals with high exposure to P . falciparum has been hypothesized to be dysfunctional , based on phenotypic similarities to analogous cells found in HIV-infected individuals . However , the functional capabilities of these cells have been poorly characterized in the setting of malaria exposure , and previous reports have been controversial regarding whether these cells produce antibody . In our study , we analyze the molecular programming of atypical memory B cells , find that they are dysfunctional in a manner similar to that observed in B cells from HIV-infected individuals , and present data that may reconcile previously conflicting studies . By delineating the transcriptional landscape of atMBCs and identifying expression of FCRL5 as a key marker of dysfunction , we provide a foundation for improving our understanding of the role of these cells in immunity to malaria .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
FCRL5 Delineates Functionally Impaired Memory B Cells Associated with Plasmodium falciparum Exposure
|
Immunological memory is a hallmark of B and T lymphocytes that have undergone a previous encounter with a given antigen . It is assumed that memory cells mediate better protection of the host upon re-infection because of improved effector functions such as antibody production , cytotoxic activity and cytokine secretion . In contrast to cells of the adaptive immune system , innate immune cells are believed to exhibit a comparable functional effector response each time the same pathogen is encountered . Here , using mice infected by the intracellular bacterium Listeria monocytogenes , we show that during a recall bacterial infection , the chemokine CCL3 secreted by memory CD8+ T cells drives drastic modifications of the functional properties of several populations of phagocytes . We found that inflammatory ly6C+ monocytes and neutrophils largely mediated memory CD8+ T cell bacteriocidal activity by producing increased levels of reactive oxygen species ( ROS ) , augmenting the pH of their phagosomes and inducing antimicrobial autophagy . These events allowed an extremely rapid control of bacterial growth in vivo and accounted for protective immunity . Therefore , our results provide evidence that cytotoxic memory CD8+ T cells can license distinct antimicrobial effector mechanisms of innate cells to efficiently clear pathogens .
Immunological memory is a central feature of the adaptive immune system and relies on the maintenance of antigen-experienced B cells and T cells . In general , a memory immune response is faster and stronger than a primary response because of the increased frequency , improved functional characteristics , and preferential localization of the memory cells in peripheral tissues [1]–[2] . Memory CD8+ T cells represent one major effector arm of the adaptive immune system that maintains long-lived protective immunity against intracellular bacteria , protozoa , and viruses [1] . Although CD8+ T cells are mainly appreciated for their capacity to kill infected cells , they can express various effector mechanisms that potentially contribute to host defense against infections [3] . Former reports have suggested that the antimicrobial potential of CD8+ T lymphocytes is also reflected in their ability to rapidly produce inflammatory cytokines such as IFN-γ , TNF-α and CCL3 that promote the control of the growth of intracellular pathogens such as Francisella tularensis , Leishmania major or Listeria monocytogenes ( Lm ) by phagocytes [4]–[6] . In the case of a memory response , both the increased numbers of recruited phagocytes and their faster or chronic activation allow for an optimized clearance of intracellular pathogens [6]–[9] . This view is based on the assumption that phagocytes usually exhibit a qualitatively and quantitatively comparable response each time the same pathogen is encountered . However , it is not known whether and how the antimicrobial activities of innate phagocytes that are expressed during a primary immune response are modulated during recall infections . Phagocytes such as macrophages , neutrophils and monocytes play a critical role during primary infections to fight intracellular pathogens . Both neutrophils and macrophages are well-established as important cellular effectors of innate immune defense , and it is clear that circulating monocytes also contribute significantly to the defense against a range of microbial pathogens [10]–[11] . Phagocytic cells recognize and engulf pathogens by expressing specific receptors of microbial products ( pattern recognition receptors , PRRs ) , of the complement system or of antibodies complexed to pathogen-derived molecules [12] . Following microbial internalization , early phagosomes undergo numerous steps of maturation that are concomitant with important changes of their associated proteins and luminal pH acidification . Acidification results from the fusion of phagosomes with pre-existing lysosomes , leading to the activation of hydrolytic enzymes which act optimally at acidic pH ( 4 . 5–5 . 0 ) to enable the degradation of micro-organisms [13] . In addition to acidification , another critical mechanism involves recruitment to the phagosomal membranes of the nicotinamide adenine dinucleotide phosphate ( NADPH ) oxidase complex subunits which in turn generate the reactive superoxide ions that accumulate in the phagosomal lumen [14] . Superoxide is further converted to several highly toxic radical oxygen species ( ROS ) that degrade phagosomal content and facilitate pathogen killing [15]–[16] . Generation of ROS also induces phagosomal pH increase which subsequently activates neutral proteases that digest and kill microorganisms . Upon activation , phagocytes generate other important antimicrobial effector molecules such as reactive nitrogen species ( RNS ) produced by the inducible nitric oxid synthase ( iNOS ) that interact with ROS to exert very toxic effects on intraphagosomal engulfed microorganisms [17] . Collectively , these antimicrobicidal effectors activities provide a toxic phagosomal environment that efficiently limits pathogen proliferation . However , some pathogens including bacteria such as Listeria , Shigella , Rickettsia and group A Streptococcus ( GAS ) have evolved strategies to quickly escape from this hostile phagosomal environment to the more hospitable milieu of the host cell cytosol [18] . After escape into the cytosol , some of these bacteria can be trapped in large cytosolic vacuoles and ultimately be degraded within autolysosomal compartments [19]–[21] . This process , known as autophagy , has been more recently described as an intrinsic host defense mechanism for recognition and elimination of intracellular cytosolic or vacuolar pathogens [19] . In a previous study using mice intravenously inoculated with Lm as an infection model , we have shown that innate phagocytes can be activated by the Macrophage Inflammatory Protein 1α ( MIP-1α/CCL3 ) that is secreted by memory CD8+ T cells upon antigen-driven reactivation [6] . CCL3 induced a rapid TNF-α secretion by innate inflammatory monocytes , which further promoted the production of ROS by both monocytes and neutrophils . We also found that ROS generation depended on CCL3 and was required for killing of Lm . In the present work , taking advantage of the same experimental settings in which the vast majority of bacteria were found inside phagocytic cells of the spleen [22]–[23] , we investigated the modulation of the antimicrobial effector activities expressed by innate phagocytes in the course of the secondary infection and the impact this have on the clearance of the bacteria from infected tissues .
CD8+ T cells usually protect against intracellular microbes by the recognition and the lysis of infected cells , via a perforin and/or a Fas/Fas-ligand-mediated mechanism of cytolysis . We recently demonstrated in the mouse Lm infection model that memory CD8+ T cells are also able to protect by orchestrating the activation of effector cells of the innate immune system via the release of the chemokine CCL3/MIP1-α [6] . Indeed , the transfer of CCL3-deficient effector [24] or memory [6] CD8+ T cells to infected mice does not confer protective immunity to these mice . Moreover , the protection achieved by wt memory CD8+ T cells infused into CCL3−/− recipient mice is totally abrogated upon CCL3 neutralization [6] . At least 2 recent reports have documented that memory CD8+ T cells induced in a bacterial and a viral infection indeed can produce high levels of CCL3 and that expression of this gene is enhanced in highly differentiated , long-lived memory cells that can confer immunological protection [25]–[26] . We further compared this mechanism to the classical CD8+ T cell-dependent effector mechanisms . For this , wt or perforin-deficient mice were immunized by infection with Lm ( then referred to as “memory mice” ) , and 3 weeks later some animals were treated with anti-CCL3 serum or anti-IFN-γ neutralizing mAb before the challenge infection ( Figure 1A ) . While animals lacking perforin exhibited only a modest increase in susceptibility ( ∼10 fold ) compared to wt memory mice [27] , none of the anti-CCL3-treated groups controlled the challenge infection . These mice had more than 106 bacteria in the spleen , a number close to that found in primary and in memory p47phox−/− infected mice that lack functional NADPH oxidase complexes and cannot produce ROS ( Figure 1B ) . Interestingly , when memory mice were treated with a neutralizing Ab against IFN-γ , an important immunomodulatory cytokine of effector and memory CD8+ T cells , only about a 10 fold loss of protection was observed in wt mice ( Figure 1A ) [28] . Thus , whereas perforin- and IFN-γ accounted for a significant degree of protection in wt memory animals , the current results revealed a substantially greater role for CCL3 and ROS in protective memory responses to Lm . To study the modulation of the effector functions of phagocytes during a memory response , we first defined the cytofluorometry gating strategies that allowed us to identify the major innate phagocyte populations of the spleen . While neutrophils ( CD11bhighF4/80lowLy-6Ghigh ) and macrophages ( CD11bmedF4/80highLy-6Gmed ) are well-defined in the literature ( Figure 2A ) , blood-monocytes that enter inflamed tissues and express antimicrobial functions have been variously described as inflammatory monocytes , [29] , TNF-α/iNOS producing-dendritic cells ( Tip-DCs ) [30] or mononuclear phagocytic cells ( MPCs ) [6] depending on the experimental context . Since all of these cells exhibit a comparable cell-surface phenotype ( CD11bmed/highF4/80medLy-6GlowCX3CR1lowCD11clowLy-6Chigh ) , produce TNF-α , RNS and ROS ( Figure 2A , S1 ) , it is likely that inflammatory monocytes , Tip-DCs and MPCs are the same cell population , which we refer to as inflammatory and/or ly6C+ monocytes . In addition to the generation of ROS , phagocytic cells also express other important antimicrobial activities within their phagosomes during activation such as the generation of reactive nitric species ( RNS ) and the rapid modulation of the pH of the phagosomal lumen [31] . Since the enzymes that produce NO ( the inducible nitric oxide synthase , iNOS ) and ROS ( the NADPH oxidase ) can be activated by the release of IFN-γ and TNF-α , and because levels of these cytokines are much higher during the memory response , we monitored the activation of both enzymatic complexes during a primary as compared to a memory immune response . Mice injected with PBS or with Lm were challenged a month later with Lm ( primary and memory responses , respectively ) , and the frequencies and mean fluorescence intensities ( MFI ) of ROS-producing and iNOS-expressing cells were measured by flow cytometry ( Figure 2B–C , S2 and S3 ) . The numbers and frequencies of ROS-producing inflammatory monocytes and neutrophils measured during a memory response ( open symbols ) were 1 . 3 to 2 . 4 higher than those of a primary response ( closed symbols ) ( Figure 2B left; 2C , upper and middle panels and S2 ) . In addition , the MFI of ROS+ ly6C+ monocytes was 1 . 2 to 1 . 5 fold increased compared to that of the ROS+ neutrophils and macrophages –at least the population of macrophages that we could recover from the spleen , suggesting that ly6C+ monocytes were intrinsically able to produce higher level of ROS than other effector phagocytes ( Figure 2B , right panels ) . Such augmented effector activities depended on memory CD8+ T cells and CCL3 since immune mice treated either with an anti-CD8 depleting mAb or an anti-CCL3 neutralizing serum exhibited a lower frequency ( ∼30–40% instead of 60–70% ) of ROS-producing inflammatory monocytes or neutrophils after challenge ( white bars ) , equivalent to that found in primary infected animals ( white and black bars ) ( Figure 2C , upper and middle panels ) . Of note , while the subset of macrophages recovered from the spleen accounted for the large majority of ROS-producing phagocytes at steady state ( data not shown ) , frequencies , numbers and MFI of these ROS+ macrophages did not vary neither during the primary nor the memory response ( Figure 2B–C bottom panels and S2 ) . Interestingly , whereas memory CD8+ T cells can promote ROS production by phagocytes , the frequencies of iNOS-expressing cells were similar in primary ( closed symbols ) and memory ( open symbols ) infected mice ( Figure S3A ) . Moreover , in contrast to oxidative burst-deficient mice ( Figure 1B & [6] ) , memory mice lacking the iNOS enzyme are equivalently protected during the secondary challenge as wt animals ( white bars ) and exhibited ∼5 , 000–6 , 500 fold less bacteria in the spleen than primary infected wt or iNOS−/− mice ( black bars ) ( Figure S3B ) . Therefore , memory CD8+ T cells allow ( i ) increased numbers of inflammatory monocytes and neutrophils to be activated and ( ii ) the differentiation of effector monocytes and neutrophils that produce higher quantities of ROS on a per cell basis . While an acidic luminal pH is essential for optimal activity of numerous microbicidal agents in macrophages and neutrophils , the NADPH-oxidase complex may also mediates the increase of the pH in phagocytic vacuoles that could lead under some experimental conditions to the activation of neutral proteases that kill microorganisms [14] , [32]–[33] . Since phagocytes exhibited an increased production of ROS during a memory response , we monitored the phagosomal pH inside ly6C+ monocytes , neutrophils and macrophages from primary and memory mice 6 hrs after the infection ( Figure 3 , S4 and not shown ) . For this , latex beads were coated with pH-sensitive and pH-insensitive fluorescent dyes and incubated with splenocytes for 20 minutes to allow uptake of beads by phagocytes . Cells were further stained for the expression of different cell surface markers expressed by phagocytes and the fluorescence intensity of the phagocytosed beads was quantified by FACS . The ratio in fluorescence intensity between the two dyes directly reflected the pH inside the phagosomes as calibrated on a standard curve [34] ( Figure S4A ) . We found that inflammatory monocytes which produced the highest levels of ROS per cell ( Figure 2B , left panel ) , exhibited increased pH ( ∼5 . 2 ) inside their phagosomes in memory as compared to those of primary infected mice ( ∼4 . 45 ) ( Figure 3 , left panel ) . As expected , the intra-phagosomal pH for neutrophils [32]–[33] , was also augmented in memory ( ∼5 . 2 ) versus primary ( ∼4 . 6 ) infected animals ( Figure 3 , right panel ) [32]–[33] . These results were in sharp contrast with the intra-phagosomal pH of the splenic macrophage subset , which remained acidic ( Figure S4B ) . Similar acidic pH values that reached pH ∼4 . 5 three hrs after phagocytosis have also been observed for the macrophage cell lines RAW264 . 7 and J774 in vitro under similar experimental conditions [35] . The higher pH of the phagosomes of inflammatory monocytes and neutrophils from infected memory mice depended on CCL3+ memory CD8+ T cells , since treatment of memory mice with an anti-CD8 depleting mAb or an anti-CCL3 neutralizing serum induced a pH drop equivalent to that of primary infected mice . Collectively , our results show that during a memory response , innate cellular effectors -inflammatory monocytes and neutrophils- exhibit an inhospitable environment inside their phagosomes that helps better clearance of bacteria and likely is more toxic for pathogens than in the course of a primary response . During Lm infection , the initial step of escape from the primary vacuole of phagocytosis to the cytosol of infected host cells is critical for bacterial survival [36]–[37] . Since Lm is rapidly trapped in such vacuoles [38] , we reasoned that the most efficient way of controlling Lm growth is to prevent its escape to the cytosol of infected phagocytes . If this hypothesis is true , Lm multiplication between primary and memory infected animals would be expected to differ as early as during the first few hrs of infection . To investigate this possibility , we monitored Lm multiplication inside the spleen of primary and memory challenged animals that were concomitantly treated with an anti-CD8 depleting Ab , anti-CCL3 neutralizing serum or control goat serum ( Figure 4 and not shown ) . As expected , primary or memory infected mice treated with anti-CD8 or anti-CCL3 did not control the infection . These mice exhibited 6 . 6 and 20 times more bacteria in the spleen respectively 9 and 12 hrs after the challenge infection than memory mice either untreated ( white circles , plain line ) or injected with control isotype matched Ab or serum ( not shown ) . At 24 hrs , between 23 and 216 times more bacteria were measured in the groups of mice that did not control the infection ( Figure 4A ) . Therefore , memory mice ( open symbols ) have already cleared 40% , 72% and 85% of the bacteria colony forming units ( CFUs ) 1 . 5 , 6 and 9 hrs after the challenge infection as compared to primary infected animals ( closed symbols ) ( Figure 4B ) . This result showed that Lm killing began very early after challenge of memory mice . Because several effector activities of phagocytes were improved at these time points , we hypothesized that killing of Lm could happen inside the primary vacuoles of phagocytes and with a higher efficacy in mice with memory responses . To determine whether the early killing of Lm indeed takes place inside such vacuoles , we took advantage of a recombinant Lm ( wt-L029 ) which expresses an antibiotic resistance to chloramphenicol upon reaching the cytoplasm of infected cells [39] . Splenocytes were incubated with chloramphenicol for a short period of time , which kills extracellular wt-L029 Lm and those that are inside vacuoles , only allowing survival of Lm that escaped to the cytosol ( Figure S5 ) . Primary and memory mice were challenged with wt-L029 Lm and the frequencies of vacuolar ( chloramphenicol-sensitive ) and cytosolic ( chloramphenicol-resistant ) viable bacteria inside phagocytes were determined 1 . 5 , 3 and 6 hrs post-infection ( Figure 4C–D ) . In primary infected mice , 60% of viable Lm localized to the cytosol and 40% to the vacuoles of cells 1 . 5 and 3 hrs later ( Figure 4D ) . In contrast , while only 5–10% of the viable bacteria were found in the vacuoles of the phagocytes from memory mice ( yellow bars ) , the vast majority ( ∼90–95% ) were recovered from the cytosol ( blue bars ) . At 6 hrs , bacterial localization was similar between primary and memory mice , with ∼90% of viable Lm localized to the cytosol and ∼10% to the vacuoles of phagocytes . Because the large majority of viable bacteria were recovered from the cytoplasm of infected cells in memory mice , these findings suggested that Lm was killed with a higher efficacy in the vacuoles of phagocytes from memory compared to primary infected mice . To provide further support to this hypothesis , we determined the absolute numbers of viable bacteria inside the vacuoles of phagocytes from primary and memory mice 1 . 5 hrs after the infection with wt-L029 Lm ( Figure 5A ) . At this early time point following challenge , memory mice already exhibited ∼30–50% fewer bacteria in the spleen than primary infected mice . Moreover , memory mice exhibited two fold less viable bacteria in the vacuoles than animals with primary infections . To formally demonstrate the improved killing of Lm inside the vacuoles of phagocytes during the memory response , primary and memory mice were challenged with listeriolysin ( LLO ) -deficient ( ΔLLO ) Lm that remained trapped inside the vacuoles , and the number of viable Lm was determined 1 . 5 hrs later . As for wt Lm , memory mice infected with ΔLLO-Lm exhibited 2 . 5 times less viable bacteria than primary infected animals ( Figure 5B ) . To assess whether the improved killing of Lm occurring inside the vacuoles of the phagocytes resulted from the enhanced activity of the NADPH-oxidase complex during the memory response , primary or memory mice lacking the inducible p47phox subunit of the NADPH-oxidase complex , or wt mice treated with either anti-CD8 , anti-CCL3 or control Abs were infected with wt-L029 Lm ( Figure 5C–D ) . p47phox−/− and wt memory mice treated with the anti-CD8 or the anti-CCL3 exhibited similar frequencies ( ∼40–60% ) of viable vacuolar Lm as primary infected wt and p47phox−/− mice . Thus , taken together our results suggested that during a memory response , phagocytes are conditioned by the memory CD8+ T cells to generate a more effective bacteria-killing environment inside their early vacuoles of phagocytosis , and that this process is directly dependent upon increased activity of the NADPH-oxidase complex . While at very early time points after the infection ( 0–3 hrs ) , Lm was most efficiently killed within vacuoles of phagocytes from memory mice , the percentage of viable vacuolar bacteria ( ∼85% ) and therefore the rate of phagosomal killing was comparable 6 hrs after the infection in primary and memory infected mice ( Figure 4D ) . However , despite equivalent phagosomal elimination of Lm , the number of viable bacteria was significantly diminished in memory as compared to primary infected animals ( Figure 4A–B ) . We therefore hypothesized that this either resulted from the more efficient early phagosomal killing of Lm or from other killing mechanisms distinct from those expressed inside the primary phagocytic vacuoles . If the first hypothesis were to be true , we expected to find a comparable distribution of the viable bacteria between the 2 groups of mice . We thus investigated the subcellular localization of Lm in infected splenocytes from primary and memory mice challenged with wt-L029 Lm ( Figure 4C–D ) . Between 9 and 24 hrs after the infection , the relative proportion of viable cytosolic bacteria inside the phagocytes from memory mice drastically decreased from ∼60% to 15% compared to that of primary infected animals which only decreased about 20% . Viable Lm inside the cytosol of infected cells ( i . e . , chloramphenicol resistant ) were rapidly lost and further recovered from vacuolar structures ( i . e . , chloramphenicol sensitive ) , suggesting that bacteria which had escaped to the cytosol of infected cells were rapidly re-engulfed into vacuoles . We reasoned that this could either result from Lm spreading from one cell to another or from the reinternalization of the free cytosolic Lm into phagosomal structures inside the same infected cell . To discriminate between the two possibilities , we used Lm lacking ActA that neither can spread from cell to cell nor infect neighboring cells . For that purpose , we generated an ActA-deficient L029 Lm strain that expressed the chloramphenicol resistance gene when reaching the cytosol of infected cells ( Figure 6A ) . Upon inoculation of wt or ActA-deficient L029 Lm , phagocytes from primary and memory mice both exhibited similar frequencies of vacuolar viable bacteria , suggesting that the appearance of such vacuole-containing bacteria did not result from Lm spreading to neighboring cells but rather from Lm reinternalization inside the same infected cell . To demonstrate that Lm relocalization from the cytosol to vacuolar structures indeed depended on CCL3+ memory CD8+ T cells , primary and memory mice were treated with an anti-CD8 depleting mAb , an anti-CCL3 neutralizing or control serum ( not shown ) , and further challenged for 24 hrs with wt-L029 Lm ( Figure 6B ) . CD8+ T cell depletion and CCL3 neutralization both prevented Lm engulfment into such vacuoles of phagocytes from memory mice that exhibited relative frequencies of cytosolic and vacuolar viable Lm similar to that of primary infected animals . Therefore , the subcellular localization of Lm was strongly modified during a memory response and this depended on memory CD8+ T cells and CCL3 . This process also correlated with the more efficient bacterial killing occurring in memory mice compared to primary infected animals , which showed 1 , 920 and 415 , 200 bacterial CFUs respectively in the spleen 24 hrs after the infection ( Figure 4 ) . Overall , these data suggest that Lm destruction in the later phase of infection involves killing mechanisms distinct from those expressed in the primary vacuole of phagocytosis . We next investigated how such de novo phagosomes that are likely key contributors to Lm destruction during recall infections were formed . Since ROS were required to clear Lm during a memory response ( Figure 1B & [6] ) , we hypothesized that the production of ROS was also involved in Lm re-engulfment to these vacuolar structures of phagocytes from memory mice . For this , wt and p47phox-/- primary or memory mice were challenged with wt-L029 Lm and 24 hrs later the distribution of viable vacuolar and cytosolic bacteria was analyzed ( Figure 6C ) . p47 phox-/- memory mice exhibited similar frequencies of viable Lm in the vacuoles and the cytosol of phagocytes compared to primary infected p47 phox-/- and wt control animals . Therefore , our results show that phagocytes from memory mice receive signals that license them to become more efficient to clear Lm by distinct mechanisms , amongst which is the engulfment of Lm into de novo formed phagosomal structures . This mechanism requires the generation of an oxidative burst promoted via the release of CCL3 by memory CD8+ T cells . Of note , we did not find any substantial differences in the frequencies of phagocytes undergoing cell death ( Annexin V+ , propidium iodide+ ) between primary and memory infected mice ( Figure S6 ) , likely ruling out increased phagocyte death in memory mice as a mechanism for controlling bacterial growth . Lm has been shown to be a target for autophagy , a process that can be promoted by the generation of ROS [20] , [40] . Since we observed that Lm is engulfed in vacuolar structures of infected phagocytes in a CCL3/ROS-dependent manner , we hypothesized that the phagocytes from memory mice were also primed to kill Lm by autophagy . We therefore monitored the conversion of the well-characterized autophagy marker LC3 inside the phagocytes . During this cellular process , cytosolic-free LC3-I is converted to lipidated LC3-I ( LC3-II ) that is covalently linked to phospholipids and associated with autophagosomes . This modification alters LC3-I gel migration properties which can be visualized by electrophoresis and western blotting . We assessed the ratio of LC3-II and LC3-I inside inflammatory monocytes , neutrophils and population of macrophages that extracted from the spleens of naïve , primary and memory mice infected for 20 hrs ( Figure 7A ) . In contrast to these macrophages which demonstrated intrinsically high levels of autophagy in naïve uninfected mice ( bottom blot ) , LC3-II/LC3-I ratios in inflammatory monocytes and neutrophils purified from the spleen of memory mice ( white bars ) were 2 and 1 . 3 times increased as compared to those from primary infected mice ( black bars ) . This increase in LC3-II/LC3-I ratios most likely reflected the direct consequence of LC3-I lipidation and site specific proteolysis , e . g . , conversion of LC3-I into LC3-II that localizes at the membranes of autophagosomes . Of note , we did not observe any significant increase in levels of LC3-I ( Figure S7 ) , suggesting that its rate of production was not affected by infection . Using electron microscopy , we looked for the presence of autophagosomes in phagocytes purified by cell sorting from memory mice infected with Lm expressing GFP for 20 hrs [22] . Figure 7B shows representative pictures of infected phagocytes containing Lm engulfed inside vacuolar structures that exhibit the characteristic double membrane feature of autophagosomes . To formally link the induction of autophagy in monocytes and neutrophils from memory mice with the generation of an oxidative burst and the presence of CCL3+ memory CD8+ T cells , wt control or p47phox−/− primary and memory mice were treated with an anti-CD8 mAb or an anti-CCL3 serum and challenged with wt Lm ( Figure S8 and not shown ) . CD8+ T cell depletion , neutralization of CCL3 or p47phox deficiency all prevented the induction of autophagy in inflammatory monocytes and neutrophils as compared to relevant controls . Collectively , our results show that innate inflammatory monocytes and neutrophils were able , with the help of the CCL3+ memory CD8+ T cells , to specifically induce efficient autophagosome formation during a memory response , and this allows the rapid control of the growth and the clearance of intracellular Lm .
Our study highlights that several effector functions of phagocytes are strongly modulated by memory CD8+ T cells in the course of a recall infection in vivo and that this leads to most efficient pathogen destruction . We show that bacterial killing involves activation of higher numbers of monocytes and neutrophils , which produce higher levels of ROS on a per cell basis during the secondary compared to the primary infection . ROS are likely inducing the rapid increase of the pH of the primary vacuoles of phagocytosis , as well as augmented levels of cellular autophagy , both allowing for Lm clearance . Collectively our results show that the cells from the innate immune system behave differently in the course of a recall infection by integrating signals from memory CD8+ T cells which stimulate them to express optimized antimicrobial effector functions . Recent studies also documented enhanced effector/memory CD4+ T cell-mediated clearance of pathogens by innate immune cells [8]–[9]; it is therefore conceivable that comparable antimicrobial protective mechanisms inside effector phagocytes are involved . Along similar lines , an elegant study documented that during lymphochoriomeningitis-induced meningitis , effector CD8+ T cells can promote the recruitment of pathogenic monocytes and neutrophils , leading to accelerated fatal outcome [41] . The current dogma is that only classical T and B lymphocytes that belong to the adaptive immune system are able to differentiate into long-lived memory cells exhibiting qualitatively improved functional properties . In fact , immunological memory is reflected by a stronger proliferation and expression of several effector functions , such as the secretion of specific cytokines and chemokines and the cytolysis of infected cells . In contrast , effector cells of the innate immune system like monocytes , macrophages and neutrophils are believed to mediate a rapid and antigen non-specific immune response that is qualitatively and quantitatively comparable each time the same pathogen is encountered . Macrophages have nevertheless been shown to undergo an adaptive-like response by acquiring distinct patterns of TLR-induced chromatin modifications that include modifications associated with the priming of antimicrobial effectors [42] . In the present work , we found that production of a chemokine by memory CD8+ T cells induced a strikingly distinct response of inflammatory monocytes and neutrophils . Indeed , we observed higher frequencies of ROS-producing inflammatory monocytes and neutrophils , an improved capacity to produce ROS and the upregulation of autophagy , a process also involved in intracellular pathogen clearance . Therefore , our data and that from others [7] , [9] , [41] promote the concept that the innate immune response during a secondary antigenic encounter can be regulated , either in a cell autonomous manner [42] or in response to lymphocyte-derived cues . Interestingly , the licensing phenomenon we describe here seems exclusively observed for inflammatory monocytes and neutrophils and likely not for the population of macrophages extracted from the spleen . One explanation could be that the primary function of monocytes and neutrophils is to circulate via the blood and the lymphatic system and patrol the body to permanently sense the tissue environment in search for danger signals [29] , [43] . Together , these cell types represent the vast majority of the blood-derived leucocytes and are rapidly attracted to damaged or infected tissues . Another possibility is that splenic monocytes , that may represent an important reservoir of resident monocytes possibly distinct from the blood monocytes , have intrinsically different functional properties [44] . Monocytes indeed exhibit a very dynamic plasticity and are able to differentiate into several functionally distinct subsets of effector cells within inflamed tissues [45] . Therefore , it is conceivable that these cells are more sensitive to cytokine/chemokine-mediated modulation of their fate in vivo compared to tissue-resident macrophages , which are terminally differentiated cells usually involved in steady-state tissue homeostasis . What mechanisms are responsible for the improved immune response of innate phagocytes during secondary Lm infection ? We recently found that during the recall infection , memory CD8+ T cells form transient “effector clusters” with inflammatory monocytes and neutrophils in the spleen [46] . These observations provide a spatio-temporal explanation to the subsequent cascade of effector functions that ultimately result in an efficient control of Lm burden during recall infection . In such clusters , memory CD8+ T cell-derived cytokines are concentrated , which likely favors the rapid and efficient integration of inflammatory signals by the phagocytes . At the molecular level , it will be interesting to investigate whether the augmented production of ROS observed in inflammatory monocytes and neutrophils results from increased phosphorylation of the cytosolic regulatory protein p47phox leading to NADPH oxidase activation [47] . The improved response of phagocytes could then be attributed to post-translational regulation mechanisms . ROS , and more specifically O2− , can act as signaling molecules to induce cellular autophagy [40] , a mechanism that is increasingly described as an important antimicrobial effector mechanism to kill pathogens . The increased level of ROS produced in the course of a memory response can also induce higher pH inside the vacuoles of inflammatory monocytes and neutrophils , and better bacterial killing . Higher pH can induce ion fluxes across the vacuolar membrane , which displaces strongly bound enzymes from the negatively charged proteoglycan granule matrix and allows them to digest and kill the microorganisms [14] . Collectively our data illustrate that the innate immune response and several antimicrobial effector functions of phagocyte populations can be regulated in the course of a memory response . ROS generation induced upon CCL3 secretion by reactivated memory CD8+ T cells likely provides the major signaling molecules promoting the expression of antimicrobial activities of both inflammatory monocytes and neutrophils . As we previously found , such processes also mediate the bystander killing of an unrelated intracellular pathogen [6] . Since we and others have shown that the modulation of the functions expressed by these phagocytes is potentially important for immunotherapeutic and vaccination strategies , it is critical to achieve a better knowledge on how such innate effector cells function in the context of distinct infections .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Committee of Animal Care and Use of the Regional Cote d'Azur . The protocol was approved by the Committee on the Ethics of Animal Experiments of the Institut de Pharmacologie Moléculaire et Cellulaire where the study was performed ( Permit Number: B-06-152-5 , delivered by the Veterinary Services of the Alpes-Maritimes Prefecture ) and by the animal use committees at the Albert Einstein College of Medicine . All efforts were made to minimize suffering and provide humane treatment to the animals included in the study . Wt BALB/cByJ ( Charles River Labs ) and C57BL/6J ( Jackson labs ) 6–8 wk-old female mice , were used in all experiments unless otherwise indicated . p47phox−/− mice were obtained from European Mouse Mutant Archive ( EMMA ) and CX3CR1+/− mice [48] on the C57BL/6 background were obtained from F . Geissmann's laboratory ( King's College London , London ) . iNOS−/− mice were purchased from CDTA ( CNRS , Orléans ) . Transgenic mice were housed and bred in our SPF animal facility . The Listeria monocytogenes ( Lm ) 10403 s wt strain was used in all experiments . The wt-GFP , ΔLLO , wt-L029 and ΔActA Lm are on the 10403 s background and have been previously described [22] , [39] . The ΔActA-L029 was generated by transfection of the pLCR plasmid that expresses the chloramphenicol resistance gene under the ActA promoter . The wt-L029 Lm and the pLCR plasmid are a kind gift from Dan Portnoy ( UCSB , CA , USA ) . The wt strains exhibit a LD50 of 3×104 in BALB/c mice . HKLM was prepared as previously described [49] . All Lm were prepared from clones grown from organs of infected mice and kept at −80°C . For Lm infections , bacteria were grown to a logarithmic phase ( OD600 = 0 . 05–0 . 15 ) in Broth Heart Infusion medium ( Sigma ) , diluted in PBS and injected i . v . into lateral tail vein . In all experiments , mice were primary immunized with a 0 . 1xLD50 of bacteria ( 3×103 ) . Secondary infections were carried out one month later with indicated bacteria . To measure Lm titers in spleen and liver , organs were dissociated in 0 . 1% X-100 Triton ( Sigma ) and serial dilutions plated onto BHI media plates . To determine the number and frequency of viable Lm localized inside the vacuole or the cytosol , spleens were dissociated in RPMI1640 ( Gibco ) 5% FCS and each cell suspension was split in two and either incubated or not with 10 µg/ml of chloramphenicol at 37°C for 45 min . After centrifugation , cell pellets were resuspended in 0 . 1% triton X-100 and serial dilutions plated onto BHI media plates . The next day , we determined the total number ( vacuolar and cytosolic ) of viable bacteria by counting the colony forming units ( CFU ) from untreated samples . The number of cytosolic bacteria was determined by counting the CFU from chloramphenicol-treated cells . By subtracting these numbers , we obtained the number of vacuolar bacteria . Lm CFUs in these cells were determined as described above . For neutralization of CCL3 , mice were treated with 75 µg of anti-CCL3 or goat IgG control i . v at the time of the secondary infection and 24 h later . For neutralization of IFN-γ , mice received one injection of 500 µg of the mAb XMG1 . 2 concomitantly to the recall infection . For depletion of CD8+ T cells , mice were injected daily 3 times with 100 µg of the anti-CD8β H-35 mAb or its control isotype i . p and further infected one day after . Anti-F4/80 ( A3-1 ) -fluorescein isothiocyanate ( FITC ) was purchased from Caltag Laboratories . The following mAbs were purchased from BD Pharmingen: , anti-CD11b ( M1/70 ) -phycoerythrin ( PE ) , -peridinin chlorophyll protein ( PcP ) or -allophycocyanin ( APC ) , anti-TCR-PE ( H57-597 ) , anti-NK1 . 1-PE ( PK136 ) , anti-CD19-PE ( MB19-1 ) , anti-Ly-6G-PE ( 1A8 ) , anti-Ly-6C-FITC and biotine ( AL-21 ) , anti-CD11c-PE and APC ( HL3 ) , anti-TNF-α-APC ( MP6-XT22 ) and control rat IgG1 mAb , AnnexinV and propidium iodide . Anti-NOS-2 ( M-19 ) polyclonal rabbit Abs were purchased from Santa Cruz Biotechnology . Goat anti-rabbit-Alexa647 was from Molecular Probes . Organs were cut in small pieces and incubated at 37°C for 20 min in HBSS medium ( Gibco ) containing 4000 U/ml of collagenase I ( Gibco ) and 0 . 1 mg/ml of DNase I ( Roche ) . Red blood cells were lysed for 2–3 min in 170 mM NH4Cl , 17 mM Tris HCl pH 7 . 4 . Cells were stained with the specified Abs in PBS 0 . 5% of BSA ( FACS buffer ) . For intracellular TNF-α staining , splenocytes were incubated at 37°C 5%CO2 for 3–4 hrs in RPMI1640 ( Gibco ) 5% FCS , 2 µg/ml Golgi Plug ( BD Pharmingen ) with or without 5×108 HKLM/ml . Cells were incubated for 20 min on ice with the indicated cell surface marker mAbs , fixed in 1% PFA FACS buffer for 20 min on ice , permeabilized for 30 min in 1XPerm/Wash ( BD Pharmingen ) . For intracellular staining of TNF-α cells were incubated for 20 min on ice in FACS buffer containing anti-TNF-α or control rat IgG1 . For intracellular staining of iNOS , cells were incubated for 20 min on ice in FACS buffer containing anti-iNOS rabbit polyclonal , or control normal goat IgG and staining was revealed using goat anti-rabbit Alexa647 mAb . In all cases , cells were washed , fixed for 30 minutes in 1% PFA FACS buffer and analyzed on a FACScalibur cytofluorometer ( Becton Dickinson , BD ) . When indicated , cells were sorted on a FACSvantage SE cell sorter ( Becton Dickinson ) . 5−10×106 splenocytes were incubated for 3 h at 37°C and 5%CO2 in RPMI1640 containing 5% FCS with 5×108 HKLM/ml and 160 µM hydroethidine ( Polysciences ) . Hydroethidine is oxidized by ROS in red fluorescent ethidium bromide ( EB ) therefore allowing for the detection of ROS-producing cells . Cells were washed in FACS buffer and stained for expression of cell surface markers . This protocol was adapted from Savina et al . [34]–[35] . 50 to 70 µl of 3 micrometers of polybeads amino ( Molecular probes ) were covalently coupled with 50 µl of 50 mg/ml FITC ( pH sensitive ) ( Fluorescein isothio isomer 1 , Sigma F7250 ) and 50 µl of 1 mg/ml FluoProbes 647 ( pH insensitive ) ( Fluo Probes , Molecular Probes FP-AK7740 ) in 400 µl of sodium hydrogen carbonate buffer at pH 8 . 0 for 2 h at room temperature . After extensively washing with glycine 100 mM , the beads were suspended in PBS . Cells were then pulsed with the coupled beads for 20 min , extensively washed in cold PBS and stained on ice for CD11b , Ly-6C or F4/40 surface markers and immediately analyzed by FACS , by selectively gating on the cells that have phagocytosed one latex bead . The ratio of the mean fluorescence intensity ( MFI ) emission between the two dyes was determined . Values were compared with a standard curve obtained by resuspending the cells that had phagocytosed beads for 20 min at a fixed pH ( ranging from pH 5 . 5 to 8 . 0 ) and containing 1XPerm/Wash ( BD Pharmingen ) . The cells were immediately analyzed by FACS to determine the emission ratio of the two fluorescent probes for each pH value . PBS-injected or wt immunized BALB/c , C57BL/6 wt or p47phox−/− mice were sacrificed and spleen cells were positively enriched using anti-CD11b-specific MACS beads ( Myltenil ) according to the standard manufacturer protocol and further flow-cell sorted on expression of CD11b , F4/80 and Ly-6G cell surface markers . Inflammatory monocytes were defined as CD11bmed/highLy-6G-F4/80med cells , neutrophils as CD11bhighLy-6GhighF4/80- cells and macrophages as CD11bmedLy-6GmedF4/80high cells ( purity>87% ) . The same amount of flow-sorted cells ( 2−4×106 ) were washed in PBS and pellets were resuspended in lysis buffer pH 8 . 0 ( Tris-HCl 50 mM , NaCl ( 200 mM ) , EDTA 5 mM , 0 . 5% Triton X-100 , 0 . 5% deoxycholic acid ) for 45 min at 4°C . Homogenates were centrifuged for 15 min at 20 , 000xg and lysate supernatants were analyzed by SDS-PAGE , transferred to PVDF membranes , blocked and probed with a rabbit anti-LC3 ( 2 µg/ml ) ( Novus Biologicals ) , a rabbit anti-actin ( 0 . 5 µg/ml ) ( Sigma ) , a anti-rabbit peroxidase ( 1∶10 , 000 ) ( Beckman Coulter ) . Western blot quantification was performed with the ImageJ software . Flow-sorted CD11b+GFP+ infected cells were fixed with 1 . 6% glutaraldehyde in 100 mM phosphate buffer pH 7 . 5 . Cells were incubated at 4°C in 100 mM cacodylate buffer pH 7 . 0 containing 1% osmium tetroxyde . Cellular pellets were washed with distilled water and incubated for 2 hrs with 0 . 5% uranyl acetate buffer at room temperature in the dark . Cells were washed in water; pellets were dehydrated in increasing acetone series and embedded in epoxy resins . Blocks were thin-sectioned using standard procedures and contrasted for the observation on a Philips CM12 electron microscope . Statistical significance was calculated using an unpaired Mann-Whitney test and Instat software , except if specified differently in the legend of the figure . All P values of 0 . 05 or less were considered significant and are referred to as such in the text .
|
The immune system comprises white blood cells that belong to the innate or the adaptive immune arms . Adaptive immune cells such as T and B lymphocytes can give rise to memory cells which mediate long-lived immunity against pathogens . During a recall infection , innate immune phagocytic cells such as monocytes and neutrophils can be critical to kill microbial pathogens inside infected tissues . Whether and how such antimicrobial features of phagocytic cells of the innate immune system are modulated during a memory response in a vaccinated host is not known . The present report shows that cytolytic memory T lymphocytes , an important subpopulation of effector T cells , can drastically enhance the functional killing capacities of monocytes and neutrophils for optimized pathogen clearance from infected hosts . These phagocytes exhibit enhanced generation of oxidative burst , increased phagosomal pH and autophagy , three mechanisms that lead to intracellular pathogen death . This result is important since it suggests that modulating innate immune cells effector activities could be an interesting strategy to enhance vaccine efficacy .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immune",
"cells",
"cytokines",
"immune",
"activation",
"immunity",
"to",
"infections",
"immunology",
"microbiology",
"lymphoid",
"organs",
"host-pathogen",
"interaction",
"animal",
"models",
"adaptive",
"immunity",
"mechanisms",
"of",
"resistance",
"and",
"susceptibility",
"model",
"organisms",
"immune",
"defense",
"bacterial",
"pathogens",
"immunotherapy",
"immunizations",
"animal",
"models",
"of",
"infection",
"t",
"cells",
"microbial",
"pathogens",
"biology",
"mouse",
"immune",
"response",
"immune",
"system",
"gram",
"positive",
"monocytes",
"immunity",
"virology",
"innate",
"immunity"
] |
2011
|
Inflammatory Monocytes and Neutrophils Are Licensed to Kill during Memory Responses In Vivo
|
The fragile X-related disorders result from expansion of a CGG/CCG microsatellite in the 5’ UTR of the FMR1 gene . We have previously demonstrated that the MSH2/MSH3 complex , MutSβ , that is important for mismatch repair , is essential for almost all expansions in a mouse model of these disorders . Here we show that the MSH2/MSH6 complex , MutSα also contributes to the production of both germ line and somatic expansions as evidenced by the reduction in the number of expansions observed in Msh6-/- mice . This effect is not mediated via an indirect effect of the loss of MSH6 on the level of MSH3 . However , since MutSβ is required for 98% of germ line expansions and almost all somatic ones , MutSα is apparently not able to efficiently substitute for MutSβ in the expansion process . Using purified human proteins we demonstrate that MutSα , like MutSβ , binds to substrates with loop-outs of the repeats and increases the thermal stability of the structures that they form . We also show that MutSα facilitates binding of MutSβ to these loop-outs . These data suggest possible models for the contribution of MutSα to repeat expansion . In addition , we show that unlike MutSβ , MutSα may also act to protect against repeat contractions in the Fmr1 gene .
The fragile X ( FX ) -related disorders ( FXDs ) are repeat expansion diseases that result from an increase in the length of a CGG/CCG-repeat tract in the 5’ UTR of the FMR1 gene ( reviewed in [1] ) . This expansion occurs from an unstable premutation ( PM ) allele that contains 55–200 repeats . The repeat is prone to expansion in germ line and somatic cells in humans and in a FXD mouse model with a targeted insertion of ~130 FX-repeats [2–4] . The molecular basis of this instability is not known . Individual strands of the FX repeat form hairpins and other atypical structures some of which may be folded and include mismatches [5–12] and current thinking is that these structures are the substrates for the expansion pathway [13] . We have previously shown that a number of different pathways that affect repeat instability are active in a mouse model of the FXDs , one that gives rise to expansions , one that results in the error-free repair of the expansion substrate and perhaps two different contraction pathways [14 , 15] . We have also shown that the mismatch repair ( MMR ) complex MutSβ , a heterodimer of MSH2 and MSH3 , is required for 98% of germ line and all somatic expansions in the FXD mouse [14] . This is consistent with what has been seen in some , but not all mouse models of other repeat expansion diseases [16–18] . MutSα , the other MSH2-containing complex present in mammals , has been shown to either have no effect or to protect against repeat instability in various mouse models [16 , 17 , 19] . For example , in a mouse model for myotonic dystrophy type 1 ( DM1 ) , MutSα protects against somatic expansions [19] and in a mouse model for Friedreich ataxia ( FRDA ) MutSα protects against both germ line expansions and contractions [17] . However , FRDA is also unique amongst the repeat expansion diseases studied thus far in that MutSα has also been shown to be involved in generating somatic expansions in the mouse model and in patient-derived induced pluripotent stem cells [20] . Whether or not this involvement of MutSα reflects some unique property of the GAA/TTC-repeats is not known . Furthermore , why MutSα protects against , rather than promotes , germ line expansions is also an open question [17] . As part of an effort to better understand the mechanisms of repeat instability in the FXD mouse model , we examined the somatic and intergenerational instability of the FX repeat in mice lacking MSH6 , the MSH2-binding protein in the MutSα heterodimer . These data , together with our biochemical studies on the binding of these complexes to CGG- and CCG-repeats have interesting implications for the mechanism of repeat instability in the FXDs .
Our previous work indicates that in the FXD mouse model expansion , contraction and error-free pathways co-exist in germ line cells [4] . This can complicate the interpretation of intergenerational transmission data since a decrease in expansions can result either because the expansion pathway has become less efficient or because a pathway that protects against contractions has been impaired , or some combination of both . However , the interpretation of somatic instability data is simpler since all evidence to date suggests that the contraction pathway is not active in adult somatic cells in this mouse model [3 , 4 , 15] . Since somatic expansion is much more extensive in males than in females in the FXD mouse model [21] , we examined the effect of the loss of MSH6 on expansion in different organs of male mice that were 6 months old . This time point was chosen since Msh6-/- male mice rarely survive beyond this age . However , since somatic expansion is clearly discernable in Msh6+/+ males at this age , any effect of the loss of MSH6 can be readily detected . To examine the effect of the loss of MSH6 on somatic instability in male mice we carried out PCR across the repeat and then resolved the PCR products by high-resolution capillary electrophoresis . Analysis of the PCR products produced from somatic tissue of Msh6+/+ animals at weaning at 3 weeks of age ( tail 1 ) or in organs that do not show somatic expansion , like heart , typically reveal a Gaussian distribution of PCR products with relatively little deviation of these products from the mean ( Fig 1A ) . These PCR profiles are indistinguishable from those obtained from samples taken at birth [3] . Some of these products represent strand-slippage products that are generated when amplifying through long repeat tracts . In particular , the PCR products smaller than the major allele that do not change with genotype , age or tissue , fall into this category . We then used these PCR profiles to determine the somatic instability index ( SII ) , a quantitative measure of the extent of repeat expansion [22] . In MSH6+/+ males the SII for heart was -0 . 1 ( Fig 1B ) . This negative number is not evidence of contractions since the SII in heart does not change with age and the PCR profile seen in old animals corresponds to the original allele size determined at birth ( [3] and Fig 1A ) . The negative value likely reflects contribution of the products of strand-slippage to the SII . In organs other than the heart , the SII was positive with the lowest SII being seen in kidney and the highest in liver and testes as we had previously observed [3] . The loss of MSH6 was associated with a significant reduction in the SII in many organs of male mice ( Fig 1B ) . The organs most affected are those with the highest level of expansion in Msh6+/+ animals , namely , the tail , brain , liver and testis . However , the distribution of products smaller than the major allele are similar in all tissues including the tail sample taken at weaning and the heart , an organ that shows little , if any instability ( Fig 1A ) . We also did not see evidence of somatic contractions in Msh2-/- mice that lack both MutSα and MutSβ [23] . Thus the failure to see evidence of contractions in Msh6-/- mice is not the result of an offsetting effect of MutSβ-mediated expansions . We can therefore conclude that the reduced SII in Msh6-/- mice is not the result of contractions that have now become apparent as a result of the loss of MSH6 . Rather the reduction must reflect either a direct or indirect role of MSH6 , and thus MutSα , in generating somatic expansions . Female Msh6+/+ mice show less somatic expansion than males [21] . This makes it difficult to see significant effects of the loss of MSH6 in young animals . We thus confined our examination of somatic instability in Msh6-/- females to the few that survived to 12 months of age . Note that despite the females being twice as old as the males , the SII in Msh6+/+ females was still lower than it was in most of the corresponding organs of 6 month old males , consistent with reduced somatic instability in females ( Fig 1B ) . Nonetheless , a role for MSH6 in generating somatic expansions was apparent in Msh6-/- females albeit only in tail and ovary ( Fig 1B ) . Thus , the loss of MSH6 , like the loss of MSH3 , reduces the extent of somatic expansion . However , while somatic expansions are completely eliminated in Msh3-/- males and females on a similar genetic background [14] , some expansion is still evident in Msh6-/- mice of both sexes . We hypothesized that the loss of MSH6 would also affect germ line expansions with loss of two copies of the gene having a larger effect than the loss of one copy . We thus examined the transmission of the PM allele on intergenerational transfer from Msh6+/+ , Msh6+/- and Msh6-/- parents . The Jonckheere-Terpstra test for ordered alternatives showed that there was a statistically significant trend towards fewer expansions with decreasing Msh6 gene dosage ( p<0 . 001 for both paternal and maternal transmission ) . Pairwise comparisons demonstrated that while the expansion frequencies in the offspring of Msh6+/- parents was not significantly different from the expansion frequency in the offspring of Msh6+/+ parents , the progeny of Msh6-/- males and females had significantly fewer expansions than the progeny of either the Msh6+/+ ( Fisher’s exact test; p = 0 . 0003 for paternal and p<0 . 0001 for maternal transmission respectively; Fig 2 ) or the Msh6+/- parents ( Fisher’s exact test; p = 0 . 008 paternal and p<0 . 0001 for maternal transmission respectively; Fig 2 ) . There was also a significant difference in the distribution of the transmitted alleles for both maternal and paternal transmissions ( Mann-Whitney U test; p<0 . 0001 for both males and females ) . There is no evidence to date to suggest that somatic and germ line expansions occur by different mechanisms in the FXD mouse . Thus the simplest interpretation of our data is that the decline in germ line expansions seen in Msh6-/- animals reflects a contribution of MutSα to the germ line expansion process . This would be above and beyond the 2% of expansions that are MSH2-dependent , but MSH3-independent that we previously attributed to MutSα [14] . However , in contrast to the 80:20 ratio of unchanged to contracted alleles seen in Msh3-/- males [14] , in Msh6-/- males the ratio was 50:50 ( 27% vs 29% of the total alleles ) . The ratio of unchanged to contracted alleles in Msh2-/- mice is intermediate between the two ( 60:40 ) consistent with the combined contribution of MutSβ and MutSα complexes to the overall distribution of residual alleles [23] . The decline in the proportion of unchanged alleles in Msh2-/- and Msh6-/- animals relative to Msh3-/- mice may reflect an additional role for MutSα in protecting against contractions which occur in the germ line , but not somatic cells . Thus the decline in expansions seen on intergenerational transmission in Msh6-/- mice may represent some combination of the reduced efficacy of the expansion pathway together with the reduced efficacy of the pathway that protects against contractions . We have previously shown that loss of MSH3 results in a change in the distribution of contraction sizes that are seen on intergenerational transmission [14] . Specifically while animals wildtype with respect to mismatch repair show a bimodal distribution of repeat sizes with the first modal class having lost 1–2 repeats and a second modal class having lost >7 repeats , Msh3-/- mice show a significant loss of alleles falling into the second modal class . Most notably in Msh3-/- males all contractions involved the loss of just a single repeat . This would be consistent with a role for MutSβ in generating larger contractions . To assess the contribution of MutSα to contractions we examined the distribution of contracted alleles in Msh6-/- animals . Msh6-/- males die young making it difficult to collect enough data on the contraction sizes of paternally transmitted alleles . Therefore we analyzed the effect of the loss of MSH6 on the distribution of contraction sizes by doing small pool PCR on sperm DNA isolated from 2 month old Msh6-/- males ( Fig 3A ) . The expansion frequency in Msh6+/+ sperm was generally lower than that observed in the live born progeny of Msh6+/+ animals . This could reflect the difference in the ages of the sperm donors ( 2 months ) versus the fathers ( 2–6 months ) , since there is a progressive increase in the proportion of expanded alleles with age [4] . A contribution of low level of contamination of the sperm sample with less expansion-prone somatic cells also cannot be completely excluded . However , the expansion frequency was also lower in the sperm of Msh6-/- mice than in the progeny of Msh6-/- males . Thus , as expected , there were fewer expansions and more contractions than in the sperm of Msh6+/+ males of the same age ( Fig 3B , Fisher’s exact test; p<0 . 0001 ) . In addition , the distribution of alleles in Msh6-/- and Msh6+/+ gametes was significantly different by the Mann-Whitney U test ( p<0 . 0001 ) . This is generally consistent with the data derived from analysis of the progeny of Msh6-/- males ( Fig 2 ) . In any event , in contrast to what is seen in Msh3-/- animals , the distribution of contractions in Msh6-/- sperm was similar to that seen in Msh6+/+ sperm ( Mann-Whitney U test; p = 0 . 14 ) . Our data thus suggest that MSH6 , and therefore MutSα does not severely impact the distribution of contraction sizes as does MutSβ [14] . It has been suggested that MSH2 partitions between available pools of MSH3 and MSH6 and thus that the loss of MSH6 should thus not lead to a decrease in MutSβ [24–26] . To verify this we compared the levels of each of the three proteins in various organs of Msh2-/- , Msh3-/- , Msh6-/- mice and mice WT for all three proteins . As can be seen in Fig 4 , the absence of MSH2 led to a complete loss of bands with the predicted mobility of MSH3 and MSH6 , consistent with previous data demonstrating that the formation of MutSα and MutSβ complexes protects their subunits from degradation [24 , 26 , 27] . The loss of MSH6 resulted in a much larger decrease in the levels of MSH2 than did the loss of MSH3 in all organs tested . However , as can be seen in Fig 4 , the levels of MSH3 are comparable in the brain and testes of Msh6+/+ and Msh6-/- mice and after normalizing to β-actin , no significant difference in the levels of MSH3 were detected . Two MSH3 bands were seen the liver and ovary of Msh6+/+ and Msh6-/- animals that were absent in extracts from Msh3-/- animals . These bands are thus likely to be MSH3-related . A similar pair of bands was seen in mouse spleen extracts using a MSH3 antibody that was directed to a similar region of the protein as the antibody we used [28] . However , in that report , only one band was detected with an antibody that recognizes a very N-terminal epitope . The N-terminal end of MSH3 known to be prone to degradation [29 , 30] and it is possible that the smaller of the two bands represents a proteolytic degradation product of MSH3 in which the N-terminus had been lost . Thus , while the original levels of MSH3 in these organs are difficult to determine unequivocally , the data from brain and testes suggests that the effect of the loss of MSH6 on somatic and germ line expansion is not due to an indirect effect on the levels of MSH3 , at least in some of the most expansion-prone organs in these animals . We have previously shown that MutSβ binds to loop-outs formed by CCG- and CGG-repeats [14] . To see whether the same was true of MutSα , we examined the binding of this protein to substrates containing a loop-out of ( CCG ) 13 or ( CGG ) 13 . These substrates were modeled on those used previously to examine MutSβ binding to CAG-repeats [31] . We also included MutSβ with a deletion of the unstructured N-terminal region of MSH3 [30] . This region of MSH3 is not involved in DNA or nucleotide binding [32] and this MutSβ complex has the same binding affinities for homoduplexes , tailed substrates and insertion/deletion loops ( IDLs ) as complexes containing the full length MSH3 protein , as well as the same rate constants and ATPase activities on these substrates [30] . Use of this MSH3 variant has the advantage of producing a DNA:protein complex with a mobility that is distinctly different from that of the DNA:MutSα complex . Both MutS complexes were of equivalent concentration as evidenced by the fact that equivalent amounts of protein contained equivalent amounts of MSH2 ( S2 Fig , panel A ) . There was also no evidence of any degradation of the subunits as evidenced by the single products detectable on western blotting with antibodies to MSH2 , MSH3 and MSH6 ( S2 Fig , panel A ) . As expected MutSα does not bind well to either homoduplex DNA or a loop-out of ( CA ) 3 , a good substrate for MutSβ-mediated but not MutSα-mediated repair ( S2 Fig , panel B ) . In contrast , MutSβ binds effectively to the ( CA ) 3 loop-out with even low concentrations of protein being able to shift almost all of the substrate . Limited binding of MutSβ to the homoduplex was also seen ( S2 Fig , panel B ) . This binding is much less extensive than the binding of MutSβ to the ( CA ) 3 loop-out as evidence by the fact that no unbound ( CA ) 3 substrate was seen at a protein concentration of 0 . 8 nM , while most of the homoduplex remained unbound even at the highest protein concentration tested ( 20 nM ) . Binding of MutSβ to homoduplexes has been previously reported where it has been attributed to end binding [33–35] . MutSα binds to a substrate containing a G•T mismatch ( S2 Fig ) and to ( CCG ) 13 and ( CGG ) 13 loop-outs ( Fig 5 ) . It also binds to ( CAG ) 13 and ( CTG ) 13 loop-outs ( S2 Fig , panel B ) . However , MutSα binds less well to the repeat substrates than to the G•T mismatch since binding of MutSα to G•T mismatch depletes all of the free probe at the highest protein concentration , while some free probe remains with all the repeat substrates . MutSα binding to the repeat substrates is also less extensive than the binding of MutSβ ( compare lanes 2 and 3 and 12 and 13 of Fig 5 ) . MutSα stimulates MutSβ binding to a canonical MMR substrate containing a 2 nucleotide insertion/deletion when present at a high MutSα:MutSβ ratio [36] . To test whether this was also true for FX repeat-containing substrates we compared the binding of MutSβ in the presence and absence of an excess of MutSα . MutSβ binding to both canonical and repeat containing substrates produced multiple DNA:protein complexes as can be seen in Fig 5 and S2 Fig . Multiple DNA:MutSβ complexes have been previously reported for both yeast and human proteins binding to canonical MutSβ substrates [34 , 35 , 37] as well as ( CAG ) 13 loop-outs [31] . The different MutSβ containing products could reflect either multiple MutSβ molecules binding to a single DNA molecule or to alternative binding modes . MutSβ binding to ( CCG ) 13 and ( CGG ) 13 substrates was increased in the presence of MutSα . This was evidenced most clearly as an increase in the amount of the DNA:MutSβ complex with the second highest mobility ( DNA:MutSβ2; compare lanes 3 and 4 and 13 and 14 of Fig 5 ) . This is not likely to be a non-specific effect since the addition of much higher concentrations of BSA do not have the same effect ( S2 Fig , panel D ) . The increase in MutSβ binding was associated with a decrease in the amount of the MutSα-shifted band . Since the substrate has not been depleted , this decrease is unlikely to reflect competition between MutSα and MutSβ for binding to the substrate . Rather , it may reflect the incorporation of MutSα into one or more higher molecular weight species . Indeed at higher MutSα concentrations , a new shifted product , indicated by the open arrowheads in Fig 5 , is apparent . This product is associated with a decline in the levels of the 2 most rapidly migrating DNA:MutSβ complexes . It is not seen in the absence of MutSβ even when very high concentrations of MutSα are used ( lanes 8 and 9 and 18 and 19 of Fig 5 ) . It thus likely represents complexes containing both MutSα and MutSβ . In reactions containing both MutSα and MutSβ a small amount of a second novel band is also seen with the CCG-substrate ( indicated by the grey arrowhead in Fig 5 , lane 17 ) . This band may represent the result of binding of multiple MutSβ and MutSα complexes to the CCG substrate or complexes in which the binding modes differ from the complex with the faster mobility . We have previously shown that MutSβ is able to increase the stability of the CCG-loop-out at physiological temperatures . To assess whether MutSα binding had the same effect , we monitored the thermal denaturation of the oligonucleotide in the presence of BSA or MutSα as previously described [14] . Since the 5’ end of the oligonucleotide was labeled with 5-carboxy-X-rhodamine ( ROX ) , a fluorescence donor and the 3’ end was labeled with IOWA Black RQ , a fluorescence acceptor , the stability of the hairpins could be assessed in the presence of protein by monitoring the effect of increasing temperature on the fluorescence at 608 nm , the ROX emission wavelength . The oligonucleotide was denatured and cooled under conditions in which the repeats are known to form hairpins [7 , 10 , 38–42] . The oligonucleotide was then mixed with either BSA or MutSα and the thermal denaturation of the oligonucleotides monitored as previously described [14] . The melting curves obtained for both protein-CCG-repeat mixtures fit a two-state model ( S3 Fig ) . While the best-fit for CCG-repeat melting in the presence of MutSα was within acceptable limits , the data suggest that the process by which the oligonucleotide melts in the presence of MutSα may be more complex than it is either in the presence of BSA or MutSβ [14] . The thermodynamic parameters derived from analysis of the melting curves are shown in Table 1 . As for MutSβ , the presence of MutSα resulted in higher apparent ∆Gs at 37˚C than is seen in the presence of BSA . This suggests that MutSα , like MutSβ , increases the stability of the CCG-repeat structure at physiological temperature . However , the significant differences in the enthalpy of melting ( ∆Hm ) of the oligonucleotide in the presence of MutSβ ( 52 . 4 ± 4 . 1 kcal/mol; [14] ) and MutSα ( 81 . 3 ± 3 . 0 kcal/mol ) suggest that the consequence of binding of these two complexes differs . This may reflect the very different modes of binding of these complexes to their substrates [30 , 34] .
We have previously shown that the loss of MSH2 eliminates all germ line and somatic expansions in the FXD mouse and that most of these expansions are MutSβ-dependent since the loss of MSH3 eliminates almost 98% of germ line expansions and all somatic ones [3 , 14] . The remaining 2% of MSH2-dependent germ line expansions are presumably the result of MutSα-dependent events . However , we show here that loss of MSH6 , and thus MutSα , reduces both the germ line and the somatic expansion frequency by much more than 2% ( Figs 1–3 ) . A comparison of the relative levels of MSH2 in Msh3-/- and Msh6-/- mice showed that the loss of MSH6 resulted in a greater decrease of MSH2 than the loss of MSH3 ( Fig 4 ) . The fact that no MSH3 or MSH6 is seen in Msh2-/- mice ( Fig 4 ) is consistent with previous reports suggesting that the levels of MSH2 , MSH3 and MSH6 are interdependent and that there is very little free MSH3 or MSH6 in cells [24 , 25] . Thus our data would be consistent with the interpretation that more MSH2 is in a heterodimer with MSH6 than is in a heterodimer with MSH3 , i . e . , that MutSα is more abundant than MutSβ in these animals . This finding is consistent with what has been reported for human cells [43 , 44] and mice with a mixed C57BL6/129/OLA/FVB background [16] , but not with what is seen in FVB mice [28] . While we did not assess the absolute amount of MSH3 , a comparison of the relative levels of MSH3 in Msh6+/+ and Msh6-/- mice showed that the loss of MSH6 did not result in a detectable decrease in MSH3 levels in expansion-prone organs like brain and testes ( Fig 4 ) . Since MSH2 levels were reduced in Msh6-/- mice it is possible that MSH6 is acting indirectly via decreasing the amount of MSH2 available to form the MutSβ complex . However , since MSH3 levels are not significantly lower in Msh6-/- mice and MSH3 is thought to be stable only when in the MutSβ complex [24 , 25] , MSH6 may well be playing a different role in the expansion process . No comparable decrease in expansions is seen in Msh6-/- mice in models of other repeat expansion diseases where MSH3 has been implicated in the expansion process and where a similar excess of MutSα was seen [16] . In addition , the loss of MSH6 in the same mouse strain or in human cells does not result in reduced repair of typical MutSβ substrates [45–49] . Thus significant redeployment of MutSβ to other sites in the genome to compensate for the loss of MutSα is also unlikely to account for the decrease in germ line and somatic expansions seen in Msh6-/- mice . However , since very few germ line expansions and no somatic ones are detected in mice that lack MutSβ [14] , MutSα is not able to efficiently substitute for MutSβ in the expansion process despite the relative abundance of MutSα in these animals ( Fig 4 ) . A contribution by both MutSα and MutSβ to repeat expansion is consistent with our previous observations that while MutSβ levels alone do not correlate well with the levels of somatic instability across 5 different organs , a better correlation is seen when the levels of both MutSβ and MutSα are considered [3] . Thus , our data suggest that the involvement of MutSα in repeat expansion is not unique to somatic expansion of GAA/TTC-repeats in FRDA , contributing to both germ line and somatic expansion in the FXD mouse . Whether MutSα acts independently of MutSβ in FRDA unknown . The nature of MutSα’s role in the expansion process in the FXD mouse is also unclear . MutSα may be acting to facilitate MutSβ-dependent expansion by increasing the stability of the expansion substrates as it does in vitro ( Table 1 ) , or by protecting the substrates from repair by another mechanism , thus allowing more time for the hairpins or other atypical structures to be processed by MutSβ to generate expansions . MutSα also promotes binding of MutSβ to the repeat substrates ( Fig 5 ) in a manner reminiscent of MutSα’s effect on MutSβ binding to a canonical MutSβ substrate [36] . This property may reflect yet another way that MutSα could facilitate MutSβ-mediated repeat expansions . A role for MutSα in repeat expansion has not been observed in mouse models of CTG/CAG-repeat expansion diseases despite the fact we have shown that MutSα binds to those repeats in vitro ( S2 Fig ) . It may be that the effect of the loss of MutSα is only apparent under certain circumstances . For example , it may be that in the FXD mouse model where the expansion frequency is high , the amount of the expansion substrate formed in the germ line exceeds the processing capacity of MutSβ acting alone . Under these conditions the effect of MutSα would become apparent . When only moderate levels of the expansion substrate are produced , an effect of the loss of MutSα might only be seen in mice with reduced MutSβ levels since the available MutSβ in Msh3+/+ animals may be sufficient to process all the expansion substrates without the assistance of MutSα . At the other end of the spectrum when the expansion substrates are present only at very low levels , either one of the MutS complexes may be sufficient to process them . This may explain the perplexing observation that in a mouse model of Huntington disease , loss of MSH2 eliminates germ line expansions but neither the loss of MSH3 nor the loss of MSH6 had any effect on the expansion frequency [18] . This idea would also be consistent with our observation that in the FXD mouse , the loss of MSH6 had more of an effect on somatic expansion in males than in females ( Fig 1 ) , since the expansion process is less extensive in females [21] and thus the requirement for MutS proteins may be lower . Thus , the different effects of MutSα and MutSβ on expansion in different models , in germ line versus somatic cells , or in males and females , may not necessarily reflect differences in the mechanisms of expansion , but rather differences in the levels of the MutS complexes relative to the substrates that potentially could be processed to generate expansions . In addition to contributing to the generation of expansions , our data suggest that MutSα may also act to protect against germ line contractions as evidenced by the reduction in the proportion of unchanged alleles in Msh2-/- [23] and Msh6-/- animals ( Fig 2 ) relative to Msh3-/- mice [14] . Protection against germ line contractions by MutSα would be consistent with what has been reported for both the GAA/TTC-mouse model [17] and a CAG/CTG-mouse model [18] . Protection against contractions by MutSα also would be consistent with a typical MMR process albeit one that is triggered by an atypical repair substrate . The nature of the FX hairpins with the high frequency of single mismatches may account for the ability of MutSα to bind to and coordinate their repair . The ability of MutSα to contribute both to error-free repair and to expansions may reflect MutSα’s ability to participate in more than one DNA repair pathway [50 , 51] . We have recently demonstrated that a hypomorphic mutation in Polβ , a key DNA polymerase involved in base excision repair ( BER ) , reduces expansion in the FXD mouse [52] . How MutSα and MutSβ interface with the BER pathway to generate expansions in these models remains an open question . One possibility is that MutSβ and MutSα act downstream of DNA damage excision to stabilize loop-outs formed during strand-slippage and strand-displacement synthesis that is mediated at least in part by Polβ . We speculate that normal signaling by MutSα results in MMR of these loop-outs resulting in error-free repair , while MutSβ , alone or together with MutSα , can channel them into a different repair pathway that results in expansions .
This work was carried according to ARAC guidelines and procedures as outlined in the Guide for the Care and Use of Laboratory Animals , U . S . Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing , Research , and Training and Public Health Service Policy on Humane Care and Use of Laboratory Animals . This work was approved by the NIDDK Animal Care and Use Committee ( ASP: K021-LMCB-12 and K021-LMCB-15 ) . Oligonucleotides were obtained from Integrated DNA technologies ( IDT , Coralville IA ) and are listed in Table 2 . Purified human MutSα was a kind gift of Drs Hsieh and Geng ( NIDDK , NIH ) . Purified human MutSβ was a kind gift of Drs Yang and Li ( NIDDK , NIH ) . This MutSβ complex contained a “trimmed” version of MSH3 containing amino acids 211–1125 . This MutSβ complex has the same binding affinities for homoduplexes , tailed substrates and IDLs as complexes containing the full length MSH3 protein , as well as the same rate constants and ATPase activities [30] . The generation of the FXD mice was described previously [2] . These mice are on a C57BL/6 background . The Msh6+/- mice were generated previously [49 , 53] and cryopreserved embryos were obtained from the NCI Mouse Repository ( Frederick , MD ) . These mice are also on a predominantly C57BL/6 background . Live born pups were generated from these embryos by implantation into the oviduct of pseudopregnant recipients using standard procedures . F2 Msh6+/- parents were bred to generate Msh6+/+ , Msh6+/- and Msh6-/- littermates . Multiple breeding pairs from the same parents were set up for each genotype . The litters for each genotype considered for this analysis had a similar parental age distribution . This was the same genetic background and breeding strategy that we had used previously to examine the effect of the loss of MSH3 on the expansion frequency [14] . Mice were maintained in accordance with the guidelines of the NIDDK Animal Care and Use Committee and with the Guide for the Care and Use of Laboratory Animals ( NIH publication no . 85–23 , revised 1996 ) . Sperm was isolated from the cauda epididymis as previously described [54] , pelleted twice by centrifugation at 500 g for 5 min and the pellet resuspended first in PBS and then in 100 μl of a solution containing a 90:10 mixture of ATL lysis buffer ( Qiagen , Valencia , CA ) and a 20 mg/ml proteinase K solution ( Invitrogen , Carlsbad , CA ) . The samples were then incubated at 55°C overnight before the addition of 30 μl of 5 M NaCl . The resultant precipitate was pelleted by centrifugation and the supernatant transferred to a new tube and mixed with 130 μl of ethanol . The DNA was then pelleted by centrifugation and dissolved in TE by incubation overnight at 55°C . This protocol results in little , if any , contamination with somatic DNA [54] . Genomic DNA from mouse tails was extracted using KAPA Mouse Genotyping Kit ( KAPA Biosystems , Wilmington , MA ) . Genomic DNA from other tissues was extracted using a Maxwell16 Mouse tail DNA purification kit ( Promega , Madison , WI ) according to the manufacturer’s instructions . Msh6 genotyping was carried out with Taq DNA polymerase in standard buffer with either the M010/M011 primer pair to detect the WT allele and M012/M013 to detect the mutant allele . The PCR parameters were 1x 94°C for 1 min . , 35x ( 94°C for 1 min . , 60°C for 2 min . and 72°C for 1 min ) , followed by 1x 72°C for 3 min . The presence of the PM allele and its repeat number was determined using a fluorescent PCR assay and FraxM4 and FraxM5 primer pair as described previously [3] . The somatic instability index ( SII ) was calculated from the GeneMapper profiles of DNA from different organs as previously described [3 , 22] and used to evaluate the extent of somatic expansion in adult mice . For small pool PCR analysis from sperm , the DNA was diluted to 3 pg/μl ( roughly 1 haploid genome equivalent/μl ) . The diluted DNA was then subjected to nested PCR . The first round of PCR was carried out using the primers FraxC and FraxF in a 25 μl PCR mix as described previously [55] . One microliter of this PCR mix was used in second round of PCR with the FraxM4 and FraxM5 primers . Roughly 50% of the reactions contained a PCR product , consistent with the idea that each positive PCR likely represents the products of amplification of DNA from a single sperm cell . An exact Jonckheere-Terpstra test of trend in ordered counts was carried out using StatXact software ( version 8; Cambridge , Massachusetts ) . Fisher’s exact test was carried out using the GraphPad QuickCalcs website ( http://www . graphpad . com/quickcalcs ) . The Mann-Whitney U test was carried out using VassarStats ( http://vassarstats . net/ ) . We set the significance level ( α ) at 0 . 050 for the pairwise comparisons . For the comparisons of WT , heterozygous and homozygous null animals this corresponds to p = 0 . 015 after adjusting for multiple testing using the ( relatively conservative ) Bonferroni correction . Hartigans’ dip test was calculated using the dip . test command in the R diptest library . Total protein extracts were prepared from flash frozen brain , liver , testes and ovary of 6-month old mice . Tissues were homogenized using a tissue homogenizer ( Precellys 24 , Bertin Technologies , Berlin , Germany ) with T-PER protein extraction reagent ( Pierce Biotechnology , Inc , Rockford , IL ) supplemented with complete , Mini , EDTA-free protease inhibitor cocktail ( Roche Applied Science , Indianapolis , IN ) . Nuclear extracts of liver proteins were prepared using the NE-PER Nuclear and cytoplasmic extraction reagents ( Pierce Biotechnology , Inc , Rockford , IL ) according to the manufacturer’s instructions . The protein concentrations were determined using a Bio-Rad protein assay kit ( Bio-Rad , Hercules , CA ) . Proteins were heated for 10 minutes at 70°C in LDS-Sample Buffer ( Life Technologies , Grand Island , NY ) , resolved by electrophoresis on either 3–8% NuPAGE Novex Tris-Acetate gels ( Life Technologies ) or 4–12% NuPAGE Novex Tris-Bis gels ( Life Technologies ) and transferred to nitrocellulose membranes using the iBlot transfer apparatus ( Life Technologies ) according to the manufacturer’s instructions . Membranes were blocked for one hour at room temperature in 5% ECL Prime blocking agent ( GE Healthcare Bio-Sciences ) in TBST , then incubated overnight at 4°C with antibodies to MSH2 ( ab70270 , Abcam , Cambridge , MA ) at a concentration of ( 1:10000 ) , MSH3 ( sc-271079 , Santa Cruz , Dallas , TX ) at a concentration of ( 1:1000 ) and MSH6 ( BD 610918 , BD Biosciences , Franklin Lakes , NJ ) at a concentration of ( 1:1000 ) . The secondary antibodies ( anti-mouse IgG , NA931V and anti-rabbit IgG , NA934V , GE Healthcare Bio-Sciences ) were both used at a dilution of 1:5000 . After addition of the ECL Prime detection reagent ( GE Healthcare Bio-Sciences ) , the blot was imaged using a Fluorchem M imaging system ( Proteinsimple , Santa Clara , CA ) . Beta-actin ( anti-mouse ab8227 , Abcam , Cambridge , MA ) was used as a loading control for total cell extracts and lamin B ( ab16048 , Abcam , Cambridge , MA ) for nuclear extracts . A representative example of a full blot of testes protein extracts showing binding to MSH2 , MSH3 , MSH6 and the loading control β-actin is shown in S1 Fig . Western blots were repeated several times and always included molecular weight markers and extracts from the appropriate null mice as negative controls . To evaluate whether the loss of MSH6 affected the levels of MSH3 in Msh6-/- mice , knowledge of the absolute levels of each protein is not necessary . Since the levels of MSH3 in each group of animals was tested with equivalent amounts of protein using the same antibody on the same gel , the avidity of the MSH3 antibody relative to the avidity of the MSH6 or MSH2 antibodies is not an issue . We thus were able to directly compare the levels of MSH3 in WT , Msh3-/- and Msh6-/- animals by determining the amount of each protein relative to β-actin ( total protein extracts ) or lamin B1 ( nuclear extracts ) using the AlphaView software for FluorChem Systems ( Proteinsimple , Santa Clara , CA ) . The levels of MSH2 and MSH6 in these animals were determined in the same way . The oligonucleotides used in EMSA were prepared as described previously [14] . The binding reactions were carried out using the Gelshift chemiluminescent EMSA kit ( Active Motif , Carlsbad , CA ) according to the manufacturer’s instructions using the indicated amounts of purified human MutSα and human MutSβ and 2 fmoles of the duplexed oligonucleotides as described previously [14] . The oligonucleotide used for thermal analysis consisted of a single strand of DNA comprised of 10 copies of CCG with the 5’ end labeled with 5-carboxy-X-rhodamine ( ROX ) and the 3’ end with IOWA Black RQ . The oligonucleotide was prepared as described previously [14] and MutSα or BSA was added to 360 nM as indicated . Thermal denaturation was monitored as described previously [14] . The melting curve was consistent with a two-state model ( S3 Fig ) and the thermodynamic parameters were thus derived from the melting curve using a two-state model ( closed and open states ) .
|
The repeat expansion diseases are a group of human genetic disorders that are caused by expansion of a specific microsatellite in a single affected gene . How this expansion occurs is unknown , but previous work in various models for different diseases in the group , including the fragile X-related disorders ( FXDs ) , has implicated the mismatch repair complex MutSβ in the process . With the exception of somatic expansion in Friedreich ataxia , MutSα has not been reported to contribute to generation of expansions in other disease models . Here we show that MutSα does in fact play a role in both germ line and somatic expansions in a mouse model of the FXDs since the expansion frequency is significantly reduced in Msh6-/- mice . However , since we have previously shown that loss of MutSβ eliminates almost all expansions , MutSα is apparently not able to fully substitute for MutSβ in the expansion process . We also show here that MutSα increases the stability of the structures formed by the fragile X repeats that are thought to be the substrates for expansion and promotes binding of MutSβ to the repeats . This , together with our genetic data , suggests possible models for how MutSα and MutSβ , could co-operate to generate repeat expansions in the FXDs .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"dna-binding",
"proteins",
"nucleotides",
"animal",
"models",
"germ",
"cells",
"model",
"organisms",
"molecular",
"biology",
"techniques",
"genotyping",
"extraction",
"techniques",
"sperm",
"research",
"and",
"analysis",
"methods",
"protein",
"extraction",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"cells",
"proteins",
"mouse",
"models",
"molecular",
"biology",
"biochemistry",
"testes",
"cell",
"biology",
"anatomy",
"oligonucleotides",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"polymerase",
"chain",
"reaction",
"genital",
"anatomy"
] |
2016
|
A MutSβ-Dependent Contribution of MutSα to Repeat Expansions in Fragile X Premutation Mice?
|
The study of dynamic functions of large-scale biological networks has intensified in recent years . A critical component in developing an understanding of such dynamics involves the study of their hierarchical organization . We investigate the temporal hierarchy in biochemical reaction networks focusing on: ( 1 ) the elucidation of the existence of “pools” ( i . e . , aggregate variables ) formed from component concentrations and ( 2 ) the determination of their composition and interactions over different time scales . To date the identification of such pools without prior knowledge of their composition has been a challenge . A new approach is developed for the algorithmic identification of pool formation using correlations between elements of the modal matrix that correspond to a pair of concentrations and how such correlations form over the hierarchy of time scales . The analysis elucidates a temporal hierarchy of events that range from chemical equilibration events to the formation of physiologically meaningful pools , culminating in a network-scale ( dynamic ) structure– ( physiological ) function relationship . This method is validated on a model of human red blood cell metabolism and further applied to kinetic models of yeast glycolysis and human folate metabolism , enabling the simplification of these models . The understanding of temporal hierarchy and the formation of dynamic aggregates on different time scales is foundational to the study of network dynamics and has relevance in multiple areas ranging from bacterial strain design and metabolic engineering to the understanding of disease processes in humans .
The network of interactions that occur between biological components on a range of various spatial and temporal scales confer hierarchical functionality in living cells . In order to determine how molecular events organize themselves into coherent physiological functions , in silico approaches are needed to analyze how physiological functions emerge from the evolved temporal structure of networks . Time scale decomposition is a well-established , classical approach to dissecting network dynamics and there is a notable history of analyzing the time scale hierarchy in metabolic networks and matching the events that unfold on each time scale with a physiological function [1]–[6] . This approach enables the identification of the independent , characteristic time scales for a dynamic system . In particular it has been possible to decompose a cell-scale kinetic model of the human red blood cell in time to show how its key metabolic demands are met through a dynamic structure-function relationship . The underlying principle is one of aggregation of concentration variables into ‘pools’ of concentrations that move in tandem on slower time scales [5] , [7] . The dynamics of biological networks characteristically span large time scales ( 8 to 10 orders of magnitude ) , which contributes to the challenge of analyzing and interpreting related models . However , there is structure in this dynamic hierarchy of events , particularly in biochemical networks in which the fastest motions generally correspond to the chemical equilibria between metabolites , and the slower motions reflect more physiologically relevant transformations . Appreciation of this observation can result in elucidating structure from the network and simplifying the interactions . The reduction in dynamic dimensionality is based on such pooling and the analysis of pooling is focused in the underlying time scale hierarchy and its determinants . Understanding the time scale hierarchy and pooling structure of these networks is critical to understanding network behavior and simplifying it down to the core interactions . Top-down studies of dynamic characteristics of networks begin with fully developed kinetic models that are formal representations of large amounts of data about the chemistry and kinetics component interactions . Network properties can be studied by numerical simulations ( that are condition-specific ) or by analysis ( that often yield general model properties ) of the model equations . Since comprehensive numerical simulation studies become intractable for larger networks and the identification of general model properties are needed for the judicious simplification of models , there is a need for analysis based methods in order to characterize properties of dynamic networks . In this study we present an in silico analysis method to determine pooling of variables in complex dynamic models of biochemical reaction networks . This method is used to study metabolic network models and allows us to identify and analyze pool formation resulting from the underlying stoichiometric , thermodynamic , and kinetic properties .
The models studied here exhibit a significant span of time scales ( Table 1 ) . A hierarchy pool formation on different time scales was found in all networks based on the calculation of all pair wise ϑij ( k ) in the models ( Figures 1C and 2 ) . The results can be presented in a symmetric correlation tiled array , where each entry can be used to represent k for a pair of concentrations . Figure 3 shows the result of such an array for the human red cell . Since the array is symmetric we can display both k and the modal coefficient ratio in the pool ( xi/xj ) for each pair of concentrations; thus The time scale ( k ) for the formation of pools and the ratio between a pair of concentrations are functions of three factors: network stoichiometry ( or topology ) , thermodynamics , and kinetic properties of the transformations in the network . Viewing the dynamics of the network in terms of the modal matrix and the pair-wise concentration correlations on progressing time scales enables one to consider the questions of ( A ) the thermodynamic versus kinetic control of concentrations within the whole network and ( B ) the delineation of kinetic versus topological decoupling in networks .
The method developed above was developed , tested , and implemented in Mathematica ( Wolfram Research , Chicago , IL ) version 5 . 2 . The models analyzed herein: the model of human red cell metabolism [20]–[22] , human folate metabolism [23] , and yeast glycolysis [24] were implemented in Mathematica . For each model , a stable steady state was identified by integrating the equations over time until the concentration variables no longer changed ( error <1×10−10 , see Table S1 ) . The Jacobian was then calculated symbolically at that steady state condition . Temporal decomposition was carried out as described in the Results/Discussion section . Briefly for a general case , a similarity transformation [8] of a square matrix , A , is given by A = DΛD−1 in which D is invertible ( by definition ) and Λ is a diagonal matrix . D is an orthogonal matrix composed of eigenvectors corresponding to the entries of Λ ( the eigenvalues ) . When the Jacobian matrix for a first order differential equation with respect to time is decomposed in this manner , the negative reciprocals of the eigenvalues correspond to the characteristic time scales for the corresponding modes [8] ( this is immediately clear upon integration of Equation 4 ) . All three of the models considered here exhibited at least one pair of complex conjugate eigenvalues at the steady states considered , hence the corresponding complex conjugate modes were combined in order to eliminate oscillating motions . The calculations for the correlations across progressive time scales were carried out as described in Results/Discussion . Once the modal matrix , M−1 , was calculated , all pairwise angles between the metabolites ( columns of the modal matrix ) were calculated ( see Equation 5 ) . The modal matrix is rank ordered from the fastest ( k = 1 ) to the slowest ( k = n ) modes . The angles between the columns of the modal matrix were recalculated n−1 more times , in which an additional row of the modal matrix is zeroed out at each iteration . For example at the third iteration ( k = 2 ) , the first two rows of the modal matrix have been zeroed out . The spectrum of correlation cut-off values for pooling were considered from 10% to 99% . Cut-off values in the range 85% to 95% resulted in pooling of variables most consistent with the known pooling structures of the human red cell [2] , [5] . A value of 90% was used as the correlation cutoff for the red cell , folate , and yeast glycolysis models . The angle between two zero vectors was classified as undefined and the angle between any zero vector and another vector with at least one non-zero element was defined as 90° . Fragmentation of the pooling structure , in the strictest sense , was identified by any 0 entry ( or <∼10−13 ) in the final row of the metabolite modal matrix . Values for the Gibbs standard free energies of formation for the metabolites in the human red cell model were used from [25] .
|
Cellular metabolism describes the complex web of biochemical transformations that are necessary to build the structural components , to convert nutrients into “usable energy” by the cell , and to degrade or excrete the by-products . A critical aspect toward understanding metabolism is the set of dynamic interactions between metabolites , some of which occur very quickly while others occur more slowly . To develop a “systems” understanding of how networks operate dynamically we need to identify the different processes that occur on different time scales . When one moves from very fast time scales to slower ones , certain components in the network move in concert and pool together . We develop a method to elucidate the time scale hierarchy of a network and to simplify its structure by identifying these pools . This is applied to dynamic models of metabolism for the human red blood cell , human folate metabolism , and yeast glycolysis . It was possible to simplify the structure of these networks into biologically meaningful groups of variables . Because dynamics play important roles in normal and abnormal function in biology , it is expected that this work will contribute to an area of great relevance for human disease and engineering applications .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"mathematics",
"biochemistry/chemical",
"biology",
"of",
"the",
"cell",
"biochemistry/bioinformatics",
"computational",
"biology/metabolic",
"networks",
"biotechnology/bioengineering",
"biochemistry/theory",
"and",
"simulation"
] |
2008
|
Top-Down Analysis of Temporal Hierarchy in Biochemical Reaction Networks
|
Endogenous small interfering RNAs ( siRNAs ) are a class of naturally occuring regulatory RNAs found in fungi , plants , and animals . Some endogenous siRNAs are required to silence transposons or function in chromosome segregation; however , the specific roles of most endogenous siRNAs are unclear . The helicase gene eri-6/7 was identified in the nematode Caenorhabditis elegans by the enhanced response to exogenous double-stranded RNAs ( dsRNAs ) of the null mutant . eri-6/7 encodes a helicase homologous to small RNA factors Armitage in Drosophila , SDE3 in Arabidopsis , and Mov10 in humans . Here we show that eri-6/7 mutations cause the loss of 26-nucleotide ( nt ) endogenous siRNAs derived from genes and pseudogenes in oocytes and embryos , as well as deficiencies in somatic 22-nucleotide secondary siRNAs corresponding to the same loci . About 80 genes are eri-6/7 targets that generate the embryonic endogenous siRNAs that silence the corresponding mRNAs . These 80 genes share extensive nucleotide sequence homology and are poorly conserved , suggesting a role for these endogenous siRNAs in silencing of and thereby directing the fate of recently acquired , duplicated genes . Unlike most endogenous siRNAs in C . elegans , eri-6/7–dependent siRNAs require Dicer . We identify that the eri-6/7–dependent siRNAs have a passenger strand that is ∼19 nt and is inset by ∼3–4 nts from both ends of the 26 nt guide siRNA , suggesting non-canonical Dicer processing . Mutations in the Argonaute ERGO-1 , which associates with eri-6/7–dependent 26 nt siRNAs , cause passenger strand stabilization , indicating that ERGO-1 is required to separate the siRNA duplex , presumably through endonucleolytic cleavage of the passenger strand . Thus , like several other siRNA–associated Argonautes with a conserved RNaseH motif , ERGO-1 appears to be required for siRNA maturation .
Small RNA pathways regulate gene expression , chromatin structure and defense against invading elements such as transposons and viruses . Three general classes of small RNAs can be distinguished in animals: piwi-interacting RNAs ( piRNAs ) , microRNAs ( miRNAs ) and small interfering RNAs ( siRNAs ) . siRNAs exist in two classes: exogenous siRNAs that are derived from an exogenously administered dsRNA and endogenous siRNAs that are naturally generated within cells without administration of a dsRNA . Endogenous siRNAs were first discovered in plants and C . elegans , but have since been identified in flies and mouse oocytes [1]–[7] . In addition to regulating gene expression , endogenous siRNAs silence transposable elements and act in chromosome segregation [8]–[10] . Endogenous siRNAs in flies and mice are derived from dsRNA hairpin precursors , from dsRNA generated upon convergent transcription , or from antisense transcription of pseudogenes . What triggers endogenous siRNA formation in C . elegans is not as well understood but small RNA deep sequencing experiments have shown that about half of all genes produce endogenous siRNAs suggesting that this regulatory axis controls a wide range of gene activities [10]–[12] . Primary siRNA biogenesis in the exogenous RNAi pathway in C . elegans and many other organisms involves enzymatic cleavage by the RNAseIII enzyme Dicer of a longer dsRNA intermediate [13] , [14] , however , only a subset of endogenous siRNAs requires Dicer ( dcr-1 in C . elegans ) [12] . siRNAs are incorporated into effector complexes , comprised of an Argonaute protein and accessory factors , which direct silencing of complementary RNAs and in certain species , such as C . elegans , recruit RNA-dependent RNA polymerases ( RdRPs ) to the target resulting in siRNA amplification [15] , [16] . In most animals , there are two sub-families of Argonaute proteins , PIWI and Argonaute , that interact with specific classes of small RNAs . In C . elegans , there is an additional sub-family of Argonautes that is worm-specific and includes eighteen members . All small RNAs act sequence-specifically through base pairing with their target mRNA , but the outcome of the small RNA:target interaction can vary from suppression of transcription to mRNA degradation or translational repression and this is likely governed in part by the specific Argonaute cofactor . The C . elegans small RNA repertoire includes a large collection of endogenous siRNAs that can be classified by the specific Argonaute they associate with , the length of the small RNA , chemical modifications and the 5′ nucleotide . These include the CSR-1-associated 22G siRNAs ( 22 nt long with a 5′G ) [9] , [10] , WAGO-associated 22G siRNAs [12] as well as the ERGO-1-associated 26G siRNAs ( 26 nt long with a 5′G ) and ALG-3/4-associated 26G siRNAs [17]–[21] that act upstream of some WAGO-associated 22G siRNAs . Whereas CSR-1-associated siRNAs function in chromosome segregation during meiosis and mitosis , the specific functions of the other three classes of siRNAs are not as clear . Genetic , molecular and biochemical analyses have identified several genes and proteins involved in endogenous siRNA formation and function . The 26G siRNAs and the corresponding downstream 22G siRNAs , collectively called the ‘ERI’ class of siRNAs , all depend on a protein complex that includes the 3′-5′ exonuclease ERI-1 , the RdRP RRF-3 , the endonuclease DCR-1/ERI-4 , and the dsRNA binding protein RDE-4 [22]–[24] . A subset of ERI class endogenous siRNAs , found in oocytes and embryos , associates with the Argonaute ERGO-1 , whereas a sperm-specific class associates with the Argonautes ALG-3 and ALG-4 [18]–[20] . The biogenesis of the downstream , secondary 22G endogenous siRNAs may be mediated by the RdRPs RRF-1 and EGO-1 , in conjunction with the helicase DRH-3 [19]–[21] . The 22G siRNAs are incorporated into complexes with one or more of twelve partially redundant worm-specific Argonautes , the WAGOs , including NRDE-3 , an Argonaute that directs cotranscriptional gene silencing in the nucleus [25] , [26] . ERI-6/7 is a Superfamily I helicase homologous to Mov10 and Mov10-like1 in mice which also act in small RNA mediated gene silencing [27] , [28] . The eri-6/7 mRNA is expressed by trans-splicing of the pre-mRNAs of the eri-6 and eri-7 genes [29] . Like eri-1 , eri-6/7 was identified as a negative regulator of exogenous RNAi , i . e . mutants of eri-6/7 display an enhanced RNAi ( Eri ) phenotype upon exposure to exogenous dsRNA [29] , a phenotype also displayed by ergo-1 , eri-4 ( dcr-1 ) and rrf-3 mutants . To characterize the role of eri-6/7 in endogenous siRNA pathways , we compared the small RNA profiles of adult and embryo staged eri-6/7 mutants as well as embryo staged ergo-1 and eri-1 mutants to wild type C . elegans . Endogenous 26 nt and 22 nt siRNAs corresponding to about one hundred target genes were missing in eri-6/7 mutants whereas the thousands of other endogenous siRNAs were normally produced . The corresponding mRNA levels of these target genes tested were dramatically up-regulated in the eri-6/7 mutant , showing that the missing endogenous siRNAs mediate the silencing of these target genes . The eri-6/7 targets comprise mostly non-conserved genes and pseudogenes and fall into groups with extensive nucleotide sequence homology , indicative of gene duplications . The poor conservation of these genes suggests they may be newly acquired genes . Thus , the results suggest that one function of endogenous RNAi pathways is to silence one or more members of recently expanded gene families , possibly providing selective pressure for one paralog over others and accelerating divergence to avoid silencing . Like eri-1 and ergo-1 [18] , [19] , eri-6/7 is required for the formation or stability of 26G primary siRNA in embryos and 22G secondary siRNAs derived from 26G siRNA targets in adults . Thus , eri-6/7 , in collaboration with other ERI class genes , initiates an siRNA cascade of these recently duplicated genes in oocytes and embryos that persists throughout development . Surprisingly , although ERI class siRNAs are Dicer-dependent , the siRNA duplex precursor lacks the canonical features of a Dicer product . Instead of containing the canonical 2 nt 3′ overhangs on each siRNA strand , the 26G siRNA strand has a 3 nt 3′ overhang and an ∼4 nt 5′ overhang . In ergo-1 mutants , the levels of 26G siRNA passenger strands are elevated relative to wild type , suggesting that ERGO-1 mediates passenger strand removal through endonucleolytic cleavage , analogous to the function of other siRNA-associated Argonautes , such as RDE-1 acting in exogenous RNAi in worms and Ago2 in flies [16] , [30] .
The ERI-6/7 helicase was identified as a negative regulator of exogenous RNAi from a genetic screen for mutants that display enhanced RNAi efficacy for exogenously administered dsRNAs corresponding to particular C . elegans genes [29] . To understand any possible roles for ERI-6/7 in natural silencing pathways , we analyzed embryo and adult RNA for the presence of endogenous siRNAs by Northern blotting ( Figure 1A ) . We found that two endogenous siRNAs from the gene K02E2 . 6 are depleted in eri-6 ( mg379 ) embryos and adults . The longer species is relatively abundant in wild type embryos while absent in adults , as seen for the oocyte/embryo-specific siRNAs that associate with the Argonaute ERGO-1 [19] , whereas the shorter species is relatively more abundant in adults . These siRNAs are so-called 26G and 22G siRNAs that depend on ERGO-1 , the exonuclease ERI-1 and several other ERI factors [18]–[21] . Using quantitative RT-PCR we specifically assayed for two more oocyte/embryo-specific 26G siRNAs in embryos and also for two sperm-specific 26G siRNAs in young adult hermaphrodites . In embryos , the 26G oocyte-specific small RNAs ( ERGO-1 class ) were reduced in all eri mutants ( Figure 1B ) , whereas the sperm-specific 26G endo-siRNAs ( ALG-3/4 class ) were unaffected in eri-6/7 mutants and ergo-1 , but greatly reduced in eri-1 ( Figure 1C ) . To more comprehensively assess the requirement for eri-6/7 in endogenous RNA silencing pathways , small RNA cDNA amplicons were prepared from both embryo and adult staged eri-7 mutant and wild type C . elegans and subjected to deep sequencing . Additionally , small RNA libraries from embryo staged eri-6 , eri-1 and ergo-1 mutants were prepared and sequenced in parallel ( Table S1 ) . In adult eri-7 mutants the small RNA size and 5′ nt distribution was similar to that of wild type , although there was a modest reduction in 22G small RNAs ( Figure 2A ) . In embryos , 26G small RNAs were largely depleted in eri-7 and eri-6 , as well as in eri-1 and ergo-1 mutants ( Figure 2A , Figure S1A ) . Together with the quantitative RT-PCR data , these data are suggestive of a role for eri-6/7 in the ERGO-1 endogenous RNAi pathway that generates 26G siRNAs in embryos and 22G siRNAs in adults [19] . To determine if classes of small RNAs other than endogenous siRNAs depend on eri-6/7 , the numbers of small RNA reads , normalized to library size , corresponding to miRNA genes and piRNA loci , in wild type versus eri-7 , eri-6 , eri-1 and ergo-1 , were analyzed . Individual miRNAs were not substantially affected in eri-7 , eri-6 , eri-1 , or ergo-1 ( Figure 2B , Figure S1C ) . piRNAs ( also called 21U RNAs in C . elegans ) appeared slightly upregulated , especially in embryo libraries ( Figure 2B and 2C , Figure S1D ) . Possibly , the lack of 26G siRNAs in embryos has modest effects on embryonic development , which could affect the relative abundance of piRNAs that are known to be more abundant in younger embryos than older embryos [31] . Over half of the genes in C . elegans produce endogenous siRNAs through multiple different pathways [11] , [12] . We analyzed which loci produce eri-6/7-dependent siRNAs . First small RNA reads for each coding gene , pseudogene and transposon were plotted as a function of library size normalized reads in wild type versus eri-7 , eri-6 , eri-1 and ergo-1 mutants ( embryos ) and in wild type versus eri-7 ( adults ) ( Figure 2B , Figure S1B ) . Using an arbitrary read threshold of 10 reads per million total reads ( RPM ) for wild type small RNA libraries , ∼80 features , primarily annotated coding genes and pseudogenes , were depleted of siRNA reads by ≥67% in both eri-7 and eri-1 mutant embryos , relative to wild type ( Figure 2B , Figure S1B , Table S3 ) . A similar , albeit more modest , reduction in siRNA reads was observed in ergo-1 mutants ( Figure 2B , Table S3 ) . A partially overlapping set of features was also depleted of siRNAs in adult eri-7 mutants ( Figure 2B ) . To account for intergenic and other non-annotated loci , we plotted the ratios of small RNA reads in eri-7 , eri-6 , eri-1 and ergo-1 mutants to wild type in 5 kb windows along 1 kb increments across each chromosome ( Figure 2C , Figure S1D ) . Loci depleted of siRNAs in eri-6 , eri-7 , eri-1 , and ergo-1 mutants largely overlapped and tended to derive from the more gene-poor arms of the chromosomes , as was observed for ergo-1 by Vasale et al . [19] ( Figure 2C , Figure S1D ) . The majority of loci depleted of siRNAs correspond to annotated coding genes and pseudogenes and predominantly yield 26 nt small RNAs in wild type embryos ( Figure 2C , Figure S1D ) . A total of 1160 individual 26G siRNAs were identified that passed a read threshold of 1 RPM and were depleted by ≥67% in both eri-7 and eri-1 mutants , relative to wild type ( Table S2 ) . In total , 26G siRNAs were depleted by >99% in eri-6 , eri-7 and eri-1 mutant embryos and by ∼96% in ergo-1 mutant embryos , relative to wild type ( Figure 2D ) . In adults , the loci depleted of small RNAs in eri-7 mutants largely overlapped with those in embryos , but were predominantly of the 22 nt size class ( Figure 2C ) and therefore likely correspond to the secondary 22G siRNAs that are thought to be downstream of 26G siRNAs [19] , [21] . 22G siRNAs derived from ERGO-1 class 26G siRNA targets are biochemically indistinguishable those derived from ALG-3/4 class 26G siRNA targets . Although our qRT-PCR results suggested that eri-6/7 is not required for ALG-3/4 class 26G siRNA accumulation , we nonetheless assessed the requirement of eri-6/7 for 22G siRNAs derived from both ERGO-1 and ALG-3/4 class 26G siRNA targets using published ERGO-1 and ALG-3/4 target datasets [18]–[20] . Consistent with a requirement for eri-6/7 specifically in the ERGO-1 class 26G siRNA pathway , 22G siRNAs derived from ALG-3/4 class 26G siRNA targets were unaffected in eri-7 mutant adults , whereas 22G siRNAs corresponding to ERGO-1 class 26G siRNA targets were depleted by ∼90% ( Figure 2E ) . Similarly , 22G siRNA reads from individual ERGO-1 class 26G siRNA targets were largely depleted and 22G siRNA reads from individual ALG-3/4 class 26G siRNA targets were unaffected in eri-7 mutant adults ( Figure 2F–2G ) . ERGO-1 class 22G siRNAs are relatively more abundant in adults versus embryos , confirming the observations made by Northern blot analysis ( Figure 1A , Figure S3 ) . The majority of 22G siRNAs , including those that associate with the Argonaute CSR-1 to direct chromosome segregation , are not dependent on either class of 26G siRNAs . These non-26G-dependent 22G siRNAs were not depleted in eri-7 mutant adults , relative to wild type ( Figure S2 ) . Thus , eri-6/7 is specifically required for ERGO-1 class 26G siRNA formation in oocytes and embryos and , possibly indirectly , for accumulation of the secondary 22G siRNAs present in the adult . From our deep sequencing datasets we identified 78 and 75 annotated coding genes , pseudogenes , and transposons that yielded ≥10 RPM and were depleted by >67% in eri-7 mutant embryos and adults , respectively , relative to wild type ( Table S3 ) . Of these , 60 are depleted of siRNAs in both embryo and adult eri-7 mutants ( Figure 3A ) . The endogenous siRNAs derived from genes and pseudogenes did not always appear antisense to the predicted exons as annotated in Wormbase . However comparison to recent modENCODE data [32] shows that for these loci intron/exon structures are either incorrectly annotated or not annotated in Wormbase . Many eri-6/7-dependent endogenous siRNAs do not match a unique sequence in the genome , as initially noticed by Vasale et al . [19] for ergo-1 target genes . To analyze the extent of homology between the eri-6/7 target genes , the sequences of all annotated coding genes and pseudogenes , including 0 . 5 kb of flanking sequence on either side , were retrieved and aligned using the blastn and discontinuous megablast algorithms . Surprisingly , over two-thirds of the genes targeted by eri-6/7-dependent siRNAs share >82% total nt identity to one or more of the other 78 eri-6/7 target genes as well as stretches of ≥27 nt with 100% identity ( Figure 3B and 3C , Table S3 , Figure S4 ) . In contrast , within a set of randomly selected genes of the same combined sequence length , only 4% share this degree of sequence identity . Whereas some target genes represent pairs of homologous genes ( Figure 3E ) , many genes form large clusters that share sequence homology ( Figure 3B and 3C ) , although within these clusters , not all genes show homology to the same sequence ( Figure 3D ) . Fourteen genes not only share sequence homology but are also in close proximity ( within 5 kb ) of each other ( Table S3 , Figure S4 ) . Some of these genes are adjacent to each other and are likely the result of recent gene duplications [33] ( e . g . W04B5 . 6 and E02H9 . 6 ) . However , seven genes are in close proximity but not homologous , suggesting that the endogenous siRNA biogenesis machinery has a preference for certain genomic regions , possibly determined by chromatin state . There are common features that may route these genes into the eri-6/7-dependent endogenous siRNA pathway . The protein sequences of target genes with eri-6/7-dependent siRNAs are poorly conserved . More than half of the eri-6/7 target genes are not conserved between C . elegans and the related nematode C . briggsae , which is about five times the genome average of 11% of genes that are not shared between the two species [34] . The 34 eri-6/7 target genes with a detectable homolog in C . briggsae rank among the least conserved between the two species; these genes had a much higher median E-value , 5 . 9E-07 as compared to a median of 5 . 8E-114 for all C . elegans gene products with a C . briggsae homolog ( Table S4 ) . The gene structure of the eri-6/7-dependent siRNA target genes also deviates from the typical C . elegans gene; the genes are shorter ( 1 . 3 kb average versus 2 . 8 kb ) than the average protein coding gene in C . elegans and contain fewer exons ( median of 3 versus 5 for all C . elegans genes ) ( Table S5 ) . The lack of conservation even in other nematodes , and lack of other indications of function for the majority of these genes suggest that they were recently acquired by C . elegans and that some of these genes may not produce functional proteins , i . e . they are pseudogenes . Indeed , in an analysis of a large scale proteomics dataset [35] we identified peptides from only 16% of eri-6/7 target genes , compared to 54% for all annotated coding genes ( Table S6 ) . The fact that many endogenous siRNAs match gene sequences repeated in the genome , suggests that the eri-6/7 pathway targets duplicated genome segments . Gene duplications occur at a relatively high frequency in C . elegans [36] , and tend to produce partial or chimeric gene duplicates [33] . Sixteen eri-6/7 target genes are among the 516 most recently duplicated genes present in the whole genome based on low synonymous substitution numbers ( Ks<0 . 1 ) between duplicated genes [36] , a seven-fold enrichment over random expectation . The physiological role of the eri-6/7 pathway remains undefined . Whereas eri-1 mutants display temperature-sensitive sterility due to defects in sperm morphology related to the ALG-3/4 endogenous siRNA pathway , eri-6/7 and ergo-1 mutants show no obvious phenotypes . However , brood size analysis at elevated temperatures indicates that down-regulation of eri-6/7 target genes may be required for optimal fecundity at higher temperatures ( Table S7 ) . Many of the eri-6/7 target genes lack indication of function; of 50 eri-6/7 target genes tested by the C . elegans community for gene inactivation phenotypes ( as annotated in Wormbase ) , only three produce RNAi-induced phenotypes ( 6% ) , whereas from the whole genome , 5 , 852 out of 20 , 808 ( 28% ) gene inactivations induce phenotypes . However , several of the target genes are homologous to nucleic acid modifying enzymes , such as the gene F55C9 . 3 ( PAZ domain ) , the genes F39E9 . 7 and F39E9 . 10 ( dsRNA binding domain ) as well as several predicted helicase genes . Possibly , upregulation of these target genes regulated by eri-6/7 contributes to the enhanced RNAi phenotype of the eri-6/7 mutants . eri-6/7-dependent siRNAs exist as two classes: the primary 26G siRNAs found in oocytes and embryos and secondary 22G siRNAs that predominate in post-embryonic stages of development . By analysis of genes shown to be enriched for siRNAs in the soma or germline [12] we determined that the eri-6/7-dependent secondary endogenous siRNAs are mostly soma-enriched ( Figure 4A ) . Four Argonaute proteins are known to act somatically [26] , including SAGO-1 , SAGO-2 and NRDE-3 , and could thus potentially interact with eri-6/7-dependent 22G siRNAs . By Northern blot analysis , an ergo-1-dependent 22G siRNA was found to be dependent on sago-1 and sago-2 [19] . NRDE-3 , an Argonaute protein that acts in co-transcriptional silencing of endogenous siRNA targets in the nucleus [25] , [38] , associates with eri-1-dependent 20–22 nt siRNAs that are required for NRDE-3 localization to the nucleus . nrde-3 is also partially required for an exogenous RNAi response: the enhanced RNAi phenotype of eri-1 mutants in response to dsRNA triggers that target nuclear RNAs is dependent on nrde-3 , but non-nuclear RNAi responses in eri-1 mutants are independent of nrde-3 [25] . To examine if eri-6/7 acts in the NRDE-3 pathway , double mutants of eri-6/7 with nrde-3 were assayed for enhanced RNAi phenotypes ( Eri ) and transgene silencing phenotypes . nrde-3 is required for the enhanced RNAi phenotype of eri-6/7 mutants in response to dpy-13 ( RNAi ) , similar to eri-1 mutants ( Table S8 ) . Like other eri mutants such as rrf-3 [39] , eri-6/7 mutants display silencing of repetitive transgenes ( Table S8 ) . nrde-3 mutants display weak silencing of repetitive transgenes , whereas nrde-3;eri-6 double mutants do not display transgene silencing ( Table S8 ) . These data suggest the existence of multiple , possibly competing , transgene silencing pathways regulated by nrde-3 and eri-6/7 . The nuclear function of eri-6/7-dependent siRNAs was further examined by NRDE-3 localization analysis in eri-6/7 mutants at post-embryonic stages . Like in eri-1 mutants , NRDE-3 fails to localize to the nucleus in eri-6/7 mutants , suggesting that a subset or all 22G endogenous siRNAs dependent on eri-6/7 direct NRDE-3 to the nucleus to mediate co-transcriptional gene silencing ( Figure 4B ) . Indeed , analysis of the NRDE-3-associated siRNAs [25] indicates that these siRNAs are depleted to 1% or less of the wild type values in eri-6/7 mutants ( Figure 4C ) . Thus , NRDE-3 requires eri-6/7-dependent siRNAs for localization to the nucleus where it mediates co-transcriptional gene silencing [25] . To assess the effects of loss of eri-6/7 on target mRNA levels , we performed quantitative RT-PCR on several eri-6/7 target genes , including ZK380 . 5 and T08B6 . 2 . We saw 10–25 fold increases in target mRNA levels in eri-6 mutant young adult samples as compared to wild type ( Figure 4D , Figure S7 ) similar to what has been seen in other eri pathway mutants [18] , [19] , [21] , [23] . mRNA levels of eri-6/7 target genes are also increased in the nrde-3 mutant as compared to wild type ( Figure S7 and [25] ) . In conclusion , eri-6/7 mutants are depleted of NRDE-3-associated siRNAs and show a disruption of NRDE-3 localization to the nucleus . In post-embryonic stages , eri-6/7-dependent siRNAs silence their targets leading to reduced mRNA levels , likely through co-transcriptional silencing via the NRDE pathway in the nucleus , although additional modes of silencing may exist . In oocytes and embryos , eri-6/7-dependent primary ( 26G ) siRNA biogenesis likely contributes to target silencing by Dicer-mediated cleavage of the target mRNA ( as discussed below ) . The mechanism of 26G siRNA formation is not well understood . In C . elegans , exogenous siRNAs are processed from a long dsRNA via processive Dicer activity generating ∼23 nt siRNAs starting at one end of the dsRNA [14] . Similarly , in plants , ( endogenous ) trans-acting siRNAs ( tasiRNAs ) are processed by Dicer in sequential 21 nt increments in phase with a miRNA-guided cleavage site [40] . We observed that 26G siRNAs are also produced in regular but variable intervals , typically ranging between 23–29 nt ( Figure 5A , Figure S5 ) , with an siRNA length of invariably 26 nucleotides . The pattern of 26G siRNA distribution does not point to a specific initiation site , except for a preference for the 5′ end to start opposite a cytosine on the complementary mRNA , nor is it consistent across a given transcript . It is likely that the regular patterning of 26G siRNAs results from a combination of a processive RdRP activity and endonucleolytic activity by Dicer ( DCR-1 ) . In support of this , DCR-1 is found in a complex with the RdRP RRF-3 , and both these factors are required for 26G siRNA formation [22]–[24] , [41] . During exogenous siRNA formation , the functional guide strand of an siRNA duplex intermediate is liberated from a passenger strand [16] , [30] , [42]–[44] . Given the requirement for DCR-1 in 26G endogenous siRNA formation , we predicted that 26G siRNAs are also liberated from a duplex intermediate . To identify potential 26G siRNA passenger strands , we searched our small RNA libraries for sequences that at least partially overlapped and were antisense to each 26G siRNA . For 1100 of the 1160 26G siRNAs examined , we were able to identify a sequence that met our criteria . Canonical Dicer products are 21–24 nt long and have 2 nt overhangs at each 3′ end of the small RNA duplex , however , the most dominant sequence antisense to each 26G siRNA was ∼19 nt long and inset by 3 nt from the 3′ end and ∼2–4 nt from the 5′ end of the 26G siRNA ( Figure 5B , 5C , Table S9 ) . The ratio of passenger strand reads to corresponding 26G siRNA reads after applying a 1 RPM threshold ranged from 0 to 6 . 4 with a median ratio of ∼0 . 05 ( Table S8 ) . Similar to the 26G siRNA strand , the passenger strand was depleted in eri-7 , eri-6 and eri-1 mutants , relative to wild type , suggesting that these factors are upstream of 26G siRNA duplex formation ( Figure 5D ) . In contrast , in ergo-1 mutants , the passenger strand was elevated by ∼2 fold , relative to wild type ( Figure 5B and 5D ) . The observation that the passenger strand is stabilized in ergo-1 mutants suggests that , similar to RDE-1 in worms and AGO2 in flies , ERGO-1 cleaves the passenger strand to liberate it from the 26G siRNA [16] . Indeed , ERGO-1 is one of only a few C . elegans Argonautes that contains the conserved RNaseH residues required for slicer activity . From a genetic screen for enhanced exogenous RNAi mutants [29] , [45] , we isolated an ergo-1 mutant , ergo-1 ( mg394 ) , that has a amino acid substitution at position 1072 in the presumable catalytic pocket defined by the D ( 852 ) D ( 930 ) H ( 1070 ) amino acids required for slicer function . This mutant has a enhanced RNAi phenotype indistinguishable from the presumable null mutant , the deletion mutant ergo-1 ( tm1860 ) . This suggests that slicer function is required for wild type ERGO-1 activity . The features of 26G siRNAs and their passenger strands suggest a novel non-canonical mechanism for Dicer activity on 26G siRNA precursors , possibly facilitated by another ribonuclease in the pathway , ERI-1 .
The ERI-6/7 helicase is a negative regulator of exogenous RNAi and as we have shown here , is required for a particular suite of endogenous siRNAs in what is now emerging as a multidimensional set of endogenous RNAi pathways . ERI-6/7 is , like the Argonaute ERGO-1 , required for the generation and/or stability of two classes of siRNAs , oocyte- and embryo-specific 26G siRNAs and the later generated somatic secondary 22G siRNAs corresponding to the same loci . These 22G siRNAs reduce target mRNA levels , similar to secondary siRNAs in the exogenous RNAi pathway . Our analysis of the ERGO-1/ERI-6/7 pathway has two major surprises: first , that this pathway targets a relatively small number of loci in the genome , a set of duplicated genes with extensive nucleic acid homology . This points to a dedicated surveillance pathway for such gene duplications . Because so many of the components of the ERGO-1/ERI-6/7 pathway are conserved across phylogeny , this duplicated gene silencing pathway is likely to be general . Secondly , the detailed deep sequencing analysis of eri-6/7-dependent small RNA revealed the presence of a passenger strand and ( imperfect ) phasing between 26G siRNAs , adding to our understanding of the mechanism of endogenous siRNA biogenesis and the role of the Argonaute ERGO-1 . The set of eri-6/7 target genes revealed by our deep sequencing analysis consists of pairs or larger groups of genes that share extensive DNA sequence homology , have a small number of introns and are poorly conserved , even in other Caenorhabditae . The poor conservation and few introns support the model that these genes have recently been acquired by C . elegans , perhaps via horizontal gene transfer , for example , from viruses . RNAi pathways have been implicated in antiviral defense , and the ERGO-1/ERI-6/7 pathway may constitute elements of such a viral surveillance pathway . Our data suggest that viral surveillance extends beyond the initial infection . Newly acquired viral genomes may tend to integrate at multiple loci so that extensive nucleotide sequence homology between disparate loci may be a signature of such genes and continue to be silenced by the eri-6/7 small RNA pathway . Alternatively , these duplicated genes may be novel DNA transposons . We did not find evidence of target site duplication at the boundaries of the homologous sequences , nor did we find terminal inverted repeats . If these are transposons , they may no longer are be active , even in eri mutants; although the level of sequence identity is high , it is not as high as of active DNA transposons in C . elegans [46] . Also , we have not found evidence of mutator activity in eri mutants . The targeting of these genes with extensive nucleotide homology suggests that the ERI-6/7 helicase , the ERGO-1 Argonaute protein and other ERI proteins must specifically generate or load siRNAs from the duplicated segments of such gene pairs . To achieve this specificity for duplicated genes , transcripts of every gene may be compared to every other gene and extensive but not perfect nucleotide homology over a distance of hundreds of nucleotides may be assessed by this system . Such a system would need to detect the distinct genome location or the small level of nucleotide sequence divergence that would distinguish surveillance of another transcript from the same gene from a transcript emerging from a duplicated distinct gene . The most obvious phenotype of the eri-6/7 or ergo-1 null mutations is enhanced response to exogenous RNAi . Our data shows that eri-6/7 mutants have a reduced brood size . The reduced fitness could be attributed to over-expression of specific target genes that act in specific pathways or by more systemic effects induced by a general accumulation of unwanted RNAs . The lack of functional annotation for the majority of eri-6/7 target genes suggests some of these genes may not be functional , though about half of the target genes are weakly conserved in C . briggsae . Thus , the surveillance of these duplicated genes does not appear to subserve any key function for development or survival in the lab . However , this surveillance program uses sufficient C . elegans small RNA machinery that when the eri-6/7 system is disabled , the ability of the animal to respond to exogenous double stranded RNA is enhanced . The small RNA demands of this pathway point to its importance to the organism . Our data provides insight into the structure of the double-stranded intermediate in 26G siRNA generation by the identification of the passenger strand and the first large scale sequence analysis of the passenger strand in a slicer-defective Argonaute mutant . An obvious candidate for duplex siRNA generation is Dicer/DCR-1 , which was shown to be required for 26G siRNA biogenesis [18] , [19] , [23] , [24]; A helicase domain mutation ( dcr-1/eri-4 ( mg375 ) ) in DCR-1 specifically abolishes endogenous siRNAs but not microRNAs [21] , [23] , [24] , indicating that the requirement for Dicer is unlikely an indirect effect via a microRNA target gene that acts in 26G siRNA biogenesis . However , the 26G siRNA duplexes are not canonical Dicer products in terms of the lengths of the antisense and passenger strands , the 5′ overhang , and the strong bias for a G as the 5′ nucleotide . Our observation of variable phasing between the 5′ ends of the 26G siRNAs within genes , suggest there is a processive activity that generates 26G siRNAs . This is most likely the RdRP RRF-3 , preferentially using a guanylate as an initiation nucleotide , that in conjuction with an endonuclease , possibly Dicer , generates a 26G siRNA duplex and continues doing so along the mRNA starting at a neighboring cytosine in the mRNA template [17] . The structure of the duplex suggests that it is modified by other enzymes , such as the ERI-1 3′ exonuclease , to produce the 19 nucleotide passenger strand ( Figure S6 ) . Our eri-1 mutant deep sequencing data did not provide evidence for the existence of longer passenger strand precursors; Possibly such precursors are not stable . eri-6/7 acts in the same pathway as the Argonaute ERGO-1 but unlike eri-6/7 , ergo-1 is not required for the passenger strand opposite of the 26G siRNA . This suggests that eri-6/7 acts , with eri-1 and other eri genes , in the production of a 26G siRNA duplex precursor , while ergo-1 acts on the duplex after biogenesis , removing the passenger strand possibly by slicing , similar to the function of another slicing-capable Argonaute RDE-1 in exogenous RNAi in C . elegans , and similar to the roles of Argonautes in flies and mammals [16] , [30] , [42] , [43] . Site-directed mutagenesis experiments of the catalytic amino acids DDH are necessary to provide more direct evidence of slicing versus other ways of passenger strand destabilization by ERGO-1 . The role of ERGO-1 in passenger strand removal versus siRNA biogenesis could also explain the weaker reduction in siRNAs seen in ergo-1 ( tm1860 ) mutants versus eri-1 and eri-6/7 mutants . Alternative explanations are that the ergo-1 ( tm1860 ) allele is a partial loss-of-function , although the deletion removes more than one third of the PAZ domain , or that other Argonautes are partially redundant with ergo-1 . The molecular function of the ERI-6/7 helicase is unclear . The homologous protein Mov10 in humans associates with Argonaute [47] and the fly homolog Armitage is required for RNA induced silencing complex ( RISC ) formation [48] . Thus it is possible that ERI-6/7 interacts with ERGO-1 and functions in the assembly of an active effector complex . eri-6/7 does not act in the sperm-specific 26G siRNA pathway that involves the Argonautes ALG-3/4 in place of ERGO-1; it will be of interest to determine if another helicase functions in 26G endogenous siRNA generation in this pathway . Vasale et al . and Gent et al . [19] , [21] have proposed a two-step model for siRNA generation in the ERGO-1 pathway . Downstream of 26G siRNAs , 22G siRNAs are generated by RNA-dependent RNA polymerases RRF-1 and EGO-1 . Our data suggests that these events are actually spaced in time , with 26G siRNA generation first in the developing embryo and subsequent RdRP-mediated 22G siRNA generation occurring post-embryonically . mRNA levels of eri-6/7 target genes are down-regulated in wild type worms compared to eri mutants . This is explained in part by routing of the endogenous siRNAs into a nuclear co-transcriptional silencing pathway that involves the Argonaute NRDE-3 . eri-6/7-dependent siRNAs are also likely to associate with other Argonautes , such as SAGO-1 and SAGO-2 , since at least two eri-6/7-dependent endogenous siRNAs , assayed by Northern blotting , were shown to associate with SAGO-1 and -2 [26] . How these Argonautes affect target gene expression is unknown , but the lack the catalytic residues required for slicing , suggests that they direct mRNA degradation by some means other than slicing or that they inhibit translation . In eri mutants , that are defective in some endogenous RNAi pathways , the exogenous RNAi pathway is more active . The opposing functions of eri-6/7 ( and other eri genes ) in exogenous RNAi and endogenous RNAi have been explained by a competition model in which the exogenous RNAi pathway competes with the endogenous RNAi pathway for limiting factors [22] , [49] . The alg-3/-4 double mutant does not show an enhanced RNAi phenotype [18] , suggesting that the limiting factors that the exogenous RNAi pathway competes for are only part of the embryo- and soma-specific ERGO-1/ERI-6/7 endogenous siRNA pathway . The observation that overexpression of the SAGO-1 and -2 proteins that interact with secondary siRNAs causes an Eri phenotype and an enhanced accumulation of endogenous siRNAs [26] , shows that these SAGOs could be the limiting factors in the exogenous RNAi pathway . It remains possible that one or more target genes regulated by the eri genes also act in RNAi pathways . Several target genes encode proteins with potential RNA modifying capability , such as a few helicases , dsRNA binding proteins and a PAZ domain protein . Whereas in mouse oocytes endogenous siRNAs are formed from antisense pseudogene transcripts , C . elegans has RNA-dependent RNA polymerases , in this case possibly RRF-3 , that can produce antisense transcripts . How RRF-3 is recruited to target mRNAs to generate antisense siRNAs is unknown . Possibly , short double stranded RNAs of mRNAs base pairing with antisense transcripts generated from homologous genes recruit the RNA-dependent RNA polymerase . Only a few eri-6/7 target genes have been annotated as pseudogenes , but it remains possible that among the annotated coding genes targeted in eri-6/7 mutants are also pseudogenes . Thus , similar to the function of some pseudogenes in mouse oocytes , pseudogenes in C . elegans may play an important role in endogenous RNAi .
All experiments were performed at 20°C unless stated otherwise . For deep sequencing analysis , wild type N2 , eri-6 ( mg379 ) ( 6× backcrossed ) , eri-7 ( mg369 ) ( 4× backcrossed ) , eri-1 ( mg366 ) ( 5× backcrossed ) and ergo-1 ( tm1860 ) ( 5× backcrossed ) were used . YY174 ( NRDE-3p::3xFLAG::GFP::NRDE-3 ) and YY178 ( eri-1 ( mg366 ) ;NRDE-3p::3xFLAG::GFP::NRDE-3 ) [25] were used in GFP::NRDE-3 sub-cellular localization analyses and crosses to eri-6 ( mg379 ) . Enhanced RNAi assays and transgene silencing were performed as described previously [29] . WM156 ( nrde-3 ( tm1115 ) ) was used in enhanced RNAi assays , transgene silencing assays and qRT-PCRs . NRDE-3::GFP localization was determined by Normarski and fluorescence imaging . GFP images were taken at 40× with a 100 ms exposure; DIC images were exposed for 59 ms . Total RNA was isolated by dounce homogenization in RNA-Bee ( Tel-Test ) followed by chloroform extraction and isopropanol precipitation . 30 µg of total RNA was run on a 15% polyacrylamide/urea gel , blotted and probed with a Starfire probe detecting an endogenous siRNA corresponding to gene K02E2 . 6 [22] . For qRT-PCR , total RNA was DNase treated using the TURBO DNA-free kit ( Applied Biosystems ) . cDNA was synthesized using RETROscript ( Applied Biosystems ) following the vendor's protocol . qPCR was done with CFX96 machine ( Bio-Rad ) using iQ SYBR Green Supermix ( Bio-Rad ) . Relative mRNA levels were calculated based on the 2−ΔΔct method using the gene Y45F10D . 4 for normalization [50] . Three technical replicates were done for each PCR . Primer sequences are listed in Table S10 . Taqman qRT-PCR and data analysis was carried out as described [18] . For ERGO-1-class and ALG-3/4-class 26G siRNAs , total RNA was extracted from embryos and 52–56 hr post hatching L4/young adult worms , respectively . Total RNA was isolated from worms grown at 20°C for ∼66–72 hrs post synchronization at the L1 stage and harvested as day one gravid adults , or from embryos . 18–28 nt small RNAs were size selected , tobacco acid pyrophosphatase treated to remove 5′ di- and triphosphate groups , ligated to 5′ and 3′ adapters and subjected to RT-PCR , according to the protocol by Gu et al . [12] . Small RNA amplicons were sequenced using an Illumina Genome Analyzer . Data analysis was done as described [11] . Briefly , sequences were parsed and mapped to the C . elegans genome ( Wormbase release WS203 ) using CASHX ( version 2 . 0 ) [51] . Small RNA reads from each library were normalized to the total number of mapped reads . The numbers of small RNA reads for small RNA sequences mapping to multiple positions in the genome were divided by the corresponding number of genomic loci . Small RNA reads were classified by genomic feature according to Wormbase annotations ( WS203 ) . Genome plots were constructed by plotting total small RNA read counts from 5 kb windows along 1 kb increments of each chromosome . siRNAs were classified using published datasets [11] , [12] , [18]–[20] . NRDE-3-associated siRNAs were identified by Guang et al . [25] . The passenger strand was identified by taking the genomic coordinates of each 26G siRNA , adding 14 nucleotides of flanking sequence at each end and extracting sequences antisense to and within this 54 bp sequence from our wild type embryo small RNA library . Based on results from our analyses , we defined the 26G siRNA passenger strand as the most abundant antisense siRNA inset by 3 nt from the 3′ end of the 26G siRNA ( Table S9 ) . Target gene sequences including 0 . 5 kb of 5′ and 3′ flanking sequence were obtained from Wormbase . Subsequently , these sequences were analyzed by discontiguous megablast and blastn for sequence identity . The best protein homologs in C . briggsae and H . sapiens for all C . elegans gene products were identified using Wormbase , with an E-value of 0 . 1 as a cut-off . Target genes that for which the intron/exon predictions are consistent between Wormbase and modENCODE data were analyzed for number of exons and gene length and compared to all genes in the genome . Peptides produced from eri-6/7 target genes were mined from Schrimpf et al . [35] . Small RNA Illumina deep sequencing data are available at the Gene Expression Omnibus ( GEO ) database ( accession no . GSE32366 ) .
|
Endogenous small interfering RNAs ( siRNAs ) are a class of small RNAs present in fungi , plants , and animals . Small RNAs , including microRNAs , are known to regulate the expression levels of genes , silence invading elements such as transposons , and act in cell division . However , the function of many endogenous siRNAs is unknown . We have found that the ERI-6/7 helicase is required for a subset of endogenous siRNAs present in the nematode Caenorhabditis elegans . The ERI-6/7 helicase acts in a pathway together with the Argonaute protein ERGO-1 to produce two types of siRNAs: a primary class of 26 nucleotides in length present in oocytes and embryos , and a class of 22 nucleotide siRNAs present in later stages of development . These siRNAs correspond to only about one hundred genes . Interestingly , we found that these genes fall into groups of genes that contain nearly identical DNA sequences . The sequences of these genes are not conserved in other organisms , not even in related nematodes . These results point to a potential function of these endogenous siRNAs: silencing of recently acquired , duplicated genes . Our work demonstrates a new role of small RNAs , different from known functions in transposon silencing and regulation of gene expression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"rna",
"interference",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"epigenetics",
"gene",
"expression",
"biology",
"molecular",
"biology",
"systems",
"biology",
"rna",
"rna",
"processing",
"nucleic",
"acids",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2011
|
The ERI-6/7 Helicase Acts at the First Stage of an siRNA Amplification Pathway That Targets Recent Gene Duplications
|
NURF is a conserved higher eukaryotic ISWI-containing chromatin remodeling complex that catalyzes ATP-dependent nucleosome sliding . By sliding nucleosomes , NURF is able to alter chromatin dynamics to control transcription and genome organization . Previous biochemical and genetic analysis of the specificity-subunit of Drosophila NURF ( Nurf301/Enhancer of Bithorax ( E ( bx ) ) has defined NURF as a critical regulator of homeotic , heat-shock and steroid-responsive gene transcription . It has been speculated that NURF controls pathway specific transcription by co-operating with sequence-specific transcription factors to remodel chromatin at dedicated enhancers . However , conclusive in vivo demonstration of this is lacking and precise regulatory elements targeted by NURF are poorly defined . To address this , we have generated a comprehensive map of in vivo NURF activity , using MNase-sequencing to determine at base pair resolution NURF target nucleosomes , and ChIP-sequencing to define sites of NURF recruitment . Our data show that , besides anticipated roles at enhancers , NURF interacts physically and functionally with the TRF2/DREF basal transcription factor to organize nucleosomes downstream of active promoters . Moreover , we detect NURF remodeling and recruitment at distal insulator sites , where NURF functionally interacts with and co-localizes with DREF and insulator proteins including CP190 to establish nucleosome-depleted domains . This insulator function of NURF is most apparent at subclasses of insulators that mark the boundaries of chromatin domains , where multiple insulator proteins co-associate . By visualizing the complete repertoire of in vivo NURF chromatin targets , our data provide new insights into how chromatin remodeling can control genome organization and regulatory interactions .
The organization of DNA in nucleosomes has a major function in controlling accessibility of DNA to the protein complexes that process genetic information . By altering nucleosome dynamics , targets for the transcription , replication and repair machineries can be rendered inaccessible or made available . A number of mechanisms exist by which chromatin states can be altered . Post-translational modification of the histone tails ( HPTMs ) can change associations between histones and DNA , altering chromatin flexibility and conformation ( reviewed in Tessarz and Kouzarides [1] ) . However , these modifications can also act as marks that can be bound by effector complexes that include ATP-dependent chromatin remodeling factors ( reviewed in Swygert and Peterson [2] ) . These multi-subunit protein complexes utilize the energy of ATP hydrolysis to alter nucleosome dynamics . They can be divided into broad families based on the core catalytic subunit and effects on nucleosomes—eviction , sliding or variant histone replacement . The imitation switch ( ISWI ) family of ATP-dependent chromatin remodeling factors mediate energy-dependent nucleosome sliding [3 , 4] . The nucleosome remodeling factor ( NURF ) is one of the founding members of this family . Although chromatin remodeling complexes based on ISWI type catalytic subunits are present in all metazoa , NURF is an innovation of the bilateria . NURF complexes are built around a large , bilaterian-conserved , NURF-specific subunit , in Drosophila Nurf301/Enhancer of bithorax ( E ( bx ) ) , in humans BPTF ( Bromodomain and PHD finger Transcription Factor ) [5 , 6] . Like other ISWI-containing complexes NURF catalyzes nucleosome sliding [5 , 6] , allowing access to transcription factor ( TF ) binding sites to be regulated and transcription controlled . Consistent with this , mutations in Nurf301/E ( bx ) were initially identified as regulators of the bithorax-complex [7] , and subsequently shown to lead to altered transcription regulation of signal cascades including the ecdysone , heat-shock responsive and JAK/STAT pathways [7–9] . Current models for NURF function propose activity at defined enhancers leading to regulation of a restricted set of gene targets . We and others have shown that Nurf301/E ( bx ) can directly interact with sequence-specific TFs that include the GAGA factor ( Trithorax-like ( Trl ) ) , ecdysone receptor ( EcR ) , and the repressor Ken [6 , 8 , 9] , suggesting that NURF remodeling is targeted to enhancer elements by direct interactions with TFs . However , transcriptional regulation in metazoa requires complex interplay between regulatory sequences that include not only upstream enhancer elements , but also core promoter regions , and flanking insulators that modulate interactions between enhancers and core promoters . In principle NURF could regulate transcription by remodelling chromatin at any of these elements . The core promoter regulates transcription initiation by serving as the recognition site for the basal transcription apparatus and determines specificity for upstream enhancers . In bilateria marked diversification of core promoter architecture occurs , with complexes including the canonical TATA-binding protein ( TBP ) containing complex TFIID , as well as complexes containing TBP-related factors ( TRFs ) such as the DREF/TRF2 complex , binding distinct core sequences [10–12] . Previous research has indicated that the DREF/TRF2 complex contains the ISWI , Caf1 and Nurf-38 subunits of NURF . However , involvement of the full NURF complex in DREF/TRF2 function is unclear . While peptides corresponding to the Nurf301/E ( bx ) subunit were not mapped to the DREF/TRF2 complex [13] , mutants lacking the Putzig subunit of DREF/TRF2 phenocopy Nurf301E ( bx ) mutants [14] , suggesting NURF interacts with DREF/TRF2 and may influence core promoter function . In turn , intra- and inter-chromosomal communication between the core promoter and enhancer regions can be controlled by insulator elements . These were first defined in Drosophila based on enhancer blocking function which can be mediated by combinations of three DNA-binding proteins—Boundary Element Associated Factor ( BEAF ) , CCCTC-Binding Factor ( CTCF ) and Suppressor of Hairy wing ( Su ( Hw ) ) –and two associated proteins , Centrosomal Protein 190 ( CP190 ) and Modifier of mdg4 ( Mod ( mdg4 ) ) [15] . To provide an unbiased assessment of regulatory elements at which NURF acts , we have used whole genome MNase-sequencing to map at base pair resolution nucleosomes that require NURF for positioning . In parallel , we have used chromatin immunoprecipitation sequencing ( ChIP-Seq ) of the Nurf301/E ( bx ) specificity subunit to identify sites of stable recruitment of NURF . To our knowledge this is the first comprehensive base pair resolution map of in vivo nucleosome targets of a metazoan chromatin remodeling enzyme . Our data indicate NURF action at three distinct transcription regulatory elements . In addition to upstream enhancer elements , we identify a novel function of NURF in orchestrating nucleosome spacing downstream of the +1 nucleosome on active genes regulated by DREF/TRF2 . Moreover , we also detect NURF remodeling and recruitment at distal insulator sites , where NURF interacts with DREF and known insulator proteins including CP190 to establish nucleosome-depleted domains .
To identify sites of NURF activity in the genome we profiled nucleosome distribution in Drosophila WT and NURF deficient primary macrophages ( hemocytes ) . As a first step normalized tag densities of uniquely-mapped reads were determined in a sliding 50 bp window across the genome and log2 fold changes between mutant and wild-type ( WT ) were used to define regions of altered read densities , we term shifts . Using this approach , we identified 47 , 000 shifted nucleosomes in Nurf301/E ( bx ) deficient larval hemocytes . Assuming an averaged female Drosophila genome size of 175 Mb [16] , this corresponded to <5% of nucleosomes , indicating that NURF remodelling was deployed locally at discrete nucleosomes and not globally over large arrays of nucleosomes . Local action by NURF was confirmed by visualizing NURF nucleosome shifts over entire chromosome arms using Hilbert plots ( S1 Fig ) , which indicated NURF nucleosome shifts were randomly distributed across both the X-chromosome and autosomes with no evidence of large domains of remodeling . The distribution of NURF nucleosome shifts relative to known gene features was then determined . Consistent with the hypothesis that NURF regulates transcription by remodelling enhancer nucleosomes , we observed significant enrichment of nucleosome shifts on 5’ regulatory elements and under-representation on coding exons and gene bodies ( Fig 1A , S2 Fig ) . However , NURF nucleosome shifts were also enriched on 5’UTRs and 3’ regulatory elements suggesting additional targets of NURF . To investigate these in greater detail , we determined the averaged distribution of NURF nucleosome shifts relative to all transcription start sites ( TSSs ) and transcription termination sites ( TTSs ) ( Fig 1B ) . This defined distinct functions for NURF in transcription regulation . Firstly , a broad enrichment of nucleosome shifts was detected extending 1 . 5 kb upstream of the transcription start site , consistent with remodelling at enhancers . Secondly , NURF-dependent remodelling was also observed downstream of the TSS , with a peak detected at the +1 nucleosome position and shifts detected above background for approximately 1 . 2 kb into the gene body , corresponding to 5–6 nucleosomes . Finally , NURF was also required to maintain nucleosome organisation at transcription terminators . These activities were defined at base pair resolution by mapping individual nucleosome dyad positions , which were then used to generate a continuous nucleosome density estimation that describes the probability of a nucleosome at each base pair in the genome . By generating nucleosome density estimates for both WT and Nurf301/E ( bx ) mutant backgrounds , nucleosomes that require NURF to maintain position could be discriminated . The validity of this mapping procedure was confirmed by analysing nucleosome positions on heat-shock loci , the targets originally used to identify NURF [17] . Nucleosome positions on heat-shock promoters are well documented , with the TF Trl binding GAGA sequences to recruit NURF , which in turn remodels nucleosomes establishing a nucleosome-depleted region [6 , 17] . Consistent with this , WT hemocyte nucleosome probabilities showed a clearly positioned nucleosome adjacent to Trl-binding GAGA elements ( Fig 1C ) . However , in Nurf301/E ( bx ) mutants this nucleosome position was not maintained and new nucleosome positions were detected ( Fig 1C and 1D ) . We next confirmed that changes in nucleosome position in Nurf301/E ( bx ) mutants corresponded to sites of NURF localisation determined by ChIP-seq . Nurf301/E ( bx ) ChIP peaks were ordered according to strength of associated signal , and sub-divided into quintiles from highest ( 1st ) to lowest ( 5th ) Nurf301/E ( bx ) signal . Analysis of nucleosome position flanking these sites demonstrated consistent changes in nucleosome position in Nurf301/E ( bx ) mutants ( Fig 1E ) . This was confirmed by peaks in nucleosome shifts flanking NURF ChIP peaks ( Fig 1F ) . In addition , microarrays were used to examine expression of other chromatin regulators in Nurf301/E ( bx ) mutant hemocytes . Analysis of expression of 667 genes that have assigned GO ( gene ontology ) molecular functions related to chromatin and transcription ( S3A Fig ) showed no substantial decreases in expression in other chromatin modifying and remodelling complexes ( S3B Fig ) . Taken together , these data suggest that the nucleosome shifts and changes in nucleosome distribution determined above were direct measures of NURF activity . NURF nucleosome remodelling flanking Trl-binding sites , and the broad peak of nucleosome shifts over the 5’ upstream regions of genes , were consistent with NURF’s proposed function of collaborating with TFs at enhancers to slide nucleosomes and control transcription . To establish if this was a specific property of Trl , or whether NURF cooperates with other families of TFs to remodel enhancer nucleosomes , we screened Drosophila TFs with known DNA-binding consensi for NURF-dependent nucleosome positioning and remodeling activity . Using DNA-binding consensi , predicted genome-wide binding sites for individual TFs were computed and averaged nucleosome density flanking these in WT and NURF-deficient hemocytes determined ( Fig 2 , S4 and S5 Figs ) . Based on flanking nucleosome organization and changes in NURF mutants , TFs could be assigned to one of five categories . Predicted binding sites for class I TFs were independent of NURF and located within nucleosomes , ( Fig 2A ) . Binding sites for the remaining classes of TFs ( Class II-V ) , were flanked by NURF-dependent nucleosome positions and revealed three distinct modes by which NURF and TFs could potentially influence enhancer nucleosome position . Thus , class II and class III predicted TF-binding sites were located either immediately adjacent to a nucleosome ( Fig 2B ) , or at nucleosome entry/exit points ( Fig 2C ) , respectively . In Nurf301/E ( bx ) mutants these positions were lost and chromatin flanking the site was more accessible , suggesting these TFs potentially could tether NURF to mediate directional nucleosome sliding . In contrast , class IV predicted TF-binding sites , exemplified by Trl ( Fig 2D ) , were located in extended nucleosome-depleted regions , which show higher nucleosome signals in Nurf301/E ( bx ) mutants . The final mode was observed flanking consensi for DNA-binding proteins including Su ( Hw ) ( Fig 2E ) . Binding sites for this class of TF were localized in a nucleosome-depleted domain flanked by organized nucleosome arrays that migrate towards the predicted binding sites in Nurf301/E ( bx ) mutants . As controls , analysis of averaged nucleosome position flanking TF sites that were clustered or isolated was broadly consistent with trends observed with all sites ( S6 Fig ) . In addition , microarray analysis of third instar larval hemocytes showed no change in TF expression in Nurf301/E ( bx ) mutants ( S3C Fig ) , excluding indirect effects of changes in expression in these factors on nucleosome organisation . However , NURF function was not restricted to remodelling nucleosomes at enhancers . Our initial comparative analysis of nucleosome tag densities detected potential nucleosome shifts not only upstream but also downstream of the TSS ( Fig 1B ) . To elucidate the basis of this , we used the calculated continuous nucleosome density estimations to compare nucleosome position downstream of all TSSs in the Drosophila genome in both WT and Nurf301/E ( bx ) mutant backgrounds . This showed the expected nucleosome distribution in WT samples with a nucleosome-depleted region flanking the TSS , a well-positioned +1 nucleosome and a regular downstream array of nucleosomes ( Fig 3A ) . In Nurf301/E ( bx ) mutants , a superficially similar distribution was observed , with a regular array of nucleosomes downstream of the TSS . However absolute nucleosome position was shifted towards the TSS for the first six nucleosomes following the TSS ( +1 to +6 nucleosome positions ) . This analysis included both active and inactive TSSs . To discriminate if NURF differentially affected transcriptionally active versus inactive promoters , we categorized active and inactive TSSs in hemocytes by profiling the hemocyte distribution of the H3K4me3 HPTM , which decorates the +1 nucleosome of active genes ( Fig 3B ) . Profiling averaged nucleosome probability relative to active ( +H3K4me3 ) and inactive ( -H3K4me3 ) TSSs showed that clearly defined nucleosome arrays were only detected downstream of active TSSs and not inactive TSSs ( Fig 3A ) . On active genes in WT cells the +1 nucleosome was located at +132 bp relative to the TSS , but was shifted closer to the TSS at +123 bp in Nurf301/E ( bx ) mutants ( Fig 3A , +H3K4me3 ) . Nucleosome shifts towards the TSS in Nurf301/E ( bx ) mutants were propagated over the next five nucleosomes after which the nucleosome array reset and limited differences between mutant and WT arrays were observed . These trends using averaged nucleosome density profiles were confirmed on randomly selected , individual H3K4me3-containing genes . For example , nucleosome positions detected on the CG10699 gene showed a 10 bp change in position of the +1 nucleosome in Nurf301/E ( bx ) mutants ( Fig 3C ) , and shifts expanding over successive nucleosomes , but limited to the first six nucleosomes . This restriction of nucleosome shifts to the first six nucleosomes downstream of the TSS , agreed well with our initial comparative analysis of nucleosome tag-density , where we only detected nucleosome shifts within the first 1 . 2 kb downstream of the TSS ( Fig 1B ) . Calculating average nucleosome repeat length downstream of the TSS on active and inactive genes in both genetic backgrounds showed that nucleosome repeat length on active genes decreased from 175 bp in WT samples to 170 bp in Nurf301/E ( bx ) mutants . In contrast , on inactive genes , the extrapolated WT average nucleosome repeat length of 185 bp was unchanged in Nurf301/E ( bx ) mutants . Similar analysis of nucleosome positions at other genomic regions , for example flanking exon-intron boundaries , also showed no change in nucleosome spacing in Nurf301/E ( bx ) mutants . Taken together we conclude that NURF has a specific nucleosome positioning and spacing function downstream of active TSSs . To establish if this NURF nucleosome spacing function depends on the absolute level of transcription , normalized WT and Nurf301/E ( bx ) mutant hemocyte gene expression levels were determined by Affymetrix microarray profiling . TSSs were binned into quintiles based on normalized WT transcript level from high ( 1st ) to low ( 5th ) , and averaged nucleosome position flanking TSSs of each quintile determined in both backgrounds . We observed that , while the WT +1 nucleosome dyad location was shifted downstream as expression level increased ( from +125 bp in the lowest expression quintile to +136 bp in the highest quintile ) , for all quintiles +1 nucleosome positions were relocated towards the TSS in Nurf301/E ( bx ) mutants ( Fig 3D ) . The extent of this movement was constant for all quintiles indicating that NURF-dependent nucleosome spacing on active genes is independent of absolute transcript level . We next examined the consequences of this spacing change on expression of active genes . Our initial assumption was that active gene expression would be reduced in Nurf301/E ( bx ) mutants . Surprisingly , real-time RT-PCR ( Fig 3E ) showed increased expression of active ( +H3K4me3 ) genes in Nurf301/E ( bx ) mutant hemocytes . Furthermore , using whole genome expression profiles to categorize TSSs with increased or decreased expression in Nurf301/E ( bx ) mutants , and profiling averaged nucleosome densities flanking the corresponding TSSs , indicated that only TSSs up-regulated in Nurf301/E ( bx ) mutants demonstrated signature NURF-dependent +1 nucleosome shifts ( Fig 3F ) . No changes in +1 nucleosome position were observed on TSSs with decreased expression ( Fig 3F ) . We conclude that NURF dampens expression of transcriptionally active genes . We demonstrated a specific NURF nucleosome spacing function downstream of active TSSs . To discriminate if NURF promoter targeting was mediated by core promoter components we classified TSSs based on associated core promoter elements . Nucleosome organisation flanking these distinct categories of TSSs was then determined in WT and Nurf301/E ( bx ) mutants . This showed clear NURF activity on promoters containing DREs and Ohler motifs 1 , 5 , and 7 ( Fig 4A , S7 Fig ) , elements enriched on TRF2-bound promoters [18] and house-keeping core-promoters [10] . This promoter class exhibited a well-defined +1 nucleosome and robust downstream nucleosome array that was shifted towards the TSS in Nurf301/E ( bx ) mutants . NURF activity on DREF-dependent promoters was confirmed by comparison of Nurf301/E ( bx ) and published DREF ChIP-Seq profiles [19] . Heatmaps indicated that DREF and Nurf301/E ( bx ) were enriched and colocalized at active TSSs ( Fig 4B ) , confirmed by co-immunoprecipitation which showed physical association of NURF and the DREF/TRF2 complex subunit Washout ( Wash ) in S2 cell extracts ( Fig 4C ) . In contrast , promoters containing TATA , INR , MTE and DPE motifs exhibited less well-defined nucleosomal arrays ( Fig 4A , S7 Fig ) . On INR- , MTE- and DPE-containing promoters the +1 nucleosome was located approximately 10 bp further into the gene body ( +142 bp ) , a location characteristic of paused promoters [20] ( S8 Fig ) . Significantly , no relative change in position of these nucleosomes was observed in Nurf301/E ( bx ) mutants ( Fig 4A , S7 Fig ) . Taken together our data indicate predominant NURF function on active promoters that are targets of DREF/TRF2 . In addition to targeting active promoters , inspection of Nurf301/E ( bx ) ChIP-Seq profiles revealed NURF localisation to isolated elements that corresponded to previously defined insulator elements , including the Fab8 and Mcp boundaries in the bithorax complex ( Fig 5A ) . This concurred with our analysis of nucleosome distribution flanking TF-binding sites ( Fig 2 ) , which showed Su ( Hw ) and BEAF insulator proteins collaborate with NURF to establish nucleosome-depleted domains . NURF targeting to insulators was verified by ChIP-Seq of the core insulator component CP190 , which showed good correlation with Nurf301/E ( bx ) at predicted insulator sites genome-wide ( Fig 5B , cdBEST Boundaries [21] ) . Inspection of boundaries in the bithorax complex demonstrated that CP190 and Nurf301/E ( bx ) co-localized in nucleosome-depleted regions in WT hemocytes , at which nucleosomes were detected in Nurf301/E ( bx ) mutants ( Fig 5A , expanded view ) . Nucleosome profiling flanking ChIP peaks for the DNA-binding insulator components CTCF , Su ( Hw ) and BEAF [22] showed that these occurred within nucleosome-depleted domains flanked by ordered nucleosome arrays . However , levels of the associated cofactor CP190 determined nucleosome reorganization at these sites . Thus clear nucleosome reorganization was observed when insulator-binding sites for CTCF ( Fig 5C , S9 Fig ) , Su ( Hw ) ( Fig 5C , S10 Fig ) , BEAF ( S11 Fig ) , and Mod ( mdg4 ) ( S12 Fig ) were ordered according to CP190 level . Segregation of binding sites for Su ( Hw ) , and BEAF into those with or without CP190 confirmed that CP190 determined nucleosome organization around insulator elements . In Nurf301/E ( bx ) mutants , nucleosome-depleted domains at CP190-containing insulator sites were reduced in size and nucleosomes repositioned . Consistent with the discriminating role of CP190 , reciprocal co-immunoprecipitation revealed that NURF and CP190 were physically associated in S2 cell extracts ( Fig 5D and 5E ) . Taken together , our data show NURF can localize to both core promoters and insulators . Our initial working assumption was that these distributions result from separate targeting of NURF either to promoters by DREF/TRF2 or to insulators by CP190 . Surprisingly , inspection of DREF ChIP profiles indicated colocalization with Nurf301/E ( bx ) and CP190 at insulator elements in the bithorax complex ( Fig 5A ) and predicted insulator sites genome-wide ( Fig 5B ) . Moreover , as we had observed for CP190 ( Fig 5C ) , levels of DREF determined nucleosome reorganization at these insulator sites . Thus , nucleosome profiling flanking Su ( Hw ) and CTCF binding sites showed that clear nucleosome-depleted domains were only detected when insulator-binding sites contained DREF ( Fig 6A ) . Those lacking DREF were occupied by nucleosomes ( Fig 6B ) . In Nurf301/E ( bx ) mutants , nucleosome-depleted domains at DREF-containing insulator sites were reduced in size . Consistent with DREF action at insulators , co-immunoprecipitation showed that the DREF/TRF2 subunit Wash and CP190 were physically associated in S2 cell extracts ( Fig 6C ) . In addition , Wash was poly ADP ribose ( PAR ) modified ( Fig 4C ) , a hallmark of destabilized chromatin [23] that occurs at insulator sites . Functional requirement for NURF at an insulator in vivo was confirmed using the known enhancer blocking function of gypsy retrotransposons . These contain binding sites for Su ( Hw ) , which can recruit insulator components to establish a functional insulator . When integrated between the wing enhancer and promoter of the cut gene in the ct6 mutation , gypsy elements suppress ct wing expression and disrupt the adult wing margin . This normally is decorated with approximately 90 mechanosensory bristles . Loss of bristles in ct6 mutants can be used as a sensitive assay for gypsy insulator function . In the absence of other mutations ct6 alleles reduce mechanosensory bristle number to 14 ( Fig 6D and 6E ) . Loss of one copy of either Nurf301/E ( bx ) or CP190 increases mechanosensory bristle number , while simultaneous removal of one copy of both shows synergistic increase in bristle number , demonstrating functional cooperation between NURF and insulator proteins at the gypsy insulator . Our data demonstrated that insulators acted on by NURF are targeted by multiple proteins including DREF and CP190 . Such co-association of multiple insulator proteins is a feature of subclasses of insulator elements that often mark the boundaries of H3K27me3 domains or topologically associating domains ( TADs ) [24 , 25] , suggesting these are the targets of NURF . Evidence for this was provided by the Drosophila even-skipped locus . In S2 cells this is spanned by a well-defined H3K27me3 domain , flanked by active H3K4me3 containing genes with peaks containing both NURF and CP190 detected at the boundaries between these domains ( Fig 6F ) . Consistent with NURF function at such boundaries , we observed negative correlation genome-wide between H3K27me3 and either NURF , CP190 , or DREF ( Fig 5B ) , and inverse correlation between NURF , CP190 , or DREF and H3K27me3 levels at H3K27me3 domain boundaries ( Fig 6G ) . Finally , as Nurf301/E ( bx ) , DREF and CP190 colocalize at insulators , and Nurf301/E ( bx ) and DREF also overlap at active promoters , we tested whether insulator components may also be similarly localized to active promoters . Comparison of CP190 and Nurf301/E ( bx ) ChIP-Seq profiles in S2 cells reveals that CP190 was present at TSSs . However , CP190 was only detected on active genes ( Fig 7B , compare +H3K4me3 and –H3K4me3 CP190 traces ) . This was confirmed by heatmaps of CP190 and NURF signals relative to H3K4me3 around TSSs , which show colocalization of signals on active promoters ( Fig 7A and 7D ) . CP190 binding to both active promoters and distant insulator elements resembles the reported association of TAF3 with CTCF that mediates looping between distal enhancer elements and the proximal promoter [26] , raising the possibility that NURF , DREF and CP190 co-localization to insulators and proximal promoters reflects functional interaction between these elements . To test whether increased expression of active genes observed in NURF mutants ( Fig 3E ) was due to impaired insulator function , we tested whether loss of CP190 similarly increased active gene expression . However , although expression of some genes increased in CP190 mutants , up-regulation was not consistently observed for all genes ( Fig 7C ) , suggesting that NURF plays additional roles distinct from insulator function at these targets .
Modulation of nucleosome dynamics by ATP-dependent chromatin remodelling enzymes has the potential to regulate all chromatin templated reactions . Key to understanding functions in diverse processes ranging from transcription to repair and replication is to elucidate in vivo targets and mechanisms by which ATP-dependent chromatin remodelling activity is deployed . Here we have used mononucleosome mapping of WT and Nurf301/E ( bx ) mutant cells to discriminate nucleosomes remodelled by NURF in vivo . We detect interactions with transcription factors and insulator components that direct NURF activity to both gene promoters and distant regulatory elements . To our knowledge this is the first comprehensive base pair resolution map of in vivo nucleosome targets of a metazoan chromatin remodeling enzyme . Our data indicate NURF remodelling occurs at discrete sites in the genome , but we speculate that by mediating local chromatin reorganization NURF can profoundly impact genome organization and long-range regulatory interactions . A central question in the ATP-dependent chromatin remodelling field is whether remodelers are highly abundant , affecting all nucleosomes and genome function generally , or more restricted factors that are recruited to discrete sites to regulate specific nuclear processes . Early estimates of Drosophila ISWI abundance [17] , and disruption of entire polytene X chromosomes in Nurf301/E ( bx ) mutants [7 , 27] , were consistent with global function of NURF . However , highly-specific changes in gene expression observed in Nurf301/E ( bx ) mutants suggest more localized effects on nucleosome positioning [7–9] . Data generated here confirm local and not global nucleosome disruption by NURF . Fewer than 5% of nucleosomes genome-wide are remodelled by NURF and large domains of nucleosome reorganisation were not detected by Hilbert curve analysis . Our data in diploid primary hemocytes is fully consistent with array-based analysis of nucleosomes at selected genome regions in ISWI mutant polytene salivary glands which fail to show global nucleosome disruption [28] . NURF targets defined here include upstream enhancers , consistent with the original isolation of NURF as a TF cofactor and initial transcriptome analysis [7–9 , 17] . Our data show NURF collaborates with TFs at enhancers to remodel nucleosomes organisation around predicted TF binding sites . We distinguish several modes of NURF-dependent nucleosome organisation at predicted TF sites: precise positioning of nucleosomes adjacent to TF sites; disruption of nucleosomes surrounding TF sites; and a barrier function in which TF sites establish a nucleosome-depleted domain and organize flanking nucleosome position . In combination these data point to distinct modalities by which TFs may deploy NURF to achieve alternate functional outcomes . A caveat , however , is that these data were generated using predicted TF-binding sites based on binding consensi . While available ChIP-Seq datasets of actual Su ( Hw ) , BEAF and Trl binding sites confirm these trends , it remains formally possible that some predicted sites may not be occupied by TFs and/or that TF-binding may be lost in Nurf301/E ( bx ) mutants . Nevertheless , these results are broadly consistent with in vitro experiments demonstrating for example how GAL4 provides a “barrier” that competes with nucleosomes for occupancy at DNA targets [29 , 30] or , alternatively , how tethering of remodelers like CHD1 can mediate directional nucleosome sliding [31] . However , our data reveal an additional nucleosome spacing function for NURF downstream of active TSSs including house-keeping targets of TRF2/DREF . The consequences of this are distinct from remodelling at enhancers , which mediates high-level changes in transcription of specific developmental/signal transduction pathways , instead exerting a general dampening effect on active gene transcription . Our data indicate that NURF shifts nucleosomes in the 3’direction away from the TSS , similar to yeast ISW1b and CHD1 remodelers [32] and also increases spacing between nucleosomes . This spacing may facilitate recruitment of factors with nucleosome-spacing dependent binding , like the Rpd3 ( S ) complex [33 , 34] . In yeast , Rpd3 ( S ) represses intragenic transcription from cryptic initiation sites [35] . One consequence of decreased spacing in NURF-deficient cells may be the failure to recruit repressors of cryptic initiation and increase in spurious transcription , consistent with our observed up-regulation of active gene expression in Nurf301/E ( bx ) mutants . NURF localisation and activity has parallels with yeast ISWI remodelling complexes , which show recruitment to TSSs but action on gene bodies [32 , 36] . We detect binding of NURF to nucleosome depleted regions ( NFRs ) upstream of the TSS , consistent with yeast ISWI localization , where NFRs provide extended linkers required for remodelling [36] . In our case , this localization is reinforced by interactions with basal transcription factors . In particular , NURF colocalizes and interacts physically and functionally with the TBP-related TRF2/DREF complex . It has been postulated that diversification of core promoter factors and evolution of TRF2 has driven the transcriptional complexity that facilitated the evolution of the bilaterian body plan [37] . It is intriguing that this is accompanied by the emergence of bilaterian-specific chromatin remodeling enzymes . Unlike the SWI/SNF2 and INO80/SWR1 complexes , which show conservation of most subunits throughout eukaryotic lineages , ISWI-complexes like NURF exhibit distinct non-catalytic subunits in bilateria . We speculate that core promoter diversification demands co-evolution of new remodeler complexes to accommodate increasing regulatory complexity . Promoter localization of NURF is likely further stabilized by NURF binding HPTMs that decorate the +1 nucleosome [38 , 39] but , also may be mediated by direct DNA binding . Nurf301/E ( bx ) contains two N-terminal AT-hook domains which bind AT-rich DNA , and have been shown to be required for full NURF nucleosome sliding activity and nucleosome-binding in vitro [6] . A precedent for this is provided by studies of the related SWR1 remodelling complex where targeting to promoter elements is mediated by the DNA-binding SWC2 subunit [40] . It has been suggested recently that remodeler peaks observed at active TSSs may be “phantom” artefacts of the ChIP procedure and should be treated with caution [41] . In our case , however , we observe nucleosome remodelling flanking NURF ChIP signals and NURF binds HPTMs that decorate the +1 nucleosome that flanks the ChIP sites [38 , 39] . It is formally possible that some changes in nucleosome positioning observed in Nurf301/E ( bx ) mutants are not the consequence of loss of NURF ATP-dependent nucleosome-sliding activity , but rather that NURF acts stoichiometrically and non-catalytically at some sites , binding and physically occupying linker DNA in a manner incompatible with nucleosome formation . Changes in nucleosome position could thus potentially be due to the lack of NURF physical presence at some sites . The ability of remodelers to engage in either “traditional” catalytic ATP-dependent nucleosome sliding versus and non-catalytic modes would allow diversification of remodeler function , and offer the potential that allosteric modulation ( by for example histone modifications ) could switch complexes between different modes of remodeling to program distinct local chromatin architectures . The use of strains containing catalytically inactive NURF induced by expressing mutant forms ISWI , in which ATPase activity is eliminated but complex assembly unaffected , could offer one approach to experimentally investigate this . It is also important to consider that these experiments were performed in cells in which function of other remodeling complexes was unaffected . Systematic profiling of yeast chromatin remodeler distributions indicates overlapping domains of function . As such it is possible that NURF is functionally redundant with other chromatin remodeling enzymes [42] . Thus , targets and nucleosome reorganization identified here may under-represent complete NURF function . There may be sites at which NURF function can be substituted by other chromatin remodeling enzymes in Nurf301/E ( bx ) mutants . Indeed initial analysis of yeast remodeling enzymes required depletion of three remodeling enzymes ( isw1 , isw2 and chd1 ) to observe substantial effects on nucleosome organisation [43] Finally , our data reveal that NURF interacts with the insulator elements and interacts with components including CP190 and DREF . Activity of NURF at insulators is consistent with our initial studies showing that Nurf301/E ( bx ) mutants modify bithorax mutations and that Nurf301 corresponds to Enhancer of bithorax E ( bx ) [7] . The bithorax mutations are caused by gypsy transposons that bind Su ( Hw ) and act as ectopic insulators to disrupt expression . NURF remodeling at insulator elements is also consistent with results showing that NURF is required for Drosophila cell-based insulator/enhancer blocking assays [44] . This appears to be a conserved function as studies in vertebrates show that NURF mediates chromatin barrier function at the chicken β-globin locus [45] and interacts with CTCF to regulate gene expression in mammals [46] . Interestingly , colocalization between Nurf301/E ( bx ) and insulator proteins is not only detected at distal insulator regions but also at active promoters . We postulate that overlap of insulator components and NURF at insulators as well as promoters reflects functional interaction between distant insulators and active promoters as has been speculated for CTCF at mammalian promoters [26 , 47] . It is tempting to speculate that functional clustering of regulatory elements provides a solution to how chromatin remodeling enzymes engage targets in the genome . Three dimensional clustering of targets in proximity would allow rapid recapture of ATP-dependent chromatin remodelers at distinct regulatory elements that require nucleosome reorganisation .
Nurf3012 , Cp1901 and Cp1902 alleles were as described [7 , 48] . The Nurf3012 allele is an EMS-induced mutation that encodes a glutamine to stop codon substitution ( aa 545 ) that truncates Nurf301 after the first PHD finger and which behaves genetically as a null allele . Unless stated flies were raised at 25°C . Third instar larvae were sexed and primary hemocytes collected from wild-type and Nurf3012 mutant female third instar larvae as described previously [9] . Briefly larvae were ripped in batches of 50 third instar larvae into HyQ-CCM3 insect medium ( Thermo Fisher Scientific ) containing protease inhibitors ( Complete , Roche ) . Cells were fixed with 1% formaldehyde in 1XPBS for 15 minutes at 25°C and pelleted at 400g for five minutes . Cells were washed three times with ice cold 1XPBS containing protease inhibitors and stored as pellets at -80°C until required . S2-DRSC cells were cultured at 25°C in Insect-XPRESS medium ( Lonza ) containing 10% FCS . S2 cells for standard ChIP were fixed with 1% formaldehyde and washed as described above . Hemocyte preparations from 1000 larvae were thawed , pooled and resuspended in buffer A ( 15 mM Tris ( pH 7 . 4 ) , 15 mM NaCl , 60 mM KCl , 0 . 34 M sucrose , 1 mM DTT , 25 mM sodium metabisulfite , 0 . 5 mM spermidine , 0 . 15 mM spermine ) . Cells were homogenized with a pellet pestle . CaCl2 was added to a final concentration of 1 mM , 800U MNase ( Worthington ) added , and the sample incubated for 12 minutes at 16°C to liberate mononucleosomes . Digestion was stopped by adding an equal volume of Stop buffer ( 0 . 1 M Tris ( pH 8 . 5 ) , 0 . 1 M NaCl , 50 mM EDTA , 1% SDS ) and samples centrifuged at 17 , 400g for 10 minutes to purify soluble chromatin . Formaldehyde cross-links were removed by addition of NaCl to 150 mM , one-tenth volume Proteinase K ( Roche ) and incubation at 65°C overnight . DNA was purified by phenol/chloroform extraction and ethanol precipitation , treated with RNase ( Promega ) at 37°C for 1 hour and then run on 2 . 2% recovery FlashGels ( Lonza ) . Mononucleosomal DNA was pipetted from recovery wells in FlashGel Recovery Buffer and DNA subjected to further round of phenol/chloroform extraction and ethanol precipitation . Libraries were generated from two biological replicates for each genotype , sequenced , and reads pooled for nucleosome mapping . For MNase-ChIP experiments , hemocytes from 1000 larvae were processed as above but soluble chromatin after Stop buffer addition was diluted in ChIP dilution buffer ( 16 . 7 mM Tris ( pH 8 . 1 ) , 167 mM NaCl , 1 . 2 mM EDTA , 0 . 1% SDS , 1 . 1% Triton X-100 ) , and processed for ChIP as described below . ChIP was performed as described in [38] with the following modifications . Samples were pre-cleared using Protein G-conjugated Dynabeads ( Invitrogen ) for 30 minutes at room temperature , followed by incubation with antibody coated Protein G-conjugated Dynabeads ( Invitrogen ) for 2 . 5 hours at room temperature . Immune complexes were recovered by magnetic selection , and washed once with low salt buffer ( 20 mM Tris ( pH 8 . 1 ) , 150 mM NaCl , 2 mM EDTA , 0 . 1% SDS , 1% Triton X-100 ) , once with High salt buffer ( 20 mM Tris ( pH 8 . 1 ) , 500 mM NaCl , 2 mM EDTA , 0 . 1% SDS , 1% Triton X-100 ) , once with LiCl immune complex wash buffer ( 10 mM Tris ( pH 8 . 1 ) , 1 mM EDTA , 0 . 25 M LiCl , 1% IGEPAL CA-630 , 1% deoxycholic acid ) and twice with TE buffer for five minutes each at room temperature . ChIP DNA was eluted using two washes of elution buffer ( 1% SDS , 0 . 1 M NaHCO3 ) for 15 minutes at room temperature . Cross-links were reversed as described above and ChIP DNA purified using 1 . 8 volumes Agencourt AMPure XP beads ( Beckman Coulter ) . The following antibodies were used: rabbit anti-NURF301 , rabbit anti-H3K4me3 ( 17–614 , Millipore ) , rabbit anti-CP190 [49] . DNA for MNase-Seq and ChIP-Seq was end-repaired and sequencing libraries prepared using a SOLiD Fragment Library Construction Kit ( Life Technologies ) . ChIP DNA was barcoded using the SOLiD Fragment Library Barcoding Kit Module 1–16 . Sequencing libraries were run on a SOLiD 4 genome analyzer . SOLiD reads for nucleosome mapping were mapped to the Drosophila genome ( BDGP R5/dm3 Assembly ) in color space using Bowtie [50] and filtered for high quality reads . Nucleosome density profiles were generated using F-seq [51] . To determine regions with altered nucleosome position between wild-type and mutant cells chromosomes were divided into 50 bp windows and read number in each bin determined . Bins with log2 fold change greater than 2 between mutant and wild-type were identified as regions with nucleosome shifts . ChIP reads were mapped to the Drosophila genome ( BDGP R5/dm3 Assembly ) using the bioscope mapping tool ( Life Technologies ) . Reads were then filtered for high quality reads where read quality was greater than 15 using Samtools [52] . ChIP peaks were called using MACS [53] and MACS ChIP wiggle tracks were then imported into Galaxy and filtered to select peak signals . Tools to generate Pearson correlation coefficients for ChIP-Seq profiles at defined genomic regions , averaged signal density relative to defined genomic regions and heatmaps of nucleosome and ChIP-Seq signals at defined regions were generated using the Cistrome package [54] . Publically available data tracks GSM762836 , GSM762837 , GSM762838 , GSM762839 , GSM762840 , GSM762841 , GSM762842 , GSM762843 , GSM762844 , GSM762845 , GSM762846 , GSM762847 , GSM762848 , GSM762849 [25] were used to define developmentally-stable insulator peaks . Nucleosome sequence files have been deposited at European Nucleotide Archive ( Study Accession PRJEB12941 ) . w1118 and Nurf3012 wandering third instar larvae were staged using the blue-gut method and hemocytes isolated in batches of 50 larvae as described previously [9 , 38] . mRNA was isolated from hemocytes isolated from the equivalent of 1000 animals using Trizol as described [9 , 38] and mRNA amplified and labeled using the GeneChIP Eukaryotic Small Sample Target Labeling Assay VII ( Affymetrix ) . Triplicate labeled mRNA samples were hybridized to GeneChip Drosophila Genome Arrays ( Affymetrix ) . Statistical analysis was carried out using R version 3 . 1 . 2 ( http://www . R-project . org ) and the gcrma and limma libraries of Bioconductor version 3 . 0 ( http://www . bioconductor . org ) . Expression values were computed using gcrma [55] . Array datasets are available through ArrayExpress ( accession number E-MTAB-4537 ) . Soluble nuclear fraction ( SNF ) was prepared from Drosophila S2 cells using a modification of the protocol of Wysocka and colleagues [56] to prepare nuclei . Nuclear pellets were extracted using extraction buffer ( 10mM HEPES ( pH 7 . 9 ) , 400 mM KCl , 3mM MgCl2 , 5% Glycerol , 0 . 5 mM DTT , 1 mM Sodium Metabisulphite , 1x Protease inhibitors ( Complete , Roche ) ) for 1 hour on ice with gentle swirling . Extracts were clarified by centrifugation at 100 , 000 g for 1 hour at 4°C . SNF was dialysed against extraction buffer adjusted to 100mM KCl . Protein G-conjugated Dynabeads ( Invitrogen ) were blocked by incubation in blocking buffer ( 1xPBS , 5mg/ml BSA , 1x Protease inhibitors ( Complete , Roche ) ) and antibodies bound by incubation overnight at 4°C . SNF was diluted into bind buffer ( 1XPBS , 5mg/ml BSA , 0 . 1% Tween-20 , 1XProtease inhibitors ) incubated with antibody-coated Protein G-conjugated Dynabeads for 2 . 5 hrs with rotation at 4°C . Beads were washed four times , 10 minutes each , using wash buffer ( 1xPBS , 0 . 1% Tween-20 ) . Bound proteins were eluted by boiling in 1XSDS-PAGE sample buffer . 5% input run as a loading control for all IPs . Additional IPs were performed using soluble chromatin prepared according to the protocol of Wysocka and colleagues [56] . Nuclei were lysed and chromatin ( predominantly mononucleosomal ) liberated by digestion with MNase ( Worthington ) . Soluble chromatin was diluted into binding buffer and incubated with antibody-coated Protein G-conjugated Dynabeads and washed as above . For RNase A , Benzonase and DNase I treatments , beads were pelleted and incubated in digestion buffer ( 1XPBS , 2 . 5mM MgCl2 , 0 . 5mM CaCl2 0 . 1% Tween-20 ) containing the respective enzymes at 37°C for 10 minutes , followed by three further washes in wash buffer . Bound proteins were eluted as described above . Antibodies used were rabbit anti-NURF301 , rabbit anti-CP190 [49] , mouse anti-Gro ( anti-Gro was deposited to the DSHB by C . Delidakis ) , mouse anti-E ( z ) ( sc-25903 , Santa Cruz Biotechnology ) , mouse anti-Wash ( P3H3-Wash was deposited to the DSHB by S . Parkhurst ) and mouse anti-pADPr ( Mab 10H , sc-56198 , Santa Cruz Biotechnology ) . For confirmation of microarray expression data , mRNA was isolated from wild-type and mutant hemocytes using μMacs columns according to manufacturer’s instructions ( Miltenyi Biotec , Auburn , CA ) and reverse transcribed by Superscript II ( Invitrogen ) at 42°C . Primer sets used are listed in S1 Table .
|
In eukaryotes DNA is folded and compacted into manageable units by wrapping around a protein spool of histone proteins to form nucleosomes . By varying the position and dynamics of nucleosomes using energy-dependent chromatin remodeling enzymes , genes can be selectively turned off or on in cells , controlling development and cellular function . Distinct sub-families of ATP-dependent chromatin remodeling enzymes have been characterised . However , their specific nucleosome targets in the genome and how they are recruited to these are not completely defined . Here we have identified nucleosome targets of the conserved higher eukaryotic chromatin remodeling enzyme NURF . Our data indicate three distinct functions for NURF during transcription . NURF organizes nucleosome positions at gene enhancer elements to regulate transcription initiation , but is also required to maintain nucleosome position downstream of the transcription start site of active genes . In addition , we detect NURF remodeling and recruitment at distal insulator sites that are required for functional organisation of the genome . We postulate that NURF function at insulators as well as promoters reflects functional interaction between distant insulators and active promoters , with functional clustering of regulatory elements providing a solution to how chromatin remodeling enzymes engage multiple targets in the genome .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"invertebrate",
"genomics",
"animals",
"animal",
"models",
"nucleosome",
"mapping",
"insulators",
"drosophila",
"melanogaster",
"model",
"organisms",
"materials",
"science",
"epigenetics",
"molecular",
"biology",
"techniques",
"chromatin",
"drosophila",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"gene",
"mapping",
"chromosome",
"biology",
"animal",
"cells",
"gene",
"expression",
"materials",
"by",
"attribute",
"molecular",
"biology",
"animal",
"genomics",
"nucleosomes",
"insects",
"hemocytes",
"arthropoda",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"genomics",
"organisms"
] |
2016
|
Genome-Wide Mapping Targets of the Metazoan Chromatin Remodeling Factor NURF Reveals Nucleosome Remodeling at Enhancers, Core Promoters and Gene Insulators
|
Somatostatin-expressing , low threshold-spiking ( LTS ) cells and fast-spiking ( FS ) cells are two common subtypes of inhibitory neocortical interneuron . Excitatory synapses from regular-spiking ( RS ) pyramidal neurons to LTS cells strongly facilitate when activated repetitively , whereas RS-to-FS synapses depress . This suggests that LTS neurons may be especially relevant at high rate regimes and protect cortical circuits against over-excitation and seizures . However , the inhibitory synapses from LTS cells usually depress , which may reduce their effectiveness at high rates . We ask: by which mechanisms and at what firing rates do LTS neurons control the activity of cortical circuits responding to thalamic input , and how is control by LTS neurons different from that of FS neurons ? We study rate models of circuits that include RS cells and LTS and FS inhibitory cells with short-term synaptic plasticity . LTS neurons shift the RS firing-rate vs . current curve to the right at high rates and reduce its slope at low rates; the LTS effect is delayed and prolonged . FS neurons always shift the curve to the right and affect RS firing transiently . In an RS-LTS-FS network , FS neurons reach a quiescent state if they receive weak input , LTS neurons are quiescent if RS neurons receive weak input , and both FS and RS populations are active if they both receive large inputs . In general , FS neurons tend to follow the spiking of RS neurons much more closely than LTS neurons . A novel type of facilitation-induced slow oscillations is observed above the LTS firing threshold with a frequency determined by the time scale of recovery from facilitation . To conclude , contrary to earlier proposals , LTS neurons affect the transient and steady state responses of cortical circuits over a range of firing rates , not only during the high rate regime; LTS neurons protect against over-activation about as well as FS neurons .
Low threshold-spiking ( LTS ) neurons are a specific subtype of interneuron in the neocortex . Their somata are located in layers 2–6 [1] , and they include the Martinotti cells of layer 5 [2] , [3] , [4] , [5] and the green fluorescent protein ( GFP ) -expressing neurons of the GIN line of transgenic mice [6] , [7] , [8] . LTS neurons express the neuropeptide , somatostatin , their action potentials have intermediate duration , and they adapt in response to suprathreshold step current injections [9] . The difference between the resting membrane potential and firing threshold of LTS cells is about 12 mV , smaller than observed in excitatory neurons or other types of inhibitory neurons [7] . LTS cells are mutually coupled by electrical synapses [10] , but inhibitory chemical synapses between them are only rarely observed [9] . Excitatory synapses from regular-spiking ( RS ) neurons onto LTS neurons show strong short-term facilitation [7] , [9] , [11] , [12] , [13] , whereas inhibitory synapses from LTS neurons onto to RS neurons usually depress [7] , [9] . LTS neurons are reciprocally coupled by depressing synapses to inhibitory neurons of the parvalbumin-expressing , fast-spiking ( FS ) type [10] , [14] . RS and FS neurons , but not LTS neurons in layer 4 , receive thalamic input [10] , [15] . There are conflicting data regarding the possibility that LTS neurons in other layers are innervated by thalamocortical axons ( see [15] , [16] ) . LTS neurons in layer 3 are excited by sensory inputs during whisking [17] ) , but these inputs could represent ascending layer 4-to-layer 3 excitation or neuromodulatory pathways . Because of the strongly facilitating nature of the RS-to-LTS excitatory synapses , rapid stimulation of a few RS neurons or , sometimes , even a single RS neuron can cause LTS neurons to fire spikes [13] . As a result , LTS neurons may mediate disynaptic inhibition between neocortical pyramidal neurons [18] , [19] , and simultaneous short bursts in four excitatory neurons are sufficient to exert disynaptic inhibition in all neighboring excitatory neurons [20] . When an RS neuron is stimulated and spikes repetitively , this disynaptic inhibition is delayed with respect to the stimulus initiation because RS-to-LTS synapses need time to facilitate before the LTS neuron can fire its own spikes . Based on their experimental results , Beierlein et al . [9] , Silberberg and Markram [18] and Kapfer et al . [19] hypothesized that LTS neurons are important for maintaining the balance between excitation and inhibition in the cortical circuit . Because the amount of excitation varies with the activity of neurons that are presynaptic to cortical neurons ( e . g . thalamic relay cells ) , maintaining this balance is a dynamic process in which LTS neurons may play an important role . For example , when the firing rate of excitatory neurons is high , facilitating excitatory input could generate a supralinear response of LTS neurons and thus prevent overactivation of excitatory neurons ( i . e . , activation beyond what is normal , leading to pathological behavior ) . This could protect the cortical network against seizures . Consistent with the idea that LTS cells serve a protective function is the observation that selective loss of somatostatin-positive dendritic-targeting interneurons ( cells similar to neocortical LTS neurons ) in hippocampus correlates with epileptic states [21] , [22] . More recently , it was suggested that LTS neurons balance excitation and prevent runaway cortical activity by decreasing the gain of pyramidal cell output [23] . The ability of LTS neurons to protect against network over-activation may be limited , however , by the depressive nature of LTS-to-RS inhibitory synapses . Furthermore , short-term synaptic plasticity can lead to firing patterns more complex than stable firing rates . The existence of two time-scales in the system dynamics — the fast time-scale of the AMPA receptor- and GABAA receptor-mediated postsynaptic potentials ( PSPs ) , and the slow time-scale of synaptic depression and facilitation processes — may , in principle , lead to various types of network oscillations or more complicated patterns . Such network oscillations were observed in previous models of excitatory and inhibitory neurons [24] , [25] , [26] , [27] , but those models did not take into account the specific physiological characteristics of LTS neurons . In this study we ask: by which mechanisms and at what firing rates do LTS neurons control the activity of cortical circuits responding to thalamic input , and how is control by LTS neurons different from that of FS neurons ? To be more specific , we compare the dynamical behavior of LTS neurons with those of FS neurons in networks with only one type of inhibitory interneuron and in networks with both inhibitory populations , to address the hypothesis of Beierlein et al . [9] , Silberberg and Markram [18] and Kapfer et al . [19] . We consider a rate model of cortical networks [28] , [29] , [30] that includes RS , LTS and FS neurons with short-term synaptic plasticity [31] , [32] , and study its responses to external inputs .
We consider a network of two populations , composed of RS and LTS neurons . To explore the role of RS-to-LTS and LTS-to-RS synapses , our first step is to study a model with these synaptic connections only , and the effect of the RS-to-RS synapses will be studied later . RS-to-LTS synapses facilitate ( τf , LR = 670 ms ) and LTS-to-RS synapses depress ( τr , RL = 1250 ms ) [18] ( see Methods and Table 2 ) . Therefore , xLR = 1 , uRL = URL , and equations 1–3 for the RS-LTS system become ( 7 ) ( 8 ) ( 9 ) ( 10 ) Absence seizures are a type of epilepsy that is considered to originate from the thalamus or at least to be driven by thalamic input [42] , [43] . Such seizures are characterized by periodic thalamic input to cortex with a frequency of about 3 Hz or somewhat higher [43] , [44] , [45] , and a duty cycle of the active phase of each thalamic cycle that is larger than 0 . 1 [46] . To investigate the response of the RS-LTS circuits to such thalamic inputs , we stimulate RS neurons by square-wave periodic input ( Figure 6A ) . Both RS and LTS neurons respond to the onset of each cycle by a brief elevation of their M followed by a deep decrease in activity and then more prolonged rebound . The integrated responses of MR and ML over a cycle ( Figure 6A ) increase with time towards their steady-state values , which are reached within about 1 sec . This behavior is similar to the evolution of MR and ML to step inputs ( Figure 3 ) . To characterize the properties of the steady-state response to the periodic , absence-seizure-like input , we define the time-averaged value , calculated for a large integration time Tinteg after the system has converged to an attractor . Similarly , we define . The values of <MR> and <ML> as functions of <IR> are shown in Figure 6B for two values of the duty cycle of the active phase , 0 . 1 and 0 . 5 ( note that the amplitude of IR during the active phase decreases with the duty cycle , to keep <IR> fixed ) . In both cases , the steady-state dependencies of <MR> and <ML> on <IR> resemble those of MR and ML on IR for constant stimuli ( Figure 2 ) . In particular , these curves become straight lines at high rates with slopes βR and βL respectively , and are shifted to the left by LTS-to-RS inhibition . As the duty cycle of the active phase of the input is reduced , the value IR , LTS , th in which LTS neurons start to fire decreases because the amplitude of the input during that active phase increases . We conclude that the RS-LTS circuit responds to absence-seizure inputs and constant thalamic inputs in qualitatively similar ways . To characterize the difference between the roles of FS and LTS neurons in the cortical circuit , we examine a network composed of RS and FS neurons . The RS-FS network is qualitatively different from the RS-LTS network in three respects [5] , [9] , [47] . First , RS-to-FS excitatory synaptic connections depress whereas RS-to-LTS connections facilitate . Second , FS neurons , but not LTS neurons , receive thalamic input . Third , FS neurons are mutually coupled by chemical synapses . We analyze the response of RS-FS circuits to constant and step inputs . Firing of excitatory neurons in cortex is controlled by inhibition from both LTS and FS interneurons , and we therefore characterize responses of the RS-LTS-FS network ( Figure 1 ) to external input that may reach the RS and FS populations . We start by describing the steady-state response of the circuit with the reference parameter set ( Table 2 ) to constant IR and IF , as summarized in the phase diagram in Figure 9 . The RS population is quiescent for small IR ( MR = 0 ) . It is active for larger IR values , and the behavioral regimes in the phase diagram are denoted by the inhibitory population ( s ) that is ( are ) silent . Just above the RS firing threshold , and when IF is small , both FS and LTS populations are quiescent ( ML = MF = 0 ) . For larger IR values and for small IF , LTS neurons fire and FS neurons are quiescent ( MF = 0 ) . For moderate values of IR and large values of IF , LTS neurons are quiescent and FS neurons fire ( ML = 0 ) . For large values of IR and IF , both populations of interneurons are active ( ML>0 , MF>0 ) . Between the last three regimes ( MF = 0 , ML = 0 , and ML>0 , MF>0 ) , there is a state of slow oscillations , on the time scale of short-term synaptic plasticity ( see below ) . This phase diagram remains qualitatively the same if the synaptic conductances are varied , except that fast oscillations , like those shown in Figure S4C , are observed for large gRR ( not shown ) . When thalamic input is varied , both IR and IF vary in a coordinated manner [15] , [36] . Therefore , we examine how the steady state firing rates of the neuronal populations vary with IR while keeping IF/IR fixed . When IF/IR = 1 . 4 ( Figure 10A; denoted by a dotted line in Figure 9 ) , the two inhibitory populations are quiescent just above the RS firing threshold . FS neurons start to fire for IR = 0 . 16 , and cause the RS firing rate to decrease . This decrease occurs because FS neurons receive independent input , IF , that increases with IR . As IR continues to increase , MR increases again because FS-to-RS synapses depress . For IR = 0 . 33 , LTS neurons start also to fire , and the RS gain decreases again before converging to βR for very large IR . When IF/IR = 0 . 75 ( Figure 10B; dashed line in Figure 9 ) LTS neurons start to fire for IR = 0 . 17 and reduce the RS gain , but do not make it negative because LTS neurons do not receive external input . Oscillations occur for 0 . 31<IR<0 . 34 . For just above IR = 0 . 34 , FS and LTS neurons fire at steady state and reduce the RS gain . This gain increases with IR and approaches βR for large IR . Similarly , the gain of FS neurons approaches βF . Note that the value of ML for IR values just above the oscillatory regime ( ) is smaller than its value just below this regime ( ) , because FS neurons fire and inhibit LTS neurons . At high rate , MR increases linearly with IR for all values of fixed IF/IR . This linear dependency is caused by LTS neurons only for low IF/IR and by both LTS and FS neurons if IF/IR is not low . Similar behavior is obtained for absence seizure thalamic input ( not shown ) . The control of seizures by the two inhibitory populations is therefore qualitatively the same . The dynamic response of three neuronal populations to step inputs IR and IF given at time t = 0 are presented in Figure 11A–D for four values of IR and IF . In all cases , RS and FS neuronal populations respond to the step initiation by a brief firing during a “window of opportunity” before settling slowly to an attractor . In Figure 11A , representing the steady-state regime “ML = 0” , those two populations increase slowly to their steady-state value after a rapidly-evolving initial response . In Figure 11B ( “MF = 0” in steady-state ) , RS and FS neurons are active during a time interval of a few tenths of seconds . Then , at about t = 0 . 35 s , LTS neurons start sharply to fire , whereas the activity of RS and FS neurons is reduced to non-zero and zero values respectively . In Figure 11C ( “MF>0 , ML>0” in steady-state ) , RS and FS are also active during an initial period of a few tenths of ms whereas the LTS neurons are silent . Here , however , the firing rate of LTS neurons increases continuously as they start to fire . The firing rates of the RS and FS neurons are reduced as a result of inhibition by LTS neurons , but both firing rates approach non-zero values at large times . The initial time courses of MR , MF , and ML in Figure 11D ( oscillations ) are similar to those in Figure 11B . At longer times , however , the time courses converge to an oscillatory state . Interestingly , the amplitude of LTS oscillations develops more gradually towards its steady-state value than the amplitudes of RS and FS oscillations . During the oscillatory state , RS neurons oscillate between a more-active phase and a less-active phase , where the firing rate in both phases is larger than zero . FS neurons fire episodes of spikes , represented by positive MF , when the RS neuronal population is in its more-active phase . They are quiescent when the RS neurons are less active . LTS neurons oscillate in opposite phase: they fire when RS neurons are in the less-active phase , and are quiescent otherwise . The oscillation frequency is on the order of a few Hz , corresponding to the time scale of short-term synaptic plasticity , and it increases as IR , and therefore IF , increases ( Figure 10B , top-right ) . The duty cycle of the more-active state is defined as the time that the RS population spends in that state ( and the FS neurons are active ) divided by the time period . This ratio varies from 0 . 2 to about 0 . 6 , and it first increases and then decreases with IR ( Figure 10B , bottom-right ) . We find that a slow oscillation state appears in our model only when it includes the two neuronal populations , whereas models of RS-LTS networks and RS-FS networks exhibit either rest states or , in restricted values of gRR , fast oscillations . What is the dynamical mechanism that leads to the slow oscillations state ? Such states are often studied using fast-slow analysis [50] , [51] , [52] , [53] , [54] , [55] . In our case , equations 1–6 for the RS-LTS-FS network ( Figure 1 ) include 8 slow variables , and it is practically impossible to analyze them . Fortunately , we find that slow oscillations still prevail in a reduced RS-LTS-FS circuit with only RS-to-LTS , LTS-to-RS , RS-to-FS and FS-to-LTS synaptic connections and without short-term plasticity properties of the depressing synapses , i . e . τr = 0 for all the synaptic connections ( Figure 12 ) . Facilitation of the RS-to-LTS synapses is , however , necessary to maintain the oscillations . The reduced system has only one slow variable , uLR , and all the other variables are much faster . We use the technique of fast-slow analysis ( See “Fast-slow analysis of slow network oscillations” in Methods ) to define the mechanism of slow oscillations . We find that in order for the slow oscillations to emerge , the fast subsystem that includes all the variables except uLR should be bistable ( Figure 13A ) . In one stable state , denoted “more active” , LTS neurons are silent and RS and FS neurons are active . In the second state , denoted “less active” , LTS neurons are active , RS neurons are active , but less than in the more active state , and FS neurons are silent . The dynamics of the full system switch rapidly back and forth between these two states . This explains the pattern of activation seen in the reduced RS-LTS-FS system ( Figure 12 ) as well as in the full system ( Figure 11D ) . Bistability ceases to exist if IF and IR are large enough , and in this case the system reaches a steady state in which both ML and MF are non-zero ( Figure 13 ) . Another condition needed to obtain oscillations is that the fixed point of the full dynamical system is not stable . This condition is broken if IF is too small , and then the system converges to a steady state where MF = 0 and ML>0 ( see “Borders of the regime of slow network oscillations in the phase diagram” in Methods ) . It is also broken if IR and IF are small and large enough respectively . In this case , the system converges to a steady state where MF>0 and ML = 0 . Qualitatively , this behavior is also shown by the original RS-LTS-FS network ( Figure 9 ) . Analysis of the reduced system also reveals that the oscillatory regime extends over a limited range of IF ( Figure 13B and Methods ) , as also found for the full model ( Figure 9 ) . The oscillatory regime of the reduced model extends over a larger IR range . This range is more limited in the case of the full model ( Figure 9 ) , probably because of the effects of synaptic depression .
Because of the facilitatory nature of RS-to-LTS connections , it was hypothesized that these neurons prevent overactivation and seizures by reducing cortical activity mostly at high rates [9] , [18] , [19] . It was also suggested that they do so by decreasing the gain of pyramidal cell output [23] . However , we find that the dynamical picture is different due to the LTS-to-RS synaptic depression . At high firing rates , LTS neurons do not change the RS gain at all , and reduce the firing rates of RS neurons by a constant value , independent of the input IR . Importantly , LTS neurons do reduce RS gain at modest firing rates , just above the LTS firing threshold , where LTS-to-RS depression is weak [38] . LTS neurons therefore have a divisive effect on the RS firing at modest rates and a subtractive effect at high rates . Their effect at high rates is therefore limited , because a divisive effect is more potent than a subtractive one during gradual elevations in cortical activity as it increases with the elevation of firing rates . Responses to absence-seizure-like inputs are qualitatively similar to the response to step inputs . Although RS-to-LTS synapses facilitate and RS-to-FS synapses depress , the two inhibitory populations reduce the firing rates of RS neurons in a similar manner at high rates . In response to input step currents , RS cells in all three networks ( RS-LTS , RS-FS and RS-LTS-FS ) respond with a brief firing epoch followed by reduced firing ( and even quiescence ) and then rebound to larger firing rates . This initial firing epoch terminates faster for FS neurons than for LTS neurons . An RS-LTS-FS network usually reaches a steady state with FS neurons quiescent for small IF , LTS neurons quiescent for small IR , and both populations active for large IR and IF . Between these behavioral regimes , there is a relatively narrow regime of slow ( few Hz ) oscillations . These oscillations are induced by the slow facilitation variable of the RS-to-LTS synapses that transfers the system alternately between two bistable states of the fast dynamics . During these oscillations , RS neurons switch from a more-active to a less active state alternately , whereas LTS and FS neurons switch alternately from an active state to a silent state . In general , FS neurons tend to follow the spiking of RS neurons closely , whereas LTS neurons follow it with delays . Inhibitory neurons can reduce the response of their targets by either shifting the target's response curve ( a subtractive effect ) or by reducing its gain ( a divisive effect ) . A simple biophysical model without synaptic depression predicts constant inhibition ( i . e . , independent of the activity of the target ) and causes a subtractive effect [29] , [56]; this result is confirmed experimentally [57] . If the activity of the inhibitory neurons is caused by the firing pattern of the excitatory target population , the effect is divisive ( Figure 2B , blue curve ) , whether the excitatory-to-inhibitory synapses facilitate or not . We show , using a rate model , that synaptic depression in the inhibitory-to-excitatory synapses exhibits similar divisive behavior at low rates , where depression effects are small . At high rates , depression causes the effect to be subtractive because the efficacy of the depressed inhibitory synapse scales as one divided by the firing frequency of its presynaptic inhibitory neuron . RS-LTS networks are different from RS-FS circuits primarily because of the facilitating nature of RS-to-LTS synapses versus the depressing nature of the RS-to-FS synapses , and because FS neurons receive strong external input [9] . As a result , tested independently of one another , the LTS and FS inhibitory populations have distinctly different effects on the input-output properties of cortical circuits that are demonstrated at steady states and low firing rates . LTS neurons do not affect the minimal input level IR above which RS neurons fire . Just above the LTS firing threshold IR . LTS , th , LTS neurons affect the RS gain most strongly , but reduce MR less strongly than at high rates . The value IR . LTS , th decreases with IL if LTS neurons receive their own thalamic input . In contrast , FS neurons , which receive substantial external input , increase the current threshold IR for RS firing , do not considerably affect the RS gain , and reduce MR effectively starting from just above this threshold . The effect of FS neurons on RS firing is therefore always subtractive . At high firing rates , both the LTS and the FS neuronal populations affect the RS firing properties in a similar manner by reducing the firing rate of RS neurons by a constant value . The reasons for this behavior , however , are different: the LTS input to RS neurons reaches a constant value at high rates because of the LTS-to-RS synaptic depression ( Figure 2 ) , whereas FS input to RS neurons is limited by the saturation of the firing rate of FS neurons themselves ( Figure 7 ) . In response to step input currents , LTS neurons respond with a delay just above IR , LTS , th ( Figure 3 ) . After the delay , LTS neurons decrease the activity of RS neurons to a minimal value , after which MR rebounds . FS neurons reduce the activity of RS neurons much more rapidly after a stimulus onset , leaving only a brief “window of opportunity” for RS initial firing ( Figure 8 ) . The RS activity then decreases to low ( even zero ) values before rebounding to its steady-state values . Interestingly , the temporal profiles of MR in the RS-LTS network with large IR and RS-FS networks are similar ( Figures 3A and 8 ) , except that the initial decay of MR in the RS-LTS network is more gradual . The temporal profiles of the activity of the two inhibitory neurons in these networks are , however , different: FS neurons , but not LTS neurons , respond with brief initial activity to the step input activity . Strong RS-to-RS connections may induce fast oscillations in both circuits ( Figures S3 , S4 ) [28] , [58] . In general , FS cells tend to track spiking of the RS cells much more closely than the LTS cells do . This behavior is seen by comparing RS-FS and RS-LTS circuits ( Figures 3 , 8 ) as well as in RS-LTS-FS circuits ( Figures 11 , 12 ) . FS and LTS neurons behave dynamically quite differently from one another . At steady state , cortical networks with active RS neurons show four types of resting states in which: only LTS neurons are active , only FS neurons are active , both interneuron populations are active , or neither is active ( Figure 9 ) . The oscillatory regime is located near the intersection of all these states . Despite the fact that it is narrower than the other states , analysis of its existence determines the structure of the other states . If the fast subsystem of variables ceases to be bistable as a parameter is varied , a state with active LTS and FS neurons is obtained . If the fast subsystem is bistable and a rest state of the full subsystem occurs on a branch with FS ( respectively LTS ) neurons quiescent , a state with a quiescent FS ( respectively LTS ) population is obtained . We show this theoretically in a reduced circuit ( Figure 13 ) and numerically in the full circuit ( Figure 9 ) . In the parameter regimes when LTS neurons are active in the steady state , the initial response to step currents is similar to that in the oscillatory regime ( Figure 11 ) . RS and FS neurons are active in the initial several tenths of one second while LTS neurons are silent . Then , LTS neurons start to fire and reduce the firing rate of FS neurons . When thalamic input varies , it is expected that IR and IF will vary proportionally [36] . Increasing the input can therefore cause non-monotonic variation of the firing rate of one of the neuronal populations , with or without passing through the oscillatory regime ( Figure 10 ) . In RS-LTS-FS circuits , as in circuits with one population of interneurons only , the gain of RS and FS neurons at high rates is not affected by the circuit . Our cortical circuit model exhibits two types of cortical oscillations . Large gRR may generate fast oscillations , as was shown in previous models of cortical circuits [28] , [59] . One type of inhibitory interneuron , either LTS or FS , is sufficient for the generation of fast oscillations , together with large ( but not extremely large ) values of gRR . The oscillation frequency is on the scale of 1/τs , about tens of Hz . Excitatory and inhibitory neurons fire nearly in phase ( Figure S4C ) , and there is a substantial time interval in each period during which both neuronal populations are quiescent . In this study , we discovered a novel type of oscillation in cortical networks that depends on RS-to-LTS synaptic facilitation and on external input to the FS neurons , and can occur without any RS-to-RS recurrent excitation . These oscillations have several characteristics . Both LTS and FS populations are necessary for generating them . The oscillation frequency , ∼1–10 Hz , is on the time scale of 1/τf , LR , the facilitation recovery time constant . RS neurons oscillate between more-active and less-active states , both with positive firing rates . FS and LTS neurons fire in phase and in anti-phase with the RS more-active state , respectively . Hence , in contrast to the fast oscillations , at least one population of neurons is active at every time point . Slow cortical oscillations have been observed during sleep , anesthesia and quiet wakefulness in vivo [60] , [61] , [62] , [63] , in vitro [64] and in computational models [26] , [65] . During these oscillations , the neurons in the network switch from an active “up” state to a silent “down” state and back . The oscillations we observe in the RS-LTS-FS model are different from those oscillations because the RS neurons during the less-active state are not silent , and because the LTS neurons fire during the less-active state . Cortical oscillations with a frequency on the order of 1 Hz , during which the network is not completely silent during the less-active state , have also been observed [66] , [67] , and spontaneous activity was observed during which neurons fired in episodes with similar frequencies [68] . Using future recordings from LTS and FS neurons in vivo [17] , or using optogenetics techniques to activate RS or FS populations selectively [15] , it will be possible to determine whether LTS neurons are active during the less-active state of the RS populations , as the theory predicts . Interestingly , the frequency range ( ∼1–10 Hz ) of the slow oscillation observed in our RS-FS-LTS model overlaps with that of absence seizures and both the tonic and clonic phases of tonic-clonic seizures [44] . While other mechanisms may contribute to these seizure components ( e . g . rhythmic thalamic input in absence seizures ) , the oscillatory pattern observed in our model could conceivably perpetuate or reinforce such pathological conditions . It remains to be seen whether FS and LTS cells alternate their firing during these conditions , as suggested by our results . Each neuronal population is represented in our model by its firing rate . Rate models can describe the properties of large networks of neurons represented by conductance-based schemes provided that the level of synchrony in the network is small , and the input is stationary or slowly modulating in time [29] . The level of synchrony in cortical networks , especially in awake animals , is often small [69] , [70] . Therefore , our rate model is expected to describe the dynamics of cortical networks that receive stationary input reasonably well in comparison to more complicated models of spiking neurons . In addition , we examine the response of networks to step or absence-seizure-like inputs . In such cases , the outcome of rate models should be regarded as a qualitative estimation of the full dynamics . In particular , neurons often show sharply transient responses to step inputs when FS neurons play a major role in the dynamics . Using rate models we can claim that such a response occurs , but cannot determine its properties on time scales of milliseconds . Our basic form of the model does not include spike-frequency adaptation and firing-rate saturation . Adaptation does not change the steady-state response of the circuits . Dynamically , with the slope of the f-I curve scaled to be equal with and without adaptation ( Equation 19 ) , a model with adaptation exhibits a stronger initial response to step inputs , whereas its subsequent long-term response is similar to that of the model without adaptation ( Figure 5 ) . Saturation reduces the activity at high rates but does not change the qualitative effects of LTS and FS inhibition on the cortical circuit . LTS neurons project mostly to distal dendrites of pyramidal neurons [4] , [5] , but their inhibitory effects are clearly observed in the soma [8] , [18] , [19] , [20] . Such effects can be described by the rate model presented here , which is based on linear summation of inhibitory PSPs in the soma [29] . Developing more elaborate rate models , that can account for spatial properties of neurons and describe LTS effects on local dendritic computation [71] , remains a challenge . We use the fast-slow analysis to determine the conditions for obtaining slow oscillations . This analysis is often used when one or several time constants in the system are much larger than the other time constants [51] , [52] . We apply the method to our reduced circuit ( Figure 13 ) with no synaptic depression , by assuming that both τf , LR is large and ULR is small . These approximations yield good fits of the predictions of the fast-slow analysis ( Figure 13 ) to the full dynamics of the reduced system , computed using numerical simulations ( Figure 12 ) . The phase diagram ( Figure 13B ) of the reduced circuit is qualitatively similar to that of the full circuit ( Figure 9 ) and displays the same behavioral regimes , but the locations of the borders between the regimes in the phase diagrams of the two circuits are quantitatively different . Most models of the response of cortical circuits ( e . g . , [28] , [58] , [59] , [72] ) to input do not consider short-term synaptic plasticity . Like our model , these models can show fast oscillations as a result of interactions between excitatory and inhibitory neurons . The contribution of LTS neurons was shown to shape the response of cortical circuits to periodic inputs [73] in a model with short-term synaptic plasticity of excitatory synapses but without considering depression of inhibitory synapses . While our model may exhibit slow oscillations with facilitation of the RS-to-LTS synapses and depression of all other synaptic connections ( Figure 11D ) , depression is not necessary for obtaining oscillations ( Figure 12 ) . In contrast , depression is essential for various slow oscillations in other models of cortical networks [24] , [25] , [26] . Facilitation of the excitatory-to-inhibitory synapses generates slow oscillations in a rate model of cortical circuits composed of excitatory and inhibitory populations [27] . Inhibitory neurons in that model receive external input and are mutually coupled by inhibitory synapses . In our model , inhibitory LTS neurons receive facilitating input from excitatory RS neurons , but do not receive external input and are not mutually coupled , according to circuit properties discovered experimentally [9] . Excitatory and inhibitory populations in the model of Melamed et al . [27] fire during the same phase interval during the cycle , whereas LTS and RS neurons in our model fire in anti-phase . Another difference is that the firing rate of excitatory neurons during the “down” state in the Melamed et al . model is zero , whereas the firing rate of the RS neurons in our model during the less-active state is positive . Roles of specific types of interneurons in diseases such as epilepsy [21] , [22] and schizophrenia [74] have been suggested . By analyzing a rate model of cortical circuits with Tsodyks-Markram kinetics for short-term synaptic plasticity , we observe that in response to high input IR , the LTS population reduces the firing rate of the RS neurons by a constant factor , independent of IR . We demonstrate this behavior specifically for a model with absence-seizure-like input . This implies that LTS neurons can help to prevent seizures in cortex , but the role of LTS cells in this task is qualitatively as limited as that of FS neurons . Indeed , selective damage to the LTS neurons ( for which there is evidence in experimental seizure models and human cortex ) may be compensated by FS neurons or by other types of inhibitory interneurons such as neurogliaform cells [23] . Our results are consistent with experimental results showing that the death of LTS interneurons does not initiate hyperexcitability in a neonatal rat model of human polymicrogyria , which is often characterized by severe seizures [75] . Whereas most of our calculations are carried out for constant or step stimuli , our results are applicable also for pulsatile thalamic input ( Figure 6 ) , at least above 3 Hz . Increasing the frequency will make the approximations of the model even more accurate . During whisking , cortical circuits receive periodic thalamic input at frequencies of about 10 Hz [76] . Similarly , visual thalamic input to cortex is often described as Poisson firing , with firing rates of about 20 Hz [77] . Since the time constants of synaptic depression and facilitation are much longer , the slow dynamics will average over the spiking process and will depend on the underlying firing rate , similar to the response to constant or slowly-varying stimuli . Therefore , our finding that LTS neurons have a strong impact on the response to modest thalamic input , and not just during high frequency activity , are valid also for the cortical response to somatosensory and visual stimuli . Our conclusion is an outcome of the depression kinetics of the Tsodyks-Markram model , where the total synaptic input reaches a saturating value as the presynaptic firing rate , M , increases . Saturation occurs because the additional postsynaptic conductance in response to one additional presynaptic spike scales as 1/M [38] . In various other types of depressing synapses characterized experimentally and using models , the response to an additional spike is larger than expected by the Tsodyks-Markram model , probably because the recovery from depression is faster at high presynaptic rates [78] , [79] , [80] . One reason we use the Tsodyks-Markram model in this work is that Silberberg and Markram fit their data for RS-to-LTS and LTS-to-RS synapses to it [18] . The theoretical results , however , suggest that the kinetics of these synapses in a broad frequency range should be measured in a more detailed manner . In this work , we observe that LTS neurons affect the gain of RS neurons at rates on the order of 10 Hz and less . These rates are comparable with the rates of LTS neurons measured in vitro during a variety of diverse activating conditions [7] , such as group I metabotropic glutamate or muscarinic cholinergic receptor agonists . Therefore , LTS neurons can affect cortical dynamics even if cortical neurons do not fire at high rates .
The parameters of the neuronal populations are provided in Table 1 . They were determined based on the experimental observations of Fanselow et al . [7] . The parameters of the synaptic connections are written in Table 2 . These parameters are used in all calculations unless otherwise stated . The parameters determining the short-term synaptic plasticity properties of LTS-to-RS and RS-to-LTS synapses are taken from Silberberg and Markram [18] who carried out experiments in layer 5 . This layer is most active in the initiation [81] and horizontal propagation of epileptiform [82] and normal [64] activity in the neocortex . Short-term plasticity parameters for the RS-to-RS synapses are taken from [83] , and those for the FS-to-RS and RS-to-FS are taken from [47] , [84] . We are not aware of any systematic research on the short-term synaptic plasticity properties of FS-to-FS , LTS-to-FS and FS-to-LTS connections , except that these synapses depress [85] . Therefore , we use the generic values τr = 400 ms and U = 0 . 3 . To simplify the analysis , we assume that τr = 0 if τr<<τf and τf = 0 if τf<<τr . The constant τs is taken for AMPA and fast GABAA excitation , and it is larger for the LTS-to-RS synapses than for the FS-to-RS synapses [9] , [18] . LTS neurons fire if sLR>θL/gLR ( Equation 14 ) . Using Equation 11 , we find that LTS neurons fire for MR>MR , th , where ( 22 ) The rate MR , th is obtained for the LTS firing threshold IR = IR , th , where ( 23 ) From Equation 12 , sRL≈URL τs , RL ML for ML<<1 . Using Equations 13 , 14 we find that just above IR , th , ( 24 ) Differentiating both sides of Equation 24 with respect to , we obtain that the RS gain , dMR/dIR , is ( 25 ) During the delay period , MR = βR ( IR−θR ) . From Equation 8 , ( 26 ) where ( 27 ) Since τf , LR>>τs , LR , sLR reaches a quasi-steady-state value , sLR = τs , LR uLR MR ( Equation 7 ) . LTS neurons start to fire when gLR sLR = θL , i . e . , when uLR reaches the value uLR , th = θL/ ( gLR τs , LR MR ) . From Equation 26 , the delay time is ( 28 ) The reduced RS-LTS-FS dynamical system ( Figure 12 ) has five dynamical variables . The four variables sLR , sRL , sFR , sLF , are fast , with τs , on the order of a few ms ( Table 2 ) . The fifth equation ( Equation 3 ) , describing the facilitation process of the RS-to-LTS synapses , is ( 29 ) We use the method of fast-slow analysis to describe the dynamics of the system for both large τf , LR and small ULR . Formally , we define C≡ULR τf , LR and study the system in the limit τf , LR→∞ and constant C . This approximation is expected to be justified for the RS-to-LTS synapses because τf , LR , 670 ms , is two order of magnitude larger than the τs's , and ULR , 0 . 09 , is an order of magnitude smaller than 1 . Using the definition of C and neglecting a term on the order of , Equation 29 becomes ( 30 ) The full dynamical system describing the network can be separated into a fast subsystem , composed of the four equations for the variables s , and a slow subsystem , that includes the variable uLR . The first step in this method is to study how the attractors of the dynamics of the fast subsystem depend on the value of uLR , taken as a time-independent parameter . In a second step , one derives the dynamics of the full system taking into account the slow variations of uLR ( Equation 30 ) . The bifurcation diagram of the fast subsystem as a function of uLR for the parameter set of Figure 12 is plotted in Figure 13A . The subsystem can settle into stable fixed points that belong to one of two branches . The upper branch is characterized by ML = 0 , MF>0 , and a high value of MR , denoted by , that does not depend on uLR . This branch exists for small uLR values and disappears for uLR = u+ at a saddle-node bifurcation [50] , where it coalesces with an unstable branch ( not shown ) . The lower branch is characterized by MF = 0 , ML>0 and a low value of MR , denoted by , that depends on uLR . This branch exists for large uLR values and disappears for uLR = u− at a second saddle-node bifurcation . The slow nullcline of Equation 30 , characterized by uLR = CMR/ ( 1+CMR ) , does not intersect with either of the stable branches . Therefore , the full system does not have any stable fixed point . Instead , it exhibits relaxation-oscillation dynamics [50] . The system converges rapidly to one of the two stable branches of the fast subsystem . If it converges to the upper branch , it will then progress slowly to the “knee” at uLR = u+ and then will move rapidly to the lower branch . On that branch , the system progresses slowly to uLR = u− and then moves rapidly to the upper branch , completing the oscillatory cycle . The trajectory of the full dynamical system with the reference parameter set that is overlaid on the bifurcation diagram in Figure 13A fits this bifurcation picture very well . This fit shows that analysis in the limit τf , LR→∞ and constant C describes well the dynamics with biologically realistic parameters . The fast-slow analysis yields three conditions that together are both necessary and sufficient for the generation of slow oscillations: We calculate u+ ( resp . u− ) , the value of uLR above ( resp . below ) where the upper ( resp . lower ) branch of the fixed points of the fast subsystem no longer exists ( Figure 13A ) . We define , and . From Equation 1 , at a steady state of the fast subsystem , ( 33 ) LTS neurons fire above u+ . According to Equation 5 , at the onset of LTS firing ( ML = 0+ ) , ( 34 ) Substituting Equation 33 in Equation 34 , we obtain for uLR = u+ , ( 35 ) Using Equations 4 , 6 and 33 , and because FS neurons are active and LTS neurons are silent on the upper branch , we obtain ( 36 ) ( 37 ) To calculate u− , the value of uLR below which the lower branch of the fixed points of the fast subsystem no longer exists ( Figure 13A ) , we note that FS neurons fire below this value . According to Equation 6 , at the onset of FS firing ( MF = 0+ ) , ( 38 ) Substituting Equation 33 in Equation 38 , we obtain for uLR = u− , ( 39 ) From Equations 4 , 5 and 33 , and because LTS neurons are active and FS neurons are silent on the upper branch , we obtain ( 40 ) ( 41 ) The parameter regime that fulfills the three conditions written above ( u+>u− and Equations 31–32 , computed using Equations 35–37 , 39–41 ) is denoted in a phase diagram in the IR–IF plane ( Figure 13B ) . Slow oscillations are observed for levels of IR that are not too small and levels of IF within a certain narrow range . This range is always below θF , such that excitation from RS neuron is needed to induce firing in the FS neurons . Above a certain value of IR ( 0 . 32 in Figure 13B ) , this IF range has an ( almost ) constant width , and its borders decrease ( almost ) linearly with IR . Outside of the oscillatory regime , the network reaches a steady state . For large IR and small IF , MF = 0 and ML>0 . For large IR and IF , MF>0 and ML>0 . For large IR and medium values of IF , MF>0 and ML = 0 . Finally , for small IR and IF , the two inhibitory neuronal populations are quiescent . Simulations were performed using the fourth-order Runge-Kutta method with a time step of 0 . 02 ms implemented as a C program or within the software package XPPAUT [86] , which was used also for computing the bifurcations of fixed points in the diagram in Figures S4A .
|
The brain consists of circuits of neurons that signal to one another via synapses . There are two classes of neurons: excitatory cells , which cause other neurons to become more active , and inhibitory neurons , which cause other neurons to become less active . It is thought that the activity of excitatory neurons is kept in check largely by inhibitory neurons; when such an inhibitory “brake” fails , a seizure can result . Inhibitory neurons of the low-threshold spiking ( LTS ) subtype can potentially fulfill this braking , or anticonvulsant , role because the synaptic input to these neurons facilitates , i . e . , those neurons are active when excitatory neurons are strongly active . Using a computational model we show that , because the synaptic output of LTS neurons onto excitatory neurons depresses ( decreases with activity ) , the ability of LTS neurons to prevent strong cortical activity and seizures is not qualitatively larger than that of inhibitory neurons of another subtype , the fast-spiking ( FS ) cells . Furthermore , short-term ( ∼one second ) changes in the strength of synapses to and from LTS interneurons allow them to shape the behavior of cortical circuits even at modest rates of activity , and an RS-LTS-FS circuit is capable of producing slow oscillations , on the time scale of these short-term changes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"circuit",
"models",
"computational",
"neuroscience",
"biology",
"neuroscience"
] |
2011
|
LTS and FS Inhibitory Interneurons, Short-Term Synaptic Plasticity, and Cortical Circuit Dynamics
|
Rabies , a fatal but preventable zoonosis , is a major public health problem in developing countries . In Cambodia the disease burden is largely underestimated because patients with encephalitis following dog bites are rarely hospitalized and die at home . Since 1998 Institut Pasteur in Cambodia ( IPC ) , Phnom Penh has been the only source of free post-exposure prophylaxis ( PEP ) and post-mortem diagnosis . The 1998–2007 data compiled by IPC was analyzed to describe all treated patients for PEP , results of human testing and confirmed rabies cases , and results of animal testing . From dog bites' characteristics , we defined a suspected rabid dog bite injury ( SRDBI ) in humans as a bite that was unprovoked , from a dog that died spontaneously , or from a dog that was reported sick . We applied a deterministic probability model to estimate 2007 rabies human mortality nationwide from the estimated incidence of rabid dog bites , the body distribution of bite wounds , and the probability of PEP access . During 1998–2007 , 124 , 749 patients received PEP at IPC ( average 12 , 470; range 8 , 907–14 , 475 ) , and 63 fatal human cases presenting with encephalitis following a dog bite were reported , in which 73% were confirmed positive for rabies by direct immunofluorescence assay or by reverse-transcriptase polymerase chain reaction . During 1998–2007 , IPC tested 1 , 255 animal brain samples; 1 , 214 ( 97% ) were from dogs including 610 ( 49% ) positive samples . In 2007 , 14 , 475 patients received PEP ( 100 PEP/100 , 000 people in Cambodia ) including 95% who resided in Phnom Penh ( 615 PEP/100 , 000 ) or five neighboring provinces . The predictive model estimated 810 human rabies deaths would occur in 2007 ( 95%confidence interval [CI] 394–1 , 607 ) , an incidence of 5 . 8/100 , 000 ( 95% CI 2 . 8–11 . 5 ) . Access to PEP is only sufficient for Phnom Penh residents . In 2007 , the estimated rabies related mortality exceeded that of malaria and that of dengue . A national rabies control program is needed to improve surveillance and access to PEP , and to initiate vaccination campaigns in dogs .
Rabies is a viral zoonotic infection of the nervous central system caused by a lyssavirus and is fatal without proper post exposure treatment [1] . Despite the existence of an effective vaccine , rabies remains a public health problem worldwide , particularly in developing countries where dogs continue to serve as the main reservoir of disease transmission to humans [2] , [3] . Globally , animal bite injuries lead to >10 million post exposure treatments per year and an estimated 55 , 000 people die of rabies each year . However , the number of deaths caused by rabies is considered largely underestimated [4] . Cambodia ( estimated 2007 population 14 . 4 million ) is a canine rabies endemic country where data are lacking to properly quantify the burden of the disease in animals and humans [5] . Since 1998 the Institut Pasteur in Cambodia ( IPC ) in the capital city of Phnom Penh ( estimated 2007 population 1 . 4 million ) has been the only source of free post exposure prophylaxis ( PEP ) and for human and animal rabies laboratory diagnosis . Unfortunately IPC does not have the capacity to take care of patients who developed rabies like symptoms . They are referred to the Calmette hospital which is situated next door . A joint collaboration between the two institutions has led to systematic reports of suspected rabid patients and the possibility for Calmette hospital's clinicians to send samples for laboratory diagnosis of rabies . In absence of national rabies control program and as surveillance of rabies related encephalitis was only phased in throughout the country in 2006–2007 , IPC has been the solely reliable source of information on human cases collected in Cambodia . We report the results of retrospective analysis of data from IPC's routine activities during 1998–2007 on ( i ) all post exposure treated patients , ( ii ) human rabies cases reported by Calmette hospital and ( iii ) human and animal specimens testing to describe the epidemiological situation of rabies in Cambodia . Information on dog abundance and rabies vaccination coverage were estimated from several surveys conducted in rural areas in 2005–2007 . Finally , we used the 2007 IPC data to estimate the incidence of rabies related human deaths in Cambodia from passively reported dog bite injuries statistics .
The 1998–2007 data compiled by IPC were analyzed to describe all treated patients , confirmed human rabies cases and laboratory results of human and animal testing . Population estimates were obtained from the 1998 national census accounting for a 1 . 81% annual population growth [5] . The 2007 number of human deaths caused by rabies [N] was calculated as the product of the incidence of suspected rabid dog bite injuries ( SRDBI ) [I] , the population at risk [Pop . ] and the probability of death following a bite from a SRDBI [Pdeath]:Relying on data collected at the PEP center , we defined a SRDBI as unprovoked biting or from a dog that died spontaneously or from a biting dog that was reported sick . To estimate I , the annual incidence of SRDBI for the country , we calculated the SRDBI incidence for Phnom Penh and extrapolated to the national level assuming that the IPC PEP was able to draw all suspected rabid dog bite injuries that occurred in Phnom Penh . We defined “Pop . ”- human population at risk for rabies as a population in which density of dogs was beyond the density threshold to sustain rabies transmission in dogs ( 9 per km2 by Knobel et al ) ( 4 ) . Density of dogs was calculated from the estimated dog∶ human ratio and human population density which derived from the Cambodian National Bureau of Statistics [5] . Finally we adapted the step by step probability model as described by Cleveland et al [14] to estimate P death , the probability of dying following a SRDBI ( Figure 1 ) . This model was based on five input parameters which were determined from Cambodian data: ( i ) proportion of signs that composed SRDBI; ( ii ) probability of confirmed rabies among suspected rabid dog , which is obtained from laboratory data by comparing positive brain dog specimens with negative ones and calculating the predictive positive values of each of the SRDBI signs when reported alone; ( iii ) distribution of single and multiple dog bite injuries on human body accounting for single and multiple locations; ( iv ) probability of developing rabies by body location of bite wounds - we used existing data on estimated likelihood of a person to develop rabies by body location of the bites accounting for rabies probabilities for multiple bite locations; and finally ( v ) the probability of PEP access which stem from the annual number of PEP center attendees over the population at risk , assuming that PEP was only provided by IPC . We believed the number of people who could afford vaccines provided by the private sector was negligible ( Figure 1 ) . We estimated the confidence limits for the total number of deaths attributable to rabies by bundling probability distributions of input parameters and running Monte Carlo simulations using STATA software version 9 . 0 ( StataCorp , College Station , TX , USA ) for 1 , 000 iterations . The value of each input parameter were chosen at random from within a defined probability distribution and the simulation program produced a probability-based distribution of the net result , which were used to report statistics such as mean and the 2 . 5th and 97 . 5th percentiles for 95% confidence intervals ( 95% CI ) . Proportions , means , medians , ratios , interquartile ranges ( IQR ) , non-parametric and chi2 tests , p values and 95% CI were calculated using STATA .
During 1998–2007 , 124 , 749 patients attended the IPC's PEP clinic because of an animal bite injury ( 99 . 1% ) or animal's licking ( 0 . 2% ) and/or scratches ( 0 . 7% ) . Of the patients with bite injuries , 9 . 0% presented with deep wounds and 49 . 0% with multiple bites . The number of patients steadily increased during this period from 8 , 485 to 14 , 475 in 1998 and 2007 respectively . In 2007 , the median age of the patients was 16 years ( range 1–92 ) with 51 . 8% males . The overall Cambodia PEP rate was 101/100 , 000 this year and varied by province from 0 . 9 to 615/100 , 000 with a median rate of 7 . 2/100 , 000 . The highest rates were observed in Phnom Penh and the five neighboring provinces ( Figure 2 . A ) which accounted for 59 . 5% and 36 . 7% of the total number of cases , respectively . Among the attendees in 2007 , 95 . 7% reported dog bite injuries of which 10 , 437 ( 75 . 3% ) met the case definition of SRDBI . Of these SRDBI , 95% of them consulted the IPC clinic within 3 days of a bite injury ( median duration 1 day; range 0–112 days ) . Adherence to prophylaxis is high; 95% of patients come back at D7 of the treatment – particularly among Phnom Penh residents compared with patients living beyond Phnom Penh ( 95 . 9% versus 94 . 4% , p<0 . 001 ) . Only 8 . 7% showed up at D28 for the fourth injection as these patients were those for whom the biting dog was killed , missing or tested positive at IPC . During 1998–2007 , among 63 patients who were admitted to the Calmette hospital with encephalitis following a dog bite ( mean 7 per year , range 0–18 patients ) , none had a reported history of vaccination against rabies . Of these , 60 had their biological samples collected and tested at IPC; 44 ( 73% ) tested positive for rabies , either by reverse-transcriptase polymerase chain reaction ( 57% , 20/35 ) of skin , brain , saliva , urine or cerebrospinal fluid , and/or by direct immunofluorescence assay on brain samples ( 96% , 24/25 ) . Among the 44 positive cases , 37% were 15 years old or younger ( median age 26 years; range , 6–67 ) and 28 ( 63% ) were males . Thirty-eight ( 87% ) patients resided within 200 km from IPC ( Figure 2 . A ) including 13 ( 20 . 6% ) and 4 ( 9% ) in Kandal province and Phnom Penh respectively . The last case who reported to have been bitten by a dog in Phnom Penh city dated from 2006 . The reported median incubation period calculated from 35 patients was 60 days ( IQR range 30 to 100 days ) . Symptoms that were observed among rabid patients included acute behavior changes or/and hyperactivity/excitability ( 98% ) , hydrophobia ( 86% ) and aerophobia ( 55% ) . Of the 15 patients presenting with encephalitis but negative for rabies , these symptoms were seen in 87% , 53% and 47% , respectively . Patients were admitted to the hospital approximately 2 days on average ( range 1–8 ) of the onset of symptoms . From 1998 through 2007 , the IPC laboratories received 1 , 255 animal heads; dog heads made 96 . 7% of the animals of which 49 . 2% tested positive for rabies antigens in brain tissue . Other animals included cats ( 17 . 6% positivity ) , bovines ( 90 . 9% ) and monkeys ( 12 . 5% ) . Except for bovines , all other heads stem from biting animals . Of the 596 laboratory confirmed rabid dogs , 67% were male and the median age was 18 months ranging from 2–120 months . Rabid dogs were confirmed in 17 Cambodia provinces; however , 95% of them were located within 200 km from Phnom Penh including 44 . 5% which lived in Phnom Penh or Kandal provinces ( Figure 2 . B ) . In 2007 , 109 ( 51% ) of 214 dogs tested positive and originated from 14 provinces . When comparing the characteristics of the rabid dogs with positive brain specimens with the negative specimens in 2007 , the positive predictive value was highest for reported sick biting dogs ( 93% , 95%CI 66%–100% ) , followed by dogs that died spontaneously 17% ( 95%CI 1%–64% ) and for biting from an unprovoked aggression ( 12% , 95%CI 6–22% ) ( Table 1 ) . Of the 1 , 538 households of 151 villages in seven rural provinces that were surveyed during 2005–2007 , 75% of the households owned at least one dog . A total of 2 , 670 owned dogs were recorded for 8 , 269 individuals surveyed yielding a ratio of 1 dog to 3 . 1 humans ( 95%CI 1∶3 . 0–1∶3 . 2 ) . Only 17 ( 1 . 4% ) dogs were reported to have been vaccinated against rabies . No significant differences in the ratios were observed between the seven provinces . In 2007 , of the 8 , 606 patients who attended the IPC's PEP rabies clinic and resided in Phnom Penh , 5 , 398 ( 62 . 7% ) were bitten by a dog and had an injury that met the case definition of a SRDBI . These reported SRDBI included inflicted by sick dogs ( 0 . 4% ) , by dogs that died spontaneously ( 1 . 0% ) or by unprovoked dog aggression ( 99 . 3% ) . Of these , most bites were located on lower limbs ( 59 . 3% ) followed by upper limbs ( 22 . 8% ) , trunk ( 10% ) and the head 5 . 7%; multiple biting ( >1 wound ) accounted for 93 . 2% . Assuming that the IPC clinic was able to capture a large majority of SRDBI that occurred in Phnom Penh , the Phnom Penh incidence of SRDBI would be at least 386 per 100 , 000 ( 95% CI: 374–394/100 , 000 ) . When extrapolating this incidence of Phnom Penh to the entire country , 53 , 732 injuries at risk for rabies ( SRDBI ) would have occurred in Cambodia . As a consequence , the model yielded a probability of rabies-related death from a suspected rabid dog bite of 1 . 51% ( 95%CI: 0 . 76–2 . 93% ) which resulted in 810 ( 95%CI: 394–1 , 607 ) deaths and a rabies incidence rate at 5 . 8/100 , 000 ( 95%CI: 2 . 8–11 . 5/100 , 000 ) for 2007 in Cambodia ( Table 2 ) .
This report is the first published national data on rabies in Cambodia and confirms that rabies remains a serious public health hazard where dog bites continue to be the main source of transmission . Clearly rabies transmission has also occurred in Phnom Penh capital city and its vicinity even though the population is more aware of the disease ( Cambodia Ministry of Health/UNICEF's unpublished data ) . Over 95% of the PEP patients come from Phnom Penh and five surrounding provinces . PEP coverage may only be satisfactory for Phnom Penh residents with rates that are similar to that of Vietnam or Thailand whereas access rates to PEP dropped out as distance from IPC increases [15] , [16] . It is likely this pattern only reflects a catchment's area of IPC instead of the true geographic distribution of rabies in Cambodia . This high attendance among Phnom Penh population combined with high adherence to treatment showed rabies still elicits fear in Phnom Penh . This may be probably true also for the rest of the country as a local name for rabies known even among children ( “mad dog” disease ) exists and is a reflection of a long history of the disease among Cambodians . Reasons for low attendance among people residing in remote provinces should be explored . Whether these people were less knowledgeable about a free PEP provided by IPC remained unclear . It is possible that despite free treatment in Phnom Penh , a series of rabies PEP injections with day intervals could deter poor people living in remote and rural areas from back-and-forth travels to IPC . This finding is of utmost importance to the health authorities as to underscore the needs for Cambodia to have more than one post exposure treatment center . Further investigations are needed to identify areas in which a second or a third treatment center would be appropriate . Our second major finding was a remarkably large population of dogs compared to that of humans; we found 1 dog for approximately every 3 humans , a ratio that was 3–4 times higher than that of neighboring countries . To our knowledge , worldwide this magnitude is second to the highest ratio observed in Sri Lanka with a ratio of 1∶2 . 4 . We estimated that the dog population at 4 . 4 million considering an 84% dog ownership [17] and 80% of Cambodians living in rural areas [5] . Commonly the dog ∶ human ratio is lower in urban areas [4] . If we assumed this ratio to be similar to that of Thailand ( two-fold lower in urban areas compared to rural areas ) , the total number of dogs in Cambodia could be up to 5 million . The abundance of dogs in Cambodia was not a surprise and likely to be explained by several factors: ( i ) as a Buddhist country , Cambodians are reluctant to kill or eat dogs; ( ii ) dogs are popular as they are useful for guarding houses and ( iii ) birth control in dogs is rarely available , particularly in rural areas . Taken together with a potentially rapid growing dog population and an inexistent canine vaccination against rabies , rabies situation in the country could become alarming . On the other hand , despite this large dog population , a relatively high frequency of dog ownership described in this country could allow a substantial proportion of dogs be accessible for canine parenteral vaccination against rabies and therefore reach the 70% herd immunity to interrupt rabies transmission in dogs [18]–[20] . The model revealed a high incidence of rabies related human deaths ( 5 . 8 per 100 , 000 ) in Cambodia , which was 15 times higher than that of the official reports . To put it into perspective , rabies caused more deaths than dengue or malaria related deaths in Cambodia ( ∼100 and ∼400 deaths on average for the past 5 years caused by dengue and malaria respectively ) . These estimates of dengue and malaria related deaths are produced by the Dengue and Malaria National Control Program from national surveillance systems [21] . This leaves Cambodia among countries with the highest incidence in the region – followed by India with an estimated incidence of 2–3 per 100 , 000 [22] . This discrepancy in the numbers of rabies related deaths between the official reports and our model was the result of the absence of surveillance and limited diagnostic capacity in Cambodia . Nevertheless , although rabies has been listed as a priority disease under surveillance since 2006 , it is not a surprise that rabies related encephalitis would still be underreported because anecdotal reports suggest that poor patients with encephalitis following dog bites are rarely hospitalized and die at home . As a consequence rabies is probably not seen by the authorities as a significant public health problem . It is important to interpret our results in light of some limitations . The model strongly relied on two major assumptions . First , we reasonably speculate that a large majority of Phnom Penh residents bitten by a dog would attend the PEP center– particularly those who were bitten by suspected rabid dogs because the IPC's PEP center is easily accessible and has become a well-known institution within Phnom Penh and rabies still generates fear in the communities . As a result , it is plausible that our estimate is close to the true incidence of SRDBI in Phnom Penh . The extent to which dog bite injuries were seen and treated by the private sector is difficult to estimate; however the number may be negligible compared to that of IPC PEP center as only a small population could afford cell culture-based vaccines against rabies at a prohibitive price . Second , the incidence of suspected rabid dog bites in Phnom Penh is supposed to be similar to that of the rest of the country – yet Cambodia is mainly rural . However , it is well recognized that incidences in rural areas are commonly higher than that of cities [4] , [23] . Therefore we believe the model has predicted a conservative estimate of rabies incidence . Indeed , our two assumptions if inaccurate , only affected the model by underestimating the true burden of rabies in rural Cambodia . Nevertheless , despite these limitations , we have shown that laboratory data can provide important information on generating predictive positive values of rabies infection for parameters that were accounted for in the model . Countries where such diagnostic capacity exists should be encouraged to collect data regarding dogs or dog bites so that each parameter needed for the decision tree model could be validated locally . Combined with additional epidemiological studies - surveys or sensitive/active surveillance system of encephalitis - to estimate the incidence of a suspected rabid dog bite injury such countries would be able to provide country level estimates needed for advocacy and setting priority in Asia and in Africa [2] , [3] . We believe that the present analysis represents an important contribution in bringing rabies to the attention of Cambodia national authorities where rabies is often perceived as a rare disease because of the lack of incidence data . Extrapolating from the model , the current PEP center for the past 10 years may have been prevented ∼1 , 000 rabies related deaths . However , only one free PEP clinic in Phnom Penh is not sufficient to handle a country in which rabies transmission occurs endemically . Since 1997 , IPC has spent approximately US$ 1 . 2 million on PEP excluding the cost of anti-tetanus vaccines and laboratory diagnostic costs ( IPC unpublished data ) . It is unlikely that this financial burden could eventually be reduced as more and more people could afford to reach the capital city in the future and if little is done to mitigate transmission of rabies virus in a plausibly growing dog population . Therefore , we strongly recommend establishing a comprehensive national rabies control program whose one of the major challenges would be to work across ministries and agencies to ensure continued political commitment and active community participation so that proper WHO recommended rabies vaccines are available and accessible to Cambodians and rabies transmission in the dog population is controlled . Progress on vaccine administration in humans and dogs , recent success in Sri Lanka or many South East Asian countries regarding rabies control program , and mounting evidence of cost effective interventions for rabies elimination should encourage Cambodia to tackle this fatal but preventable disease [24]–[29] .
|
In Cambodia , rabies still elicits fear in the communities . Since 1998 the Institut Pasteur in Cambodia ( IPC ) , Phnom Penh has been the only source of free post-exposure prophylaxis ( PEP ) and post mortem diagnosis . During 1998–2007 , on average ∼12 , 400 patients received PEP annually at IPC ( range 8 , 907–14 , 475 ) and 63 fatal human cases presenting with encephalitis following a dog bite were reported including 73% who tested positive by fluorescent-antibody test on brain samples or/and by reverse-transcriptase polymerase chain reaction on skin , cerebrospinal fluid , or urine . In 2007 , 14 , 475 patients received PEP ( 100 PEP/100 , 000 people in Cambodia ) including 95% who resided in Phnom Penh city ( 615 PEP/100 , 000 ) or five neighboring provinces . Using a step-by-step probability model , we estimated that 810 human rabies deaths would occur in 2007 ( 95% confidence interval [CI] 394–1 , 607 ) ; an incidence of 5 . 8/100 , 000 ( 95%CI 2 . 8–11 . 5 ) . As a result , despite high attendance at the IPC's PEP center most Cambodians living in peripheral provinces in Cambodia may not have adequate access to PEP . Finally , the model generated one of the highest incidences of rabies worldwide . A national rabies control program is needed to improve surveillance and access to PEP , and to initiate vaccination campaigns in dogs .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"health",
"and",
"epidemiology/infectious",
"diseases"
] |
2009
|
Rabies Situation in Cambodia
|
Chikungunya virus ( CHIKV ) is a mosquito-transmitted alphavirus that causes high fever , rash , and recurrent arthritis in humans . It has efficiently adapted to Aedes albopictus , which also inhabits temperate regions , including Europe and the United States of America . In the past , CHIKV has mainly affected developing countries , but has recently caused large outbreaks in the Caribbean and Latin America . No treatment or licensed CHIKV vaccine exists . Here , we have identified determinants in the CHIKV cell-attachment protein E2 that facilitate cell binding . The extracellular part of the E2 gene is subdivided into the three domains , A , B , and C . These domains were expressed in E . coli and as Fc-fusion proteins generated from HEK293T cells and used for cell-binding assays . Domains A and B bound to all cells tested , independently of their permissiveness to CHIKV infection . Domain C did not bind to cells at all . Furthermore , CHIKV cell entry was promoted by cell-surface glycosaminoglycans ( GAGs ) and domain B interacted exclusively with GAG-expressing cells . Domain A also bound , although only moderately , to GAG-deficient cells . Soluble GAGs were able to inhibit CHIKV infection up to 90%; however , they enhanced the transduction rate of CHIKV Env pseudotyped vectors in GAG-negative cells . These data imply that CHIKV uses at least two mechanisms to enter cells , one GAG-dependent , via initial attachment through domain B , and the other GAG-independent , via attachment of domain A . These data give indications that CHIKV uses multiple mechanisms to enter cells and shows the potential of GAGs as lead structures for developing antiviral drugs .
The Chikungunya virus ( CHIKV ) is a mosquito-transmitted alphavirus that causes high fever , rash , and recurrent arthritis in humans . The majority of symptoms disappear after about one week . However , in about 30% of cases , arthritis can last for months or even years , which may cause substantial economic losses [1] , [2] . The virus has been endemic in Sub-Saharan Africa , the Indian Ocean islands , India , and Southeast Asia . However , the virus spread to the Caribbean in late 2013 and is now responsible for a large , still-ongoing outbreak there and in Latin America with 1 . 9 million suspected cases as of December 2016 ( www . paho . org/hy/ ) . The mortality rate is very low ( 0 . 1% ) , but the infection rates are high ( sometimes 30% ) and asymptomatic cases are rare ( about 15% ) . Due to climate change , globalization , and vector switching , the virus will most likely continue to cause new , worldwide outbreaks . Additionally , more temperate regions of the world like Europe or the USA , which have recently reported their first cases , will likely become targets [3] , [4] . Alarmingly , no specific treatment or vaccination against CHIKV is available so far . CHIKV is a ( + ) single-stranded RNA virus . Like other alphaviruses , it enters cells by receptor-mediated endocytosis and a subsequent pH-dependent fusion step . CHIKV has two surface proteins that mediate cell entry: the transmembrane glycoproteins E2 and E1 . E2 mediates cell attachment and E1 is a class II viral fusion protein [5] , [6] . E2 and E1 associate as trimers of heterodimers ( E2–E1 ) on the particle surface [7] , [8] , [9] . The E2 protein contains two N-glycosylation sites at position 263 and 345 . The E2 envelope protein consists of domain C , located close to the viral membrane , domain A , in the center of the protein , and domain B , at the distal end , prominently exposed on the viral surface [7] , [8] . These domains are promising sites of interaction with the target cell . Potential interaction partners of viruses on the cell surface are glycosaminoglycans ( GAGs ) , which are ubiquitously present on the surfaces of all animal cells and are an essential part of the extracellular matrix ( ECM ) [10] , [11] , [12] . They consist of long linear chains of disaccharide units ( 30–60 per chain ) . These disaccharides are sulfated to different degrees and are thus negatively charged . GAGs that are covalently linked to a core protein are called proteoglycans ( PGs ) . They differ depending on the carbohydrates that form the disaccharide units . The best characterized GAGs linked to core proteins on human cells are heparan sulfate ( HS ) , chondroitin sulfate ( CS ) , and dermatan sulfate ( DS ) [12] . Since GAGs are ubiquitously present on the cell surface , many pathogens exploit them to cross the cell membrane barrier and use them for initial cell attachment or as entry receptors . These pathogens include several bacteria , parasites , and viruses [10] , [13] . Cell surface HS , the most extensively studied GAG , promotes attachment and/or entry of herpes simplex virus type 1 ( HSV-1 ) , human immunodeficiency virus ( HIV ) , hepatitis C virus ( HCV ) , vaccinia virus ( VACV ) , dengue virus ( DENV ) , and adeno-associated virus isolate 2 ( AAV-2 ) into cells [13] . Binding of an alphavirus , the eastern equine encephalitis virus ( EEEV ) , to cell surface HS , thereby enhancing its neurovirulence , has also been reported [14] . So far , the role of GAGs in CHIKV replication has only been studied in the context of viral attenuation . Point mutations within domain A of the E2 protein ( e . g . , E79K or G82R ) have been found in attenuated vaccine strains that were cell culture adapted and showed enhanced GAG dependency but reduced in vivo replication [15] , [16] , [17] , [18] . Additionally , it was reported that cell-surface PGs promote replication of some CHIKV strains , but this replication was not inhibited by the presence of soluble GAGs [15] . Furthermore , one CHIKV strain has been shown to be not influenced at all by the presence or absence of PGs [17] . The E2 domains A and B have been suggested before to be putative receptor binding sites [7] , [8] . Since the cell entry process of CHIKV is not understood in detail [19] , we examined the role of GAGs in CHIKV cell entry and analyzed the binding properties of the E2 domains A , B and C , and their dependency on GAGs . The two surface-exposed E2 domains , A and B , both bound to cells expressing GAGs . Domain C was not involved in cell binding at all . We could show that CHIKV entry is enhanced in cells expressing GAGs and that domain B binds exclusively to GAGs . Domain A also bound to cells that do not express GAGs . Our results suggest that CHIKV uses at least two entry mechanisms , one GAG-dependent , via attachment through E2 domain B , and the other GAG-independent , via domain A .
All cells used for this study were cultured at 37°C under 5% CO2 . HEK 293T ( ATCC: CRL-1573 ) cells were incubated in Dulbecco’s modified Eagle’s medium ( DMEM; Lonza , Verviers , Belgium ) . Jurkat ( ATCC: TIB-152 ) and BHK-21 ( CCL-10 ) cells were grown in Roswell Park Memorial Institute medium ( RPMI; Biowest , Nuaille , France ) . CHO-K1 and pgsA-745 cells were grown in Ham’s F-12 medium ( Life technologies , Darmstadt , Germany ) and 2 mM glutamine . Media were supplemented with 10% FBS ( v/v; PAA , Pasching , Austria ) and 5% L-glutamine ( 200 mM; Lonza , Verviers , Belgium ) . The glycosaminoglycans chondroitin sulfate , dermatan sulfate , heparan sulfate , heparin and dextran sulfate were purchased from Sigma-Aldrich ( Taufkirchen , Germany ) . The codon-optimized CHIKV E3-E1 gene ( based on isolate “S27” ) was synthesized by GeneArt ( Life Technologies , Darmstadt ) and cloned into the plasmid pIRES2-eGFP ( Clontech/Takara , Saint-Germain-en-Laye , France ) as described previously [20] . The E2 domain A ( including the β-ribbon connector ) , B ( aa 172–231 ) , and C ( aa 271–341 ) genes were cloned into the bacterial expression vector pET-15b . The same was done with the entire extracellular part of the E2 protein ( E2ex ) . Cloning was achieved by adding the restriction sites NdeI and BamHI via primers by PCR ( template DNA: pIRES2-EGFP-CHIKV E3-E1 ) , cutting the DNA products and the vector , and subsequent ligation . For domain A , two fragments were derived via PCR . One fragment contained domain A itself and the first part of the β-ribbon connector ( C-terminus of domain A ( aa 1–171 ) ) . The other fragment contained the second half of the connector C-terminus of domain B ( aa 231–270 ) . The fragments were cloned into the pET-15b vector by a triple ligation ( via NdeI and BamHI ) . The two fragments were linked via a shared SmaI restriction site ( at the C-terminal part of fragment one and the N-terminal part of fragment two , respectively ) . By this procedure , E2 domain B was bypassed and replaced with the sequence G4PG5 . The constructs also contained an N-terminal poly-histidine-tag for purification . The primers used for cloning were: The Fc-fusions were constructed by cloning E2 fragments via Apa I , Nhe I sites introduced by PCR , into the vector pCMV2 . 5-hIgG1Fc-XP ( kind gift of Stephan Dübel , TU Braunschweig ) . This generated expression vectors that express E2-Fc fusion proteins without any tags . The following primers were used to amplify the E2 fragments from the E . coli expression vectors as template: Domain A fw 5’ AAAAGGGCCCAGCACCAAGGACAACTTCAAC , rev 5’AAAAGCTAGCCTTGGGCACCATGCAGGTC; domain B fw 5’ AAAAGGGCCCCCCGACACCCCCGATAGAA , rev 5’AAAAGCTAGCGGTCACGGCGGCGTGGC; domain C fw 5’ AAAAGGGCCCGCCCGGAACCCTACCGTG , rev 5’AAAAGCTAGCCTGGGGCCAGTACTTGTAGG; domain A-ß fw 5’ AAAAGGGCCCAGCACCAAGGACAACTTCAAC , rev 5’AAAAGCTAGCGGGGTCGTGGTGGAAGGG . All mutations were introduced by site directed mutagenesis . Proteins were expressed in BL21-CodonPlus ( DE3 ) -RIPL competent cells ( Agilent Technologies , Böblingen , Germany ) transformed with the pET-15b plasmid containing construct A , B , C , or E2ex . Bacteria were inoculated into 100 ml of LB medium containing ampicillin ( 0 . 1 mg/ml ) and grown overnight ( 37°C , 220 rpm ) . After 16 hrs , 2 l of LB medium were inoculated with the 100 ml overnight culture . The bacteria were grown to an OD600 of 0 . 5–0 . 7 , and then protein expression was induced by the addition of 1 mM IPTG . After another 2 . 5 hrs of incubation , cells were harvested and the pellets were frozen at –20°C . The recombinant proteins were purified from the bacterial pellets under native ( B , C ) or denaturing ( A , E2 ) conditions using HisTrap FF Crude columns ( GE Healthcare , Freiburg , Germany ) and the ÄKTA system ( GE Healthcare , Freiburg , Germany ) as described by [21] . For A and E2 , ion-exchange chromatography was additionally performed to remove contaminating bacterial proteins . After purification , proteins were dialyzed against PBS using Slide-A-Lyzer Dialysis Cassettes 3 . 5K MWCO ( Pierce , Thermo Scientific , Bonn , Germany ) and concentrated with Ultra-4 3 kDa Centrifugal Filter Units ( Merck Millipore , Schwalbach , Germany ) . The protein concentration was determined by SDS-PAGE with marker proteins , following staining with Coomassie ( Bio-Rad , Munich , Germany ) . Proteins were then quick-frozen with liquid nitrogen and stored at –80°C . For experiments , proteins were thawed in a 37°C water bath . The empty Fc control protein and the E2 domain-Fc-fusion proteins were produced in HEK293T cells by transient transfections . Transiently transfected HEK293T cells were grown in DMEM containing 10% FCS . After 48 h , supernatants of the transfected cells were harvested two times at 24 h intervals . The secreted Fc-fusion proteins were purified by affinity chromatography with protein A-agarose , eluted at a pH 2 . 5 , neutralized with 1M Tris , pH 9 . 0 and dialyzed against PBS pH 7 and stored at -80°C . The plasmid pCHIKV-mCherry-490 [22] was in vitro-transcribed with T7 RNA polymerase after NotI linearization . The mRNA was transfected into BHK-21 cells using Lipofectamine 2000 ( according to the manufacturer’s protocol; Life Technologies ) . Virus-containing supernatants were harvested 48 hrs later and used to reinfect fresh BHK-21 cells for virus amplification . Infected cells showed a clear red fluorescence . Supernatants were collected and stored at –80°C or used to determine the viral titer . For CHIKV infections , 293T cells were seeded onto a 24 well plate . Cells were incubated at 37°C for 16–24 hrs and counted ( about 3 × 105 cells ) . Subsequently , the mCherry-tagged CHIKV ( CHIKV-mCherry-490 ) [22] was added at a multiplicity of infection ( MOI ) of 1 . After 6 hrs , the cells were collected in medium , washed and resuspended in 2% paraformaldehyde in PBS , and analyzed by flow cytometry . At least 10 , 000 events were acquired with an LSRII instrument ( BD Biosciences ) and analyzed using FACS Diva software . Lentiviral vector particle production was performed as described previously [20] . Briefly , 293T cells were seeded in 10 cm dishes in 10 ml DMEM . Cells were cotransfected 16 hrs post seeding with the plasmids pRRLsinhCMV-GFP-pre ( a lentiviral vector genome encoding GFP ) or pCSII-Luc ( a lentiviral vector genome encoding luciferase ) , pMDLg/pRRE , pRSVrev , and pHIT-G or pIRES2-eGFP-CHIKV E3-E1 using Lipofectamine 2000 ( according to the manufacturer’s protocol; Life Technologies ) . After 24 hrs incubation , the medium was discarded and replaced with 5 ml of fresh DMEM . Another 24 hrs later , the supernatant containing the vector particles was harvested , sterile filtered with 0 . 45 μm filters ( Sartorius , Göttingen , Germany ) , and frozen at -80°C . Cells were transduced with pseudotyped lentiviral vector particles in 384-well plates as described previously [20] . Briefly , 6000 293T cells per well were seeded ( using a MultiFlo Microplate Dispenser; BioTek , Bad Friedrichshall , Germany ) in 20 μl DMEM in white CELLSTAR 384-well microtiter plates ( Greiner Bio-One , Frickenhausen , Germany ) and incubated for 16–24 hrs at 37°C . The same was done for CHO-K1 and pgsA-745 cells using Ham’s F-12 medium and 3000 cells per well were seeded . Soluble glycosaminoglycans ( GAGs ) were serially diluted ( three-fold ) four times in DMEM containing vector particles in 96-U-well plates ( Thermo Scientific , Rockford , IL , USA ) , and incubated at 4°C for 1 h . This resulted in equal amounts of vector and serially diluted compound . The vector particle mixtures were then added to the cells using a Matrix Multichannel Equalizer Electronic Pipette ( Thermo Scientific , Rockford , IL , USA ) , transferring 20 μl each to three wells of the 384-well plate out of one well of the 96-well plate . This resulted in a final concentration of GAGs/dextran sulfate ranging from 500 to 6 . 2 μg/ml . Cells were incubated with the vector particle mixtures for another 16–24 hrs . Afterwards , 20 μl of BriteLite substrate ( PerkinElmer , Rodgau , Germany ) was added . After 5 minutes incubation at room temperature , the luciferase signal was detected using the PHERAstar FS microplate reader ( BMG LABTECH , Ortenberg , Germany ) . Statistical analyses were done using the GraphPad Prism 5 . 04 software ( La Jolla , CA , USA ) . The p-values were determined by the unpaired two-tailed t-test . The recombinant proteins purified from E . coli or E2-Fc-fusion proteins were incubated for 30 min at 4°C with cells in PBS/2% FCS . The cells were then washed and bound protein was detected via an anti His-tag antibody ( Dianova , Hamburg , Germany ) and an anti-mouse IgG-FITC antibody or an FITC coupled anti-human IgG antibody ( Fc-fusion proteins ) followed by flow cytometry . Binding was detected as the mean change in fluorescence . In addition , the CHIKV E2-derived protein sA ( containing the surface exposed regions of domain A connected by linkers ) [23] was used as a negative control for the binding of E . coli derived proteins and the Fc-protein served as negative control for Fc-fusion proteins and the values are given as fold increase in binding compared to Fc .
First it was analyzed if GAGs play a role in viral entry . For this investigation , 293T cells , CHO-K1 cells , and the CHO-K1 derived cell line pgsA-745 which , due to an enzymatic defect , is not able to produce GAGs , were transduced with CHIKV Env- or VSV-G-pseudotyped lentiviral vectors encoding GFP . Transduction was determined as the number of GFP-positive cells and was standardized as percentage of GFP-positive cells obtained after transduction with VSV-G-pseudotyped vectors . The results displayed in Fig 1A show that CHIKV cell entry into 293T and CHO-K1 cells was almost equally efficient . However , the transduction rate of pgsA-745 cells was significantly reduced by more than 50% in comparison to the parental cell line . Thus , cell entry of CHIKV Env-pseudotyped vectors into GAG-deficient cells is strongly and significantly reduced in comparison to those carrying cell-surface GAGs . However , cell entry was only reduced by about 50% , indicating the existence of at least one more entry pathway . To confirm the relevance of the above experiments , the dependency of CHIKV infections on cell-surface GAGs was studied . For this , CHO-K1 and pgsA-745 cells were both infected with the recombinant CHIKV-mCherry-490 using an MOI of 1 . This virus contains an mCherry gene within the nsP3 gene of CHIKV and has growth characteristics similar to the wild-type virus [22] , [24] . Viral replication was determined at 6 and 24 hrs post-infection by flow cytometry . The infection rate at 6 hrs post-infection was significantly reduced ( 9 . 9-fold ) in pgsA-745 cells compared to CHO-K1 cells ( Fig 1B ) . At 24 hrs post-infection , the difference decreased to a 1 . 5-fold higher , yet still significantly different , infection rate in CHO-K1 cells compared to pgsA-745 cells . Thus , both the cell entry of CHIKV Env-pseudotyped vectors and the replication of CHIKV were reduced on pgsA-745 cells lacking cell-surface GAGs , but not fully inhibited , indicating that GAGs enhance , but are not essential for entry . CHIKV cell entry into GAG-deficient cells was reduced; accordingly , the presence of soluble GAGs might inhibit CHIKV entry into GAG-expressing cells or influence entry into GAG-free cells . Therefore , 293T , CHO-K1 , and pgsA-745 cells were transduced with CHIKV Env-pseudotyped vectors in the presence of different amounts of soluble GAGs . Dextran sulfate ( DX ) , which consists of long chains of highly sulfated glucose units , was used as a control for the soluble GAGs , in addition to HP . It has a similar charge to HP and the other GAGs , but its structural background is built up of entirely different carbohydrates . The experiment was carried out in a 384-well plate format with vectors encoding firefly luciferase [20] . Transducing 293T and CHO-K1 cells with CHIKV Env-pseudotyped vectors transferring a luciferase gene in the presence of GAGs generally resulted in dose-dependently reduced transduction efficiencies compared to the control without GAGs ( Fig 2 , top ) . DX and HP were the most potent inhibitors on both cell lines . On 293T cells , the cell entry could be inhibited to about 10–20% of the untreated control at a concentration of 500 μg/μl GAGs . On CHO-K1 cells , transduction was reduced to about 20–30% of the GAG-free control maximally ( Fig 2 , top ) . Cell entry of CHIKV Env-pseudotyped vectors into pgsA-745 cells was , on the contrary , dose-dependently enhanced by the addition of rising GAG concentrations ( Fig 2 , bottom ) . The highest values reached , at 500 μg/μl , were between 121 ( DX ) and 177% ( HP ) of the untreated controls . Substantial inhibition of transduction with values lower than 100% was only observed for DX at concentrations lower than 500 μg/μl . In conclusion , the data indicate that cell entry of CHIKV Env-pseudotyped vectors is enhanced by GAGs , since low level transduction is still possible in pgsA-745 cells and in the presence of soluble GAGs . In addition , the entry into GAG-deficient pgsA-745 cells could be enhanced with increasing amounts of GAGs , indicating an activation of CHIKV Env-pseudotyped vectors . As shown above , CHIKV replication in GAG-deficient cells was diminished . Accordingly , this raised the question of whether infection of cells is inhibited in the presence of soluble GAGs . 293T cells were infected with CHIKV-mCherry ( MOI 1 ) in the presence of 500 μg/ml of the respective soluble GAGs ( but 500 U/ml HP ) . Six hours later , cells were analyzed by flow cytometry . Fig 3A shows that all GAGs reduced viral replication significantly by at least 76 . 3% ( HS ) . HP was most effective , reducing replication by 90 . 7% compared to the GAG-free control . To determine whether the inhibition of CHIKV replication by GAGs occurs at the attachment/entry step of the viral life cycle , CHIKV and the different GAGs were incubated with target cells at 4°C to allow viral attachment to the cells but to avoid the subsequent endocytosis step . After 30 minutes , unbound virus and the GAGs were washed away and the cells were incubated for 6 hrs at 37°C in fresh medium . Infection rates were measured via flow cytometry . Fig 3B shows , similar to the previous experiment , a significant reduction of cell attachment/entry by 62 . 1 ( HS ) to 82 . 4% ( HP ) . In conclusion , replication of recombinant CHIKV-mCherry-490 in 293T cells is significantly inhibited by the addition of soluble GAGs . Attachment and/or endocytosis are the critical steps in the viral life cycle where this inhibition occurs . The E2 protein is the cell-binding moiety of CHIKV and has been frequently described to contain determinants recognized by neutralizing antibodies [7] , [8] , [25] . Therefore , one could speculate that parts of E2 are components that bind the unidentified cellular receptor of CHIKV . To analyze which domains of the E2 protein are involved in cell binding , the sequences encoding domain A ( including the β-ribbon connector , aa 1–171 and 231–270 ) , domain B ( aa 172–231 ) and domain C ( aa 271–341 ) were cloned into the bacterial expression vector pET-15b . The same was done for the entire extracellular part of the E2 protein ( E2ex ) , which served as a positive control . The constructs also contained an N-terminal poly-histidine-tag for purification . The proteins were expressed in E . coli , and partially purified by Ni2+ affinity chromatography under native ( E2 domains B and C ) and denaturing ( E2 domain A and E2ex ) conditions . For domain A and E2ex additional ion-exchange chromatography was performed to remove bacterial protein contaminants . Fig 4 shows a Coomassie-stained SDS-PAGE separation of the purified proteins . Domain A has a molecular mass of 26 . 5 kDa , B of 8 . 5 kDa , C of 10 . 7 kDa , and E2ex of 40 . 4 kDa . The purified proteins migrated at the expected size . Their cell binding was analyzed with CHO-K1 , the GAG deficient pgsA-745 , 293T and Jurkat cells . 293T cells show good transduction efficiencies with CHIKV Env-pseudotyped vectors . In contrast , Jurkat cells have revealed very low transduction efficiencies [20] . The recombinant proteins were incubated with cells at 4°C . Bound protein was detected by flow cytometry using an anti-His-tag antibody and an anti-mouse IgG-FITC antibody . Data are presented as fold increase in the mean FITC values of the sample in comparison to those of the control ( cells only treated with staining antibodies ) . In addition , the protein sA , which contains only surface-exposed domains of A and did not induce neutralizing anti-E2 antibodies upon vaccination [23] and was used as a negative control for the binding studies . Fig 5 reveals that domains A and B , and protein E2ex showed a significant difference in the mean FITC signal compared to the control sample , indicating binding of these proteins to 293T and CHO-K1 cells . In contrast , there was no significant cell binding detectable for the negative control , sA , and domain C . An identical pattern was obtained for Jurkat cells , although at a generally lower binding level . Conducting the experiment with pgsA-745 cells resulted in weak , yet still significant , binding of domain A and E2ex , and no binding of domain B or C ( Fig 5 ) . Accordingly , these data indicate that binding of domain B to cells is enabled by cell-surface GAGs , while that of domain A is only partly dependent on these molecules . To further prove that cell binding of the E2 protein domains is GAG dependent , inhibition of their cell binding by addition of soluble GAGs was analyzed . The recombinant E2 proteins and different cell lines were again incubated at 4°C for binding; however , this time in the presence of 500 μg/ml soluble GAGs . Heparin ( HP ) was used as a control to ascertain the role of charge in the cell binding . HP is structurally derived from heparan sulfate ( HS ) , but is more heavily sulfated and thus has a higher negative charge density . Fig 6 shows the results of soluble GAG-induced protein-binding inhibition and indicates that binding of domain A to 293T cells was not reduced in the presence of any of the soluble GAGs , except HP ( about 2 . 5-fold ) . In contrast , the presence of all GAGs reduced the cell binding of domain B up to 3 . 0-fold . Again , HP showed the most efficient inhibition ( 4 . 5-fold ) . Inhibition of protein E2ex binding was in between that of domains A and B . A similar picture was observed for CHO-K1 cells ( Fig 6 , middle ) . Here , in contrast to 293T cells , HP did not strongly inhibit domain A binding . Minor inhibition of domain A binding was detectable after addition of HS and chondroitin sulfate ( CS ) . The same experiment was performed with pgsA-745 cells ( Fig 6 , bottom ) . Here , domain C was used as an additional control . However , the binding of the recombinant proteins to pgsA-745 cells was not inhibited or enhanced in the presence of soluble GAGs . The viability of 293T cells incubated with 500 μg/ml of the different GAGs ( the maximum concentration used in all experiments ) was tested using MTT assays . None of the GAGs were significantly cytotoxic at this concentration ( data not shown ) . In summary , binding of E2 domain A to 293T and CHO-K1 cells was not or weakly inhibited by the addition of soluble GAGs . Only the strong negatively charged HP inhibited the interaction of domain A with 293T cells . In contrast , binding of domain B to both cell lines was massively decreased in the presence of HP , HS , CS , and dermatan sulfate ( DS ) . The addition of GAGs to domains A , B , or C had no influence on the binding properties of these proteins towards GAG-deficient pgsA-745 cells . There was a serious concern that the E2 domains expressed in E . coli might not be correctly folded or have incorrect disulfide bond patterns . Therefore the E2 domains were expressed as Fc-fusion proteins in a soluble form in eukaryotic cells by transient transfections of 293T cells . Supernatants containing these proteins were partially purified by protein-A Sepharose chromatography and used for binding assays ( Fig 7A ) . Western blot analysis of the proteins under native conditions confirmed that the proteins were Fc-dimers , as expected for antibodies ( Fig 7B ) . This implies that disulfide bonds were formed and might be present not only in the Fc part , but also in the E2 fragments . First the three subdomains A , B and C were tested for their cell binding ability as Fc-fusion proteins towards CHO-K1 and pgsA-745 cells . The Fc protein was used as negative control . Two concentrations were tested and “low” represents one sixth of the original sample . Again , domain A and B bound to CHO-K1 cells ( Fig 7C ) . Domain C bound only marginally to CHO-K1 cells . This distribution is similar to the one determined with proteins expressed by E . coli ( Fig 5 ) , although domain A expressed from eukaryotic cells had a higher binding affinity compared to domain B , than domain A derived from E . coli . Analysis of cell binding to pgsA-745 cells revealed that domain A shows residual cell binding in the absence of GAGs . Domain C did not bind to pgsA-745 cells . Point mutations in domain A have been described before to increase CHIKV binding to GAGs [15] , [16] , [18] , therefore we generated two mutants of domain A and expressed them as Fc-fusion proteins , domain A with a mutation at position 79 ( Fc-CHIKV-E2-A-E79K ) and position 166 ( Fc-CHIKV-E2-A-166K ) . In addition , we generated a construct containing domain A without the ß-ribbon connector ( A-ß ) ( Fig 8A ) . Again , the proteins run in SDS-PAGE under native conditions as dimers ( Fig 8B ) . The two point mutants in domain A showed a higher binding affinity to CHO-K1 cells compared to domain A ( Fig 8C ) and their binding to pgsA-745 cells was only weakly increased . This indicates that introducing the point mutations in domain A increased its binding affinity to GAGs . Deletion of the ß-ribbon connector decreased domain A binding to CHO-K1 cells , however binding to pgsA-745 cells remained unaltered compared to domain A containing the ß-ribbon connector . This indicates that the ß-ribbon connector partially contributes to GAG binding and that the GAG independent binding site is located in domain A .
The alphavirus CHIKV enters cells by receptor-mediated endocytosis and a subsequent pH-dependent fusion step [26] . Host factors that are required for CHIKV entry are still only poorly understood . First hints have emerged from genome-wide RNAi screens , where downregulation of archain1 , fuzzy homologue or TSPAN9 inhibited CHIKV infection but also that of alphaviruses in general [27] . On the viral side , the CHIKV E2 glycoprotein mediates cell attachment; however , the detailed mechanism has not been studied well so far [5] , [6] . The structures of the CHIKV envelope proteins have been solved by X-ray crystallography [7] , [8] . The extracellular part of E2 has three immunoglobulin ( Ig ) -like extracellular domains called A , B , and C [7] . Here , we expressed the extracellular part of E2 ( E2ex ) and its three domains A , B and C in E . coli . Cell-binding experiments revealed a significant binding of domains A and B , and E2ex , but not domain C , to all cells tested . Domain C is found close to the viral membrane and is followed by the stem region , which is a linker to the transmembrane region composed of hydrophobic amino acids [7] . Since domain C is not surface accessible and does not contain epitopes for neutralizing antibodies [7] , it was not surprising that this protein lacked cell-binding activity . Domain B is located at the tip of the protein protruding from the viral surface and is linked to domain A via a long β-ribbon connector ( containing an acid-sensitive region [ASR] ) . Domains A and B are prominently exposed on the viral membrane and it has previously been speculated that the cellular receptor mainly interacts with domain A [7 , 8 , 28] . The recombinant proteins containing domains A and B bound to cells independently of the cells’ ability to allow cell entry of CHIKV [20] . Binding to non-permissive Jurkat cells was reduced in comparison to binding to highly permissive 293T cells , which indicates that the binding moiety might be present on all cells but at different densities . Further analysis of the possible cellular attachment factor was performed with the help of CHO-K1 and pgsA-745 cells . The pgsA-745 cells are derived from CHO-K1 cells and lack glycosaminoglycans ( GAGs ) on the cell surface . Binding assays revealed that domains A and B bound to CHO-K1 , but only domain A bound to pgsA-745 cells , however markedly reduced but still significant . Furthermore binding of domain B to GAG-expressing cells in the presence of soluble GAGs was highly reduced . This suggests that domain B binds GAGs and thereby facilitates CHIKV cell binding . In contrast , domain A binding to GAG-deficient pgsA-745 cells and competition with soluble GAGs only marginally decreased domain A cell binding , indicating GAG-independence . These data were confirmed with Fc-fusion proteins containing the E2 domains A , B and C , suggesting that the E . coli derived proteins have a native structure . In addition , point mutations in domain A showed the previously observed increase in GAG binding . Deletion of the ß-ribbon connector decreased the GAG binding of domain A , indicating that the ß-ribbon connector is partially responsible for GAG binding . A GAG binding consensus sequence is located in the ß-ribbon connector ( -DRKGK- amino acid 251–255 ) [29] . GAG independent binding was not affected and might be mediated by domain A . These data were substantiated by transduction of pgsA-745 cells with CHIKV Env-pseudotyped vector particles , which was significantly reduced by over 50% in comparison to transduction of the parental CHO-K1 cells , and soluble GAGs inhibited transduction of GAG-expressing cells . Remarkably , the transduction rates could not be reduced to less than 10% of the soluble GAG-free control , indicating that at least one GAG-independent entry pathway must exist . The presence of soluble GAGs did not decrease , but rather enhance transduction of pgsA-745 cells with CHIKV Env-pseudotyped vectors . This might be simply due to a local acidic environment generated by soluble GAGs , which may induce a fusion competent conformation of the glycoprotein . Alternatively , a pre-activation of the CHIKV envelope proteins through binding to the soluble GAGs might occur . Soluble GAGs did not enhance the binding of domains A or B to pgsA-745 cells; therefore , it is tempting to speculate that GAGs might induce conformational changes within the envelope proteins that allow them to bind more effectively to structures on the cell surface , which then might promote viral uptake . Such an activation of the virus has been described for the human papillomavirus type 16 ( HPV-16 ) in the presence of HP , which allowed HPV-16 infection in the absence of cell-surface GAGs [30] . Cell-surface HS is an important attachment factor for HPV-16 [31] . For AAV-2 particles , slight structural rearrangements on the viral surface have been described upon HP binding [32] . Furthermore , it has been proposed that initial structural rearrangements on the alphavirus surface occur directly after binding to the cell surface [9] , based on the observation that transitional epitopes of the Sindbis virus ( SINV ) became accessible to antibodies upon cell binding [33 , 34 , 35] . Replication of CHIKV was also significantly reduced in pgsA-745 cells in comparison to CHO-K1 cells at 6 hrs post infection . However , this effect was less pronounced 24 hrs post infection , although still significant . CHIKV replication in 293T cells was significantly reduced in the presence of soluble GAGs . The lack of GAG-dependency at a later time point during infection might be explained by potential direct cell-to-cell transmission , as it has been observed for CHIKV in cell culture [36] , or could just be due to the saturation of infection . For example , the majority of cells could already be infected 18 hrs post-infection and the viral titer might already be at a plateau , regardless of whether the cell entry is less efficient or not . However , an enhancement of infection by soluble GAGs was not observed . This might indicate that higher concentrations of GAGs are needed , which are possibly within their toxic range . So far , the role of GAGs in CHIKV replication has mainly been studied with regard to viral attenuation . Point mutations within the E2 domain A ( e . g . , E79K , G82R or E166K ) have been found in attenuated vaccine strains that were cell-culture adapted and showed enhanced GAG-dependency but reduced in vivo replication [15] , [16] , [18] . These mutations increase the positive charge in domain A [16] , and as shown here for the first time , directly affect its binding affinity . In addition , recently it has been shown that CHIKV binding to GAG receptors on mammalian cells enhances replication in those cells and cell binding was influenced by the N-glycosylation pattern of the viral envelope proteins [37] . There are two glycosylation sites in E2 , one in the ß-ribbon connector ( N-263 ) and another one in the stem region , outside the region that was analyzed here . Deletion of the ß-ribbon connector and expression of domain A in E . coli decreased domain A’s cell binding , indicating that glycosylation of N-263 may partially contribute to GAG binding . In contrast to domain A , domain B has mostly been described to be associated with covering the E1 protein and thus antibodies binding to domain B mainly inhibit the movement of domain B during fusion , but not cell attachment [25] . Our findings reveal a unique role for cell-surface GAGs during CHIKV infection , in which they are not absolutely necessary for CHIKV replication , but undoubtedly promote viral entry and replication . There is at least one GAG-independent entry pathway , as CHIKV entry into GAG-deficient cells is still possible , and soluble GAGs cannot fully block CHIKV cell entry . This additional pathway ( s ) consequently include different cell surface receptor ( s ) . The proposed entry pathways are most likely mediated by different binding sites on the E2 protein . The GAG-dependent pathway would be characterized by binding of the prominently exposed domain B most likely in combination with domain A to cell-surface GAGs , inducing a conformational change of the CHIKV Env molecules which might allosterically spread to the other CHIKV Env molecules on the particle surface [9] . This activation might enable the envelope molecules to bind to a second molecule on the cell surface , possibly via domain A . Following this binding , the virus is taken up by receptor-mediated endocytosis and the further pH-induced conformational changes occur . The proposed key role for domain B is supported by experiments with chimeric viruses which revealed that domain B is the critical target of neutralizing antibodies in humans and mice [38] . The second , GAG-independent pathway in this model possibly involves direct binding of domain A to another cell-surface molecule which might not play a role in the GAG-dependent pathway . GAGs on the cell surface are thus not absolutely required for CHIKV cell entry , but they are part of a strategy CHIKV employs in order to enter cells . They might be comparable with T-cell immunoglobulin and mucin ( TIM ) membrane proteins that promote , although are not absolutely required for , the cell entry of a number of viruses , including CHIKV [39] . The fact that CHIKV is able to infect a wide range of , and evolutionary only distantly related , species , and to infect many different cell types and organs within one organism [6] , makes the scenario plausible , in which the virus can utilize the ubiquitously expressed GAGs for entry [40] , but additionally exploits other opportunities and receptors to get into the host cell . Finally , GAG mimetics might thus be promising antiviral candidates for the treatment of CHIKV infections however they will not inhibit the GAG-independent entry pathway [41 , 42] . Because the E2 domains B and A contain cell binding moieties , they might be promising targets for vaccine development [23] .
|
The chikungunya virus ( CHIKV ) glycoprotein E2 mediates cell attachment and consists of three domains A , B and C . Since the cell entry process of CHIKV is not understood in detail , we analyzed the binding properties of the three E2 domains with proteins expressed in E . coli or as Fc-fusion proteins and the role of glycosaminoglycans ( GAGs ) on E2 cell binding and CHIKV entry . The two surface-exposed E2 domains , A and B , both bound to cells and domain B bound only to cells expressing GAGs . Domain A bound additionally to GAG-deficient cells and domain C did not bind to cells . CHIKV-pseudotyped lentiviral vector and CHIKV entry were enhanced in cells expressing GAGs . Our results suggest that CHIKV uses at least two entry mechanisms , one GAG-dependent , via attachment through E2 domain B , and the other GAG-independent , via binding of domain A . These data give indications that CHIKV uses multiple mechanisms to enter cells and shows the potential of GAGs as lead structures for developing antiviral drugs . In addition , it shows that domain A and B might constitute good targets for vaccine development .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"cell",
"binding",
"flow",
"cytometry",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"viral",
"transmission",
"and",
"infection",
"293t",
"cells",
"pathogens",
"biological",
"cultures",
"tropical",
"diseases",
"microbiology",
"alphaviruses",
"viruses",
"chikungunya",
"virus",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"spectrum",
"analysis",
"techniques",
"viral",
"entry",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"recombinant",
"proteins",
"cell",
"lines",
"spectrophotometry",
"biochemistry",
"cytophotometry",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"organisms"
] |
2017
|
Identification of Functional Determinants in the Chikungunya Virus E2 Protein
|
Neurocysticercosis ( NCC ) , a helminth infection of the brain , is a major cause of seizures . The mediators responsible for seizures in NCC are unknown , and their management remains controversial . Substance P ( SP ) is a neuropeptide produced by neurons , endothelial cells and immunocytes . The current studies examined the hypothesis that SP mediates seizures in NCC . We demonstrated by immunostaining that 5 of 5 brain biopsies from NCC patients contained substance P ( SP ) -positive ( + ) cells adjacent to but not distant from degenerating worms; no SP+ cells were detected in uninfected brains . In a rodent model of NCC , seizures were induced after intrahippocampal injection of SP alone or after injection of extracts of cysticercosis granuloma obtained from infected wild type ( WT ) , but not from infected SP precursor-deficient mice . Seizure activity correlated with SP levels within WT granuloma extracts and was prevented by intrahippocampal pre-injection of SP receptor antagonist . Furthermore , extracts of granulomas from WT mice caused seizures when injected into the hippocampus of WT mice , but not when injected into SP receptor ( NK1R ) deficient mice . These findings indicate that SP causes seizures in NCC , and , suggests that seizures in NCC in humans may be prevented and/or treated with SP-receptor antagonists .
Neurocysticercosis ( NCC ) is a parasitic infection of the human central nervous system that is caused by the pig tapeworm Taenia solium . NCC is the major cause of acquired seizures worldwide and now accounts for up to 10% of emergency room visits for seizures in the southwestern United States [1]–[7] . Live T . solium cysts in the brain of NCC patients are surrounded by little or no inflammation . Seizures are thought to result from the granulomatous host response initiated by dead or dying cysts rather than mediators produced by the parasite itself [5] , [8]–[11] . Since antihelmintic medications kill live cysts , a dilemma arises regarding treatment with these agents . The mediators responsible for inducing seizures in NCC are not known; their identification may lead to more effective strategies for prevention and/or treatment of seizures in this disease . Substance P ( SP ) is a neuropeptide within the tachykinin family produced by neurons , endothelial cells and immunocytes such as lymphocytes and macrophages . Receptors for SP ( NK1R ) are expressed by cells within and outside the central nervous system including neurons , endothelial cells and immunocytes [12]–[16] . SP signaling is involved in nociception [17] and neuropathic inflammation . A few reports implicate SP in the pathogenesis of seizures including one in which SP amplified seizure responses in rats [18] , [19] . The studies reported herein were undertaken to examine the hypothesis that SP is the epileptogenic agent in NCC . Our findings in brain biopsies of patients with NCC and in rodent models of NCC indicate that SP , produced within cysticercosis granulomas , causes seizures and suggest that seizures in NCC patients can be prevented and/or treated with SP receptor antagonists currently available in the clinic .
To determine if SP-producing cells can be detected in the brains of NCC patients with seizures , particularly within the region of granulomatous inflammation surrounding the dead or dying cyst , we performed immunohistochemistry on brain tissue specimens from patients with and without NCC . Included were 5 patients with NCC who underwent craniotomy and brain biopsy to remove intraparenchymal cysts and 2 individuals who died from non-neurological causes ( Figure 1 ) . SP peptide expression was readily detected within cells adjacent to parasite remnants in each of the 5 NCC brain biopsy specimens; the level of expression scored an average of 2±0 . 71 . In contrast , no SP+ cells were found distant from the parasite in the NCC biopsy specimens that included sufficient brain tissue; these distant areas scored 0 . 5±0 ( p< . 05 , Mann-Whitney test ) . Autopsy specimens from patients without NCC displayed no SP+ staining cells ( score = 0 . 5±0; p<0 . 05 , Mann-Whitney test ) . These findings are consistent with the hypothesis that SP is produced within the granulomatous inflammation that surrounds dead or dying cysts in NCC . To determine if SP alone is capable of inducing seizures , we injected SP directly into the hippocampus of rats and examined them for seizure activity using direct observation and EEG monitoring ( Figure 2 ) . Injection of SP ( 10 nanomoles ) resulted in severe behavioral seizures ( Racine grade 4 . 5±0 ) and increased electrical activity on EEG ( 1 . 06±0 . 1 mV ) , which lasted 5 . 0±1 . 23 sec . In contrast , injection of PBS elicited essentially no seizure activity assessed by these parameters ( p≤0 . 001 , Mann-Whitney test , all comparisons ) . Thus , SP alone was epileptogenic in rats . We previously demonstrated in the T . crassiceps murine model of cysticercosis that early and late stage granulomas form around the dead or dying parasites and can be isolated from peritoneal cavity [20] , [21] . Early stage granulomas contained parasite remnants and produced Th1 cytokines and SP peptide , while late stage granulomas did not contain parasite remnants or SP and produced IL-4 in addition to Th1 cytokines [20] , [21] . Injection of extracts of early , but not late stage granulomas into the hippocampus of rats induces seizure activity [22] . Since the ability of early granuloma extracts to induce seizures could not be attributed to the distinct cytokine profiles within the two stages [23] , we explored the hypothesis that SP was responsible for seizures in this model of NCC , by examining the ability of extracts of early vs . late granulomas to induce seizures and determining whether or not a correlation existed between seizure activity and SP peptide level ( Figure 3 ) . Seizure activity in rats was induced in all cases following intrahippocampal injection of extracts of early stage granulomas ( 12 of 12 ) . In contrast , seizure activity was induced in only 1 of 7 extracts of late granulomas ( p< . 005 , Fishers exact test ) . In addition to the frequency of seizure induction , the severity of seizure activity was greater in rats receiving early vs . late granuloma extracts as determined by behavioral seizure grade ( 3 . 17±0 vs . 0 . 14±0 . 13; p≤0 . 001 , Mann-Whitney test; Figure 3C ) , amplitude of EEG electrical activity ( 1 . 45±0 vs . 0 . 07±0 . 1 mV; p≤0 . 001 , Mann-Whitney test; Figure 3E ) and total duration of seizure activity ( 6 . 43±1 . 11 sec vs . 0 . 74±0 . 65 sec; p≤0 . 01 , Mann-Whitney test; Figure 3G ) . To determine the relationship , if any , between seizure activity and levels of SP peptide within extracts , we measured SP peptide in the extracts and correlated this with the intensity of seizure responses . There was a significantly positive correlation between SP levels and the behavioral seizure grade ( R2 = 0 . 789 , p<0 . 001 , Pearson Correlation Coefficient; Figure 3D ) , the amplitude of seizure activity ( R2 = 0 . 444 , p<0 . 05 , Pearson Correlation Coefficient; Figure 3F ) and duration of seizure activity ( R2 = 0 . 537 , p≤0 . 05 , Pearson Correlation Coefficient; Figure 3H ) . These findings add additional support to the hypothesis that SP contributes to the ability of early granuloma extracts to cause seizures . SP binds to its cognate receptor , NK1R , to mediate its effects . To support the contention that SP within early granulomas mediated the seizures , we examined the effects of pre-treatment with aprepitant , a SP receptor ( NK1R ) antagonist , on seizures induced by early stage granuloma extracts ( Figure 4 ) . Injection of aprepitant ( 1 µg ) 30 min before injection of early stage granuloma extracts resulted in complete abrogation of EEG seizure activity ( p≤0 . 001 , Mann-Whitney test , all comparisons ) . To further support the above observation that SP was important for induction of seizure activity , experiments were performed using transgenic mice devoid of NK1R ( NK1R−/− ) . We thus examined the effect of NK1R deletion on seizure activity induced by extracts of early granulomas ( Figure 5 ) . Similar to the results in rats , extracts of early granulomas obtained from wild type mice induced seizures when injected into the hippocampus of mice generating seizures with behavioral seizure grades of 2 . 5±0 , electrical amplitude of 1 . 54±0 . 2 mV , and total duration of 4 . 63±2 . 13 sec . In contrast , no seizure activity was observed when extracts were injected into NK1R−/− mice ( p≤0 . 05 , Mann-Whitney test , all comparisons ) . To further establish that SP is the tachykinin within extracts of early granulomas that binds to NK1R within the brains of rodents to induce seizures , we examined whether early granulomas obtained from SP precursor knockout ( SPP−/− ) mice were capable of inducing seizures when injected into the hippocampus of wild type mice ( Figure 5 ) . Twelve granulomas from infected SPP−/− mice were examined—7 early stage and 5 late stage . Neither early nor late granuloma extracts from these mice induced seizure activity .
We demonstrated that SP-producing cells are present in the brain of NCC patients and are localized specifically to areas of inflammation adjacent to the degenerating worms . Also , severe seizures occurred in rats after intrahippocampal injection of SP alone , as well as after injection of SP-containing extracts of granulomas obtained from T . crassiceps-infected wild type ( WT ) mice . In addition , seizure activity correlated with SP levels within extracts and was prevented by intrahippocampal pre-injection of aprepitant , a SP receptor antagonist . Furthermore , studies using mice deficient in SP precursor or SP receptor ( NK1R ) demonstrated that SP precursor was required to generate epileptogenic granuloma extracts , and that the SP receptor ( NK1R ) was required to respond to these extracts with seizure activity , respectively . Taken together these findings strongly suggest that SP produced within cysticercal granuloma causes seizures in rodent models of NCC and together with the SP localization findings in NCC patients suggest that seizures can be prevented and/or treated in human NCC with SP receptor antagonists . SP peptide is encoded by the preprotachykin A gene . The protein product of this gene , preprotachykinin A , is cleaved to form two active neuropeptides , SP and neurokinin A . The SP precursor ( preprotachykinin ) knockout mice used in our studies are deficient in both SP and neurokinin A . Consequently , our results in SPP-knockout mice can be attributed to a deficiency in either SP or neurokinin A . The possibility that neurokinin A , and not SP , is mediating seizures in NCC is unlikely for several reasons . First , the SP antibody used in the brain immunohistochemistry studies was specific for SP and did not cross-react with neurokinin A . Also , while neurokinin A binds to NK1R , it binds with much lower affinity than SP; while , the opposite is true for NK2R , NK2R binds neurokinin A with much greater affinity than SP . Our results demonstrated a complete absence of seizures in NK1R-knockout mice , which express NK2R . If neurokinin A was the mediator of seizures in our models of NCC , we would have anticipated persistent epileptogenic activity within granuloma extracts in these mice . The mouse model of cysticercosis due to Taenia crassiceps used in these studies has previously been exploited by us and others as a model to study the molecular pathogenesis of NCC [24]–[27] . Similar to T . solium infection in humans , live T . crassiceps cysts in mice cause little or no inflammation , while dead or dying parasites initiate a granulomatous reaction . The inflammatory response that occurs around dead or dying T . crassiceps cysts can be separated into early stage and late stage granulomas based on the extent of destruction of the parasite , their cytokine profile , and their pattern of SP peptide production [20] , [21]; early stage granulomas contain parasite remnants , produce Th1 cytokines , and SP peptide while late stage granulomas contain neither parasite remnants nor SP and produce IL-4 in addition to Th1 cytokines [20] , [21] . Cytokines such as TNF-α , IL-1β and IL-6 [28] , [29] have been shown to induce seizures , However , in previous studies we were unable to attribute the epileptogenicity of early granuloma extracts to their distinct cytokine profile [20] , [23] . Our current findings provide definitive support to the conclusion that it is not the repertoire of cytokines present within early granuloma that induces seizures , rather , the presence of SP peptide . In addition to demonstrating that SP within cysticercal granulomas causes seizures in a rodent model of NCC , this is the first demonstration that SP alone causes seizures . SP previously had been shown to evoke epileptiform responses in neurons [30] . Also , mice with disruption of the preprotachykinin A gene showed a reduction in the duration and severity of seizures induced by intrahippocampal administration of kainic acid or pentylenetetrazol [19] . In addition , Liu et al demonstrated that while administration of SP ( 10 pmol ) alone into the hippocampus of rats had no effect , SP administration combined with stimulation of the perforant path within the hippocampus resulted in severe self-sustaining status epilepticus ( SSSE ) and generated a pattern of acute hippocampal damage resembling that seen in status epilepticus in patients [18] . Our findings establishing that SP alone causes seizures raises the possibility that it contributes to seizures seen in other infectious and/or inflammatory diseases involving the brain . These studies have potential implications for treatment and prevention of seizures in the setting of NCC especially seizures induced by antihelmintic treatment of NCC patients with viable cysts . Current management options for patients with moderate infections and viable cysts include antihelmintic treatment along with administration of corticosteroids to reduce inflammation and the predisposition to seizures initiated by dying parasites . Antiepileptic drugs are used to treat NCC patients with spontaneously-occurring seizures [31] , [32] and to prevent future attacks; many patients must resign themselves to taking these drugs for the rest of their lives [32] . However , both corticosteroids and antiepileptic drugs can have severe side effects . Our finding that SP causes seizures in NCC precipitated by the granulomatous response to dead or dying cysts suggest the possibility of using SP receptor antagonist for seizure prophylaxis during periods of antihelmintic treatment and , perhaps , as an adjuvant or replacement for anti-epileptic drugs for the treatment and/or prevention of spontaneous seizures in NCC . In addition to being induced by early granuloma surrounding dead or dying cysts , seizures in NCC have been associated with calcified lesions , which presumably represent end-stage granuloma or scaring . Of note , calcified lesions with surrounding edema , in particular , have been associated with episodic seizure activity [33]–[35] . Based on our findings , it is tempting to speculate that SP within the inflammatory response leading to edema around calcified cysts causes seizures in this setting as well . However , support for this hypothesis awaits further studies .
Human studies were approved the Institutional Review Board of Baylor College of Medicine . Human subjects underwent neurosurgical procedures as part of their clinical standard of care and they provided written informed consent to their treating physicians prior to surgery . After diagnostic histopathology evaluation was done , remaining tissues were archived and would ultimately be discarded . Human Institutional Review Board of Baylor College of Medicine approval was obtained to use de-identified paraffin sections from this archival material . Human brain tissue sections derived from 5 patients who were diagnosed with NCC associated seizures were used to probe for SP peptide by immunohistochemistry . Human brain tissue from 2 normal individuals who had died due to other brain-related causes was used as control . Immunoperoxidase staining was performed on 5 µm thick paraformaldehyde-fixed human brain sections by using the avidin-biotin method , an automated immunostainer ( Biogenex ) , and polyclonal rabbit antibody to SP ( 1∶2500; Chemicon , Temecula , Calif . ) or control rabbit serum . Slides were scored on the basis of positive staining demonstrated within the cytoplasm of cells at a level above the level of the nonspecific signal in tissue cells . Five-to-ten high-power fields ( 1000× ) of each slide were scored by an experienced microscopist blinded to the study design . The slides were graded on a scale of 0 to 4+ as follows , 0–0 . 5+- no positive cells , faint diffuse staining; 1+ , 1–10% of cells positive , 2+ , >10–20% of the cells positive; 3+ , >20–30% of the cells positive; and 4+ , >30% of the cells positive . All studies with animals were approved by the Institutional Animal Care and Use Committee of Baylor College of Medicine . Use of all animals involved in this project were carried out according to the provisions of the Animal Welfare Act , PHS Animal Welfare Policy , the principals of the NIH Guide for the Care and Use of Laboratory Animals , and the policies and procedures of Baylor College of Medicine . All surgery was performed under DEA Schedule III anesthesia , and all efforts were made to minimize suffering . Six-week-old WT C57/BL6 mice were purchased from Jackson Laboratories . Tachykinin 1 or SP precursor knockout mice ( SPP−/− ) mice were generated , as described [36] . Briefly , a targeting vector containing neomycin resistance and thymidine kinase genes was used to disrupt the Tac1 gene , replacing the SP exon 3 with neo and deleting the neurokinin A exon 6 . Chimeric animals were backcrossed into the C57BL/6J for 15 generations . SPP−/− mice are viable , fertile , normal in size and do not display any gross physical or behavioral abnormalities . Nociceptive pain responses and neurogenic inflammation are absent . SPP−/− male and female mice were obtained from Dr . Julio Fontan , Southwestern Medical Center , Dallas , Texas and bred within the BCM animal facility . NK-1R−/− were generated as described [37] . Briefly , a targeting vector was generated in which exon 1 of the NK-1R gene was partially deleted , including the initiating methionine codon , and replaced with a cassette encoding lacZ and neomycin resistance . Chimeric animals were backcrossed into the C57BL/6J for at least 15 generations . The homozygous NK1R−/− mice expressed no detectable NK1 receptor peptide , were grossly normal developmentally , were fertile , and appeared to be healthy under barrier isolation conditions . NK1R−/− mice were originally obtained from the laboratory of Dr . Joel Weinstock , Tufts University School of Medicine , Boston , Massachusetts . Adult Sprague Dawley rats ( 125–175 grams ) were purchased from Harlan Laboratories ( Houston , Texas ) . Female wild type and SPP−/− mice ( 15–20 mice per group ) were infected by intraperitoneal inoculation with 10 cysts of the ORF strain of T . crassiceps , as previously described [21] , [38] . All infected animals were housed in a BSL2 facility at Baylor College of Medicine . Three months following infection , the mice were euthanatized by cervical dislocation under anesthesia . Granulomas associated with parasites were identified visually , removed from the peritoneal cavity , portioned into 3 parts and used for different experiments as follows . One part of the granulomas was fixed in 4% paraformaldehyde to be utilized for histological staging , the second part of the granuloma was frozen in liquid nitrogen to be used for quantitation of SP peptide . The remainder was washed once , followed by homogenization in phosphate buffered saline to generate extracts for study of epileptogenic responses . The portion of each granuloma fixed with 4% paraformaldehyde was paraffin-embedded and cut into 5-µm sections , which were then stained with Giemsa and examined microscopically . We separated the granulomas into early and late stages based on the histological appearance of the degenerating parasite according to previously described methods [20] . Early and late stage granulomas were homogenized in 1% trifluoroacetic acid ( TFA ) ( 1 ml/gram of tissue ) and centrifuged at 17 , 000 g for 15 minutes at 2–8°C . A Sep-Pak C18 Cartridge ( Waters , Associates , Milford , MA ) was prewetted with 100% acetonitrile followed by 1% TFA in water . The supernatant was then passed through the cartridge , followed by a wash with 10–20 ml of 1% TFA . Protein was then eluted with 3 ml of a 60∶40 solution of acetonitrile∶1% TFA and dried using a centrifugal concentrator under vacuum . The dried samples were then reconstituted in assay buffer containing protease inhibitor , aprotinin ( 500 KIU/ml , Sigma ) and quantitated using ELISA kit from R&D Biosystems . Results are expressed as picograms of SP in 1 mg of total protein . Total protein was quantitated using the Bradford method ( cat no . 500-0006 , Bio-Rad , Hercules , CA ) . To prepare extracts for seizure experiments , each granuloma was homogenized in phosphate buffer saline . The homogenate was centrifuged at 10 , 000 rpm for 10 minutes and the supernatant used for seizure experiments as underlined below . Total protein in the supernatant was quantitated using the Bradford method ( cat no . 500-0006 , Bio-Rad , Hercules , CA ) . Twenty-four hours after intrahippocampal injection , animals were anesthetized with urethane ( 1 . 2 g/kg ) and perfused through the heart with 4% paraformaldehyde . The brain tissues were equilibrated in 30% sucrose and 35 µm horizontal sections were Nissl stained and examined for the needle track/injection site . Statistical differences were determined using the Mann-Whitney test , Student's t-test or Pearson's correlation test , as indicated .
|
Neurocysticercosis ( NCC ) , is a helminth infection of the brain that is caused by Taenia solium . NCC is the major cause of acquired seizures worldwide . Live Taenia solium parasites in the brain of NCC patients are surrounded by little or no inflammation . Seizures are thought to result not from parasitic infection per se , but from the chronic granulomatous host response initiated by dying cysts . Antiparasitic drugs can be used to kill the parasites , but the symptoms may worsen due to the host inflammatory responses being stimulated by the dying parasites . The mediators that are responsible for mediating seizures in NCC are not known; identification of the seizure mediator ( s ) may lead to prevention/treatment of seizures with specific antagonists . In this important study , we demonstrated that Substance P , a neuropeptide and pain transmitter is responsible for seizures in NCC . These studies have potential implications for treatment and prevention of seizures in the setting of NCC .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"global",
"health",
"parasitic",
"diseases"
] |
2012
|
Substance P Causes Seizures in Neurocysticercosis
|
Plasmodium vivax and Hepatitis B virus ( HBV ) are globally outspread in similar geographic regions . The concurrence of both infections and its association with some degree of protection against symptomatic and/or severe vivax malaria has been already described . Nevertheless , data on how host response to both pathogens undermines the natural progression of the malarial infection are scarce . Here , a large cohort of vivax malaria and HBV patients is retrospectively analyzed in an attempt to depict how inflammatory characteristics could be potentially related to the protection to severe malaria in coinfection . A retrospective analysis of a databank containing 601 individuals from the Brazilian Amazon , including 179 symptomatic P . vivax monoinfected patients , 145 individuals with asymptomatic P . vivax monoinfection , 28 P . vivax-HBV coinfected patients , 29 HBV monoinfected subjects and 165 healthy controls , was performed . Data on plasma levels of multiple chemokines , cytokines , acute phase proteins , hepatic enzymes , bilirubin and creatinine were analyzed to describe and compare biochemical profiles associated to each type of infection . Coinfected individuals predominantly presented asymptomatic malaria , referred increased number of previous malaria episodes than symptomatic vivax-monoinfected patients , and were predominantly younger than asymptomatic vivax-monoinfected individuals . Coinfection was hallmarked by substantially elevated concentrations of interleukin ( IL ) -10 and heightened levels of C-C motif chemokine ligand ( CCL ) 2 . Correlation matrices showed that coinfected individuals presented a distinct biomarker profile when compared to asymptomatic or symptomatic P . vivax patients , or HBV-monoinfected individuals . Parasitemia could distinguish coinfected from symptomatic or asymptomatic P . vivax-monoinfected patients . HBV viremia was associated to distinct inflammatory profiles in HBV-monoinfected and coinfected patients . The findings demonstrate a distinct inflammatory profile in coinfected patients , with characteristics associated with immune responses to both pathogens . These host responses to P . vivax and HBV , in conjunction , could be potentially associated , if not mainly responsible , for the protection against symptomatic vivax malaria .
Malaria still rises major concerns in public health worldwide . The burden caused by the disease is noticeable , as it leads to more than 200 million cases and billions of dollars invested each year [1] . Despite all the investments and increased interest in the pursuit of new interventions [1 , 2] , there was an elevation in the number of estimated cases in the successive years of 2016 and 2017 [1] . Moreover , P . vivax , which is the most widespread of the five main species of Plasmodium [1 , 3 , 4] , have been increasingly associated with severe disease presentations and mortality [1 , 5–8] . Hepatitis B virus ( HBV ) infections are no less of a problem , with more than 250 million chronic cases estimated in 2015 [9] . Incidence of HBV has been reduced since the introduction of the vaccine , however approximately 815 , 000 deaths were accountable to HBV infections and its chronic complications in 2016 [10] . Both hepatitis B and vivax malaria are mainly outspread in tropical countries [1 , 9] , and there is overlapping occurrence of these diseases [11] . HBV-associated tissue damage is described to be directly related to the host inflammatory response against infection [12 , 13] . The immune responses in chronic HBV infections are characterized by decreased T-cell proliferation potential and exhaustion [14–18] . Although not completely understood , these events are thought to be related to a higher release of HBsAg particles ( hence , viral load ) in the circulation , expression of co-inhibitory receptors , and production of IL-10 [14–21] . In vivax malaria , intensity of immune activation is associated with worse clinical outcomes [5 , 6 , 22–24] . On the converse , cases of asymptomatic P . vivax infection are hallmarked by a less pronounced pro-inflammatory response , with increased IL-10 levels in peripheral blood [6 , 11 , 22] . Thus , at first glance , both HBV and P . vivax infections seem to drive similar profiles of systemic inflammation in distinct clinical settings . Nevertheless , no previous study has performed a detailed characterization of systemic immune activation profile in HBV-malaria comorbidity . Another similarity between HBV and malarial biology is the participation of the liver as a key organ part of the immunopathogenesis in both infections [4 , 7] . Notably , severe vivax malaria is associated with remarkable hepatic involvement [5–7 , 22 , 25] , whereas tissue damage is determinant for the presentation of cirrhosis and hepatocellular carcinoma in chronic HBV infections [12 , 16 , 26] . Counterintuitively , HBV infection has been shown to lead to a distinct systemic inflammatory response in Plasmodium infections , resulting in increasing odds for asymptomatic malaria [11] . The present study expands the current knowledge as it examines in detail a rich interplay of cytokines , chemokines and acute phase proteins in a large number of patients infected with P . vivax , HBV or both . These analyzes demonstrate the nuances of different inflammatory responses in confluence , which culminates in an intense but balanced immune response to both pathogens , with key participation of relevant biomarkers as TNF-α , IL-4 , IL-10 and CCL2 .
Written informed consent was obtained from all participants or their legally responsible guardians , and all clinical investigations were conducted according to the principles expressed in the Declaration of Helsinki . The project was approved by the institutional review board of the Faculdade de Medicina , Faculdade São Lucas , Rondônia , Brazil , where the study was performed . The present study is based on analyses performed retrospectively in databank containing immunological , clinical and epidemiological data from 601 subjects , including uninfected controls , recruited between 2006 and 2007 from the state of Rondônia , in the Brazilian Amazon . Multiple investigations have been reported from the project which this study is a part of [5 , 6 , 11 , 22–25 , 27–30] . Patient investigation included both active case detection in the municipalities of Buritis and Demarcação ( Rondônia , Brazil ) and passive case detection from individuals who sought care at Brazilian National Foundation of Health ( FUNASA ) diagnostic centers or at the municipal hospital in Buritis ( Rondônia , Brazil ) . Malaria diagnosis was conducted through microscopic examination of thick smears and nested polymerase chain reaction ( PCR ) evaluation in whole blood samples ( 20mL ) , with control for cross-contamination , performed at the Instituto Gonçalo Moniz ( Fiocruz-BA ) , Salvador , Bahia , Brazil , as previously reported [5 , 6 , 22–24] . Individuals who tested positive through PCR evaluation and persisted without the presentation of fever ( axillary temperature >37 . 8°C ) and/or sweating , chills , jaundice , myalgia , arthralgia , asthenia , nausea , and emesis for 30 days were considered asymptomatic . Patients , which parasitological tests were positive , presenting any symptom listed above , were considered symptomatic . HBV diagnosis was conducted employing the AXSYM automatic ELISA system ( Abbott , Wiesbaden , Germany ) , HBSAg , HBeAg , total anti-HBS , total anti-HBc , anti-HBc IgM and anti-HBe IgG were screened , according to the most updated protocols published by the Brazilian Ministry of Health at the time of study enrollment , and no acute HBV infection was detected ( HBSAg+ , anti-HBS- , anti-HBc IgM+ ) . All the measurements were performed right at the study enrollment and diagnosis of malarial and/or HBV infection , meaning that the collections were performed before the initiation of antimalarial or HBV-specific therapy . Information regarding the number of previous malaria episodes and years that the patients resided in the area at the time of study enrollment were obtained directly from the patients in the interview part of the medical examination . For the present study patients with both symptomatic ( n = 179 ) and asymptomatic ( n = 145 ) P . vivax monoinfection , ongoing HBV infection ( n = 29 ) , concurrent P . vivax and HBV infections ( n = 28 ) and healthy controls ( n = 165 , from which 152 had all the epidemiological data available ) were included . The exclusion criteria for the present study were: patients with documented P . falciparum or HIV infections , tuberculosis , cancer , or use of immunosuppressant drugs . For the analyses of biochemical markers , patients presenting P . vivax monoinfection who were previously infected by HBV were excluded , in order to avoid interferences on the inflammatory profile . In addition , for part of these analyses , P . vivax monoinfected individuals , independently of clinical status ( symptomatic or asymptomatic ) , were considered as a single group ( n = 268 ) , to compare and attest if the factors involved in the clinical presentation from coinfected subjects would also differ P . vivax-HBV coinfection from the P . vivax monoinfection overall . Clinical , demographic and epidemiological characteristics of the participants included in the current study are described in Tables 1 and 2 and S1 . Plasma levels of cytokines IL1-β , IL-4 , IL-6 , IL-10 , IL-12p70 , IFN-γ , tumor necrosis factor ( TNF ) -α , C-C motif chemokine ligand ( CCL ) 2 , CCL5 , C-X-C motif chemokine ligand ( CXCL ) 9 , and CXCL10 were measured using the Cytometric Bead Array—CBA ( BD Biosciences Pharmingen , San Diego , CA , USA ) , according to the manufacturer’s protocol . The measurements of aspartate amino-transferase ( AST ) , alanine amino-transferase ( ALT ) , total bilirubin , direct bilirubin , creatinine , fibrinogen and C-reactive protein ( CRP ) were performed at the Pharmacy School of the Federal University of Bahia and at the clinical laboratory of Faculdade São Lucas . The median values with interquartile ranges ( IQR ) were used as measures of central tendency and dispersion . Chi-square test was used to compare frequencies between the study groups . Continuous variables were compared between the study groups using the Mann-Whitney U test ( 2-group comparisons ) , or the Kruskall-Wallis test with Dunn’s multiple comparisons ad hoc test ( between 3 or more groups ) . Hierarchical cluster analyzes were performed using the Ward’s method with bootstrap ( 100X ) . Spearman tests were performed to analyze correlations and to build the correlation matrices , which assessed markers in each study group . Only correlations with Spearman rank ( r ) values above 0 . 6 were plotted in the matrices . A p-value below 0 . 05 after adjustment for multiple measurements ( false discovery rate of 1% ) was considered statistically significant . The statistical analyzes were performed using Graphpad Prism 7 . 0 ( GraphPad Software Inc . , San Diego , CA , USA ) , and JMP 12 . 0 ( SAS , Cary , NC , USA ) .
The baseline characteristics of the study population are shown in Table 1 . The study groups were similar with regard to sex . Among P . vivax-infected individuals , asymptomatic patients were older than symptomatic patients and those coinfected with HBV ( median age: 43yrs , IQR: 34–52 vs . 29yrs , IQR: 19–42 vs . 31yrs , IQR: 23-46yrs , respectively ) ( Table 1 ) . In addition , asymptomatic malaria patients were older than healthy endemic controls but with similar median age than those with HBV monoinfection ( Table 1 ) . Of note , referred number of previous malaria episodes was lower in patients presenting with symptomatic malaria compared to those with asymptomatic malaria , malaria-HBV coinfection and those with HBV monoinfection ( Table 1 ) . Individuals with symptomatic P . vivax infection more frequently referred that they had lived for shorter time in the malaria endemic area when compared with the other clinical groups ( Table 1 ) . As expected according to previous reports [11] , parasitemia , assessed in thick blood smears , was substantially lower in individuals with HBV-malaria comorbidity compared to those with symptomatic P . vivax monoinfection ( median: 753 parasites/μL , IQR: 444 . 3–4 , 262 vs . 6 , 324 , IQR: 913 . 5–60 , 623 , P = 0 . 0004 ) , whereas asymptomatic malaria patients predominantly did not exhibit detectable number of parasites in peripheral blood using microscopic examination ( Table 1 ) . In addition , frequency of P . vivax-HBV coinfection was significantly higher in asymptomatic individuals ( n = 25 ) . Overall , 18 ( 13 . 04% ) asymptomatic and 38 ( 21 . 23% ) symptomatic vivax malaria patients presented serological status compatible with previous history of HBV infections ( HBSAg- , anti-HBS+ , anti-HBc+ ) . Furthermore , all cases of HBV infection , with or without malaria co-infection , presented serological status of chronic infection . Among the 179 symptomatic vivax malaria patients , eighteen presented severe/complicated vivax malaria , and six individuals eventually died from the disease . Detailed information on symptoms presented by individuals from each clinical group is available in S1 Table . Median values of all biochemical markers per group were log-transformed and z-score normalized for hierarchical cluster analysis . Using this approach , three clusters of markers were identified ( Fig 1A ) . Fold differences of the circulating levels of all biomarkers were then calculated to assess which parameters were differentially expressed in all four main subpopulations against healthy controls ( Fig 1A ) . Asymptomatic vivax malaria patients presented a similar number of variables with significant concentrations increases ( mainly IFN-γ , IL-10 and direct bilirubin ) and decreases ( such as CXCL10 , IL-1β , IL-4 and indirect bilirubin ) . Patients with HBV monoinfection presented the same number of variables which concentrations were increased ( mainly IFN-γ , TNF-α , IL-6 and CXCL9 ) , when compared to asymptomatic vivax malaria monoinfection , but only significant decrease of one variable ( also IL-4 ) . Furthermore , symptomatic vivax patients presented augmented levels of almost every analyzed analyte , except for IL-10 , CCL5 , CXCL10 and specifically CCL2 , which levels were diminished in comparison to uninfected controls ( Fig 1A ) . P . vivax-HBV coinfected individuals presented multiple significant elevations in biomarker values when compared to healthy controls , although not as many as found in symptomatic malaria monoinfection . Noteworthy , coinfected patients exhibited an impressive 22-fold elevation in IL-10 levels when compared to healthy controls , and remarkable decreases in IL-8 concentrations ( Fig 1A ) . Other significant differences are illustrated in Fig 1A . Fig 1B and 1C show Venn’s diagrams to further illustrate and depict these differences and similarities initially demonstrated by the subpopulations . These results delineate the systemic inflammatory profile associated with this comorbid condition . Detailed information of the laboratorial results and analysis in the subpopulations are available in Table 2 . Fold differences of the circulating levels of all biomarkers were also calculated to assess which parameters were differentially expressed in coinfected individuals in comparison to other main study groups ( coinfected vs . asymptomatic or symptomatic P . vivax monoinfected patients , and HBV monoinfected individuals ) . Patients with malaria-HBV coinfection presented elevated concentrations of multiple variables such as IFN-γ , TNF-α , IL-4 , IL-10 , CCL2 , and reduced levels of IL-12 and creatinine levels when compared to those with asymptomatic vivax malaria monoinfection ( Fig 2A ) . When compared to symptomatic vivax malaria patients ( Fig 2A , ) , coinfected individuals presented elevated levels of IFN-γ , IL-10 and CCL2 , and diminished plasma concentrations of multiple variables as TNF-α , IL-6 , IL-12 , and CRP . When compared to those with HBV monoinfection , coinfected patients presented elevated levels of IL-4 , IL-10 , CCL2 , CRP , fibrinogen , and direct bilirubin , and reduced concentrations of TNF-α and IL-8 ( Fig 2A ) . Other significant differences are illustrated in Fig 2A . Thus , in summary , IL-10 and CCL2 were the only variables which coinfected patients presented with elevated concentrations in comparison to all the other three main study groups . Then , considering the immunoregulatory nature of IL-10 and the dimension of its elevations in coinfected individuals , the next step was to analyze the behavior of the biomarkers in comparison to IL-10 levels in all main study groups . Coinfected individuals presented reduced IL-10 ratios for all variables ( S1 Fig ) , with the exception of IFN-γ , IL-4 and CXCL10 ( Fig 2B ) . The IFN-γ/IL-10 and CXCL10/IL-10 ratios could not distinguish coinfected and asymptomatic vivax patients . In addition , HBV-monoinfected and P . vivax-HBV coinfected patients could not be distinguished by their IL-4/IL-10 ratio values . These results highlight similar biosignatures that may be reminiscent from each respective infection in P . vivax-HBV coinfected individuals . Multiple correlation matrices were inputted into a network analysis to assess the profile of associations between cytokine levels in each study subpopulation ( Fig 3A ) . It was noticeable the decreased number of significant connections ( which represent statistically significant correlations ) in the network of asymptomatic vivax patients ( Fig 3A ) when compared to the networks calculated from the other groups . This tendency is also maintained when such networks were compared to that from uninfected controls ( S2 Fig ) . In addition , the correlation matrix of individuals with symptomatic P . vivax monoinfection showed an increase of significant positive connections between the biochemical parameters ( Fig 3A ) . Interestingly , while displaying an increased number of significant positive correlations between variables , P . vivax-HBV coinfected individuals ( Fig 3A ) also exhibited the tendency of negative connections presented by HBV-monoinfected patients ( Fig 3A ) . HBV patients presented negative correlations between IL-4 and IFN-γ , IL-1β , and IL-12p70 concentrations , between CCL2 and IFN-γ or IL-1β levels , and between total bilirubin and creatinine levels . In coinfected individuals , IL-4 and IL-12p70 levels were again negatively correlated , but also between IL-8 and AST , fibrinogen , direct bilirubin , and total bilirubin concentrations . In addition , IFN-γ was only positively correlated with CXCL9 and CXCL10 , whereas IL-1β levels were negatively correlated with concentrations of CCL5 , ALT and total bilirubin ( Fig 3A ) . The further step was to analyze the correlations between the biomarkers and viral load , which has been previously associated with a downregulated proinflammatory response in individuals chronically infected with HBV [14 , 15] . Overall , concentrations of ten biomarkers were significantly correlated with viremia levels in patients with HBV monoinfection or in those with HBV-malarial coinfection . Furthermore , TNF-α did not presented the same correlation pattern in the two groups , which was the case of IFN-γ concentrations , for example ( Fig 3B ) . In HBV-monoinfected individuals , TNF-α levels was not correlated with the viral load , while presenting negative correlation with viremia in coinfected patients . The other significant correlations to viral load are shown in S2 Fig . We next performed additional analyses in which symptomatic and asymptomatic P . vivax monoinfected subjects were considered as a single group ( S3 Fig ) . When compared with uninfected controls , monoinfected P . vivax individuals presented the characteristic significant reduction of CXCL10 and CCL2 levels and increases of IL-10 and TNF-α levels ( S3A Fig ) found in previously separated groups ( Fig 1A ) . In addition , when compared against coinfected patients , monoinfected P . vivax individuals presented significant elevations of AST , ALT , CRP , IL-8 and IL-12 ( S3A Fig ) , and significant reductions in IFN-γ , CXCL10 , CCL2 , IL-4 and IL-10 ( S3A Fig ) . The IFN-γ/IL-10 and CXCL10/IL-10 ratios could distinguish coinfected patients from both groups of P . vivax and HBV-monoinfected patients ( S3B Fig ) . S3C Fig shows the distribution of the patients based on their parasitemia values . When compared with those presenting HBV coinfection , P . vivax-monoinfected patients presented with significantly reduced parasite counts ( S3D Fig ) .
In the present study , we performed novel analyses of multiple inflammatory biomarkers related to key immune and inflammatory responses associated with disease progression in the context of HBV and Plasmodium vivax infections . These expanded analyses provide deeper comprehension of the immune response against P . vivax-HBV coinfection , which culminates with reduced odds of severe disease and progression of vivax malaria [11] . In the study population , asymptomatic vivax and coinfected patients presented distinct epidemiological profiles . Elevated number of previous malaria episodes and more advanced age are well-known to be associated with milder and asymptomatic vivax malaria [6 , 11 , 31] . However , while referring a similarly increased number of previous malaria episodes , coinfected patients were predominantly younger than asymptomatic vivax patients ( Table 1 ) . Furthermore , coinfected individuals presented similar median age to symptomatic P . vivax malaria patients ( Table 1 ) . Moreover , asymptomatic monoinfected vivax patients predominantly presented with undetectable parasitemia examined by thick smears , whereas coinfected individuals , predominantly asymptomatic from a malarial perspective , presented an elevated values of parasite counts ( S3C Fig ) . In fact , coinfected individuals presented significantly increased parasite counts when compared to P . vivax monoinfection ( S3D Fig ) . Hence , these distinct epidemiological and serological characteristics of asymptomatic vivax and coinfected patients argues that other factors may have influenced malarial presentations in coinfection with HBV . Moreover , the parasitemia results are in line with previous hypothesis that mainly the host response to infection , and not the parasite load alone , are responsible for clinical presentations in vivax malaria [5 , 6 , 22–24] . Multiple cytokines , chemokines and acute phase proteins were then profiled to further analyze the mechanisms associated with disease presentation in P . vivax-HBV coinfected patients . Coinfection was hallmarked by extremely elevated concentrations of IL-10 , as well as heightened levels of CCL2 , in comparison to the distinct clinical presentations of P . vivax infections or HBV monoinfection . IL-10 is an immunoregulatory cytokine and its levels were previously reported to be closely associated with disease progression and outcomes in both hepatitis B and vivax malaria . Patients with severe vivax malaria have been shown to present unbalanced concentrations of IL-10 against levels of proinflammatory biomarkers , when compared to individuals with uncomplicated vivax malaria [5 , 6 , 22] . In addition , these individuals with uncomplicated or asymptomatic P . vivax infections have been shown to express relatively augmented concentrations of IL-10 when compared to those with symptomatic or severe vivax malaria [5 , 6 , 31 , 32] . In viral infections , IL-10 levels are associated with diminished T-cell activation , which may already start to occur rapidly after infection [20] . Furthermore , HBV actively suppresses immune responses [14] and augmented IL-10 levels are closely associated with viral persistence [16–18 , 33] . Herein , as expected , HBV-infected patients presented increased IL-10 levels when compared to uninfected controls . Furthermore , only symptomatic vivax malaria patients could not be distinguished from uninfected controls by their IL-10 levels . Although coinfected patients presented almost an 8-fold increase in IL-10 concentrations when compared to asymptomatic vivax-monoinfected individuals , they presented undistinguishable values of IFN-γ/IL-10 ratios , which highlights a similar tendency of immune balance in this aspect of inflammatory response . A higher baseline concentration of IL-10 , associated with the condition of antiviral response , and hence a distinct overall inflammatory profile , could then be responsible for the difference in absolute IL-10 levels identified between coinfected and asymptomatic vivax patients . Both study groups also presented similar CXCL10/IL-10 ratio values . CXCL10 is an IFN-γ induced protein which acts in chemotaxis , apoptosis , cell proliferation and angiogenesis [34] . Thus , similar results of CXCL10/IL-10 and IFN-γ/IL-10 are not surprising . Even with these solid findings , experimental models are still necessary to further define and ratify whether these similarities were directly carried from responses associated to HBV persistence and T-cell exhaustion , or from the antimalarial response , or from concurrent responses to both pathogens . CCL2 , the other biomarker which concentrations are augmented in coinfected individuals in comparison to all other study groups , is an important chemokine involved in recruitment of monocytes [35 , 36] and NK cells [36] . CCL2 is reported to be produced by hepatocytes under acute HBV stress [37] and was previously associated with uncomplicated P . vivax infections [31] . However , without the weight of an acute and heavy antiviral response , an infection with a pathogen known to induce acute hepatocyte damage as P . vivax [6] could possibly trigger this local chemokine production . ALT and AST levels could reinforce this hypothesis as they were found to be augmented in symptomatic vivax malaria patients when compared to HBV-monoinfected individuals ( Table 2 ) , while only being correlated to viremia in coinfected P . vivax-HBV patients and not in HBV-monoinfected subjects ( S2 Fig ) . Herein , coinfected individuals , which presented predominantly asymptomatic malarial infection , had a 2 . 2-fold increase in CCL2 concentrations when compared to patients with symptomatic P . vivax infections . Therefore , these results may reinforce the association of CCL2 with uncomplicated malaria . It is also reported that CCL2 influences and directs CD4+ T lymphocytes to a more biased response towards IL-4 production [38] . Herein , coinfected individuals presented significant elevations IL-4 levels when compared to healthy controls ( Fig 1A ) , asymptomatic vivax or HBV-monoinfected subjects ( Fig 2A ) . Although IL-4 levels could not distinguish coinfected and symptomatic vivax malaria patients , correlations with IL-4 concentrations were completely different in both study groups . IL-4 concentrations were negatively correlated to IL-12p70 levels in coinfected patients , while being positively correlated with multiple other proinflammatory cytokines in symptomatic vivax patients ( Fig 3A ) . Hence , these antagonic tendencies suggests that different mechanisms , and not just the antimalarial response in this case , could be responsible for the elevation of IL-4 concentrations in coinfected and symptomatic vivax malaria patients . In addition , IL-4/IL-10 ratio values could not distinguish coinfected and HBV-monoinfected individuals ( Fig 2B ) . This similar profile displayed in both groups of patients infected by HBV suggests that antiviral or responses associated with hepatocyte stress , possibly under CCL2 influence in this hypothesis , could be responsible for these elevations of IL-4 levels in coinfected individuals . In practical terms , this immune response of coinfected individuals with augmented IL-4 concentrations happens without much proinflammatory pressure , as IL-10 immunoregulatory mechanisms should be expected to bring them a more balanced inflammatory response , oppositely to what occurs in symptomatic vivax individuals ( Fig 1A ) . This augmented production of IL-10 alongside IL-4 heightened levels can directly downregulate key proinflammatory cytokines such as TNF-α [39 , 40] , and thus have a protective effect against severe malaria presentations . Herein , TNF-α concentrations were found to be significantly reduced in coinfected individuals , when compared to both HBV or symptomatic vivax malaria patients ( Fig 2A ) . Furthermore , TNF-α concentrations were negatively correlated to viremia only in patients with P . vivax-HBV coinfections ( Fig 3B ) , which could be read as a possible effect of these previously reported mechanisms in the patients from the present study . These results are compatible with the hypothesis that coinfection drives reduction of systemic inflammation , which we previously published [11] . Therefore , these confluent events from responses to both pathogens ( increased production of IL-10 , CCL2 protective role in malaria , as well as combined effects of IL-4 and IL-10 ) could enable the host to respond properly without unbalanced inflammation . Thus , this proper response creates an environment unfavorable for the Plasmodium to thrive and induce detectable symptoms . Our study presented some limitations . We did not have data available from follow-up of the HBV-infected patients and their antiviral treatments , as they were referred to a specialized service . Although we collected information regarding number of previous malaria episodes , these data were expressed by the patients , and not extracted from official documents of the health centers . Thus , this fact further limits the analysis and evaluation of relapses in the study patients , and if these events could be related to an association between the chronic HBV infection and hypnozoite activation . Experimental models and biopsies would have helped with the assessment of T-cell exhaustion , the impact of liver involvement into inflammatory responses , and cytokine evaluation at tissue level . Therefore , further longitudinal and experimental studies are still necessary to completely understand the events associated with P . vivax-HBV coinfection . However , despite some limitations , the present study was successful to analyze several biomarkers and their associated biomechanisms , and link them to the known protective effect of chronic HBV infections in vivax malaria . In conclusion , the results presented here represent a translation of an increased demand and pressure caused by the acute P . vivax infection on the immune system of a chronically HBV-infected host . Hence , there is an augmented presence of inflammatory biomarkers as IFN-γ and CRP , counterbalanced with the immunoregulatory mechanisms discussed here . In summary , coinfection was hallmarked by substantially increased levels of IL-10 and augmented concentrations of CCL2 . CCL2 is expressed by hepatocytes during acute injury , reportedly leads to IL-4 increases , while IL-10 is directly related to viral persistence and T-cell exhaustion , and both cytokines are associated with protection in P . vivax infections . Thus , these results argue that distinct mechanisms associated with antiviral and antimalarial activity are due to changes in cytokine balance , and lead to the known increased odds of asymptomatic vivax malaria in coinfected HBV-P . vivax patients . This knowledge of responses to both pathogens counteracting proinflammatory responses helps to depict the pathophysiology associated with the coinfection , and could prove relevant to future studies and approaches with immunotherapy in cases of severe malaria or HBV infection .
|
The determinants of the diverse clinical presentations of Plasmodium vivax malaria are not completely understood . Previous studies have reported that P . vivax-HBV coinfection is associated with increased odds of presenting with asymptomatic malaria , but little is known about the immune mechanisms driving such association . To illuminate host pathways associated with protection against malaria , we analyzed multiple cytokines , chemokines and acute phase proteins in groups of patients from the Brazilian Amazon with different presentations of vivax malaria monoinfection , HBV monoinfection , and P . vivax-HBV coinfection . The results indicate that coinfection is hallmarked by a conjunction of immune responses , related to each one of the monoinfections , that results in a balanced inflammation associated with clinical immunity and absence of symptoms . In biological terms , the readouts are that the combined responses to each pathogen would induce a distinct profile of systemic immune activation , with the hallmarked activity of IL-10 , a classical immunoregulatory cytokine , in confluence mainly with CCL2 and IL-4 activity . These multiple pathways would prevent the unbalanced proinflammatory activity associated with symptomatic and/or severe vivax malaria . Moreover , these findings highlight the importance of the immune system in driving disease presentation , raise discussion of immunotherapy in vivax malaria , and how these approaches have the potential to influence clinical outcomes .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"plasmodium",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"biomarkers",
"parasitic",
"diseases",
"hepatitis",
"b",
"virus",
"parasitology",
"viruses",
"developmental",
"biology",
"apicomplexa",
"signs",
"and",
"symptoms",
"molecular",
"development",
"infectious",
"diseases",
"inflammation",
"medical",
"microbiology",
"microbial",
"pathogens",
"hepatitis",
"viruses",
"immune",
"response",
"immune",
"system",
"biochemistry",
"diagnostic",
"medicine",
"physiology",
"viral",
"pathogens",
"co-infections",
"biology",
"and",
"life",
"sciences",
"malaria",
"organisms"
] |
2019
|
Chronic hepatitis B virus infection drives changes in systemic immune activation profile in patients coinfected with Plasmodium vivax malaria
|
Envenoming induced by Bothrops snakebites is characterized by drastic local tissue damage that involves an intense inflammatory reaction and local hyperalgesia which are not neutralized by conventional antivenom treatment . Herein , the effectiveness of photobiomodulation to reduce inflammatory hyperalgesia induced by Bothrops moojeni venom ( Bmv ) , as well as the mechanisms involved was investigated . Bmv ( 1 μg ) was injected through the intraplantar route in the right hind paw of mice . Mechanical hyperalgesia and allodynia were evaluated by von Frey filaments at different time points after venom injection . Low level laser therapy ( LLLT ) was applied at the site of Bmv injection at wavelength of red 685 nm with energy density of 2 . 2 J/cm2 at 30 min and 3 h after venom inoculation . Neuronal activation in the dorsal horn spinal cord was determined by immunohistochemistry of Fos protein and the mRNA expression of IL-6 , TNF-α , IL-10 , B1 and B2 kinin receptors were evaluated by Real time-PCR 6 h after venom injection . Photobiomodulation reversed Bmv-induced mechanical hyperalgesia and allodynia and decreased Fos expression , induced by Bmv as well as the mRNA levels of IL-6 , TNF-α and B1 and B2 kinin receptors . Finally , an increase on IL-10 , was observed following LLLT . These data demonstrate that LLLT interferes with mechanisms involved in nociception and hyperalgesia and modulates Bmv-induced nociceptive signal . The use of photobiomodulation in reducing local pain induced by Bothropic venoms should be considered as a novel therapeutic tool for the treatment of local symptoms induced after bothropic snakebites .
Bothropic envenomation is characterized by severe local manifestation associated with oedema , myonecrosis , hemorrhage and intense pain [1–4] caused by the toxic action of venom components and aggravated by induced-inflammation . The local effects induced by bothropic venoms are the result of multifactorial and synergistic actions of toxins , which are still poorly understood . Bothrops moojeni is a venomous snake responsible for most of the snakebites in the Central region of Brazil [5] . Despite the medical importance , there are only a few studies related to the local inflammatory reaction caused by Bothrops moojeni venom ( Bmv ) . In this sense , the literature shows that in the accidents caused by these snakes serious local complications occur , including a prominent edema formation , intense pain , swelling and pallor , which may develop into more severe outcomes such as muscle mass loss , neuropathy , and amputation [6 , 7] . Currently , the most effective treatment for Bothrops snakebites accidents is the antivenom therapy ( AV ) . However , although AV has proven to be effective in reversal the systemic response , its administration does not prevent local effects and resultant disabilities [3] . Consequently , there is a need to find therapeutic approaches associated with AV treatment that can be effective in reducing the local effects caused by Bothrops snakes envenoming in order to minimize or prevent the progression to a severe clinical status observed after Bothrops snakebites [8 , 9] . Photobiomodulation is a form of light that triggers biochemical changes within cells , where the photons are absorbed by cellular photoreceptors and triggers chemical alterations [10] . The mechanisms of photobiomodulation essentially rely on particular visible red and infrared light waves in photoreceptors within sub-cellular components , particularly the respiratory chain within mitochondrial membranes due to the activation of various transcription factors by the immediate chemical signaling molecules produced from mitochondrial stimulation [11] . The most important of these signaling molecules are thought to be Adenosine Triphosphate ( ATP ) , cyclic-AMP , nitric oxide ( NO ) and Reactive Oxygen Species ( ROS ) [12] . Many studies have demonstrated analgesic and anti-inflammatory effects provided by photobiomodulation in both experimental [13 , 14] and clinical trials [15 , 16] . Photobiomodulation has also proven to be an interesting and efficient complementary alternative for the treatment of local effects caused by bothropic venom through the ability of decreasing the observed local effects , such as myonecrosis [17 , 18]; inflammation [19–22] hemorrhage [21] and pain [20 , 23] . In this context , we have recently demonstrated that photobiostimulation with LLLT and light emitting diode ( LED ) reverse edema formation , local hemorrhage and inflammatory hyperalgesia induced by Bohtrops moojeni venom ( BmV ) in mice [18 , 24] . Although some studies have demonstrated the effectiveness of photobiomodulation in reducing hyperalgesia and allodynia induced by bothropic venom , the mechanism involved in this effect still remains unknown . In this context , the present experiments were designed to investigate the antinociceptive effect of photobiomodulation on BmV-induced allodynia and hyperalgesia and to explore possible underlying mechanisms .
Male Swiss mice weighing 20–25 g , age-matched , were used throughout this study . Animals were maintained under controlled light cycle ( 12/12 h ) and temperature ( 21 ± 2°C ) with free access to food and water . All animal experimentation protocols received the approval by the Ethics Committee on the Use of Animals at of Hospital Sírio-Libanês ( Protocol no . ( CEUA 2010/01 ) , in agreement with Brazilian federal law ( 11 . 794/2008 , Decreto n° 6 . 899/2009 ) . We followed institutional guidelines on animal manipulation , adhering to the “Principles of Laboratory Animal Care” ( National Society for Medical Research , USA ) and the “Guide for the Care and Use of Laboratory Animals” ( National Academy of Sciences , USA ) . Bothrops moojeni venom ( Bmv ) was supplied by the Serpentarium of the Center of Studies of Nature at UNIVAP . Bmv was lyophilized , kept refrigerated at 4°C and diluted in sterile saline solution ( 0 . 9% ) immediately before use . Bmv was injected into the subplantar surface of the right hind paw at the concentration of 1 . 0 μg/50 μL . Equine antivenom ( AV ) used in the experiments was a polyvalent Bothrops AV ( lot# 990504–18 ) raised against a pool of venom from B . alternatus , B . jararaca , B . jararacussu , B . cotiara , B . moojeni and B . neuwiedi obtained from the Butantan Institute ( São Paulo , SP , Brazil ) . AV was injected through the intravenous route ( 0 . 2 μL of AV diluted in saline; final volume of 50 μL , considering that 1 mL of AV neutralizes 5 mg of Bothropic venom [25] 30 min after BmV injection . Hyperalgesia and allodynia of the hind paw were assessed as described by Takasaki et al . [17] . Mice were placed individually in plastic cages with a wire bottom , which allowed access to their paws . To reduce stress , mice were habituated to the experimental environment one day before the first measurement . At the day of the test , the animals were placed in the cages 30 min before the beginning of each measurement and received an injection of 1 . 0 μg of crude Bmv diluted in 50 μL of sterile saline into the subplantar surface of the right hind paw . Control group animals received the same volume of sterile saline . Von Frey filaments with bending forces of 0 . 407 g ( 3 . 61 filament—allodynia stimulus ) , 0 . 692 g and 1 . 202 g ( 3 . 84 and 4 . 08 filaments—hyperalgesia stimulus ) were pressed perpendicularly against the plantar skin and held for 5 s , at 1 , 3 , 6 and 24 h after venom injection . A stimulation of the same intensity was applied three times to each hind paw at intervals of 5 s . The responses to these stimuli were ranked as follows: 0 , no response; 1 , move away from von Frey filament and 2 , immediate flinching or licking of the hind foot . The nociceptive score was calculated as follows: Nociceptive score ( % ) = Σ ( average score of each animal ) x 1002 x number tested animals Animals were returned to their home cages with free access to food and water between the 1 and 3 h , 3 and 6 h and 6 and 24 h measurements . A low-level semiconductor Ga-As laser , Theralaser D . M . C . ( São Carlos , SP , Brazil ) , operating with a wavelength of red 685 nm , was used through the experiments with a beam spot of 0 , 2 cm2 and an output power of 30 mW , energy density of 2 . 2 J/cm2 and exposure time of 15 s . Laser doses , low enough to avoid any thermal effect , were chosen on the basis of previous study from our laboratory [18] . Animals were gently manually restrained and the LLLT was applied to the same area as the injection of Bmv or saline solution . A control group was treated using the same experimental procedure but with the laser turned off . Animals were irradiated 30 min and 3 h after subplantar injection of either Bmv or saline and were immediately returned to their home cages with free access to food and water after each application . Experiments were conducted in an environment with partial obscurity to not suffer interference from external light . The output power of the laser equipment was measured using the Laser Check1power meter ( MM Optics , São Carlos , Brazil ) . Six hours after the intraplantar ( i . pl . ) injection of Bmv or saline , mice were deeply anesthetized with ketamine hydrochloride ( 100 mg/kg ) and xylazine ( 10 mg/kg ) and transcardially perfused with phosphate-buffered saline and 4% paraformaldehyde in 0 . 1 M phosphate buffer ( PB; pH 7 . 4 ) . The spinal cord ( L4 and L5 ) was removed , left in the same fixative for 5–8 h and then cryoprotected overnight in 30% sucrose . Thirty μm frozen sections were immunostained for Fos expression . The spinal cord sections were incubated free floating with a rabbit polyclonal antibody against the nuclear protein which is the product of the early response gene c-fos ( Ab-5; Calbiochem , CA/USA ) , and diluted 1:1000 in PB containing 0 . 3% Triton X-100 plus 5% of normal goat serum . Incubation with the primary antibody was conducted overnight at 24°C . After three washes ( 10 min each ) in PB , the sections were incubated with biotinylated goat anti-rabbit sera ( Vector Labs , Burlingame , CA ) diluted 1:200 in PB for 2 h at 24°C . The sections were washed again in PB and incubated with the avidin-biotin-peroxidase complex ( ABC Elite; Vector Labs ) . After the reaction with 0 . 05% 3–3’ diaminobenzidine and a 0 . 01% solution of hydrogen peroxide in PB and intensification with 0 . 05% osmium tetroxide in water , the sections were mounted on gelatin- and chromoalumen-coated slides , dehydrated , cleared , and coversliped . The material was then analyzed on a light microscope , and digital images were collected . A quantitative analysis of the immunolabeled material was analyzed using a light microscope and the NIS Elements F3 . 0 Image analysis system ( Nikon Instruments Inc . , USA ) . A quantitative analysis was performed on the density of nuclei representative of thle immunoreactivity for Fos ( Fos-IR ) in: a ) the dorsal horn of the spinal cord ( DHSC; superficial laminae-I to IV according to the classification of Rexed . Measurements were taken from 10 different sections for each animal analyzed , including areas that were defined for each structure by using a 20 x objective for the DHSC . Measurements were performed with the program Image J and the operator was blinded to the animal treatment group . Total RNA was isolated from subplantar muscles and spinal cord by TRIzol reagent ( Gibco BRL , Gaithersburg , MD ) , according to the manufacturer's protocol . RNA was subjected to DNase I digestion , followed by reverse transcription to cDNA , as previously described [26] . PCR was performed in a 7000 Sequence Detection System ( ABI Prism , Applied Biosystems , Foster City , CA ) using the SYBRGreen core reaction kit ( Applied Biosystems ) . Primers used are described in Table 1 . Quantitative values for IL-6 , IL-10 , TNF-α , kinin B1 and B2 receptors , CAPDH and mRNA transcription were obtained from the threshold cycle number , where the increase in the signal associated with an exponential growth of PCR products begins to be detected . Melting curves were generated at the end of every run to ensure product uniformity . The relative target gene expression level was normalized on the basis of GADPH expression as endogenous RNA control [27] . Results are expressed as a ratio relative to the sum of GAPDH transcript levels as internal control . Results were expressed as the mean±SEM . Statistical analyses of data were generated by using GraphPad Prism , version 4 . 02 ( GraphPad ) . A value of p<0 . 05 indicated a significant difference . Statistical comparison of more than two groups was performed using analysis of variance ( ANOVA ) , followed by Bonferroni’s test . Statistical comparison for treatment over time was performed using two way ANOVA followed by Bonferroni’s test .
We initially investigated the effects of photobiomodulation on the allodynia and hyperalgesia induced by Bmv . We found that animals injected with Bmv showed significant mechanical allodynia and hyperalgesia when compared with baseline measurement taken before the test , as indicated by basal threshold in response to stimulation by von Frey filaments observed from 1st h after Bmv injection up to 24 h ( Fig 1 ) . Photobiomodulation treatment applied 30 min and 3 h after Bmv injection reversed mechanical allodynia of mice in all evaluated times ( Fig 1A ) . Regarding hyperalgesia , LLL was able to interfere with mechanical sensitivity evaluated by 3 . 84 filament in all evaluated times ( Fig 1B ) however , for the 4 . 08 filament the reversion of hyperalgesia was observed only at the 3rd h of evaluation ( Fig 1C ) . AV treatment itself did not interfere with mechanical sensitivity of mice ( Fig 1 ) . As demonstrated in Fig 2 , intraplantar administration of Bmv induced a significant increase of Fos immunoreactivity observed in the dorsal horn of the spinal cord of animals injected with Bmv ( 42 . 75 ± 3 . 26 ) when compared to the saline group ( 10 . 65 ± 1 . 61 ) . Photobiomodulation treatment significantly decreased Fos expression ( 26 . 58 ± 3 . 58; Fig 2 ) . Cytokine production was evaluated on samples obtained from either spinal cord or footpad of animals previously evaluated at the nociceptive tests . As shown in Fig 3 , the mRNA concentrations of IL-6 and TNF-α increased significantly at 6 h after Bmv injection in the footpad of mice when compared with control group ( Fig 3A and 3B ) . After laser treatment , a significant reduction of both IL-6 and TNF-α mRNA levels was found . Moreover , treatment with AV did not significantly interfere with either IL-6 or TNF-α mRNA levels . However , concomitant treatment of mice with AV and phtobiomodulation decreased both IL-6 and TNF-α mRNA levels ( Fig 3A and 3B ) . Furthermore , no changes on IL-6 and TNF-α were observed in samples from spinal cord of mice ( Fig 3D and 3E ) . IL-10 mRNA levels were decreased after Bmv injection on both footpad and spinal cord of mice . Photobiomodulation treatment increased IL-10 levels in both footpad and spinal cord samples ( Fig 3C and 3F ) . AV treatment did not interfere with IL-10 levels , however it prevented the decrease of this cytokine on samples from spinal cord ( Fig 3F ) . A significant increase on mRNA expression of kinin B1 receptors was observed on Bmv-treated mice when compared to the control group ( Fig 4A ) . LLLT , AV and the association of LLLT and AV induced a significant decrease of mRNA levels of kinin B1 receptors when compared with Bmv-treated animals ( Fig 4A ) . Kinin B2 receptors mRNA expression was also significantly increased in envenomed mice paw when compared to control group ( Fig 4B ) . Once again , LLLT or AV treatment decreased mRNA levels of B2 kinin receptors . More interestingly , the combination of LLLT and AV was more effective in decreasing B2 levels when compared with AV itself ( Fig 4B ) .
The most effective treatment for venomous snakebites accidents is antivenom therapy . However , it is well known that such therapy is effective in neutralizing only the systemic effects of envenomation , without interfering with the severe local effects induced by these venoms [3] . Thus , given the importance framework triggered by local envenoming caused by bothropic venom and the incapacity of the antivenon to neutralize them , it is essential to investigate alternative therapies , with the greatest effectiveness in delaying progression and decreasing local symptoms of envenomed victims . Among clinical symptoms induced by bothrops snakebites , local pain is a common and clinically relevant manifestation to the patient [28 , 29] . Therefore , in this study , we investigated the capacity of photobiomodulation in reducing the nociceptive response caused by Bmv in mice footpad as well as the mechanisms involved . Herein , the intraplantar injection of Bmv induced mechanical allodynia and hyperalgesia . These results are in accordance with previous data demonstrating that Bmv induces potent mechanical allodynia and hyperalgesia in mice [24 , 30] . Photobiomodulation applied 30 min and 3 h after Bmv reversed both mechanical allodynia and hyperalgesia . From these data , we confirmed that photobiomodulation , in fact , is effective in reducing Bmv-induced local pain . In our study , as in previous studies [24 , 30] , we observed that antinociception was not related to AV treatment , since it was not able to interfere with mechanical sensitivity of mice . Also , the association of LLLT and AV did not modify the effect of LLLT alone , reinforcing the therapeutic potential of LLL in treating local effects induced by bothrops venoms . To better understand the capacity of photobiomodulation to decrease nociception , we evaluated the expression of Fos protein in the dorsal horn of the spinal cord of mice . The expression of proto-oncogenes from the c-fos , c-jun , and erg-1 family are extensively used as tools for the expression of enhanced activity of nociceptive neurons [20 , 21] . Our results demonstrate that the intraplantar administration of Bmv induced a significant increase of Fos expression , observed in the dorsal horn of the spinal cord , which is characteristic of nociceptor activation . According to the results of this study , photobiomodulation not only significantly inhibited Bmv-induced mechanical allodynia and hyperalgesia , but also decreased nociceptor activation at the spinal level . More interestingly , we showed here that photobiomodulation is able to interfere with the transmission of Bmv-induced pain message to the central nervous system , reducing nociceptor activation at the central level . This result reveals sensory neurons as an important cellular target for photobiomodulation in the context of pain . In addition to nociceptor-mediated effects , other mechanism ( s ) may also take part in the antinociception observed in our experimental model . We hypothesized that photobiomodulation may reduce the inflammatory cytokines in the paw and spinal cord . Therefore , the next experiment was designed to further validate the proposed hypothesis . It is commonly believed that proinflammatory cytokines such as TNF-α and IL-6 are involved in the pain process and that their peripheral and central levels are up-regulated in many pain models [31 , 32] . In addition , as described in previous studies , Bothropic venom induces the accumulation of pro-inflammatory IL-6 and TNF-α cytokines in the local of venom injection , which contributes to the enhancement of local tissue damage [1 , 33 , 34] . Moreover , some studies suggest that the analgesic effect of LLLT may be due to the anti-inflammatory activity by the inhibition of inflammatory mediators [13 , 35 , 36] . Hence , to further analyze the mechanism by which photobiomodulation reduces nociception of mice induced by Bmv , the expression of pro-inflammatory IL-6 and TNF-α cytokines was evaluated on samples obtained from either footpad or spinal cord of animals . Our results showed that photobiomodulation was able to reduce IL-6 and TNF-α gene expression in the footpad of animals . Also , we showed that associated treatment of AV and LLLT induced the same decrease on IL-6 and TNF-α mRNA levels as the observed with LLLT alone . Moreover , no changes on IL-6 and TNF-α mRNA levels were observed in samples from spinal cord of mice , thus suggesting that inhibition of hyperalgesia depends on a peripheral inhibition of inflammatory cytokines . This result corroborates the study of Ferreira et al . ( 2005 ) [13] that proposed that the analgesic effect of LLLT involves the inhibition of hyperalgesic mediators . Regarding IL-10 , we observed that Bmv injection decreased IL-10 mRNA levels on both footpad and spinal cord samples . Also , LLLT increased IL-10 mRNA levels in both footpad and spinal cord . AV treatment did not interfere with IL-10 levels on samples from footpad of mice . However it prevented the decrease of this cytokine on samples from spinal cord . From these data , we confirmed that AV prevents systemic effects induced by Bmv however it did not protect against local hyperalgesia . IL-10 is considered a regulatory cytokine , related to the control of the inflammatory process due to its capacity of inhibiting the proinflammatory cytokine secretion [37] . Results presented herein suggest that laser irradiation was able to modulate the expression of this regulatory cytokine , both in the local of venom injection and in the spinal cord , and it appears likely that this modulation plays a role in the anti-nociception observed after bothropic venom in response to photobiomodulation . To further analyze the mechanism by which photobiomodulation reduced Bmv-induced nociception , we evaluated the kinin receptors levels in the footpad of mice . Both kinin B1 and B2 receptors , evaluated here , play a central role in the pathophysiology of inflammation [38] . Kinin B2 receptors are broadly and constitutively expressed in most tissues , whereas B1 receptor is weakly expressed in most tissues under basal conditions but strongly upregulated following inflammation [39] . The involvement of bradykinin on Bmv-induced hyperalgesia and edema has been demonstrated [7 , 40] . In addition , it was already demonstrated that the kinin B2 receptors are involved in hyperalgesic response induce by B . jararaca and B . asper venoms [22 , 41] . Our results demonstrate that both B1 and B2 kinin receptors are increased in the footpad of animals injected with Bmv . Among the treatments , we found that both LLLT and AV were able to reduce the expression of B1 and B2 kinin mRNA levels . However , the association of LLLT and AV showed greater effectiveness in reducing B2 kinin receptors . Considering that kinin receptors are important mediators on Bothrops-induced hyperalgesia [22 , 23] it is feasible to suggest that photobiostimulation reverses Bmv-induced hyperagesia , at least in part , by modulating bradikinin receptors involved in the process . We conclude that photobiomodulation with low level laser is effective in decreasing nociceptor activation at the spinal level . Moreover LLL is effective in modulating pro- and anti-inflammatory cytokines as well as kinin receptors at mRNA transcriptional level . These effects , at least in part , contribute to the decrease of hyperalgesia observed after Bmv . Photobioestimulation with the parameters used herein should be considered as a potential therapeutic approach for the treatment of local effects of Bothrops snakebite .
|
Envenoming caused by Bothrops snakes is characterized by drastic local tissue damage involving hemorrhage , blistering , myonecrosis , prominent inflammatory response and intense pain . The most effective treatment for Bothrops snakebites is antivenom therapy , which is very efficient in reversing systemic effects of envenomation but not the severe local effects . Thus , there exists a need to find novel complementary therapies that may further assist in the prevention or even counteract the severe local effects of bothrops snakebite . Several studies have shown the effectiveness of photobiomodulation in reducing local effects induced by Bothropic venoms , however its mechanisms still remain unknown . In this study , we analyzed the effectiveness of photobiomodulation in reducing BmV-induced mechanical allodynia and hyperalgesia as well as part of the mechanisms involved in such effect . Results demonstrate that photobiomodulation reduces venom-induced mechanical allodynia and hyperalgesia and this effect depends on a decrease of nociceptor activation at the spinal cord level and by a modulation of pro- and anti- inflammatory cytokines as well as kinin receptors at mRNA transcriptional levels . These findings make phtobiomodulation a promising candidate to be associated to antivenom therapy for the treatment of the local response induced by Bothrops venoms .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"toxins",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"cytokines",
"engineering",
"and",
"technology",
"nervous",
"system",
"lasers",
"immunology",
"tropical",
"diseases",
"hyperalgesia",
"neuroscience",
"toxic",
"agents",
"toxicology",
"developmental",
"biology",
"signs",
"and",
"symptoms",
"molecular",
"development",
"neglected",
"tropical",
"diseases",
"optical",
"equipment",
"snakebite",
"venoms",
"spinal",
"cord",
"inflammation",
"immune",
"response",
"immune",
"system",
"pain",
"management",
"diagnostic",
"medicine",
"neuroanatomy",
"anatomy",
"equipment",
"physiology",
"biology",
"and",
"life",
"sciences",
"allodynia"
] |
2016
|
Analgesic Effect of Photobiomodulation on Bothrops Moojeni Venom-Induced Hyperalgesia: A Mechanism Dependent on Neuronal Inhibition, Cytokines and Kinin Receptors Modulation
|
Champagne coat color in horses is controlled by a single , autosomal-dominant gene ( CH ) . The phenotype produced by this gene is valued by many horse breeders , but can be difficult to distinguish from the effect produced by the Cream coat color dilution gene ( CR ) . Three sires and their families segregating for CH were tested by genome scanning with microsatellite markers . The CH gene was mapped within a 6 cM region on horse chromosome 14 ( LOD = 11 . 74 for θ = 0 . 00 ) . Four candidate genes were identified within the region , namely SPARC [Secreted protein , acidic , cysteine-rich ( osteonectin ) ] , SLC36A1 ( Solute Carrier 36 family A1 ) , SLC36A2 ( Solute Carrier 36 family A2 ) , and SLC36A3 ( Solute Carrier 36 family A3 ) . SLC36A3 was not expressed in skin tissue and therefore not considered further . The other three genes were sequenced in homozygotes for CH and homozygotes for the absence of the dilution allele ( ch ) . SLC36A1 had a nucleotide substitution in exon 2 for horses with the champagne phenotype , which resulted in a transition from a threonine amino acid to an arginine amino acid ( T63R ) . The association of the single nucleotide polymorphism ( SNP ) with the champagne dilution phenotype was complete , as determined by the presence of the nucleotide variant among all 85 horses with the champagne dilution phenotype and its absence among all 97 horses without the champagne phenotype . This is the first description of a phenotype associated with the SLC36A1 gene .
Many horse breeders value animals with variation in coat color . Several genes are known which diminish the intensity of the coloration and are phenotypically described as “dilutions” . Two of these are a result of the Cream ( CR ) locus and Silver ( Z ) locus . The molecular basis for Cream is the result of a single base change in exon 2 of the SLC45A2 ( Solute Carrier 45 family A2 , aka MATP for membrane associated transport protein ) on ECA21 [1] , [3] . This change results in the replacement of a polar acidic aspartate with a polar neutral asparagine in a putative transmembrane region of the protein coded for by this gene [3] , [2] . CR has an incompletely dominant mode of expression . Heterozygosity for CR dilutes only pheomelanin ( red pigment ) whereas homozygosity for CR results in extreme dilution of both pheomelanin and eumelanin ( black pigment ) [4] . The Silver dilution is the result of a missense mutation of PMEL17 ( Premelanosomal Protein ) on ECA6 . The base change causes replacement of a cytosolic polar neutral arginine with non-polar neutral cysteine in PMEL17 [2] . In contrast to CR , the Z locus is fully dominant and affects only eumelanin causing little to no visible change in the amount of pheomelanin regardless of zygosity . The change in eumelanin is most apparent in the mane and tail where the black base color is diluted to white and gray [5] . The coat color produced by the CH locus is similar to that of the CR locus in that both can cause dilution phenotypes affecting pheomelanin and eumelanin . However , the effect of CH differs from CR in that; 1 ) CH dilutes both pheomelanin and eumelanin in its heterozygous form and 2 ) heterozygotes and homozygotes for CH are phenotypically difficult to distinguish . The homozygote may differ by having less mottling or a slightly lighter hair color than the heterozygote . Figure 1 displays images of horses with the three base coat colors chestnut , bay and black and the effect of CH upon each . Figure 2 shows that champagne foals are born with blue eyes , which change color to amber , green , or light brown and pink “pumpkin skin which acquires a darker mottled complexion around the eyes , muzzle , and genitalia as the animal matures [6] . Foals with one copy of CR also have pink skin at birth but their skin is slightly darker and becomes black/near black with age . The champagne phenotype is found among horses of several breeds , including Tennessee Walking Horses and Quarter Horses . Here we describe family studies that led to mapping the gene and subsequent investigations leading to the identification of a genetic variant that appears to be responsible for the champagne dilution phenotype .
Table 1 summarizes the evidence for linkage of the CH gene to a region of ECA14 . The linkage phase for each family was apparent based on the number of informative offspring in each family . Recombination rates ( θ ) were based on the combined recombination rate from all families . Four microsatellites showed significant linkage to the CH locus: VHL209 ( LOD = 6 . 03 for θ = 0 . 14 ) , TKY329 ( LOD = 3 . 64 for θ = 0 . 10 ) , UM010 ( LOD = 5 . 41 for θ = 0 . 04 ) and COOK007 ( LOD = 11 . 74 for θ = 0 . 00 ) . Figure 3 identifies the haplotypes for offspring of a single sire showing recombination between the genetic markers and the CH locus . Pedigrees of the three sire families and haplotype information are provided in Figure S1 and Table S1 respectively . The CH locus maps to an interval between UM010 and TKY329 with microsatellite . No recombinants were detected among 39 informative offspring between the CH and COOK007 locus . Candidate genes were selected on the basis of proximity to the marker COOK007 and as genes previously characterized in other species as influential in the production or migration of pigment cells . SPARC was located closest at ∼90 kb downstream from COOK007 and is coded for on the plus strand of DNA . It has been implicated in migration of retinal pigment epithelial cells in mice [7] . SLC36A family members are solute carriers and other solute carrier families have been found to play a role in coat color . SLC36A1 is located ∼250 kb downstream from COOK007 . It is the first and most proximal to COOK007 of three genes in this family and is coded for on the minus strand of DNA . SLC36A2 and SLC36A3 are coded for on the plus strand of DNA and are approximately 350 k and 380 k downstream from COOK007 respectively . A2 and A3 have been found to be expressed in a limited range of tissues in humans and mice [8] . RT-PCR ( reverse transcription-polymerase chain reaction ) was used to determine if SLC36A1 , SLC36A2 or SLC36A3 were expressed in skin . SLC36A1 and SLC36A2 were expressed in skin and their genomic exons were sequenced . SLC36A3 was not detected in skin and therefore not investigated for detection of SNPs . Results for RT-PCR of these three genes are shown in Figure 4 . All 9 exons of SPARC were sequenced . Three SNPs were found in exons but none showed associations with the champagne phenotype and are shown in Table S2 . SLC36A2 was sequenced with discovery of 9 SNPs in exons . None of the SNPs showed associations with CH . These SNPs and all other variations detected are described in Table S2 . SLC36A1 was sequenced . Only one SNP was found , a missense mutation involving a single nucleotide change from a C to a G at base 76 of exon 2 ( c . 188C>G ) ( Figure 5 ) . These SLC36A1 alleles were designated c . 188[C/G] , where c . 188 designates the base pair location of the SNP from the first base of SLC36A1 cDNA , exon 1 . Sequencing traces for the partial coding sequence of SLC36A1 exon 2 with part of the flanking intronic regions for one non-champagne horse and one champagne horse were deposited in GenBank with the following accession numbers respectively: EU432176 and EU432177 . This single base change at c . 188 was predicted to cause a transition from a threonine to arginine at amino acid 63 of the protein ( T63R ) . Figure 6 shows the alignment of the protein sequence for exons 1 and 2 of SLC36A1 for seven mammalian species with sequence information from Genbank ( horse , cattle , chimpanzee , human , dog , rat and mouse ) . Alignment was performed using AllignX function of Vector NTI Advance 10 ( Invitrogen Corp , Carlsbad , California ) . The alignment demonstrates that this region is highly conserved among all species . At position 63 , the amino acid sequence is completely conserved among these species , with the exception of horses possessing the champagne phenotype . This replacement of threonine with arginine occurs in a putative transmembrane domain of the protein [9] . The distribution of c . 188G allele among different horse breeds and among horses with and without the champagne phenotype was investigated . Table 2 is a compilation of the population data collected via the genotyping assay . All dilute horses ( 85 ) which did not have the CR gene , tested positive for the c . 188G allele with genotypes c . 188C/G or c . 188G/G . No horses in the non-dilute control group ( 97 ) possessed the c . 188G allele . The horses used for the population study were selected for coat color and not by random selection; therefore measures of Hardy-Weinberg equilibrium are not applicable and were not calculated .
Family studies clearly showed linkage of the gene for the champagne dilution phenotype within a 6 cM region on ECA14 [10] ( Table 1 ) . Based on the Equine Genome Assembly V2 as viewed in ENSEMBL genome browser ( http://www . ensembl . org/Equus_caballus/index . html ) this region spans approximately 2 . 86 Mbp [11] . Within that region , four candidate genes were investigated; one based on known effects on melanocytes ( eg . SPARC ) and three for their similarity to other genes previously shown to influence pigmentation ( eg , SLC36A1 , A2 , and A3 ) . While SNPs were found within the exons of SPARC , none were associated with CH . Of the other 3 candidate genes , only SLC36A1 and SLC36A2 were found to be expressed in skin cells . Therefore , the exons of those two genes were sequenced . A missense mutation in the second exon of SLC36A1 showed complete association with the champagne phenotype across several breeds . While SNPs were found for SLC36A2 , none showed associations at the population level for the champagne dilution phenotype . This observation is the first demonstration for a role of SLC36A1 in pigmentation . Orthologous genes in other species are known to affect pigmentation . For example , the gene responsible for the cream dilution phenotypes in horses , SLC45A2 ( MATP ) , belongs to a similar solute carrier family . In humans , variants in SLC45A2 have been associated with skin color variation [12] and a similar missense mutation ( p . Ala111Thr ) in SLC24A5 ( a member of potassium-dependent sodium-calcium exchanger family ) is implicated in dilute skin colors caused from decreased melanin content among people of European ancestry [13] . The same gene , SLC24A5 is responsible for the Golden ( gol ) dilution as mentioned in the review of mouse pigment research by Hoekstra ( 2006 ) [14] It is proposed , here , that the missense mutation in exon 2 of SLC36A1 is the molecular basis for champagne dilution phenotype . While this study provides evidence that this is the mutation responsible for the champagne phenotype , the proof is of a statistical nature and a non-coding causative mutation can not be ruled out at this point . SLC36A1 , previously referred to by the name PAT1 ( proton/amino acid transporter 1 ) in human and mouse [15] , is a proton coupled small amino acid transporter located and most active in the brush border membranes of intestinal epithelial cells . This protein has also been characterized in rats under the name LYAAT1 ( lysosomal amino acid transporter 1 ) . LYAAT1 is localized in the membrane of lysozomes in association with LAMP1 ( lysosomal associated protein 1 ) and in the cell membrane of post-synaptic junctions . In lysozomes it allows outward transport of protons and amino acids from the lysozome to the cytosol [16] . During purification and separation of early-stage melanosomes LAMP1 is found in high concentrations in the fraction containing stage II melanosomes [17] , . Perhaps SLC36A1 plays a role in transitions from lysozome-like precursor to melanosome . Since organellular pH affects tyrosine processing and sorting [18] , an amino acid substitution in this protein may affect pH of the early stage melanosome and the ability to process tyrosine properly . There must be an increase in pH , before the tyrosinase can be activated . The cytosolic pH gradient must also be maintained for proper sorting and delivery of the other proteins required for melanosome development [19] . Thus , the pH gradient of the cell may be altered by this mutation . This variant , discovered in association with a coat dilution in the horse , is the first reported for the SLC36A1 gene . The phenotype resulting from this mutation , a reduction of pigmentation in the eyes , skin and hair , illustrates previously unknown functions of the protein product of SLC36A1 . Furthermore , now that a molecular test for champagne dilution is established , the genotyping assay can be used in concert with available tests for cream dilution and silver dilution to clarify the genetic basis of a horse's dilution phenotype . This will give breeders a new tool to use in developing their breeding programs whether they desire to breed for these dilutions or to select against them .
Three half-sibling families , designated 1 , 2 and 3 , were used for mapping studies . Family 1 consisted of a Tennessee Walking Horse ( TWH ) stallion , known heterozygous at the Champagne locus ( CH/ch ) , and his 17 offspring out of non-dilute mares ( ch/ch ) . Family 2 consisted of an American Paint Horse stallion ( CH/ch ) and his 10 offspring out of non-dilute ( ch/ch ) mares . Family 3 consisted of a TWH stallion ( CH/ch ) , 23 offspring and their 12 non-dilute dams ( ch/ch ) and 1 dilute ( buckskin ) dam ( ch/ch , CR/cr ) . To investigate the distribution of the gene among dilute and non-dilute horses of different horse breeds , 97 non-champagne horses were chosen from stocks previously collected and archived at the MH Gluck Equine Research Center . These horses were from the following breeds: TWH ( 20 ) , Thoroughbreds ( TB , 35 ) , American Paint Horses ( APHA , 32 ) , Pintos ( 5 ) , American Saddlebreds ( ASB , 2 ) , one American Quarter Horse ( AQHA ) , one pony , and one American Miniature ( AMH ) Horse . Hair and blood samples from horses with the champagne dilution phenotype were submitted by owners along with pedigree information and photographs showing the champagne color and characteristics of each horse . Samples were collected from the following breeds ( 85 total ) : American Miniature Horse ( 9 ) , American Cream Draft cross ( 1 ) , American Quarter Horse ( 27 ) , American Paint Horse ( 13 , in addition to the family ) , American Saddlebred ( 2 ) , Appaloosa ( 1 ) , ASB/Friesian cross ( 1 ) , Arabian crossed with APHA or AQHA horses ( 3 ) , Missouri Foxtrotter ( 4 ) , Mule ( 2 ) , Pony ( 1 ) , Spanish Mustang 1 ) , Spotted Saddle Horse ( 1 ) , Tennessee Walking Horse ( 20 , in addition to the families ) . To be characterized as possessing the champagne phenotype , horses exhibited a diminished intensity of color ( dilution ) in black or brown hair pigment and met at least two of the three following criteria: 1 ) mottled skin around eyes , muzzle and/or genitalia , 2 ) amber , green , or light brown eyes , or 3 ) blue eyes and pink skin at birth [6] . This was accomplished by viewing photo evidence of these traits or by personal inspection . Due to potential confusion between phenotypes of cream dilution and champagne dilution , all DNA samples from horses with the dilute phenotype were tested for the CR allele and data from those testing positive were not included in the population data . DNA from blood samples was extracted using Puregene whole blood extraction kit ( Gentra Systems Inc . , Minneapolis , MN ) according to its published protocol . Hair samples submitted by owners were processed using 5–7 hair bulbs according to the method described by Locke et al . ( 2002 ) . The hair bulbs were placed in 100 µl lysis solution of 1× FastStart Taq Polymerase PCR buffer ( Roche , Mannheim , Germany ) , 2 . 5 mM MgCl2 ( Roche ) , 0 . 5% Tween 20 ( JT Baker , Phillipsburg , NJ ) and 0 . 01 mg proteinase K ( Sigma-Aldrich , St Louis , MO ) and incubated at 60°C for 45 minutes , followed by 95°C for 45 min to deactivate the proteinase K . The genome scan was done in polymerase chain reaction ( PCR ) multiplexes of 3 to 6 microsatellites per reaction . The 102 microsatellite markers used are listed in Table S3 . Primers for these microsatellites were made available in connection with the USDA-NRSP8 project [20] . Two additional microsatellites were used; TKY329 [21] was selected based on its map location between two microsatellites used for genome scanning ( UM010 and VHL209 ) and COOK007 was developed in connection with this study based on DNA sequence information from the horse genome sequence viewed in the UCSC genome browser [8] in order to investigate linkage within the identified interval . Primers for COOK007 were designed using Primer 3 software accessed online ( Forward , 5′- 6FAM-CATTCCAAACACCAACAACC - 3′ ) , ( Reverse , 5′ – GGACATTCCAGCAATACAGAG – 3′ ) [22] . The initial scan was conducted on a subset of samples from Family 3; including sire 3 , five non-champagne offspring and five champagne offspring . When the microsatellite allele contribution from the sire was not informative , ( e . g . the sire and offspring had the same genotype ) , dams from family 3 were typed to determine the precise contribution from the sire . When the inheritance of microsatellite markers in family 3 appeared to be correlated with the inheritance of the CH allele , then the complete families A , B and C were typed and the data analyzed for linkage by LOD score analysis [23] . Amplification for fragment analysis was done in 10 µl PCR reactions using 1× PCR buffer with 2 . 0 mM MgCl2 , 200 µM of each dNTP , 1 µl genomic DNA from hair lysate , 0 . 1 U FastStart Taq DNA polymerase ( Perkin Elmer , Waltham , MA ) and the individual required molarity of each primer from the fluorescently labeled microsatellite parentage panel primer stocks at the MH Gluck Equine Research Center . Samples were run on a PTC-200 thermocycler ( MJ research , Inc . , Boston , MA ) at a previously determined optimum annealing temperature for each multiplex . Capillary electrophoresis of product was run on an ABI 310 genetic analyzer ( Applied Biosystems Inc . ABI , Foster City , CA ) . Results were then analyzed using the current version of STRand microsatellite analysis software ( http://www . vgl . ucdavis . edu/informatics/STRand/ ) . PCR template for sequencing was amplified in 20 µl PCR reactions using 1× PCR buffer with 2 . 0 mM MgCl2 , 200 µM of each dNTP , 1 µl genomic DNA from hair lysate , 0 . 2 U FastStart Taq DNA polymerase ( Perkin Elmer ) and 50 nM of each primer . Exon 2 of SLC36A1 was sequenced with the following primers: Forward ( 5′-CAG AGC CTA AGC CCA GTG TC-3′ ) and Reverse ( 5′-GGA GGA CTG TGT GGA AAT GG-3′ ) at an annealing temperature of 57°C . Primers used to sequence the other SLC36A1 exons and primers for sequencing genomic exons of SLC36A2 are provided in parts 1 and 2 respectively of Table S4 . Template product was quantified on a 1% agarose gel , then amplified with BigDye Terminator v1 . 1 cycle sequencing kit according to manufacturer's instructions ( Applied Biosystems ) , cleaned using Centri-Sep columns ( Princeton Separations Inc . , Adelphia , NJ ) , and run on and ABI 310 genetic analyzer ( Applied Biosystems ) . Six samples were initially sequenced: 2 suspected homozygous champagnes ( based on production of all champagne dilution offspring when bred to at least 10 non-dilute dams ) , 2 heterozygotes , and 2 non-dilute horses . The results were analyzed and compared by alignment using ContigExpress from the Vector NTI Advance 10 . 3 software package ( Invitrogen Corporation , Carlsbad , California ) . RT-PCR was performed in 25 µl reactions a Titan One Tube RT-PCR Kit ( Roche ) according to enclosed protocol with the primers listed in part 3 of Table S4 . RNA from different tissues of non-dilute horses was used to acquire partial cDNAs containing the first two exons for SLC36A1 , first three exons SLC36A2 and first 4 exons of SLC36A3 . The cDNA acquired was sequenced and the resulting sequences were verified for their respective genes with a BLAT search using the equine assembly v2 in ENSEMBL ( http://www . ensembl . org/Equus_caballus/index . html ) genome browser . RT-PCR was also performed utilizing RNA extracted from skin , kidney and testes of non-dilute animals currently in lab stocks . SLC36A1 cDNA was produced from the skin and blood using 50 ng RNA per reaction . SLC36A2 cDNA was produced from testes using 1 mRNA per RT-PCR reaction then following up with a nested PCR for shorter product . SLC36A2 cDNA was produced from skin using 50 ng mRNA per RT-PCR reaction . Nested PCR was not necessary . SLC36A3 cDNA was produced from testes using 1 ug mRNA per reaction . 9 µl of initial reaction was visualized on a 2% agarose gel to check for visible bands of product . When product was not initially detected an additional 20 µl PCR was performed in reactions as outlined above using 5 µl of RT product in the place of hair lysate per reaction . Detected product was then sequenced with the protocol listed above . Sequences were then used in a BLAST search using equine genome assembly 2 on ENSEMBL genome browser to verify the correct cDNA was amplified . A Custom TaqMan SNP Genotyping Assay ( Applied Biosystems ) was designed for c . 188C/G SNP in filebuilder 3 . 1 software ( Applied Biosystems ) to test the population distribution of the SLC36A1 alleles . A similar assay was also designed to test for the cream SNP . These assays were run on a 7500HT Fast Real Time-PCR System ( Applied Biosystems ) . All dilute horses tested for SLC36A1 variants were concurrently tested for SLC45A variants . Horses testing positive for CR alleles were not used in the dataset to avoid any confusion over the origin of their dilution phenotype .
|
The purpose of this study was to uncover the molecular basis for the champagne hair color dilution phenotype in horses . Here , we report a DNA base substitution in the second exon of the horse gene SLC36A1 that changes an amino acid in the transmembrane domain of the protein from threonine to arginine . The phenotypic effect of this base change is a diminution of hair and skin color intensity for both red and black pigment in horses , and the resulting dilution has become known as champagne . This is the first genetic variant reported for SLC36A1 and the first evidence for its effect on eye , skin , and hair pigmentation . So far , no other phenotypic effects have been attributed to this gene . This discovery of the base substitution provides a molecular test for horse breeders to test their animals for the Champagne gene ( CH ) .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics/gene",
"discovery",
"genetics",
"and",
"genomics/comparative",
"genomics",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics"
] |
2008
|
Missense Mutation in Exon 2 of SLC36A1 Responsible for Champagne Dilution in Horses
|
H1 linker histones facilitate higher-order chromatin folding and are essential for mammalian development . To achieve high-resolution mapping of H1 variants H1d and H1c in embryonic stem cells ( ESCs ) , we have established a knock-in system and shown that the N-terminally tagged H1 proteins are functionally interchangeable to their endogenous counterparts in vivo . H1d and H1c are depleted from GC- and gene-rich regions and active promoters , inversely correlated with H3K4me3 , but positively correlated with H3K9me3 and associated with characteristic sequence features . Surprisingly , both H1d and H1c are significantly enriched at major satellites , which display increased nucleosome spacing compared with bulk chromatin . While also depleted at active promoters and enriched at major satellites , overexpressed H10 displays differential binding patterns in specific repetitive sequences compared with H1d and H1c . Depletion of H1c , H1d , and H1e causes pericentric chromocenter clustering and de-repression of major satellites . These results integrate the localization of an understudied type of chromatin proteins , namely the H1 variants , into the epigenome map of mouse ESCs , and we identify significant changes at pericentric heterochromatin upon depletion of this epigenetic mark .
In all eukaryotes , nuclear DNA is packaged into chromatin by its association with histones [1] . The nucleosome , the building block of chromatin , consists of an octamer of four core histones ( H2A , H2B , H3 and H4 ) wrapped by 147 bp of DNA [2] . Linker histone H1 binds to DNA entering and exiting nucleosome core particles as well as the linker DNA between nucleosomes , facilitating folding of chromatin into higher order structure [1] , [3]–[5] . The H1 histone family is the most divergent group among the highly conserved histone proteins . To date , 11 different H1 variants have been characterized in mammals , including somatic H1 variants ( H1a to H1e ) , the replacement H1 ( H10 ) , germ cell specific H1s ( H1t , H1T2 , HILS1 and H1oo ) , as well as the recently characterized variant H1x [6] . Deletion of three major somatic H1 variants ( H1c , H1d and H1e ) together leads to a 50% reduction of the total H1 level and embryonic lethality at midgestation , demonstrating that H1 level is critical for mammalian development [7] . H1 variants are conserved from mouse to human , and differ in their biochemical properties and expression patterns during development and malignant transformation [8]–[11] . Although none of the H1 variants tested is essential for mouse development [12]–[15] , they have been shown to regulate specific gene expression in various cell types [6] , [16]–[18] . However , the mechanisms by which H1 variants modulate chromatin structure and gene expression remain under-explored . Mapping of the precise genomic localizations of different H1 variants in vivo is likely to provide significant insights , but has been challenging due to the lack of high quality antibodies that could accurately distinguish different H1 variants . Pluripotent embryonic stem cells ( ESCs ) can differentiate into cells of all three germ layers , offering great potential in regenerative medicine . The epigenome is suggested to play a critical role in stem cell fate determination , and genome-wide mapping studies have revealed that ESCs have characteristic epigenetic landscapes that differ from differentiated cells [19] , [20] . However , despite significant efforts to characterize the chromatin features of human and mouse ESCs , both by individual labs [19] , [21]–[23] and by large consortia ( ENCODE [24] , Roadmap Epigenomics [25] ) , the landscapes of linker histone H1 variants have not been described on a genome-wide scale . In this study , we have achieved high resolution mapping of H1d , H1c and H10 in ESCs by chromatin immunoprecipitation followed by massive parallel sequencing ( ChIP-seq ) . H1d and H1c are among the most abundant linker histones in mouse ESCs , accounting respectively for 32 . 6% and 16 . 4% of total H1 , whereas the differentiation associated H1 , H10 , accounts for 2% of H1 in undifferentiated ESCs [26] , [27] . These three variants differ significantly in terms of their residence time on chromatin and their ability to promote chromatin condensation in vitro [28] , [29] . They also display different expression patterns during mammalian development and in exponentially growing cells vs . quiescent cells [8] , [10] , [30] . Here , we have generated FLAG-tagged H1d knock-in ESCs , Myc-tagged H1c knock-in ESCs , as well as FLAG-tagged H10 overexpressing ESCs , designated as respective H1dFLAG , H1cMyc , and fH10 cells . We demonstrate that tagged H1 variants maintain the biochemical properties of the endogenous H1s in vivo and that FLAG-H1d can substitute for H1d during mouse development . High resolution mapping reveals that H1d and H1c occupancies are highly correlated , both enriched at AT-rich regions , but also possess different binding specificity . Both H1d and H1c largely co-localize with H3K9me3 , but show an inverse correlation with GC% or H3K4me3 . Importantly , we discover that H1d and H1c are highly enriched at major satellite elements , which display a longer nucleosome repeat length than bulk chromatin in ESCs . Finally , we show that H1 depletion leads to chromocenter clustering and increased expression of major satellites independent of multiple epigenetic marks at these regions .
Efforts to generate high resolution genome-wide maps of H1 variants were hampered by the lack of H1 variant specific antibodies of sufficient quality for ChIP-seq . Here , we established knock-in mouse ESC lines in which H1d or H1c variant was N-terminally tagged with an epitope ( FLAG or Myc ) for which highly specific antibodies exist . An H1dFLAG cell line was created by inserting the FLAG tag coding sequence at the endogenous H1d locus through homologous recombination ( Figure 1A ) . H1c/H1e double knockout mice develop normally , yet H1c/H1d/H1e triple knockout ( H1 TKO ) mice are embryonic lethal [7] . Thus , ESCs with H1dFLAG allele in H1c+/−H1e+/− background could be used to produce H1c−/−H1dFLAG/FLAGH1e−/− mice to determine whether FLAG-tagged H1d ( FLAG-H1d ) functions equivalently to endogenous H1d by assessing if the tagged H1d can rescue the embryonic lethality of H1 TKO mutants . Toward this end , we generated both H1c+/−H1d+/FLAGH1e+/− ( “H1dFLAG” ) and H1c+/−H1dFLAG/−H1e+/− ( “H1d-trans” ) ESC lines by transfection of the FLAG-H1d targeting vector ( Figure 1A ) into the cis triply targeted H1c+/−H1d+/−H1e+/− ESCs established previously [7] . ESC clones with either cis or trans configuration of the H1dFLAG allele with the H1c and H1e KO allele were identified and verified by Southern blotting ( Figure 1B ) . As expected , FLAG-H1d was located in the nuclei of the H1dFLAG cells ( data not shown ) . Analysis of histone extracts of chromatin prepared from cis-targeted H1dFLAG cells by HPLC and immunoblotting indicated that FLAG-H1d was associated with chromatin and eluted in the same fraction as the endogenous H1d , suggesting that FLAG-H1d has the same hydrophobicity as the endogenous H1d ( Figure 1C and 1D ) . The ratio of somatic H1 variants , H1 a–e , to nucleosome ( H1/nuc ) of H1dFLAG cells was nearly identical to that of H1c+/−H1d+/+H1e+/− ( cehet ) cells , indicating a similar expression level of FLAG-H1d as the endogenous H1d ( Figure 1E ) . As expected , the protein level of differentiation associated H10 variant was minimal in undifferentiated ESCs . We injected cis-targeted H1dFLAG cells into mouse blastocysts and produced chimeric mice which gave germline transmission of the H1dFLAG allele . H1c+/−H1d+/FLAGH1e+/− mice were intercrossed to generate H1c−/−H1dFLAG/FLAGH1e−/− homozygous mice ( designated as H1dFLAG/FLAG mice ) ( Figure S1Ai ) . These homozygotes were viable , fertile and developed normally as H1c/H1e double null ( ceKO ) mice , demonstrating that FLAG-H1d can substitute for the endogenous H1d to fully rescue the lethal phenotype of H1 TKO mutants . HPLC , mass spectrometry and immunoblotting demonstrated that H1dFLAG/FLAG mice had full replacement of H1d by FLAG-H1d ( Figure S1Aii and S1Aiii ) and that the H1/nuc ratio of spleen chromatin from H1dFLAG/FLAG mice was 0 . 7 , comparable to that of ceKO mice ( Figure S1Aiv ) . Taken together , these results demonstrate that FLAG-H1d maintains the expression level and properties of the endogenous H1d in vivo . Using a similar knock-in strategy , we generated H1c+/MycH1d+/−H1e+/− ESCs ( H1cMyc ) by transfection of the H1cMyc targeting construct into the cis triply targeted H1c+/−H1d+/−H1e+/− ESCs and selected ESC clones that underwent homologous recombination at H1c locus ( Figure S1Bi and S1Bii ) . Similar to FLAG-H1d , the N-terminally Myc tagged H1c ( Myc-H1c ) colocalized with Hoechst stained nuclear regions in H1cMyc cells ( data not shown ) , and Myc-H1c was eluted in the same fraction as the endogenous H1c protein from HPLC analysis ( Figure S1Biii and S1Biv ) . H1cMyc cells had a H1/nuc ratio of 0 . 38 , comparable to the ratio of 0 . 36 in cehet cells ( Figure 1E , Figure S1Biii ) , indicating that like FLAG-H1d , Myc-H1c has the same expression level and biochemical properties as the endogenous H1c . To achieve high resolution mapping of H1d and H1c variants in mouse ESC genome , we performed ChIP-seq in cis-targeted H1dFLAG and H1cMyc ESCs using anti-FLAG and anti-Myc antibodies , respectively . In each ChIP-seq library , approximately 80–90% of reads were mappable to the mouse genome ( mm9 ) using the Bowtie aligner [31] ( Table S1 ) . While sonicated chromatin input control libraries on average had 65% vs . 22% of reads mapped to unique positions and multiple positions respectively , the H1c ChIP-seq libraries had 44% vs . 45% mapped to unique vs . multiple positions , suggesting that a higher proportion of H1c resides on repetitive sequences . Similarly , an overrepresentation of multi-match sequence reads ( 39% of mapped reads ) occurred in H1d ChIP-seq libraries . A survey of sequencing signal intensities indicated that H1d and H1c were generally depleted from gene rich regions with the deepest dips around transcription start sites of active genes ( examples shown in Figure 2A and Figure S2A ) . ChIP-seq with the anti-FLAG antibody in control ESCs not containing FLAG-H1d generated minimal random background signals ( data not shown ) , and examination of H1c ( anti-Myc ) signals showed no enrichment at c-Myc target genes , such as Oct4 , Nanog and Sox2 [32] ( Figure S2A ) , indicating no cross-reactivity for these antibodies . To compare H1 occupancy with other histone marks , we performed ChIP-seq of an active histone mark , H3K4me3 , and two repressive histone marks , H3K9me3 and H3K27me3 , in murine ESCs . Visual examination of the track files revealed that H1 dips often coincided with H3K9me3 dips or H3K4me3 peaks and that H1 displayed higher signals at gene poor regions with high AT% ( low GC% ) ( Figure 2A ) . H3K27me3 , enriched at Hox gene clusters ( Figure S2B ) as expected , did not show obvious pattern correlation with H1 ( Figure 2A and Figure S2A ) . These observations suggest possible correlations of H1d and H1c with H3K9me3 , H3K4me3 , gene distribution and GC content in vivo . We next investigated the relationship between H1 occupancy and gene expression levels at a 10 kb region centered around transcription start sites ( TSSs ) as well as a 10 kb region centered around transcription termination sites ( TTSs ) using GenPlay software [33] . Such metagene analysis revealed that H1 signals were always lower than chromatin input control within these regions ( IP-IN<0 ) ( Figure 2B ) , suggesting a general depletion of H1 at gene containing regions . Both H1d and H1c were especially depleted around the TSSs with dips much deeper at highly active genes than at silent genes ( Figure 2B ) . Interestingly , except at TSSs and promoters , H1 signals remained largely constant throughout the gene encompassing regions and the signal intensity was higher at the silent genes than that at active genes , suggesting that H1 is underrepresented at surrounding regions of active genes as well ( Figure 2B ) . Indeed , for genes highly depleted of H1 variants at promoters , the signal values of H1s , although gradually increased toward distal regions , remained diminished up to 200 kb from TSS ( Figure S3 ) , suggesting that H1s are depleted from broad domains at these regions in the genome . H3K4me3 is known to be peaked around TSS of active genes [34] , [35] , and metagene H3K4me3 curves displayed an opposite pattern to that of H1 ( Figure 2B ) , further indicating that H1 is absent at active promoters . H3K9me3 exhibited a very similar distribution pattern to that of H1d and H1c , whereas H3K27me3 did not show similar profiles to that of H1 variants ( Figure 2B ) . Metagene analysis of H1 and histone marks on genes finely partitioned by expression levels ( each group with 20% of genes ) over a 10 kb region ( −5 kb to +5 kb of TSS ) further corroborated their distinctive patterns at TSSs as a function of gene expression ( Figure 2C ) . To better define the correlation of H1 occupancy with histone marks around TSSs and promoters , metagene analysis of H1 signals was performed for genes partitioned into 5 groups according to their levels of H3K9me3 , H3K4me3 , or H3K27me3 , which displayed characteristic profiles around TSS ( Figure 2D , 2E , 2F and Figure S2C ) . H1 signals displayed positive and negative correlations with respective H3K9me3 and H3K4me3 signals , having the deepest dip for promoters and TSSs with the lowest H3K9me3 levels ( Figure 2D ) or highest H3K4me3 signals ( Figure 2E ) . On the other hand , H1 signals showed no correlation with H3K27me3 levels and no difference among the 5 groups of genes partitioned according to H3K27me3 levels ( Figure 2F ) . Interestingly , H1 was also depleted at the promoters of genes bound by H3K4me3 and H3K27me3 bivalent marks [21] but not at H3K4me3-free promoters , regardless of the presence or absence of H3K27me3 ( Figure 2G ) . Although most H1d and H1c signals appeared universally distributed , we identified regions enriched for H1 binding using SICER [36] and GenPlay software . Identified H1d and H1c enriched regions often formed broad domains ( examples shown in Figure S4A ) . Annotation of H1d- and H1c- rich regions using CEAS [37] , a software designed to characterize both sharp and broad ChIP-seq enrichment , indicated that , similar to H3K9me3 , both H1d and H1c “peaks” were over-represented in distal intergenic regions and under-represented at promoters and 5′UTR , which were highly enriched with H3K4me3 peaks as reported previously ( Figure S4B and [34] ) . We next performed genome-wide correlation analysis to determine if the similarity and/or contrast of H1 variants with GC% and histone marks at TSSs also extend to a genome-wide scale . Indeed , the distribution of H1d and H1c were highly correlated throughout the genome ( R = 0 . 7866 ) ( Figure 3A ) , and both variants were negatively correlated with GC% ( R = −0 . 4182 and −0 . 4140 for respective H1d and H1c ) , indicating that H1d and H1c were enriched or depleted at similar regions . Both H1d and H1c were correlated negatively with H3K4me3 ( R = −0 . 2640 and −0 . 3317 respectively ) , but positively with H3K9me3 ( R = 0 . 5732 , 0 . 5790 ) ( Figure 3B ) , suggesting their enrichment at heterochromatin . On the other hand , these two variants showed no obvious correlation with H3K27me3 ( R = −0 . 08 for both variants ) ( Figure 3B ) . Correlation analysis of sequencing signals on enriched or depleted regions gave similar coefficients as the respective genome-wide coefficients ( data not shown ) . It is interesting to note that the coefficients of H1 vs . H3K4me3 on sex chromosomes were dramatically different from those of autosomes ( Figure S5A ) . This result echoes the previous finding that sex-chromosome genes are overrepresented among genes with altered expression levels by triple H1 deletion in ESCs [26] , suggesting that H1 may play a role in regulating higher order chromatin structures of sex chromosomes . To gain a comprehensive view of the DNA features of H1d- and H1c- rich regions , we selected the regions highly enriched for H1 variants and histone marks , and performed cross-comparison of genome attributes using the statistical analysis software EpiGRAPH [38] . Such analysis ( Figure 3C and Figure S5B ) revealed that: a ) H1d/H1c common peaks ( regions highly enriched for both H1d and H1c ) appeared similar to H3K9me3 peaks in genome attributes , except for satellite DNA which was relatively overrepresented in H1 peak regions; b ) H1d/H1c common peaks were enriched at AT-rich sequences , satellite DNA , and chromosome G-bands but were absent from GC-rich regions , and genes or exons when compared with H3K4me3 or H3K27me3 peaks; c ) comparison of H1d/H1c common peaks with H1d/H1c unique peaks ( regions highly enriched for H1d or H1c but not both ) showed similar features as the comparison of H1d/H1c common peaks with H3K4me3 or H3K27me3 peaks; d ) comparison of H1d vs . H1c specific peaks indicated that H1d unique peaks were relatively enriched at GC-rich sequences and LINEs , whereas H1c unique peaks were more enriched at AT-rich sequences , Giemsa positive regions and satellite DNA; e ) the overrepresentation analyses between H1d ( or H1c ) unique peak regions and histone mark peak regions exhibited similar features as comparisons using H1 common peaks . These results define common and unique features for H1d and H1c enriched regions . The EpiGRAPH overrepresentation analysis indicated that peak regions of H1d and H1c were enriched for satellite repeats . Indeed , examination of the top ranked H1 peak regions with especially high binding signals revealed that these regions overlap perfectly with major satellite sequences ( examples shown in Figure 4A ) . This finding and the above observation of overrepresentation of multi-match sequence reads in H1 ChIP-seq libraries prompted us to perform a thorough mapping study of sequence reads to a database of repetitive sequences . We aligned sequence reads of H1d , H1c , H3K9me3 , H3K27me3 and H3K4me3 ChIP-seq libraries to Repbase Update , a comprehensive database of repetitive elements from diverse eukaryotic organisms [39]–[41] . We found that both H1d and H1c were significantly enriched at repetitive sequences , with H1d and H1c ChIP-seq libraries having on average percent mapped repeats respective 2 . 3- , and 2 . 8-fold of that of chromatin input-seq libraries ( Figure 4B ) . H3K9me3 , H3K27me3 and H3K4me3 ChIP-seq libraries had an average respective percent mapped repeats 1 . 4- , 0 . 7- , and 0 . 9- fold compared with input controls ( Figure 4B ) , suggesting an overrepresentation of H3K9me3 , yet not as dramatic as H1d and H1c , at repetitive sequences . Importantly , we found that the increased proportion of total reads of H1 libraries mapped to repetitive sequences was predominantly caused by overrepresentation on the major satellite sequences on which the levels of H1d and H1c occupancy were enriched on average 4 . 0- and 5 . 6-fold compared with the chromatin input control ( Figure 4B ) . This level of H1 enrichment appeared to be specific to major satellites because we did not observe H1d and H1c enrichment among other abundant repeats , except for a moderate increase of H1d and H1c occupancy at minor satellites . qChIP-PCR results confirmed the preferential binding of these two H1 variants to major satellites ( Figure S6 ) . Sequencing results showed that H1d and H1c levels on most of other less abundant classes of repetitive elements , such as L1 , IAP LTR retrotransposons , SINE , non-LTR retrotransposons , and DNA transposons , were similar or lower compared with the input control ( Figure 4B and Figure S7 ) . H3K4me3 was highly enriched at 5′end of a subset of LINE L1 sequence ( Figure S7 ) , consistent with the abundant expression of L1 detected in multiple cell types [42]–[44] , whereas H3K9me3 was enriched at major satellite repeats and LTR transposons , such as IAP particles , with similar levels as previously reported [34] , [45] ( Figure 4B and Figure S7 ) . Enrichment of H1 variants at major satellites was also confirmed by calculating the normalized “IP-IN” signals at major satellite regions in mouse genome mm9 assembly ( July 2007 ) annotated by RepeatMasker ( http://repeatmasker . org ) ( Figure S8 ) . Analysis of ChIP-seq libraries of FLAG-H1d in H1d-trans ESCs , which had similar levels of FLAG-H1d and total H1/nuc ratio as the cis H1dFLAG ESCs , also showed similar level of enrichment at major satellites as H1dFLAG ESCs ( Figure S9 ) . The level of H1 has been shown to be a determinant of nucleosome repeat length ( NRL ) with a higher level of H1 correlating with a longer NRL [46] , [47] . To validate the enrichment of H1 variants at major satellites and to investigate its impact on the local chromatin structure at these regions , we measured the NRL of bulk chromatin and that of the pericentromeric ( major satellites ) and centromeric ( minor satellites ) regions with a time-course micrococcal nuclease ( MNase ) digestion assay . Southern blotting images revealed that chromatin at major satellites was more resistant to MNase digestion than bulk chromatin and minor satellites ( Figure 5A ) . Consistent with previous studies [26] , the bulk chromatin of mouse ESCs displayed a NRL of ∼187 bp ( Figure 5B ) . However , the NRL at major satellites had a value of 200 bp , which was ∼13 bp and ∼8 bp longer than the NRLs of respective bulk chromatin and minor satellites in ESCs ( Figure 5B ) . These results suggest that the enrichment of H1d and H1c at major satellite repeats may contribute to the increase of NRL in the pericentromeric region compared with bulk ESC chromatin . Analysis of H1c/H1d/H1e triple knockout ( H1 TKO ) ESCs established previously , which have an H1/nuc ratio of 0 . 25 in bulk chromatin compared with that of 0 . 46 in WT ESCs [26] , indicated that H1 depletion caused a proportional decrease of NRLs in bulk chromatin , major satellites and minor satellites ( Figure S10 ) . Consistently , qChIP analysis using a pan-H1 antibody showed total H1 levels were reduced at major and minor satellites by H1 depletion ( Figure S10D ) . Major satellite repeats at pericentric heterochromatin from different chromosomes tend to cluster together and form the chromocenter , a nuclear compartment that plays an important role in structural maintenance of the chromosomes [48] , [49] . Several chromatin proteins such as MeCP2 , MBD2 , DNMT3a , DNMT3b , and UHRF1 have been shown to contribute to chromocenter clustering [50]–[52] , however , the role of H1 in chromocenter formation has not been studied to date . Since both H1d and H1c are markedly enriched at major satellites , we set out to determine the effects of H1 depletion on chromocenter clustering in WT and H1 TKO ESCs by fluorescence in situ hybridization ( FISH ) using a major satellite specific probe . The chromocenter numbers in H1 TKO ESCs ( median = 8 , n = 160 ) were significantly lower than WT cells ( median = 17 , n = 206 ) ( Figure 6 ) , and the size of chromocenters in H1 TKO ESCs on average was bigger than that in WT ESCs ( Figure S11 ) , demonstrating a previously unnoticed defect in the pericentromeric chromatin structure caused by H1 depletion . Analysis of “rescue” ( RES ) cells established previously [53] showed that overexpressing H1d in H1 TKO cells effectively restored the size and the numbers of chromocenters to the levels comparable to WT cells ( Figure 6 and Figure S11 ) . Similarly , H1dFLAG and H1cMyc cells displayed normal chromocenter clustering as WT ESCs ( Figure S15 ) . These results indicate that the increased chromocenter clustering is likely due to the dramatic decrease of total H1 levels in H1 TKO ESCs . Pervasive transcription of repetitive sequences contributes to genome regulation , and aberrant regulation of the expression of satellite sequences interferes with heterochromatin assembly and chromosome segregation [49] , [54]–[56] . To further examine the effects of H1 depletion on major satellites , we analyzed several repetitive sequences for expression and epigenetic marks in WT and H1 TKO ESCs . Quantitative reverse transcription-PCR ( qRT-PCR ) analysis showed that the expression levels of major satellites were 3 . 5-fold higher in H1 TKO ESCs than in WT ESCs , whereas the expression levels of minor satellites and LINE L1 were not significantly changed ( Figure 7A ) . Such de-repression of major satellites by H1 depletion was dramatically curbed in RES cells ( Figure 7A ) as well as in H1dFLAG and H1cMyc ESCs ( Figure S16 ) , indicating that the levels of H1s have a direct impact on transcriptional regulation of major satellites . Notably , the levels of multiple epigenetic marks , such as repressive marks H3K9me3 , H3K27me3 , and H4K20me3 , the active mark H3K4me3 , as well as DNA methylation all remained unchanged at the analyzed repeats in H1 TKO ESCs compared with WT ESCs ( Figure 7B and 7C ) . The lack of significant changes in the histone marks and DNA methylation at these repetitive sequences suggests that the increase in expression levels at major satellites may be due to an effect of local chromatin decondensation caused by H1 depletion in H1 TKO ESCs . We note that the level of H10 , the replacement H1 variant , was increased significantly in TKO ESCs compared with that in undifferentiated WT ESCs where H10 was minimal [26] , [53] . To examine if the increased chromocenter clustering and expression of major satellites in H1 TKO ESCs could be attributed to an increase in H10 levels , we generated “fH10” cells by over-expressing FLAG-H10 in WT ESCs , and selected cell lines that expressed FLAG-H10 at a similar level to that of H10 in H1 TKO ESCs ( Figure S12 and [26] ) . As expected , FLAG-H10 was eluted in the same fraction as endogenous H10 . ChIP-seq of H10 in fH10 cells with an anti-FLAG antibody indicated that , despite its different biochemical properties and unique expression patterns [6] , [8] , [57] , H10 shared similar distribution features to that of H1d and H1c in ESCs , including depletion at active promoters and enrichment at major satellites ( Figure S3 , Figure S8 , Figure S13 , and Figure S14 ) . Similar to H1d and H1c , H10 also displayed overall positive correlation with H3K9me3 and inverse correlations with GC% and H3K4me3 , although the level of correlation was to a lesser extent ( data not shown ) . Furthermore , H10 enriched regions were significantly under-represented in gene regions but over-represented in distal intergenic regions with 80 . 1% of H10 peaks located in these regions ( data not shown ) . Beside major satellites , H10 also appeared to be enriched at minor satellites and , to a lesser extent , at LINE L1 elements as determined by ChIP-seq and ChIP-PCR ( Figure S14B and S14C ) , suggesting differential binding preferences of H10 compared with H1d and H1c . Analysis of fH10 ESCs by FISH and qRT-PCR indicated that the chromocenter numbers were not reduced compared with WT ESCs ( Figure S15 ) and that expression of major satellites remained at low levels ( Figure S16 ) , excluding the possibility of H10 upregulation being responsible for chromocenter clustering and upregulation of major satellite transcription in H1 TKO ESCs . Collectively , these results demonstrate increased chromocenter clustering and major satellite transcription by H1 depletion , and suggest important roles of the dominant H1 variants in ESCs in maintaining pericentric chromatin properties .
H1 Linker histones are abundant chromatin binding proteins that facilitate the formation of higher order chromatin structures [1] , [2] . The existence of multiple mammalian H1 variants which are differentially regulated during development presumably offers additional levels of modulation on chromatin structure and function . Despite many efforts , the in vivo localization and function of individual H1 variants in genome organization remain elusive . Chromatin plays critical roles in stem cell fate determination and reprogramming , and the epigenome of ESCs has been intensively studied . However , the genome-wide maps of one group of the major chromatin proteins , H1 variants , have not been established . Here , we have filled both gaps by generating high resolution maps of three H1 variants in mouse ESCs , identified unique H1 binding features , and discovered an unusual enrichment and function of H1 variants at major satellites . We have established a knock-in system to stringently test the functions of the tagged H1s and to facilitate the generation of high resolution maps of H1 variants in ESCs by ChIP-seq . Our results demonstrate that , when tagged at the N-terminus , the short FLAG and Myc tags , with respective 8 and 13 amino acids , do not alter the biochemical and cellular properties of H1 proteins in vivo . The strategy of homologous recombination ensures that the expression of tagged H1 variants is comparable to that of their endogenous counterparts . FLAG-H1d fully rescues the lethal phenotype of H1d deletion on H1c/H1e double knockout genetic background , further demonstrating the functional equivalence of the tagged H1 and the respective endogenous H1 variant in vivo . Although Myc-H1c was not tested in mice , it is anticipated to mimic the endogenous H1c based on all the other assays performed . These data provide a technical demonstration on how highly similar protein variants can be analyzed differentially and on a genomic scale using in vivo validated knock-in mice . On the H1 genome-wide maps we have generated here , H1d and H1c are highly correlated and display similar binding patterns in the ESC genome . Both variants are enriched at AT-rich regions , gene deserts and major satellites , but are depleted at GC-rich , gene-rich regions and especially at active promoters . Thus , despite their differences in compacting DNA in vitro and the expression patterns during development [8] , [10] , [28] , H1d and H1c are quite similar in overall distribution in the genome , which we surmise contributes to the redundancy among the major somatic H1s as suggested from previous studies of single or double H1 variants knockout mice [7] , [14] . Nevertheless , analyses of the regions that are uniquely enriched for H1d or H1c reveal some differences in sequence features ( Figure 3C and Figure S5B ) . H1c has a higher enrichment at major satellites than H1d but is relatively depleted from LINE sequences ( Figure 3C and Figure 4B ) . In addition , H1c enriched regions have a higher proportion in gene bodies and proximal regions compared with H1d peak distribution ( Figure S4B ) . These differences may account for an additional level of modulation and fine-tuning of genome function by the presence of multiple H1 variants in mammals . H10 , the H1 variant associated with differentiation , has unique expression pattern and biochemical properties . It is highly basic , expressed in differentiated cell types , and more similar to histone H5 in avian red blood cells than any other somatic variants [53] , [57] . However , overexpressed H10 ( in fH10 cells ) shares the distinctive features of H1d and H1c in ESCs in genome-wide occupancy . It is worth noting , though , that endogenous H10 proteins are present at very low levels in undifferentiated WT ESCs and the genome-wide localization of H10 in ESCs may differ significantly from its binding patterns in differentiated cells . It would be interesting to systematically determine the genome-wide maps of histone variants in different cell types , particularly in light of a recent study reporting a distribution pattern change of H1 . 5 in cellular differentiation [58] . The cell lines and mouse models generated in this study will greatly facilitate these future studies . The prevalent H1 variants binding with local troughs at active promoters we observed here in the mouse ESC genome is reminiscent of the previous results when ChIP-chip and a pan-H1 antibody were used to map H1 on a portion of the human genome in MCF-7 cells [59] or when DamID method was used to map H1 in Drosophila cells [60] . The depletion of H1 at TSSs of active genes observed in three systems suggests that this feature is common to all H1s and evolutionarily conserved . However , our study differs from the two previous studies and offers more opportunities for high resolution and in-depth analysis because the knock-in system generated in this study allows for robust and highly specific mapping of H1 variants and deep-sequencing covers the entire genome including the repetitive genome . Furthermore , we have found that the depletion of H1 at active genes is not restricted to regions around the TSS , but also expands to the entire gene encompassing domain ( Figure 2B and 2C ) . Such phenomena suggests that a wide-spread change in higher order chromatin structure may be associated with gene expression and that gene-rich domains may adopt an overall decondensed chromatin structure with less H1 occupancy . Correlation analyses indicate that H1d and H1c are inversely correlated with GC content , H3K4me3 mark , but positively correlated with H3K9me3 mark across the mouse ESC genome ( Figure 3B ) . Our finding that the common peaks of H1d and H1c are enriched with AT-rich DNA sequences in vivo resonates with the previous observation that H1 is preferentially associated with scaffold associated regions ( SAR ) [61] , which are also AT-rich sequences [62] . This binding feature may reflect a higher affinity of H1 to AT-tracts observed in in vitro studies [63] , [64] . The GC content has been suggested to be an intrinsic factor for nucleosome occupancy [65] , and our data suggest that it may also have an impact on H1 binding . It is also noteworthy that , compared with gene expression levels , H3K4me3 and H3K9me3 correlate better with H1 levels at TSS . For example , we did not observe dips of H1d and H1c around promoters of 40% genes when partitioned by H3K4me3 or H3K9me3 signals , whereas a small H1 signal dip exists even for the 20% genes with lowest expression values ( Figure 2C , 2D , and 2E ) . It is possible that the steady state level of RNA messages ( expression ) may not faithfully reflect the active/inactive state of the promoters which may correlate better with the status of histone marks . It has been reported that promoters of many genes with low expression have high H3K4me3 levels [21] , and we surmise that H1 may be absent from these gene promoters as well . The co-localization of H1d and H1c with H3K9me3 suggests that these two variants are enriched at heterochromatin and may facilitate the maintenance of constitutive heterochromatin structure . Such association may be mediated through HP1 , the heterochromatin protein binding to H3K9me3 and H3K9 methyltransferase Suv39h and facilitating spreading of heterochromatin marks [66]–[68] . Indeed , H1 has been shown to interact in vitro with HP1α [69] , [70] . On the other hand , localization of HP1 is impaired in H1 depleted Drosophila [71] , suggesting that H1 may also contribute to the proper targeting of HP1 . Surprisingly , we found that , at major satellite sequences , H1d and H1c signals are dramatically overrepresented , and this accounts for almost all the increased proportion of H1 sequence reads at repetitive sequences . The levels of H1d and H1c at major satellites are much higher than H3K9me3 ( Figure 4B ) , a repressive histone mark also enriched at these repeats [34] . The overrepresentation of H1 at major satellites in ESCs is also supported by a longer NRL , which suggests a higher local H1 level than bulk chromatin and minor satellites . Consistent with previous observations [49] , [51] , we find that major satellites are more resistant to MNase digestion than bulk chromatin and minor satellites in ESCs ( Figure 5 ) , suggesting that pericentromeric regions may adopt special higher order chromatin structure as indicated by sucrose sedimentation assay [72] . High resolution mapping in this study identifies major satellites as the dominant preferential binding sites for H1 variants in ESCs , suggesting that H1 may play an important role in mediating the formation of distinct chromatin structure at pericetromeric regions . This is further supported by the effects of H1 depletion on chromocenter clustering and expression of major satellites . We note that a higher NRL in major satellites than bulk chromatin is also present in H1 TKO ESCs ( Figure S10 ) , suggesting a possible enrichment of the remaining H1 variants at major satellite sequences in H1 TKO ESCs . Consistently , we find that overexpressed H10 also appear to preferentially accumulate at satellite sequences in ESCs ( Figure S14 ) . The enrichment of H1 at major satellites could not be solely attributed to the relatively high affinity of H1c and H1d to AT-rich sequences . Major and minor satellites sequences contain approximately 65% of A and T , with a ratio of A∶T being respective 2 . 6∶1 and 1 . 8∶1 . This could result in major satellites having more A-tracts to which H1 might have a higher affinity . Phased nucleosome positioning observed at the major satellites [73] , [74] could also contribute to the preferential binding of H1 at this region because different nucleosome positioning patterns have been shown to differentially affect H1 binding in vitro [75] . Mouse major satellites , constituting the pericentromere [76] , [77] necessary for chromosome structure and function , are shown to form clusters/chromocenters , exhibit distinct heterochromatin features and adopt a more stable and condensed chromatin conformation than the bulk chromatin [49] , [72] . Our findings of the preferential binding of H1 at major satellites and chromocenter clustering ( reduced number of chromocenters ) in H1 TKO ESCs suggest that H1 contributes to and may be required for the proper formation of pericentric heterochromatin . The rescue of the clustering effects by overexpressing H1d in H1 TKO ESCs or in H1dFLAG and H1cMyc cells compared with H1 TKO ESCs indicates that the total H1 level , rather than a specific H1 variant , is a key determining factor of chromocenter clustering . This conclusion is further supported by our finding that overexpressing H10 level to 3 . 5 fold of that of endogenous H10 in WT ESCs has little effect on chromocenter numbers or major satellite expression . In vitro studies have shown highly cooperative binding of H1 globular domain to DNA [78] , a property which we speculate could contribute to increased chromocenter clustering in the face of marked reduction of H1 levels in H1 TKO ESCs . A larger nucleosome spacing ( 200 bp ) ( Figure 5 ) together with a higher local H1 level at major satellites could be important for efficient compaction of pericentromeric chromatin because nucleosome arrays with a NRL of 197 bp are able to form 30 nm fiber structure in vitro in the presence of linker histones whereas arrays with a short NRL are only able to form thinner and less compact structures [5] . The effects of H1 on major satellites are not restricted to chromatin structure and heterochromatin formation . Loss of H1c , H1d and H1e causes a dramatic increase in transcripts from major satellites , but does not change the levels of the repressive epigenetic marks , H3K9me3 , H4K20me3 , H3K27me3 , or DNA methylation at these sequences . This suggests that the increase in expression of major satellites in H1 TKO ESCs is not mediated by loss of these repressive epigenetic marks , but rather caused by reduced binding of H1 per se or the potential decondensation of local chromatin structure . The phenomenon of changes in chromocenter organization independent of H3K9me3 is reminiscent of results from deletion of UHRF1 [52] , a histone binding protein or overexpression of MeCP2 in mouse myoblasts [50] . Chromocenter organization is likely to be independent of H3K9me3 pathway because double deletion of Suv39h1 and Suv39h2 has minimal effects on the number and size of chromocenters in mouse cells [79] , [80] . The expression changes in major satellites in H1 TKO ESCs are also not due to potential changes in cell cycle since H1 TKO ESCs have similar growth rate [26] and cell cycle profiles ( data not shown ) to WT ESCs . The reduction in expression levels of major satellites detected in RES cells compared with H1 TKO cells further supports that the drastic decrease in H1 levels causes de-repression of major satellites . Noncoding major satellite transcripts have been shown to be important for proper chromocenter formation [81] , thus we speculate that the increased levels of major satellite transcripts contribute to chromocenter clustering in H1 TKO cells . In light of previous findings that ESCs null for DNA methyltransferases displayed chromocenter clustering [51] , similar to what we observed in H1 TKO ESCs , we surmise that H1 and DNA methylation may act cooperatively in the proper maintenance of chromocenter structure . In summary , we report high resolution maps of two abundant somatic H1 variants and the replacement H1 variant in mouse ESCs , connecting this important yet under-explored repressive mark with the well-studied ESC epigenome . The enrichment and effects of H1d , H1c and H10 on major satellites highlight an important role of these H1 variants in the maintenance of chromosome architecture and function . The cell lines and mouse strains we generated using the knock-in system also provide valuable tools for studying H1 variant specific functions both in vitro and in vivo . Genome-wide distribution studies of other H1 variants as well as in differentiated cell types are likely to lead to a better understanding of the role of H1 and higher order chromatin folding in gene expression and chromatin function .
The H1dFLAG knock-in targeting vector containing H1d 5′ and 3′ homology regions flanking the N-terminal FLAG-tagged H1d and the SV40-Blasticidin resistant gene was transfected into ESCs as described previously [14] . 200 ESC clones resistant to 20 µg/ml Blasticidin ( Life Technologies ) and 2 µM gancyclovir ( Sigma-Aldrich ) were picked , and 5 clones with homologous recombination were identified by Southern blotting using the probe shown in Figure 1A . Two cis-targeted clones were injected into C57BL/6 recipient blastocysts to produce chimeric mice , which gave germline transmission . H1c+/−H1d+/FLAGH1e+/− mice were intercrossed to generate H1c−/−H1dFLAG/FLAGH1e−/− ( H1dFLAG/FLAG ) mice . All animal work was performed according to procedures approved by the Institutional Animal Care and Use Committee ( IACUC ) at Georgia Institute of Technology . Nuclei and chromatin of ESCs and mouse tissues were prepared and analyzed according to protocols described previously [82] , [83] . Histones were extracted from chromatin with 0 . 2 N sulfuric acid and 50–100 µg of total histone preparations were injected into a C18 reverse phase column ( Vydac ) on an ÄKTA UPC10 system ( GE Healthcare ) . The effluent was monitored at 214 nm ( A214 ) , and the peak areas were recorded and analyzed with ÄKTA UNICORN 5 . 11 software . The A214 values of the H1 and H2B peaks were adjusted by the number of peptide bonds in each H1 variant and H2B . The H1/nucleosome ratio was determined by dividing the A214 of all H1 peaks by half of the A214 of the H2B peak . Fractions corresponding to different H1 variants from HPLC analysis were collected , lyophilized and analyzed with silver staining , Coomassie staining and Western blotting . The following antibodies were used in this study: anti-FLAG ( Sigma-Aldrich F3165 ) , anti-DYKDDDDK tag ( Cell Signaling #2368 ) , anti-Myc-tag ( Cell Signaling #2272 ) , anti-H3K4me3 ( Millipore 07-473 ) , anti-H3K9me3 ( Abcam 8898 ) , anti-H3K27me3 ( Millipore 07-449 ) , anti-H4K20me3 ( Millipore 07-463 ) , anti-H10 ( Santa Cruz 56695 ) , anti-H1 ( Milipore 05-457 ) and IgG ( Millipore 12-370 ) . ChIP assays were performed as described previously [26] with the following modifications: 20 µl of Dynabeads Protein G ( Life Technologies ) were incubated with 2 µg of antibody for 4 hours , followed by incubation with 40 µg of sonicated soluble chromatin overnight at 4°C . Dynabeads were washed , immunoprecipitates were eluted , and DNA-protein complexes were incubated overnight at 65°C to reverse crosslinks . DNA was purified with a DNA Isolation column ( Qiagen ) . Input control DNA was prepared from reverse-crosslinked soluble chromatin prior to immunoprecipitation . Quantitatitve PCR on ChIP samples for major satellites , minor satellites , LINE L1 , IAP LTR and Hprt was performed with primers published previously [45] , [84] . The libraries for massive parallel sequencing were prepared with the ChIP-seq Sample Preparation Kit ( Illumina ) according to the manufacturer's instructions . Briefly , 10 ng of immunoprecipiated DNA or input DNA were end repaired , 3′ adenylated and ligated with adapter oligos supplied by the manufacturer . DNA fragments within the range of 120∼500 bp were purified following gel electrophoresis and amplified with primers provided by the manufacturer . Library DNA was subsequently purified with a Qiagen DNA Isolation column , quantified and submitted for sequencing . Sequencing was performed with Illumina Genome Analyzer II and Illumina HiSeq 2000 systems , and raw sequence reads containing more than 30% of ‘N’ were removed and adaptor sequences were trimmed . Clean sequences were aligned against mouse genome , mm9 ( UCSC website ) , and 2 , 669 categories of mammalian repeats from RepBase version 14 . 07 [39] , [40] using Bowtie aligner software ( http://bowtie-bio . sourceforge . net/index . shtml ) . The first 40 bp ( for alignment to mm9 ) or the first 35 bp ( for alignment to RepBase ) of the reads were used as seed sequences with up to two mismatches allowed for the alignment , and aligned number of reads were scored . Reads with multiple alignment positions were mapped randomly to one of the possible position . Reads for each ChIP-seq or input-seq library aligned to mm9 were normalized to 10 million reads , and IP-IN signals were calculated in each 100 bp sliding window by subtraction of normalized read counts per 10 million mappable reads of ChIP-seq library by that of its corresponding input-seq library using GenPlay software ( http://genplay . einstein . yu . edu/wiki/index . php/Documentation ) [33] ) . Percentage of reads for each repeat mapped to RepBase was calculated by dividing reads mapped to the respective repeat by the total reads in the library , and the fold enrichment for the respective repeat was subsequently calculated as the ratio of the percent of reads of ChIP-seq library to that of the input-seq library . Read length and read counts of each library are listed in Table S1 . Representative ChIP-seq libraries with the most sequencing reads mapped to mm9 were utilized for genome browser visualization and metagene analysis , and all replicate ChIP-seq libraries were included in repetitive sequence analysis . Sequencing data have been deposited in NCBI's Gene Expression Omnibus database and assigned GEO Series accession number GSE46134 . The sum of signals ( IP-IN ) for each 1000 bp window ( normalized to 10 million reads ) was used to calculate the correlation coefficients of H1 variants with GC% and different histone markers . Genome-wide and chromosome-wide correlation coefficients were calculated , and the scatter-plots were generated using Matlab . Significantly enriched regions were identified using GenPlay or SICER v1 . 1 [36] at the following parameter settings: window size = 200 , gap size = 600 , E-value = 1000 , an effective genome size of 80% of the entire mouse genome , and q-value ( FDR ) = 0 . 001 . In order to optimize the gap size for H1 variants , the gap size was varied from 0 to 3 times the window size ( 0 , 200 , 400 , 600 ) and the best value was chosen according to the criteria as previously described [36] . Distribution of peak regions relative to gene regions was analyzed by CEAS [37] . Top 10% of enriched regions for each ChIP-seq library were selected to identify the overrepresented features using EpiGRAPH ( http://epigraph . mpi-inf . mpg . de/WebGRAPH/ ) [38] . 2214 H1d/H1c common peaks , 1939 H1d unique peaks , 433 H1c unique peaks , 1891 H3K9me3 peaks , 4778 H3K27me3 peaks , and 3446 H3K4me3 peaks were analyzed by EpiGRAPH . ESC nuclei were extracted and MNase digestion was performed as described previously [26] . Briefly , 2 . 5×106 nuclei were resuspended in 200 µl of MNase digestion buffer ( 0 . 32 M sucrose , 50 mM Tris-HCl pH 7 . 5 , 4 mM MgCl2 , 1 mM CaCl2 , 0 . 1 mM PMSF ) and digested at 37°C with 20 units of micrococcal nuclease ( MNase ) ( Worthington ) for time course analysis or 2 units of MNase ( Worthington ) for 5 min in analysis shown in Figure S10A . Nuclei were lysed and DNA was subsequently purified and analyzed by electrophoresis . Southern blotting was performed using major or minor satellite specific probes as described previously [26] . The NRL at each time point was calculated using the regression line generated with size ( bp ) of polynucleosomes [7] , [26] , and the values at time “0” were extrapolated as described previously [72] . FISH was performed as described previously [85] . The major satellite probe was biotin-labeled , denatured and hybridized to the slides overnight . The nuclei were incubated with FITC-Avidin for 1 hour , and counterstained with DAPI . Signals were detected with an Olympus Epifluorescence Microscope ( Olympus , Inc . ) equipped with an Olympus QCLR3 cooled digital camera . The experiments were repeated three times , and the number of chromocenters for each cell line was counted by three researchers as blind tests . Statistical analysis was performed using a Mann-Whitney U nonparametric test . Areas of chromocenters were quantitated using AxioVision software V4 . 8 . 2 . 0 and presented as pixel2 . The conversion factor of pixel/micron was 18 . 7 pixels per micrometer . 1 µg of total RNA extracted from ESCs was treated with RNase free DNaseI ( Sigma-Aldrich ) and reverse transcribed using a SuperScript first-strand cDNA synthesis kit with random hexamers ( Life Technologies ) . Triplicate PCR reactions using the iQ SYBR Green Supermix ( Bio-Rad ) were analyzed in a MyIQ Real-Time PCR Detection System ( Bio-Rad ) . All samples were typically analyzed in two independent experiments . Relative expression units were calculated by subtracting the mock reverse-transcribed signals ( RT− ) from reverse transcribed signals ( RT+ ) and normalizing the adjusted values with signals of the housekeeping gene GAPDH . The qRT-PCR primers for repetitive sequences are the same as in qChIP , and the primers for GAPDH are as described previously [53] . 1 µg of DNA extracted from ESCs was treated with the CpGenome DNA modification Kit ( Millipore ) according to the manufacturer's manual . 20 ng of treated DNA was used in each PCR reaction as previously described [26] . The primers used to generate PCR products from the bisulfite-converted DNA are specific for the converted DNA sequence of the analyzed regions . The PCR products were subsequently cloned using the TOPO TA Cloning kit ( Life Technologies ) , and colonies containing the converted DNA inserts were picked . DNA inserts were sequenced and analyzed with BiQ Analyzer [86] . Primers for major and minor satellites were as previously described [87] .
|
Embryonic stem cells ( ESCs ) possess unique chromatin and epigenetic signatures , which are important in defining the identity and genome plasticity of pluripotent stem cells . Although ESC epigenomes have been extensively characterized , the genome localization of histone H1 variants , the chromatin structural proteins facilitating higher-order chromatin folding , remains elusive . Linker histone H1 is essential for mammalian development and regulates the expression of specific genes in ESCs . Here , by using a knock-in system coupled with ChIP–seq , we first achieve the high resolution mapping of two H1 variants on a genome-wide scale in mouse ESCs . Our study reveals the correlations of this underexplored histone family with other epigenetic marks and genome attributes . Surprisingly , we identify a dramatic enrichment of H1d and H1c at major satellite sequences . H10 , mapped using an overexpressing ESC line , shows similar features at active promoters but differential binding at repetitive sequences compared with H1d and H1c . Furthermore , using mutant ESCs that are deficient for multiple H1 variants , we demonstrate the role of H1 in chromocenter clustering and transcriptional repression of major satellites . Thus , these results connect this important repressive mark with the well understood ESC epigenome and identify novel functions of H1 in mammalian genome organization .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"chromosome",
"structure",
"and",
"function",
"gene",
"regulation",
"dna",
"transcription",
"gene",
"function",
"histone",
"modification",
"genome",
"sequencing",
"developmental",
"biology",
"stem",
"cells",
"epigenetics",
"molecular",
"genetics",
"chromatin",
"chromosome",
"biology",
"embryonic",
"stem",
"cells",
"gene",
"expression",
"biology",
"genetics",
"genomics",
"stem",
"cell",
"lines"
] |
2013
|
High-Resolution Mapping of H1 Linker Histone Variants in Embryonic Stem Cells
|
Human movement is a key behavioral factor in many vector-borne disease systems because it influences exposure to vectors and thus the transmission of pathogens . Human movement transcends spatial and temporal scales with different influences on disease dynamics . Here we develop a conceptual model to evaluate the importance of variation in exposure due to individual human movements for pathogen transmission , focusing on mosquito-borne dengue virus . We develop a model showing that the relevance of human movement at a particular scale depends on vector behavior . Focusing on the day-biting Aedes aegypti , we illustrate how vector biting behavior combined with fine-scale movements of individual humans engaged in their regular daily routine can influence transmission . Using a simple example , we estimate a transmission rate ( R0 ) of 1 . 3 when exposure is assumed to occur only in the home versus 3 . 75 when exposure at multiple locations—e . g . , market , friend's—due to movement is considered . Movement also influences for which sites and individuals risk is greatest . For the example considered , intriguingly , our model predicts little correspondence between vector abundance in a site and estimated R0 for that site when movement is considered . This illustrates the importance of human movement for understanding and predicting the dynamics of a disease like dengue . To encourage investigation of human movement and disease , we review methods currently available to study human movement and , based on our experience studying dengue in Peru , discuss several important questions to address when designing a study . Human movement is a critical , understudied behavioral component underlying the transmission dynamics of many vector-borne pathogens . Understanding movement will facilitate identification of key individuals and sites in the transmission of pathogens such as dengue , which then may provide targets for surveillance , intervention , and improved disease prevention .
Historically epidemiologists have viewed human movement from the perspective of populations of susceptible hosts moving into high risk areas or infected hosts moving into susceptible populations as explanation for disease occurrence and spread . Indeed , across different scales and diseases , movements of hosts affect pathogen transmission in a variety of ways . Thirty years ago Prothero [9] provided a typology to facilitate study of the role of human movements in epidemiology based on his experience in Africa . Drawing on geography literature concerned with understanding human movement [10] , [11] , [12] , Prothero highlighted the difference between circulatory movements , where individuals return home after some period , and migratory movements , which tend to be permanent changes of residence ( see Figure 1 in [11] ) . He further characterized movements by their ‘spatial scale’ , which he categorized in terms of a rural-urban gradient , and temporal scale based on the time and timing of displacements . He qualified these categories in terms of their relevance to public health . For instance , seasonal movements from one rural area to another for agriculture could potentially expose individuals to different ‘ecological zones’ where the risk of malaria or African trypanosomiasis is high [13] . His argument was that knowing something about the nature of such movements would help explain the incidence and prevalence of disease in a population and provide informed options for control [9] . In Figure 1 we generalize Prothero's typology in terms of the spatial and temporal scale ( sensu [14] ) of human movement and extend it to include most vector-borne disease contexts . At broad spatial scales ( e . g . , national , international ) individual movements drive pathogen introduction and reintroduction ( far right , Figure 1 ) . Global spread of dengue virus via shipping routes was characterized by periodic , large , spatial displacements [15] . Globalization and air transportation have changed the dynamic of pathogen spread by dramatically shortening the time required to travel around the globe [16] , [17] , [18] . The recent chikungunya epidemic in the Indian Ocean that subsequently spread to Italy is an example [19] . At finer scales ( e . g . , regional , urban-rural , intra-urban; far left of Figure 1 ) , movement associated with work , recreation , transient migration , and other phenomena is important to patterns of pathogen transmission and spread [9] , [20] . Movements into high-risk areas not only lead to individual infection , but can also contribute to local transmission when infected hosts return home and infect competent vectors . For example , in the Chocó region of Colombia most malaria transmission occurs in rural areas and many cases diagnosed in the city of Quibdó are due to travel to these areas [21] . Transmission also occurs locally within Quibdó [22] , however , most likely because of infected travelers returning and infecting competent vectors . Understanding the origin of infections and the relative importance of human movement at different scales to both local and regional transmission dynamics would increase effectiveness of disease prevention programs by , for example , identifying individuals at greatest risk of contracting and transmitting pathogen . Generally , a key significance of human movement for vector-borne disease at any scale lies with exposure to vectors . Exposure is local in space and time and variation in exposure due to individual host movement could strongly influence the transmission dynamics of pathogens . For instance , circulatory movements associated with working in rural areas and variation in movement patterns among cultures may explain heterogeneous patterns of onchocerciasis incidence . While men in Cameroon and Guatemala both experience similar parasite loads reflecting exposure to vectors when working in fields , women in the 2 countries show different patterns of infection partly due to differences in exposure [23] . The type of movement most relevant for exposure will depend on site specific differences , the ecology of the arthropod vector , human behavior , and the relative scale of host and vector movement . For pathogens transmitted by vectors able to move long distances in search of a host , fine scale host movements may not be important , while they are for pathogens transmitted by sessile vectors . Aedes aegypti—the principal vector of dengue virus—bites during the day [24] , disperses only short distances [25] and is heterogeneously distributed within urban areas [26] , [27] . Conversely , humans move frequently at local scales ( bottom-left of Figure 1 ) , allocating different amounts of time to multiple locations on a regular basis . This will influence individual risk of infection with dengue virus [28] and thus overall patterns of transmission [29] , [30] , [31] .
The dynamics of human movement , the locations used and the paths between them , is conceptualized by the ‘activity space’ model developed in the 60's by human geographers [12] , [32] , [33] . This model , much like the ‘home-range’ concept in ecology , is effective because organisms exhibit habitual behavior in their use of space [34] . For our purposes of studying dengue , the ‘activity space’ refers to those few locations where humans commonly spend most of their time [32] , [35] and ‘movement’ refers to the use of these locations . Thus , exposure to host-seeking female Ae . aegypti is the sum of exposure across an individual's activity space . For other vectors and pathogens , human movements per se ( e . g . , walking between the house and a water source ) and/or visits to less common destinations could be relevant for the transmission of other pathogens ( e . g . , African trypanosomiasis ) depending on the behavior of the vector and the relative scales of vector and host movement . The activity space model represents movement associated with the regular activity of individuals [36] . We present a version of this model in Figure 2 for understanding how movements within an urban area might contribute to risk of exposure . Risk at locations within an individual person's activity space will vary depending on the number of infected , host seeking vectors present and their biting behavior . For instance , visits to locations during the day are of minimal risk for bites from nocturnal An . gambiae , but are relatively high for day active Ae . aegypti ( Figure 2 ) . Exposure to vector bites may also depend on how long a person stays at a given location . If vectors are stimulated by the arrival of an individual to a location ( as may be the case for Ae . aegypti and Aedes albopictus [37] , [38] ) , then a bite is most likely to occur early after arrival ( i . e . the cumulative probability of a bite during a visit , e ( t ) , accumulates rapidly ) . Alternatively , for vectors like triatomine bugs , which are less opportunistic than mosquitoes , long visits will be expected to pose a higher risk of host-vector contact ( e ( t ) slowly accumulates over time ) . How vectors respond to hosts arriving at a site is important because it weights the risk of visits differently depending on their frequency and duration . If a vector is stimulated to host seek by the arrival of a host , then multiple short visits to that site will carry greater risk than a single long visit of equivalent total duration . In summary , a person's risk of exposure to an infective vector can be represented with a simple exposure model for indirectly transmitted disease: ( 1 ) Here , the risk of exposure ( i . e . , being bitten by a vector ) for individual i , ri , over some observation period is simply the sum across sites visited , j , of vector abundance , Vj , conditioned on the time and duration of all visits to that site , k , as determined by vector behavior ( where K is the total number of visits during the observation period ) . The biting rate , ak , is the number of bites expected per visit and is drawn from the day biting rate distribution for the times of the visit . ( 2 ) How vectors respond to the appearance of a host at a site is captured by ek , the cumulative probability of a bite given the time spent in the site , and is bounded by the unit interval . ( 3 ) Visits , k , are defined by an arrival time , t0 , and a departure time , t1 , in hours and are in reference to a single day . At the limit ( where t1−t0 = 24 hours ) , ak becomes the day biting rate , a , and ek goes to 1 and we recover the model often assumed for vector-borne diseases where exposure occurs in the household . Note that although we imply here that a site comprises a household or other edifice because of our focus on dengue , in truth it simply demarcates a location where the abundance and activity of vectors is independent of other locations and is defined by the scale of vector movement . Site-specific exposure risk is calculated as: ( 4 ) and has units of bites*humans for the observation period . Note that in this formulation , risk among individuals using the same site is assumed to be independent ( i . e . , the expected number of bites at a site is the product of humans present and vector activity ) . This may not be realistic if hosts occupy a site at the same time , which would be expected to dilute the number of bites individual hosts receive , and can be corrected ( see below ) by incorporating the actual amount of time individual humans spend in a location . The estimate of risk , rj , can be used to estimate the transmission rate , R0 , which is the number of secondary infections expected from the introduction of a single infective individual into a wholly susceptible population . Woolhouse et al . ( 1997 ) use the following approximation for R0: ( 5 ) where vj is the proportion of vectors at site j , hj is the proportion of hosts living in site j , and J is the total number of sites . Risk as estimated above is incorporated by replacing vj with site associated risk , rj , discounted by the proportional use of that site within some interval by people , hj: ( 6 ) For example , if a site is used by 2 individuals for 6 hours each over a week , hj = ( 2 humans * 6 hours ) / ( 24 hours/day * 7 days ) = 0 . 07 humans . The activity space model elaborated here illustrates that host and vector behavior are very important for determining who gets bitten and has the greatest risk of contracting or transmitting a pathogen .
The activity space model when coupled with our knowledge of vector behavior provides a tool for determining what human movements are important for transmission ( e . g . , Figure 1 ) . Specifically , it allows us to identify places and individuals that contribute disproportionately to pathogen transmission dynamics . For example , consider the following scenario depicted in Figure 3 for a human population at risk for dengue virus infection like the one we are studying in Iquitos , Peru ( Figure 3 , Text S1 and Table S1 ) . Briefly , individuals spend their time at a number of different sites , both commercial and residential , during their regular weekly activities ( Sites , first column in Figure 3 ) . Sites have different numbers of female mosquitoes and are visited at different rates and for different durations . We can estimate the risk of exposure to host-seeking female mosquitoes ( ri ) for each person ( columns 1–13 in Figure 3 ) at each site ( rows in Figure 3 ) and then estimate R0 . In this particular example , R0 as approximated when accounting only for the home ( eq . 5 ) is 1 . 3 and the site with the highest estimated risk is house 5 ( in bold in column under R0 ) . If we account for exposure at all locations in addition to the home and assume the biting rate at night is 10% of the rate during the day [39] , our estimate of R0 ( eq . 6 ) jumps nearly 3-fold and the most important site is 13 , a clinic ( in bold under R0e ) . This latter result arises because of the relatively large number of bites per person expected at that site , determined largely by the significant amount of time a single person spends there ( e . g . , their workplace ) . In this example , all individuals except individual 10 experience the greatest exposure to bites in their homes because that is where they spend the most time . Individual 10 , however , experiences the highest risk at site 4 , which represents their workplace . This individual is also at the greatest risk in the host population . This example illustrates that the key sites are not necessarily those of greatest vector abundance , as is commonly assumed . For this example scenario , R0j increases monotonically with vector abundance when transmission is assumed to occur only in the home ( Figure 4 ) . When exposure rates are accounted for , however , there is no relationship between R0j and vector abundance ( Figure 4 ) . Similarly , people living where vector abundance is greatest are not necessarily at greater risk . Human movement and subsequent variation in exposure thus becomes more important than vector density per se . Because heterogeneity in contact patterns has a large influence of the rate of pathogen transmission , variation in exposure rates due to individual movement patterns could have considerable impact on disease dynamics [40] , [41] .
As an aid to future research , in the remainder of this article we discuss key issues and considerations for designing studies of human movement based on our experiences with dengue . Because patterns of contact between pathogens and susceptible hosts are heterogeneous , disease interventions can be made more effective and efficient by targeting the key points or ‘nodes’ of transmission [3] . Even where heterogeneous patterns are clearly documented , not knowing the factors driving such patterns impedes one's ability to effectively target control . Is a biting preference toward young adults [60] because they are intrinsically more attractive to a host-seeking mosquito or , because of their behavior , they are more likely to be exposed to mosquitoes ? Although many different causes of host-vector contact heterogeneity have been proposed ( summarized by [6] ) , variation in exposure due to human behavior is likely to be key across disease systems . The role of other risk factors ( e . g . , host-preference ) will always be conditioned by exposure rates . The study of human movement is thus critical to the identification of key individuals and key locations . Nevertheless , movements have largely been neglected in studies of indirectly transmitted disease even though it is becoming increasingly easy to measure . Quantifying and describing human movements promises more than just characterization of key heterogeneities . Quantification of the collective dynamics of human populations provides information necessary for models intended to predict disease outbreak and spread and to evaluate control alternatives to halt epidemics [8] , [35] , [51] . Buscarino et al . [61] , for instance , predict that movements within a population have an important effect on the epidemic threshold , lowering this as individuals move over larger distances more frequently . Additionally , quantifying movements and applying that information to a variety of diseases creates the opportunity to identify common places where infection occurs across diseases and , thus , the potential to leverage public health programs by allowing limited resources to be targeted to the most important locations for more than one disease . Rigorous examination of the role of human movement across different scales will significantly improve understanding of pathogen transmission , which will be critical to increasing the effectiveness of disease prevention programs . As transmission rates are reduced through intervention efforts , we expect the importance of heterogeneity in exposure to increase and to play an even more important role in pathogen persistence . Characterization of movements will thus not only facilitate the elimination of disease , it will help to prevent its return .
|
Vector-borne diseases constitute a largely neglected and enormous burden on public health in many resource-challenged environments , demanding efficient control strategies that could be developed through improved understanding of pathogen transmission . Human movement—which determines exposure to vectors—is a key behavioral component of vector-borne disease epidemiology that is poorly understood . We develop a conceptual framework to organize past studies by the scale of movement and then examine movements at fine-scale—i . e . , people going through their regular , daily routine—that determine exposure to insect vectors for their role in the dynamics of pathogen transmission . We develop a model to quantify risk of vector contact across locations people visit , with emphasis on mosquito-borne dengue virus in the Amazonian city of Iquitos , Peru . An example scenario illustrates how movement generates variation in exposure risk across individuals , how transmission rates within sites can be increased , and that risk within sites is not solely determined by vector density , as is commonly assumed . Our analysis illustrates the importance of human movement for pathogen transmission , yet little is known—especially for populations most at risk to vector-borne diseases ( e . g . , dengue , leishmaniasis , etc . ) . We outline several important considerations for designing epidemiological studies to encourage investigation of individual human movement , based on experience studying dengue .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases",
"ecology/behavioral",
"ecology",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/viral",
"infections",
"public",
"health",
"and",
"epidemiology/social",
"and",
"behavioral",
"determinants",
"of",
"health",
"ecology/spatial",
"and",
"landscape",
"ecology",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] |
2009
|
The Role of Human Movement in the Transmission of Vector-Borne Pathogens
|
Genetic variants underlying complex traits , including disease susceptibility , are enriched within the transcriptional regulatory elements , promoters and enhancers . There is emerging evidence that regulatory elements associated with particular traits or diseases share similar patterns of transcriptional activity . Accordingly , shared transcriptional activity ( coexpression ) may help prioritise loci associated with a given trait , and help to identify underlying biological processes . Using cap analysis of gene expression ( CAGE ) profiles of promoter- and enhancer-derived RNAs across 1824 human samples , we have analysed coexpression of RNAs originating from trait-associated regulatory regions using a novel quantitative method ( network density analysis; NDA ) . For most traits studied , phenotype-associated variants in regulatory regions were linked to tightly-coexpressed networks that are likely to share important functional characteristics . Coexpression provides a new signal , independent of phenotype association , to enable fine mapping of causative variants . The NDA coexpression approach identifies new genetic variants associated with specific traits , including an association between the regulation of the OCT1 cation transporter and genetic variants underlying circulating cholesterol levels . NDA strongly implicates particular cell types and tissues in disease pathogenesis . For example , distinct groupings of disease-associated regulatory regions implicate two distinct biological processes in the pathogenesis of ulcerative colitis; a further two separate processes are implicated in Crohn’s disease . Thus , our functional analysis of genetic predisposition to disease defines new distinct disease endotypes . We predict that patients with a preponderance of susceptibility variants in each group are likely to respond differently to pharmacological therapy . Together , these findings enable a deeper biological understanding of the causal basis of complex traits .
Genome-wide association studies ( GWAS ) have considerable untapped potential to reveal new mechanisms of disease[1] . Variants associated with disease are over-represented in regulatory , rather than protein-coding , sequence; this enrichment is particularly strong in promoters and enhancers[2–4] . There is emerging evidence that gene products associated with a specific disease participate in the same pathway or process[5] , and therefore share transcriptional control[6] . We have recently shown that cell-type specific patterns of activity at multiple alternative promoters[7] and enhancers[3] can be identified using cap-analysis of gene expression ( CAGE ) to detect capped RNA transcripts , including mRNAs , lncRNAs and eRNAs[3 , 5] . In the FANTOM5 project , we used CAGE to locate transcription start sites at single-base resolution and quantified the activity of 267 , 225 regulatory regions in 1824 human samples ( primary cells , tissues , and cells following various perturbations ) [8] . Unlike analysis of chromatin modifications or accessibility , the CAGE sequencing used in FANTOM5 combines extremely high resolution in three relevant dimensions: maximal spatial resolution on the genome , quantification of activity ( transcript expression ) over a wide dynamic range , and high biological resolution–quantifying activity in a much wider range of cell types and conditions than any previous study of regulatory variation[2 , 4] . Since a majority of human protein-coding genes have multiple promoters[5] with distinct transcriptional regulation , CAGE also provides a more detailed survey of transcriptional regulation than microarray or RNAseq resources . Heritability of traits studied by some GWAS is substantially enriched in these FANTOM5 promoters[9][10] . Genes that are coexpressed are more likely to share common biology[11 , 12] . Similarly , regulatory regions that share activity patterns are more likely to contribute to the same biological pathways[5] . We have previously shown transcriptional activity of regulatory elements ( both promoters and enhancers[3] ) is associated with variable levels of expression arising at these elements in different cell types and tissues[5] . Informative regulatory networks can be derived from predicted transcription factor interactions with FANTOM5 regulatory regions[6] . We therefore use transcript expression here as a surrogate for transcriptional regulatory activity . In contrast to previous studies[6 , 13 , 14] , we sought to explore the similarities in activity at disease-associated sets of regulatory regions , rather than genes , and independent of transcription factor binding predictions . In order to determine whether coexpression of regulatory elements can provide additional information to prioritise genome-wide associations that would otherwise fall below genome-wide significance , we developed network density analysis ( NDA ) . The NDA method combines genetic signals ( disease association in a GWAS ) with functional signals ( correlation in promoter and enhancer-associated transcript levels measured by CAGE across numerous cell types and tissues , Fig 1 ) , by mapping genetic signals onto a pairwise coexpression network of regulatory regions , and then quantifying the density of genetic signals within the network . Every expressed regulatory region that contains a GWAS SNP associated with a given trait is assigned a score quantifying its proximity in the network to every other regulatory region containing a GWAS SNP for that trait . We then identified specific cell types and tissues in which there is preferential activity of regulatory elements associated with selected disease-related phenotypes , thereby providing appropriate cell culture models for critical disease processes .
For the purpose of this analysis , promoters identified in the FANTOM5 dataset were defined as the region from -300 bases to +100 bases from a transcription start site[15] . Previous analysis demonstrated that this covers the areas of maximal sequence conservation across species and the core region of transcription factor binding . Enhancers are widely transcribed across the human genome ( eRNAs ) . Since eRNA TSS are considerably longer than promoter TSS ( median length ( IQR ) 272 ( 173–367 ) vs 15 ( 9–26 ) ) , enhancers were defined by the range covered by eRNA transcription start sites . For each GWAS study , SNPs were identified that lie within either a functional promoter or enhancer . Any promoter or enhancer that contained a variant putatively associated with a given phenotype was considered to be candidate phenotype-associated regulatory region . A pairwise matrix was then generated from the full FANTOM5 dataset of promoters and enhancers , in which each node is a regulatory region , and edges reflect the similarity in activity ( expression ) patterns arising at these regulatory regions , across different cell types and tissues . To test the hypothesis that regulatory regions genetically associated with a given phenotype are more likely to share activity patterns , we devised the NDA method , which quantifies the strength of coexpression among a chosen pool of putative phenotype-associated regulatory regions . This approach avoids arbitrary cut-offs between clusters ( or “communities” ) of nodes , and yields a single value for each node , quantifying the closeness with all other nodes in a particular subset ( network density ) . NDA was used to integrate the putative association between a regulatory sequence and the phenotype of interest ( indicated by the presence of a phenotype-associated SNP ) , with the coexpression similarity between this node with other nodes that are also putatively associated with the same phenotype . NDA integrates information from two distinct and independent sources: the relationships between nodes in the network , and the choice of subset . In the present work , nodes are regulatory regions , the subset is those regulatory regions that contain variants associated with a particular phenotype . Spearman’s rank correlation was chosen to quantify pairwise relationships , in view of the robustness of this measure in a variety of different distributions . However , the NDA approach is generalisable to any network of pairwise relationships . Within a network of all possible pairwise relationships between nodes , a subset of nodes is selected that share a particular characteristic . Within this subset of nodes , every pair of nodes is considered . Each relationship between two nodes is expressed as the –log10 of the empirical probability of a relationship at least as strong occurring between the chosen node and another , randomly-chosen , node from anywhere in the network . These probabilities are specific to each node and are directional . The NDA score is the sum of the –log10 ( p ) values for a node in the chosen subset and all other nodes within the subset . The NDA score therefore quantifies the density of this subset of nodes in network space . The purpose of using the empirical probability of a correlation , rather than the raw correlation metric , is to control for bias in favour of highly-connected nodes , as would occur if one expression profile were very common . Finally , the NDA score is assigned its own p-value by comparison to that obtained using randomly permuted subsets ( see below ) . If the network contains no additional information about this subset of nodes , then the relationships between nodes in the chosen subset will be no stronger than the relationships seen in permuted subsets . From the set of all nodes in a network , a subset is selected because they share some characteristic . In the case of the genomic analyses reported here , the nodes are TSS , and the subset of interest is those TSS that contain a variant that has some evidence of association with a particular trait . Throughout this paper , we have defined the set of phenotype-associated transcription start sites , R , as follows: the set of regulatory elements associated with phenotype-associated single nucleotide polymorphism within 300bp ( promoters ) or 0bp ( enhancers ) upstream from a FANTOM5 transcription start site ( TSS ) and 100bp ( promoters ) or 0bp ( enhancers ) downstream . In order to enable the detection of new associations , we use a deliberately permissive threshold . We define as “putatively-significant” a SNP-phenotype association of p < 5 × 10−6 . Let the integer variable i be used to index the base pairs ( bp ) of the genome . For a given trait , the set of input SNPs , K , are those that have a putatively-significant association with that trait at our chosen threshold . If we let TSSstart equal the base pair index 300bp ( promoters ) or 0bp ( enhancers ) upstream from a FANTOM5 transcription start site ( TSS ) and TSSend 100bp ( promoters ) or 0bp ( enhancers ) downstream , the set , P , of putative trait-associated promoters is given by: P={i:i∈K , TSSstart−300≤i≤TSSend+100} and the set E of enhancers containing a putative trait-associated SNP is given by: E={i:i∈K , TSSstart≤i≤TSSend} giving a total set of regulatory regions: R=P∪E Input SNPs from GWAS results tend to be in LD with nearby variants . There is therefore a risk of spurious coexpression , since nearby regulatory regions are also likely to share regulatory influences , such as chromatin accessibility , enhancers , and lncRNAs . One solution to this would be to filter input SNPs by LD . However this would require that LD relationships for all SNPs be known for all of the populations from which SNP association data were derived , which is not the case . It would also risk removing functionally important regulatory regions from the analysis , by choosing only one SNP per LD block . In order to overcome these problems , we sought to identify those regulatory region-associated SNPs within a given region that are most likely to contribute to a given subnetwork of putative phenotype-associated regulatory regions . By the definitions described above , these will be those regulatory regions with the highest NDA score . Regulatory regions are considered for combination if they are separated by 100 , 000bp or less . If any regulatory region within this range has a correlation p -value of less than 0 . 1 with any other regulatory regions in the range , they are combined . A single representative regulatory region is then chosen—the regulatory region with the largest NDA score in the group , derived from a network comprised of all other groups . In order to confirm that spurious coexpression signals are not being generated solely because of LD , we used the ENSEMBL Perl API for the 1000 genomes phase 3 data ( CEU ) to search for variants in LD with each SNP lying within the chosen regulatory region for each group . Variants in LD with a variant in any other chosen regulatory region are reported . A is defined as the set of all nodes in the whole network . Each member of A is a node in an interaction network . For each i ∈ R , Spearman’s rank correlation , x , is calculated with each other node in R . The probability , p , of a correlation as strong as , or stronger than , the index correlation , x , arising by a chance pairing between the index node and any other node ( n ( r>x ) ) is inferred from the empirical distribution of all correlations ( r ) of the index node in A . For every node in the set R , a score s is calculated to summarise the strength of interactions with all other nodes in R . Since the only thing that the elements of R have in common is that they are TSS identified by the set of input SNPs , unexpectedly strong inter-relationships between elements of R are taken as indirect evidence of a relationship between the input SNPs themselves . The NDA score , s , is defined as the sum of –log10 ( p ) values for interaction strength within the matrix . Raw p-values are calculated from the empirical distribution of values of s for 10000 permuted networks . The Benjamini-Hochberg method is used to estimate false discovery rate ( FDR ) . Significant network density scores are taken as those with FDR < 0 . 05 . In order to enable comparison of coexpression scores between different analyses , the raw coexpression score ( s ) is corrected by dividing by the total number of independent groups of regulatory regions included in each analysis , nres , yeilding a corrected coexpression score , ccs: ccs=s/nres The node in the network with the highest NDA score has , by definition , numerous strong correlations with other nodes in the subset R . The NDA scores assigned to these other nodes are therefore inflated by their association with the stongest node . This inflation may reflect biological reality , since both TSS have a putative genetic association with the phenotype of interest , and both share strong links . However , there is a risk that TSS sharing a chance association with a strongly coexpressed TSS will be spuriously inflated to significance . For this reason , we have applied a stringent correction in order to ensure that we have confidence in each significantly coexpressed TSS independently of all TSS with stronger coexpression in the network: the NDA score for each TSS is calculated after removing all TSS with stronger NDA scores from the network . Of 267 , 225 robust promoters and enhancers identified by FANTOM5 , 93 , 558 ( 50 . 6% ) were promoters within 400 bases of the 5′ end of a known transcript model . These were annotated with the name of the transcript . Alternative promoters were named in order of the highest transcriptional activity . Where necessary , coordinates for GWAS SNPs were translated to hg19 coordinates using LiftOver , or coordinates were obtained for SNP IDs from dbSNP version 138 . A circular permutation method was devised to prevent systematic bias by maintaining the underlying structure of GWAS SNP data . The NDA score for a given regulatory region was compared with NDA scores obtained from randomly permuted subsets of genes to give an empirical p-value for coexpression . If permuted networks consist of randomly-selected regulatory regions , then this p-value quantifies coexpression alone; if the permuted networks are generated by mapping randomly-selected SNPs to regulatory regions , then the final p-value is a composite of two measures: coexpression , and the enrichment for true GWAS hits in regulatory sequence . Pre-mapping permutations use a random set of SNPs generated by rotation of the input set of SNPs , K , on a concatenated circular genome . The choice of background is critical—some more recent GWAS studies consider only a subset of variants with a high probability of association with a given trait . In the present analyses , background data were chosen to reflect as accurately as possible the pool of variants included in the original study . For this reason , results are presented only for phenotypes for which the the entire summary dataset was available , including a p-value for every SNP , so that the background used to generate permuted networks is exactly the same background from which the real dataset is drawn . In order to quantify the effect of coexpression alone ( i . e . eliminating the inflation of NDA scores that occurs due to enrichment of trait-associated SNPs in regulatory regions ) , permuted networks were generated after mapping to TSS regions . This is analogous to randomly reassigning the labels in the network , but aims to preserve the local relationships between regulatory regions , since we cannot assume that regulatory regions are randomly distributed on the genome , and since regional regulatory events , such as chromatin reorganisation , are expected to lead to coexpression between nearby regulatory regions . Where A is defined as a list of regulatory regions comprising the whole set of FANTOM5 TSS , post-mapping permutations select a subset of A in a similar circular manner , by displacing the members of the set R by a random number of places on the list . Where the displacement pushes members of R off the end of the list , they are re-entered at the beginning . This process generates a pool of variants that are likely to be grouped in a similar distribution on the genome to the input set . If the input set contains a large group of TSS regions in close proximity to each other on the genome , it is likely that this group of TSS regions will be joined as a single unit ( see above ) for analysis . During generation of permutations , the same number of consecutive TSS regions elsewhere on the genome may not be in sufficient proximity ( and expression correlation ) to be grouped together . This would create extra network nodes , potentially inflating the NDA scores in the permuted sets . To mitigate against this , those TSS from each permutation that do not conform to the input set distribution are re-entered into a further circular permutation until an identical distribution is found . If no matching grouping is found after 8 repeat permutations , additional regulatory regions are added from consecutive positions above and below whichever group is nearest in size to the relevant group in the original input dataset . False discovery rates ( FDR ) are calculated using the Benjamini-Hochberg method . The enrichment for GWAS hits from a pooled resource comprising the NCBI GWAS catalog and the GWASdb database ( observed SNPs per Mb: expected SNPs per Mb ) was quantified at increasing search window sizes upstream and downstream from the transcription start site ( TSS ) . A table of GWAS hits for a broad range of phenotypes was obtained from the NCBI GWAS catalog and from a larger , less selective catalog of GWAS p-values meeting permissive criteria for genome-wide significance , GWASdb . The GWASdb dataset is less curated than the NCBI GWAS catalog , but contains a much greater range of SNPs since it does not restrict inclusion to the strongest associations , or to putative causative variants . Because both databases are limited by the variation in reporting , and quality , of the original GWAS studies from which data are drawn , this analysis was restricted to variants meeting genome-wide significance at a widely-accepted threshold ( p < 5 × 10−8 ) . These catalogues were combined and filtered to remove duplicate entries . Data were obtained from: Overlapping phenotypes , such as “urate” and “uric acid” were manually merged . Phenotypes that were considered to be too broad to be informative were excluded , as were those that were not related to human disease . A complete table of phenotypes in GWASdb and NCBI GWAS catalog , showing mergers and inclusion/exclusion in the present work , is provided in a supplementary file ( S2 Table ) . Strong anti-correlation between pairs of TSS associated with the same phenotype may have biological importance , such as down-regulation at one TSS but expression at another , or negative regulation of a signalling pathway on which expression of a TSS is dependent . For this reason , anti-correlations may improve detection of true associations in this analysis . However , in order to confer an overall improvement on the performance of the algorithm , true inverse expression relationships between phenotype-associated TSS would need to be sufficiently common to overcome the noise added by incorporating all strong anti-correlations into the NDA score . Anti-correlations do not contribute any net improvement to the NDA scores for a training set ( Crohn’s disease , 50% of all SNPs , chosen at random ) , and were therefore excluded . Full GWAS or meta-analysis data , reporting every SNP genotyped or imputed in a given study , are required in order to permute subsets against the appropriate background for a given study . These were obtained from the following sources: In order to better understand the pathophysiological implications of disease variants in regulatory regions , we sought to identify whether these regions exhibit unexpectedly specific expression in any given cell types or tissue samples . In order to reduce noise , technical and biological replicates were averaged for this and subsequent analyses . The full table of samples in FANTOM5 , showing which samples were averaged as technical replicates , and which were excluded , is in S2 Table ( S2 Table ) . For a given trait , we took the subset of regulatory regions for which a significant coexpression pattern was detected for that trait ( coexpression FDR ≤ 0 . 05 ) . For each regulatory region , we created a list of all cell types in which that region was active , ranked by expression level . We then combined the cell type lists for each regulatory region using a robust rank aggregation ( RRA ) . There are several possible sources of bias in this raw measurement . For example , some cell types have more cell-type specific transcriptional activity , perhaps because these cell types fulfil a specialised role; other cell types are particularly well-represented in the FANTOM5 samples . We therefore controlled for the probability that a given cell type would be highly ranked in the initial RRA analysis , by permuting RRA results for at least 100 , 000 random selections of n regulatory regions . We then calculated the empirical p-value for each cell type , i . e . the probability that this cell type would be assigined a raw RRA p-value at least as strong by random chance . We then corrected for multiple comparisons using the Benjamini-Hochberg method to estimate false discovery rate ( FDR ) . Computer code required to run the NDA method , specifically for the detection of coexpression in FANTOM5 regulatory regions , can be obtained from https://github . com/baillielab/coexpression/
Our initial evaluation demonstrated that coexpression is stronger among regulatory regions containing variants with low GWAS p-values ( Fig 2 ) . The coexpression signal obtained for the test input set was evaluated using different subsets of FANTOM5 samples ( cell lines , timecourses following a perturbation in primary cells or selected cell lines , tissue samples , primary cells , or various combinations of these ) , and different types of regulatory region ( enhancers , promoters assigned to annotated genes , other promoters , or all regulatory regions combined ) ( Fig 3 ) . The strongest coexpression is seen in the combined sample set . A “minimal detail” sample set was also tested , comprising a single average value for each of the timecourses , primary cell types and tissue types , and excluding data from unstimulated cell lines . The complete dataset , including all cell types and tissues , provided the strongest signal , demonstrating that there is additional biologically-relevant information contained in the expression profiles from all sample subsets ( Fig 3 ) . The difference between the distributions of NDA scores derived from pre- and post-mapping permutations reveals the different components of the measure . When compared to a random pool of SNPs ( pre-mapping permutations ) , two factors inflate the NDA scores for real GWAS data: firstly , more regulatory regions are identified because true GWAS hits are enriched within regulatory regions; secondly , the coexpression signal itself is greater for real data . In contrast , post-mapping permutations have precisely the same number of regulatory regions included as the real dataset , so there is no component of inflation due to enrichment in regulatory regions . The effects of these different components are shown in Fig 4 , which reveals the NDA score to be a composite measure of both signals . Similar expression profiles are often seen arising from regulatory regions that are close to each other on the same chromosome , which may also span linkage disequilibrium blocks . The effect of this on the coexpression signal was mitigated by grouping nearby ( within 100 , 000bp ) regulatory regions into a single unit , unless they have notably different expression patterns . SNPs in nearby regulatory regions are also more likely to be in linkage disequilibrium , and these regulatory regions themselves are more likely to share cis- ( or short-range trans- ) regulatory signals in common . We checked for significant linkage disequilibrium between regulatory regions assigned to independent groups . At a threshold of r2 > 0 . 8 , there is no linkage disequilibrium between significantly coexpressed groups; three examples of weaker linkage relationships were detected with 0 . 08 ≤ r2 ≤ 0 . 6 ( Supplementary results ) . Regulatory regions around individual TSS with higher coexpression scores contain variants with stronger GWAS p-values ( Fig 5A ) , indicating that this independent signal provides additional information that may be used for fine-mapping causative loci ( Fig 6; Supplementary results ) . Where data are available , we have compared our results to the recent fine mapping study by Huang et al , who use high-resolution genotyping in 67 , 852 subjects with inflammatory bowel disease to quantify the probability that a given variant is causal . A total of 9 variants with a causal probability > 0 . 1 lie within 150 , 000bp of a significantly coexpressed region; of these , 7 lie immediately adjacent to the most significantly coexpressed promoter/enhancer in the region . In order to enable the detection of new regulatory regions with strong coexpression relationships , we chose a permissive threshold at GWAS p < 5 × 10−6 . GWAS data for Crohn’s disease[16] were used for initial optimisation of the NDA approach . Of the 8 GWAS datasets for phenotypes that were not used in algorithm development ( i . e . all apart from Crohn’s disease ) , 6 showed evidence of significant coexpression ( Table 1 ) . Among these , between 17 and 24% of regulatory regions identified as containing a GWAS SNP were found to be significantly coexpressed with other regulatory elements associated with the same phenotype ( FDR < 0 . 05 , compared with 100 permuted subsets of equal size; see Methods ) . Although many coexpressed regulatory regions are not promoters for annotated genes ( supplementary results; Fig 3 ) , we compared the named genes in our results with gene-level burden of significance scores from PASCAL[17] analysis of the original GWAS studies . Since the coexpressed regulatory regions were detected due to the presence of a variant with a low p-value , it is expected that the genes with coexpressed promoters will be highly ranked in a gene-level analysis . However , the weak but significant correlation ( Spearman r = 0 . 30; p = 1 . 9 × 10−5 ) between the approaches provides further evidence that the coexpression signal itself provides additional information which successfully prioritises regulatory regions ( Fig 5B ) . For a given disease , regulatory regions containing GWAS variants are coexpressed if they share similar activity patterns ( i . e . similar expression patterns among transcripts arising from these regulatory regions ) with other regulatory regions implicated in that disease . Fig 7A shows significant coexpression superimposed on a two-dimensional representation of the entire network of pairwise correlations . Since activity ( transcript expression ) was measured in many samples , the true proximity of regulatory regions to one another cannot be accurately represented in two dimensions–a perfect representation would require as many dimensions as there are unique samples . In contrast , the NDA method quantifies proximity of regulatory regions in true network space without artificial dimensionality reduction . Thus significantly coexpressed elements are detected even if they are not directly adjacent on a two-dimensional representation of the network ( Fig 7 ) . We saw no evidence of spurious coexpression due to genomic proximity with shared regulatory influences ( see Methods ) . In each of the GWAS analyses for which significant coexpression was detected , strong coexpression links were seen between loci that were widely separated on the genome ( Fig 4; supplementary results ) . The coexpression signal essentially combines the signal for association in a GWAS with the location and activity pattern of regulatory regions on the genome . We deliberately chose a permissive GWAS p-value threshold in order to enable the detection of new signals that did not achieve genome-wide significance in the original studies . For example , we found that coexpressed transcripts for both LDL and total cholesterol ( TC ) arise from promoters for well-studied genes such as APOB[18] and ABCG5[19] , but also from regulatory regions not previously associated with cholesterol levels . A promoter for SLC22A1 , which encodes an organic cation transporter , OCT1[20] , is strongly coexpressed among elements associated with LDL and TC ( Supplementary results ) . OCT1 transcription is regulated by cholesterol[21] and the transporter regulates hepatic steatosis through its role in thiamine transport[22] . This action of OCT1 is inhibited by metformin[22] , an oral hypoglycaemic agent whose cholesterol-lowering effect[23] is not well understood[24] . Full results of coexpression analyses are in the supplementary results , and online at http://baillielab . net/coexpression . The significantly-coexpressed networks detected here could be regarded as revealing the signature expression profile , at least within the FANTOM5 dataset , for a given disease or trait . We next explored whether these signature expression patterns reveal cell types or biological processes that may contribute to the trait or disease susceptibility . We therefore ranked cell types and tissues by transcriptional activity for each of the significantly-coexpressed loci for each trait , and combined the rankings using a robust rank aggregation[25] . By first detecting the characteristic expression signature associated with a given phenotype using only high-resolution GWAS data , and then detecting the cell type and tissue activity profiles that underlie this signature , we improve on the statistical power of previous methods that have attempted to detect cell-type specific signatures of disease[4 , 6 , 26] . Signals that are strong enough to be detected in previous , less powerful studies are highly significant in our analysis; for example genetic loci associated with cholesterol are transcriptionally active in hepatocytes and liver tissue[6] ( Supplementary results ) .
The FANTOM5 atlas is accessible from http://fantom . gsc . riken . jp/data/ An online service running the coexpression method is available at http://baillielab . net/coexpression Code delivering the NDA coexpression method is available at https://github . com/baillielab/coexpression
|
We discover that genetic variants associated with specific diseases have more in common with each other than we have previously seen . Specifically , variants associated with the same disease tend to be in parts of the genome that are turned on or off in similar complex patterns across many different cell types . We discover that genetic variants associated with specific diseases are found within regulatory elements that share patterns of expression . Specifically , variants associated with the same disease tend to be in parts of the genome that are turned on or off together in similar complex patterns across many different cell types . Knowing this helps us to find new variants associated with some diseases , and to better understand the genetic causes of other diseases . Furthermore , we discover that the genetic causes of inflammatory bowel disease fall into two distinct patterns , indicating that two aetiologically-distinct endotypes of this condition exist . Unlike other methods to learn about disease mechanisms from genetic information , our approach does not require any knowledge or assumptions about the genes themselves–it depends only on the patterns in which parts of the genome are activated in different cell types .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"crohn's",
"disease",
"gene",
"regulation",
"immunology",
"ulcerative",
"colitis",
"colitis",
"cell",
"signaling",
"clinical",
"medicine",
"mathematics",
"gastroenterology",
"and",
"hepatology",
"genome",
"analysis",
"inflammatory",
"bowel",
"disease",
"discrete",
"mathematics",
"combinatorics",
"genomic",
"signal",
"processing",
"transcriptional",
"control",
"gene",
"expression",
"signal",
"transduction",
"permutation",
"cell",
"biology",
"clinical",
"immunology",
"genetics",
"autoimmune",
"diseases",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"genetics",
"of",
"disease",
"computational",
"biology",
"human",
"genetics"
] |
2018
|
Shared activity patterns arising at genetic susceptibility loci reveal underlying genomic and cellular architecture of human disease
|
Functional genomics screens using multi-parametric assays are powerful approaches for identifying genes involved in particular cellular processes . However , they suffer from problems like noise , and often provide little insight into molecular mechanisms . A bottleneck for addressing these issues is the lack of computational methods for the systematic integration of multi-parametric phenotypic datasets with molecular interactions . Here , we present Integrative Multi Profile Analysis of Cellular Traits ( IMPACT ) . The main goal of IMPACT is to identify the most consistent phenotypic profile among interacting genes . This approach utilizes two types of external information: sets of related genes ( IMPACT-sets ) and network information ( IMPACT-modules ) . Based on the notion that interacting genes are more likely to be involved in similar functions than non-interacting genes , this data is used as a prior to inform the filtering of phenotypic profiles that are similar among interacting genes . IMPACT-sets selects the most frequent profile among a set of related genes . IMPACT-modules identifies sub-networks containing genes with similar phenotype profiles . The statistical significance of these selections is subsequently quantified via permutations of the data . IMPACT ( 1 ) handles multiple profiles per gene , ( 2 ) rescues genes with weak phenotypes and ( 3 ) accounts for multiple biases e . g . caused by the network topology . Application to a genome-wide RNAi screen on endocytosis showed that IMPACT improved the recovery of known endocytosis-related genes , decreased off-target effects , and detected consistent phenotypes . Those findings were confirmed by rescreening 468 genes . Additionally we validated an unexpected influence of the IGF-receptor on EGF-endocytosis . IMPACT facilitates the selection of high-quality phenotypic profiles using different types of independent information , thereby supporting the molecular interpretation of functional screens .
Genome-scale functional genetics screens using technologies such as RNA interference ( RNAi ) have recently started to generate high-dimensional datasets by measuring either the same parameter in different cell lines [1] , [2] or different features in the same cell line [3]–[5] . Such high-dimensionality improves the phenotypic specificity but , at the same time , increases the complexity of the analysis: the knock-down of two genes may have a similar phenotype on one parameter but yield different results on another . This poses a substantial challenge for the mechanistic interpretation of such screens [6] , [7] . Furthermore , it has been noticed that targeting the same gene with different siRNAs can lead to conflicting results [3] . This ambiguity is caused by the additive influence of noise in the assay and off-target effects ( OTEs ) . OTEs occur when the detected phenotype is due to interactions between the silencing molecules and genes other than the intended target [8] , [9] . Thus , OTEs complicate the functional interpretation of RNAi screens and may lead to spurious gene annotation . Even though OTEs can be reduced in small-scale studies ( e . g . by gene rescue experiments ) , it is very difficult to completely avoid them in large-scale genomic screens [10] . Consequently , it is often impossible to unambiguously assign the assay readout to a target gene without considering additional information . Note that frequently even replicate measurements using the same siRNA can be inconsistent , which is not necessarily an indication of bad experimental skills , but rather a problem intrinsic to the complexity of genome-wide screens [11] , [12] . Previous work has shown that integrating independent information , such as protein interaction networks with RNAi screening data removes noise and improves the elucidation of molecular mechanisms [7] , [4] , [13]–[18] . These approaches exploit the fact that phenotypes that are observed consistently across a set of interacting genes are less likely to be noise . Hence , interaction data can be used to filter for genes that are more likely true positives . However , existing studies have not sufficiently addressed the problem of high-dimensional phenotypes nor the ambiguity of results from different siRNAs [7] , [13]–[16] . The issue of multiple profiles per gene is relevant for studies performing replicate measurements with the same siRNAs , using different siRNAs per gene , as well as studies conducting functional assays on cells from multiple individuals/different cell lines . Further , published studies often rely on first defining an arbitrary cut-off value for selecting ‘hit genes’ and subsequently interpreting their phenotypes using prior information [15] , [16] , [19] . Such approach is problematic because genes falling just below the threshold may be rejected even though their phenotype is consistent with interacting genes . Instead , it has been suggested to infer sets of relevant genes by first integrating the phenotype data with network information without any threshold and then simultaneously accounting for strength of the phenotype and its consistency in the network [20] . Two classes of such methods exist: methods of the first class assume one phenotype score per gene ( e . g . the strength of the phenotype ) and search for network regions enriched for high-scoring genes [16] , [17] , [21] , [22] . The second class works on multi-dimensional phenotypic profiles , and assesses the similarity of them between genes being close in the network [7] , [23] , [24] . In these cases , multiple measurements are available to describe the loss of function phenotype , such as the number of objects , their average size , the average intensity of a marker protein and so on . We could not find a method integrating multiple phenotype vectors per gene with interaction data . Thus , there is a need for new computational methods allowing for the integration of multi-parametric phenotypic data with molecular interaction information . Here , we present a computational framework called IMPACT ( Integrative Multi Profile Analysis of Cellular Traits ) that integrates high-dimensional , quantitative phenotypic profiles with independent data like protein interactions . We devised two algorithms operating on two different types of prior information: sets of related genes ( IMPACT-sets ) and network information ( IMPACT-modules ) . This framework offers several advantages: first , it can handle multiple phenotypic profiles per gene; second it avoids a priori definition of ‘hit genes’ based on score thresholding; third , it allows to rescue genes that do not have a significant phenotype based on the RNAi data alone , but show a behavior consistent with their interacting partners . Further , it can cope with many potential biases , e . g . caused by the different frequency of phenotype patterns in the screen , by the structure of the network , or due to variable numbers of knock-down experiments per gene . We validated both methods using a multi-parametric genome-wide RNAi screen on endocytosis [3] leading to new insights into the underlying molecular pathways . Implementations of IMPACT-sets and IMPACT-modules as well as the data used in this publication are freely available at http://cellnet . cecad . uni-koeln . de/impact . html . The source code is available at https://github . com/SimeoneMarsico/IMPACT .
We designed a general framework that combines data from quantitative multi-parametric measurements with protein interaction information ( Figure 1 ) . We refer to the set of parameters measured after each knock-down experiment as ‘phenotypic profile’ . Given several profiles from different si-/esi-RNAs targeting the same gene , our aim was to identify the most likely ‘authentic’ profile , i . e . selecting those profiles that are least affected by noise and OTEs . IMPACT exploits that profiles being similar across interacting genes/proteins are more likely true ( Figure S1 ) . For this filtering process , we developed two methods using two types of gene-gene relationships: sets of genes and binary network information ( Figure 1 b ) . We applied our methods to an image-based , genome-wide RNAi screen assessing the role of genes in transferrin ( TF ) and epidermal growth factor ( EGF ) endocytosis in human HeLa cells [3] ( Figure 1a and Input Data in Methods ) . Forty quantitative parameters describing various aspects of cargo uptake and propagation along the endocytic pathway , such as endosome number , size and intracellular distribution , were extracted by image analysis [3] , [29] ( Table S1 ) . On average about 7 si-/esi-RNA per gene were screened . Ideally , one would expect a high correlation between the phenotypic profiles of different siRNAs targeting the same gene . However , those profiles were often not significantly correlated ( Figures S1 & S2 ) . Such inconsistency is neither caused by technical or biological variation in the screen , nor by different silencing potency of the siRNAs [3] , but mainly due to siRNA-specific OTEs [3] , [8] , [30] , [31] . In order to systematically and quantitatively assess the performance of recovering genes involved in endocytosis , we compiled a set of known endocytosis-related genes as positive controls ( Figure 1a ) . This selection is based on relevant Gene Ontology ( GO ) terms and exclusively using experimentally inferred gene annotations ( in total 387 genes annotated for the terms reported in Table S2 ) . The negative control set ( 21 , 585 genes ) was assembled considering genes that are annotated with functions other than endocytosis ( i . e . genes without any annotation were excluded from this analysis ) . We ranked the genes based on the p-values of the protein complexes or network modules they belong to , and tested whether known endocytosis related genes ( i . e . genes from the positive set ) rank higher than the negative set genes . We used Receiver Operator Characteristic ( ROC ) , precision-recall ( PR ) curves , and balanced accuracy ( BACC ) [32] curves ( Figure 2 ) for visualizing to what extent IMPACT distinguishes known endocytosis-related genes from the negative set . We also computed the Area Under the ROC Curve ( AUC , [33] ) to quantitatively compare the overall performances of different search parameters and across different methods . In order to also experimentally validate that our approach improves the phenotype selection , we rescreened 468 genes from the most significant protein complexes and network modules ( Table S5 ) using an improved set of 4 siRNAs per gene ( see Methods ) . The siRNAs used for this rescreen represented a new , independent set of reagents from a different provider , produced with newer technology , which improves the knock-down efficiency , induces less toxicity , and lowers off-target effects [36] . In order to independently confirm the improved quality of the new siRNAs we validated that both , individual parameters as well as phenotypic profiles are more reproducible using the new set of siRNAs ( Figures S3 & S4 ) . Importantly , profiles of different siRNAs targeting the same gene are more similar in the rescreen compared to the primary screen . Therefore , the new profiles are expected to be closer to the true phenotype . The profiles selected by IMPACT are thought to be closer to the true phenotype of the genes than the rejected ones and , thus , should also be more similar to the rescreen data . Indeed , we observed that the pairwise correlation of the selected profiles to the new profiles is significantly higher than the correlation between rejected and new profiles ( Figure 3 ) . Furthermore , the reference profiles ( i . e . , the median of selected profiles per set or network module ) are even more similar to the rescreen data than the selected profiles ( Figure 3 ) . Although being significant , the improvement is not dramatic: this is partly due to the fact that the phenotypic data for the new set of oligonucleotides are better but still noisy ( Figure S3 and S4 ) . To confirm this notion , we selected a few strong examples where the set of new oligo profiles show high intra-similarity within the rescreen ( suggesting low noise ) . The similarity of the profiles selected by IMPACT is much higher to this new set than to the old ones for the same gene . Among those , we had some genes important for endocytosis ( PDPK1 , Furin , MLC1 ) and for signaling ( ERBB2 , IGF1R ) ( Figure S19 ) . These data demonstrate that our analysis successfully selected profiles that are more reproducible in the rescreen and likely better reflect the true function of the genes . Furthermore , the reference profiles , representing the consensus phenotype of a protein complex or network module , were even less affected by noise . In order to visualize the phenotypes of the analyzed complexes and network modules we created a ‘phenotype map’ representing the strength and specificity of the phenotype for transferrin or EGF ( Figures 4 & 5 ) . This visualization groups phenotypically related complexes and network modules and it also shows simplified representations of the profiles , thus , facilitating the interpretation of the findings . Even though the analysis above already showed that our method improved phenotype selection , we also verified the validity of our results by focusing on proteins and protein complexes with known functions related to endocytosis . Our analysis rescued several genes that did not score in the initial analysis [3] , like RAB4A , SARA ( ZFYVE9 ) , APPL1 , RAB11FIP1 , VPS28 , VAMP8 , VIT1A , STX2 and SNX1 ( see Table S12 for full list of 91 endocytic genes selected by IMPACT and missed in the previous analysis ) . Genes selected by IMPACT were enriched also for other endocytosis-related functional terms from the KEGG and GO annotations ( DAVID analysis , [37] , [38] ) , such as endocytosis ( p = 1 . 9e-3 , modified Fisher's Exact Test ) , phosphatidylinositol signaling system ( p = 1 . 3e-14 ) and inositol phosphate metabolism ( p = 1 . 4e-8 ) for KEGG; membrane enclosed lumen ( p = 1 . 7e-26 ) and membrane bounded vesicle ( p = 1 . 5e-5 ) for cellular compartment ( GO CC ) ; membrane fusion ( p = 1 . 8e-3 ) , invagination ( p = 6 . 5e-2 ) and docking ( p = 9e-2 ) for GO biological processes ( GO BP ) . Moreover , our method selected expected phenotypes for several known cellular machineries . The AP2 complex , for instance , is known to be primarily involved in transferrin endocytosis [39] . Even though phenotypic profiles of individual AP2 subunits were ambiguous , our method correctly identified the transferrin-specific phenotype as being enriched in this complex ( Figure 4 and S1 ) . The integrative analysis allowed us to reveal subtle phenotypic differences between closely related machineries . Two examples are the families of SNARE and ESCRT complexes ( Figure 4 and S13 ) . The reference profiles extracted for those complexes through our method again suggest possible insights into molecular mechanisms , therefore posing the basis for focused experimental testing ( Text S1 ) . Internalization and trafficking of signaling molecules such as membrane receptors is crucial for many signaling pathways . Whereas the importance of endocytosis for signaling is well established , much less is known about how signaling pathways control endocytosis [40] , [41] . The network analysis allowed us to gain insights into this process by identifying several signaling pathways over-represented in statistically significant network modules , such as the ErbB and Insulin signaling pathway , the focal adhesion and actin pathway and pathways involved in diseases , particularly cancer ( Table S11 ) . Also , our analysis further elucidated how the position of a protein in a pathway relates to its phenotype . For example , we detected two transforming growth factor beta ( TGF-beta ) related network modules with distinct phenotypic profiles ( “Activins” and “SMADs-Notch” , Figure 5 ) . Consistent with the fact that Activins and SMADs act in the same pathway , our algorithm assigned related phenotypes to them , both showing a reduction of transferrin and EGF uptake , as already reported [3] with a stronger impact on transferrin than EGF ( Figure 5 ) . However , our analysis also uncovered significant differences between these two parts of the TGF-beta pathway . The first module contains several Activin receptors ( ACVR1B , ACVR2B , ACVR2A , ACVR1 and AXVRL1 ) that are known to modulate and transform signals for the TGF-beta superfamily of ligands . The second module links the TGF-beta and Notch pathways [42] . This SMADs-Notch module has a core consisting of SMAD2 , SMAD3 and NOTCH1 , which in turn are associated with several transcriptional regulators ( Figure 5 ) . SMAD3 and NOTCH1 were missed in the initial screen hit list and have been rescued by the integrative analysis . Knock-down of genes in both , the Activins and SMADs-Notch sub-networks , significantly reduced the number of endosomes ( G1 ) , underlining the importance of these pathways for endocytosis . However , knock-down of the Activin module reduces cargo uptake ( G2 ) , whereas knock-down of the SMADs-Notch module increases cargo uptake for transferrin endosomes . The difference between the Activins and SMADs-Notch modules underlines that upstream and downstream components of the same signaling pathway ( i . e . the TGF-β pathway in this case ) can have different effects on endocytosis . Thus , the position of proteins in the pathway seems to critically affect the impact on the assay's readout . We evaluated the general applicability of IMPACT in three different ways: first , we applied IMPACT-modules to the same RNAi screening data , but using a different network as a prior . Second , we ran it on another siRNA screen with autophagy as an endpoint [51] and finally , we used IMPACT to analyze a CRISPR-Cas9 knockout screen in human cells [52] ( see respective paragraphs in Text S2 ) . We run IMPACT-modules for the endocytosis screen on different interaction networks derived from the STRING database [53] . STRING incorporates diverse types of information , such as co-expression , experimentally validated protein binding , and text mining , to predict the functional relationships between genes . Importantly , it can be used to evaluate the importance of these individual feature types for the phenotype prediction . This analysis revealed that the choice of the network strongly affects the quality of the phenotype prediction . Specifically , we noticed that 1 ) the performance deteriorates when considering co-expression data only ( AUC = 0 . 505 ) ; 2 ) experimentally validated interaction networks yield better classification ( AUC = 0 . 6483 for the HPRD-Intact-KEGG combined network and 0 . 603 for STRING experimental ) than networks allowing also non-experimental interactions such as database and text mining predictions ( AUC = 0 . 553 ) . See paragraph Other sources of prior information , Text S2 . Thus , this analysis confirmed that using high-quality , experimentally confirmed protein interaction data maximally reduced noise from the RNAi data . Importantly , both experimental networks ( our combined and STRING-experimental ) gave results that were better than random . Next , we run IMPACT-modules on a siRNA autophagy screen in the human HEK293 cell line [51] where 3 replicates of the entire screen were acquired and 3 different image-based parameters were measured . We analyzed the recovery of the known autophagy genes reported in the human autophagy database ( www . autophagy . lu , [54] ) . The original screen analysis identified 25 known genes ( out of the 175 autophagy genes screened ) among the 1'000 reported hits ( enrichment p-value = 0 . 04 ) ; IMPACT-modules identified 1'332 significant genes , of which 46 were autophagy annotated ( out of the 161 mapping on the network; enrichment p-value = 2e-3 ) . Also , IMPACT performed better than the ranking measure considered in the screen for classifying the known autophagy genes ( AUC = 0 . 563 , p = 2e-4 for IMPACT; AUC = 0 . 4949 , p = 0 . 59 for hit ranking ) . See paragraph Analysis of the siRNA autophagy screen in Text S2 . Finally , we run IMPACT-modules on a CRISPR-Cas9 knockout screen in the human melanoma cell line A375 [52] , where the authors investigated the effect of gene loss upon treatment with vemurafenib , a therapeutic drug inhibitor of BRAF , by measuring cell viability in 4 different conditions ( vehicle versus drug , 7 and 14 days ) . IMPACT identified 1'659 significant genes ( p<0 . 05 ) . Gene enrichment analysis of over-represent GO biological processes and KEGG pathways ( DAVID [37] , [38] ) revealed interesting insights into the mechanism of action of the drug . Pathways involved in cancer and related to BRAF activity , such as “Melanoma” , “MAPK” , “Pathways in cancer” were strongly enriched ( fold enrichment of 2 . 35 , 1 . 84 , 2 . 06; p-value of 3e-7 , 8e-11 , 2e-20 respectively ) . Also , biological processes related to phosphorylation , kinase activity , cell migration , cell proliferation and cell death were strongly enriched ( p-values ranging from 1e-12 to 1e-6 ) . The screen hit list derived using RIGER [55] identified overall GO and KEGG terms with higher p-values and lower fold enrichment ( i . e . less significant ) , related mainly to “Oxidative phosphorylation” ( p = 1e-4 ) , transcription ( 2e-4 ) and histone modification ( 1 . 2e-3 ) . See paragraph Analysis of the CRISPR-Cas9 knockout screen in Text S2 . Thus , we conclude that IMPACT improves the analysis of functional genomics screens beyond RNAi screens .
We chose to use only protein interaction data to support mechanistic interpretation of the phenotypes in terms of molecular machineries . Co-expression and co-functionality data have a broader coverage of the genome , however they do not necessarily imply molecular interactions and would therefore not satisfy the purpose . Instead of using IMPACT-sets , one may also transform protein complexes into interaction networks by considering all pairwise interactions instead of protein sets . In this case , one could use IMPACT-modules to perform the analysis . We tested this possibility ( IMPACT-modules on sets ) and calculated the classification performance ( AUC ) in identifying known endocytic genes . Running IMPACT-modules on sets still performs better than the other approaches considered in this study , but its performance is worse compared to IMPACT-sets ( Table S6 ) . Two reasons may explain this phenomenon: first , the statistical analysis of IMPACT-sets may be better suited for the analysis of sets . Second , the conversion of sets into networks is questionable . Thus , considering set information , when available , rather than splitting it into binary interactions can be advantageous . The complexity of our approach arises in part from the fact that we considered all phenotypic profiles separately and selected the profiles with an enriched pattern among other genes in the set or interacting genes in the network . A simpler approach would have been to combine all profiles of a gene first ( e . g . averaging ) and then assessing the consistency in the networks . However , this method would not take into account that different siRNAs often produce very different profiles due to their heterogeneous off-target signatures . In fact , sometimes less than half of the siRNAs yield profiles resembling the correct phenotype ( Figure S18 ) and averaging would result in profiles that are more strongly affected by OTEs and noise . We indeed observed that applying IMPACT after averaging either decreases performances ( Figures S6b and S7b; Table S4 ) or reduces the number of selected genes ( Table S5 ) , with a stronger effect on network module identification . The classification analysis showed improvements relative to other approaches , both considering ( JAM , MATISSE ) and ignoring ( Chi-square ) prior information . However , we were hoping for an even better performance: using IMPACT the AUCs never exceeded 0 . 65 for endocytosis GO terms and 0 . 67 for the Rab5 effectors and proteins with endocytic domains . Our analysis showed that noise in both , the RNAi screening data and the network can significantly compromise the performance . However , using a network-prior improves the signal-to-noise ratio , which is even more important when data is noisy . Also , reducing network coverage led to lower performances , suggesting that IMPACT or related methods relying on prior information can further improve as new protein-protein interactions will be discovered . Importantly , whichever method is used , the quality of the results is always limited by the quality of the input data . Thus , all method assessment should be regarded as relative comparisons of alternative approaches . One concern when comparing methods relying on prior information ( such as IMPACT , JAM , Matisse ) to methods relying on phenotypic data only is that the performances of these methods may be inflated by the fact that interacting genes tend to share the same annotation used for evaluation , which may result in circular reasoning . This problem cannot be completely solved since equally annotated , connected genes could be at the core of molecular machineries , which makes it difficult to distinguish a circularity bias from the reality of the underlying biology . We have tried to address this problem in two ways: first , we evaluated the enrichment of not just any GO term among top-scoring genes , but terms specific for endocytosis . A network-based method running on random data may still identify sub-networks or protein complexes that are enriched for certain functions , but not necessarily for the functions relevant for the screen . Second , we have performed extensive new experimental validation , which is independent of any reported annotation in public databases . Application of our two methods to an endocytosis RNAi screen led to improved recovery of known endocytosis-related genes compared to the analysis of the primary screen data alone . Further , based on several performance measures , our network-based method performs better than published methods that are either mono-parametric or do not assume multiple profiles per gene . Importantly , using only one ( averaged ) profile per gene also reduces the performance of our method , which demonstrates that considering multiple profiles is crucial . The importance to assess sets and modules based on individual-siRNA profiles was also confirmed by the rescreen , which showed that the selected profiles were significantly better correlated with the new independent data than the rejected profiles . Moreover , the partial coverage that prior information based approaches can offer ( in our case , 9 , 715 out of the 17 , 730 genes in the screen were mapped on the interaction network ) suggests as a sensible solution the use of IMPACT as a complement rather than an alternative to other methods relying only on phenotypic information , to combine the strengths of both , i . e . identifying protein machineries and elucidating their function with IMPACT , while still assessing the effect of genes without interaction information . The computation of the significance of network modules controls for potential biases , such as the topology of the sub-networks , that were ignored in previous work [16] , [21] , [23] , [56] . To address this , our approach takes into consideration the number of neighbors ( degree ) and the number of profiles ( siRNAs ) of each gene , as well as the frequency of the enriched phenotypic profile across all genes in the network . Not considering these factors may lead to inflated or deflated significance estimates . Note that global permutation testing [16] , [56] would neither account for the specificities of a given phenotypic profile nor for the local topology of the network . Whereas our network search is based on a heuristic greedy search , other methods [57] , [58] provide exact solutions using constraint programming for module search . However , these approaches can only deal with a single score per gene; further development will be necessary to devise similar strategies for multi-parametric measurements and multiple profiles per gene . Recent work already suggested feedback regulation of signaling pathways onto endocytosis [40] , [59][3] . The findings in this study ( Table S11 ) underline the tight bonds between the endocytic machinery and signaling networks . Additionally , our network analysis revealed that different modules within the same signaling pathway can exert diverse effects on trafficking ( e . g . Activins- and SMADs-containing modules within the TGF-beta pathway ) , whereas components of different signaling pathways ( e . g . TGF-beta and Notch ) can have similar effects on endocytosis . Thus , there is no trivial relationship between a gene's membership in a signaling pathway and effects on cellular machineries . Our computational analysis suggested that IGFR might impact specifically on EGFR trafficking , an observation that has not been described so far . This effect might be mediated by direct binding between activated IGFR and EGFR [44] or by the IGFR signaling pathway . Our experiments confirmed that IGF stimulation induced faster endosomal accumulation of EGF , probably by accelerating early endosome fusion . The redistribution of transferrin was not surprising , since the two cargos extensively colocalize in endosomes at early time points [60] . However , IGF-1 stimulation affected specifically EGF trafficking kinetics by inducing both faster uptake and decay ( consistent with degradation ) without affecting the overall uptake and recycling kinetics of transferrin . Importantly , since these effects were relatively weak , IGFR was not detected as a hit gene in the initial analysis [3] . Only the integrated analysis exploiting the network context revealed its effect on EGF endocytosis . Other mechanisms of crosstalk among IGFR and EGFR involving receptor cross-activation and heterodimerization [44] or cross-transcriptional regulation [43] have previously been proposed . Our findings extend those reports by uncovering novel aspects of the integration of these signaling pathways at the level of the trafficking system . The application of the screen to other sources of prior information and other phenotypic data revealed important aspects of the presented method . First , the choice of prior information can significantly affect the quality of the results . Molecular interaction data ( as opposed to functional relationships such as co-expression ) helped best to reduce noise and improve the mechanistic interpretation of the results . For the CRISPR screen application , it is difficult to compare the performance of our method to the RIGER screen analysis due to the absence of a positive set . However , we have shown that our method can be successfully applied to diverse kinds of data and can lead to interesting hypotheses to further explore . IMPACT did not reveal modules of genes involved in drug resistance ( i . e . increasing cell viability ) , because they either lack network context or the phenotypic effect is not conserved among interactors . However , it succeeded in identifying molecular machineries responding to the drug ( i . e . decreasing cell viability ) , which could have potential therapeutic applications for the design of co-inhibitors of other genes in the pathway to overcome drug resistance in melanoma . Despite the aforementioned advantages , there are limitations to this approach . Our network covers less than half of the human protein coding genes . Genes outside the network are ‘inaccessible’ to our analysis . We anticipate that future projects will use other information ( such as predicted protein-protein interactions ) and also improve the statistical framework . This work therefore just represents the beginning of a gradually more integrated analysis of high-dimensional functional screens in conjunction with network data .
Both IMPACT-sets and IMPACT-modules work on a high-dimensional dataset generated by functional screens ( e . g . knock-down , knock-out , gene editing screen ) . Here , the phenotype of each gene g is measured m times , as for instance after knock-down with m different oligonucleotides . ( Alternative scenarios are for instance targeting the same gene in different individuals or cell lines . ) For each single knock-down experiment , the phenotype is described quantitatively by the phenotypic profile p , which is an N-dimensional vector where each element is a parameter measured . The parameters measured ( and thus also the dimensionality of p ) must be the same for all genes , whereas the number of measurements per gene ( m ) can vary between genes . The phenotype information D is therefore represented as a PxN matrix , where rows represent different perturbations ( e . g . siRNAs ) and columns the different parameters . N is the number of parameters and P is the total number of phenotypic profiles for all genes , with , where mi is the number of different perturbation experiments ( e . g . oligonucleotide knock-downs ) for the ith gene and Ng is the total number of genes screened . Importantly , both methods can work with single phenotypic profiles per gene ( mi = 1 ) as well as with multiple oligo profiles per gene ( mi> = 1 ) . Genes in the same data set can have different number of profiles . The screen data D needs to be normalized ( e . g . z-score or similar ) , so that parameters have similar impact on the similarity metric during the method search . We considered as phenotypic data for our analysis a high-dimensional image-based RNAi screen performed in human HeLa cells [3] . This screen aimed to characterize the loss of function phenotype of each gene involved in the endocytosis of two cargo molecules , transferrin ( TF ) and the epidermal growth factor ( EGF ) . To this purpose , 40 parameters ( Table S1 ) were quantitatively measured to assess the effect of multiple oligonucleotide knock-downs per gene , with an average of about 7 different si-/esi-RNA reagents per gene . The prior information we used to guide the method search is detailed in the paragraphs “Protein Complexes” and “Protein-Protein Interaction Network” . The gene-set-based approach tests for the enrichment of a set of related genes for a specific phenotypic profile . The set of genes can be defined based on functional relationships ( e . g . pathway co-membership ) or physical association ( protein complexes ) . The algorithm consists of two main steps: First , the algorithm identifies a ‘common’ phenotypic pattern that is shared by a maximum number of genes in the set . Then , the statistical assessment is performed via randomizations: for each set with an enriched profile , we generate random sets of the same size and with the same number of profiles per gene in each set . The resulting empirical distribution is used for computing p-values . For the network-based analysis of the phenotypic data , we have implemented a greedy search algorithm ( Figure S17 ) . This search method may operate on any network representing genes as nodes and any kind of relationship between genes as ( undirected ) edges . The method can be applied to phenotypic data where either multiple profiles are available for each gene or when there is a single profile per gene . The algorithm works in three main steps: Protein complexes were taken from CORUM [25] , which contains 2 , 083 experimentally verified mammalian protein complexes ( Table S3 ) of which 1930 had phenotype data from our RNAi screen . Orthologous complexes from non-human species were mapped using the ENSEMBL orthology information . We assembled an interaction network combining experimentally validated protein-protein interactions from three public sources: HPRD [26] in vivo interactions ( interactions that are validated in in vivo assays ) , IntAct [27] and physical protein interactions from KEGG [28] . After removing genes without phenotype information the combined network contains 9 , 642 nodes and a total of 49 , 827 interactions . ROC ( Receiver Operating Characteristic ) curves report the true positive rate ( TPR , y-axis ) as a function of the false positive rate ( FPR , x-axis ) . , where TP is the number of true positives and P is the total number of positive , i . e . the sum of true positives plus false negatives ( FN ) . , where FP is the number of false positives and N is the total number of negatives , i . e . the sum of false positives and true negatives ( TN ) . Sensitivity is a synonym for TPR; specificity is . Precision recall ( PR ) curves report precision ( y-axis ) versus recall ( x-axis ) . Precision is defined as ( see definition above ) . Recall is another name for true positive rate ( TPR , see above ) . Balanced accuracy is defined as . In presence of unequal sized classes and different classification performance on positive or negative sets , the balanced accuracy is a better measure than accuracy [32] . In case of balanced classes it reduces to conventional accuracy . The balanced accuracy curve shows the balanced accuracy value ( y-axis ) as a function of the p-value threshold ( x-axis ) . The Area Under the Curve ( AUC ) is the integral under the ROC curve , calculated by the trapezoidal numerical approximation method . The standard error ( sem ) was estimated as reported in [33] . To test if the AUC is significantly better than the random case ( i . e . , AUC = 0 . 5 ) , we performed the z-test on the quantity , as in [61] . The statistical assessment of the comparison between two AUCs ( Table S10 ) was performed through stratified bootstrapping ( N = 1000 ) by calculating the quantity , where A1 and A2 are the two AUCs and A1_r and A2_r are the bootstrapped AUC values , described in [62] . A subset of genes ( n = 468 ) selected with the integrative analysis have been rescreened with 4 new , independent siRNAs ( Stealth Select RNAi from Invitrogen ) and compared to the primary screen data . All the genes considered for the analysis belong to statistically significant modules and complexes . We used exactly the same cell line and conditions as in the primary screen [3] and we assayed the knock-downs in the same way . For each gene , four groups of profiles have been considered: ( 1 ) the module reference profile , ( 2 ) the oligo profiles selected by our method , ( 3 ) the oligo profiles excluded by our method , and ( 4 ) the new oligo profiles in the rescreen . We computed the distribution of the pairwise Pearson correlations between each one of the groups ( 1 ) - ( 2 ) - ( 3 ) and group ( 4 ) and compared the three cumulative distributions of the correlation values . Two different non-parametric tests , the Kolmogorov-Smirnov and the Mann-Whitney U test , were used to assess the statistical significance of the differences between pairs of distributions . HeLa cells were grown in DMEM supplemented with 10% FCS and 24 h prior to the experiment cells were plated in 96 well plates to reach approximately 80% confluence on the day of the experiment . Each experimental condition was repeated twice in the plate layout , and each experiment was repeated 4 and 5 times for the co-pulse-chase and the co-pulse , respectively .
|
Genome-scale functional genomics screens are important tools for investigating the function of genes . Technological progress allows for the simultaneous measurement of multiple parameters quantifying the response of cells to gene perturbations such as RNA interference . Such multi-dimensional screens provide rich data , but there is a lack of computational methods for interpreting these complex measurements . We have developed two computational methods that combine the data from multi-dimensional functional genomics screens with protein interaction information . These methods search for phenotype patterns that are consistent among interacting genes . Thereby , we could reduce the noise in the data and facilitate the mechanistic interpretation of the findings . The performance of the methods was demonstrated through application to a genome-wide screen studying endocytosis . Subsequent experimental validation demonstrated the improved detection of phenotypic profiles through the use of protein interaction data . Our analysis revealed unexpected roles of specific network modules and protein complexes with respect to endocytosis . Detailed follow-up experiments investigating the dynamics of endocytosis uncovered crosstalk between the cancer-related EGF and IGF pathways with so far unknown effects on endocytosis and cargo trafficking .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"systems",
"biology",
"signal",
"transduction",
"functional",
"genomics",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology"
] |
2014
|
Revealing Molecular Mechanisms by Integrating High-Dimensional Functional Screens with Protein Interaction Data
|
Due to the risk of severe vaccine-associated adverse events , yellow fever vaccination in Brazil is only recommended in areas considered at risk for disease . From September 2008 through June 2009 , two outbreaks of yellow fever in previously unvaccinated populations resulted in 21 confirmed cases with 9 deaths ( case-fatality , 43% ) in the southern state of Rio Grande do Sul and 28 cases with 11 deaths ( 39% ) in Sao Paulo state . Epizootic deaths of non-human primates were reported before and during the outbreak . Over 5 . 5 million doses of yellow fever vaccine were administered in the two most affected states . Vaccine-associated adverse events were associated with six deaths due to acute viscerotropic disease ( 0 . 8 deaths per million doses administered ) and 45 cases of acute neurotropic disease ( 5 . 6 per million doses administered ) . Yellow fever vaccine recommendations were revised to include areas in Brazil previously not considered at risk for yellow fever .
Yellow fever is an acute viral hemorrhagic disease transmitted by mosquitoes . Severity ranges from self-limited febrile illness to hemorrhagic syndrome with jaundice , multiple organ failure and death; severe cases are more likely to be detected and reported to passive surveillance systems [1] , [2] . Yellow fever is considered endemic in tropical regions of Africa and South America . For endemic countries , the World Health Organization recommends vaccination of persons living in areas at-risk for yellow fever , as well as for epidemic control [3] . According to the 2010 revised yellow fever risk map , Brazil is one of 11 South American countries with endemic or transitional areas for yellow fever , and one of seven in which vaccine is recommended in only part of the territory [4] . In Brazil , yellow fever virus transmission is maintained in tropical forests in a sylvatic cycle first described in the 1930s [5] , involving non-human primates and several species of tree-dwelling mosquitoes . Since several New World monkeys species develop fatal disease following yellow fever viral infection [2] , sudden die-offs of non-human primates may signal yellow fever virus circulation and a potential exposure risk to humans . Surveillance for epizootic disease is recommended by the Pan American Health Organization [6] , and is conducted in several Brazilian states as an early warning system for viral circulation to inform preventive vaccination [7] . In Brazil , sporadic human cases may also occur as a result of recreational or occupational exposures to jungle areas [8] , [9] . Yellow fever vaccine developed from attenuated viral strains has been used in Brazil since 1939 [10] . Yellow fever vaccine recommendations must weigh the risk of exposure to yellow fever virus against the rare occurrence of fatal adverse events among vaccinated individuals [11] , [12] . The Brazilian Ministry of Health recommends vaccination against yellow fever for persons who reside in or visit areas where transmission of yellow fever virus occurs [13] . However , yellow fever virus may also be transmitted from human to human by Aedes mosquitoes , resulting in urban epidemics . Although Brazil successfully controlled urban transmission in the 1940s through vector control and vaccination [14] , re-establishment of Aedes aegypti in urban areas has resulted in recurrent epidemics of dengue fever and poses a risk for outbreaks of urban yellow fever [15]–[17] . Brazil's national yellow fever control strategy seeks to prevent human disease by identifying areas where the virus circulates , as well preventing re-introduction of urban epidemics through mass vaccination [11] . Since 1999 , yellow fever has re-emerged in parts of Brazil that had been silent for several decades [9] , [18]–[21] , challenging prevention strategies and resulting in frequent revision of yellow fever vaccine recommendations [13] . Beginning in 2008 , following yellow fever outbreaks in central and southeastern Brazil , northeastern Argentina and Paraguay [4] , the Ministry of Health initiated enhanced surveillance for human and epizootic yellow fever viral activity during the usual seasonal period from October to June [8] . We describe two yellow fever outbreaks that occurred during the first year of enhanced yellow fever surveillance in Brazil , during the 2008–2009 season .
This study involved analysis of routinely collected surveillance data and did not require ethical review according to the Brazilian National Committee for Ethics in Research . Personally identifiable information ( patient name and information included on case report form ) was available only to surveillance officers and was not used in this study . Yellow fever is a notifiable disease in Brazil . The national passive surveillance system of the Brazilian Ministry of Health receives reports of suspected cases of yellow fever and epizootic events from state and municipal health departments . Beginning in 2008 , the Ministry of Health took several actions to increase sensitivity of surveillance and timeliness of response vaccination during the seasonal period of highest risk of yellow fever from October to June . Enhanced surveillance includes raising awareness of yellow fever among health workers , mandatory notification of persons with ictero-hemorrhagic syndromes , investigation of human deaths due to unknown causes , intensification of epizootic surveillance in non-human primates and immediate communication of investigation findings . The national yellow fever surveillance system coordinates epidemiological surveillance for human cases and epizootic events , as well as communication between health departments and public health laboratories and [13] . Suspected cases are defined as individuals presenting with fever accompanied by jaundice or hemorrhagic symptoms , who within the previous 15 days were exposed to areas considered at-risk of yellow fever or with evidence of yellow fever virus circulation . Suspected cases in persons who received their first yellow fever vaccination within 10 days prior to symptom onset are classified for surveillance as unvaccinated . Persons presenting with signs and symptoms of yellow fever ( jaundice , abnormal laboratory values , hemorrhage or neurologic symptoms ) who experienced symptom onset within 60 days of receipt of yellow fever vaccine identified through enhanced surveillance for ictero-hemorrhagic syndromes were investigated following a clinical and laboratory protocol as possible adverse reactions to vaccination [22] . Adverse events were classified as confirmed , probable , suspected , discarded or inconclusive , according to U . S . Centers for Disease Control and Prevention criteria [23] . Clinical , epidemiological and laboratory data , as well as vaccination history , are reviewed by expert committee that provides recommendation for final classification as previously described [22] . Laboratory confirmation for human cases includes the presence of yellow fever virus-specific antibodies detected by IgM-capture ELISA [24] or immunohistochemistry [25] , [26] , detection of yellow fever virus by reverse-transcriptase PCR [27] or isolation of yellow fever virus in cell culture [28] . Nucleotide sequencing of yellow fever virus is used to differentiate between wild-type viral infections and vaccine virus [27] , [29] . An epizootic is defined as a sudden die-off of non-human primates in a small geographic area . Sightings of sick and dying monkeys or carcasses are reported to local health departments , which conduct investigations and collect specimens for laboratory testing . Epizootics of yellow fever are confirmed if: 1 ) yellow fever virus is detected by immunohistochemistry or reverse-transcriptase PCR or isolated from animal specimens [7] , [20] , [30] , or 2 ) if multiple non-human primate deaths are clustered within a short time period or linked by similar hydrographic features or vegetation to neighboring areas with documented circulation of yellow fever virus . A yellow fever outbreak is defined as two or more confirmed human cases with a common probable location of infection . In accordance with International Health Regulations , evidence of yellow fever virus circulation in an area in which the resident population has not been vaccinated against yellow fever is classified by the Brazilian Ministry of Health as a Public Health Emergency of National Importance , requiring immediate evaluation of the risk of viral dissemination and need for intervention [31] , [32] . Recommendations for vaccination against yellow fever virus may be extended to “affected areas” ( municipalities with confirmed yellow fever human cases or epizootics , or detection of yellow fever virus in mosquito vectors ) and bordering municipalities . During yellow fever outbreaks , vaccination may be recommended beginning at 6 months of age; yellow fever is not recommended in infants younger than 6 months of age [3] , [23] . Surveillance data were tabulated in TabWin ( version 3 . 2 , Datasus , Brazilian Ministry of Health , Rio de Janeiro , Brazil ) and maps were created in TerraView ( version 3 . 2 . 1 , INPE , São José dos Campos , Brazil ) . Analyses were conducted in EpiInfo ( version 6 . 04d , Centers for Disease Control and Prevention , Atlanta , USA ) . Incidence of yellow fever was calculated as the number of confirmed cases divided by the total state population , stratified according to areas with or without yellow fever vaccination recommendations prior to the 2008–2009 outbreaks . Rates of adverse events were calculated per million doses administered in each state during the period .
During the epidemic period from 28 September 2008 through 6 June 2009 , 270 suspected yellow fever cases were reported to the Brazilian Ministry of Health . Of the suspected cases , 50 cases were classified as confirmed yellow fever based on laboratory criteria ( n = 46 ) or epidemiologic linkage ( n = 4 ) . In the southernmost state of Rio Grande do Sul , there were a total of 21 confirmed cases ( 2 . 1 cases per million residents ) with 9 deaths ( case-fatality , 43% ) , including 3 cases ( 5 . 0 per million residents ) in areas in which yellow fever vaccine was recommended and 18 cases ( 1 . 9 per million residents ) in areas without vaccine recommendation prior to the outbreak ( Figure 1a ) . In the most populous state of São Paulo , there were 28 confirmed cases ( 2 . 7 per million residents ) with 11 deaths ( case-fatality , 39% ) , all in areas without vaccine recommendations prior to the outbreak ( Figure 1c ) . During the same period , only 1 ( non-fatal ) confirmed case ( 0 . 3 cases per million residents ) resulting from sporadic exposure was reported from Mato Grosso state in central Brazil . Among other suspected cases , 196 ( 89 . 1% ) were negative for antibodies to yellow fever and were classified as non-yellow fever cases and another 24 ( 10 . 9% ) suspected cases had no epidemiological link with areas of transmission of yellow fever or had evidence of disease due to other causes that included dengue , leptospirosis , hantavirus infection and sepsis . Among confirmed case patients , 35 ( 70% ) were male; median age was 31 years ( range , 3 days to 73 years ) . Characteristic symptoms of jaundice and hemorrhage were recorded on case report forms for only 17 ( 34% ) and 18 ( 36% ) confirmed cases , respectively , although transaminase levels were markedly elevated ( Table 1 ) . 37 ( 74% ) were hospitalized for more than 24 hours and overall case-fatality was 40% . Age and gender distribution was similar among confirmed cases in the two most affected states . One case patient with laboratory-confirmed yellow fever had received yellow fever vaccine 9 years earlier; none of the other case patients was considered effectively vaccinated . However , three individuals had received vaccine 1 to 2 days prior to experiencing symptom onset and were investigated as possible vaccine-associated adverse events; wild-type yellow fever infection was confirmed in two case patients by nucleotide sequencing while wild-type infection was considered most likely in the third case patient based on epidemiologic linkage to a laboratory-confirmed case . Four additional case patients received yellow fever vaccine after the onset of symptoms; identification of yellow fever virus confirmed wild-type infection in three case patients while the fourth case patient was classified as confirmed based on exposure to areas with yellow fever virus circulation prior to symptom onset . A total of 46 ( 92% ) confirmed cases occurred in areas in which vaccination against yellow fever had not been recommended prior to the 2008/2009 season ( Figure 2 ) . Probable locations of exposure for all confirmed cases were forested or rural areas , except for one case of perinatal exposure in an 8-day old infant [33] . The majority of cases were clustered in a small number of municipalities , including five in São Paulo state ( Piraju [n = 11] , Sarutaiá [n = 7] , Buri [n = 5] , Avaré [n = 4] and Tejupa [n = 1] ) and nine in Rio Grande do Sul ( Santa Cruz do Sul [n = 7] , Vera Cruz [n = 4] , St . Angelo [n = 3] , Pirapó [n = 2] , Bossoroca [n = 1] , Espumoso [n = 1] , Jóia [n = 1] , Ijuí [n = 1] , and Augusto Pestana [n = 1] ) . Of 14 municipalities in the two states identified as probable location of exposure of confirmed human cases , 12 were not previously considered at-risk for yellow fever and vaccination of residents had not been recommended prior to the epidemic . Yellow fever vaccination was recommended prior to the outbreak in two municipalities in northwestern Rio Grande do Sul following detection of yellow fever virus circulation in 2002 [30] . Epizootic activity involving deaths of non-human primates preceded human cases in both states that experienced yellow fever outbreaks ( Figure 1 ) . In Rio Grande do Sul , epizootic surveillance registered 950 reports of deaths among non-human primates during the period of enhanced surveillance; 947 [99%] involving howler monkeys of the genus Alouatta . Yellow fever virus circulation was confirmed in 173 ( 67% ) of 259 events with samples available for testing , and 7 were confirmed by epidemiological linkage ( Figure 1b ) . In Sao Paulo , 125 epizootics were reported and totaled 146 dead animals . Of these , 67 were of the genus Alouatta ( 45 . 9% ) , 56 Callithrix ( 38 . 3% ) , 14 Cebus ( 9 . 6% ) and 9 were not identified ( 6 . 2% ) . Biological specimens were collected for testing from 64 ( 44% ) of 146 animal carcasses , including tissue specimens from 58 animals and blood specimens from 23; the presence of yellow fever virus was detected in specimens from 2 separate epizootics ( Figure 1d ) . Although 238 epizootic events were reported from other states , only 1 from Paraná state was confirmed as yellow fever . In Rio Grande do Sul , epizootics were reported from all but one municipality identified as probable locations of exposure of confirmed human cases , and circulation of yellow fever virus among non-human primates was detected in the state 9 weeks before the occurrence of human cases ( Figures 1a and 1b ) . In São Paulo , despite reports of epizootic activity throughout the epidemic period , yellow fever virus circulation among non-human primates was not confirmed until late March , 2009 , following occurrence of human cases in late February 2009 ( Figures 1c and 1d ) . In all , circulation of yellow fever virus was documented in 78 municipalities: 71 ( 91% ) with confirmed epizootic activity ( including 7 with both human and animal disease ) and only 7 ( 9% ) with confirmed human cases without confirmed epizootics . Prior to the outbreak in Rio Grande do Sul , yellow fever vaccination had been recommended in 59 municipalities in the northwest corner of the state ( 595 , 346 inhabitants ) ; during and following the outbreak , vaccination was extended to 462 municipalities ( 93% of all municipalities in the state , with 9 , 963 , 267 residents ) , leaving only a small region along the Atlantic coast without vaccine recommendation . Following extension of yellow fever vaccine recommendations to affected areas , 3 , 636 , 722 doses of yellow fever vaccine were administered in Rio Grande do Sul ( Figure 1a ) , reaching approximately 39% vaccination coverage in previously unvaccinated populations . In São Paulo state , the number of municipalities with yellow fever vaccine recommendations increased from 332 ( with 7 , 584 , 215 residents ) to 452 ( with 10 , 469 , 327 residents ) , covering 70% of the state and approaching metropolitan São Paulo . A total of 1 , 869 , 960 vaccine doses were administered in previously unvaccinated areas ( Figure 1c ) , reaching 64% of the resident population . In all , 5 , 506 , 682 million doses ( 69% of all doses administered in Brazil during the period ) were administered in areas in which vaccination against yellow fever had not been recommended before the epidemic . In both states , numbers of doses administered peaked after the confirmation of human cases ( Figure 1 ) , although municipalities were included in the area with yellow fever vaccine recommendation at different times as the epidemic spread to new areas . During the same period , Brazil's national immunization program received 97 reports of severe adverse events among individuals who received yellow fever vaccine . Of these , 51 were classified by the national vaccine safety committee as associated with yellow-fever vaccine , including 6 cases of acute viscerotropic disease ( incidence , 0 . 8 cases per million doses administered in Brazil ) , all of which were fatal , and 45 cases of acute neurologic disease ( 5 . 6 per million doses administered in Brazil ) with no deaths . Of these , 40 cases ( 89% ) were classified as aseptic meningitis; three cases ( 7% ) were reported as encephalitis , one case was diagnosed as meningoencephalitis and one case as a right peripheral facial paralysis . However , two of the patients who had meningoencephalitis and encephalitis developed neurological sequelae that included difficulty walking and decreased visual acuity , respectively . Among the 6 confirmed cases of yellow fever vaccine-associated acute viscerotropic disease , median age was 31 years ( range 4–44 years ) and 2 ( 33% ) case patients were male . Rates of vaccine-associated viscerotropic disease in the three states that reported cases ranged from 0 . 4 to 1 . 6 cases per million doses administered ( Table 2 ) . Among the 45 confirmed cases of yellow fever vaccine-associated acute neurologic disease , median age was 21 years ( range 22 days - 66 years ) with 26 ( 58% ) cases among males . Rates of neurologic disease in the three states ranged from 0 . 8 per million doses administered in Santa Catarina to 11 . 0 per million in Rio Grande do Sul ( Table 2 ) . Two events were classified as vaccine-associated neurologic disease resulting from secondary transmission to breastfed infants [34] , [35] .
These two yellow fever outbreaks in unvaccinated populations in the Brazilian states of Rio Grande do Sul and São Paulo during the 2008–2009 epidemic season challenged control strategies and resulted in revised vaccination guidelines for Brazil . Despite the relatively small number of confirmed cases and deaths , both outbreaks occurred in geographic areas without yellow fever vaccination recommendations , in which circulation of yellow fever virus had not been identified for four decades [9] . Although control vaccination was rapidly implemented following identification of human cases and epizootic events , the virus spread more quickly than expected and human cases continued to occur in newly affected areas . The experience in Rio Grande do Sul demonstrates the importance of active surveillance for yellow fever epizootics among non-human primates to inform vaccine recommendations [7] . Mass vaccination in previously unvaccinated populations may have prevented additional cases , but most vaccination occurred after the peak of the outbreaks . Yellow fever vaccine was associated with six deaths and multiple severe vaccine-related adverse events . Additional strategies are needed to prevent yellow fever outbreaks in unvaccinated populations until safer vaccines are available . Mass vaccination has been associated with increased detection of adverse events since the first description of yellow fever vaccine-associated viscerotropic disease during intensified yellow fever vaccination in Brazil [22] , [36] , [37] . Vaccination of adults without prior immunity may increase rates of severe events , since risk is greatest with first vaccination and appears to increase with age [38] , [39] . In addition , although detection rates of viscerotropic disease in Brazil are lower than estimated incidence based on United States surveillance data ( 0 . 4 per 100 , 000 doses administered ) [39] , improved surveillance during mass vaccination likely contributes to increased detection of adverse events [22] . The 2008–2009 outbreaks were the first in Brazil to identify rates of vaccine-associated neurotropic disease similar to those reported from adverse events surveillance in the United States ( 0 . 8 per 100 , 000 doses administered ) [39] , suggesting that enhanced surveillance and laboratory testing ( specifically , real-time PCR for detection of yellow fever virus RNA in cerebrospinal fluid specimens ) improved detection of neurologic events . Brazilian authorities have proposed universal childhood immunization against yellow fever to decrease risk of severe vaccine-associated adverse events later in life [11] , based on the lower risk of viscerotropic disease in young children . However , due to uncertainty about the true risk , yellow fever vaccine is included in routine childhood immunizations only in Brazilian municipalities where vaccine is recommended for the entire population [40] . Factors associated with emergence of yellow fever are still poorly understood . It is unclear why yellow fever re-emerged in two non-endemic areas more than 1000 kilometers apart with limited evidence of viral circulation in other parts of Brazil . There had been widespread evidence of virus circulation in central Brazil during the preceding epidemic season ( October 2007—June 2008 ) , with confirmed human cases in 8 states and reports of epizootics in 14 states [8] . Circulation of yellow fever virus had been confirmed in northern São Paulo state and the western part of Paraná state ( between Rio Grande do Sul and São Paulo ) in early 2008 . The role of climatic events ( increased temperatures and high rainfall ) , high densities of susceptible non-human primate hosts and human exposure to mosquito vectors in forested areas may all have contributed [9] . Conditions may have favored viral spread to populations of susceptible non-human primate hosts during the interepidemic period , seeding outbreaks in the two previously unaffected areas . Better understanding of factors affecting viral spread and the dynamics of viral transmission would help to focus preventive immunization in unvaccinated populations . As a result of the 2008–2009 outbreaks , Brazil's yellow fever risk map was revised in 2010 to include large areas in which the population had not previously been vaccinated ( Figure 3 ) . This followed a change in 2008 by the Brazilian Ministry of Health to simplify classification of municipalities as those in which yellow fever vaccination is recommended or those without recommendation [13] . The 2008 revision harmonized state and federal recommendations , and the 2010 revision added vaccine recommendations in 334 municipalities in four states , with an estimated population of 8 . 6 million residents . In 2013 , WHO revised yellow fever vaccine recommendations based on evidence that a single dose of YF vaccine confers life-long immunity against YF disease in most vaccine recipients , suggesting that booster doses may not be necessary [3] . However , cases of yellow fever have infrequently been documented in vaccinated individuals [3] , as observed in one individual during the 2008–2009 outbreaks [13] . The Brazilian Ministry of Health continues to recommend re-vaccination against yellow fever every 10 years in areas considered at-risk for yellow fever , although priority is given to primary vaccination of previously unvaccinated persons in these areas . Vaccine recommendations may be revised as data become available from ongoing studies on the duration of immunity provided by yellow fever vaccine in adults and children . During the past decade , from 2000 to 2010 , the majority of human cases have been associated with exposures outside the Amazon River basin as yellow fever re-emerged in previously silent areas and unvaccinated individuals entered natural environments where yellow fever viruses circulate . As a result of changes in the epidemiology of yellow fever , the Brazilian Ministry of Health adopted new surveillance strategies , including enhanced monitoring of yellow fever virus activity during the epidemic season [13] . During the 2008–2009 epidemic season , enhanced surveillance contributed to improved laboratory diagnosis of suspected cases , detection of epizootic activity in affected areas and identification of vaccine-associated adverse events , especially neurologic disease . Earlier detection and treatment or more sensitive surveillance may be associated with lower case-fatality observed during 2008–2009 ( 40% ) compared to the 2007–2008 season , with 49 confirmed cases and 28 ( 57% ) deaths ( Brazilian Ministry of Health , unpublished data ) . However , epizootic surveillance can only be effective in preventing human cases if there is adequate time to vaccinate or alert the population at-risk to avoid exposure of susceptible individuals . While active surveillance informed preventive vaccination in Rio Grande do Sul , epizootic events near São Paulo were not reported until after the outbreak had occurred . Additional strategies prioritized by the Ministry of Health included monitoring vaccination coverage in areas with yellow fever vaccine recommendations , syndromic surveillance for febrile icterohemorrhagic diseases for early case detection and surveillance for adverse events following yellow fever vaccination . To prevent outbreaks as well as sporadic cases , public health authorities need to intensify efforts to ensure that individuals at highest risk of exposure are vaccinated . Unvaccinated individuals traveling to areas where vaccination is recommended should be vaccinated at least 10 days prior to travel . Public education is needed about the risk of disease and indications for vaccination , including contraindications and precautions for persons who might be at increased risk of severe adverse events . Efforts should also continue to develop a safer vaccine [41] .
|
Yellow fever is a viral hemorrhagic disease transmitted by mosquitos , endemic in tropical regions of Africa and South America . Large urban outbreaks of yellow fever have been eliminated in the Americas , where most yellow fever cases result from human exposure to jungle or forested environments . Vaccination is effective but carries a risk of potentially fatal adverse events in a small number of vaccinees . In a large country such as Brazil , vaccination is recommended only in areas where there is a risk of exposure to yellow fever virus . We describe two outbreaks of yellow fever in areas without yellow fever vaccine recommendations . Numerous epizootics , or die-offs of non-human primates , were reported from areas with human cases . In response to the outbreaks and epizootic activity , over five million doses of vaccine were administered in previously unvaccinated populations , resulting in vaccine associated adverse events , six of which were fatal . The outbreaks resulted in expansion of areas with yellow fever vaccine recommendations , and highlight the need for safer yellow fever vaccines .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"yellow",
"fever",
"immunizations",
"medicine",
"infectious",
"diseases",
"veterinary",
"diseases",
"veterinary",
"virology",
"zoonotic",
"diseases",
"viral",
"diseases",
"public",
"health",
"veterinary",
"science"
] |
2014
|
Yellow Fever Outbreaks in Unvaccinated Populations, Brazil, 2008–2009
|
Nucleos ( t ) ide analog therapy blocks DNA synthesis by the hepatitis B virus ( HBV ) reverse transcriptase and can control the infection , but treatment is life-long and has high costs and unpredictable long-term side effects . The profound suppression of HBV by the nucleos ( t ) ide analogs and their ability to cure some patients indicates that they can push HBV to the brink of extinction . Consequently , more patients could be cured by suppressing HBV replication further using a new drug in combination with the nucleos ( t ) ide analogs . The HBV ribonuclease H ( RNAseH ) is a logical drug target because it is the second of only two viral enzymes that are essential for viral replication , but it has not been exploited , primarily because it is very difficult to produce active enzyme . To address this difficulty , we expressed HBV genotype D and H RNAseHs in E . coli and enriched the enzymes by nickel-affinity chromatography . HBV RNAseH activity in the enriched lysates was characterized in preparation for drug screening . Twenty-one candidate HBV RNAseH inhibitors were identified using chemical structure-activity analyses based on inhibitors of the HIV RNAseH and integrase . Twelve anti-RNAseH and anti-integrase compounds inhibited the HBV RNAseH at 10 µM , the best compounds had low micromolar IC50 values against the RNAseH , and one compound inhibited HBV replication in tissue culture at 10 µM . Recombinant HBV genotype D RNAseH was more sensitive to inhibition than genotype H . This study demonstrates that recombinant HBV RNAseH suitable for low-throughput antiviral drug screening has been produced . The high percentage of compounds developed against the HIV RNAseH and integrase that were active against the HBV RNAseH indicates that the extensive drug design efforts against these HIV enzymes can guide anti-HBV RNAseH drug discovery . Finally , differential inhibition of HBV genotype D and H RNAseHs indicates that viral genetic variability will be a factor during drug development .
Hepatitis B virus ( HBV ) is a hepatotropic DNA virus that replicates by reverse transcription [1] . It chronically infects >350 million people world-wide and kills up to 1 . 2 million patients annually by inducing liver failure and liver cancer [2]–[4] . Reverse transcription is catalyzed by a virally-encoded polymerase that has two enzymatic activities: a DNA polymerase that synthesizes new DNA and a ribonuclease H ( RNAseH ) that destroys the viral RNA after it has been copied into DNA [1] , [5] . Both activities are essential for viral replication . HBV infections are treated with interferon α or one of five nucleos ( t ) ide analogs [6]–[8] . Interferon α leads to sustained clinical improvement in 20–30% of patients , but the infection is very rarely cleared [1] , [3] , [9] . The nucleos ( t ) ide analogs are used more frequently than interferon . They inhibit DNA synthesis and suppress viral replication by 4–5 log10 in up to 70–90% patients , often to below the standard clinical detection limit of 300–400 copies/ml [10]–[12] . However , treatment eradicates the infection as measured by loss of the viral surface antigen ( HBsAg ) from the serum in only 3–6% of patients even after years of therapy [10]–[13] . Antiviral resistance was a major problem with the earlier nucleos ( t ) ide analogs , but resistance to the newer drugs entecavir and tenofovir is very low [6] , [14] , [15] . This has converted hepatitis B from a steadily worsening disease into a controllable condition for most individuals [16] . The cost of this control is indefinite administration of the drugs ( probably life-long; [7] ) , with ongoing expenses of $400–600/month [17] , [18] and unpredictable adverse effects associated with decades-long exposure to the drugs . The key form of the HBV genome in cells that must be eliminated to clear the infection is the nuclear episomal covalently-closed circular DNA ( cccDNA ) that is the template for transcription of all HBV RNAs [19] . Following reverse transcription in the cytoplasm , newly synthesized genomes can either be enveloped and secreted from the cell as virions , or they can be transported into the nucleus to replenish the cccDNA pool ( Fig . 1 ) [19] , [20] . Transfer of newly synthesized viral genomes into the nucleus via “recycling” is the default pathway , and virion secretion occurs only if the cccDNA pool is large enough to support adequate synthesis of the HBsAgs . The cccDNA pool is very stable , but nucleos ( t ) ide therapy can suppress cccDNA levels in the liver by ∼1 log10 after 1–2 years [21]–[23] . The indefinite persistence of the cccDNA even in patients whose HBV titres in serum have been suppressed below the limit of clinical detection by the nucleos ( t ) ide analogs is due to residual viral replication , leading to replenishment of the cccDNA pool by a combination of intracellular recycling and low-level infection of new cells [24] , [25] . The sequential accumulation of resistance mutations during nucleos ( t ) ide therapy confirms that cccDNA maintenance by residual viral replication occurs in the absence of clinically detectable viremia [15] , [26] , [27] . A recent genetic analysis of HBV DNA in the liver explicitly demonstrated that low levels of cccDNA replenishment occurs even when nucleos ( t ) ide analog therapy has reduced viral titres below the clinical detection limit [24] . RNAseH enzymes hydrolyze RNA in an RNA:DNA heteroduplex [28] . They belong to the nucleotidyl transferase superfamily whose members share a similar protein fold and presumably have similar enzymatic mechanisms [29] . This family includes E . coli RNAseH I and II [30] , DNA transposases including the Tn5 transposase [31] , retroviral integrases including the HIV integrase [32] , the RuvC Holliday junction resolvase [33] , the Argonaute RNAse [34] , and human RNAseH 1 and 2 [35] , [36] . The canonical RNAseH structure contains about 100 aa including four conserved carboxylates ( the “DEDD” motif ) that coordinate two divalent cations [37] . The RNAseH mechanism is believed to involve both divalent cations [29] , [38] , [39] , although a one-ion mechanism has also been proposed [40] , [41] . The HBV RNAseH domain shares low but recognizable ( ∼20% ) sequence identity with the RNAseH domains of reverse transcriptases and other retro-elements [42]–[44] . Manually optimizing alignment of the HBV RNAseH and the HIV-1 RNAseH yielded 23% identity and 33% similarity ( Fig . 2 ) . A similar alignment between the HBV RNAseH and the HIV integrase revealed 19% identity and 33% similarity . The HBV RNAseH is encoded at the carboxy-terminus of the viral polymerase protein that also encodes the viral DNA polymerase activity ( reverse transcriptase ) . The high hydrophobicity of the HBV polymerase and its existence as a complex with host chaperones [45] have severely restricted study of the HBV RNAseH . Furthermore , we demonstrated that the RNAseH in its native context within the polymerase protein is unable to accept exogenous heteroduplex substrates [46] , analogous to the inability of the DNA polymerase active site to engage exogenous primer-templates [47] . Consequently , most of our limited knowledge of the RNAseH comes from mutational studies of the viral genome in the context of viral replication conducted by us and others [48]–[53] . These restrictions have prevented biochemical characterization of the RNAseH and blocked biochemical screens for anti-HBV RNAseH drugs to date . A few reports of recombinant forms of the hepadnaviral RNAseH exist . Wei and co-workers [54] expressed the HBV RNAseH domain in E . coli and purified it by denaturing nickel-affinity chromatography . Following refolding , they found an RNAse activity . Lee et al . [55] expressed the HBV RNAseH domain in E . coli as a dual maltose-binding protein/hexahistidine fusion and purified soluble protein by two-step affinity chromatography; this enzyme had RNAseH activity . Choi and co-workers [56] expressed the intact duck hepatitis B virus polymerase in yeast and reported that it had a weak RNAse activity . Finally , Potenza et al . [57] expressed the HBV RNAseH domain as a synthetic gene in E . coli . Following purification from inclusion bodies and refolding , this enzyme had RNAse activity . However , no follow-up reports have appeared with any of these systems , possibly due to the technical difficulties associated with the purification protocols and/or contamination challenges with host RNAseH or other RNAse classes . Human Immunodeficiency Virus ( HIV ) reverse transcription also requires a virally encoded RNAseH activity [58] , and consequently the RNAseH has attracted much attention as a potential drug target [38] , [59]–[78] . Over 100 anti-HIV RNAseH compounds have been reported , typically with inhibitory concentration-50% ( IC50 ) values in the low µM range . Most of the compounds inhibit HIV replication in culture , typically with effective concentration-50% ( EC50 ) values that are ∼10-fold higher than the biochemical IC50 values . These compounds are often modestly cytotoxic , leading to therapeutic indices ( TI ) that are usually <10 . Second-generation inhibitors with substantially improved efficacy have been reported , but their TI values were not necessarily improved markedly [68]–[70] . Despite these limitations , compounds with efficacy and TI values appropriate for a drug exist [68] , [75] . Most of the compounds inhibit the RNAseH by binding to the enzyme and chelating the divalent cations in the active site [64] , [65] , [69] , [70] , [73] , [76] , but compounds that appear to inhibit the RNAseH by altering the enzyme's conformation or its interaction with nucleic acids have also been reported [63] , [75] . As predicted from their common membership in the nucleotidyl transferase superfamily , some anti-HIV RNAseH compounds can inhibit the HIV integrase , and some anti-integrase compounds can inhibit the RNAseH [59] , [68] , [70] , [74] , [79] . The ability of the nucleos ( t ) ide analog drugs to profoundly suppress HBV in most patients and to cure HBV infection in a few patients indicates that they can push the virus to the brink of elimination . This presents an opportunity to cure many more patients by suppressing HBV replication further , but achieving a cure will require novel drugs against targets other than the DNA polymerase active site . These drugs would be used in combination with the nucleos ( t ) ide analogs to suppress viral replication below the level needed to maintain the cccDNA . A logical target would be the second of HBV's two enzymatic activities , the RNAseH . Here , we report production of enzymatically active recombinant HBV RNAseH suitable for low throughput antiviral drug screening . Using this novel reagent , we demonstrated that the HIV RNAseH and integrase are similar enough to the HBV RNAseH to allow information derived from HIV RNAseH and integrase inhibitors to guide identification of anti-HBV RNAseH compounds .
The HBV DEDD residues have been implicated to be D702 , E731 , D750 , and D790 ( numbering for HBV strain adw2 ) by sequence alignments against other RNAseHs ( Fig . 2 ) , but only D750 has been experimentally confirmed to be essential for RNAseH activity [48] . Therefore , we introduced D702A , E731A , D750V , and D790A mutations into the predicted DEDD motif residue in an HBV genomic expression vector . The wild-type and mutant genomes were transfected into Huh7 cells , five days later intracellular viral capsids were purified , and then HBV DNAs within the particles were detected by Southern analysis . All four mutants supported DNA synthesis and hence could be analyzed by this approach . The signature of an RNAseH-deficient enzyme is production of RNA:DNA heteroduplexes that migrate like double-stranded DNAs on native gels but as faster-migrating single-stranded DNAs of multiple lengths following digestion of the capsid-derived nucleic acids with exogenous RNAseH . DNAs produced by the wild-type genome were unaffected by treatment with RNAseH prior to electrophoresis ( Fig . 3 ) . Mutating each of the four predicted RNAseH DEDD residues blocked production of the slowest-migrating double stranded forms ( mature relaxed-circular DNAs ) and led to accumulation of smaller forms that migrated similar to the less-mature relaxed-circular DNAs produced by the wild-type genome . Treatment of the nucleic acids from the mutant genomes with exogenous RNAseH collapsed the double-stranded forms to single-stranded forms ( Fig . 3 ) . Therefore , all four mutants were RNAseH deficient . We expressed HBV RNAseH sequences from the HBV isolate employed by Potenza et al . [57] in E . coli as a carboxy-terminally hexahistidine tagged recombinant protein , but we moved the amino terminus nine residues upstream to residue 684 of the HBV polymerase because we felt this site was more probable to yield soluble protein ( HRHPL; Fig . 4A ) . As a negative control , we mutated two of the DEDD active site residues ( D702A and E731A ) . These constructs were expressed in E . coli , soluble lysates were prepared , and the lysates were subjected to nickel-affinity chromatography . Five proteins of approximately 80 , 70 , 26 , 14 , and 11 kDa detectable by Coomassie staining were recovered following chromatography , none of which correlated with the predicted mass of 18 . 9 kDa for HRHPL ( Fig . 4B ) . Mass spectrometry identified the dominant 26 kDa band as the E . coli prolyl isomerase SlyD . Concentrating the samples seven-fold did not increase the RNAseH to levels detectable by Coomassie staining . Western analysis with anti-polyhistidine antibodies revealed a large number of cellular bands but failed to unambiguously identify HRHPL . This was presumably due to the presence of histidine-rich regions in the bacterial proteins that promoted their binding to the nickel-affinity resin ( e . g . , SlyD ) . However , western analysis with the anti-HBV RNAseH domain antibody 9F9 ( [80]; Santa Cruz Biotechnology ) revealed a small amount of recombinant HBV RNAseH that migrated close to its predicted mass plus a larger amount of the protein that migrated as a doublet near 15 kDa ( Fig . 4B ) . The doublet is presumably due to proteolysis near the protein's N-terminus because the antibody epitope and hexahistidine tag are at the C-terminus . The sizes of the truncation products imply that they were cleaved near HRHPL residue 36 , which would remove the essential D702 carboxylate ( HRHPL residue 20 ) and inactivate the protein . These experiments indicate we could express and enrich small but detectable amounts of soluble recombinant HBV RNAseH . We tested activity of the recombinant HBV RNAseHs in a DNA oligonucleotide-directed RNA cleavage assay . In this assay , a DNA oligonucleotide is annealed to a uniformly-labeled RNA to create an RNA:DNA heteroduplex . Cleavage of the RNA in the heteroduplex yields two RNA fragments of predictable size that are resolved by electrophoresis and detected by autoradiography ( Fig . 5A ) . We employed the 264 nt RNA ( DRF+ ) used in our previous RNAseH assays [46] in combination with two DNA oligonucleotide pairs . One oligonucleotide in each pair was the correct polarity to anneal to the DRF+ RNA and the other was its inverse complement as a negative control . Oligonucleotide-directed RNAseH assays were conducted with wild-type HRHPL enzyme and the RNAseH-deficient D702A mutant . The RNA was not cleaved when the non-complementary oligonucleotides were employed in the reactions ( Fig . 5B ) , demonstrating that the enzyme preparations did not contain non-specific RNAse activity . Use of complementary oligonucleotide #1 ( D2507− ) led to complete cleavage of the DRF+ RNA by E . coli RNAseH into products of 154 and 94 nt , and to partial cleavage of the RNA at the same site by wild-type HRHPL ( Fig . 5B ) . The large majority of this RNAseH activity was due to the HBV enzyme because mutating DEDD residues D702A and/or E731A sharply reduced cleavage of the RNA . Note that although the relative yield of full-length mutant RNAseH was less than the wild-type enzyme in Fig . 4 , in other preparations the amount of mutant RNAseH exceeded the amount of wild-type enzyme ( e . g . , Fig . 6 ) . In all cases , the enzymatic activity associated with the mutant RNAseH preparations was far lower than in the wild-type preparations . The residual cleavage products in reactions with the mutant enzymes appear to be non-specific breakdown products from the RNA substrate and/or digestion products from trace contamination with bacterial RNAseH . The RNA products shifted sizes as expected when complementary oligonucleotide #2 ( D2543M-Sal , which anneals 33 nt closer to the 3′ end of the RNA ) was employed in the RNAseH assays ( Fig . 5B ) : the larger fragment became larger ( 187 nt ) and the smaller fragment became smaller ( 61 nt ) . These data demonstrate that the RNAse activity in HRHP is specific for RNA annealed to the DNA oligonucleotides , and hence confirm that it is an RNAseH activity . Finally , we synthesized a quenched fluorescent RNA:DNA chimeric hairpin oligonucleotide substrate ( RHF1 ) to confirm RNAseH activity with a different assay . RHF1 has fluorescein at its 5′ end , 20 nt of RNA , a 4 nt DNA hairpin , 20 nt of DNA complementary to the RNA , and an Iowa Black FQ quencher at the 3′ terminus . The hairpin brings the fluorescein and quencher into close proximity , and digesting the RNA frees the fluorescein and increases its fluorescence ( Fig . 5C ) . RHF1 was terminally digested with E . coli RNAseH , the reactions were terminated with 10 mM EDTA , and fluorescence was measured . This digestion amplified the fluorescence of RHF1 22-fold , indicating a 95% quenching efficiency . RHF1 was then employed in an RNAseH assay with buffer alone , wild type HBV RNAseH ( HRHPL ) , and HRHPL-D702A/E731A . RNAseH activity for HRHPL was about 2-fold higher than the no-enzyme control , and mutating the RNAseH active site eliminated this activity ( Fig . 5D ) . This weak signal ( 7% of the maximal signal in this assay ) appears to be due to poor binding between the small substrate and the RNAseH in the relatively high ionic strength of the reactions because detection of RNAseH activity required reducing the NaCl concentration from 190 to 130 mM . These data indicate that we can readily detect HBV RNAseH activity in the enriched bacterial extracts despite the fact that the HBV RNAseH is a minor component of the mixture . The optimal enzymatic conditions for the HRHPL HBV RNAseH were determined by systematically varying the reaction components in the oligonucleotide-directed RNAseH assay ( Table 1 ) . Recombinant HBV RNAseH was active over a wide range of pH values but was most active near 8 . 0 . Its activity maximum was at 190 mM NaCl and it became able to digest single-stranded RNA below ∼100 mM NaCl . The RNAseH required ∼5 mM Mg++ for maximal activity; increasing Mg++ beyond ∼7 mM suppressed RNAseH activity , and inclusion of Mn++ in the reactions led to nonspecific degradation of single-stranded RNA . The enzyme became inactive at low reductant concentrations , but it could tolerate up to 2% DMSO . It was stable upon storage in liquid nitrogen , and only marginal loss of activity was observed following five sequential freeze-thaw cycles . HBV has eight genotypes ( A–H , plus provisional identification of genotypes I and J ) that differ by >8% at the sequence level [81] . We cloned HBV RNAseH domains for genotype A , B , C , and H isolates using the same structure as the HRHPL construct ( genotype D ) to determine whether HBV's genetic diversity leads to variable sensitivity to inhibitors that must be taken into account during drug development ( Fig . 6A ) . The protein profile detectable by Coomassie staining following expression and nickel-affinity enrichment for all additional constructs was the same as for HRHPL . Western blotting with antibody 9F9 detected the genotype B , C , and D RNAseHs , with the genotype C enzyme appearing primarily as the full-length protein ( Fig . 6B ) . The failure to detect the genotype A and H RNAseHs was due either to lack of accumulation of the proteins or to amino acid variations in the C-terminus of the protein where the antibody epitope is located [80] . The genotype A , B , C , D , and H RNAseH extracts were assessed with the oligonucleotide-directed RNAseH assay ( Fig . 6C ) . The genotype A and B enzymes were inactive , activity of the genotype C RNAseH ranged from undetectable to modest in replicate experiments , and activity of the genotype H enzyme was similar to that of the genotype D RNAseH . The [NaCl]- , [Mg++]- , temperature- , and pH-profiles of the genotype H RNAseH were very similar to those of the genotype D enzyme ( data not shown ) . Therefore , we can express recombinant HBV genotype B , C , D , and H RNAseH proteins that are detectable by enzymatic assays and/or western blotting , but only the genotype D and H proteins are consistently active . We hypothesized that the HBV RNAseH may be inhibited by antagonists of the HIV RNAseH based on the similarity of the reactions they catalyze . We identified 10 compounds known to inhibit the HIV RNAseH or that were predicted by chemical structure-activity relationships to do so ( Table 2 and Supplementary Fig . S1 ) . We further hypothesized that anti-HIV integrase compounds may inhibit the HBV RNAseH because the integrase and RNAseH are both members of the nucleotidyl transferase superfamily and because some anti-HIV RNAseH and integrase compounds can cross-inhibit their target enzymes [59] , [68] , [70] , [74] , [79] . Consequently , we also obtained 11 compounds either known to inhibit the HIV integrase or predicted to do so by chemical structure-activity relationships ( Table 2 and Supplementary Fig . S1 ) . We first measured the effect of irrelevant compounds ( tryptophan , sucrose , and IPTG ) on the RNAseH assay . These compounds reduced RNAseH activity of HRHPL to 52±9% relative to the DMSO vehicle control ( Figs . 7 and 8A ) . This allowed us to define the mean of the residual activity in the presence of the irrelevant compounds minus two standard deviations of the irrelevant controls as a threshold reduction of the RNAseH activity that must be exceeded before we considered inhibition by the test compounds to be relevant . Using this threshold , 12 of the 21 compounds inhibited the HBV genotype D RNAseH at 10 µM ( Fig . 7 , Table 2 , and Supplementary Table S1 ) . These 21 compounds were also screened against the HBV genotype H RNAseH using the oligonucleotide-directed RNAseH assay . The unexpectedly high frequency of inhibition of the genotype D enzyme led us to question the mechanism ( s ) by which it was inhibited by the compounds . We addressed this in two manners . First , RNAseH inhibitors usually block the HIV enzyme by interfering with the divalent cations in the active site [64] , [65] , [69] , [70] , [73] , [76] . Consequently , we asked whether the compounds act non-specifically by chelating Mg++ . Isothermal calorimetry demonstrated that compounds #5 , 6 , and 8 did not bind Mg++ in the absence of the protein extracts ( data not shown ) . This is consistent with their inability to significantly inhibit poly-G synthesis by the Hepatitis C virus ( HCV ) RNA polymerase which is also active in 5 mM Mg++ [82] ( Fig . 8B ) . Second , we titrated selected compounds from 50 to 0 . 5 µM to examine dose-responsiveness of inhibition ( Fig . 8C ) . Compound #12 had a typical inhibition curve with an IC50 of 2 . 5 µM in this experiment; similar smooth dose-response curves were observed for compounds #39 and 40 ( data not shown ) . In contrast , inhibition by compound #6 plateaued at 20–30% between 3 and 40 µM but then increased to 75% at 50 µM . Compound #8 was ineffective below 5 µM , it inhibited the enzyme by 40–85% between 10 and 30 µM , and caused aberrant migration of the RNA at 40 and 50 µM . These data indicate that some compounds behaved as predicted from their mechanism of action against HIV , but that inhibition by other compounds may have been due to alternative effects , possibly including interaction with the RNA and/or aggregation of the enzyme . A likely cause of cellular toxicity for anti-HBV RNAseH drugs would be inhibition of human RNAseH1 because it is responsible for about 80% of the RNAseH activity in human cells [83] , [84] . Therefore , we cloned the human RNAseH1 with an N-terminal hexahistidine tag , expressed it in E . coli , and enriched the protein by nickel affinity chromatography . The same spectrum of contaminating E . coli proteins as was observed for the other RNAseH preparations was detectable by Coomassie staining , but RNAseH1 could be detected at its predicted mass of 32 kDa ( Fig . 9A ) . This enzyme was active in the oligonucleotide-directed and fluorescent RNAseH assays ( Fig . 9B and data not shown ) . To determine how inhibition of human RNAseH1 compared to inhibition of the HBV RNAseH , we titrated RNAaseH1 to yield similar levels of activity as the HBV enzyme , and then we directly compared the ability of compounds #8-12 to inhibit human RNAseH1 and HRHPL at 10 µM . All five compounds inhibited the HBV RNAseH . Compound #8 inhibited RNAseH1 well , #9 and 12 inhibited it weakly , and #10 and 11 had no effect on RNAseH1 . Therefore , it is possible to inhibit the HBV RNAseH without inhibiting human RNAseH1 . Finally , we asked whether HBV RNAseH inhibitors could block HBV replication in culture . Huh7 cells were transfected with genomic expression vectors for HBV genotype A or D isolates , the cells were treated with 10 or 50 µM compounds , and viral nucleic acids were isolated from intracellular HBV capsids after four days . Replicate nucleic acid aliquots were mock treated or treated with DNAse-free E . coli RNAseH to destroy RNA:DNA heteroduplexes , and then HBV DNAs were detected by Southern blotting . The signature of RNAseH inhibition is accumulation of RNA:DNA heteroduplexes that migrate as double-stranded species without exogenous RNAseH treatment but as faster-migrating single-stranded DNAs following RNAseH treatment . The mobility of the DNAs synthesized in cells containing the wild-type genotype A genome was unaffected by exogenous RNAseH treatment ( Fig . 10 ) . Ablation of RNAseH activity by the D702A mutant altered migration of the double-stranded forms , and treatment of these samples with RNAseH collapsed the double-stranded forms to single-stranded DNAs ( Fig . 10 left panel ) . The mobility of HBV DNAs from cells replicating HBV genotype A treated with DMSO was unaffected by RNAseH digestion ( Fig . 10 center panel ) , but treatment of cells with compound #12 at 10 µM blocked production of the slowest-migrating double-stranded forms and led to accumulation of RNA:DNA heteroduplexes whose mobility increased upon removal of RNA . Treatment of cells with 3 to 50 µM compound #12 revealed that the degree of inhibition was proportional to the concentration of the compound ( data not shown ) . Plus-strand preferential real-time PCR across the gap in the minus-polarity viral DNA revealed that 10 µM compound #12 reduced plus-strand DNA accumulation to 7 . 3% of the DMSO-treated control ( data not shown ) . None of the other compounds reproducibly inhibited HBV genome synthesis ( Table 2 ) , but compound #14 ( 25 µM ) inhibited HBV replication in one experiment and #40 ( 50 µM ) inhibited replication in another experiment . Overt cellular toxicity was not observed for any of the compounds at 10 µM . Toxicity was often observed at higher concentrations; this led to the reduced yield of HBV DNA from cultures treated with 50 µM compounds #5 , 6 , and 8 in Fig . 10 . The effect of the compounds on replication of a genotype D isolate was tested to evaluate the generality of the results with the genotype A isolate . Treatment of capsid-derived nucleic acids from the DMSO control cells with exogenous RNAseH led to partial conversion of the double-stranded molecules to single-stranded forms . Therefore , RNA:DNA heteroduplexes accumulated in capsids even in the absence of RNAseH inhibitors . This indicates that the RNAseH activity during reverse transcription was incomplete for this isolate . Very few of the most slowly-migrating double-stranded nucleic acids accumulated in cells treated with 10 µM compound #12 , and many of the duplex DNAs collapsed to single-stranded forms upon treatment with exogenous RNAseH . Therefore , the inefficient HBV RNAseH in this isolate created a high background , but we were able to detect suppression of the HBV RNAseH activity above background by compound #12 . None of the other compounds tested against the genotype D isolate detectably inhibited HBV replication ( Table 2 ) . Therefore , compound #12 inhibited replication of HBV genotypes A and D in cells at low µM concentrations by blocking RNAseH activity , with the anti-RNAseH effect being somewhat less pronounced than complete ablation of the activity by mutating the RNAseH active site .
Nucleos ( t ) ide analog therapy has turned chronic HBV infection into a disease that can be controlled indefinitely , with enormous benefits to patients [6] , [7] , [85] . However , the infection is very rarely cleared , so treatment is essentially life-long , very expensive , and may be associated with unpredictable long-term side effects . Despite these limitations , the ability of protracted nucleos ( t ) ide analog therapy to slowly suppress cccDNA and HBsAg and to cure a small minority of HBV patients [10]–[13] , [21]–[23] indicates that the nucleos ( t ) ide analogs can push the virus to the brink of elimination . This implies that many more patients could be cured by employing a new drug against a novel HBV target in combination with the nucleos ( t ) ide analogs to further suppress HBV replication . Here , we report production of recombinant HBV RNAseH suitable for low throughput antiviral drug screening and demonstrate that chemical structure-activity relationships based on HIV RNAseH and integrase inhibitors can guide identification of compounds likely to inhibit the HBV enzyme . Production of soluble recombinant HBV polymerase or domains of the polymerase is notoriously difficult , and our experience with the HBV RNAseH domain was no exception . Soluble HBV RNAseH accumulated to low levels in E . coli and was a minor component of the extracts even after nickel-affinity enrichment ( Fig . 4 ) . Much of the RNAseH was apparently cleaved near its N-terminus , and these cleavage products are unlikely to be active because their sizes imply that they lack D702 . Although the concentration of the intact enzyme was very low , its specific activity was high enough to yield readily detectable signals in both radioactive and fluorescent RNAseH assays ( Fig . 5 ) . Potenza et al . [57] previously expressed recombinant HBV RNAseH that was very similar to HRHPL ( genotype D ) , but their expression conditions led to accumulation of the enzyme in inclusion bodies , necessitating refolding following purification under denaturing conditions . The refolded enzyme possessed RNAse activity , but this activity was not demonstrated to be an RNAseH . Differences between the assays employed here and in Potenza's study prevent comparison of the specificity and specific activity of the enzyme prepared under native and denaturing conditions . The optimal reaction conditions for the recombinant HBV RNAseH ( Table 1 ) were typical for nucleic-acid modifying enzymes and were similar to conditions in which recombinant hepadnaviral reverse transcriptase is active [86]–[88] . Its activity was dependent upon a divalent cation , but it became active against single-stranded RNA in addition to RNA in a heteroduplex when Mn++ was substituted for Mg++ ( data not shown ) . This is similar to the reduced fidelity of nucleic acid polymerases ( including the duck HBV polymerase ) in the presence of Mn++ [89]–[91] . The RNAseH had a relatively high NaCl optimum of 190 mM and it lost specificity for heteroduplex RNA at low ionic strength ( data not shown ) . Importantly given that a primary goal of this study was to produce enzyme suitable for antiviral drug screening , recombinant HBV RNAseH was stable upon storage in liquid nitrogen , could be repeatedly frozen and thawed , and was fully active in up to 2% DMSO . Therefore , enzyme suitable for low-throughput anti-HBV RNAseH drug screening has been produced . The HIV RNAseH is a very active target of ongoing antiviral drug discovery [38] , [59]–[78] , but to our knowledge none of the anti-HIV RNAseH compounds have entered clinical trials yet . This is primarily due to the relatively low therapeutic indexes of most known anti-HIV RNAseH compounds . Similar challenges were faced by the HIV integrase field in the early stages of development of anti-integrase drugs . Many inhibitors were discovered , but clinical development did not begin until strand transfer inhibitors , active site metal binders , etc . were discovered . The failure to advance to HIV RNAseH inhibitors to clinical trials may also be partially due to the large number , high potency , and diverse profile of existing anti-HIV drugs . In contrast , current anti-HBV therapies are primarily based on a single class of inhibitors , nucleos ( t ) ide analogs . Hence , inhibitors of a new HBV enzymatic function would address the current challenges of limited efficacy and cross-resistance among the nucleos ( t ) ide analogs , and this would allow meaningful combination therapies for HBV similar to HAART that dramatically changed the landscape of anti-HIV therapy . The ability to template HBV RNAseH drug discovery on the HIV experience would greatly accelerate anti-HBV efforts . The HIV data could narrow the chemical space to be assessed during screening , compounds synthesized during anti-HIV RNAseH screening would be available for immediate screening against HBV , and the toxicity profile of some of these compounds is known . Templating anti-HBV RNAseH drug development on HIV efforts would be analogous to the development of the anti-HBV nucleos ( t ) ide analogs , which was greatly facilitated by the parallel development of anti-HIV nucleoside analogs [92] . Twenty-one candidate RNAseH inhibitors were selected due to their similarity to known inhibitors of the HIV RNAseH or integrase . Twelve of these compounds ( 57% ) inhibited the HBV RNAseH at 10 µM to below the threshold defined by control reactions with irrelevant compounds ( Fig . 7 and Table 2 ) . Importantly , 10 of 11 compounds analogous to anti-HIV integrase compounds inhibited the HBV RNAseH , including both approved anti-HIV integrase drugs , raltegravir ( compound #11 ) and elvitegravir ( #10 ) . This is consistent with the membership of both the RNAseH and integrase in the nucleotidyl transferase superfamily of enzymes . Therefore , there is enough similarity between the HBV RNAseH and the HIV RNAseH and integrase active sites to guide screening for anti-HBV RNAseH compounds . Most anti-HIV RNAseH inhibitors bind to the enzyme and chelate the divalent cations in the active site [64] , [65] , [69] , [70] , [73] , [76] . Similarly , anti-HIV integrase compounds that target the active site typically do so by binding to the enzyme or the enzyme plus DNA and chelating the active site divalent cations [93] . The compounds tested here were selected for the ability to bind to Mg++ ions oriented as they are in the HIV RNAseH or integrase active sites , and hence inhibition of the HBV enzyme is predicted to be through binding to the active site and interfering with the Mg++ ions . The mechanisms by which the HBV RNAseH inhibitors function have not been determined , but IC50 curves reveal at least two patterns . The profiles for compounds #12 , 39 , and 40 were consistent with the predicted competitive inhibition mechanism ( Fig . 8C and data not shown ) . In these cases , inhibition appears to be specific . Other compounds , such as #6 and #8 , had inhibition profiles with one or more broad plateaus that were inconsistent with simple competitive binding to the active site . In addition , the electrophoretic mobility of the RNA was retarded at high concentrations of compound #8 ( Fig . 8C ) , implying that this compound may react with the RNA substrate . The compounds employed here were selected by structure-activity relationships with the goal of testing whether these relationships could predict biochemical inhibition of the HBV RNAseH . The compounds were not selected to have other properties necessary for a drug , such as the ability to enter cells . Nevertheless , compound #12 inhibited HBV replication in cell culture at 10 µM without extensive cellular toxicity ( Fig . 10 ) . The reduction in mobility following treatment of capsid-derived nucleic acids with E . coli RNAseH demonstrates that RNA:DNA heteroduplexes accumulated in the viral capsid in the presence of compound #12 , confirming that these compounds blocked HBV RNAseH activity in culture . Therefore , it is possible to pharmacologically inhibit the HBV RNAseH in cells , and identification of anti-HBV compounds that are active in cells can be achieved employing structure-activity relationships based on anti-HIV compounds . Furthermore , the ability of compounds identified by screening against recombinant genotype D and H enzymes to inhibit both genotype A and D isolates in culture demonstrates that it is possible to identify RNAseH inhibitors that are active against a range of HBV isolates . The sensitivity profile of the HBV genotype D and H RNAseHs to the inhibitors was not the same ( Table 2 ) . This has two implications . First , the genotype H RNAseH may be a better candidate for primary drug screening than the genotype D enzyme because its inhibition profile more accurately predicted inhibition of HBV replication in culture . Second , the variable sensitivity of the genotype D and H enzymes to the compounds indicates that HBV's high genetic diversity is likely to be an important issue during development of anti-HBV RNAseH drugs . The key HBV molecule that must be eradicated to cure patients is the viral cccDNA ( Fig . 1 ) [25] . Ideally , clearing the cccDNA would be achieved by simultaneously suppressing its synthesis rate with the existing nucleos ( t ) ide inhibitors and increasing its degradation rate with a new drug . The problem with this approach is that we do not know how to safely destabilize the cccDNA , so the approach that has the most realistic chance of clearing HBV in the foreseeable future is to further suppress its synthesis rate . Importantly , pharmacological suppression of viral genomic synthesis may not need to completely eradicate the cccDNA by itself because the latter stages of viral clearance may be assisted by the immune system . HBV's proteins , including HBsAg [94]–[99] , HBeAg [100] , [101] , and the polymerase [102]–[104] , have immunosuppressive activities . Consequently , if viral genomic replication can be suppressed far enough to inhibit cccDNA synthesis rather than just virion secretion ( Fig . 1 ) as is usually achieved with the nucleos ( t ) ide analogs , levels of the cccDNA would drop . This reduction in the transcriptional template would reduce production of HBV's proteins , presumably weakening HBV's immunosuppression and promoting immune-mediated viral clearance . Three challenges remain prior to beginning full-scale antiviral drug screening against the HBV RNAseH . First , the majority of HBV's disease burden is caused by genotypes B and C , and we have been unsuccessful to date in generating consistently active recombinant RNAseH from these genotypes . This challenge is likely to be surmountable because only a few isolates of these genotypes have been tested for activity and because compound #12 identified by screening against genotypes D and H inhibited replication of HBV genotype A in culture , confirming that cross-genotype inhibition is possible . Second , the existing tissue culture and biochemical assays are sufficient for low throughput drug screening , but anti-HBV RNAseH drug development is anticipated to require screening many thousands of compounds even when the chemical search space is constrained by prior studies with HIV . Therefore , full-scale drug screening and subsequent mechanistic assessment of hit compounds will require improving the yield and purity of the biochemical RNAseH assay . This challenge should be met by further optimizing the induction and extraction conditions , expanding the bacterial induction cultures beyond the 100 ml scale used in this study , adding a second purification step such as ion-exchange chromatography , and expanding efforts to control proteolysis of the enzyme . We are optimistic this goal can be achieved because recent improvements to the induction and extraction conditions have increased the specific activity of the enzyme approximately four-fold , and initial scale-up experiments have not met with difficulty . Finally , the HBV RNAseH assay must be adapted to a format suitable for high throughput screening . This challenge should also be surmountable because fluorescent RNAseH assays have been widely employed to screen for anti-HIV RNAseH inhibitors and because the signal∶background ratio for the first-generation HBV RNAseH fluorescent assay in Fig . 5 should be improved by increasing the concentration of the RNAseH and/or by optimizing the substrate structure .
pCMV-HBV-LE- ( CMV-HBV ) is an HBV over-length genomic expression vector containing 1 . 2 copies of the HBV ( adw2 ) genome ( Genbank X02763 . 1 ) downstream of the CMV promoter cloned into pBS ( Promega ) . Surface protein expression from this vector is ablated by mutating the preS and S open reading frames . pCMV-HBV ( genotype D ) is an analogous HBV genomic expression construct and was a gift from Dr . Shuping Tong . For bacterial expression , codon-optimized cDNA sequences for HRHPL ( genotypes A , B , C , D , and H ) were cloned by gene synthesis ( Genscript ) between the NcoI and EcoRI sites into pTrcHis2B ( Invitrogen ) with a C-terminal hexahistidine tag . HRHPL contains HBV genotype D ( Genbank V01460 ) polymerase residues 684–845 . The human RNAseH1 gene ( NP_002927 . 2 ) was cloned with an N-terminal hexahistidine-tag between the BamHI and XhoI sites of pRsetB ( Invitrogen ) by gene synthesis . HRHPL and human RNAseH1 were expressed in E . coli BL21 codon+ cells ( Invitrogen ) . Saturated overnight bacterial cultures were diluted 4-fold into 100 ml fresh medium and protein expression was induced with 0 . 5 mM IPTG at 30°C for six hours . The cells were lysed by sonication in lysis buffer [50 mM HEPES pH 8 . 0 , 800 mM NaCl , 0 . 1% NP40 , 27 . 5% glycerol , 2 mM DTT 20 mM imidazole , and protease inhibitor cocktail ( Sigma ) ] . RNAseH proteins were enriched by nickel–agarose affinity chromatography , eluted with 350 mM imidazole , dialyzed into 50 mM HEPES pH 7 . 3 , 300 mM NaCl , 20% glycerol , and 5 mM DTT , and stored in liquid nitrogen . For the oligonucleotide-directed RNAseH cleavage assay [46] , 6 µl protein extract ( typical protein concentration 2 . 8 mg/ml ) was mixed with 0 . 5 µg internally 32P-labeled DRF+ RNA ( nucleotides 2401–2605 of the duck HBV genome plus 60 nucleotides of flanking sequences from pBluescript ) and 3 µg oligonucleotide D2507− or its corresponding negative control D2526+ on ice in 20 µl under the conditions in Table 1 . Some reactions in Fig . 5 employed oligonucleotide D2543M-Sal or its D2453+ negative control as indicated . The reactions were incubated at 42°C for 90 min . and terminated by addition of Laemmli protein loading buffer and boiling . The samples were resolved by 12% SDS-PAGE , the gels were stained with Coomassie blue to monitor protein loading , and labeled RNA was detected by autoradiography . Candidate inhibitors were dissolved in DMSO and added at the indicated concentrations during assembly of the reactions . Control reactions lacking the compounds contained DMSO as a vehicle control . The RNAseH autoradiograms were scanned and quantified with ImageJ . The oligonucleotides were: D2526+ ( CCACATAGGCTATGTGGAAC ) , D2507− ( GTTCCACATAGCCTATGTGG ) , D2453+ ( CCGCCTGATTGGACGGCTTTTCC ) , and D2543M-Sal ( GCAACTGTGTCGACAGCAGCTCCGAAGGAGA ) . For the fluorescent RNAseH assay , the DRF+ RNA and DNA oligonucleotides were omitted from the RNAseH reactions and replaced with 20 µM of the quenched fluorescent chimeric RNA:DNA oligonucleotide RHF1; the reaction conditions were identical to the oligonucleotide-directed reactions except that the NaCl concentration was reduced to 130 mM . The reactions were incubated in the dark at 42°C for 90 min . prior to termination by addition EDTA to 10 mM and detection of fluorescence at 520 nM on Synergy 4 plate reader ( Biotec , Inc . ) . The sequence of the RHF1 substrate ( IDT , Inc . ) was: 5′-56-FAM/rCrCrArCrArUrArGrGrCrUrArUrGrUrGrGrArArCTTTTGTTCCACATAGCCTATGTGG/3IBkFQ/-3′ . The RNA:DNA heteroduplex in the RHF1 substrate was the same as the heteroduplex formed by oligo D2507− annealed to DRF+ . Huh7 cells were maintained in Dulbecco's modified Eagle's medium with 10% fetal bovine serum at 37°C in 5% CO2 . Cells were seeded into 60 mm dishes and transfected at 70% confluency with 2 . 6 µg of plasmids using TransIT-LT1 ( Mirus , Inc . ) . Test compounds were added the morning following transfection at 10 or 50 µM , and fresh medium containing the compounds was provided every 1–2 days . Four or five days post-transfection HBV cores were isolated by lysis of the cells in 10 mM Tris pH 7 . 5 , 1 mM EDTA , 0 . 25% NP40 , 50 mM NaCl , and 8% sucrose followed by sedimentation through a 30% sucrose cushion as described [105] . Viral DNAs were isolated from cytoplasmic core particle preparations by proteinase K digestion followed by phenol/chloroform extraction as described [46] . Duplicate aliquots of the nucleic acids were treated with 2 U E . coli DNAse-free RNAseH ( Invitrogen ) at 37°C for 30 min . or were mock treated . The nucleic acid samples were resolved by electrophoresis on 1 . 2% agarose gels and detected by Southern blotting with 32P-labeled HBV DNA as a probe .
|
Current therapy for HBV blocks DNA synthesis by the viral reverse transcriptase and can control the infection indefinitely , but treatment rarely cures patients . More patients could be cured by suppressing HBV replication further using a new drug in combination with the existing ones . The HBV RNAseH is a logical drug target because it is the second of only two viral enzymes that are essential for viral replication , but it has not been exploited , primarily because it is very difficult to produce active enzyme . We expressed active recombinant HBV RNAseHs and demonstrated that it was suitable for antiviral drug screening . Twenty-one candidate HBV RNAseH inhibitors were identified based on antagonists of the HIV RNAseH and integrase enzymes . Twelve of these compounds inhibited the HBV RNAseH in enzymatic assays , and one inhibited HBV replication in cell-based assays . The high percentage of compounds developed against the HIV RNAseH and integrase that were also active against the HBV RNAseH indicates that the extensive drug design efforts against these HIV enzymes can be used to guide anti-HBV RNAseH drug discovery .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"drugs",
"and",
"devices",
"viral",
"enzymes",
"microbiology",
"hepatitis",
"pharmacology",
"infectious",
"diseases",
"proteins",
"hiv",
"biology",
"recombinant",
"proteins",
"viral",
"replication",
"drug",
"discovery",
"biochemistry",
"hiv",
"diagnosis",
"and",
"management",
"hepatitis",
"b",
"drug",
"research",
"and",
"development",
"virology",
"viral",
"diseases",
"antivirals"
] |
2013
|
The Hepatitis B Virus Ribonuclease H Is Sensitive to Inhibitors of the Human Immunodeficiency Virus Ribonuclease H and Integrase Enzymes
|
Diseases of protein folding arise because of the inability of an altered peptide sequence to properly engage protein homeostasis components that direct protein folding and function . To identify global principles of misfolding disease pathology we examined the impact of the local folding environment in alpha-1-antitrypsin deficiency ( AATD ) , Niemann-Pick type C1 disease ( NPC1 ) , Alzheimer's disease ( AD ) , and cystic fibrosis ( CF ) . Using distinct models , including patient-derived cell lines and primary epithelium , mouse brain tissue , and Caenorhabditis elegans , we found that chronic expression of misfolded proteins not only triggers the sustained activation of the heat shock response ( HSR ) pathway , but that this sustained activation is maladaptive . In diseased cells , maladaptation alters protein structure–function relationships , impacts protein folding in the cytosol , and further exacerbates the disease state . We show that down-regulation of this maladaptive stress response ( MSR ) , through silencing of HSF1 , the master regulator of the HSR , restores cellular protein folding and improves the disease phenotype . We propose that restoration of a more physiological proteostatic environment will strongly impact the management and progression of loss-of-function and gain-of-toxic-function phenotypes common in human disease .
The transition from protein folding to misfolding , in both normal physiology and disease , is dynamically managed by multiple proteostatic pathways [1] , [2] . The heat shock response ( HSR ) is a central signaling pathway managing the malleable composition of the proteostasis network ( PN ) of folding and degradation machineries . The cellular PN environment contributes to what we refer to as the quinary ( Q ) state of the protein fold [3]–[6] , which emphasizes that the structure of a protein is tightly integrated with a dynamic proteostatic system to direct structure–function relationships in health and address challenges in response to disease [1] , [5] , [71]–[11] . Q-state managers of each protein fold draw from the proteostasis pool of molecular chaperones ( Hsp40s , Hsc70s , Hsp70s , and Hsp90 ) , small heat shock proteins , and ubiquitin-based degradation components [3] , [5] , [12]–[14] . These managers are responsive to multiple signaling pathways including the unfolded protein response ( UPR ) [15] , controlling compartmentalized folding , and the heat-shock response ( HSR ) , controlling cytoplasmic/nuclear folding [8] . The importance of an integrated Q-state is exemplified in the function of coupled protein synthesis and folding machineries [16] , linked cargo-specific folding and trafficking machineries [1] , [5] , [7]–[9] , [17] , and the activity of cytoplasmic Q-bodies that actively monitor the health of each protein in the cytoplasm [6] , [18] . Together , these proteostasis machineries operate as integrated sensors of individual protein structure–function relationships that now need to be understood [3] , [6] , [19]–[24] . The HSR is controlled by the heat shock transcription factor 1 ( HSF1 ) , with the chaperone Hsp90 regulating its activation [8] , [25] . Transient stimulation of the HSR pathway , based on the heat shock paradigm [26] , is generally beneficial in that it alters the composition of proteostasis components in the cytosol to provide immediate , but temporary , protection to misfolded proteins in the face of divergent stress insults [8] , [9] , [27] . Consistent with this view , transcriptional profiling , in response to acute heat shock , revealed that approximately 500 genes are up-regulated , whereas more than 1 , 000 genes are repressed [28]–[30] , conditions that , if sustained , could negatively impact cell viability . Our understanding of these complex gene expression changes and their impact on protein structure–function relationships in response to chronic folding insults remains to be elucidated . Diseases of protein folding arise due to the inability of an altered peptide sequence to properly engage the prevailing local proteostasis components . Gain-of-toxic-function diseases such as Alzheimer's ( AD ) [31] and inherited loss-of-function diseases such as alpha-1-antitrypsin deficiency ( AATD ) [32] , [33] , Niemann-Pick type C1 disease ( NPC1 ) , and cystic fibrosis ( CF ) [34] present a unique challenge to cells because of the chronic nature of the insult [34] , [35] . A current paradigm in disease biology is that stress pathways are not sufficiently activated to provide the necessary protection . Therefore , activation of these pathways , such as the HSR , should improve folding and/or clearance of disease-related proteins . Indeed , HSF1 activation has been shown to provide partial correction for some misfolding diseases [36] , however , the in vivo benefits for the chronic activation of HSF1 have not been investigated . Recently , HSR activation has been shown to exacerbate the aggregation of mutant huntingtin protein ( htt-Q91 ) in a cellular model of Huntington's disease ( HD ) [37] . Moreover , sustained HSR activation promotes proliferation of cancer cells [28] , [38] , a pathologic disease state leading to reduced human lifespan . In cancer cells , HSF1 drives a distinct transcriptional program from the classical HSR , implying a more complex function than previously anticipated [39] . We have recently suggested that HSF1 activators that partially promote correction of CF do so by activation of unknown cellular pathways [40] , which we now need to understand in the context of the prevailing proteostasis biology to provide new insights into the evolution of chronic disease management by the cell [1] , [5] . Herein we have studied four misfolding disorders to address central principles in managing chronic protein folding stress in human disease: ( 1 ) the deletion of phenylalanine 508 ( F508del ) variant of the cystic fibrosis transmembrane conductance regulator ( CFTR ) ( F508del-CFTR ) , a multi-membrane–spanning protein with large cytoplasmic domains , which fails to traffic to the plasma membrane and is responsible for 90% of CF cases [10] , [34]; ( 2 ) the Z-variant of alpha-1-antitrypsin ( Z-AAT ) , which accumulates as a misfolded polymer in the early secretory endoplasmic reticulum ( ER ) compartment , leading to liver disease and chronic obstructive pulmonary disease ( COPD ) /emphysema because of its failure to be secreted and delivered to the lung [32] , [33]; ( 3 ) the I1061T variant of NPC1 , key component in lipid and cholesterol homeostasis in the late endosome/lysosome ( LE/L ) compartment , which fails to traffic from the ER to the LE/L in human disease , resulting in the lysosomal storage disease NPC1; and ( 4 ) AD , which arises from aberrant Alzheimer precursor protein ( APP ) processing and trafficking , resulting in accumulation of extracellular Aβ amyloid aggregates [31] , [41] . Although our primary focus is on the correction of CF disease , we now show that the long-term expression of disease-causing misfolded proteins results in what we refer to as a maladaptive stress response ( MSR ) , a state reflecting the sustained activation of the HSR pathway , which contributes to disease progression by undermining the normal folding capacity of cells . We provide evidence that silencing of HSF1 alleviates the MSR and improves the multiple disease phenotypes , suggesting a general principle that chronic alteration of the prevailing PN contributes to the progression of inherited diseases , a step that will now require active management to mitigate pathophysiology [1] , [6] .
CF is caused by mutations in the multi-membrane–spanning protein CFTR , a chloride channel responsible for ionic and fluid homeostasis in the lung [34] . The F508del variant of CFTR is characterized by misfolding , ER accumulation , and removal by ER-associated degradation ( ERAD ) [34] . F508del-CFTR is retained in the ER in a Hsp70/90-containing chaperone trap , a step that wild-type ( WT ) -CFTR and temperature-corrected F508del ( 30°C ) are able to navigate [42] . We therefore focused our attention on the HSR pathway that manages cytoplasmic chaperone biology . To assess the effect of HSR activation on the folding environment , we first heat shocked bronchial epithelial cells ( CFBE41o- ) expressing WT- or F508del-CFTR and monitored its impact on CFTR protein stability and trafficking . CFTR stability and trafficking is monitored by Western blot , in which the ER-localized ( band-B ) and post-ER glycoforms ( band-C ) exhibit a differential migration pattern . Whereas WT-CFTR remained mostly unaffected , more than 90% of F508del-CFTR was degraded after 60 min of heat shock ( HS ) ( Figure 1A , B ) . HS activation was confirmed by increased HSF1 phosphorylation of Serine-326 ( HSF1-P at S326 ) . Since F508del-CFTR is sensitive to alterations in temperature , we determined whether the destabilization of F508del was caused by HSR activation and not simply elevated temperature . For this purpose , we overexpressed a constitutively active variant of HSF1 ( ΔHSF1186–201 ) [43] , [44] with F508del-CFTR in CFBE41o- cells . Overexpression of active ΔHSF1186–201 , confirmed by elevated levels of HSF1-P and the stress-inducible Hsp70 ( I-Hsp70 ) , also led to destabilization of F508del-CFTR ( Figure 1C ) . These data support the conclusion that activation of the HSR pathway results in destabilization of F508del-CFTR rather than correcting the stability and trafficking defect associated with this disease variant . We also observed that in the absence of HS , F508del-expressing cells already exhibited increased HSF1-P relative to that seen in WT-expressing cells ( Figure 1A , 1D ) , revealing that the HSR pathway was already hyperactive in these cells . To confirm this observation , we compared additional markers of HSF1 activation , including HSF1 trimerization and expression of I-Hsp70 . Cells expressing F508del-CFTR exhibited a significant increase in total , trimerized , and phosphorylated HSF1 , as well as increased I-Hsp70 levels relative to WT-expressing cells ( Figure 1D–1F ) . We also observed a significant increase in mRNA levels of the HSF1-responsive genes , HspA1A ( I-Hsp70 ) , Hsp90α ( I-Hsp90 ) , and DNAJB1 ( I-Hsp40 ) , relative to that seen in WT-expressing and in isogenic cells lacking CFTR ( CFTR −/− ) ( Figure 1G ) . Additionally , silencing of F508del-CFTR led to a significant decrease in HSF1 and HSF1-P expression ( Figure S1A , S1B ) , suggesting that the observed HSR activation is directly related to the expression of this misfolded CFTR variant . Temperature correction of F508del , which corrects its associated stability and trafficking defects , also led to a reduction in HSF1 and HSF1-P to levels seen in WT-expressing and CFTR−/− cells ( Figure 1E ) . Altogether , our results suggest that the HSR activation observed in F508del-expressing cells at physiological temperature is a direct consequence of the expression of the misfolded F508del-CFTR . To address whether the observed HSR activation was in response to the immortalized CFBE41o- cell line phenotype , we also examined these markers on patient-derived human bronchial epithelia ( hBE ) homozygous for WT- or F508del-CFTR . Consistent with the findings observed in cystic fibrosis bronchial epithelial ( CFBE ) cells , F508del-expressing hBE cells also showed elevated HSF1-P and I-Hsp70 protein levels , as well as increased I-Hsp70 ( HspA1A & A6 ) , I-Hsp90 , and I-Hsp40 mRNA levels , relative to that seen in WT-expressing hBEs ( Figure 1H–1J ) . No differences were observed in mRNA levels of non-classical HSF1-responsive genes , previously shown to be increased in cancer cells ( CKS2 , LY6K , and EIF4A2 ) ( Figure 1J ) [28] , suggesting activation of the classical HSF1 pathway . In order to quantify the magnitude of this HSR activation , we compared the up-regulation of the I-Hsp40 and I-Hsp70 protein levels seen in F508del-expressing cells to that seen after HS . We observed a 1 . 5- and 2-fold increase in I-Hsp40 and I-Hsp70 , respectively , in F508del-expressing CFBE cells relative to that seen in WT-expressing cells , whereas a 3 . 5- and 4-fold increase in I-Hsp40 and I-Hsp70 , respectively , was observed after acute HS ( Figure 1K ) . Thus , the level of HSR activation seen in response to chronic expression of F508del-CFTR represents approximately 50% of that seen during acute HS , indicating the presence of a subacute , chronic activation of the HSR pathway . The transcriptional changes reported to occur in response to HSR activation [28]–[30] are likely to have a global impact on cellular function . Thus , we monitored the folding of firefly luciferase ( FLuc ) , a sensor of folding stress in the cytosol [34] , [45] that has also been used to monitor both ER and oxidative stress [46]–[50] . Here , we used the FLuc reporter not as an absolute measure of protein folding , but as a sensor for relative cytoplasmic folding stress when comparing control with diseased cells . Importantly , F508del-expressing cells exhibited a 50% reduction in the specific activity of FLuc compared to WT-expressing cells , which was restored to WT-like levels in response to F508del silencing ( Figure 1L ) . Since the chronic activation of the HSR , observed in response to the expression of a misfolded protein , negatively impacts the folding of other cellular proteins as reported by FLuc , a state which is likely to impact multiple cellular function ( s ) ( Figure 1M ) , we refer to this altered PN environment as a maladaptive stress response ( MSR ) . Given the increased activation of HSF1 in cells expressing F508del-CFTR , we next examined the impact of the Hsp90 co-chaperone , p23 , an important regulator of HSF1 activity [8] , [51] , [52] . Since the MSR is a chronic response , we performed all small interfering RNA ( siRNA ) interventions for a total of six days to allow for appropriate rebalancing of the PN environment . P23 silencing significantly reduced HSF1 activation in response to HS , as exemplified by a reduction in the level of HSF1-P ( Figure 2A ) , confirming its central role in the activation cycle of HSF1 . At physiological temperature , p23 silencing in F508del-expressing CFBEs also resulted in a significant decrease in HSF1 and HSF1-P protein levels , as well as I-Hsp70 mRNA and protein levels ( Figure 2B , 2C ) , to a level similar to that seen in WT-expressing cells ( Figure 2D ) . Furthermore , abrogation of the MSR following p23 silencing led to a concomitant restoration of FLuc folding in F508del-expressing CFBEs ( Figure 2E ) . Silencing of p23 had no effect on HSF1 mRNA level ( Figure S2A ) nor on HSF1 stability , determined by pulse-chase ( Figure S2B , S2C ) . However , we did observe a reduction in the amount of labeled HSF1 in the pulse-phase ( Figure S2B , S2D ) , indicating a reduction in HSF1 translation in response to p23 silencing . In contrast , p23 silencing had no impact on HSF1 levels in WT-CFTR expressing cells , in which no MSR is detected ( Figure S2E ) , suggesting that p23 plays a critical role in modulating the MSR induced in F508del-CFTR expressing cells . Since p23 silencing reduced the MSR state in F508del-expressing cells , we assessed its effect on F508del-CFTR biogenesis . P23 silencing resulted in a significant increase in F508del ER stability ( band-B ) and trafficking ( band-C ) compared to control siRNA treatment ( Figure 3A ) . It also resulted in an increase in the trafficking index , defined as the ratio of band-C to band-B ( C/B ) [53] , an indicator of its post-ER stability ( Figure 3A ) . These results suggest that a reduction of the MSR , which restores a WT-like PN state ( Figure 2D ) , supports the increased trafficking efficiency of F508del similar to what is observed following 30°C correction ( Figure 3B ) , providing significant benefit to the CF phenotype . P23 silencing did not increase WT stabilization or trafficking ( Figure 3C ) , indicating that its effect on F508del correction occurs in response to alleviation of the MSR exclusively seen in F508del-expressing cells . CFTR pulse labeling in response to sip23 revealed a significant increase in the synthesis of F508del-CFTR but not of WT-CFTR ( Figure S3B , S3D ) , consistent with the results presented above for the steady-state levels of CFTR ( Figure 3A , 3C ) . This differential synthesis could be due to change in transcription , translation and/or post-translational stability of F508del-CFTR . Although p23 down-regulates the transcription of the glucocorticoid and thyroid hormone receptors [54] , [55] , we did not observe any changes in WT- or F508del-CFTR mRNA levels ( Figure 3D ) . However , p23 silencing did significantly reduce the degradation rate of F508del-CFTR but not that of WT-CFTR ( Figure S3A , S3C ) , suggesting that p23 specifically affects the stability of nascent F508del-CFTR . Increased F508del stability was not due to altered proteasome activity , since combining sip23 with the proteasome inhibitor , MG132 , resulted in an additive effect on F508del stability and trafficking ( Figure S3E ) . In support of this conclusion , the levels of ubiquitinated F508del following sip23 also remained unchanged ( Figure S3F ) . p23 silencing also promoted a significant reduction of Hsc/p70 and Hsp90 ( Hsp90α and Hsp90β ) levels recovered in F508del-CFTR immunoprecipitates ( Figure 3F ) , indicating that abrogation of the MSR allows F508del-CFTR to properly navigate early folding intermediates known to contribute to the ER retention of F508del-CFTR [42] . Given the observed correction of the F508del-CFTR trafficking defect by p23 silencing , we assessed whether the corrected pool of F508del was functional . F508del-expressing cells treated with sip23 exhibited a significant increase in channel activity , as determined by iodide efflux ( Figure 3E ) and short circuit current ( Isc ) recordings ( see below , Figure 4B ) . Our results show that abrogation of the MSR by p23 silencing promotes trafficking of a functional F508del-CFTR to the cell surface . Since the expression of F508del-CFTR results in chronic activation of HSF1 , which not only affects F508del biogenesis but also the activity/folding of other cellular proteins ( Figure 1L ) , we tested whether HSF1 silencing would also correct the trafficking defect associated with F508del-CFTR . HSF1 silencing resulted in a significant increase in ER stability ( band-B ) , maturation ( band-C ) and trafficking index for F508del-CFTR ( Figure 4A and Figure S4A ) . Furthermore , it also led to increase in F508del function by Isc recordings to the level seen with siHDAC7 , a validated siRNA target for correction of CF [56] , and with VX809 , a CF corrector currently in clinical trials for the treatment of F508del homozygous patients ( Figure 4B ) [57] , [58] . In order to determine whether the MSR observed in CF is a general phenomenon associated with protein misfolding diseases , we monitored the HSR activation state in models of AATD , NPC1 , and AD . In AATD , the G342K mutation in AAT , referred to as the Z-variant , results in ER-retention , polymerization , and degradation of this normally secreted enzyme , the loss of which leads to COPD [32] , [59] . Cells expressing the Z-variant exhibited higher levels of HSF1-P compared to WT-AAT expressing cells ( Figure 4C and Figure S4B ) , suggesting , once again , the existence of a MSR . Furthermore , HSF1 silencing resulted in increased maturation ( AAT-M ) and secretion ( AAT-S ) of the mutant Z-AAT ( Figure 4D and Figure S4C , S4E ) ; however , no changes in the polymerization state of the Z-AAT variant were observed ( Figure S4D ) . This result is consistent with the effect of other correctors , such as suberoylanilide hydroxamic acid ( SAHA ) , in which increased maturation and secretion of Z-AAT is observed without changes in polymerization [33] . Thus , MSR abrogation also provides benefit to a protein misfolding disease found in the ER lumen [9] , [33] , suggesting a link between ER stress biology [15] and cytoplasmic stress management by HSR . This observation is consistent with previous results suggesting a crosstalk between these pathways [60] . In addition , genomic analysis has revealed that transcriptional targets of HSF1 found in the secretory pathway are also induced by UPR [61]–[63] , providing a mechanism by which silencing of HSF1 could be beneficial for AATD . We next investigated whether a MSR arose in response to the I1061T variant of the NPC1 protein responsible for the lysosomal storage disease Niemann-Pick type C1 , which , like CF , is characterized by protein misfolding and ERAD-mediated clearance [64] . An analysis of human primary fibroblasts from homozygous I1061T NPC1 patients and healthy donors ( WT ) reveals that cells expressing the I1061T variant exhibit an elevation in the levels of HSF1-P , suggesting the presence of a MSR ( Figure 4E and Figure S4F ) . Here HSF1 silencing also improved the trafficking defect of I1061T-NPC1 , as exemplified by the increased endo H resistance reflecting modification of its N-linked oligosaccharides by Golgi enzymes , relative to that seen with control siRNA ( Figure 4F ) . We then examined a Caenorhabditis elegans model of cytoplasmic amyloid aggregation . C . elegans expressing the β-amyloid-42 ( Aβ42 ) peptide fused to CFP ( Aβ42-CFP ) under the control of a muscle-specific unc-54 promoter forms CFP-positive Aβ aggregates in the cytoplasm of muscle cells ( Figure S5A , S5B ) . The C . elegans model has been extensively used in the field of misfolding diseases and is a validated tool to study the impact of amyloid disease in organismal models [19] , [21] , [65] , [66] . Here we observed an increase in I-Hsp70 level in Aβ42 worms ( ∼150-fold , Figure S5C ) , which was not further up-regulated after HS as seen in WT worms . Up-regulation of I-Hsp70 was reduced in response to HSF1 silencing or reduction of Aβ42 expression ( Figure S5D ) , indicating that the misfolding stress caused by Aβ42 expression also induces a MSR state . Accumulation of cytosolic Aβ42 aggregates led to paralysis in 75% of diseased worms relative to its WT counterparts , which was significantly reduced by silencing of not only Aβ42 ( silencing of yellow fluorescent protein- [siYFP] ) but also in response to I-Hsp70 and HSF1 silencing ( Figure S5E , S5F ) . Conversely , HSF1 overexpression resulted in increased Aβ42 induced proteotoxicity with an approximately 30% increase in paralyzed worms ( Figure S5G ) . To extend these observations to a neurodegenerative model of Aβ42 amyloid aggregation , we examined the expression levels of HSF1 and HSF1-P ( phosphorylated at T142 ) [67] in brain homogenates of WT and AD mice ( AβPP Tg ) at three different ages ( approximately 4 mo , 9 mo , and 16 mo old ) . We observed a significant increase in both HSF1 and HSF1-P expression in all AD mice compared to their age-matched WT counterparts ( Figure 4G ) . The toxic Aβ42 amyloid species ( 4 kDa monomer and 6-12 kDa multimers ) [68] , [69] , previously characterized in this AβPP Tg mice model [70] , were detected in brain homogenates from AD mice but not in that of WT mice . The accumulation of Aβ42 amyloid in AD mice was also age dependent ( Figure 4H ) , consistent with previously published studies showing age-dependent increase in Aβ plaques , and mean plaque size on these mice [70] . Despite the age-related increase in toxic amyloid , we did not observe an age-dependent increase in HSF1-P in the AD mice , a result consistent with the known decline of proteostatic capacity as has been previously documented in aging organisms in the face of increasing cellular stress [71]–[73] . The MSR is a chronic state transferring the misfolding challenges to all aspects of cellular folding biology managed by proteostasis components impacting the activity of the Q-state of F508del [42] . Thus , we examined in more detail the impact of HSF1 silencing , which in our CF cell model resulted in increased stability and trafficking of F508del-CFTR at steady state ( Figure 4A ) . To address whether the observed increased in F508del stability reflected an increase in global protein synthesis , we compared the level of S35-labeled proteins in cellular lysates from F508del-expressing cells in the presence or absence of siHSF1 to that seen in WT-expressing cells . Strikingly , we first observed that MSR-affected F508del-expressing cells exhibited a drastic decrease in total protein synthesis , representing less than 50% of that seen in healthy WT-expressing cells ( Figure 5A ) . This highlights the negative impact of MSR activation on the proteome and is consistent with attenuation of protein synthesis seen in numerous types of stress [74] . HSF1 silencing had no impact on the level of total protein synthesized in F508del-expressing cells ( Figure 5A ) ; however , we did observe an increase in F508del synthesis after pulse labeling , followed by increased stability of de novo synthesized F508del band-B in the chase phase of the experiment ( Figure 5B ) . We also observed increased stability of band-C after inhibition of de novo protein synthesis by cycloheximide ( CHX ) treatment ( Figure 5C ) . To determine if band-C stability resulted from increased band B to C trafficking following CHX treatment , we used brefeldin A ( BFA ) to block ER to the Golgi trafficking and track the stability of rescued F508del-CFTR ( rF508del ) band-C by preventing egress to the cell surface . The half-life ( T1/2 ) of band-C in temperature-rescued F508del ( rF508del ) was approximately 2 h , whereas HSF1 silencing significantly increased the stability of the rF508del pool , exhibiting a T1/2 of 6 h , a value similar to that seen for WT-CFTR ( Figure 5D ) . These data suggest that alteration of the MSR by siHSF1 increases the stability of rF508del band-C , possibly as a result of improved protein folding . To directly address whether we have achieved improved protein folding following siHSF1 , we used limited trypsin proteolysis , a method previously shown to distinguish between the stable and destabilized fold of the WT and F508del variants , respectively [75] . We used antibodies specific for the first nucleotide binding domain ( NBD1: 18D1 ) and second nucleotide binding domain ( NBD2: M3A7 ) of CFTR , to assess the susceptibility of these domains to resist proteolysis . HSF1 silencing leads to a significant stabilization of both NBD1 and NBD2 , exhibiting a more pronounced stabilizing effect to that seen with temperature correction ( Figure 5E ) . It also led to the appearance of an approximately 25 kDa band in NBD1 , which has previously been described to represent the stable core fragment seen in WT-CFTR , but not in the F508del variant [75] . HSF1 silencing also restored the folding of the FLuc reporter to a level comparable to that seen in sip23-treated F508del-expressing cells and WT-expressing cells ( Figure 5F ) . Overall our results suggest that alleviation of the MSR by siHSF1 generates a more permissive cellular environment for productive folding , not only improving the CF phenotype but also that of other protein misfolding diseases by restoring a WT-like proteostasis environment . To understand the impact of HSF1 silencing on F508del-CFTR stability , we performed gene expression analysis . Here we found that HSF1 or p23 silencing leads to a reduction in the expression of HSF1-responsive genes , such as I-Hsp70 , HSPB1 , and I-Hsp40 . However , they had no effect on the transcription of CFTR itself ( Figure S6A ) , nor the expression levels of markers for other PN cellular pathways , including ubiquitin proteasomal system ( UPS ) , autophagy , and oxidative stress ( NRF2 pathway ) ( Figure S6B ) . In addition to the alleviation of the HSR , both siHSF1 and sip23 also decreased the expression of UPR-related genes ( Figure S6B ) . UPR but not the oxidative stress pathway was up-regulated in F508del-expressing cells in comparison with WT-expressing or CFTR−/− cells ( Figure S6C ) , suggesting a link between HSR and UPR activation , as previously described [60] . Finally , we used the proteasomal inhibitor MG132 and the autophagy inhibitor 3-methyladenine ( 3-MA ) to examine the impact of proteasome and autophagic pathways on the FLuc folding sensor . Whereas HS of F508del-expressing cells further reduced FLuc activity and folding from a level of 50% to 25% of that of WT-cells ( Figure S6D ) , neither MG132 nor 3-MA impacted FLuc folding in F508del-expressing cells . These results suggest that blocking proteasomal activity or autophagy is not sufficient to rescue FLuc folding in an environment already affected by the MSR . Given the impact of the MSR on the recovery of F508del function , we tested the effect of chemical inhibition of HSF1 in F508del-expressing CFBEs , using the previously characterized HSF1 inhibitor , triptolide [76] . Triptolide reduced the HS-induced up-regulation of I-Hsp70 and I-Hsp90 mRNA levels ( Figure S7A ) , confirming its ability to block HSF1 transactivation , consistent with previously published data [76] . Treatment of F508del-expressing cells with triptolide resulted in an increase in band-B stability as well as trafficking to band-C ( Figure 6A ) . It also restored cell surface channel activity shown by quenching of the halide sensing YFP-H148Q/I152L ( Figure 6B ) , to a level similar to that seen with VX809 ( Figure 6B ) . Since misfolding diseases present a chronic challenge to the cell , we next assessed the benefit a chronic dosing regimen of triptolide on correcting the F508del-CFTR trafficking defect . Chronic treatment resulted in a time-dependent increase in stabilization and trafficking of F508del-CFTR over the course of 96 h ( Figure 6C ) . The effect of triptolide was dependent on HSF1 , since combining triptolide and siHSF1 did not result in additivity for F508del stability , trafficking , and function ( Figure S7B ) , further supporting the conclusion that suppression of HSF1 hyper-activation promotes F508del correction . Since down-regulation of the MSR provides a favorable environment for protein folding and trafficking of F508del-CFTR , we re-assessed the potency of existing correctors of F508del-CFTR in combination with triptolide or siHSF1 . Treatment of F508del-expressing cells with VX809 or triptolide alone led to a moderate restoration of F508del-CFTR activity ( Figure 6B ) . In contrast , combining both drugs had a synergistic effect on F508del-CFTR trafficking and channel activity ( Figure 6D , 6E and Figure S7C ) . Similar results were also observed with siHSF1 in combination with VX809 and other CF correctors ( Figure S7D ) , showing that alleviation of the chronic stress improves the potency of clinically relevant correctors of F508del trafficking and function . We next examined the effect of triptolide treatment in patient-derived bronchial hBE cells homozygous for F508del . Treatment with triptolide resulted in a modest 1 . 4-fold increase in short-circuit current ( Isc ) relative to that seen with vehicle treatment ( Figure 7A ) . Maximal correction was obtained when triptolide was combined with VX809 , resulting in a 7-fold increase in Isc over the basal current ( Figure 7A , 7B ) , synergizing with the VX809 effect , which achieved a 3 . 5-fold increase in Isc . To address whether this effect was tissue specific , we also tested the effect of triptolide using primary CF intestinal organoids derived from two F508del CF patients [77] . In this assay , increased organoid swelling is indicative of restored F508del function . Although we did not observe any effect with triptolide alone , we did observe an approximate 50% increase in organoid swelling when VX809 was combined with triptolide as compared to that seen with VX809 alone , revealing a synergistic response in both CF patient codes ( Figure 7C , 7D ) , similar to that seen in primary hBE and CFBE cells . These results highlight the beneficial impact of MSR abrogation and its ability to improve the potency of existing therapeutics , consistent with our hypothesis that restoration of a WT-like folding environment could be a critical factor in managing human misfolding disease [5] , [9] .
Our results demonstrate that the long-term expression of disease-causing misfolded proteins can lead to an abnormal , chronic stress response that we now refer to as the maladaptive stress response ( MSR ) . This altered Q-state [3]–[6] , which emphasizes that the structure of a protein is tightly integrated with a dynamic proteostatic system [1] , [5] , [6] , [9] , negatively impacts the folding of disease-associated proteins , such as F508del-CFTR [42] , leading to a self-propagating proteotoxic crisis ( Figure 8 ) . We have found that targeting the MSR can significantly alleviate disease progression , thereby improving the disease phenotype in different disease models of protein folding . In CF , this is consistent with the view that folding of CFTR is a multi-step , vectorial process involving sequential folding intermediates that must be therapeutically managed for effective correction [42] , [78] , [79] . We now suggest that restoration of the native cellular proteostasis-state could represent a critical first line of therapeutic intervention to more effectively achieve the correct structure–function relationship necessary to restore cellular function . Our results show that the proteostatic biology of F508del-expressing cells is different than that seen in WT-expressing cells , characterized by a subacute increase in heat shock protein expression , reduced protein synthesis , and altered protein folding , phenomena contributing to the disease phenotype that we have referred to in the past as the chaperone trap [42] . These results are consistent with previous observations where elevated levels of heat shock proteins were observed in postmortem brain tissue of AD patients [80]–[83] , and in lung tissue of COPD patients [84] . Our proposed paradigm shift in how to address protein misfolding diseases leads us to suggest that , unlike the well-documented protective benefit of HSR activation to solve acute and transient protein misfolding problems ( see below ) [40] , the MSR is counterproductive when chronically activated , attempting to repeatedly manage a misfolding problem that it cannot solve . This condition thereby exacerbates the disease rather than relieving it , emphasizing the importance of first managing the disease from the perspective of proteostasis by mitigating the chronic folding stress problem . We propose that abrogation of the MSR , either by directly stabilizing the initiating misfolding intermediate [34] , [85] , [86] or , as suggested herein , through restoration of a WT-like Q-state [5] , [6] , [9] , could provide substantial benefit to counter the proteotoxic crisis found in chronic disease ( Figure 8 ) . It is becoming increasingly evident that there exists a fine balance between protection and toxicity in the function of the protein folding environment in eukaryotic cells [2] , [5] , [6] , [22] , [87] . On one hand , the beneficial impact of HSR activation in preventing proteotoxicity in worm and mouse models of HD and AD [88]–[90] and in promoting cell survival in the face of diverse stress insults has been well documented [8] , [9] , [27] . Additionally , HSF1 activators and overexpression of select chaperones have been shown to be neuroprotective [36] , [91]–[93] . However , the mechanism of action of such compounds and the chronic effect of HSF1 activation in vivo remain to be elucidated . Proteostasis regulators shown to activate HSF1 and to provide benefit in HD have also been shown to affect other stress pathways , including oxidative stress and UPR , which could contribute to disease management [36] . HSF1 overexpression has also been shown to exacerbate mutant Htt aggregation in a cellular model of HD [37] . On the other hand , Hsps are known to be actively involved in disease progression [80] , [82] , [83] . For example , in tau pathology , Hsp90 binding promotes tau misfolding and aggregation [94] , not unlike the chaperone trap state found in CF [42] , [78] , [79] , a result consistent with the dynamic state of the disordered tau protein and its interaction with Hsp90 in disease [95] . Moreover , chaperone balance is disrupted upon overexpression of polyQ aggregates through sequestration of low-level expression regulatory co-chaperones required for protein folding [96] . While Hsp90 inhibitors , which indirectly activate HSF1 , show promise in treating neurodegenerative diseases [97] , [98] , the beneficial effect was shown to be directly due to Hsp90 inhibition , which , in the case of tauopathies , reduces the functional cycling of kinases and thereby tau phosphorylation , minimizing its aggregation and toxicity [99] , [100] . Thus , while the mechanism of action of HSF1 activation is poorly understood , perhaps reflecting experimental conditions where a ‘brief’ burst of chaperones provides temporary relief to the misfolding problem , there is limited evidence in vivo that chronic activation of HSF1 provides long-term disease benefit . Indeed , proliferation of cancer cells is also dependent on a MSR characterized by sustained HSR activation and elevated levels of proteostatic components that sustain invasive survival [28] , [38] , a pathological condition leading to reduced human lifespan . The global proteotoxic crises that arise in protein misfolding diseases may be a consequence of an amplifying cascade of misfolding challenges as disease progresses , a view consistent with reports of reduced longevity in worms following chronic overexpression of misfolded proteins [35] , [90] , [101] , [102] . Alternatively , disease progression could reflect either the loss of proteostatic capacity associated with aging [4] , [8] , [21] , [73] , [103]–[105] or an overload of the cellular PN capacity . In the latter case , since Hsc/Hsp70 and Hsp90 represent at least 0 . 5% and 1% of total cellular protein , respectively , and cells exhibiting a MSR have reduced global protein synthesis , it is unlikely that the chaperone capacity per se is saturated , but this remains to be tested directly , given the complexity of the folding environment and lack of understanding of chaperone capacity in each cell type and/or disease environment . However , we have observed that the silencing of key proteostatic chaperones leads to a partial rescue of F508del-CFTR cell surface channel activity ( Figure S7E ) [106] , arguing against a possible overload of the chaperone capacity , at least in CF disease . Indeed , the reduced specific activity of the FLuc sensor suggests a significant challenge to the overall cellular folding environment , a result that is consistent with the recent observation that overexpression of the Hsp40/70 system decreases the fraction of protein that achieves a functional fold using activity-based profiling [11] . These observations underline the importance in understanding folding mis-management by the chronic MSR that exceeds a set-point defined by chaperone/co-chaperone balance normally required for a healthy cell . It is clear that this new principle of short-term acute versus long-term chronic proteostatic set-points now needs to be considered as an important contributor to the onset and progression of misfolding diseases such as CF , AATD , NPC1 , and AD . For example , the activity of FLuc , a sensor of the folding environment of the prevailing PN [45] , [46]–[50] , in cells chronically expressing the misfolded F508del-CFTR was reduced in response to elevated HSF1 activity , but restored to WT-levels upon MSR abrogation by siHSF1 , sip23 or , importantly , following removal of the misfolded F508del-CFTR . Here , we suggest that p23 , acting in concert with Hsp90 in protein folding and transcriptional activation of HSF1 , accentuates the activity of the chaperone trap components , engaging F508del in an inappropriate attempt to resolve progression along the folding pathway [42] . Consistent with this conclusion , we observed HSF1 phosphorylation and I-Hsp70 levels , in response to sip23 , reduced to the levels seen in WT-expressing cells , thereby restoring a WT-like PN that would be expected to be optimized for CFTR biogenesis and proteome function . While abrogation of the MSR by siHSF1 did not affect CFTR transcription , global protein synthesis , or other tested PN pathways ( UPS , autophagy , and oxidative stress ) , it specifically abrogated both the HSR and UPR activation , restoring function . It also improved folding and activity of the FLuc reporter sensor . Thus , we now suggest that early translation-linked events could be critical determinants of HSR , disease onset and/or progression promoting the MSR , a conclusion consistent with the increasing regulatory complexity of the HSR at the level of transcription [96] , [107] , [108] . Why does the HSR work acutely but trigger a maladaptive state when chronically active in misfolding disease , triggering MSR ? One possibility is that during evolution , the HSR pathway evolved strategies to manage long-term proteostasis states that are necessary for optimizing stemness [21] , [105] , [109] and/or direct long-term development , differentiation and multi-organ genesis , required for integrated organismal function , and to extend lifespan [24] , [110] . Such a finely tuned Q-state in higher eukaryotes may be less permissive to fluctuations in PN biology in response to inherited variants in human disease that become out of reach of the normal proteostasis buffering capacity , and therefore more prone to maladaptation [5] , [28] , [39] , [111] . Curiously , maladaptation not only includes the role of the HSF1-Hsp90 axis in supporting proliferation of cancer cells , a pathogenic state [28] , [38] , [39] , [112] , but also the propagation and resistance of viral pathogens to host defenses that can impact human health [113]–[115] . We would now propose maladaptation as a potent force in evolvability [21] , [105] , [109] , contributing to improved survival and fitness [5] , [6] , [18] , [116] , highlighting an important principle applicable to correction and increased survival in response to chronic human disease , perhaps through epigenetic mechanisms that , we now appreciate , play a central role in HSF1 management [107] , [108] and correction of human disease [28] , [33] , [56] . We now suggest that an appreciation of the impact of maladaptation on protein folding dynamics managed by the Q-state [4] , [5] could provide insight into how to effectively manage the vast array of chronic protein misfolding states affecting human disease [1] .
Human bronchial epithelial cells CFBE41o- stably expressing F508del-CFTR or WT-CFTR were cultured as previously described [56] . IB3 cells expressing WT-AAT or Z-AAT were cultured as previously described [33] . For all temperature-corrected experiments , F508del-CFTR expressing CFBE cells were transferred to 30°C for 24 h . Hela cells stably expressing WT or I1061T-NPC1 were cultured in Dulbecco's Modified Eagle Medium ( DMEM ) containing 10% ( v/v ) fetal bovine serum ( FBS ) , 2 mM L-glutamine , 3 µg/ml puromycin , and 600 µg/ml G418 . Primary fibroblasts derived from healthy donors ( WT ) or patients homozygous for the I1061T mutation of NPC1 were cultured in DMEM containing 10% ( v/v ) FBS , and 2 mM L-glutamine . Cells were obtained from Scott Randall at UNC Chapel Hill . Cells were plated on PureColl100 coated plates and grown in bronchial epithelial growth media ( BEGM ) + bullet kit ( Lonza ) + 1 µM all-trans retinoic acid with daily media changes until cells reached 90% confluence . Cells were harvested with Accutase at 37°C for 10 min and pelleted at 500 g for 5 min . Cells were re-suspended in BEGM and plated on human placental collagen coated 12 mm Costar snapwell filters ( Corning ) at a density of 5×105 cells/filter . Cells were grown in liquid/liquid culture for the first 96 h with daily media changes as previously described [117] on the apical ( 0 . 5 ml ) and basolateral ( 2 ml ) chambers . Cells were subsequently switched to air/liquid culture and basolateral media , changed every day for the first 7 days and three times a week for 4–6 weeks until differentiation was complete . Culture of organoids was performed as previously described [77] . Briefly , biopsies were washed with cold , complete chelation solution and incubated with 10 mM EDTA for 30 ( small intestine ) or 60 ( rectum ) min at 4°C . Crypts were isolated by centrifugation and embedded in Matrigel ( growth factor reduced , phenol free; BD bioscience ) and seeded ( 50–200 crypts per 50 µl Matrigel per well ) in 24-well plates . The Matrigel was polymerized for 10 min at 37°C and immersed in complete medium ( DMEM/F12 with penicillin and streptomycin , 10 mM HEPES , Glutamax , N2 , B27 [Invitrogen] , 1 µM N-acetylcysteine [Sigma] ) and the following growth factors: 50 ng/ml mouse epidermal growth factor ( mEGF ) , 50% Wnt3a-conditioned medium and 10% noggin-conditioned medium , 20% Rspo1-conditioned medium , 10 µM nicotinamide ( Sigma ) , 10 nM gastrin ( Sigma ) , 500 nM A83-01 ( Tocris ) , and 10 µM SB202190 ( Sigma ) . Medium was changed every 2–3 days . Organoids were passaged every 7–10 days , and passages 1–10 were used for confocal live-cell imaging . The gene coding for the eYFP fluorescent protein was fused at the C-terminus of the WT Firefly luciferase gene ( FLuc ) and cloned into the lentivirus vector , pLVX-Puro ( Clontech ) . CFBE cells stably expressing WT- or F508del-CFTR were infected with 5×106 PFU of pLVX-Puro-eYFP-FLuc lentivirus . Cells expressing eYFP-FLuc fusion protein were sorted by FACS to generate WT- or F508del-CFTR CFBE cell lines stably expressing eYFP-FLuc . siRNA transfections and preparation of cell lysates and Western blots was done as previously described [56] . For overexpression experiments , cells were plated at 70% confluency in a 12-well plate and transfected using 1 µg of DNA , 2 µl of P3000 per µg of DNA , and 1 . 5 µl of lipofectamine 3000 in Opti-MEM containing 5% FBS ( Life Technologies ) . Cells were washed and fed on the next day and lysed 48 h after transfection . qRT-PCR was performed using the iScript One-Step RT-PCR kit with SYBR green ( Bio-Rad ) . RNA was standardized by quantification of beta-glucuronidase ( GUS ) mRNA , and all values were expressed relative to GUS . Statistical analysis was performed on three independent technical replicates for each RNA sample , where error bars represent SD or SEM . For each immunoprecipitation ( IP ) , 1 mg of total protein was used . CFTR IP was performed as previously described [118] . For HSF1 IP , cells were lysed in 20 mM Tris-HCl pH 7 . 4 , 130 mM NaCl , 10 mM Na2MoO4 , 1 mM EDTA , 5 µM ATP , 0 . 5% NP-40 , and 2 mg/ml of complete protease inhibitor cocktail . Lysates were incubated with 3 µl of HSF1 antibody ( Abcam , ab52757 ) for 18 h , and complexes were recovered with 30 µl of γ-bind beads incubated at 4°C for 90 min . The beads were washed three times with lysis buffer and eluted with 10% SDS and 20% Tris-HCl pH 6 . 8 . For total protein synthesis , cells were starved in methionine-free MEM ( Sigma ) for 30 min and subsequently pulse labeled for 1 h with 35S-methionine ( 0 . 1 mCi per well in a 6-well plate ) . Lysates were loaded in a 4%–20% gradient gel , with the amount of lysate normalized for number of cells in each condition . CFTR or HSF1 processing efficiency was measured by pulse-chase . Analysis of CFTR stability by pulse-chase was performed as previously described [56] . For HSF1 pulse-chase , cells were starved in methionine-free MEM ( Sigma ) for 30 min , pulse labeled for 4 h with 35S-methionine ( 0 . 1 mCi per well in a 6-well plate ) , and chased for a total of 24 h . Cells were lysed and HSF1 IP performed as described above . The recovered radiolabeled proteins were then visualized by autoradiography . CFBE cells were seeded in 60 mm dishes at a density of 4×105 one day prior to transfection . Iodide efflux assay was performed as previously described [119] . CFBE41o- cells stably expressing the halide sensitive YFP-H148Q/I152L [120] ( CFBE-YFP ) , were dosed with compounds 24 h before the YFP-assay , which was performed as previously described [40] . Primary human bronchial epithelial ( hBE ) cells were dosed every 24 h for a total of 96 h with the indicated concentration of DMSO , VX809 , or triptolide . Cells were mounted in modified Ussing chambers , and the cultures were continuously short-circuited with an automatic voltage clamp . Transepithelial resistance , RT , was measured periodically from the current required to apply a 2 . 5 mV bipolar voltage pulse . RT was calculated from Ohm's law . The basolateral bathing Ringer solution was composed of ( 137 mM NaCl , 4 mM KCl , 1 . 8 mM CaCl2 , 1 mM MgCl2 , 10 mM HEPES , and 10 mM glucose ) . NaCl concentration of the apical bathing solution was reduced by replacing NaCl with equimolar Na-gluconate . The chambers were maintained at 37°C and gassed continuously with a mixture of 95% O2 , 5% CO2 . Sodium currents were blocked by addition of the sodium channel blocker amiloride ( 10 µM ) to the apical solution . Subsequently , the cAMP agonist , forskolin ( 10 µM; both chambers ) , the CFTR potentiator genistein ( 50 µM; apically ) , and the CFTR channel blocker CFTRInh-172 ( 10 µM; apically ) were added sequentially to determine cAMP-stimulated CFTR currents . Organoids from a 7-day-old culture ( 20–80 organoids ) were seeded in a 96-well plate ( Nunc ) in 5 µl Matrigel and 100 µl of medium [77] . One day after seeding , organoids were incubated with 100 µl of medium containing 10 µM calcein green ( Invitrogen ) for 60 min . Then 5 µM forskolin was added , and organoids were directly analyzed by confocal live-cell microscopy ( LSM710 , Zeiss , ×5 objective ) . Three wells were analyzed per condition , and up to 60 wells per experiment . Organoids were pre-incubated for 24 h with 3 µM VX809 , 25 nM triptolide , or a combination of both . For CFTR potentiation , 3 µM VX770 was added with forskolin . Organoid surface area was automatically quantified using Volocity imaging software ( Improvision ) . The total organoid surface ( XY plane ) increase relative to that at T = 0 of stimulus was calculated and averaged from two individual wells per condition . Results are shown as mean ± SD , and p value determined by two-tailed t-test using DMSO as a control reference . HSF1 cross-linking to monitor HSF1 trimerization status was performed at room temperature with 1 mM final concentration of disuccinimidyl suberate ( DSS ) for 30 min with gentle mixing , and quenched by addition of 50 mM Tris-HCl pH 7 . 5 for 15 min . Prior to the luciferase ( Luc ) assay , cells were lysed and 15 µg of total protein loaded on 8% SDS-PAGE gel to perform immunoblots for Luciferase and actin control to assess Luc expression level . Immunoblots were quantified to ensure that the same amount of Luc was analyzed in the activity assay for each sample . 20 µg of Luc was incubated with Steady-Glo luciferase assay reagent ( Promega ) for 5 min , and luminescence was read at 562 nm to measure Luc activity . All results are presented as specific FLuc activity , which represents FLuc activity normalized to the amount of FLuc expressed in each condition . Three hours before measurement of AAT secretion kinetics , cells were washed with PBS and incubated with 350 µl ( 12-well plate ) of FBS-free culture medium . After the 3 h incubation , cells were harvested , and the corresponding media centrifuged at 1500 rpm for 30 min at 4°C to separate cells and medium . After lysis , AAT immature and mature forms in the lysate or secreted into the culture media were analyzed by SDS-PAGE or Native gel for analysis of AAT polymer formation . For native gel electrophoresis , 25 µg of protein in the lysate or 30 µl of cell media was separated on a 3%–20% native gel according to the manufacturer's instructions ( Expedeon Inc ) . Loading of the media was normalized to protein concentration in the lysate for each sample . Native gels were transferred and probed for AAT using the anti-AAT antibody ( Immunology Consultants Laboratory ) . AD mice , referred to as the AβPP Tg mice model , express the hAPP751 cDNA containing the London ( V717I ) and Swedish ( K670M/N671L ) mutations under the regulatory control of the murine ( m ) Thy-1 gene ( mThy1-hAPP751 ) . Mice were generated as previously described [121] . For this study , the APP line 41 mice ( C57/BI6 ) were utilized , as they produce high levels of Aβ42 and develop synaptic damage and memory deficits . Young ( approximately 4 mo old ) , middle aged ( approximately 9 mo old ) , and old ( approximately 16 mo old ) , WT and AD mice pairs were humanely killed , and tissue was frozen for analysis . Posterior half of mouse hemibrains were homogenized in 500 µl of PDGF buffer ( 1 mM HEPES , 5 mM Benzamidine , 2 mM 2-Mercaptoethanol , 3 mM EDTA , 0 . 5 mM Magnesium Sulfate , 0 . 05% Sodium Azide , 2 mg/ml Protease Inhibitor cocktail [Roche] , and 1 tablet of PhosSTOP phosphatase Inhibitor cocktail [Roche] per 10 ml of buffer , pH 8 . 8 ) , using a tissue homogenizer . Samples were spun at 5 , 000 g for 5 min at 4°C , and the supernatant centrifuged at 100 , 000 rpm for 1 h at 4°C to separate the cytosolic and particulated fractions . Pellets were resuspended in 150 µl of PDGF buffer and homogenized by sonication ( 10% for 10 s ) . Protein concentration was determined by Bradford , and 20 µg protein from cytosolic fractions were loaded in SDS-PAGE for immunoblotting . To detect Aβ monomer and multimers , 40 µg particulate fractions of brain homogenates were loaded in a 4%–12% bis-tris gel and immunoblots were incubated with 6E10 Aβ specific antibody ( Covance ) . CFBE41o- cells expressing WT- or F508del-CFTR at the indicated treatment were lysed for 30 min at 4°C with lysis buffer ( 50 mM Tris-HCl pH 7 . 4 , 150 mM NaCl , 1% Triton X-100 , 2 mg/ml Protease Inhibitor cocktail [Roche] ) , and harvested at 20 , 000 g for 20 min at 4°C . Total protein concentration of pre-cleared lysates was determined by Bradford . Proteolysis was performed by incubating 80 µg of total protein with increasing concentration of Trypsin in PBS ( 0 . 01–0 . 25 mg/ml ) at 4°C for 15 min . Proteolysis was stopped by adding 1 mM of PMSF and 6x SDS-PAGE sample buffer . Samples were equally divided and loaded onto two 12% SDS-PAGE for separation of the proteolytic fragments and probed with CFTR antibodies for NBD1 ( 18D1: epitope 536-545 ) and NBD2 ( M3A7 ) . Hela cells expressing WT- or I1061T-NPC1 were transfected for 72 h , lysed in RIPA buffer ( 10 mM Tris-HCl pH 8 . 0 , 140 mM NaCl , 1 mM EDTA , 1% NP-40 , 0 . 1% SDS , 0 . 1% Na-deoxycholate , 2 mg/ml Protease Inhibitor cocktail [Roche] ) , and harvested at 20 , 000 g for 15 min at 4°C . NPC1 was immunoprecipitated using 400 µg of total protein and 2 µg of NPC1 antibody , for 18 h at 4°C . Complexes were recovered with 40 µl of γ-bind beads incubated at 4°C for 2 h . The beads were washed two times with lysis buffer , and one time with PBS , and eluted with 36 µl of denaturing buffer ( NEB ) for 10 min at 90°C . Elutions were divided in two tubes , one without and other with 1 µl of endo-H enzyme , and incubate for 1 h at 37°C . Samples were run on 4%–20% gradient gel and immunoblotted for NPC1 . The data represents densitometric analysis of immunoblots using an Alpha Innotech Fluorochem SP . The error bars represent the SEM ( n≥3 ) or the SD of the mean . In all panels asterisks indicate a p-value <0 . 05 as determined by a two-tailed t-test using the control as the reference .
|
The function of all proteins is dependent on achieving the correct folded state , a process referred to as protein homeostasis or proteostasis . Cellular proteostasis is maintained by diverse signaling pathways , including the heat shock response ( HSR ) , which protects proteins in the face of acute stress . However , genetic disorders are a challenge to cells , since the mutated protein will often fail to fold properly and function correctly . We have discovered that the chronic expression of such disease-causing proteins can trigger the sustained activation of the HSR in a failed attempt to correct the associated misfolding defect . Such chronic HSR activation presents an unanticipated challenge to the cell by initiating a sustained state of stress management , which leads to a general protein-folding deficiency . This in turn further exacerbates the disease phenotype—a condition we have termed maladaptive . We show that down-regulation of this maladaptive stress response ( MSR ) restores cellular protein folding and improves the disease condition in loss-of-function disorders such as cystic fibrosis , Niemann-Pick disease and alpha-1-antitrypsin deficiency , as well as gain-of-toxic-function diseases such as Alzheimer's disease . MSR management therefore potentially represents an important therapeutic first step in regulating the progression of human disease associated with chronic protein misfolding .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"heat",
"shock",
"response",
"cellular",
"stress",
"responses",
"medicine",
"and",
"health",
"sciences",
"cell",
"biology",
"proteins",
"protein",
"structure",
"biology",
"and",
"life",
"sciences",
"cell",
"processes",
"cystic",
"fibrosis",
"transmembrane",
"conductance",
"regulator",
"molecular",
"cell",
"biology",
"protein",
"misfolding"
] |
2014
|
Modulation of the Maladaptive Stress Response to Manage Diseases of Protein Folding
|
The development of a functional nervous system requires tight control of neurite growth and guidance by extracellular chemical cues . Neurite growth is astonishingly sensitive to shallow concentration gradients , but a widely observed feature of both growth and guidance regulation , with important consequences for development and regeneration , is that both are only elicited over the same relatively narrow range of concentrations . Here we show that all these phenomena can be explained within one theoretical framework . We first test long-standing explanations for the suppression of the trophic effects of nerve growth factor at high concentrations , and find they are contradicted by experiment . Instead we propose a new hypothesis involving inhibitory signalling among the cell bodies , and then extend this hypothesis to show how both growth and guidance can be understood in terms of a common underlying signalling mechanism . This new model for the first time unifies several key features of neurite growth regulation , quantitatively explains many aspects of experimental data , and makes new predictions about unknown details of developmental signalling .
The development of the nervous system requires the precise wiring of billions of cells . To achieve this considerable feat , growing axons navigate over long distances to reach their synaptic targets , and hence establish appropriate patterns of connectivity . Dysregulation of this process contributes to developmental and neurological disorders [1 , 2] , and the inability to recapitulate early growth events hinders nerve injury repair [3] . One major regulator of axonal growth ( trophism ) and guidance ( tropism ) is signalling by extracellular chemical cues . A large number of these are known , with nerve growth factor ( NGF ) being perhaps the best studied [4–8] . The effects of NGF on growth and guidance are exerted via tight control of signalling along the axon for extension and turning , and at the cell-body to coordinate synthesis and supply of raw materials . As might be expected , neither growth nor guidance occurs at very low concentrations . More puzzling is that they are also both inhibited at higher concentrations , exhibiting a biphasic dose response that peaks in an intermediate concentration regime [9–18] . Such tight constraints on both growth and guidance are important since they increase the challenge for therapeutically effective interventions . Existing theories of the biphasic effects of NGF on growth depend on mechanisms intrinsic to single cells [16 , 19–21] , or a collective effect in which growth within aggregates of cells is hindered by increased fasciculation ( the grouping of neurites into bundled fibres ) [15] . However , these theories remain purely qualitative and lack thorough experimental tests . Two fundamental mechanisms underlie guidance: turning and differential growth . Turning largely occurs for steep gradients . The decline in its sensitivity with overall concentration can therefore be explained by saturation of finite numbers of receptors [18] . However , for shallow gradients , ( for NGF , < 1% concentration change per 10 μm ) guidance is remarkably sensitive , and depends on preferential growth towards higher concentrations [17 , 22] . In this paradigm of chemotactic response , neurites do not exhibit biased turning , but grow more quickly or more slowly when extending up or down a gradient , respectively . The biphasic effects of NGF on guidance by differential growth therefore inherit the puzzle of those on growth in general . Intriguingly , however , even in the decreasing portion of the dose response curve for growth , where lower concentrations should generate more growth than higher concentrations , differential growth remains biased up the gradient [22] . Thus , guidance involves a true detection of the gradient , yet as it is also inextricably linked to growth , this tropic response cannot be fully understood in isolation . To address these issues we first test previous proposals for NGF growth inhibition at high concentrations , and find they do not explain the biphasic response of dorsal root ganglia ( DRG ) explants in collagen gels . Second , inspired by a reanalysis of the extensive shallow gradient data set of ref . [18] , we propose a novel signal transduction mechanism which resolves the apparent contradictions introduced above . In this , the growth at the neurite tip is promoted by the local concentration of NGF , but , critically , is also inhibited by a somatically-computed signal that results from NGF-dependent signalling among the collection of cell bodies . Anterograde transport from the soma modulates growth by supply of signalling components , whereas retrograde transport from the tip provides the soma with information about the distal concentration . The inhibition implements the decrease in growth at high concentrations , and provides the normalisation that allows differential growth to be a suitably sensitive guidance mechanism . The model represents a new signalling paradigm for understanding nervous system development and repair , and makes testable predictions applicable to NGF and other growth and guidance cues .
There are two main hypotheses as to how high concentrations of NGF inhibit growth . Based on experiments on chick thoracic DRGs , ref . [15] proposed that an NGF-dependent increase in neurite fasciculation hinders growth from aggregates of cells , and reported no effect in single cells . Conversely , others have proposed that inhibition acts at the single-cell level , such as by TrkA receptor saturation or downregulation [16 , 19 , 21] or signalling via the low affinity receptor p75 [20] . To test these competing theories , we measured the NGF response of early postnatal rat DRG explants and dissociated cells , grown together at low density in collagen gels for 48 h . Previous studies have quantified explant outgrowth by manual measurements of radial extension [15 , 16 , 19] , semi-automated measurement of area and density [17 , 18 , 22 , 23] , and least-squares fitting of ellipses to explant shape [24] . However , manual measurements become impractical for large data sets , and previous semi-automated methods provide only coarse descriptions of collective outgrowth that are difficult to interpret at a lower level . To gain a more detailed understanding of growth patterns , we instead used a Fourier decomposition that is capable , in principle , of capturing arbitrary patterns of neurite extension , and is directly related to specific features of the response . Briefly , we fitted boundary curves to the central cell-body region of the explant and to the outer limit of neurite outgrowth , and parameterised the distance between the curves with an angular variable . The Fourier coefficients of this radial outgrowth function quantified the average radial outgrowth a0 ( average distance between explant body and limit of neurite extension ) , outgrowth bias in orthogonal image axes a1 , b1 , and other higher-order features ( Methods ) . The average radial outgrowth of the explants exhibited the expected biphasic response curve ( Fig 1A–1C ) . Explant outgrowth peaked at 0 . 3nM NGF concentration ( mean 817 μm ) , and was comparatively reduced ( mean 491 μm ) at 10 nM ( p = 1 × 10−4 , Mann-Whitney U-test for difference between 0 . 3 nM and 10 nM conditions , n = 15 explants per condition , 8 animals from 4 separate experiments ) . By contrast , recording the length of the longest neurite of each dissociated cell ( Fig 1D–1F ) , we observed no evidence of growth inhibition at high NGF concentrations ( p = 0 . 35 , Mann-Whitney U-test for difference between 0 . 3 nM and 10 nM conditions , n = 126 and n = 156 cells respectively ) . Comparing dissociated cells in the 0 . 1 nM condition , which exhibited the highest median neurite length , with the 10 nM condition gave a similar result ( p = 0 . 2 , Mann-Whitney U-test , n = 71 cells for 0 . 1 nM ) . The results of both comparisons were robust to the removal of outliers ( defined as values lying more than 1 . 5 times the interquartile range above the third quartile; Fig 1F ) . To test whether the lack of observed effect in dissociated cells may be due to a growth latency caused by the dissociation procedure , we repeated the experiment with an extended period of 96 h total growth . Consistent with the 48 h results , we observed a pronounced difference in explant outgrowth after 96 h , but no detectable difference in dissociated-cell neurite length distributions ( S1 Fig ) . Thus , inhibition of growth at high NGF concentrations is a property of intact ganglia , and not of isolated single cells . To assess the effect of NGF on fasciculation ( cf . ref [15] ) , we acquired higher resolution images of explants in the 0 . 3 nM and 10 nM conditions , and performed an automated image analysis to compute distributions of neurite bundle widths ( Fig 1G–1I ) . As a positive control , we tested the ability of our method to discriminate distributions from sample patches containing mainly thick or thin bundles as judged by eye ( taken from both NGF conditions , see S2 Fig for examples ) . The difference between control samples , corresponding to a 1 − 2 μm increase in bundle widths , was easily detected ( p = 5 × 10−30 , one-tailed Mann-Whitney U-test , n = 353 segments and n = 414 segments for thick and thin respectively ) . Applying the automated analysis to the test conditions , we found no detectable increase in bundle widths at 10 nM compared with 0 . 3 nM ( p = 1 , one-tailed Mann-Whitney U-test , n = 1687 segments and n = 2546 segments for 10 nM and 0 . 3 nM respectively , from 8 explants each ) . Thus , in our system , increased fasciculation does not explain the biphasic NGF response . This suggests the correlation between fasciculation and growth inhibition observed by ref . [15] is not a causal relationship , nor is it a general property of NGF-dependent growth regulation . As the results of our experiments contradict previous suggestions [15 , 16 , 19–21] , we propose an alternate mechanism for neurite growth regulation . A fundamental difference between an intact ganglion and a single dissociated cell is the central mass of neuronal cell bodies and support cells that comprise the ganglion body . This suggests the possibility that an NGF-dependent signal within the collection of cell bodies plays an inhibitory role in ganglion outgrowth . This may be mediated , for instance , by cell-cell interactions in which a paracrine factor is secreted from the cell bodies and inhibits neighbouring cells ( as shown for regulation of cell survival [25] ) . Another possibility is that satellite glial cells within the ganglion , which also carry NGF receptors [26] , communicate an inhibitory signal to the neural cell bodies that they ensheathe . By analogy with other systems [27 , 28] , we further propose that this inhibitory mechanism permits sensitive gradient detection , and may thus explain the guidance by differential growth observed by ref . [22]—there termed guidance by growth rate modulation . We develop this hypothesis with a mathematical model , and thus build a quantitative and predictive description of inhibitory growth and guidance signalling . To constrain the model , we first applied our explant image analysis to the NGF gradient data set of ref . [18] , which documents the growth of DRG explants after 48 h in very shallow NGF gradients in collagen gels . The data set comprises 3460 images of explants in which the ganglion body region had been manually segmented from the neurite region . Gradient parameters in the experiments varied between 0 − 0 . 3% concentration change per 10 μm , and background concentrations ≈ 0 . 001 − 100 nM , making it the most extensive record of NGF growth regulation yet compiled . An example image of an explant is shown in Fig 2A , here displaying a pronounced bias in neurite outgrowth in the direction of increasing NGF concentration . Applying the image analysis to the data , only two Fourier coefficients , a0 and b1 , varied systematically with the gradient parameters . We thus quantified the growth response by the average radial outgrowth a0 , and directional bias b1/a0 , which gives the fractional increase in neurite extension on the up-gradient side of the explant , relative to the average ( or fractional decrease on the down-gradient side ) . Explants with less than 100 μm average radial outgrowth ( 125/3460; 4% ) were excluded from the analysis of directional bias . Consistent with the results of ref . [18] , and Fig 1C , the average radial outgrowth exhibited a biphasic NGF dependence , with a peak at 0 . 3 nM and inhibition at higher concentrations ( Fig 2B ) . Outgrowth was biased in the direction of the gradient for background concentrations between 0 . 01 − 1 nM ( Fig 2C , S1 Table ) . By contrast , analysing the explant-body boundary curves alone , we found that the average explant body boundary was well-approximated by a circle of radius RE = 300 μm . We found no correlations in shape properties within or between the outgrowth and explant body regions ( S2 Table ) . The peak directional bias was observed at ≈ 0 . 1 nM in a 0 . 3% gradient , in which neurites facing directly up the gradient extended ≈ 15% further than the average over the explant . With these gradient parameters and the measurements in Fig 2B , this implies a ≈ 100 μm increase in growth has resulted from a maximum concentration difference of only 0 . 03 nM across the full length of the neurites . Observed over two orders of magnitude of background concentrations , this remarkable response involves a tropic growth modulation [22] , and places a strong constraint on any proposed mechanism . We construct a model that assumes an inhibitory NGF-dependent signal within the ganglion , integrated with known signalling components of NGF growth promotion , and that satisfies the constraints of the experimental data . Our approach is motivated by the seminal signalling models of ref . [28] , who demonstrated that a network architecture that combines competing activating and inhibitory pathways with upstream signal amplification is sufficient to explain perfect adaptation and high sensitivity in amoebae and neutrophils . We find that , in a different region of parameter space , a similar network structure also embodies the minimal ingredients required to explain growth and gradient sensing in our system . We begin by treating a neurite as a single well-mixed compartment that receives two NGF-dependent inputs . One input is assumed to be transduced by receptors at the growth cone , and the other by the proposed inhibitory signal at the cell body . Although in reality this system is more complex , involving , for instance , transport along an extending neurite , our first objective was to determine sufficient processes by which the two primary inputs can be integrated to explain the data . With these simplifications , we initially construct a model that accounts for the biphasic NGF dependence of ganglion outgrowth , but is unable to simultaneously satisfy the requirements of gradient detection . We then construct a signalling network for gradient detection that sensitively compares two concentrations , independent of their magnitude and associated signal saturation . Finally , we couple these two modules together in a two-compartment model , in which we explicitly include transport of signalling components between growth cone and cell body . Simulating this network in neurites extending in a gradient , we account for all competing demands of the experimental data . We consider activating ( A ) and inhibitory ( I ) signals that interact within a cell to regulate the conversion of a substrate G to a growth promoting active form G* ( Fig 3A ) . We construct the activating pathway as a coarse-grained representation of known NGF/TrkA receptor signalling . The signal A represents the binding occupancy of TrkA receptors at the growth cone , which drives growth in response to the local NGF concentration c1 . We assume saturable binding , such that the steady-state of A is given by the standard expression A ¯ ( c 1 ) = A T c 1 c 1 + K A , ( 1 ) with AT the total number of receptors on the growth cone and KA the dissociation constant . Consistent with measured values of ∼ 0 . 01 − 1 nM [29–31] , we fix KA = 0 . 1 nM , and use an order of magnitude estimate of AT = 1000 total receptors . For the proposed inhibitory pathway , we do not model possible processes of secretion or cell-cell communication explicitly . Working under the assumption that these are short-range effects between neighbouring cells , we simply model the signal I as responding to the local NGF concentration c2 at the cell body within the ganglion . As , at some stage , this must be transduced by receptor binding , we coarse-grain this signal into a similar form to that of A , I ¯ ( c 2 ) = I T c 2 c 2 + K I . ( 2 ) We assume for simplicity that I has the same maximal intensity as A , achieved straightforwardly in the model by setting IT = AT , and leave KI as a free parameter . We return to discuss the interpretation of this signal in Discussion . Integration of the activating and inhibitory signals is modelled as a simple push-pull reaction . The substrate is produced in the inactivated form at a constant rate , activated in proportion to A and inactivated in proportion to I , and decays exponentially in either form . The output of the model is the concentration of protein in the active form G* , which we assume acts linearly to control neurite extension . We provide the governing differential equations and parameters of the model in Methods . We compute the steady state of the network under the assumption that the input concentrations remain fixed ( ignoring for now the change in growth-cone concentration during neurite extension in a gradient ) . Expressed in terms of the signals of Eqs ( 1 ) and ( 2 ) , the steady-state output is given by G*¯=k0A¯k1+k2A¯+I¯ , ( 3 ) where the constants k0 , k1 and k2 are combinations of rate parameters . When c1 = c2 , fitting of parameters ki , along with KI from Eq ( 2 ) , yields a response in good agreement with the NGF dependence of explant outgrowth ( Fig 3B ) . Setting c2 = 0 , to remove the influence of inhibitory signalling , gives a simple model for dissociated cell growth that exhibits the saturating response of Fig 1F . Thus , the mechanism we propose , expressed as a very simple model , quantitatively accounts for the results of our experiments . By itself , however , this model lacks the gradient sensitivity implied by the experiments of ref . [18] . We illustrate this in Fig 3B , in which we plot the response of the constrained model with a 50% asymmetry in inputs ( c1 = 1 . 5c2; red line ) . Although this is nearly double the maximum value experienced by neurites in the experiments of ref . [18] , only a modest increase in response compared to the uniform condition is observed . The obstruction to gradient sensitivity in Model 1 is the saturable form of the signals A and I , combined with the parameter requirements of a biphasic growth response . Within the framework of the model , sensitive gradient sensing requires a comparison of A and I while both are in the linear regime with respect to concentration . In this case , A/I ≈ c1/c2 , and a highly effective gradient detector can be constructed . Indeed , a central assumption of the models of ref . [28] is that both activating and inhibitory signals are far from saturation . Here , however , a biphasic steady-state response ( Fig 3B ) requires that activation occurs at much lower concentrations than inhibition . This imposes the necessary condition that KA ≪ KI , precluding a direct comparison in respective linear regimes when the difference in input concentrations is small . Independent of the model , TrkA activation by NGF is indeed saturable [29–31] , yet remarkable gradient sensitivity was observed experimentally over two orders of magnitude of concentration ( Fig 2C ) . Thus , both the model and experimental data point to a mechanism for gradient detection that operates with high sensitivity , despite receptor saturation . How can the effects of signal saturation be overcome ? A common theoretical assumption is that cells can perform the necessary computations to invert expressions such as ( 1 ) and ( 2 ) , and thus access the original input variables [32–36] . In this way , a system that depends on the ratio of inputs could be constructed by forming the expression K A A ¯ ( I T - I ¯ ) K I I ¯ ( A T - A ¯ ) = c 1 c 2 , ( 4 ) providing a possible means for sensitive gradient detection . However , to the best of our knowledge , no biologically realisable implementation of this operation has been derived . To make our hypothesis concrete , we construct a network that performs the algebraic manipulations required of Eq ( 4 ) ( motivated by ref . [37] ) , and thus present an explicit gradient sensing mechanism . To do so requires only minor modifications of Model 1 , sharing both the input pathways and basic network structure ( Fig 3C ) . A dual negative regulation , induced by interaction between A and I , provides the necessary processing to unpack both saturating signals simultaneously . In this network , the activating and inhibitory pathways are integrated indirectly through downstream effector molecules X and Y . The two effectors enzymatically convert a target protein between an inactive F and active form F* . Inhibitors ZX and ZY are produced through upstream interaction between A and I , and thus degrade the effectors in proportion to the product AI . Intuitively , this can be understood as a form of mutual inhibition where A acts to suppress the activity of I , when I is present , and vice versa . We provide the governing differential equations and parameters of the model in Methods . At steady-state , and assuming that degradation of ZX and ZY is much slower than the inhibitory reactions , X ¯ ≈ k 3 A ¯ ( k 4 - I ¯ ) and Y ¯ ≈ k 5 I ¯ ( k 6 - A ¯ ) , ( 5 ) where the ki are combinations of rate parameters , and equality holds when the rates of inhibitor degradation go to zero . Assuming general Michaelis-Menten kinetics for activation and inactivation of F , the steady-state output of the network F * ¯ is given by the Goldbeter-Koshland function [38] , which depends only on X ¯ and Y ¯ through the ratio X ¯ / Y ¯ . Thus , with appropriate choice of parameters ki , by comparison with Eq ( 4 ) , the network output is a function of the concentration gradient , and independent of background concentration and associated receptor saturation . Moreover , if the enzyme kinetics are assumed to operate in the zero-order regime , the network can be made arbitrarily sensitive to small differences in concentrations , while remaining bounded in the case that c2 = 0 . Functionally , this is approximately equivalent to the response , F*¯ ( A ( c1 ) , I ( c2 ) ) ∼ ( c1/c2 ) h1+ ( c1/c2 ) h , ( 6 ) with tunable Hill coefficient h . To perfectly extract the gradient signal through this network requires an appropriate choice of the parameters that appear in Eq ( 5 ) , such that k4 = IT , k6 = AT and k3/k5 = KA/KI . To test the robustness of the output to changes in these values we computed the steady-state with random perturbations to parameters . For individual pairs of inputs ( c1 , c2 ) spanning 0 . 001 − 100 nM , a 10% multiplicative , uniformly distributed noise term η ∼ U ( 0 . 9 , 1 . 1 ) was applied independently to each parameter in Eq ( 5 ) . We performed this procedure 100 times for each pair of concentrations , and computed the average output over trials 〈 F * ¯ ( A ( c 1 ) , I ( c 2 ) ) 〉 ( Fig 3D ) , representing the average response of a collection of neurites with some cell-cell variability in intrinsic parameters . For concentrations between 0 . 001 − 1 nM , a sharp separation persists between up-gradient ( c1 > c2 ) and down-gradient ( c1 < c2 ) conditions , with an eventual loss of discriminability at higher concentrations . Further simulations revealed k6 to be the most sensitive parameter , as keeping this value fixed extended the sharp boundary to concentrations up to 100 nM . We have shown how an inhibitory cell-body signal can be integrated with TrkA activation at the growth cone to produce two distinct outcomes . The push-pull network of Model 1 reproduces the biphasic ganglion outgrowth response , whereas the dual negative regulation of Model 2 yields an adaptive sensor that is highly sensitive to small differences in input concentrations . We now couple these motifs together to produce a model of NGF signalling that quantitatively accounts for the experiments of refs . [18 , 22] . For the coupled system , we model a neurite as two well-mixed compartments , representing the growth cone and cell body . The network of Model 1 is localised to the growth cone , whereas the network of Model 2 is localised to the cell body ( Fig 4 ) . Communication between compartments occurs via retrograde transport of activated receptors A to form a cell body population Ac , and anterograde transport of the inhibitory signal I to produce a copy at the growth cone Ig . Similar to the signal amplification by substrate supply of ref . [28] , when the output of the cell-body compartment F* regulates the synthesis and transport of growth cone substrate G , neurites extend preferentially in the direction of a gradient . We tested the model against the gradient data set by simulating 48 h ganglion outgrowth in shallow exponential gradients ( Methods ) . Fitting the parameters of the model ( Table 1 ) yielded good agreement with the data over all concentration and gradient conditions tested . The average radial outgrowth of simulated explants follows the characteristic biphasic NGF dependence , and is independent of the gradient steepness ( Fig 5A ) . The directional bias of simulated outgrowth also closely matches that of the experimental data , exhibiting a large asymmetry in outgrowth for background concentrations of 0 . 01 − 1 nM ( Fig 5B ) . Thus , for the first time , we have provided a mechanistic and quantitative explanation of these two fundamental features of neurite growth control . Modular processing , coupled by protein transport and supply , permits powerful integration of antagonistic signals and fine tuning of collective growth . There are two key structural features of the model that contribute to the NGF response . The first is our central hypothesis that NGF-dependent inhibition arises from signalling within the ganglion body . The second is that detection of shallow gradients occurs via transport and comparison of signals between the growth cone and cell body , thus maximising the concentration differences being sensed . We describe two experiments which could test these claims , and simulate the model to predict the observable response . The role of inhibitory signalling within the ganglion can be tested by growing explants in compartmentalised chambers that separate neurite and cell-body regions , similar to the assay of ref . [39] . With a fixed NGF concentration at the cell bodies , the model predicts an absence of growth inhibition when high concentrations are applied to the distal neurites ( Fig 5C ) . Moreover , due to the sensitivity of gradient detection , the model predicts a switch-like transition to this regime . In the absence of gradient sensing ( setting the output of this component of the model F* to a constant ) , the transition is less steep , but the predicted outgrowth remains uninhibited as the distal neurite concentration is increased ( Fig 5C ) . As NGF/TrkA retrograde transport is slow compared with receptor binding and the rate of neurite growth [40] , the necessity of this mechanism for gradient detection can be tested with temporal manipulations . Fig 5D shows a simulation of the model in a uniform 0 . 1 nM concentration which was transiently increased to 0 . 2 nM with a 500 min pulse of NGF . Because there is a time delay in transporting newly bound receptors from the growth cone , the concentration at the cell-body is initially perceived as higher , creating an artificial negative gradient . The model therefore predicts , counterintuitively , that a uniform concentration increase will transiently decrease the rate of neurite outgrowth ( Fig 5D ) . Similarly , a concentration decrease is predicted to have the opposite effect . The predicted out of phase response is robust to the precise temporal regulation of NGF , requiring mainly that the timescale is equivalent to that of retrograde transport , or slower ( ≈400min or more ) .
What is the molecular basis of the mechanism we describe ? As growth inhibition was observed only for explants , but not for dissociated cells , our experiments provide strong evidence against any pathway in which inhibitory effects are mediated by direct binding of NGF to cell-surface receptors . This includes the proposal of ref . [16] that excess NGF stabilises the population of TrkA receptors in an inactive configuration , as well as the suggestion that TrkA receptors undergo activity-dependent down regulation analogous to chemotactic receptors in leukocytes [21] . Similarly , the lack of effect in dissociated cells precludes other possible hypotheses such as direct inhibition by the low-affinity NGF receptor p75 , or toxicity from overstimulation of downstream TrkA pathways . This led us to propose the involvement of paracrine signalling within the ganglion by NGF-dependent secretion and subsequent binding of an inhibitory factor . In a period of competitive survival in sympathetic neurons , NGF signalling leads to cell-body secretion of brain derived neurotrophic factor , which promotes apoptosis of neighbouring cells through the receptor p75 [25] . Although p75 is also an antagonist of NGF/TrkA growth signalling , whether it plays a general role in inhibition in high concentrations is unclear; genetic knockout of p75 had no observable effect on DRG explants in bath applications of NGF [19] , whereas outgrowth from trigeminal ganglia was no longer repelled from NGF-coated beads [20] . However , p75 belongs to the broader tumour necrosis factor receptor superfamily , of which many members influence neurite growth in critical developmental stages [41–47] . Shared signalling pathways among this family of receptors suggest a potential general basis for paracrine growth regulation . In DRGs , tumour necrosis factor receptor-1 ( TNFR1 ) is localised to neuronal cell bodies , and strongly antagonises NGF/TrkA responses when stimulated by tumour necrosis factor alpha ( TNFα ) [48] . A source of secreted TNFα in DRGs is the satellite glial cells that closely ensheathe neuronal cell bodies , and express both p75 and TrkA receptors [26] . Although the role of NGF receptors in satellite glial cells is only beginning to be understood [26 , 49] , it is plausible that high NGF concentrations could produce the glial cell activation that elicits TNFα release , and thus a cell-body inhibitory signal through TNFR1 . Consistent with this prediction , genetic knockout of either TNFα or TNFR1 yielded a threefold increase in embryonic DRG explant outgrowth at an NGF concentration of ≈ 2 nM [48] . Integration of NGF and TNFα signalling could occur via the Akt pathway , analogous to DRG neurons stimulated with related neurotrophin insulin-like growth factor ( IGF ) [50] . High concentrations of TNFα antagonise Akt activation , growth associated protein 43 ( GAP43 ) expression and neurite growth promoted by IGF , in a phosphatidylinositol 3-kinase ( PI3K ) -dependent manner [50] . PI3K/Akt signalling is a primary pathway of retrograde TrkA activity [51 , 52] , and GAP43 activation by TrkA at the growth cone promotes cytoskeletal assembly and growth [53] , suggesting possible candidates for the substrates F and G in the model . Thus , an interpretation of the signalling network of Fig 4 is that TrkA and TNFR1 control activation of Akt at the cell body ( F ↔ F* ) , regulating expression of GAP43 , which is then activated at the growth cone ( → G → G* ) . A second pathway of TNFR1 signalling involves a cascade of several caspases . Caspase- 3 , 6 and 9 , in particular , are key effectors of the axonal degeneration that accompanies NGF withdrawal [54 , 55] . GAP43 is a substrate of caspase-3 [56] , thus providing a link between TNFR1 activation at the cell body , and growth inhibition at the growth cone ( G ← G* ) . These molecular candidates provide further means by which the proposed role of TNFα/TNFR1 signalling can be tested experimentally . The sensitivity of explant outgrowth to shallow gradients observed experimentally by refs . [17 , 18 , 22 , 57 , 58] demonstrates the exquisite chemosensory ability of developing neurites . In our model , this is explained by a comparison of concentrations between growth cone and cell body , and a chemical computation that overcomes the deleterious effects of receptor saturation . Our approach was inspired by the study of ref . [28] . There it was shown how the chemosensory system of a single cell can exhibit a steady-state response to uniform extracellular concentration changes that is independent of the concentration ( perfect adaptation ) , yet maintain a persistent asymmetry of response across the cell when presented with a gradient . Perfect adaptation arises through competing processes of activation and inhibition , similar to Fig 3A in the case that c1 = c2 , and non-saturating kinetics that permit a cancellation of concentration-dependent components in the steady-state output . Given the adaptation constraint , sensitive gradient detection was achieved in ref . [28] through a global signal conveyed by diffusion of the inhibitor , and feedback that increases the supply of protein . The behaviour and constraints of our model are rather different; ours is an extended multicellular system , exhibits a biphasic dose response rather than perfect adaptation , and requires that signals are transduced sensitively by receptors that are known to saturate within the regime of interest . Yet , consistent with the prediction of ref . [28] that the principles underlying their work would be conserved across systems , we found that a similar network design and substrate supply mechanism also provided a compatible model structure . The presence of receptor saturation in our model is the key difference that introduces concentration dependence into the steady-state response , which has the appropriate biphasic form when KI ≫ KA . However , this constraint also demands a distinct treatment of gradient detection as it limits the effectiveness of directly comparing activator and inhibitor . By introducing a dual negative regulation of opposing enzymes to counter saturation , we showed how tropic sensitivity can , in principle , persist over a wide range of concentrations . Although serving a different purpose in our model , we note that this additional interaction also permits perfect adaptation , and thus provides a simple generalisation of the work of ref . [28] to encompass the wider concentration regime in which signals saturate . Whether this interaction is indeed a feature of NGF signalling is unknown . Intriguingly , PI3K and the opposing enzyme phosphatase and tensin homology ( PTEN ) , are subject to a dual positive regulation by regulatory subunit p85 [59] . This suggests a possible implementation of a variant of the motif by sequestration of p85 , though to pursue this idea is beyond the scope of the present work . To allow comparison of concentrations from different points in space , our model assumes trafficking of components along the neurite , rather than the intracellular diffusion of inhibitor assumed by ref . [28] . Importantly , this also allows the growth and guidance components of the model , which have competing computational demands , to be processed in distinct regions of the cell . Here , the substrate supply mechanism does not amplify these signals through feedback , but couples them to provide the fine modulation required of guidance by differential growth . Although the spatial structure of a neuron largely precludes intracellular diffusion as a means of relaying signals in our system , another possibility is that this could occur extracellularly . The cell-cell interactions within our model are assumed to be of short range , with an inhibitory signal that depends only on the NGF concentration local to the cell body ( Eq ( 2 ) ) . However , a multicellular counterpart of the global inhibition mechanism of ref . [28] was recently proposed in which an inhibitory factor diffuses throughout a collection of cells [60 , 61] . In this , collective signalling permits a robust response to a gradient , even at sensory limits at which the stimulus cannot be detected by an isolated cell . Thus , although global signalling is not required for gradient detection within our model , long-range diffusion within a ganglion could serve as an alternate or additional means of amplifying a weak gradient across the collection of cells . For long diffusion length scales , such a mechanism would also likely yield a dependence of trophic response on explant body size , as the volume from which the hypothetical inhibitor would be secreted grows more quickly with the radius than the boundary over which the inhibitor would escape by diffusion . However , we observed no correlation between outgrowth and explant body size in the experimental data ( S2 Table ) , arguing against this possibility . To examine these effects more precisely requires detailed modelling of diffusion within the crowded environment of a ganglion , suggesting another avenue for future work , should our key predictions ( Fig 5C and 5D ) be confirmed . The guidance mechanism we propose is of a very different nature to that originally posited for shallow gradients by refs . [18 , 22] . These studies consider a Bayesian model in which guidance decisions are drawn from a probability distribution , determined by computation of gradient direction from noisy receptor binding . Here , we do not consider binding noise; given the 48 h duration of the shallow gradient experiments , quantified by measurements that average over many neurites , we believe our deterministic model gives a suitable account of collective outgrowth , into which noise can be absorbed . Although our model supports the suggestion of ref . [22] that shallow gradients are sensed along the neurite length , our approach leads to a different conclusion regarding the decline in tropic response magnitude with high background concentrations . The interpretation of refs . [18 , 22] was that this is due to receptor saturation limiting the signal to noise ratio for reliable detection . However , our model yields a similar response , while explicitly removing the effect of receptor saturation . Instead , we suggest that in background concentrations with low average outgrowth , neurites simply do not extend far enough through the gradient to develop a concentration asymmetry along their length sufficient to drive a large tropic response . The biphasic average outgrowth constraint was not enforced in simulations that employed the Bayesian model [22 , 58] , meaning this additional dependence may not have been observed . What is the developmental benefit of this mode of tropic response ? Navigation over long distances towards a chemoattractant source may require an early phase of growth in gradients too shallow for a reliable turning response . One possibility for overcoming this problem is that growth cones synergistically process gradients of multiple cues [62] . Another is that a comparison of growth cone and cell-body concentrations may initially orient collective extension toward the source , and thus avoid expensive outgrowth in the wrong direction . As the gradient becomes steeper , closer to the source , turning responses could then take over to direct individual neurites at finer spatial scales . Even at these later stages , a cell-body signal could serve as a useful reference concentration , acting to halt the growth of neurites that have strayed too far off course . Our aim in this study was to understand better the origins and implications of NGF regulation of neurite growth and guidance . We provided experimental evidence that demands a revision of current theories of growth , and argued that theories of guidance in shallow gradients should address the coupling to the underlying growth response . The specific hypothesis we proposed is centred around a role for inhibitory signalling at the neural cell bodies , and quantitatively explains multiple features of the observed responses of developing neurites . The predictions of our theory , which can be tested both with macroscopic manipulations of growth conditions and targeting of specific candidate molecules , motivate new experiments to unravel the complex processes of nervous system development and repair .
Experiments involving animals were conducted in accordance with the Animal Care and Protection Act Qld ( 2002 ) , and the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes , 8th edition ( 2013 ) . Ethics approval was obtained from the University of Queensland Anatomical Biosciences Ethics Committee , approval QBI/548/16 . Thoracic and lumbar DRGs were extracted from P1-P2 rat pups into Leibovitz medium on ice . Excess axonal tissue was trimmed off . For the explants , DRGs were digested for 12 min in 0 . 25% trypsin at 37°C and then washed three times in Leibovitz and kept at 4°C until use . For the dissociated cells , DRGs were digested for 45 min in 0 . 25% trypsin at 37°C and then triturated through a fire-polished pipette . Cells were washed three times in opti-MEM and concentrated by centrifugation to a smaller volume ( ≈ 200 μL ) . Collagen was prepared , on ice , with the following concentrations: 0 . 2% rat tail collagen type 1 ( Corning ) , 0 . 1% sodium bicarbonate , 1×opti-MEM and 1×penicillin/streptomycin . After addition of NGF to the collagen , the dissociated cells were added to the collagen and mixed thoroughly . 750 μL of the collagen was spread on a 35 mm petri dish and allowed to set . A second layer of 750 μL of collagen was added and 5 − 12 DRG explants were added within this second layer . After the collagen set , the dishes were transferred to an incubator for 2 days ( 37°C and 5% CO2 ) . Explants and dissociated cells embedded in collagen were fixed with 4% paraformaldehyde/0 . 1% Triton-X 100 in PBS overnight . Plates were washed five times with PBS with 1 hour between washes , and then incubated overnight at 4°C with β-III-tubulin antibody TuJ1 ( 1:500; R&D Systems ) . After five washes with PBS of 1 hour each , plates were incubated overnight at 4°C with secondary antibody Alexa Flour 488-conjugated goat anti-mouse IgG ( 1:500 , Invitrogen ) . Plates were washed five times in PBS for 1 hour each before acquisition of images using Apotome imaging on a Zeiss Z1 microscope . Images were acquired as z-stacks and flattened by average intensity projection for explants , and by maximum intensity projection for dissociated cells . For measurements of fasciculation , additional z-stacks were acquired at higher resolution ( 0 . 645 μm per pixel ) , and flattened by average intensity projection . Explant images were manually segmented ( using ImageJ ) to separate the cell-body and neurite outgrowth regions , and thresholded by pixel intensity to form binary masks , as described in ref . [18] . For each image , boundary curves were fitted to the cell-body mask and the outer boundary of the largest connected component of the outgrowth mask using the MATLAB function ‘bwboundaries’ . The curves were smoothed with a moving average filter of width 150 pixels , and then parameterised by polar coordinates with N = 360 discrete angles θ n = 2 π n N about an origin defined as the centroid of the cell-body mask . In the event that a ray from the origin intersected a boundary at multiple points , the closest point to the origin was selected . The radial outgrowth function , R ( θn ) , was defined as the distance between the cell-body and neurite region boundaries at each θn . We extended this to a continuous representation by performing a discrete Fourier transform to give R ( θn ) in terms of frequency components , R ( θn ) =Σk=0N−1R^k·e2πikn/N , and then folding about the Nyquist frequency to determine equivalent Fourier coefficients as a 0 = Re ( R^0 ) , a k = 2 Re ( R^k ) and b k = - 2 Im ( R^k ) , 1 ≤ k ≤ 180 . Examples of the image processing steps are shown in S3 Fig . We found that the first five coefficients were sufficient to reconstruct the major features of explant outgrowth via R ( θ ) ≈ a 0 + a 1 cos ( θ ) + b 1 sin ( θ ) + a 2 cos ( 2 θ ) + b 2 sin ( 2 θ ) . ( 7 ) The coefficient a0 determines the average radial outgrowth , a1 and b1 determine the bias in outgrowth in orthogonal image axes , and a2 and b2 capture the polarised growth exhibited by some explants ( likely resulting from growth hotspots at the sites of axotomy ) . For data analysis , we used the coefficient a0 to quantify average outgrowth ( in units of μm ) . For the NGF gradient data set , we used the normalised coefficient b1/a0 as a dimensionless measure of outgrowth bias up the gradient . Neurite growth from dissociated cells was quantified by manual tracing using the ImageJ plugin NeuronJ [63] . We recorded the length of the longest neurite of each cell as an analogue of the extent of radial outgrowth recorded for the explants . Neurite bundle widths were measured using the ImageJ plugin Ridge Detection [64 , 65] . For each explant , we applied the analysis to four 650 μm × 100 μm image strips . The strips were arranged at distances of 150 μm from the explant body , with the short axes parallel to the predominant direction of neurite extension ( approximately forming a square about the body ) . Detected neurite segments of length less than 20 μm were excluded . These restrictions limited double counting of neurites that were identified by Ridge Detection as a collection of broken segments , and artefacts from the regions of dense growth near the explant body . Examples of the segmentation are shown in S2A Fig . Pooling across explants in each condition ( 0 . 3 nM and 10 nM NGF ) we constructed distributions of neurite bundle widths for statistical comparison . As a positive control , we constructed sample image sets of patches containing mostly thin or thick bundles ( S2B and S2C Fig ) , which were easily discriminated by the image analysis . We used the width parameter σ = 1 . 37 pixels in the Ridge Detection algorithm ( which sets an effective range for detection and width estimation ) , and confirmed that changing this parameter did not effect the difference in distributions of tested conditions , nor positive controls . Ganglion outgrowth was simulated as a deterministic extension of neurites from the boundary of a disc of radius RE = 300μm , representing average neurite trajectories in an experimental image plane ( cf . Fig 1A ) . The boundary of the disc was seeded with N = 360 model neurites , projecting radially at angles θ n = 2 π n N . The rate of extension of each neurite was determined from the system of ordinary differential equations for the signalling model ( Eqs ( 25 ) – ( 31 ) ) and linear growth rule ( Eq ( 33 ) ) . For comparison with the NGF gradient data set , outgrowth was simulated with the explant body disc centred in an exponential gradient , given in polar coordinates by c ( r , θ ) = c0 exp ( sr sin ( θ ) ) . Here , c0 denotes the background concentration and s denotes the gradient steepness ( a 0 . 1% per 10 μm gradient corresponds to s = 1 × 10−4 ) . The growth-cone NGF concentration c1 and cell-body concentration c2 , which drive the signalling model , were determined for each neurite dependent on its length and angle of projection . For a neurite of length R at angle θ , the cell-body concentration remained fixed at c2 = c ( RE , θ ) , whereas the growth-cone concentration was updated as the neurite extended as c1 ( t ) = c ( RR + R ( t ) , θ ) . The model was initialised at steady-state values , and 48h outgrowth was simulated . The results were analysed using the Fourier decomposition described above . All simulations were performed in MATLAB ( Mathworks ) with custom written code and the solver ode15s .
|
For the brain to become wired up during development , growing nerve fibres use molecular guidance factors to navigate over long distances and find their appropriate targets . However , the ability of nerve fibres to do this is severely limited by the loss of both growth and guidance as the concentration of guidance factors increases—a phenomenon that has never been fully explained . We propose a mathematical model that couples growth and guidance at the level of signal transduction , and show that it can , for the first time , quantitatively explain the largest current dataset of precisely controlled measurements . This finding impacts on understanding of both the causes of neurodevelopmental disorders , and repair after brain or spinal injury .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"signaling",
"networks",
"neurites",
"neuroscience",
"collagens",
"developmental",
"biology",
"signal",
"inhibition",
"signs",
"and",
"symptoms",
"network",
"analysis",
"molecular",
"development",
"neuronal",
"dendrites",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"proteins",
"biological",
"tissue",
"immune",
"system",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"diagnostic",
"medicine",
"cell",
"biology",
"anatomy",
"fasciculations",
"ganglia",
"neurons",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling"
] |
2018
|
Control of neurite growth and guidance by an inhibitory cell-body signal
|
Salmonella Typhimurium ( S . Tm ) is a common cause of self-limiting diarrhea . The mucosal inflammation is thought to arise from a standoff between the pathogen's virulence factors and the host's mucosal innate immune defenses , particularly the mucosal NAIP/NLRC4 inflammasome . However , it had remained unclear how this switches the gut from homeostasis to inflammation . This was studied using the streptomycin mouse model . S . Tm infections in knockout mice , cytokine inhibition and –injection experiments revealed that caspase-1 ( not -11 ) dependent IL-18 is pivotal for inducing acute inflammation . IL-18 boosted NK cell chemoattractants and enhanced the NK cells' migratory capacity , thus promoting mucosal accumulation of mature , activated NK cells . NK cell depletion and Prf-/- ablation ( but not granulocyte-depletion or T-cell deficiency ) delayed tissue inflammation . Our data suggest an NK cell perforin response as one limiting factor in mounting gut mucosal inflammation . Thus , IL-18-elicited NK cell perforin responses seem to be critical for coordinating mucosal inflammation during early infection , when S . Tm strongly relies on virulence factors detectable by the inflammasome . This may have broad relevance for mucosal defense against microbial pathogens .
The intestinal mucosa is a key site limiting microbial access to the body [1 , 2] . Nonetheless , some enteropathogenic bacteria , including Salmonella enterica subspecies 1 serovar Typhimurium ( S . Tm ) , have the capacity to overcome the mucosal defenses and utilize the gut as a port of entry . It is still not well understood how defenses are mounted and coordinated to limit infection . The innate immune system provides formidable protection against the vast majority of invading microbes . Its chemo-sensors , termed danger recognition receptors , are detecting conserved microbial products ( including key virulence factors ) and tissue damage inflicted by the infection , boost antimicrobial defense and recruit phagocytic cell populations to eliminate the pathogen and cellular debris [3–5] . It appears that some "successful" pathogens have evolved mechanisms to evade these innate responses [6 , 7] . This is thought to complicate the analysis of many important recognition events . For example , S . Tm can downregulate expression of flagella and the SPI-1 type III secretion system at systemic sites and may thereby evade ( at least partially ) the detection by danger recognition receptors , e . g . the NAIP/NLRC4 inflammasome [8–10] . Such "stealthy" behavior is not an option at sites where the respective virulence factors ( or other recognized components ) are needed by the pathogen for performing an important step of the infection cycle . Based on these considerations , enteropathogenic bacteria may be particularly prone to detection by the innate immune system when they arrive at the mucosal surface and have to deploy virulence factors to achieve host cell manipulation and invasion . Indeed , recent work has identified the epithelial NAIP/NLRC4 inflammasome as a central player in the initial defense against Citrobacter rodentium and S . Tm , initiating the expulsion of infected enterocytes into the gut lumen [11–13] . However , it had remained unclear whether and how this is coordinated with other defenses at this critical site . S . Tm is a key cause of diarrhea worldwide [14] . The streptomycin mouse model for Salmonella Typhimurium diarrhea is used to study the pathogen's virulence factors and the mucosal responses mounted upon infection [15 , 16] . In the gut lumen , S . Tm relies on flagella to reach the epithelial surface and the TTSS-1 to bind and invade the intestinal epithelium [12 , 17–21] . In fact , enterocyte invasion was already described in pioneering work employing guinea-pig and bovine models [22 , 23] . While numerous mucosal immune defense mechanisms have been proposed [16 , 24] , the critical responses coordinating the mucosal defenses during the initial wave of pathogen invasion have remained elusive . Only recently , first mechanistic insights have been obtained in the mouse model . This has established a critical function of the epithelial NAIP/NLRC4-inflammasome to detect enterocyte invasion by S . Tm during the first hours of infection [12] . This inflammasome-dependent defense appears to be two-pronged . By facilitating the expulsion of infected epithelial cells into the gut lumen , the innate defense was shown to reduce epithelial pathogen loads by as much as 100-fold . However , this fails to completely clear the pathogen from the tissue . In parallel , the inflammasome response triggers a pro-inflammatory program that leads to overt tissue inflammation characterized by crypt abscesses , tissue damage and leukocyte infiltration—typical hallmarks of acute S . Tm diarrhea [16 , 25] . However , it has remained unclear how inflammasome activation and induction of mucosal inflammation are linked in this disease . Here , we have analyzed the mechanisms switching the naïve mucosa into overt inflammation in response to S . Tm infection . Our work identifies a key role of caspase-1 ( not -11 ) mediated IL-18 production , IL-18-mediated accumulation of mature natural killer ( NK ) cells in the mucosa and suggests that NK-cell perforin is important for mounting gut inflammation within 12h of the infection .
We set out to characterize the initiation of a mucosal immune response and the subsequent cecum inflammation upon acute S . Tm infection using the streptomycin mouse model [15] . As inflammasome recognition of S . Tm has already been identified as an important component of the early mucosal defense [12] , we postulated that the inflammasome dependent IL-1 family cytokines IL-1β and IL-18 could play a role in shaping the early immune response and the onset of inflammation . Indeed , both cytokines showed elevated mature protein levels in the infected mucosal tissue already early during infection ( 5x107 CFU S . Tm , 12h p . i . ; Fig 1a ) . In case of Il1b this went along with a transcriptional upregulation ( S1a Fig ) , as observed previously [26 , 27] . In contrast , the IL-18 response occurred mostly at the post-transcriptional level , as Il18 transcript levels remained unchanged at least at this early stage of the infection ( S1a Fig ) . To study the functional importance of IL-1β and IL-18 in the onset of mucosal inflammation , we infected mice deficient in IL-1α/β or IL-18 and appropriate littermate controls and measured the degree of mucosal inflammation using a well-established histopathological scoring scheme that assesses the degree of submucosal edema , the integrity of the epithelial layer , the abundance of mucin-filled goblet cells in the epithelial lining and the infiltration by polymorphonuclear leukocytes ( according to [15] ) . At 12h p . i . , S . Tm had reached equally high pathogen densities in the cecal lumen of all experimental groups ( S1b Fig ) . IL-1α/β-deficient animals and controls featured equivalent levels of pronounced cecum pathology ( Fig 1b and 1c and S1c Fig ) . In contrast , the cecal mucosa of the IL-18-deficient mice was much less inflamed ( Fig 1b and 1c and S1c Fig ) . This provided a first indication that IL-18 is functionally important for initiating gut mucosal inflammation . However , the underlying molecular and cellular processes remained to be established . Depending on the disease model , IL-18 can be processed and released via caspase-1 and/or caspase-11 inflammasomes [12 , 28–30] . To identify which caspase is required for IL-18 processing in our model , we infected mice deficient in caspase-1 and -11 or deficient in caspase-11 alone and corresponding littermates and assessed mucosal IL-18 levels as well as cecal pathology . While caspase-1/11-deficient animals displayed reduced levels of mature IL-18 in the infected cecal mucosa ( p<0 . 05; Fig 1d , left side ) , we observed no such reduction in the caspase-11-deficient mice ( Fig 1d , right side ) . Correspondingly , caspase-1/11-deficient animals , but not caspase-11-deficient mice , showed reduced cecum mucosal inflammation ( Fig 1e ) . This pheno-copied the IL-18-deficient mice ( Fig 1b and 1c right side ) and suggested that caspase-1 ( not caspase-11 ) is required for IL-18 processing and the subsequent elicitation of gut inflammation during the first hours of S . Tm infection . The dependency of mucosal inflammation on IL-18 was further explored in a time course experiment , infecting IL-18-deficient mice and littermates for 6h , 12h , 18h or 36h with S . Tm . Quantification of mature IL-18 protein in the cecum mucosa of the IL-18 proficient littermate controls revealed elevated protein levels during the first 6h-18h p . i . , which returned to basal expression at 36h p . i . ( S1d Fig ) . Importantly , colitis was clearly reduced in IL-18-deficient mice up to 18h p . i . ( S1f Fig ) , while cecum luminal colonization appeared stable throughout the course of infection ( S1e Fig ) . By 36h p . i . , the knockout mice featured an equivalent degree of inflammation as their littermate controls ( S1f Fig ) . Therefore , IL-18 deficiency causes a delayed onset , but not a complete blunting of mucosal inflammation in response to S . Tm infection . It had remained unclear , if the observed delay in cecal pathology is explained either by an important function of IL-18 during early infection , or by a general alteration of mucosal homeostasis in the knockout mice . To address the underlying mechanism , we examined if neutralization of IL-18 during acute S . Tm infection would might reduce cecum pathology by 12h p . i . . To this end , we artificially increased the concentration of a naturally occurring inhibitor of IL-18 , by injecting C57BL/6 wild-type mice ( WT ) with recombinant IL-18BP ( 2mg/kg i . p . ) or PBS as a control . While the treatment did not affect cecum luminal colonization ( S1h Fig ) , IL-18 inhibition significantly reduced mucosal pathology ( Fig 1f ) . Cytokine neutralization yielded in fact an equivalent phenotype as genetic IL-18 ablation ( compare Fig 1b and 1f ) . This is in line with a direct regulatory function of IL-18 during acute infection rather than a homeostatic imbalance in the genetically deficient animals . Next , we examined the effect of increased IL-18 concentrations . As IL-18 accelerates the progress of inflammation ( see S1f Fig ) , artificially raising IL-18 levels should enhance inflammation during the first hours of infection . Indeed , the injection of recombinant IL-18 ( 120μg/kg i . p . ) enhanced mucosal inflammation by 8h p . i . , an early time point when untreated mice showed only minor signs of pathology ( Fig 1g ) . Based on this observation , we tested if IL-18 would be sufficient to drive mucosal inflammation even in the absence of invasive bacteria . To this end , we infected mice with an avirulent isogenic S . Tm mutant ( S . Tmavir; ΔinvG; sseD::aphT; [31] ) , which colonizes the gut lumen but fails to invade the cecum tissue and does not elicit mucosal pathology within 4 days of infection [31 , 32] . However , injection of rIL-18 during infection with S . Tmavir did not induce mucosal inflammation ( S1k Fig ) . From these findings , we conclude that IL-18 is required for , and modulates the potency of the acute inflammatory response towards invasive S . Tm . Yet , it is insufficient to induce inflammation in the absence of pathogen tissue invasion . Cytokine and chemokine response patterns can provide cues about underlying immunological processes . To elucidate how IL-18 affects the mounting of gut inflammation , we therefore collected infected cecum tissue samples from IL-18-deficient mice and littermate controls ( 12h p . i . , 5x107 CFU S . Tm ) and performed transcriptional profiling by RNA-Seq . In total , mRNA levels for more than 200 genes were reduced in IL-18-deficient mice compared to the littermate controls ( for a complete list of significantly regulated genes , see S1 Table ) . Many of those differentially expressed genes belonged to pathways involved in general pro-inflammatory responses , immunity to infection as well as chemokine and cytokine signaling . One group of these IL-18 dependent chemokines is known to coordinate the recruitment of neutrophils to sites of infection ( Fig 2a , depicted in green ) . The accumulation of neutrophils in the infected mucosa is a hallmark of S . Tm-induced tissue inflammation [16 , 33] and represents a key line of pathogen defense in the cecum lumen , the cecum tissue as well as at systemic sites at later stages of the infection [34–39] . However , their function in eliciting mucosal inflammation had remained unclear . We speculated that reduced neutrophil recruitment in IL-18-deficient mice could explain their delayed onset of mucosal pathology . Thus , we have analyzed the role of neutrophils in the initial phase of the infection . S . Tm-mediated induction of two candidate chemokines ( CXCL1 and CXCL2 ) as well as their reduced induction in IL-18-deficient animals was confirmed by qRT-PCR ( Fig 2b ) . In line with this expression pattern , we observed a significant accumulation of Ly-6G+CD11b+CD45+ cells ( i . e . neutrophils ) in the infected mucosal tissue compared to naïve animals ( 12h p . i . , 5x107 CFU S . Tm; Fig 2c and 2d ) . In contrast , the lamina propria of infected Il18-/- mice featured significantly decreased amounts of Ly-6G+CD11b+CD45+ cells compared to their littermate controls ( Fig 2e and 2f ) , although recruitment was not completely blunted . This verified that IL-18 affects neutrophil recruitment to the infected cecum mucosa already early in S . Tm infection , presumably via the induction of neutrophil-recruiting chemokines . In order to determine , if the recruited neutrophils functionally contribute to the onset of mucosal inflammation , we depleted neutrophils using a combination of anti-G-CSF ( 0 . 4mg/kg ) and anti-Ly-6G ( 6mg/kg ) antibodies . Mice were infected with S . Tm for 12h and enteropathy was assessed by pathoscoring . All mice showed equivalent levels of cecum luminal S . Tm colonization ( S2a Fig ) . Interestingly , neutrophil depletion did not significantly alter the mucosal pathology ( Fig 2g and 2h , S2b Fig ) . In spite of the blunted neutrophilic influx into the tissue and lumen , tissue pathology was comparable between neutrophil-depleted animals and non-depleted controls . As neutrophils have been described to limit S . Tm loads in the cecum lumen and at systemic sites at later stages of the infection [38 , 39] , we hypothesized that they might still limit S . Tm tissue loads at 12h p . i . . To address this hypothesis , we infected mice depleted of neutrophils and non-depleted controls with S . Tm carrying a reporter expressing GFP from a SPI-2 promoter once the pathogen has invaded host cells ( pssaG-GFPmut2; [31] ) and enumerated the GFP-positive S . Tm in the epithelial layer . Surprisingly , we could observe no difference in S . Tm tissue loads between mice depleted of neutrophils and WT animals ( S2c Fig ) . In conclusion , our data identify an early recruitment of neutrophils to the infected LP , which is partially dependent on the presence of IL-18 . However , during this early state of infection , neutrophils do not seem contribute to the onset of tissue pathology or the control of S . Tm tissue-loads . We conclude that reduced neutrophil counts cannot explain the delayed tissue pathology of the IL-18-deficient animals that we had observed in the experiments , above ( Fig 1 ) . In addition to Neutrophil recruiting chemokines , a second group of chemokines was significantly differentially expressed in our RNA-Seq dataset , which is known to coordinate the recruitment of NK cells to sites of infection ( Fig 3a , depicted in red;[40] ) . NK cells are early effectors in the mucosal defense against several viruses , bacteria , protozoa and fungi [41] . They respond to a wide range of chemokines , are rapidly mobilized in response to danger signals and therefore quickly recruited to sites of inflammation and disease [42] . There , NK cells are activated either indirectly by cytokines or directly via the recognition of stressed and infected cells [43–45] . During systemic bacterial infections , activated NK cells can limit tissue infection and prevent systemic spread of the pathogen through direct lysis of target cells or by releasing GM-CSF , TNF and IFNγ to orchestrate further responses [41] . The secretion of such pro-inflammatory cytokines by NK cells and other early effector cells orchestrates and amplifies the local immune response , inducing a full-blown mucosal inflammation to boost pathogen elimination . Our findings provided a first hint that NK cells might be important coordinators initiating gut inflammation . For most of those chemokines , we could confirm their S . Tm-mediated induction and their IL-18 dependence by qRT-PCR ( Fig 3b and 3c ) . In line with this , the cecum mucosa of infected WT control mice featured ~10-fold higher densities of NK1 . 1+ CD3- CD45+ cells than that of uninfected animals ( 12h p . i . , 5x107 CFU S . Tm; Fig 3d and 3e ) . Time course experiments verified the recruitment of NK1 . 1+ CD3- CD45+ cells during the initial 12-18h p . i . ( S3a Fig ) . In contrast , IL-18-deficient animals harbored reduced numbers of NK1 . 1+ CD3- CD45+ cells in the infected cecal mucosa ( Fig 3f and 3g ) . These data suggest that IL-18-dependent chemokine responses control the accumulation -and most likely also the function- of NK1 . 1+ CD3- CD45+ cells in the cecum tissue during the initial phase of the pathogen-host interaction . We further characterized the accumulating NK1 . 1+ CD3- cells according to their surface expression of KLRG1 , NKp46 , CD122 , TCRγδ and Thy1 ( Fig 3h ) . Most NK1 . 1+ CD3- cells were identified as KLRG1+ NKp46+ CD122+ Thy1+ TCRγδ- NK1 . 1+ CD3- NK cells . As expected for NK cells , this population expressed high levels of the transcription factors Eomes and Tbx21 as well as Prf-1 and Sell transcripts , while mRNA-levels of transcription factors Rorc and Gata3 were not significantly affected ( Fig 3i ) . To further verify the IL-18 function in NK cell recruitment , we performed experiments on caspase-1/11-deficient mice . As these mice produced reduced levels of mature IL-18 protein in response to mucosal infection ( see Fig 1d ) , we reasoned that these animals should feature reduced NK cell numbers in the infected cecum tissue . 12h infection experiments with caspase-1/11-deficient animals and their littermate controls verified that this is indeed the case ( Fig 3j and S3b Fig ) . In contrast , caspase-11-deficient mice featured equivalent mucosal NK cell numbers as their littermate controls ( Fig 3k and S3c Fig ) . This provided further evidence supporting a link between mucosal IL-18 induction and the accumulation of NK cells during S . Tm infection . In order to assess their functional importance , we depleted NK cells using an α-asialo GM1 antiserum and infected mice with S . Tm ( 5x107 CFU S . Tm , 12h p . i . ) . The depletion efficiency was verified by flow cytometry ( S3d Fig ) . Both experimental groups showed equal bacterial loads in the mLN , suggesting that initial pathogen translocation kinetics were not accelerated in the absence of NK cells . The cecal pathology was assessed by pathoscoring . Strikingly , depletion of NK cells during the S . Tm infection reduced levels of mucosal pathology compared to mock-depleted controls ( Fig 3l and 3m , S3e Fig ) . Equivalent observations were made when using an α-NK1 . 1 antibody for cell depletion ( clone PK136 , 10mg/kg , i . p . ; Fig 3m and S3e Fig ) . This antibody is less specific than the α-asialo GM1 antiserum , as it recognizes not only NK cells , but also additional NK1 . 1+ cell populations . Nonetheless , both NK cell depletion strategies recapitulate the delayed mucosal inflammation observed in IL-18-deficient mice ( compare Figs 1b and 3m ) . These data provided a first indication that IL-18 dependent NK-cell recruitment is central for initiating the inflammatory response . So far , we could show that the initiation of mucosal inflammation depends on IL-18 , that this leads to changes in the cellular composition of the mucosa ( including increased NK cell abundance ) and that NK cells are required for mounting the disease . However , it had remained unclear how NK cells are engaged in the process . At least three different mechanisms might be involved , i . e . enhanced NK-cell immigration , increased NK cell proliferation in the infected tissue and/or elevated NK-cell activation within the responding tissue . First , we analyzed the IL-18 receptor dependency of NK cell accumulation . Here , it was important to find out if NK-cell immigration is driven by direct IL-18 signaling ( via the NK-cell's IL-18 receptor ) or by the NK-cell recruiting chemokines produced ( by other cells ) in the infected mucosa ( Fig 3a–3c ) . It should be noted that our initial experiments in IL-18 deficient animals could not distinguish between these two mechanisms , as IL-18 deficiency also reduced tissue inflammation and chemokine production ( other than IL-18 ) . Therefore , it remained unclear whether a defective NK cell response can be directly attributed to IL-18 signaling . To address this issue , we reconstituted lethally irradiated CD45 . 1 WT mice with a 1:1 mixture of CD45 . 1 WT and CD45 . 2 Il18r1-/- bone marrow . These mice were infected with S . Tm and we analyzed NK cell numbers from both genotypes in the infected cecal LP . In this setting , we could directly compare both genotypes within the same mucosa . Compared to WT ( CD45 . 1+ ) NK1 . 1+ cells , the Il18r1-/- ( CD45 . 2+ ) NK1 . 1+ cells accumulated in significantly lower numbers in the infected mucosa ( Fig 4a ) . In contrast , WT and mutant cells were present at equivalent frequencies in the blood ( S4a Fig ) . This suggests that IL-18 directly , and not the altered inflammatory environment of the mucosa , affects the accumulation of NK cells in the mucosa during the first hours of S . Tm infection . Notably , the direct effect of IL-18 implies that reduced mucosal NK cell accumulation in absence of IL-18 ( Fig 3d–3i ) is likely attributable to both , reduced NK cell chemokine levels and a direct effect of IL-18 via the IL-18R of NK cells . There are at least two conceivable modes of action for IL-18 in this scenario . Either it stimulates proliferation of the NK cells in the infected mucosa , thereby expanding the population in situ , or it enhances the migratory capacity of NK cells , thus boosting NK cell recruitment . To address the first scenario , we analyzed the proliferation of LP NK cells using an in vivo EdU incorporation assay . In contrast to the clear increase of NK1 . 1+ cell abundance in the infected mucosa , the fraction of EdU+ cells within this subset remained virtually unchanged ( Fig 4b ) . As control , we measured in parallel the EdU incorporation in CD11b+ NK1 . 1- cells , which should comprise different myeloid subsets known to proliferate in inflamed tissue [46 , 47] . In contrast to the NK1 . 1+ cells , the infected mucosa featured highly increased fractions of EdU+ CD11b+ NK1 . 1- cells ( S4b Fig ) . This argues against an in situ proliferation of NK cells in response to IL-18 . To verify that IL-18 has an impact on the migratory behavior of NK cells , isolated NK cells ( purity ~95% , S4c Fig ) were stimulated ex vivo with rIL-18 ( 100ng/mL rIL-18 , 3h ) and examined in 2D Transwell migration experiments using CXCL9 , a classical NK cell recruiting chemokine [40] . Indeed , stimulation with IL-18 increased the migratory efficiency of NK cells , in particular at lower CXCL9 concentrations ( 50 or 250 ng/ml; Fig 4c and S4d Fig ) . This increased migratory potential was clearly dependent on IL-18 signaling , as IL-18R-deficient NK cells were unresponsive to the stimulation and showed a migration comparable to unstimulated WT NK cells ( S4e Fig ) . As IL-18-stimulated NK cells displayed an increased migratory potential , we examined if this can be attributed to an up-regulated surface expression of the CXCL9 receptor , CXCR3 . However , rIL-18 stimulation affected neither the number of CXCR3-expressing NK cells , nor the amount of CXCR3 surface expression on stimulated NK cells ( Fig 4d ) . This suggested that IL-18 enhances CXCL9/CXCR3 signaling downstream of the receptor ( CXCR3 ) . In summary , these data support that IL-18 increases the migratory capacity of NK cells ( by engaging the NK-cell's IL-18 receptor ) , thereby enhancing NK cell recruitment to the infected mucosal tissue . Throughout the body , tissue NK cells are featuring distinct functions and maturation stages [48] . By convention , the surface expression of CD11b and CD27 defines four maturation stages of murine NK cells [49] . These correspond to the NK cells' capacity to produce cytokines and their cytotoxic potential . Immature double negative CD27- CD11b- NK cells follow the maturation profile -> CD27+ CD11b- -> CD27+ CD11b+ double positive -> CD27- CD11b+ NK cells . Of these developmental stages , both CD11b+ NK cell populations are considered as mature as they exert typical NK cell effector functions [49–51] . The phenotype of the NK cells accumulating in the S . Tm infected gut mucosa remained to be established . The cecal mucosa of non-infected C57BL/6 mice harbored only a small number of NK cells and these mainly expressed immature phenotypes ( CD27- CD11b- and CD27+ CD11b-; Fig 5a and 5b ) . During the first 12h of infection , CD27+ CD11b+ NK cells accumulated in the infected mucosa ( Fig 5a and 5b ) , indicating that NK cells not only increase in abundance , but also display a higher degree of maturation . In contrast , in the infected mucosa of IL-18-deficient mice , NK cell remained scarce and mainly exhibited immature phenotypes ( CD27- CD11b- and CD27+ CD11b-; Fig 5c and 5d ) . In fact , the maturation state of NK cells in infected IL-18-deficient animals resembled that of uninfected WT mice . These data suggest that IL-18 affects not only the recruitment , but also the maturation state of NK cells in the infected cecal mucosa . As our previous analysis has revealed that the accumulated NK cells in the infected cecal LP are phenotypically mature , we next wanted to address the NK cell effector function , contributing to the onset of cecal inflammation . NK cells can affect defense via ( at least ) two different mechanisms , i . e . their cytotoxic function and the production of effector cytokines that boost antimicrobial defenses of other cell types [41 , 52] . Our RNA-Seq data suggested a decreased expression of the three major NK cell-derived effector cytokines TNF , GM-CSF and IFNγ ( Fig 6a , depicted in red ) . Therefore , we investigated the potential contribution of those three cytokines in the induction of early mucosal pathology after S . Tm infection . Although RNA-Seq analysis had shown a clear downregulation of GM-CSF transcripts , GM-CSF protein was not yet detectable in the cecal mucosa at 12h p . i . ( S5a Fig ) rendering it an unlikely candidate for promoting pathology at this initial phase of the infection . Other than GM-CSF , TNF protein levels were induced in the cecal LP by 12h p . i . and markedly reduced in IL-18-deficient mice ( S5b and S5c Fig ) . However , flow cytometric analysis of TNF-producing cells in the cecal LP uncovered that the protein was not produced by NK cells but rather by cells from the myeloid compartment , at least at this early stage of the infection ( S5d Fig ) . This excluded TNF as a likely NK cell effector cytokine and prompted us to focus on IFNγ . RT PCR analyses confirmed that transcripts of Ifng and IFNγ-regulated genes known to be important in diverse innate and adaptive IFNγ-dependent antibacterial responses were significantly reduced in IL-18-deficient mice ( Iigp1 and Cxcl10 shown as examples , Fig 6b ) . In addition , the IFNγ protein levels did not rise above the detection limit in the infected mucosa of IL-18-deficient mice ( Fig 6c ) . Flow cytometry of cecum LP cells revealed that the absence of IFNγ is attributable to decreased populations of IFNγ-producing cells in the Il18 knockout animals ( Fig 6d ) as well as to reduced IFNγ levels per cell ( S5e Fig ) . These populations could be partially rescued by the injection of rIL-18 ( 120μg/kg , i . p . ) into infected IL-18-deficient mice ( S5f Fig ) . Flow cytometric analysis confirmed that the majority of IFNγ+ cells were CD45+ CD3- NK1 . 1+ lymphocytes ( Fig 6e and 6f ) expressing CD11b , Thy1 , NKp46 and Eomes ( S5g Fig ) , and are likely identical with the NK cells identified above ( see Fig 3h and 3i ) . In line with earlier work , some CD3-positive cells ( likely T-cells ) can also produce IFNγ in response to IL-18 ( Fig 6f ) [53 , 54] . However , our NK cell depletion assays suggest that they do not affect the kinetics of mucosal inflammation . Taken together , these data support that IL-18 induces IFNγ production by NK cells during the early phase of the mucosal response to S . Tm infection . However , it had remained unclear , if this IFNγ is functionally required to drive the tissue inflammation observed by 12h p . i . . To address if IFNγ enhances the inflammatory pathology by 12h p . i . , we infected mice deficient in IFNγ signaling ( Ifng or Ifngr1 knockout mice ) and appropriate littermate controls with S . Tm . In contrast to IL-18-deficient or NK cell-depleted animals , IFNγ- and IFNγR-deficient mice showed equivalent levels of cecal pathology as the littermate controls ( Fig 6g and S5h Fig , compare with Figs 1b and 2m ) . This indicated that a functional IFNγ response by NK cells is not required to initiate mucosal inflammation . It should be noted that IFNγ is well-known to exert important functions limiting bacterial growth at systemic sites at later stages of typhoid-fever-like disease [55] . To verify this in our infection model , we infected IFNγ- and IL-18-deficient mice for 72h with S . Tm , a time point when the pathogen has spread from the mucosal tissue to systemic organs [15 , 20] . Indeed , in the mesenteric lymph nodes of both , IFNγ- and IL-18-deficient mice , we detected significantly elevated S . Tm loads at 72h p . i . ( Fig 6h ) . This verified that IL-18 induced IFNγ is dispensable for mounting mucosal inflammation , but important for the subsequent restriction of pathogen spread to systemic sites . Besides the production of pro-inflammatory cytokines , NK cells can exert their effector function by inducing cell death of target cells [41] , either by inducing target cell apoptosis by death ligand signaling via TRAIL or FasL or by releasing cytotoxic granules , containing proteases called granzymes and perforins [56–58] . IL-18 can prime this cytotoxicity [59] . Upon release of the granules in close proximity to the target cell , perforin forms pores in the plasma membrane , enabling granzyme uptake into the target cell and subsequent induction of apoptotic cell death or osmotic cell lysis . These mechanisms are well known to eliminate virus-infected host cells [60] . A role in bacterial infections in vivo ( i . e . liver infection by Chromobacterium violaceum and Citrobacter rodentium infection in the colon ) has only recently been identified [6 , 61] . Other systemic infections by intracellular pathogens ( e . g . S . Tm , Listeria monocytogenes ) are not affected in perforin-deficient mice , presumably due to down-regulation of inflammasome/IL-18 stimulating ligands at these sites [6 , 39 , 62] . However , it remained unclear if NK-cell mediated cytotoxicity might be involved in the initial phases of S . Tm gut infection . Based on the increased levels of mature IL-18 in the infected mucosa ( Fig 1a ) and the accumulation of matured NK-cells by 12h p . i . , we hypothesized that this might indeed be the case . To this end , we infected perforin-deficient animals and littermate controls for 12h ( 5x107 CFU S . Tm by gavage ) and assessed cecal pathology . Strikingly , perforin-deficient mice showed a much lower degree of mucosal pathology than their littermate controls ( Fig 7a and 7b , S6b Fig ) , while luminal pathogen loads were not affected ( S6a Fig ) . In fact , perforin-deficiency fully recapitulated the delayed mucosal pathology observed in IL-18-deficient mice and NK cell-depleted animals ( compare Fig 7a to Figs 1b and 2m ) . This provided a first hint suggesting that perforin is an important NK cell effector mechanism that accelerates disease kinetics in the early phase of S . Tm infection . NK cells are not the only effector cell type capable of inducing perforin-dependent cell death . Clearly , cytotoxic T cells could represent an alternative perforin source during infection . Therefore , we addressed a possible contribution of cytotoxic CD8+ T cells in the IL-18 driven mucosal immune response . However , the LP of infected Il18-/- mice ( 5x107 CFU S . Tm by gavage; 12h p . i . ) featured similar amounts of CD8+ T cells as their littermate controls ( S7a–S7c Fig ) . In addition , depletion of CD8+ cells using an anti-CD8 antibody ( 200μg/mouse , i . p . ; S7e Fig ) did not significantly alter mucosal pathology upon S . Tm infection ( 5x107 CFU S . Tm , 12h; S7f and S7g Fig ) . These data suggest that cytotoxic CD8+ T cells are not involved in early S . Tm-induced mucosal inflammation and suggest that other cell types like the NK-cells are the relevant source of perforin . The latter would be in line with observations from NK-cell transfer experiments ( S6d–S6f Fig ) . It has been shown recently that NK cells are able to control Chromobacterium violaceum loads in the liver and Citrobacter rodentium loads in the colon by perforin-dependent cytotoxicity [6 , 61] . Therefore , we wanted to analyze if this mechanism of host protection is also involved in mucosal S . Tm infection . To this end , we infected IL-18- and Perforin-deficient animals for 12h with S . Tm carrying a reporter expressing GFP from a SPI-2 promoter ( 5x107 CFU S . Tm-GFP; 12h ) and enumerated bacterial tissue loads ( Fig 7c and 7d ) . However , we could not observe any significant differences for tissue resident bacteria comparing Il18-/-mice , Prf-/-mice and littermate controls . This suggests that gut epithelial S . Tm loads are controlled independently of NK cell mediated killing , at least within the first 12h of the infection . A second possible mechanism explaining how perforin-mediated cell cytotoxicity might affect disease pathology could reside in the release of soluble pro-inflammatory mediators from dying cells . ATP would be a likely mediator of such a response , as ATP is released from dying cells and a potent stimulus initiating the recruitment of inflammatory monocytes , macrophages and neutrophils [63–65] . In order to test any involvement of ATP signaling in the induction of tissue pathology , we treated infected C57BL/6 mice with the pan-P2 Receptor antagonist Suramin during the infection and assessed the degree of pathology at 12h p . i . . Suramin-treated animals did indeed feature reduced mucosal pathology ( S6h–S6j Fig ) . This is consistent with a general involvement of ATP signaling in the onset of early pathology . As the host expresses numerous different P2 receptors on a broad variety of cell types , it remains to be verified if Suramin affects indeed the same pathway as triggered via perforin . Overall , however , these data indicate that NK-cells are a relevant source of perforin and that perforin-mediated NK cell cytotoxicity may affect disease kinetics by the release of host cellular molecules . In summary , this work identifies a key signaling cascade in the cecal mucosa whereby S . Tm-induced IL-18 drives the recruitment of activated NK cells to the infected LP and suggests perforin mediated cytotoxicity as a novel mechanism eliciting gut inflammation .
The mucosal defense program that commences inflammation had remained incompletely understood . Here , we investigated how the naive intestinal mucosa mounts the initial response to S . Tm infection . This identified caspase-1-dependent IL-18 as a pivotal cytokine in the process . Our data establish IL-18-promoted migration and accumulation of mature NK cells and perforin-mediated NK cell cytotoxicity as a key axis driving mucosal inflammation . The IL-18/NK cell perforin axis establishes a second arm of the innate immune defense that is elicited upon recognition of S . Tm virulence factors ( i . e . flagella , TTSS-1;[13] ) by the mucosal caspase-1 inflammasome . The first mechanism , expulsion of infected enterocytes , reduces the epithelial pathogen loads by about 50-100-fold by 12-18h p . i . , but cannot completely clear the pathogen [12] . A complementary arm is identified in this paper . It is activated by the release of mature IL-18 , which triggers pronounced cytokine and chemokine responses that lead to accumulation of matured NK cells that seem to promote inflammation via perforin . It is well established that inflammation results in a generalized antimicrobial state in the mucosa , featuring elevated numbers of phagocytic leukocytes , augmented production of antimicrobial peptides , and a boosting of adaptive immune responses [16 , 33] . How does gut inflammation affect pathogen loads ? Strikingly , the IL-18/perforin-axis is required for mounting gut inflammation within 12h , while S . Tm loads in the gut tissue and in the gut lumen remain unaltered . Thus , during these first 12h , pathogen loads seem to be mainly controlled via NLRC4-dependent ( but IL-18 independent ) expulsion of infected enterocytes [12 , 13] . Only at later stages of the infection , macrophages , granulocytes and other parts of the innate immune defense seem to kick in and reduce pathogen tissue loads . In the gut lumen of streptomycin-pretreated mice ( as used in the present study ) , the gut luminal pathogen loads are only affected at day 2 or later [38 , 66] . In conclusion , the host's IL-18 dependent innate defenses and overt tissue inflammation seem to start controlling pathogen loads only after the initial 12h of infection . During the first days of S . Tm infection , numerous cytokines are produced by the intestinal mucosa , including IL-1β and IL-18 [67 , 68] . The release of mature IL-1β is induced in response to infection and it is known to increase host resistance to systemic spread during later stages of the disease [36 , 69] . In contrast , IL-1β seems to have little ( if any ) role during the first 12-18h of the gut infection . Mice deficient in IL-1α/β signaling controlled the intraepithelial S . Tm load equally well as littermate controls [12] and exhibited wild-type cecum inflammation kinetics . Nonetheless , this does not formally exclude a redundant function for IL-1β in this initial phase or in other , non-redundant processes that may manifest only at later time points . IL-18 on the other hand is constitutively and highly expressed in the intestinal mucosa , especially by IECs [70–73] . It is thereby ideally positioned to mediate first-line responses to infection . Indeed , Il18-/- mice featured a delayed onset of inflammation . This was not attributable to perturbed homeostasis . Rather IL-18 was necessary for the initiation of mucosal inflammation . IL-18 levels of the infected cecal mucosa did critically affect disease kinetics , as decreasing or increasing IL-18 concentrations delayed or accelerated the mucosal response , respectively . IL-18 alone was , however , insufficient to elicit gut inflammation in the absence of stimuli from invasive wild type S . Tm . Thus , IL-18 must exert its function together with other signals whose nature remains to be elucidated . To address the cellular source of Il-18 , we have performed experiments in bone marrow chimeras ( S8 Fig ) . The data suggest that IL-18 from epithelial/stromal as well as bone-marrow derived cells may play a role . This would be well in line with enterocytes and resident lamina propria cells as sources contributing to the IL-18 driven inflammatory response . As an epithelial NAIP/NLRC4/caspase1-inflammasome appears chiefly responsible for early recognition of invading S . Tm [12] ) , it seems likely that infected enterocytes are a key source of this cytokine which may be further supplemented from myeloid sources ( e . g . resident macrophages and dendritic cell population; [74] ) . Our data suggest that NK cells are critical effectors for the IL-18 dependent mounting of gut inflammation . IL-18 stimulated this NK cell response in at least two ways , i . e . via IL-18R dependent recruitment and by enhancing NK cell activation . It is a common theme that cytokine and chemokine responses act in concert to recruit leukocytes to sites of infection , thereby enhancing a local inflammatory response [75] . In our system of acute bacterial mucosa infection , IL-18 a ) upregulated NK cell recruiting chemokines at the site of infection and b ) stimulated the migratory capacity of NK cells . Yet , the underlying molecular mechanism of the enhanced migratory phenotype seen in IL-18 stimulated NK cells still needs to be elucidated . One explanation could reside in IL-18-mediated regulation of NK cell surface receptor expression [76] . However , our data exclude a direct effect of IL-18 on the chemokine receptor CXCR3 expression . Nevertheless , one could envision a synergy of cytokine and chemokine signaling , where the priming stimulus of the cytokine amplifies the downstream signaling events of the chemokine receptor complex , leading to increased sensitivity and an enhanced migratory capacity of the stimulated cell [77–80] . In particular the increased migratory potential at lower chemokine concentrations may indicate that is indeed the case [77] . Besides an increased NK cell accumulation , we observed a higher maturation state of mucosal NK cells in presence of IL-18 , which is in line with previous findings addressing IL-18 function during NK cell responses upon viral infections [81] . The experimental design could not distinguish between an IL-18 dependent recruitment of pre-activated NK cells and an IL-18 induced in situ maturation of recruited NK cells . In any case , mature CD11b+ NK cells are generally associated with increased cytotoxicity and cytokine production [49 , 50] . Indeed , IFNγ production by NK cells was reduced to background levels in the absence of IL-18 . If also other NK cell activating cytokines ( especially IL-12; [82 , 83] ) may synergize with IL-18 in IFNγ production by NK cells [84] during early mucosal S . Tm infection , remains to be established . Surprisingly , early mucosal inflammation developed independent of IFNγ . This is in contrast to the important role of IFNγ during other bacterial gut infections as well as in later stages of S . Tm infection , where pathogen resistance , colitis and systemic pathogen restriction are coupled to a functional IFNγ response [85–87] . How do NK-cells promote the gut inflammation ? Our data implicate a functional NK cell Perforin response in the elicitation of cecal inflammation during the early mucosal phase of the disease . NK cell mediated cytotoxicity is a well-known effector mechanism in anti-viral defense [88] . Recent evidence suggests that NK cells can also limit the susceptibility to C . rodentium infection as well as systemic loads of some intracellular bacteria via Perforin-dependent effector mechanisms [6 , 61 , 89] . However , the control of systemic S . Tm infection ( at least at 12h p . i . ) does not require perforin , whereas Chromobacterium violaceum infection is efficiently limited [6] . Similarly , Perforin-mediated NK cell cytotoxicity does not significantly alter mucosal S . Tm loads ( at least at 12h p . i . ) . Much rather , it seems to promote the induction of mucosal inflammation . Of note , as shown by the NK cell transfer into Rag-/-γc-/-animals , NK cells alone may not able to fully restore mucosal pathology ( i . e . to WT C57/BL6 levels ) . This suggests that NK cells are necessary but may alone not be sufficient to induce full-blown inflammation in response to infection . Other cell types may also contribute to the process . First preliminary results might suggest that Perforin-mediated killing is able to enhance the inflammatory state of the cecal mucosa by inducing the release of soluble pro-inflammatory mediators like ATP from dying cells . This in turn could lead to the recruitment of other effector cells , like inflammatory monocytes , macrophages or neutrophils , to the infected mucosa [63–65] and subsequently induce mucosal tissue pathology . In addition to this ATP-dependent recruitment , previous studies have described a role for IL-18 in the recruitment of inflammatory monocytes to the infected LP via the induction of chemokine production from NK cells [90] . Moreover , we and others could show that IL-18 is able to stimulate the recruitment and activation of neutrophils [91 , 92] , a hallmark cell type in the infected mucosa with prominent effects during S . Tm infection [16 , 38 , 93–95] . Although our data suggests that neutrophils alone are not the driving force for early mucosal inflammation , they could play a redundant role in combination with other effector cells in the amplification of mucosal pathology . The IL-18/NK cell perforin axis may thus coordinate the recruitment and activity of a whole range of antimicrobial leukocyte responses . A detailed molecular analysis of how Perforin-mediated NK cell cytotoxicity fuels gut inflammation is an important topic for future work . Is the IL-18/NK cell/perforin axis of general importance for initiating gut mucosal defense ? So far , only few studies have focused on the initiating events and none has explored the entire mechanism . Nevertheless , there is accumulating circumstantial evidence suggesting a broader relevance . The importance for the NAIP/NLRC4/caspase-1 inflammasome in mucosal defense has been observed in S . Tm and Citrobacter rodentium infection models [11 , 12] . Pro-IL-18 is highly expressed under steady state conditions in the intestinal mucosa and therefore a likely source of mature IL-18 during the first hours of pathogen attack . However , it had remained unclear how IL-18 switches the mucosa to inflammation . In DSS models , the gut inflammation is affected by IL-18 . However , depending on the experimental protocols and the time points analyzed , IL-18 has enhancing or ameliorating effects [71 , 96 , 97] . This suggests a role for IL-18 in establishing inflammation or in regulation of the steady state defense , but does not answer if NK cell perforin responses are involved . Toxoplasma gondii gut infection has also provided evidence for the role of IL-18 in the mucosal response [98 , 99] . In this model , IL-18 produced by bone-marrow-derived cells was critical for mucosal pathology at days 3–5 of infection and the disease was ameliorated by daily application of the IL-18 inhibitor IL-18BP [98] . However , due to the long time required to establish overt disease pathology , one cannot discern primary inductive mechanisms from tonic , multipronged effects on the global tissue responsiveness . The S . Tm infection data described here is the first evidence implicating perforin in mounting mucosal inflammation ( which will likely help to control pathogen loads at later phases of the pathogen-host interaction ) . This would be in line with recent evidence from systemic Chromobacterium violaceum infection and colonic Citrobacter rodentium infection that implicated IL-18 elicited NK cell perforin as an essential effector mechanism limiting pathogen loads [6 , 61] . Thus , bacterial pathogens other than S . Tm are also affected by the IL-18/NK cell perforin axis . Overall , the available evidence suggests that the caspase-1 elicited IL-18/NK cell perforin responses may be of general importance for coordinating defenses , including the initial stages of microbial gut infection . In conclusion , our findings provide important new insights into the mounting of mucosal inflammation in the infected gut . In particular , they reveal the caspase-1/IL-18/NK cell axis as a central regulator of the initial disease kinetics and demonstrate the role of activated NK cell accumulation . These activated NK cells seem to promote inflammation via perforin-mediated cytotoxicity . It seems likely that this defense axis is of relevance for other enteropathogenic infections and that it coordinates additional branches of the mucosal defense . Our findings are an important step towards deciphering the multi-layered responses that protect the intestinal mucosa against microbial attack .
Salmonella Typhimurium SL1344 ( SB300 , SmR [100] ) was used as WT . S . Tm-pssaG-GFPmut2 has been previously used ( e . g . [101] ) , S . Tmavir ( M557 , ΔinvG; sseD::aphT [31] ) is an isogenic derivative of SL1344 . For infections , the bacteria were grown in LB/0 . 3M NaCl for 12h , subcultured at a 1:20 dilution for 4h , spun down and resuspended in PBS , pH 7 . 4 . All mice were maintained as specific pathogen-free in individually ventilated cages at the Rodent Center RCHCI ( ETH Zürich ) or the ETH Phenomics Center EPIC ( ETH Zürich ) . C57BL/6Ptprcb mice ( congenic marker Ly5 . 2+ ) originating from Charles River ( Sulzfeld , Germany ) were used as wild-type mice . For generation of bone marrow chimeras , B6 . SJL-PtprcaPepcb mice ( congenic marker Ly5 . 1+ ) were used as wild type . Knockout mouse lines Casp1/11-/- ( B6 . 129S2-Casp1tm1Sesh [102] ) , Casp11-/- ( B6 . Casp11tm1 [103] ) Il1ab-/- ( B6 . D-IL1atm1Yiw/IL1btm1Yiw [104] ) , Il18-/- ( B6 . 129P2-Il18tm1Aki [105] ) , Il18r1-/- ( B6 . 129P2-Il18r1tm1Aki [106] ) , Ifng-/- ( B6 . 129S7-Ifngtm1Ts [107] ) , Ifngr1-/- ( B6 . 129S7-Ifngr1tm1Agt [108] ) , Prf1-/- ( C57BL/6-Prf1tm1Sdz [56] ) and Rag2-/-γc-/- ( C57BL/6-Rag2tm1Fwa . Il2rgtm1Cgn[109 , 110] ) were all of C57BL/6 background . For experimentation , wild type ( +/+ ) , heterozygous ( +/- ) and homozygous knockout ( -/- ) littermates were obtained by backcrossing into C57BL/6 and genotypes were verified by PCR . S . Tm infections were performed as described previously [15] . In brief , 7–12 week old mice were pretreated with 25mg/animal streptomycin sulfate ( Applichem ) by gavage . 24 h later mice were infected with 5×107 CFU S . Tm by gavage . Infections were performed for 6h , 8h , 12h , 18h , 36h or 72h , as indicated . Bacterial loads in gut luminal content , mLN , spleen and liver were determined by plating . For in vivo treatment with murine rIL-18 , mice were injected intraperitoneally with a single dose of murine r-IL18 ( 120μg/kg , MBL ) at the time of infection . For in vivo neutralization of IL-18 , mice received intraperitoneal injections of human r-IL18BP ( 2mg/kg , Life Technologies Europe ) at the time of infection . NK1 . 1+ cells were depleted by intraperitoneal injection of anti-NK1 . 1 ( 10mg/kg , clone PK136 , BioXCell ) on two consecutive days , starting one day prior to infection . In addition , NK cells were depleted by intraperitoneal injection of anti-asialo GM1 antiserum ( 50uL/mouse; Wako ) ; mice were injected 3 times in total , starting 3 days prior to infection . Neutrophils were depleted by a daily intraperitoneal injection of anti-G-CSF ( 0 . 4mg/kg , clone 67604 , R&D Systems ( Abingdon , UK ) ) starting one day prior to infection in combination with a single dose of anti-Ly6G ( 6mg/kg , clone 1A8 , BioXCell ) at one day prior to infection . CD8+ cells were depleted by intraperitoneal injection of anti-CD8 ( 8mg/kg , clone , BioXCell ) 3 days and 1 day prior to infection . For adoptive transfer experiments , splenic NK cells from WT BL6 or Prf1-/- mice were isolated by MACS using the murine NK cell isolation Kit II . Recipient Rag2-/-γc-/- animals received 5x105 purified NK cells by i . v . injection , infections were conducted 7 days post transfer . Generation of bone marrow chimeras has been described before [111] . Briefly , Il18-/- , Il18r1-/- Ly5 . 2 and C57BL/6 Ly5 . 1 donor mice were euthanized and bone marrow was extracted from tibia and femur . Recipient mice ( C57BL/6 Ly5 . 1 or Il18-/- ) were γ-irradiated ( 1000rad ) and reconstituted intravenously with 5x106 WT , 5x106 Il18-/- or a 1:1 mixture of 2 . 5x106 Il18r1-/- Ly5 . 2 and 2 . 5x106 C57BL/6 Ly5 . 1 bone marrow cells . Mice were kept on Borgal ( Veterinaria AG ) for 3 weeks and were infected 8 weeks after reconstitution . Reconstitution efficiency was controlled by flow cytometry ( Ly5 . 1/CD45 . 1 , Ly5 . 2/CD45 . 2 staining ) on LP and blood cells . Cecum tissue was frozen in OCT ( comp ) and stored at -80°C . 5μm cryosections were mounted on glass slides , air-dried and stained with hematoxylin and eosin . Histopathology was evaluated in a blinded manner as described previously [15] , scoring submucosal edema , epithelial integrity , goblet cell number and polymorphonuclear leukocyte infiltration , resulting in a total pathological score between 0 ( uninflamed ) and 13 ( maximally inflamed ) . Mice were sacrificed and the cecum was opened longitudinally and washed in ice cold PBS to remove the remaining cecal content . In order to dislodge the epithelial cells , the cecum tissue was cut into small pieces , and placed in two subsequent rounds into PBS containing 5mM EDTA , 15mM HEPES and 10% FCS , incubated for 20min at 37°C , and shaken mildly . Samples were washed in RPMI supplemented with 30% FCS and digested for 1h in RPMI containing 1mg/mL Collagenase VIII ( Sigma ) and 0 . 2mg/mL DNase I ( Roche ) . Cells were filtered through a 70μm cell strainer and rinsed in RPMI . Isolated cells were loaded onto a NycoPrep 1 . 077 matrix ( Progen ) and centrifuged for 30 min at 400g . Cells were collected from the interface and washed in RPMI . Cell suspensions were stained in PBS containing 1% FCS and 0 . 02% sodium azide . For intracellular cytokine staining , isolated lamina propria cells were incubated for 5h in RPMI containing 5% FCS and 10μg/mL BrefeldinA ( Sigma ) at 37°C/5%CO2 . After surface staining , cells were fixed in fix/perm solution ( eBioscience ) for 30min at 4°C and stained intracellular in wash/permeabilization buffer ( eBioscience ) . Antibodies were either from Biolegend , i . e . CD45 ( clone 30-F11 ) , CD3 ( clone 17A2 ) , CD4 ( clone GK1 . 5 ) , CD8α ( clone 53–6 . 8 ) , CD8β ( clone YTS156 . 7 . 7 ) , CD27 ( clone LG3 . A10 ) , IFNγ ( clone XMG1 . 2 ) , CD11b ( clone M1/70 ) , CD90 . 2 ( clone 30-H12 ) , NKp46 ( clone 29A1 . 4 ) , CD45 . 1 ( clone A20 ) , CD45 . 2 ( clone 104 ) , TCRγδ ( clone GL3 ) , CD122 ( clone TM-β1 ) , KLRG1 ( clone 2F1/KLRG1 ) , Eomes ( clone Dan11mag ) or from BD Biosciences i . e . NK1 . 1 ( clone PK136 ) and TCRb ( clone H57-597 ) . For detection of 5-ethynyl-2’-deoxyuridine ( EdU ) incorporation , the Click-iT EdU Alexa Fluor 488 Flow Cytometry Assay Kit ( Invitrogen ) was used . Data were acquired on a LSRII ( BD Biosciences ) and analyzed using FlowJo software ( TreeStar ) . Cell sorting was performed on a FACS Aria III ( BD Biosciences ) and equal amounts of CD45+ CD3- NK1 . 1+ and CD45+ CD3- NK1 . 1- cells were collected separately in two flow tubes . In vivo NK cell proliferation was measured by intraperitoneal injection of EdU ( 400μg/mouse ) 12h prior to sacrifice . Single cell suspensions were obtained from cecal tissue by cecal lamina propria isolation as described above . Incorporated EdU was detected by fluorescent-azide coupling reaction according to the manufacturer’s protocol ( Invitrogen ) and analyzed by flow cytometry . NK cells were isolated from spleens of naïve C57BL/6 mice by MACS using the murine NK cell isolation Kit II . Isolated spleenocytes were cultured for 3h in presence or absence of 100ng/mL rIL-18 and cell migration was assessed by the 24-well Transwell Systems and polycarbonate filters with a pore size of 5μm ( Corning Costar ) . Briefly , 1x105 cells were allowed to migrate for 3h to the lower compartment containing different concentrations of CXCL9 . Migrated cells were harvested and cell numbers were determined by flow cytometry . Cecum tissue was fixed in 4% paraformaldehyde/4% sucrose , saturated in PBS/20% sucrose , embedded in optimum cutting temperature medium ( Tissue-Tek ) , flash-frozen in liquid nitrogen , and stored at −80°C; 20 μm cryosections were air-dried , rehydrated with PBS , permeabilized ( PBS/0 . 5% Triton X-100 ) , and blocked ( PBS/10% Normal Goat Serum ) . Antibody stainings included anti-ICAM-1/CD54 ( clone 3E2 , Becton Dickinson ) , anti-CD18 ( clone M18-2 , Biolegend ) , appropriate secondary reagents , AlexaFluor647-conjugated phalloidin ( Molecular Probes ) , and DAPI ( Sigma Aldrich ) . Samples were mounted with Mowiol ( Calbiochem ) . A Zeiss Axiovert 200 m microscope with 10×–100× objectives , a spinning disc confocal laser unit ( Visitron ) , and two Evolve 512 EMCCD cameras ( Photometrics ) were used for imaging . Postcapture processing and analysis used the Visiview ( Visitron ) and Image J ×64 . For quantification of S . Tm in the epithelium , 20 μm cross-sections were stained for ICAM-1 , phalloidin , and DAPI and intracellular S . Tm-GFP+ were manually enumerated blindly in six to nine nonconsecutive sections/mouse . All data represent averages/section . For analysis of mRNA expression , total RNA was extracted from homogenized cecum tissue using the RNeasy Mini Kit ( Qiagen ) . Total RNA from sorted cells was extracted using the RNeasy Micro Kit ( Quiagen ) . For reverse transcription , 1μg total RNA was transcribed using the RT2 HT First Strand Kit ( Qiagen ) . RT-qPCR was performed using Custom RT2 Profiler Arrays ( Qiagen ) or RT2 qPCR Primer Assays ( Qiagen ) with RT2 SYBR Green ROX FAST ( Qiagen ) on an Applied Biosystems 7900 HT Fast Real-Time PCR Cycler . Relative mRNA expression was calculated using the ΔCt method , using Actb as reference gene [112] . Cecum tissue was washed in ice-cold PBS to remove remaining luminal content . The tissue was homogenized in PBS containing 0 . 5% Tergitol and lysates were cleared by centrifugation . IL-1β and IFN-γ concentrations were determined using the respective Cytometric Bead Array Mouse Flex Sets ( BD Biosciences ) with the CBA Mouse/Rat Soluble Protein Master Buffer Kit ( BD Biosciences ) . IL-18 concentrations were measured by employing the IL-18 ELISA Kit ( MBL ) according to the manufacturer's protocol . All animal experiments were subject to the Swiss animal protection law ( TschG ) and therefore reviewed independently by a dedicated cantonal committee ( Tierversuchskommission ) and approved by the "Kantonales Veterinäramt , Zürich" ( licenses 223/2010 and 222/2013 ) .
|
Salmonella Typhimurium is a common cause of foodborne diarrhea . The disease symptoms arise already a few hours after infection . However , it had remained unclear how the immune system can mount the responses eliciting the disease symptoms so quickly . Earlier work in a mouse model had shown that the gut epithelium expresses a sensor , called NAIP/NLRC4/caspase-1 inflammasome that can detect the pathogen and mount a defense by 12-18h p . i . However , it has remained uncharacterized how inflammasome sensing drives the initial gut inflammation . Here , we found that the caspase-1 inflammasome triggers the production of IL-18 , a pro-inflammatory cytokine that appears essential for the early onset of inflammation . IL-18 is driving the accumulation of NK cells into the infected mucosa , via the upregulation of NK cell chemoattractants and by the stimulation of their migratory capacity . Mature NK cells seem to induce mucosal inflammation via a perforin-mediated cytotoxic response . These data suggest that the inflammasome/IL-18/NK cell axis is a driver of early mucosal inflammation via a perforin-dependent cytotoxic NK cell response . Future work will have to address , if this mechanism is equally potent in the human gut and may contribute to ramping up the host's response during the first hours of infection . This may have implications for other gut infections and might provide leads for developing therapies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"flow",
"cytometry",
"anatomical",
"pathology",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"cytokines",
"immunology",
"cytopathology",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"signs",
"and",
"symptoms",
"molecular",
"development",
"neutrophils",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"mouse",
"models",
"immune",
"response",
"spectrophotometry",
"immune",
"system",
"gastrointestinal",
"tract",
"cytophotometry",
"diagnostic",
"medicine",
"cell",
"biology",
"anatomy",
"nk",
"cells",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"spectrum",
"analysis",
"techniques",
"cecum"
] |
2016
|
An NK Cell Perforin Response Elicited via IL-18 Controls Mucosal Inflammation Kinetics during Salmonella Gut Infection
|
Effective transboundary conservation of highly migratory marine animals requires international management cooperation as well as clear scientific information about habitat use by these species . Populations of leatherback turtles ( Dermochelys coriacea ) in the eastern Pacific have declined by >90% during the past two decades , primarily due to unsustainable egg harvest and fisheries bycatch mortality . While research and conservation efforts on nesting beaches are ongoing , relatively little is known about this population of leatherbacks' oceanic habitat use and migration pathways . We present the largest multi-year ( 2004–2005 , 2005–2006 , and 2007 ) satellite tracking dataset ( 12 , 095 cumulative satellite tracking days ) collected for leatherback turtles . Forty-six females were electronically tagged during three field seasons at Playa Grande , Costa Rica , the largest extant nesting colony in the eastern Pacific . After completing nesting , the turtles headed southward , traversing the dynamic equatorial currents with rapid , directed movements . In contrast to the highly varied dispersal patterns seen in many other sea turtle populations , leatherbacks from Playa Grande traveled within a persistent migration corridor from Costa Rica , past the equator , and into the South Pacific Gyre , a vast , low-energy , low-productivity region . We describe the predictable effects of ocean currents on a leatherback migration corridor and characterize long-distance movements by the turtles in the eastern South Pacific . These data from high seas habitats will also elucidate potential areas for mitigating fisheries bycatch interactions . These findings directly inform existing multinational conservation frameworks and provide immediate regions in the migration corridor where conservation can be implemented . We identify high seas locations for focusing future conservation efforts within the leatherback dispersal zone in the South Pacific Gyre .
Leatherback turtles ( Dermochelys coriacea ) in the eastern Pacific ( EP ) have exhibited population declines of up to 90% during the past two decades [1 , 2] . These declines have been driven by a number of factors , including incidental mortality in fisheries , loss of nesting habitats , and unsustainable egg harvest [1 , 3] . Of the extant leatherback nesting beaches in the EP , Playa Grande in Parque Nacional Marino Las Baulas ( PNMB ) , Costa Rica , supports the largest nesting colony [1] . After the nesting period ( approximately 60 d ) , EP leatherbacks perform long-distance migrations from breeding areas to feeding areas , where they remain for 2 to 7 y [4] . Therefore , while protection of nesting habitat is important to enhance recruitment into the population , an improved understanding of the at-sea distribution and movements of EP leatherbacks is vital to ensuring their long-term survival . In particular , long-range tracking studies using electronic tags can inform conservation efforts by identifying high-use areas for leatherbacks in time and space , as well as environmental influences on leatherback behavior [5] . Leatherback turtles globally undertake long-distance migrations over thousands of kilometers [6–14] . Morreale et al . [6] first described the movements of EP leatherbacks from the tracks of eight turtles ( durations 3–87 d ) and identified a persistent southbound migration corridor from PNMB toward the Galápagos Islands . Additional tagging efforts at a nesting beach in Mexiquillo , México , about 965 km north of Costa Rica , revealed that leatherbacks traveled routes that shared the same directional heading and general high seas habitats in the eastern South Pacific as those traveled by Costa Rican turtles [7] . In contrast , leatherbacks from other populations demonstrate inter-individual behavioral variation with respect to post-nesting migration routes [8–10 , 13 , 14] . The apparent persistence of the EP leatherback migration pattern provides a unique opportunity to generate a cohesive conservation management approach for this endangered population . Conservation of highly migratory marine species requires international cooperation for implementation of transboundary management strategies . Specifically , information on movements and distributions of large marine predators collected by electronic tracking devices can provide guidance to the development of national and multinational fisheries management strategies and bycatch mitigation efforts , as well as support related policy efforts [15] . One such framework is the Eastern Tropical Pacific Seascape ( ETPS ) initiative [16] , which is a multinational coordination of marine resource management within the combined exclusive economic zones of Costa Rica , Panama , Colombia , and Ecuador . The ETPS is an area that is home to several marine protected areas ( MPAs ) ( e . g . , PNMB ) and World Heritage sites ( e . g . , Cocos Island , Coiba Island National Park , Malpelo Island , Galápagos Islands and Marine Reserve ) . Thus , the ETPS represents a framework through which habitat use and movement data for migratory animals , such as leatherbacks , can be translated into tangible management actions . Here we present the largest multi-year tracking data set collected for this species , based on 46 individuals satellite-tagged during 2004–2007 at PNMB . Our approach is consistent with a recent review [17] , which emphasized the importance of tracking large sample sizes and an interdisciplinary approach integrating oceanographic cues with behavior . These data enabled us to ( 1 ) describe the distribution and horizontal movements of leatherbacks in the EP , ( 2 ) examine the influence of oceanic currents on leatherback migrations , ( 3 ) assess leatherback high-use habitats , ( 4 ) confirm and elucidate a leatherback migration corridor from the nesting beach to 5 °S , and ( 5 ) describe leatherback movements beyond 10 °S into the South Pacific . In addition , these data identify critical areas for directed conservation efforts to ensure the survival of this species in the EP .
We tagged 46 female leatherback turtles during oviposition , resulting in 12 , 095 tracking days spanning 21 January 2004–5 July 2007 , with a mean track duration of 263 d , a distance of 8 , 070 km , and a travel speed of 37 . 7 km d−1 ( Table 1 ) . Movements by cohorts from a given year displayed cohesion , even though initiation of the post-nesting migration among individuals differed by up to several weeks ( Figure 1 ) . Only one individual tagged in 2005 ( tag ID 56280 ) remained in coastal waters off Costa Rica and Panama for the entire tag duration ( Figure 1A ) . Upon completion of nesting activity , leatherbacks embarked on rapid ( 42 . 9 km d−1 , standard deviation ( sd ) = 27 . 7 km d−1 ) directed southward migrations through the equatorial region . Once south of 5 °S , the turtles dispersed throughout the South Pacific Gyre following slower ( 23 . 8 km d−1 , sd = 16 km d−1 ) , meandering paths , and remained there through the duration of the tracking period ( Figure 2A–2C ) . Across their migrations , turtles experienced a wide range of surface temperatures ( 11 . 2–32 . 7 °C , mean = 25 . 2 °C , sd = 3 . 2 °C; Table 1 ) . They encountered areas of high–eddy kinetic energy ( EKE ) in the equatorial region ( >100 cm2s−2 ) , and areas of very low EKE ( <50 cm2s−2 ) in the dispersal region ( Figure 2B ) . Likewise , chlorophyll-a ( CHL ) concentrations were highest in the equatorial region ( >0 . 3 mg m−3 ) , and lowest in the South Pacific Gyre ( <0 . 1 mg m−3 ) ( Figure 2C ) . Swimming speed was significantly higher in areas of high CHL and vice-versa ( linear regression: β = 0 . 964 ± 0 . 057 , F1 , 9577 = 281 , p < 0 . 001 , r2 = 0 . 029 ) , although the association between these two variables was weak . Ocean current energetics had a major impact on the turtles' migration route . Between latitudes 12 °N and 5 °S , southbound turtles negotiated the strong alternating eastward-westward flows of the equatorial current system , whose strength can be of comparable magnitude to turtle travel speeds ( Text S1 and Figure S1 ) . Turtles initially moved rapidly WSW ( mean speed = 63 . 8 km d−1 , sd = 31 . 8 km d−1; mean heading = 247° , sd = 40° ) through a narrow zone of low mean kinetic energy ( MKE ) near 10 °N between the southern edge of the Costa Rica Dome ( CRD ) and the Costa Rica Coastal Current ( CRCC ) ( Figure 3A–3C , Text S1 , and Figure S1 ) . They then crossed the energetic flow along the southern edge of the CRD between 8 °N and 6 °N on a SE heading ( mean speed = 49 . 9 km d−1 , sd = 27 . 8 km d−1; mean heading = 173° , sd = 42° ) . Once outside the CRD , turtles turned WSW ( mean speed = 50 . 2 km d−1 , sd = 26 . 4 km d−1; mean heading = 225° , sd = 43° ) over another area of low MKE near 4 °N and continued rapidly on this course aided by the westward-flowing northern branch of the South Equatorial Current ( SEC ) near 3 °N . Between 1 °N and 2 °S , turtles turned southward ( mean speed = 41 . 7 km d−1 , sd = 22 . 9 km d−1; mean heading = 186° , sd = 46° ) , as they rapidly crossed the Equatorial Undercurrent ( EUC ) by again increasing their southward speed , even while being advected eastward by the EUC ( Figure 3A–3C ) . A final SW turn ( mean speed = 43 . 5 km d−1 , sd = 17 . 8 km d−1; mean heading = 196° , sd = 37° ) occurred as the turtles crossed the much weaker southern branch of the westward SEC between 3 °S and 5 °S . Examination of the ratio of turtle meridional velocity to current zonal velocity in the 12 °N–5 °S region revealed that in areas of strong currents , the turtles responded by increasing their southward velocity regardless of flow direction ( i . e . , ratios consistently close to zero in the 8 °N–6 °N and 4 °N–1 °N latitudinal bands; Figure 3D ) . After the effect of the currents was removed ( Figure 3F ) , the tracks appeared much straighter for all years , showing a consistent SSW heading between 12 °N and 1 °N ( mean = 193° , sd = 30° ) and a southward heading afterward . The contours of geomagnetic force were generally oriented NE-SW while the contours of geomagnetic inclination were generally oriented NW-SE , forming a grid pattern in the 12 °N–5 °S region ( Figure 3F ) . The current-corrected tracks generally crossed these magnetic gradients from north to south ( Figure 3F ) .
This multi-year dataset confirmed the existence of a persistent migration corridor for leatherbacks spanning from the Pacific coast of Central America , across the equator and into the South Pacific . The turtles traveled along a predominantly southwesterly heading , which was strongly influenced by ocean currents . An earlier telemetry study hypothesized a leatherback migration corridor between Costa Rica and Galápagos that could be influenced by environmental factors such as ocean fronts , bathymetric features , currents , or geomagnetic cues [6] . We examined each of these hypotheses with our larger sample size , and found no relationship between the southward turtle movements and the most prominent frontal feature in the region , the Equatorial Front , which runs east to west just north of the equator ( Text S1 ) . We also found no consistent association between leatherback tracks and the Cocos Ridge , the dominant bathymetric feature in the region , even after current correction ( mean turtle heading = 193° versus 224° if they had followed the orientation of the Ridge ) . Instead , the turtles' movements over the Cocos Ridge were correlated with the current strength of the southern edge of the CRD , which deflected them over portions of the Ridge each year ( Figure 3A–3C ) . Once the influence of currents was removed , it was apparent that the turns observed in the tracks in the corridor region ( 12 °N–5 °S ) were current-induced . For these reasons , we conclude that navigation through the complex and highly energetic equatorial region supports the existence of a compass sense , possibly guided by the geomagnetic map ( Figure 3F ) formed by the force and inclination fields in the region , as has been documented for sea turtles in other parts of the world [18] . Our results demonstrate that leatherbacks responded to strong zonal currents by increasing their southward speed , probably to maintain their SSW headings and to avoid being pushed too far eastward or westward from their destination in the South Pacific Gyre . Inter-annual variability in current location and strength was a major force shaping the turtles' migration routes . In each year , the migration corridor was initially constrained to the zone of lowest MKE associated with the center of the CRD ( Figure 3A–3C ) and , ultimately , the breadth of their dispersal within the South Pacific Gyre ( Figure 2B ) was determined by the strength of the equatorial currents through which they migrated . This was particularly evident in 2005 , when the currents were weaker , and in 2007 , with stronger currents ( Figure 3B and 3C ) . This interaction between post-nesting EP leatherbacks and currents contrasts with that of South African leatherbacks [19] , whose variable long-distance movements suggest passive drift with prevailing currents in the Southwest Indian Ocean . Leatherbacks moved rapidly through the productive equatorial region [20 , 21] and then dispersed in the most oligotrophic region of the Pacific Ocean [22] . The slow , meandering movements by the turtles in the South Pacific Gyre suggest that post-nesting female leatherbacks probably migrate there to forage . Leatherback turtles do not feed directly on phytoplankton but on large gelatinous zooplankton [23] , and despite its low phytoplanktonic biomass , the South Pacific Gyre ecosystem sustains an ample mesozooplanktonic forage base and a substantial longline tuna fishery [24 , 25] . Therefore , we suggest that following the energetic demands of egg production , it may be more efficient for post-nesting EP leatherbacks to forage in an oceanographically quiescent region within which high water clarity could enhance prey detection [26] while requiring minimal swimming effort . A further possible explanation for the consistency in migration routes followed by EP leatherbacks is that present-day migration patterns do not reflect the historic diversity of migration strategies , such as that observed in other leatherback populations [13 , 14 , 19 , 27] . The leatherback tracks presented here , which document the first 12–18 mo of the entire ∼4-y remigration interval , suggest that post-nesting EP turtles almost exclusively occupy oceanic areas . Eckert and Sarti [7] also tracked EP leatherbacks to oceanic areas , but a few of these turtles moved into coastal areas off South America . A single turtle in this study ( tag ID 56280 , tagged during 2005 ) occupied exclusively nearshore foraging habitats along the coast of Central America throughout the entirety of its tracking duration ( 562 d ) . Previous reports have indicated substantial leatherback bycatch in nearshore fisheries in the EP , specifically in swordfish driftnets off Chile and Peru [7 , 28 , 29] , and leatherbacks continue to interact with fisheries in Peruvian [30] and Chilean waters [31] . Given that coastal areas in the EP represent highly productive areas when compared with oceanic areas , Saba et al . [32] hypothesized that this bycatch could have essentially extirpated a “coastal” migratory phenotype in this population . Sustained tracking efforts on the EP population , including continuous tracking studies on previously tagged remigrant turtles , and tagging of turtles in foraging habitats are necessary to test this hypothesis . Characterization of spatio-temporal habitat use is a fundamental element of effective biodiversity conservation management strategies . Our results have enabled us to define two high-use areas for leatherback turtles in the EP: ( 1 ) an oceanic post-nesting migration corridor shaped by currents , and ( 2 ) and putative foraging grounds in the South Pacific Gyre . The data provide compelling new strategies for conservation of Pacific leatherbacks ( Box 1 ) , which could also benefit other marine species . First , we encourage enhanced regional and international cooperation in management of leatherbacks and their migration corridor occurring within existing MPAs ( i . e . , PNMB , Cocos Island , Coiba Island , Galápagos Islands ) and conservation initiatives ( i . e . , ETPS ) . Because much of the leatherback dispersal region occurs within international waters , multinational organizations and policy instruments ( i . e . , Inter-American Convention for the Protection and Conservation of Sea Turtles , Inter-American Tropical Tuna Convention , Convention on Highly Migratory Species , Comisión Permanente del Pacífico Sur , United Nations Convention on the Law of the Sea , and South Pacific Regional Fisheries Management Organization ) should be leveraged to achieve turtle management and conservation outcomes on the high seas . The leatherback migration corridor , which occurs during a period of a few months ( February–April , Figure 3H ) , is largely contained within the boundaries of the ETPS , which also includes PNMB and the coastal areas used by one of the tagged turtles . This affords an opportunity for each of the governments involved in these initiatives to actively participate in the spatio-temporal management of leatherbacks as they occupy the network of MPAs in this region ( Figure 4 ) . Second , we strongly recommend enhanced and increased collection of fisheries-dependent bycatch data and fisheries-independent habitat use data throughout the EP . In particular , expanded and improved observer coverage in EP fisheries would be an important step to characterize leatherback interactions with fisheries operations in the EP , because currently available information on leatherback bycatch is inconsistently collected for most fisheries . Collection of fisheries bycatch data for leatherbacks is critical to evaluating relative effects of distinct fisheries on leatherback mortality . Leatherback interactions with small-scale fisheries ( i . e . , artisanal , traditional , subsistence ) may be especially critical in coastal habitats where extremely high sea turtle bycatch rates have been observed [7 , 33] . For fisheries-independent information , further electronic tagging efforts on EP leatherbacks and other highly migratory marine species that share similar high seas habitats and face common human threats would improve effectiveness of adaptive conservation schemes . Our data represent only the initial segment of the entire nonreproductive period for female leatherbacks , leaving much of their at-sea behavior and habitat use unexplored . Therefore , a priority for future leatherback tagging studies should be to focus on foraging ground behavior and movements throughout the entire nonreproductive period . Third , improved data on fisheries bycatch and leatherback habitat use in the EP as outlined above would inform planning of dynamic time-area closures and/or appropriate gear modifications intended to reduce turtle interactions with fisheries . A current illustration of this approach is in the USA California/Oregon-based drift-gillnet and longline fisheries , where a time-area closure was implemented based on temporal and spatial patterns of leatherback distributions; leatherback bycatch was reduced to zero following implementation of this measure [34] . This tracking study defines at least two regions within the Pacific where strategic time-area closures could be a useful management tool for protecting leatherbacks within high-use habitats . The first area is the post-nesting migration corridor spanning an open-ocean region from the Pacific coast of northwest Costa Rica ( approximately 12 °N ) to approximately 5 °S , within which turtles are seasonally concentrated , in predictable patterns , during the period of February–April . The second high-use area is defined by the predictable association of turtles during their dispersal phase ( through putative foraging habitats ) with the low EKE region of the South Pacific Gyre . Time-area closures could be applied to protect post-nesting turtles when they are seasonally concentrated during migration ( i . e . , while moving through the ETPS ) and within international waters on the high seas during foraging periods within low EKE regions of the South Pacific Gyre . Another specific opportunity to manage turtles during dispersal occurs when they pass through the oceanic island territories of Chile ( i . e . , Easter Island , Juan Fernandez Islands , and the Desventuras Islands ) . Management actions within each of the above regions should be based upon spatio-temporal overlap of the leatherback high-use areas and areas of high bycatch . In addition to establishing time-area closures , new technologies such as vessel monitoring and tracking systems ( VMS ) combined with continuous satellite tagging of EP turtles ( for near–real time high-use data ) could further mitigate human interactions with leatherbacks . The implementation of time-area closures to protect leatherbacks in the South Pacific would provide parallel conservation benefits for other marine species [35–42] whose movements through the ETP region have also been revealed by satellite tracking and other data . For example , recent satellite tracking studies on green turtles ( Chelonia mydas ) from Galápagos have indicated that February–April closures to protect green turtles would also benefit migrating leatherbacks that use similar migratory and foraging habitats with the EP [42] . Large-scale electronic tagging studies will increasingly play an important role in informing spatio-temporal management activities of coastal and pelagic habitats for threatened marine species [5] . Our results elucidate the oceanic behaviors of leatherback turtles , and are applicable to existing and future conservation strategies that promote the recovery of EP leatherbacks ( Box 1 ) . In the future , animal movement models derived from satellite-tag data , in combination with real-time oceanography [20 , 32 , 43] will provide managers with the ability to predict the movement patterns of leatherback turtles and to take effective conservation actions to protect them at sea .
Leatherback sea turtles ( n = 36 ) were instrumented with Sea Mammal Research Unit ( SMRU ) Satellite Relay Data Logger ( SRDL ) tags during 2004 ( n = 17 ) , 2005 ( n = 8 ) , and 2007 ( n = 11 ) . The SRDL tags were programmed to collect and transmit position , temperature , dive data , and tag diagnostic information [24] . We tagged ten additional turtles in 2004 with Wildlife Computer Smart Position Only ( SPOT ) tags , which were programmed to provide position data . We mounted the satellite transmitters on the turtles during oviposition using a harness technique [44] . Data from the tags were transmitted via the ARGOS satellite system [45] . We extracted tag-derived surface temperature measurements from the temperature-at-depth data transmitted by the SRDL tags . Surface was considered to be the first depth bin ( mean = 5 . 1 m , sd = 0 . 7 m ) . A total of 5 , 787 temperature measurements were available after discarding 105 records because the first depth was missing , had a negative value , or had spurious position values . We generated final position estimates at regular 6-h intervals using state-space models ( SSMs ) [46 , 47] that were applied to the raw unfiltered satellite data to improve position accuracy and to align with SMRU summary dive data . The application of a switching SSM provided the capacity to discern between two behavioral modes based on a first-difference correlated random walk . The location of the switch between these two behavioral modes was used to objectively define the transition from inter-nesting ( “mode 2” ) to the post-nesting migration ( “mode 1” ) [47] . In cases where a clear switch was not present , we used a sudden change in the travel speed to determine the cut-off . For this paper , we only used the post-nesting portion of the tracks . Median daily speeds and headings were calculated from the interpolated tracks via first differencing consecutive points . The observed track of an animal at any given time is the result of the animal's movement ( swimming ) plus the displacement caused by ocean currents ( drift ) . The true behavior of a turtle can be thus obtained by removing the influence of currents on the animal's trajectory . We used surface current estimates obtained from the sum of the geostrophic and Ekman components , as measured by satellite [48] , and removed them from the turtle movements at the locations generated with the SSMs . Within the equatorial band ( 4 °N–4 °S ) , a β-plane solution was applied [49] . We produced gridded utilization distribution maps [50] using a mesh size of 100 km2 and a fixed-kernel search radius of 0 . 5° for all years combined . The 95% utilization contour was used to define turtle high-use regions throughout the eastern tropical and South Pacific and the 75% contour to delineate the migration corridor between latitudes 12 °N and 5°S . We characterized the energetics of large-scale currents and their mesoscale fluctuations in the eastern tropical and South Pacific using merged satellite altimeter measurements of absolute dynamic topography and associated sea-level anomalies [51] . These data are generated by the Archiving , Validation , and Interpretation of Satellite Oceanographic data ( Aviso ) project at 1/3° resolution . Within five degrees of the equator , the Aviso product applies a β-plane solution [49] to obtain velocity and velocity anomaly vectors . We computed MKE from the mean u- and v- components of the geostrophic velocity as MKE = 0 . 5* ( <u2> + <v2> ) . These calculations were performed separately for the February–April period of each tracking year , since the emphasis was on assessing the impact of inter-annual variability in geostrophic current strength on turtle migration while crossing the equatorial region . On the other hand , we computed EKE as a long-term mean for the period 14 October 1992–18 April 2007 from the mean geostrophic velocity anomalies ( u′ and v′ ) , as EKE = 0 . 5* ( <u′2> + <v′2> ) . In this case , the emphasis was on examining turtle distribution in relation to a region of low mesoscale variability in the South Pacific Gyre . The distribution of phytoplankton standing stock is a useful indicator of biogeography and ecosystem structure [24] . Near-surface CHL concentration , a proxy for phytoplankton standing stock , was obtained from Sea-viewing Wide Field-of-view Sensor ( SeaWiFS ) satellite ocean-color observations at 9-km resolution . We computed a long-term mean for the period September 1997–March 2007 for comparison of turtle movements in relation to phytoplanktonic biomass distribution throughout their range . Individual 8-d averages were also obtained for each turtle median daily position . The relationship between CHL and the turtles' median daily speed was investigated using linear regression , after log- and square-root-transformation , respectively , to meet normality assumptions . We extracted bathymetry from the global sea-floor topography of Smith and Sandwell [52] , version 8 . 2 ( November 2000 ) ( http://topex . ucsd . edu/WWW_html/mar_topo . html ) . This dataset combines all available depth soundings with high-resolution marine gravity information provided by the Geosat , ERS-1/2 , and TOPEX/Poseidon satellite altimeters , and has a nominal resolution of 2 arc min ( ∼4 km ) . The 2000-m isobath was extracted from this dataset to obtain the outline of the Cocos Ridge , the most prominent bathymetric feature in the migration corridor region ( 12 °N–5 °S ) running northeast ( ∼43° azimuth ) for ∼1 , 200 km between Galápagos and Central America . Data on Earth's magnetic field ( force and inclination ) in the study area were calculated using the software GeoMag 6 . 0 , available from the NOAA National Geophysical Data Center ( http://www . ngdc . noaa . gov/seg/geom_util/geomutil . shtml ) , and the most recent ( 2005 ) International Geomagnetic Reference Field 10th generation ( IGRF-10 ) coefficients .
|
Highly migratory marine animals routinely cross international borders during extensive migrations over thousands of kilometers , thus requiring conservation strategies with information about habitat use and movement patterns . Critically endangered leatherback turtles ( Dermochelys coriacea ) in the eastern Pacific have suffered a severe population decline in recent years . In this study , we present the largest multi-year satellite tracking data set for leatherback turtles ( n = 46 turtles , 12 , 095 days ) to describe the migrations , habitats , and dispersal of female leatherbacks tagged at Playa Grande , Costa Rica . Leatherbacks followed a migration corridor southward from Costa Rica into the South Pacific Gyre in each year of our study . In the equatorial region , leatherbacks experienced strong ocean currents that influenced the direction of their movements; leatherbacks responded to current deflection with rapid , directed movements to maintain their southward heading . After passing through this equatorial current field , turtles dispersed broadly within a low-energy , low-productivity region of the South Pacific . Our analyses revealed that ocean currents shaped the migration corridor and influenced the scope of turtle dispersal in the South Pacific—results that provide a biological rationale for the development of multi-scale conservation strategies . These strategies could involve improved and enhanced monitoring of leatherback–fisheries interactions as well as dynamic time-area fisheries closures and protected area designations within the high seas of the South Pacific .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"ecology",
"science",
"policy",
"evolutionary",
"biology"
] |
2008
|
Persistent Leatherback Turtle Migrations Present Opportunities for Conservation
|
Viral RNA-host protein interactions are critical for replication of flaviviruses , a genus of positive-strand RNA viruses comprising major vector-borne human pathogens including dengue viruses ( DENV ) . We examined three conserved host RNA-binding proteins ( RBPs ) G3BP1 , G3BP2 and CAPRIN1 in dengue virus ( DENV-2 ) infection and found them to be novel regulators of the interferon ( IFN ) response against DENV-2 . The three RBPs were required for the accumulation of the protein products of several interferon stimulated genes ( ISGs ) , and for efficient translation of PKR and IFITM2 mRNAs . This identifies G3BP1 , G3BP2 and CAPRIN1 as novel regulators of the antiviral state . Their antiviral activity was antagonized by the abundant DENV-2 non-coding subgenomic flaviviral RNA ( sfRNA ) , which bound to G3BP1 , G3BP2 and CAPRIN1 , inhibited their activity and lead to profound inhibition of ISG mRNA translation . This work describes a new and unexpected level of regulation for interferon stimulated gene expression and presents the first mechanism of action for an sfRNA as a molecular sponge of anti-viral effectors in human cells .
The critical roles of type I interferon ( IFN ) in detecting and clearing a wide range of viral infections have been well established [1] . IFNs are produced and released into the extracellular space by virtually all cell types upon recognition of pathogen-associated molecular patterns . Secreted IFNs act on the producing and neighboring cells to induce transcriptional activation of hundreds of antiviral IFN-stimulated genes ( ISGs ) , establishing an antiviral state that rapidly targets viruses at various steps of their life cycle . While the transcriptional regulation of ISGs has been long defined , post-transcriptional events have recently emerged as critical regulators of the amplitude and specificity of the response . Regulation of mRNA stability [2] , [3] , translation [4] , [5] or ubiquitination [6] , [7] were shown to be critical for IFN-mediated antiviral effects . Nevertheless , the relative contribution of these post-transcriptional regulators and how they fine-tune the IFN system remain poorly understood . Like other cytoplasmic RNA viruses , flaviviruses are highly sensitive to the antiviral effects of IFNs and as a result have evolved a wide array of countermeasures to avoid their action [8] . Described mechanisms include concealing double-stranded RNA replication intermediates in virally-induced ER membranes to decrease activation of innate immune sensors [9] , cap methylation to mimic cellular mRNAs [10] , degradation of regulators of IFN activation by the viral protease NS2B/3 [11] , [12] , or destabilization of transcription factor STAT2 by viral NS5 protein to dampen transcriptional activation of ISGs [13] More recently a ∼0 . 5 kb , abundant non-coding RNA derived from incomplete degradation of the viral 3′ untranslated region ( 3′UTR ) by the cellular 5′-3′ exonuclease XRN1 and produced by all flaviviruses ( termed sfRNA for subgenomic flaviviral RNA ) was reported to be required for viral pathogenicity in a mouse model of the attenuated Kunjin strain ( KUNV ) of West Nile virus ( WNV ) [14] . Follow-up studies determined that KUNV sfRNA counteracted IFN antiviral activity [14] , [15] . Strikingly , while the majority of genomes synthesized during infection are processed into sfRNA , it is dispensable for RNA replication in IFN-incompetent cells , arguing for an important , conserved role in antagonizing immune defenses . The only possible mechanism for the anti-IFN action of the sfRNA was suggested by a recent report that suggests the Japanese encephalitis virus ( JEV ) sfRNA inhibits IFN production by blocking the phosphorylation of IRF-3 [16] . The observation suggesting a decrease in IFN production by transfecting JEV sfRNA was not properly controlled by the use of other RNAs and thus we believe that to date the anti-IFN mechanism of sfRNA remains unknown . As the flaviviral positive-strand 11 kb RNA genome encodes only 10 viral proteins , it is not surprising that host proteins , especially RNA binding proteins ( RBPs ) , play a critical role in viral replication and pathogenicity . In a screen for host proteins interacting with dengue virus 2 ( DENV-2 ) RNA , we identified ubiquitous , multifunctional RBPs , G3BP1 , G3BP2 and CAPRIN1 [17] . G3BP1 and CAPRIN1 had been reported as proviral factors in vaccinia virus ( VACV ) and respiratory syncytial virus ( RSV ) infection [18] , [19] . On the other hand , G3BP1 and G3BP2 had antiviral activity against poliovirus ( PV ) and alphaviruses [20] , [21] , suggesting a variety of possible mechanisms of action in viral infections . In this study , we investigated their role in DENV-2 infection . We found that these proteins have a potent antiviral action against flaviviruses , linked to a previously unknown role in regulating translation of ISG mRNAs . We further demonstrate that this activity is targeted by DENV-2 sfRNA , which binds G3BP1 , G3BP2 and CAPRIN1 and prevents their function , protecting viral replication against IFN-mediated antiviral effects .
G3BP1 , G3BP2 and CAPRIN1 were initially discovered as interacting with DENV-2 3′UTR in an RNA affinity chromatography screen performed in our laboratory [17] . The variety of cellular functions described for these proteins in control of mRNA translation and stability , regulation of cell signaling pathways , and in the integrated stress response [22]–[26] prompted us to examine their role in DENV-2 infection . RNAi-mediated knockdown and overexpression studies indicated that G3BP1 , G3BP2 and CAPRIN1 had a modest but significant antiviral activity against DENV-2 NGC in HuH-7 cells ( Figure S1A–D ) . Importantly , transfection of control and specific siRNAs did not induce IFN mRNA accumulation , which could have influence DENV-2 replication as an off target effect , [27] ( Figure S1E ) . The antiviral effect was not restricted to the laboratory adapted DENV-2 NGC as G3BP1 , G3BP2 and CAPRIN1 had antiviral activity against the clinical DENV-2 isolate PR1940 as well as the related yellow fever vaccine strain ( YFV-17D ) ( Figure S1F ) . Nonetheless , G3BP1 , G3BP2 and CAPRIN1 depletion had no effect on the DENV-2 strain PR6913 , which exhibited low replication levels ( Figure S1F ) . Since the type I IFN response has been long established as a critical mediator of anti-flaviviral innate immunity [8] and low levels of flaviviral replication have been shown to correlate with lower induction of type I IFN response [28] the data above suggested that G3BP1 , G3BP2 and CAPRIN1 could play a previously unidentified role in the innate immune response . In order to investigate a link between IFN-mediated antiviral activity and that of G3BP1 , G3BP2 and CAPRIN1 , cells depleted of these three proteins ( Figure 1A ) were pretreated with IFN-β before infection with DENV-2 NGC . As in previous studies [29] , 100 UI/ml IFN-β , which is comparable to that observed in sera of DENV infected patients [30] , abrogated formation of viral replication complexes ( Figure 1B ) . Strikingly , the IFN effect was dramatically diminished in G3BP1 , G3BP2 and CAPRIN1-depleted cells ( Figures 1B ) , indicating that G3BP1 , G3BP2 and CAPRIN1 are required for IFN-β antiviral effects . Next , DENV-2 infectivity was measured over a range of IFN concentrations , revealing that treatment with two independent sets of siRNAs targeting G3BP1 , G3BP2 and CAPRIN1 ( siG12C#1 and siG12C#2 ) resulted in a 4- to 5-fold increase in IFN-β IC50 ( Figure 1C ) . This effect was even more pronounced at the level of infectious progeny virus formation and accumulation of viral RNAs ( Figures 1D and 1E ) , consistent with IFN-β targeting various steps of the viral life cycle . Notably , the magnitude of the G3BP1 , G3BP2 and CAPRIN1-mediated antiviral activity in the presence of IFN-β was much greater than its ∼2-fold effect in the absence of IFN-β , ( Figure S1B ) , suggesting that the main antiviral role of these proteins occurs through IFN action . It should be noted that HuH-7 cells secrete low levels of IFN-β upon DENV-2 infection ( [31] and see below ) , and therefore conditions without exogenously added IFN-β should be consider to have low levels of IFN . Interestingly , G3BP1 , G3BP2 and CAPRIN1 antiviral activity was redundant since depletion of all three proteins was required to observe an effect on DENV replication ( Figure S2 ) , which was consistent with previous studies on other functions of these RBPs [18] , [21] , [25] , [32] . The requirement of G3BP1 , G3BP2 and CAPRIN1 in IFN-mediated antiviral activity was observed in YFV-17D infection ( Figure 1F ) . Taken together , these data indicate that G3BP1 , G3BP2 and CAPRIN1 have an unexpected and important role in mediating the anti-flaviviral activity of IFNs . We next investigated the mechanism of action of this unexpected role of G3BP1 , G3BP2 and CAPRIN1 in the IFN response . Since G3BP1 , G3BP2 and CAPRIN1 had previously been determined to be critical regulators of stress granules ( SG ) assembly , another aspect of the cellular response to infection , we examined SG formation in infected cells and upon IFN treatment . Indeed , a recent study of SG dynamics in HCV infection revealed that treatment of HCV-infected cells with IFN-α triggered potent SG formation [33] , suggesting that SG could mediate IFN antiviral effects . We monitored SG formation during DENV-2 infection in the presence or absence of IFN-β and found no evidence of increased SG formation in DENV-2 infected cells , even following treatment with IFN-β that effectively reduced viral replication ( Figure S3 ) . These data indicate that bona fide SG formation , which is defined microscopically by the appearance of cytoplasmic granules containing a set of protein markers , was not required to mediate the IFN anti-DENV-2 effects . Our data do not exclude an important role for SG and SG-associated proteins in DENV-2 infection , but they suggested that G3BP1 , G3BP2 and CAPRIN1 could play an alternative role in the IFN response to DENV-2 infection . To gain insight into such alternative modes of action , we examined the integrity of the pathway leading to the establishment of the IFN induced antiviral state . Binding of IFNs to their receptor on the cell surface activates the JAK-STAT signaling cascade , leading to the transcriptional activation of hundreds of IFN-stimulated genes ( ISGs ) , which have specific antiviral activities [34] . We selected a representative panel of ISGs and measured their induction in response to IFN-β in control versus G3BP1 , G3BP2 and CAPRIN1 depleted HuH-7 cells . IFN-inducible IFITM2 , RIG-I/DDX58 ISG15 and STAT1 have been reported to have anti-DENV-2 activity; however , the dsRNA-activated kinase PKR/EIF2AK2 and MX1 do not affect DENV-2 replication [35]–[38] . Quantitative real-time RT-PCR analysis showed no significant decrease of control mRNA levels ( GAPDH or ACTINB ) or ISG mRNA induction in response to increasing IFN-β concentration in G3BP1 , G3BP2 and CAPRIN1-depleted cells ( Figures 2A to 2D and S4A to S4D ) . Strikingly , expression of all six ISG proteins , however , was significantly reduced in G3BP1 , G3BP2 and CAPRIN1-depleted cells ( Figures 2E , 2F and S4E ) . PKR protein accumulation , as measured by densitometry analysis of the western blot data shown in Figure 2E , was reduced 4 . 05 and 3 . 21 fold compared to siGFP control in siG12C#1 and siG12C#2-treated cells , respectively . In this experiment , RIG-I accumulation was reduced 1 . 86 and 1 . 80 fold , and ISG15 2 . 15 and 1 . 60 fold respectively . STAT1 protein levels were reduced 4 . 50 and 4 . 98 fold , and MX1 2 . 46 and 11 . 2-fold respectively in cells depleted of G3BP1 , G3BP2 and CAPRIN1 ( Figure S4 ) . Because determination of protein levels from HRP-based western data is semi-quantitative , we performed analysis of IFITM2 levels using fluorophore-conjugated secondary antibodies in the Licor Odyssey system . Quantification of western blots by fluorescence intensity in three independent experiments revealed a 3- to 5-fold reduction in IFITM2 levels , normalized to ACTINB levels , after stimulation with 100 or 1000 UI/ml IFN-β ( Figure 2F ) . These data indicate a general and robust effect of G3BP1 , G3BP2 and CAPRIN1 on establishment of the antiviral state through post-transcriptional control of multiple ISGs . We propose that many more ISGs will be affected , and the cumulative effect would likely explain the dramatic drop in IFN antiviral potency observed in the previous experiments . Given our data above it was possible for G3BP1 , G3BP2 and CAPRIN1 to influence ISG mRNA splicing , transport or translation , or ISG protein stability . Because of previous reports on these proteins [23] , [24] , [26] , [39] , [40] , we first addressed their action on ISG mRNA translation . We focused on ISGs IFITM2 and PKR to delineate the role of G3BP1 , G3BP2 and CAPRIN1 . Using 35S metabolic labeling , we established that depletion of G3BP1 , G3BP2 and CAPRIN1 depletion did not downregulate global cellular translation ( Figure 3A ) . Consistent with a lack of a profound global effect on translation , polyribosome fractionation revealed no significant difference in rRNA profiles in G3BP1 , G3BP2 and CAPRIN1-depleted cells ( Figure 3B ) . Polyribosome association profiles of several cellular mRNAs ( ELF2 , GAPDH and BIP/GRP78 mRNAs ) showed minimal or no change upon G3BP1 , G3BP2 and CAPRIN1 depletion and IFN-β treatment ( Figures 3C–E ) . Indeed , although GAPDH and BIP mRNA slightly shifted to lighter fractions , the fraction containing the majority of the mRNA remained unchanged , indicating that association of these mRNAs with polyribosomes was not dramatically affected . Polyribosome-association of IFITM2 and PKR mRNAs , however , was strongly impaired in the absence of G3BP1 , G3BP2 and CAPRIN1 ( Figures 3F–G ) , suggesting that the strong effect of G3BP1 , G3BP2 and CAPRIN1 depletion was specific for ISG mRNA translation . Importantly , depletion of G3BP1 , G3BP2 and CAPRIN1 did not affect polyribosome-association of DENV-2 genomic RNA ( Figure 3H ) , supporting the previous hypothesis that G3BP1 , G3BP2 and CAPRIN1 antiviral action is principally mediated by the IFN system rather than by a direct role in viral translation . In order to confirm the specificity of G3BP1 , G3BP2 and CAPRIN1 in regulating ISG mRNA translation , we established stable cell lines expressing firefly luciferase reporters under the transcriptional control of a minimal promoter and an ISRE to provide IFN induction , and including the ELF2 , GAPDH , IFITM2 or PKR UTRs with the first and last 30 nucleotides of the coding sequence ( ELF2-Fluc , GAPDH-Fluc , IFITM2-Fluc and PKR-Fluc Figure 4A ) . We observed that IFN-induction of GAPDH-Fluc , IFITM2-Fluc and PKR-Fluc mRNAs was modestly reduced , although the effect was not significant for IFITM2-Fluc and PKR-Fluc . The induction of ELF2-Fluc mRNA was robustly and significantly reduced in the absence of G3BP1 , G3BP2 and CAPRIN1 ( Figures 4B–E ) . However , the significance of these observations is not clear since depletion of G3BP1 , G3BP2 and CAPRIN1 did not affect absolute levels of endogenous GAPDH , IFITM2 or PKR mRNAs ( see Figures 2 and 3 ) . Importantly , IFN induction of ELF2-FLuc luciferase activity was not inhibited by knockdown of G3BP1 , G3BP2 and CAPRIN1 , however , IFN induction of IFITM2-Fluc and PKR-Fluc activity was robustly inhibited ( Figure 4F , H and I ) . While induction of GAPDH-Fluc activity was significantly reduced in these conditions , the effect could be fully explained by the aforementioned effect on GAPDH-Fluc mRNA levels ( compare Figures 4C and G ) . Indeed , a calculation of the relative translation efficiency , the ratio of protein induction ( derived from luciferase activity ) relative to mRNA induction , clearly revealed that both ELF2 reporter translation was increased 6 . 3-fold in G3BP1 , G3BP2 and CAPRIN1-depleted cells compared to control siGFP , while the relative translation efficiency of IFITM2-Fluc and PKR-Fluc was reduced 2 . 54- and 1 . 3-fold , respectively ( Table 1 ) . In the case of GAPDH-Fluc , the relative translation efficiency was slightly reduced ( 1 . 07-fold ) , which correlates with the modest shift observed in GAPDH mRNA distribution in polyribosomes fractions in the absence of G3BP1 , G3BP2 and CAPRIN1 ( see Figure 3E ) . Taken together , these results show that G3BP1 , G3BP2 and CAPRIN1 differentially affect reporter mRNA translation and that elements in the IFITM2 and PKR mRNA UTRs and/or the first and last 30 nucleotides of their coding sequence render translation of these messengers specifically dependent on G3BP1 , G3BP2 and CAPRIN1 . This suggest that G3BP1 , G3BP2 and CAPRIN1 can , as previously described in the literature , play various roles in cellular mRNA metabolism , but are specifically required for translation of ISG mRNAs . While the precise mechanisms of translational regulation and how these proteins achieve selectivity remain to be investigated , several hypotheses will be proposed in the discussion . Flaviviruses , like other viruses , have been reported to interfere with the host IFN response by hijacking a large variety of cellular factors required for establishment of the antiviral state . In the case of DENV-2 , all previously described evasion strategies affect signaling pathways upstream of ISG transcriptional activation [11] , [13] , [41] , [42] However , these mechanisms are not completely efficient since ISG mRNA upregulation is observed widely in response to DENV-2 infection [43] . Data presented above suggested that ISG mRNA translation could be targeted by DENV-2 and this would not have been detected in previous studies measuring ISG mRNA induction as a surrogate for efficient IFN response . In DENV-2 infected cells , viral RNA replication resulted in a 24-fold increase in IFN-β mRNA between 24 and 48 h post infection , which was accompanied by a 7-fold increase in IFITM2 mRNA ( Figures 5A to 5C ) . No induction of IFITM2 was detected up to 72 h post infection ( Figures 5D and 5E ) , indicating that ISG expression was indeed controlled at a post-transcriptional level during DENV-2 infection . Importantly , polyribosome fractionation analysis showed that in DENV-2 infected cells and in G3BP1 , G3BP2 and CAPRIN1-depleted cells , IFITM2 and PKR mRNA translation was strongly impaired ( Figures 5F to 5I ) . Interestingly , DENV-2 infection inhibited IFN induction of both PKR mRNA and protein ( Figure S5 ) , indicating that this virus can regulate some ISGs via multiple mechanisms to keep the IFN response under check . Most importantly the data indicate that DENV-2 interfered with IFITM2 and PKR mRNA translation , which phenocopied G3BP1 , G3BP2 and CAPRIN1 depletion , and suggested that DENV-2 gene product ( s ) target these RBPs . Previously G3BP1 had been reported to be antagonized in poliovirus infection , where it is cleaved by a viral protease [20] . DENV-2 infection however , did not decrease the levels of G3BP1 , G3BP2 or CAPRIN1 ( Figure S5 ) , suggesting a mechanism other than proteolytic degradation . We have shown previously that G3BP1 , G3BP2 and CAPRIN1 each interact with the 3′UTR of DENV-2 RNA [17] , a region included in the 3′UTR-derived non-coding sfRNA . These interactions , together with the fact that the sfRNA from a related flavivirus , Kunjin virus ( KUNV ) , interferes with the IFN response [15] , made DENV-2 sfRNA an ideal candidate for targeting G3BP1 , G3BP2 and CAPRIN1 . Therefore , we hypothesized that DENV-2 sfRNA would bind G3BP1 , G3BP2 and CAPRIN1 and inactivate their antiviral effect . In order to determine whether DENV-2 sfRNA interacts with G3BP1 , G3BP2 and CAPRIN1 , we first performed co-localization experiments using in situ hybridization for DENV-2 RNAs and immunofluorescence for G3BP1 in infected cells . In situ probes detecting the viral 5′UTR , which interrogate only the gRNA , and 3′UTR , which detect both gRNA and sfRNA , were both found to colocalize with G3BP1 during infection ( Figure S6 ) . To test and quantify an interaction between viral RNAs and the three RBPs , we used RNA-immunoprecipitation and a real-time PCR strategy designed to discriminate between gRNA and sfRNA , which is identical to the last 428 nucleotides of the genome [44] ( Figures 6A and S7A–E ) . As suggested previously [44] , we found that DENV-2 sfRNA was 5–10 times more abundant than the gRNA during infection of HuH-7 cells ( Figure S7G ) . Both DENV-2 gRNA and sfRNA were found enriched in G3BP1-immunoprecipitates from infected cells ( 3- and 6-fold relative to GAPDH RNA , respectively ) , but were not found to interact with KSRP , an unrelated host RBP ( Figures 6B and 5C ) . DENV-2 gRNA and sfRNA were also enriched in G3BP2 and CAPRIN1 immunoprecipitates , while c-Myc mRNA was not enriched in these ( Figure S8 ) , confirming the specificity of the interaction . Taken together , these data indicate that G3BP1 , G3BP2 and CAPRIN1 interact with DENV-2 gRNA and sfRNA during infection . Having established that G3BP1 , G3BP2 and CAPRIN1 interacted with DENV-2 sfRNA in infected cells , we sought to examine which sequence or structural elements were required for this interaction . The DENV-2 3′UTR contains a series of highly conserved secondary structures ( Figure 6A ) , which have been proposed to serve as platforms of interaction for host RBPs [17] [45] . We designed sfRNA variants containing various deletions and point mutations ( Figure 6D ) and tested their ability to interact with G3BP1 , G3BP2 and CAPRIN1 by RNA affinity chromatography . We found that stemloop II ( SL-II ) , but not the predicted pseudoknot PKSL-II , was required for G3BP1 , G3BP2 and CAPRIN1 binding to DENV-2 3′UTR ( Figure 6E ) . We also tested the ability of the 3′UTR of related flaviviruses to interact with these RBPs and observed that only 3′UTRs of clinical isolates from DENV-2 , but not DENV-3 , the attenuated WNV subtype KUNV or the YFV vaccine strain 17D were able to pull-down G3BP1 , G3BP2 and CAPRIN1 ( Figure S9 ) . Notably , in these mutants and isolates , the three proteins shared the same binding requirements , suggesting that these proteins interact as a complex . In order to determine whether DENV-2 sfRNA was able to inhibit G3BP1 , G3BP2 and CAPRIN1 activity and impair ISG expression , we transfected increasing amounts of DENV-2 3′UTR RNAs , as sfRNA-mimics , into cells and measured ISG mRNA and protein expression upon IFN-β treatment . To control for effects mediated by functions of the sfRNA unrelated to G3BP1 , G3BP2 and CAPRIN1 binding , we constructed a mutant unable to bind these RBPs but containing all other sequence elements of DENV-2 sfRNA . Since the structure required for binding , SL-II , has been implicated in formation and stability of flaviviral sfRNAs [46] , we sought to minimize the effects of its deletion by replacing it with the equivalent structure from YFV-17D , SLE ( Figures 7A and 7B ) , which did not interact with G3BP1 , G3BP2 and CAPRIN1 in vitro ( Figure S9 and Ward et al , unpublished data ) . This hybrid mutant , DENV-2 3′UTR YFSLE , exhibited 5-fold decreased binding to G3BP1 ( Figure 7C ) , and additional point mutations in DENV-2 SL-IV , whose secondary structure resembles SL-II ( indicated on Figure 7B ) , further decreased the interaction to background levels ( DENV-2 3′UTR YFSLE-ST4 , Figure 7C ) . When increasing amounts of in vitro transcribed RNAs were transfected into cells followed by treatment with 100 UI/ml IFN-β , we observed that DENV-2 3′UTR , but not the control DENV-2 3′UTR YFSLE RNA , was able to decrease in a dose-dependent manner expression of ISGs IFITM2 ( Figures 7D to 7G ) and PKR ( Figure S10 ) . Both RNAs accumulated to similar levels and had no effect on ISG mRNA induction levels ( Figures 7D , 7E and S10 ) , indicating that DENV-2 sfRNA is able to post-transcriptionally interfere with ISG expression and that this activity depended on G3BP1 , G3BP2 and CAPRIN1 binding . As observed for G3BP1 , G3BP2 and CAPRIN1 depletion , ectopic expression of DENV-2 3′UTR interfered with ISG mRNA association with polyribosomes , while GAPDH and ELF2 mRNA were minimally or not affected in the same conditions ( Figure S11 ) . Taken together , these data show that ectopic expression of DENV-2 sfRNA mimics inhibited IFITM2 and PKR mRNA translation through G3BP1 , G3BP2 and CAPRIN1 binding . We showed that interaction of DENV-2 sfRNA with G3BP1 , G3BP2 and CAPRIN1 was able to downregulate expression of ISGs , which is consistent with the sfRNA acting as a decoy for these host RBPs . To further test this hypothesis and determine the importance of this mechanism during infection and for viral evasion of the IFN response , we constructed mutant DENV-2 replicons unable to sequester G3BP1 , G3BP2 and CAPRIN1 . We used the established DENV-2 replicon system [47] , whose biphasic reporter activity examines translation of input RNAs and subsequent replication and translation steps independently , to evaluate the effect of G3BP1 , G3BP2 and CAPRIN1 binding on translation , replication and sensitivity to inhibition by IFN-β . We modified the DENV-2 replicon 3′UTR deleting the SL-II and introducing point mutations in SL-IV described before ( D2Rep-dSLII-ST4 , Figure 8A , S12A ) and confirmed that replicon RNAs bearing these mutations had reduced binding to G3BP1 ( Figure 8B ) . While SLII was reported to be required for sfRNA formation in some flaviviruses , we did not measure a decrease in sfRNA formation in dSLII-ST4 mutants ( Figure S12B ) , ruling out the possibility that the effect of the mutation could be linked to SL-II functions mediated by other regions of the sfRNA . Indeed this finding is consistent with in vivo results in the recent report by Liu et al [48] . We electroporated these reporters into HuH-7 . 5 cells , which were derived from HuH-7 cells and harbor a point mutation in RIG-I that renders them deficient in IFN production through this pathway [49] and the parental HuH-7 cells . Importantly , we detected no difference in luciferase activity between D2Rep-WT and D2Rep-dSLII/ST4 at any time after electroporation ( Figure 8C ) , indicating that reduced binding to G3BP1 , G3BP2 and CAPRIN1 had no effect on replicon translation or replication . In HuH-7 cells the dSLII-ST4 mutation did not alter very early luciferase activity , a measure of translation of input RNAs ( Figure S12C ) , however luciferase activity of the D2Rep-dSLII/ST4 was reduced by an average of 4 . 4-fold compared to the WT replicon at 72 h post-electroporation ( Figure 8D ) . The different effects in HuH-7 . 5 and HuH-7 cells suggested that G3BP1 , G3BP2 and CAPRIN1 binding to viral RNAs was required for viral replication in the context of a functional innate immune response . To examine the effect of adding exogenous IFN on D2Rep-WT and D2Rep-dSLII-ST4 activity we electroporated these in HuH-7 cells and treated these with 50 UI/ml IFN-β at 4 h post-electroporation . The modest deleterious effect of the dSLII/ST4 mutation was strikingly enhanced by IFN treatment , with luciferase activity reduced 16 . 9-fold compared to the D2Rep-WT at 72 hr post-electroporation ( Figure 8E ) . Overall , addition of exogenous IFN-β inhibited D2Rep-WT activity by 3 . 2-fold at 72 h post-electroporation while D2Rep-dSLII-ST4 was inhibited 12 . 3-fold ( Figure 8F ) , indicating that the dSLII/ST4 mutation renders replicons more sensitive to the antiviral effects of IFNs . Finally , we analyzed the rates of accumulation of luciferase reporter between 48 and 72 h post electroporation . We observed no significant difference between the rates of D2Rep-WT and D2Rep-dSLII/ST4 in the absence of exogenously added IFN-β; however the D2Rep-dSLII/ST4 was significantly impaired in the presence of exogenously added IFN ( Figure 8G ) . On the one hand , in the presence of low levels of endogenous IFN , which we expect with HuH-7 but not HuH-7 . 5 cells , after an initial delay the mutant replicon is still able to surmount IFN-mediated inhibition . On the other hand in the presence of higher levels of IFN the D2Rep-dSLII-ST4 is persistently inhibited . The results above convincingly argue that anti-DENV-2 action of G3BP1 , G3BP2 and CAPRIN1 is mediated by their pro-IFN activity and support the hypothesis that the DENV-2 sfRNA antagonizes the IFN response in part by sequestering these host RBPs .
In this study we make two new and important observations in the understanding of host innate antiviral measures and their inhibition by viral countermeasures . First , we identified G3BP1 , G3BP2 and CAPRIN1 , three conserved , multifunctional RNA-binding proteins , as critical positive regulators of the antiviral IFN response . This unexpected role was mediated through the specific activation of antiviral ISG mRNA translation . Second , we described the DENV-2 sfRNA as an antagonist to their antiviral effect , providing the first mechanism of action for this abundant , non-coding flaviviral RNA ( Figure 9 ) . Although G3BP1 , G3BP2 and CAPRIN1 have been shown to have a large variety of cellular functions , this report associates them for the first time with innate immunity . We show that these three RBPs were required for an antiviral IFN response against several isolates of DENV-2 and YFV-17D . Our data indicate that G3BP1 , G3BP2 and CAPRIN1 regulate the expression of ISGs known to have broad antiviral activity: PKR , RIG-I , IFITM2 , ISG15 , STAT1 and MX1 [34] . Therefore , while the full spectrum of ISG targets and the individual contributions of the three RBPs remain to be determined , we posit their antiviral activity will be conserved against a wide array of viruses . Indeed previous evidence suggested this: the poliovirus ( PV ) protease degrades G3BP1 [20]; the core protein of Japanese encephalitis virus ( JEV ) , another flavivirus , was identified as an important CAPRIN1 antagonist [50] , and the nsP3 protein of Chikungunya virus ( CHIKV ) , an alphavirus , as G3BP1 and G3BP2 opponent [51] . All these interactions were shown to be required for optimal viral replication , supporting our conclusion that G3BP1 , G3BP2 and CAPRIN1 are major regulators of the cellular immune response . The fact that these RBPs were not previously identified as IFN-related antiviral factors can be explained by two reasons . First , previous studies usually focused on SG formation and were performed in absence of exogenously added IFN . Second , the role of these proteins was examined independently , in ways that would not unearth their redundant functions in the IFN system . Interestingly , the direct antiviral role of SG formation is intuitive but has not been formally demonstrated given the challenges in differentiating the role of the granules themselves from the role of their numerous individual components . While the relative contributions and connections of these two branches of the innate immune response remain to be determined , our study suggests that activity of G3BP1 , G3BP2 and CAPRIN1 against DENV-2 is primarily through the IFN system . Perhaps our most unexpected finding was that G3BP1 , G3BP2 and CAPRIN1 are critical for ISG mRNAs translation , a step previously understudied in the IFN response . Although the dogma is that establishment of the IFN-mediated antiviral state is primarily controlled by transcriptional activation , recent evidence suggests that additional layers of control regulate the amplitude and specificity of the response . For instance , a screen for host proteins implicated in IFN-mediated inhibition of hepatitis-C virus identified a large number of splicing factors [52] . This result implicates post-transcriptional mechanisms , which could regulate splicing or the proteins could moonlight in other aspects of RNA metabolism . Control of the stability of IFN-β mRNA by KSRP and STAT mRNA by PCBP2 were equally able to modulate IFN-mediated inhibition of viral replication [2] , [3] . A recent study implicates , but does not directly address , the importance of ISG translational regulation [53] in antiviral signaling and underscores the importance of our findings . The precise mode of action of G3BP1 , G3BP2 and CAPRIN1 in ISG translational regulation and especially how specificity for ISG mRNAs is achieved remain to be elucidated . Several hypotheses could be considered . G3BP1 , G3BP2 and CAPRIN1 could bind to ISG mRNA UTRs and recruit translation initiation factors , recruit ISG mRNAs to subcellular localizations where translation is more efficient in conditions of stress , or relieve miRNA-mediated inhibition of ISG mRNA translation . The RBPs could also act indirectly either activating or repressing mRNAs coding for positive or negative regulators of ISG mRNA translation . Alternatively , G3BP1 , G3BP2 and CAPRIN1 could modulate signaling events leading to translational activation in the IFN response , such as the PI3K/Akt or Mnk pathways that are required for ISG mRNA translation [5] , [54] . Finally , the RBPs could be involved in a stress response induced by IFNs that while not inducing bona fide SG would generally repress many mRNAs and by mass action enhance the translation of ISG mRNAs . In all above scenarios though , the cis-acting elements in the ISG mRNA UTRs conferring dependency on G3BP1 , G3BP2 and CAPRIN1 will be a critical feature to determine . In the second part of our study , we show that the DENV-2 abundant , non-coding sfRNA interacts with G3BP1 , G3BP2 and CAPRIN1 , inactivates them and thus mediates inhibition of ISG expression . The sfRNA - G3BP1 , G3BP2 and CAPRIN1 interaction that we propose as a decoy mechanism was conserved for DENV-2 clinical isolates indicating its potential relevance for DENV-2 pathogenicity . While the antagonism of the immune response by viral non-coding RNAs has been well described , few mechanisms of action have been uncovered . Sequestration of host proteins has been widely hypothesized but only in a few instances was it formally demonstrated . The adenovirus VA RNAs was shown to bind and antagonize PKR and a similar role was proposed for Epstein Barr virus EBER RNAs [55]–[57]; the Sendai virus trailer RNA was hypothesized to sequester the RBP TIAR to subvert apoptosis [58]; the interaction between Kaposi's sarcoma-associated herpesvirus ( KSHV ) PAN RNA and PABP was suggested to participate in the host translational shutoff effect [59] , [60] . Here we demonstrate a role for DENV-2 sfRNA as a molecular sponge or decoy for G3BP1 , G3BP2 and CAPRIN1 resulting in a crippled IFN response . While the sfRNA-G3BP1 , G3BP2 and CAPRIN1 interaction was conserved for all DENV-2 viruses tested , no binding was detected for DENV-3 , KUNV or YFV-17D 3′UTR . This suggests that although the IFN antagonist action of the sfRNA is conserved among flaviviruses , the precise mechanisms diverge for different viruses [15] . This is not unexpected since specific tactics for viral evasion of the IFN response have been shown to vary widely between related viruses , even strains of the same virus [8] , [34] . For instance , NS4B proteins from some DENV-2 clinical isolates , but not from others , were able to interfere with IFN signaling [61] . Furthermore the sfRNA includes the so-called variable region ( VR ) , which , while generally conserved in RNA secondary and tertiary structure , diverges significantly in primary sequence among flaviviruses . The VR is therefore a propitious platform for rapid evolution of new host RBP binding sites providing this viral genus with a wide array of tactical solutions to counter host innate defenses . It is thus conceivable that KUNV sfRNA , although not binding to G3BP1 , G3BP2 and CAPRIN1 , could target different subsets of host RBPs to prevent establishment of the antiviral state . To conclude , it is widely accepted that host immune measures and pathogen countermeasures evolve rapidly , leading to remarkable diversity on both sides . Here , we propose that RBPs such as G3BP1 , G3BP2 and CAPRIN1 are critical mediators of the antiviral state and that antagonizing them is a strategy employed by many viruses , including DENV-2 . Equally , we believe that targeting different subsets of host RBPs is a pan-flaviviral anti-IFN strategy , for which many targets remain to be uncovered .
HuH-7 hepatocellular carcinoma cells were maintained in DMEM supplemented with 10% FBS . BHK-21 cells , which were used for virus titration , were maintained in RPMI supplemented with 10% FBS . DENV-2 strain NGC and YFV-17D were propagated in Aedes albopictus C6-36 cells . All infections were carried at a multiplicity of infection of 1 ( MOI = 1 ) for 24 h unless otherwise indicated . Infectivity was measured using indirect immunofluorescence detection of viral antigens or dsRNA in infected cells , quantitative real-time RT-PCR analysis of viral genomes , or quantification of infectious particles released by focus forming assay , as previously reported [17] . 25 nM of the indicated siRNA duplexes or 75 nM control siRNA ( see supplementary materials ) were transfected into cells at 50% confluency twice at 48 hr intervals ( day 1 and 3 ) with Lipofectamine RNAiMax ( Invitrogen ) following manufacturer's instructions . Human IFN-β ( PBL Interferon Source ) was added to cells 24 hrs post-transfection and incubated for 16 h prior harvesting or infection with DENV-2 . Lysates were collected 24 hrs post infection ( day 6 ) and analyzed by western blotting for knockdown efficiency and ISG expression , and quantitative real-time RT-PCR for RNA levels ( see supplementary methods ) . Polyribosome fractionation was performed as previously described [62] with minor modifications: cells were harvested by trypsinization and 50 µg/ml cycloheximide was added into polyribosome lysis buffer . Individual mRNA levels in each fraction were measured by quantitative real-time RT-PCR and expressed as percentage of total for this mRNA in all the gradient fractions . pcDNA3 . 1 constructs containing the firefly luciferase open reading frame flanked by IFITM2 , PKR , GAPDH , or ELF2 5′ and 3′UTRs and driven by an ISRE promoter ( for cloning details refer to supplementary materials ) were transfected in HuH-7 cells and selected for stable expression in DMEM supplemented with 1500 µg/ml G418 ( Gibco ) . siRNA-mediated knockdown was performed as described and cells stimulated with 1000 UI/ml IFN-β for 10 h . Firefly luciferase activity was assessed using the Dual luciferase reporter assay system ( Promega ) . Firefly luciferase mRNA levels were measured by quantitative real-time RT-PCR . RNA-immunoprecipitations were performed using the MAGNA-RIP kit ( Millipore ) following manufacturer's recommendations . The levels of RNA in IP were determined by quantitative real-time RT-PCR and normalized to GAPDH mRNA levels and control rabbit IgG IP following the formula:Tobramycin RNA affinity chromatography was carried out as described previously [17] . DENV-2 gRNA and sfRNA levels were quantified using a differential quantitative real-time RT-PCR assay designed based on the sfRNA mapping in Liu et al [44] ( see Figure 5 and S7 ) . One primer , annealing upstream of the stop codon in which one pair of primer recognizes specifically gRNA while a second pair of primers amplifies sequences shared between gRNA and sfRNA . Briefly , RNA extracted from experimental samples was reverse transcribed and parallel reactions set-up . Primer QG-FOR ( 5′ CCATGAAAAGATTCAGAAG 3′ , annealing upstream of the stop codon ) was used to detect gRNA only while primer QGSF-FOR ( 5′ GTG AGC CCC GTC CAA GG 3′ , annealing downstream of the start of sfRNA ) detected both gRNA and sfRNA . The reverse primer QGSF-REV ( 5′ GCTGCGATTTGTAAGGG 3′ annealing downstream of DB2 ) was shared , leading to products of 309 and 184 bp , respectively . In order to determine the relative sfRNA/gRNA ratio in a given sample , 1–2 µg of total cellular RNA ( 1–10 ng of in-vitro transcribed RNA ) were incubated at 70°C for 5 min and reverse transcribed using the ImPromII kit ( Promega ) following manufacturer's recommendation . Triplicate wells containing 100–200 ng of cDNA , 300 pmol of each primer ( QG-For or QGSF-For and QGSF-Rev ) and Biorad SYBR Green reagent following manufacturer's recommendation were set up in a total of 25 µl . Reactions were run on a Biorad CFX96 quantitative real-time RT-PCR with the following parameters: 90°C 5 min , 40 repeats of 90°C for 30 s , 55°C for 30 s and 72°C for 30 s . Fluorescence detection was performed during the 72°C elongation step at each cycle . For each reaction the molar amount of template ( n ( G ) and n ( GSF ) ) was calculated from the CT value using a standard curve generated from serial dilutions of reverse transcribed purified full-length D2Rep RNA . sfRNA levels were inferred by subtracting the molar amount n ( GSF ) – n ( G ) . A similar strategy was designed for analysis of YFV-17D gRNA and sfRNA levels , in this case primers were based on sfRNA mapping in Silva et al [63] . ( YFV-G-For 5′ GGATGGAGAACCGGACTCC 3′ , YFV-GSF-For 5′ GCTAAGCTGTGAGGCAGTGC 3′ , YFV-GSF-Rev 5′ CGTCTTTCTACCACCACGTG 3′ ) . Templates for synthesis of control DENV-2 3′UTR YFSLE , in which the DENV-2 SL-II sequence ( DENV-2 nt 10306–10348 ) was replaced by YFV 17D SLE sequence ( YFV17D nt 10530–10611 ) , were custom synthesized by GenScript . DENV-2 3′UTR and 3′UTR YFSLE templates were PCR amplified from stock plasmids to add a T7 promoter immediately upstream of the DENV-2 stop codon ( T7-VR-For ) , and in vitro transcribed using the MegaScript kit ( Ambion ) . 10 , 100 or 1000 ng/ml RNA were transfected in cells at 50% confluency using Lipofectamine RNAiMax for 4 h . Cells were washed and incubated with complete medium containing 100 UI/ml IFN-β for 4 h before analysis of protein and mRNA contents . The DENV-2 reporter replicon system ( D2Rep ) , based on DENV-2 strain 16681 ( U87411 . 1 ) , has been described before [47] . Detailed experimental procedures are available in supplementary materials . All results are presented as mean ± SEM of at least 3 independent experiments , unless otherwise indicated . Data were analyzed using unpaired , two-tailed Student's t-test and considered significant if p<0 . 05 ( *p<0 . 05; **p<0 . 01; ***p<0 . 005 ) . The following reference sequences were used to design oligonucleotides throughout the study: DENV-2 NGC ( AF038463 . 1 ) ; DENV-2 PR1940 ( GQ398308 . 1 ) ; DENV-2 PR5344 ( GQ398283 . 1 ) ; DENV-2 EDEN 05K3295 ( EU081177 . 1 ) ; DENV-3 EDEN 05K802 ( EU81184 . 1 ) ; DENV-3 EDEN 05K4454 ( EU081222 . 1 ) ; YFV-17D ( X03700 . 1 ) ; G3BP1 ( NM_005754 . 2 ) ; G3BP2 ( NM_203505 . 2 ) ; CAPRIN1 ( NM_005898 . 4 ) ; IFITM2 ( NM_006435 . 2 ) ; ISG15 ( NM_005101 . 3 ) ; MX1 ( NM_01144925 . 2 ) ; DDX58/RIG-I ( NM_014314 . 3 ) ; EIF2AK2/PKR ( NM_002759 . 3 ) ; STAT1 ( NM_007315 . 3 ) ; GAPDH ( NM_002046 . 3 ) ; ELF2 ( NM_201999 . 2 ) ; GRP78/BIP ( NM_005347 . 4 ) .
|
Dengue virus is the most prevalent arbovirus in the world and an increasingly significant public health problem . Development of vaccines and therapeutics has been slowed by poor understanding of viral pathogenesis . Especially , how the virus subverts the host interferon response , a powerful branch of the innate immune system remains the subject of debate and great interest . Dengue virus produces large quantities of a non-coding , highly structured viral RNA , termed sfRNA , whose function in viral replication is elusive but has been linked in related viruses to inhibition of the interferon response . Nonetheless the mechanisms involved are yet to be characterized . Here , we show that dengue virus 2 sfRNA targets and antagonizes a set of host RNA-binding proteins G3BP1 , G3BP2 and CAPRIN1 , to interfere with translation of antiviral interferon-stimulated mRNAs . This activity impairs establishment of the antiviral state , allowing the virus to replicate and evade the interferon response . While this particular mechanism was not conserved among other flaviviruses , we believe it is highly relevant for dengue virus 2 replication and pathogenesis . Taken together , our results highlight both new layers of complexity in the regulation of the innate immune response , as well as the diversity of strategies flaviviruses employ to counteract it .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"biology",
"and",
"life",
"sciences",
"immunology",
"microbiology",
"medicine",
"and",
"health",
"sciences"
] |
2014
|
G3BP1, G3BP2 and CAPRIN1 Are Required for Translation of Interferon Stimulated mRNAs and Are Targeted by a Dengue Virus Non-coding RNA
|
As scientific advances in perturbing biological systems and technological advances in data acquisition allow the large-scale quantitative analysis of biological function , the robustness of organisms to both transient environmental stresses and inter-generational genetic changes is a fundamental impediment to the identifiability of mathematical models of these functions . An approach to overcoming this impediment is to reduce the space of possible models to take into account both types of robustness . However , the relationship between the two is still controversial . This work uncovers a network characteristic , transient responsiveness , for a specific function that correlates environmental imperturbability and genetic robustness . We test this characteristic extensively for dynamic networks of ordinary differential equations ranging up to 30 interacting nodes and find that there is a power-law relating environmental imperturbability and genetic robustness that tends to linearity as the number of nodes increases . Using our methods , we refine the classification of known 3-node motifs in terms of their environmental and genetic robustness . We demonstrate our approach by applying it to the chemotaxis signaling network . In particular , we investigate plausible models for the role of CheV protein in biochemical adaptation via a phosphorylation pathway , testing modifications that could improve the robustness of the system to environmental and/or genetic perturbation .
Biological systems in general show various types and degrees of robustness to environmental changes , meaning that they continue to function even when changes in the environment occur . This imperturbability is often accompanied by robustness to genetic perturbations , meaning that progeny function even though their genotype is not identical to the parent genotype [1]–[4] . Both features play an important role in evolutionary biology . While the former is a direct outcome of selection , the relationship between evolution and genetic robustness is likely to be indirect for low functional mutation rates [5]–[7] since selection acts only on the phenotype of an organism and not its genotype [8] . It has been argued that the ability of an organism to withstand genetic mutations improves its ability to evolve [8]–[11] . However , the rationale for selection for genetic robustness is still controversial [5]–[8] , [12]–[14] . A correlation between the evolution of environmental and genetic robustness has been proposed [1] , [8] , [15] , [16] based on examples observed in many biological systems such as in yeast [1] , bacterial sncRNAs [2] , segment polarity in the fruit-fly [3] , bacterial chemotaxis [4] , [17]–[24] , heat-shock proteins [25] , [26] , and miRNA stem-loop structures in various species [27] and based on numerical models of evolution under varying fitness conditions [15] , [16] . Similarly , it has been shown that metabolic networks evolving under fluctuating environments acquire robustness to the loss of certain genes as well , while those evolving under stable environments do not [28] . However , there is no general mathematical proof for this correlation [8] . In this study , we develop a computational experiment to investigate the plausibility of this hypothesis , that there is a general correlation between environmental and genetic robustness , and provide a quantitative measure of the degree of correlation , if any . In more detail , we shall show that the presence of a specific dynamic network characteristic in networks is associated with a better correlation between genetic and environmental robustness than found in networks where it is absent . Rather than focusing on a particular system in a specific organism , we choose one function of interest: The ability to attain steady state output for constant input . If a network capable of carrying out this function is robust to external environmental perturbations , what is the probability that it is also robust to internal ( e . g . , genetic ) disruption ? To be specific , we define environmental robustness of a biological network as the ability to maintain an output in the face of input perturbations . Genetic robustness is defined as the ability of a biochemical system to maintain the same output in the face of genetic mutations represented as rate constant changes in the equations representing it . This representation of a mutation as a jump from one set of parameters to another is a standard assumption [29] . For mathematical convenience , we restrict our discussion to Michaelis-Menten type networks as they are likely to reach a steady state under constant inputs relative to general networks without sigmoidal saturation . Such networks were also used in the analysis of three node biochemically adaptable networks by Ma et al . [30] . The sensitivity of biochemical kinetic models to parameter perturbations has been intensively investigated [29] , [31]–[35] as a mathematical model of a biological system should be able to reproduce the function of interest or fit experimental data with a minimal need for parameter fine-tuning [35] , [36] . Systems of biochemical adaptation [30] , [37]–[40] have been of interest in particular . Defining a topology to be a graph of interactions independent of parameter values , we test a large number of random N-node topologies for networks capable of reaching a steady state both under constant input concentrations and after a persistent step change in these input concentrations . We define a network as a topology with a specific set of parameters . Each network is given a numerical value for its level of robustness to input and parameter perturbations . The level of robustness of the topology is determined by averaging over this value obtained from its corresponding networks . In particular , we differentiate between networks that show a transient response to a step change in input and those that do not . We find that there is a statistically significant model II regression between the level of robustness to input of a topology and its level of robustness to parameter perturbations that has a steeper slope in networks with a transient response . Our results may be relevant to the discussion about the relationship between the need to survive in a constantly changing environment and the evolution of genetic robustness . There is a large literature on functional motifs that are necessary for a biological system to carry out specific tasks [30] , [41]–[49] . Here , we test all possible 3-node topologies to find the particular motifs that are of use in achieving both robustness to input and parameters . Having established the correlation between environmental and genetic robustness , we ask if there are topologies sharing certain sets of motifs/architectures that show stronger correlations than others . Ma et al [30] computationally explored all possible topologies of 3-node Michaelis-Menten enzymatic networks for motifs that can best accomplish biochemical adaptation . Using our results on this correlation between different sets of architectures we refine the list of motifs of biochemical adaptations previously published [30] . Our approach can be used to select/reject plausible/improbable models of a system of interest . We demonstrate this via a comparative study of bacterial chemotaxis signaling systems . Chemotaxis is a process generally used by bacteria to sense changes in their chemical environment [4] , [17]–[24] . Chemotactic signaling is a well-studied system , but most of the focus has been on the chemotaxis network of the Escherichia coli ( E . coli ) bacterium [4] , [17]–[20] despite the fact that chemotactic signaling pathways differ between species [21]–[24] . For instance , CheV is a chemotaxis protein found in many bacteria but not in E . coli . In many species , it was shown that CheV , or a variant of it , plays a role in biochemical adaptation during chemotaxis via its phosphorylatable receiver domain [24] , [50] , [51] . However , the exact mechanism is still not known [24] . Here , we compare the coarse-grained network of E . coli chemotaxis with several others involving CheV phosphorylation . We draw conclusions based on the resultant values of robustness to both input and parameter perturbations and the correlation between them . In summary , we provide extensive evidence for a mathematical principle stating that , statistically speaking , dynamical systems that are biochemically adaptable are also genetically robust . We apply this knowledge to search for topological categories and subcategories within 3-node networks that show a particularly strong correlation and a linear relationship between their robustness to input and to parameter perturbations , and to shed more light on the chemotactic signaling pathways in bacteria . This method of searching for motifs can be extended to other functions and to bigger networks in order to find motifs that combine more complex functions necessitating larger numbers of nodes .
We define and derive ( see Methods ) two quantitative measures of input and parameter robustness for each topology : , , , and . and are the values of input and parameter robustness of TR networks ( networks that passed the Pearson test ) while and are the values of input and parameter robustness of NP networks ( networks that did not pass the Pearson test ) . A topology is perfectly robust to input perturbations if is very small , and similarly is perfectly robust to parameter perturbations if is very small ( i . e . , has a very large negative value ) . For topologies with more than 3 nodes we sample over at least 50000 different ones of each size ( 5 , 10 , 15 , and 30-node topologies ) while 3-node topologies are exhaustively sampled . The different topologies ( that have more than 3-nodes ) are sampled randomly as described in Selection Criteria in the Methods section . We reject topologies with a low fraction of TR networks ( , where is the ratio of the number of TR networks to the total number of networks ) and exclude them from any further analysis . We chose 2 . 3% to be the cutoff on as it is the minimal value of that removes clusters and outliers ( Fig . S1 ) . With this , we are left with 2445 , 7847 , 18300 , 19264 , and 16589 3-node , 5-node , 10-node , 15-node , and 30-node topologies respectively . The networks sampled from each of these topologies are qualified as TR ( Fig . 3 ) or NP ( Fig . 4 ) and separated accordingly . We find that over the parameter space of a topology , and can span a wide range of values . Within both TR and NP networks , we find a significant ( p≅0 . 0 ) linear correlation between and . A comparison of the slope of the linear regression ( using model II regression , in particular the ordinary least square bisector method described in [52] ) shows a clear and systematic pattern between topologies of different sizes and TR and NP networks of the same size . We find that the slope increases as the size of the network increases: slope = 0 . 62±0 . 09 for 3-node ( Fig . 3A ) , 0 . 64±0 . 05 for 5-node ( Fig . 3B ) , 0 . 76±0 . 04 for 10-node ( Fig . 3C ) 0 . 80±0 . 05 for 15-node ( Fig . 3D ) , and 0 . 96±0 . 13 for 30-node topologies ( Fig . 3E ) . The marginal error is taken as the 95% confidence interval where the variance of the slope is calculated using its estimate for OLS bisector regression derived by Isobe et al [52] . Similarly , for NP networks we obtain: slope = 0 . 50±0 . 04 for 3-node ( Fig . 4A ) , 0 . 44±0 . 02 for 5-node ( Fig . 4B ) , 0 . 54±0 . 02 for 10-node ( Fig . 4C ) , 0 . 59±0 . 02 for 15-node ( Fig . 4D ) , and 0 . 70±0 . 03 for 30-node topologies ( Fig . 4E ) . As above , the marginal error here is taken as the 95% confidence interval . The confidence intervals show that , for all sizes , the values of slopes for TR networks are consistently higher than that for NP networks of the same size and that the difference between the two slopes is significant . The values of the Pearson correlations within TR and NP networks show no clear pattern . This is mainly due to the variability introduced by parameters whose robustness stays invariant and reducible topologies within N>3 N-node networks ( see Text S1 and Discussion ) . Due to these caveats we are cautious about drawing conclusions based on the values of the Pearson correlation . Sampling over all 39 possible topologies , our results show only 4153 topologies have associated TR networks . Within these topologies we find a significant linear correlation before ( Fig . S2 A ) and after ( Fig . 3 A ) introducing the cutoff , as discussed in the previous section . In what follows we show how we can extract motifs ( i . e . , basic topologies that may be more likely to appear in biological systems ) by examining the slope of the linear regression between and . Here , we show that motifs can be extracted from topologies representing the basic backbones shared by a set of topologies showing a stronger relation between environmental and genetic robustness as follows . We first consider two known motifs , the incoherent feedforward motif ( IFF ) and the negative feedback loop motif ( NFL ) and examine their corresponding relations . IFF ( Fig . 5A ) is a topology wherein the output node is affected by the input receiving node via two paths , one direct and the other indirect , such that , collectively , one path is activating and the other is deactivating . This implies four subcategories denoted IFF1–IFF4 . NFL ( Fig . 5A ) is a topology wherein a node is activated/deactivated by another node , and node is deactivated/activated back by node either directly ( NFL1 , NFL2 ) or indirectly ( NFL3–NFL10 ) . We find that the majority of TR topologies have IFF , NFL , or both IFF and NFL motifs . Only a few have neither IFF nor NFL; these are , however , robust to neither input nor parameter perturbations ( Fig . 6A ) and they all have low fractions of successful trials , indicating that TR networks generated from these topologies are sparse . All 4 subcategories of IFF are fairly robust to both input and parameter perturbations ( results not shown ) . Though NFL only topologies are generally less robust than those containing IFF , a small group of them ( green cluster at the bottom left of Fig . 6A ) have low numbers of successful trials but are highly robust within their small TR space . When not coexisting with other robust motifs , only 4 out of the 10 categories of NFL ( NFL1 , NFL2 , NFL4 , and NFL6 ) are robust to both input and parameter perturbations ( Fig . 6B ) . Seeing how NFL1 topologies show separate groups in Fig . 6B , we examine the distribution of all topologies containing NFL1 according to its 8 types ( Fig . 5B ) . We find that NFL1 type1 topologies are strongly correlated ( r = 0 . 92 , p≅0 ) while NFL1 type2 show separate clustering ( Fig . 6C ) . Thus , we further divide NFL1 type2 into two subtypes ( Fig . 5C ) , type2a ( the output node deactivates itself ) and type2b ( all others ) . While both subtypes show strong correlation between their and values ( Fig . 6D , type2a: r = 0 . 97 and p = 10−28 , type2b: r = 0 . 97 and p = 10−51 ) , type2b shows a much steeper slope ( 1 . 12 for type2b , 0 . 33 for type2a , ttest = 3 . 8 and p = 0 . 0002 ) . This steeper slope may be advantageous for specific biological functions , though both types show strong correlation between the two types of robustness . In the presence of IFF , the two types show no correlation ( p = 0 . 14 and 0 . 20 for type2a and type2b respectively ) . In this section , we answer the following questions: ( 1 ) What is the reason for the large variation around the regression lines in Figs . 3 and 4 ? ( 2 ) How does the distribution of and values and their correlation relate to correlations in and values of the networks within the individual topologies ? To answer the first question , we speculated that since clearly each of the parameters in a topology will have different robustness values , we might be able to separate the parameters into different categories such that the regression along each category leads to different slope values . If we show this to be true , then as the number of possible categories increases , one expects larger variation in the value of for a given value . If , in addition , the number of categories is proportional to the number of nodes , then the observed variation would increase for a bigger network , as evident in Fig . 3 . In what follows , we investigate this possibility within 3-node networks . Consistent with the 5- , 10- 15- , 30-node analysis above , we remove topologies with a low fraction of TR networks ( ) and are left with 2534 topologies to work with . . Next , we separate the parameters of each topology into 7 categories ( Fig . 7 ) . Parameters belonging to categories 1 or 2 are those associated with links affecting ( i . e . , directed towards ) the input receiving node , node 1 . Those belonging to categories 3 or 5 are associated with links affecting the buffer node , node 2 . The rest ( in categories 4 , 6 , 7 ) are associated with links affecting the output node , node 3 . Then , for each category j of a network , we evaluate the value which takes into consideration only robustness to perturbations in parameters belonging to category j ( Eq . 26 in Methods ) . The corresponding value for the topology , is ( Eq . 28 in Methods ) . We find that indeed the regression on each of the 7 categories results in a different slope and different correlation strengths . The results of the overall linear regression are shown in Fig . 8A . For the separate categories , we find that the strongest correlation is between and robustness to perturbations in parameters belonging to category 1 , ( Fig . 8B , slope = 0 . 97 , r = 0 . 97 , p = 0 ) , followed by category 2 ( Fig . 8C , slope = 1 . 01 , r = 0 . 78 , p = 0 ) . Conversely , and show no correlation ( Fig . 8H , slope = 1 . 0 , r = 0 . 01 , p = 0 . 81 ) . In fact , the strength of the correlation between and decreases in the following order: j = 1 ( r = 0 . 97 ) , 2 ( r = 0 . 78 ) , 3 ( r = 0 . 44 ) , 5 ( r = 0 . 42 ) , 4 ( r = 0 . 32 ) , 6 ( r = 0 . 12 ) , and 7 ( r = 0 . 01 ) . The second question is related to whether within each topology the parameter subspace corresponding to input robustness is positively correlated with that corresponding to parameter robustness . If they are not correlated , then the two subspaces could be disjoint and the collective/coarse-grained correlation ( i . e . , the correlation between the and ) does not support our hypothesis . We follow the same procedure as above and separate the parameters into the 7 categories depicted in Fig . 7 . The aim is to be able to compare the results with those in Fig . 8 . For each topology , we perform a linear regression on the relationship between and for each category j . The results of the correlation strength and slopes are represented by their corresponding square of the Pearson correlations and slopes , for . The relationship between and is shown in Fig . 9 while that between and is shown in Fig . 10 . As above , the strongest correlations and steepest slopes are found between and of parameters belonging to category 1 , . For all the topologies , remains ≥0 . 9 ( Fig . 9A ) and ≥0 . 8 ( Fig . 10A ) . A weaker fine-grained correlation indicates a less collective robustness as indicated by the increase in ( i . e . , decrease in parameter robustness ) as decreases , for ( Fig . 9A , B , C , E ) . This pattern does not appear for ( Fig . 9D , F , G ) , which is consistent with the results in Fig . 8 where categories 4 , 6 , and 7 show the weakest correlations between and compared to the other categories . In particular , most of the values are very small , less than 0 . 2 , which is consistent with the results in Fig . 8H where no correlation is found ( as indicated by the high p value ) . Furthermore , one can map the different clusters appearing in Fig . 8B–H into the clusters that appear in Fig . 10A–G . For example , the set of topologies showing a can be mapped to the cluster in Fig . 9G with ranging between 0 and 0 . 4 and that in Fig . 10G with ranging between 1 . 0 and 1 . 5 . Similarly , in Fig . 8D , the separate two sets of topologies showing a low parameter robustness value ( between −0 . 2 and 0 ) can be mapped to the two clusters in Fig . 9C on the top left side with ranging between 0 and 0 . 3 for one , and between 0 . 2 and 0 . 4 for the other , and the two clusters in Fig . 10C with negative values of . Further investigation of the set of topologies corresponding to the different clusters goes beyond the scope of the work presented here . The main proteins/receptors involved in E . coli chemotaxis are CheA , CheW , CheB , CheR , CheZ , and CheY . E . coli uses an anticlockwise rotation of its flagella to move forward . A decrease or increase in the concentration of nutrients ( chemo-attractants ) or harmful chemicals ( chemo-repellents ) , respectively , provokes a change to a clockwise rotation which causes the E . coli to tumble and thus change direction . This signal to the flagella is controlled by the chemotaxis protein CheY . A stimulus ( i . e . , a change in the chemical concentration in the environment ) is sensed by periplasmic binding proteins which couple to CheA in the inner membrane with the help of CheW . An increase in chemo-attractant concentrations inhibits the phosphorylation of the receptor complex CheA-CheW ( RC-P ) ( Fig . 11A ) while a chemo-repellent enhances it ( Fig . 11B ) . RC-P gives its phosphate group to both CheY and CheB ( CheY-P , CheB-P ) . CheB-P demethylates glutamate residues while CheR enhances methylation . In turn , methylated glutamate ( M ) enhances the phosphorylation of the receptor complex . The chemotaxis protein CheZ helps speeding the autodephosphorylation of CheY-P [19]–[21] ( Fig . 11A–B ) . For simplicity , we further coarse-grain this network such that M and RC-P interact via a negative feedback loop ( Fig . 11C–D ) . In the supplementary material ( Fig . S3 ) , we demonstrate that there is no significant difference in the results between the topologies shown in Fig . 11A and its coarse-grained equivalent shown in Fig . 11C ( slope = 0 . 79 and 0 . 77 , respectively ) , though coarse-graining improves the Pearson correlation as it removes the redundant link leading to additional variability . The topology under the influence of a chemo-repellent has a much lower fraction of TR networks and shows no correlation ( r = −0 . 01 , p = 0 . 85 ) in its un-coarse-grained form ( Fig . 11B ) . It was important to remove the redundancy to obtain a significant correlation ( Fig . 11D , slope = 0 . 37 , r = 0 . 45 , p = 10−14 ) . Chemotaxis in many other bacteria is more complex and involves more proteins . One such protein is CheV which generally contains a phosphorylatable domain [24] . Here we consider all possible coarse-grained interactions between phosphorylated CheV ( CheV-P ) , RC-P , and M . The only assumption we make is that RC-P gives its phosphate group to CheV in addition to CheB and CheY ( Fig . 11E–F ) . With this , we obtain 33 possible sets of signed directed edges as listed in table 1 , where we are considering all 3 possibilities ( i . e . , activation , deactivation , or no link ) for the 3 suggested links . For each of the 27 topologies , we compute the and values ( Figs . 12 , 13 ) and the slopes of the regression between and values of their corresponding TR networks ( Figs . 12B , 13B ) . We compare the results with that of the E . coli topology both under positive ( Fig . 12 ) and negative ( Fig . 13 ) stimuli . Topologies 1–3 , 5–7 , 10–12 , 16 , and 19 are highly improbable as they have no significant number of TR networks within the sampled parameter space when chemo-repellents are the stimulus ( Fig . 13 ) . Topologies 4 , 8 , 18 , 25–27 are also eliminated as they either show either a negative or no correlation ( Fig . 12C–D , 13C–D ) under either a chemo-attractant or a chemo-repellent . Topologies 9 , 14 , 18 , 21 , 23 are less likely than the rest ( 13 , 15 , 20 , 22 , 24 ) as they have a weaker correlation between and than E . coli as deduced from the lower p values ( Fig . 12C , 13C ) . Topologies 20 and 22 are less robust to input perturbation than Ecoli when chemo-repellents are the stimulus ( Fig . 13 ) , and 24 has a significantly smaller slope . Finally 13 is more robust to parameter perturbations than 15 . The distributions of , for each topology are shown in Figs . S4 , S5 , S6 .
In this work , we demonstrated that there is a general positive power-law correlation between environmental and genetic robustness in TR networks , and a statistically significant trend to a directly proportional linear relationship between the two in the limit of large networks . Conversely , monotonically responsive and non-responsive ( NP ) networks show a weaker relationship than TR ones . Furthermore , this distinction between the two classes becomes more prominent as the size of the networks increases . Therefore , this relationship associated with TR may be relevant to the evolution of biochemical networks . While other factors have played a role in the evolution of genetic robustness , our results show that , for TR networks , as the system evolves to withstand external environmental perturbations , it will , with high probability , concomitantly become robust to certain genetic perturbations . We speculated that the inverse of the slope is proportional to where is the number of nodes . We performed the corresponding regression and obtained for and ( r = 0 . 9439 , p = 0 . 008 ( 1-tailed ) , p = 0 . 016 ( 2-tailed ) ) . To confirm our results , we performed a Bayesian analysis for the model with a uninformative flat prior on the parameters and obtained and from the second moments of the posterior . Thus the Bayesian analysis confirms the linear regression . For NP networks , the same regression gives for and ( r = 0 . 7608 , p = 0 . 07 ( 1-tailed ) , p = 0 . 13 ( 2-tailed ) ) . The value of for TR networks in the limit of N large is thus 0 . 99±0 . 08 while that of NP networks is 0 . 68±0 . 21 . While the latter's regression is not significant at the p = 0 . 05 level , the value of the intercept did not significantly change for different power values ( we tried and ) . The statistically significant regression for TR networks implies that as a network evolves to be more robust to input perturbations it will also evolve to be robust to parameter perturbation ( and vice versa ) at a faster rate . Most importantly , as the size of TR networks becomes larger , the linear relationship between the logarithms quantifying robustness to input and that to parameter perturbations implies that for larger TR networks , is , with statistical significance , and within the computed uncertainty , proportional to while for larger NP networks , tends to be proportional to . As standard in the analysis of power-law relationships , we computed the regression using logarithms . For specific biological situations , it may be conceptually more appropriate to compute a direct fit , but for general random networks , we know of no such principle . An exponential fit between and for different numbers of nodes would be difficult to interpret as the power-law is changing with the number of nodes , tending to a constant only as the number of nodes becomes large . A drawback of our method is that the random generation of large networks does not account for reducible topologies which can introduce more variability and thus more error and a lower correlation between the two robustness measures . This makes a comparison between the correlation coefficients of topologies of different sizes a trifle problematic . However , the space of topologies grows so rapidly with the number of nodes that the likelihood of randomly selecting a reducible network decreases precipitously . Similarly , the averaging method does not distinguish between links contributing to the robustness of either input or parameters and those that do not . A method that could pinpoint such links would be useful in this context . Our results on the adaptability of 3-node motifs differ somewhat from Ref [30] due to our use of a qualitative test , the Pearson shape correlation , for assessing the transient response property of a network . We are not aware of a biologically plausible rationale for an explicit cutoff on the size or speed of a response as biological examples can exhibit both extremes of size or duration of transients . The general motifs shown in the literature [30] need further qualification to be deemed biochemically adaptable . For example , many topologies containing NFL are nonresponsive . Conversely , we show that a subcategory of NFL , NFL1 type2b is particularly robust and exhibits a strong correlation between robustness to input and parameter perturbations ( Fig . 6D ) . Our results are consistent with biological networks described in the literature . For example , we show that the coarse-grained network topology of E . coli chemotaxis , as described in the literature [17]–[21] , is NFL1 type2b ( Fig . S7C ) , as follows . When the receptor complex is activated , it causes the phosphorylation of the response regulator CheY leading to increased probability of tumbling . An increase in the chemo-attractant level ( I ) suppresses the activity of the complex and , in turn , the phosphorylation of CheY ( Fig . S7A ) . If I is the input ( which we set to always activate the input-receiving node in our computations , for consistency ) , then we can define the concentration of the input-receiving node as that of the deactivated complex , X1 ( i . e . , the activated complex represent X1 in its deactivated form ) . In this case , X1 deactivates CheB which inhibits methylation ( M ) . M activates the complex which is equivalent to deactivating X1 . The latter inhibits the phosphorylation of CheY ( the output ) and thus decreases the probability of tumbling ( Fig . S7B ) . An example of IFF is the Ras model of MAPK cascades discussed in Ref [53] . The input simultaneously activates two factors , SOS and RasGAP which activate and deactivate Ras , respectively and simultaneously . The model is shown [53] to be responsive only when the activation of SOS is faster than that of RasGAP . Thus , one can further coarse-grain it by removing the intermediate node between Ras and the input node ( Fig . S8 ) . This reduces to an IFF1 topology . In Ref [30] , all NFL topologies wherein the output node directly affects the input receiving node were found to be not robust or transiently responsive . While consistent with our results showing that NFL7–NFL10 are not robust , note that when the negative feedback loop has a direct and an indirect path , the outgoing and incoming links of the input receiving node must have the same sign for adaptability and parameter robustness to be achieved ( see NFL4 and NFL6 as opposed to NFL3 and NFL5 in Fig . 6B ) . Our work goes beyond pointing out general motifs . We refine subcategories within these motifs and show that , in fact , they do vary in their biochemical adaptation properties . Traditionally , network motifs represent subgraph topologies that appear in biological networks much more often than one would expect in a randomly constructed network [49] , and specific functions were assigned to different types of motifs [41] , [46]–[48] . The validity of this approach has been questioned as the frequency of occurrence of these motifs was not statistically significant when compared with corresponding ( i . e . same degree ) randomly constructed networks [54] . It was argued that one cannot analyze subgraphs independently of the rest of the network as interactions will drastically change the functions assigned to the particular topology [55] . In our work , a motif does not represent a subgraph , rather the topology of the backbone of ( possibly much ) bigger networks . We use our approach to differentiate between plausible models of the role of the CheV-P protein in bacterial chemotaxis . We find that there are only a few possible ways that CheV-P can be linked to RC-P and M . We suggest that while there are at most 9 possible topologies , the most plausible one has M enhancing the phosphorylation of both CheV and the receptor complex . Some specific network features have been associated with robustness to environmental variation in bacterial gene expression . Insulating gene expression by different modes of control , from activation to repression depending on the required high or low activity , has been suggested as a general control feature [56] . Our approach to motif discovery can be extended to networks with backbones with more than 3 nodes . While exhaustive enumeration of small motifs with desired functions is fascinating [30] , [41]–[43] , it is neither immediately evident nor has it been demonstrated in any context that such motifs could be put together to make systems with multiple functions while preserving the robustness or responsiveness properties of the separate motifs . To get to the point where we can plausibly discuss architectural principles in biology , it seems necessary to find general characteristics of classes of networks of all sizes that could perform functions of biological interest . Our work is a step towards this goal .
Following the same initial setup as in Ref [30] , a biochemical network is represented with a directed signed graph wherein the nodes of the network represent the enzymes . The latter can either be active or inactive and are able to interconvert between the two states . Thus , the elements of the corresponding adjacency matrix can take the values , implying that node deactivates node , has no effect on , or activates , respectively . No parallel links going in the same direction are allowed , i . e . , cannot be >1 . We divide the nodes into two types , varying nodes and fixed nodes . The latter correspond to inputs and basal enzymes which are added to each network to ensure that each node has at least one activating and one deactivating link . Thus , for an N-node network with inputs and basal enzymes , is an matrix where . These concentration values are represented by an vector ( 1 ) where is the concentration of the active form of the enzymes at time , ( 2 ) and are the time-independent concentrations of the inputs and basal enzymes , respectively . Assuming that the enzymes are non-cooperative and hence that they obey the Michaelis-Menten kinetics , the rate equations governing the dynamics of the network take the following compact form ( 3 ) where is a unit step function defined as ( 4 ) and are the catalytic and Michaelis-Menten rate constants for the regulation of enzyme by enzyme , for and . In equation ( 3 ) , the total concentration of each enzyme is kept constant and normalized ( i . e . , the concentration of the active form of an enzyme plus that of its inactive form is always equal to one ) . Thus , for . For all simulations presented here we use only one input , . This particular choice of input concentration should not have a significant effect on our qualitative results , as we have checked explicitly while formulating our hypothesis . Networks are allowed to reach steady state before the concentration of the input is perturbed . We are only concerned with the relative change in steady state concentrations . N-node networks are identified with directed signed graphs representing their topology and a set of parameters , for , where is the total number of sampled topologies , excluding those wherein one or more nodes have a total degree of zero or the output node cannot be reached from the input receiving node ( Fig . 1 ) . Each topology , in turn , is sampled over a large number of random networks , i . e . , a large number of randomly chosen parameter sets , for , where is the total number of sampled networks ( sets of parameters ) for topology . The total number of parameters in each set ( i . e . , length of the vector ) varies depending on the topology . The order of magnitude of increases exponentially with the size of the networks . For example , values for are around 20 , around 40 for , and 500 for . Similarly , the number of iterations ( i . e . , number of sampled networks ) required ( see Selection Criterion below ) also increases with . For example , for values range between 104 and 105 , while for , values range between 106 and 107 . Typically , an iteration takes less than 10−3 seconds of CPU time for small networks ( ) , thus 1 to 2 minutes to test each topology . On the other hand , for large networks ( ) , an iteration typically takes around 0 . 04 seconds of CPU time , 3 to 5 days for testing each topology . We define a TR network as one whose output dynamics has a non-monotonic transient between two steady states as a response to input change ( i . e . , the steady state values before input perturbation and that after input perturbation ) . We find the transition time ( i . e . , the time at which the concentration is maximal/minimal before it starts decreasing/increasing again ) and enzyme concentrations , and ( i . e . , concentrations at ) , by solving for the turning point ( 5 ) where is the concentration of node k at steady state . We use a Pearson test to determine if a given network is TR . First , we define two functions and as model functions of perfect adaptability and non-adaptability ( a monotonically changing network ) , respectively ( Fig . 2 ) : ( 6 ) ( 7 ) Define corresponding Pearson shape correlations and as ( 8 ) ( 9 ) where , and are the mean values of , and . With this , a network is deemed TR if . Comparing the absolute values or and instead of the actual values is necessary . Even though our definitions of and will most likely lead to positive values , this is not always the case . The reason is that eq . ( 6 ) and ( 7 ) assume the perfect case where the differences in the concentrations from the initial steady state have always the same sign ( as in Fig . 2 E and F ) . If instead the difference in concentrations at the transition point is smaller than that at the final steady state ( i . e . , post-perturbation steady state ) , then and/or will have negative values . However , that does not matter since we are only interested in the shape of the time-course ( see Fig . 2G , for example ) . Note that the mean values are taken as the average over all the discretized time-steps; for example , for time steps . The size of the time-step , , is the same for all networks ( ) , but this is not the case for the number of time steps , , as the length of the time-course of each network varies depending on how long the network needs to reach a new steady-state ( i . e . , the rate equations in eq . ( 3 ) for all nodes reach zero again after input perturbation . Of course , computationally , the run will stop when the rate equation for all nodes is less than 10−10 ) . For example , in Fig . 14A and 14B we show the time-courses ( in blue ) of two different networks . The network in Fig . 14A needed around 130 seconds ( ) to reach a steady state , while that shown in Fig . 14B needed around 150 seconds ( ) . Simulations that take too long to reach a steady state ( ) are thrown away and not considered in the analysis ( i . e . , are thrown away without performing the Pearson test ) . This cutoff on the maximal number of time-steps allowed is chosen for computational efficiency . Preliminary results showed that for most networks , if a steady state was not reached within 2000 time-steps , it is unlikely it will be reached for a long time . Since we are only interested in the statistical results and since networks are chosen randomly , there is no reason to insist on including a network that takes a lot computational time to reach a steady state . We chose because we were looking for the largest time-step ( to improve computational time ) that does not change the statistical results . In preliminary runs , we compared the results of 3-node networks when using and . The finer time-step allowed more topologies to pass as TR . However , these topologies had very low fraction of TR networks and were removed after the cutoff . Moreover , the statistical results were the same both before and after the cutoff . As mentioned in Experimental Setup above , large networks ( 30-node ) take 3 to 5 days of CPU time for each topology . Using would increase this simulation time to over a month for each topology which is impractical . We test the robustness of the Pearson test described above by comparing the results from the 3-node simulations to those employing instead the Spearman correlation using the same definition of and ( Fig . S9A ) . Both are also compared to simulations using a different definition , and as follows: ( 10 ) ( 11 ) where is chosen here to be . This new definition allows and to get to the transition concentration , , at a slower rate , then after the transition point , , coincides with while relaxes back to the pre-perturbation steady state , at a much slower rate ( Fig . 14A , B ) . In all cases we find no significant difference between the results for 3-node simulations ( Fig . S9 ) . This does not mean that there are no variations within individual networks . For example , in Fig . 14 , we show the and values of the Pearson test for all the networks corresponding to a typical 3-node topology using both definitions ( Fig . 14C ) . We find that there are 90 out of 21579 networks that were deemed TR in one but NP in the other . A typical time-course where the outcome of the Pearson tests differ or agree are shown in Fig . 14A and Fig . 14B , respectively . In general , most networks do not fall into this category where the values of and are very close such that different definitions of and lead to different outcomes . In Fig . S9 , we verify that this change does not affect any statistical observations . To quantify the degree of robustness to input and parameter perturbations of a particular network , we calculate the relative change in the steady state concentrations of the output node due to perturbing the input and parameter values , respectively . Let and be the average of the sensitivity of the steady state concentration of the output to each input and each parameter , respectively . Then , ( 12 ) ( 13 ) where the node is the output node , is the set of and corresponding to each reaction/link ( i . e , the non-zero values ) , , is the total number of links , and is the set of steady state concentrations of the output node for input and parameter set . Defining the degree of input and parameter robustness of a network as inversely proportional to the values of and ensures all the inputs and parameters of the network are taken into consideration . Analyzing the rate functions of equation ( 3 ) ( see Steady State Analysis in the Text S1 for the detailed derivation ) , we obtain ( 14 ) ( 15 ) where , , and are the Jacobians with respect to the node concentrations , input , and parameters , respectively . A robust topology is one that gives rise to robust networks with a higher probability when tested with a large number of parameter sets . Quantitatively , the degree of robustness to input perturbations of a given topology is taken to be the geometric average of over all . Similarly , the degree of parameter robustness is the geometric average of . A TR topology is one that has a statistically significant number of TR networks . Topologies that do not have enough TR networks are rejected and excluded from any further analysis . With this , we are left with topologies out of the sampled ones . For each topology , we define two quantitative measures each for input ( and ) and parameter ( and ) robustness . and are the values of input and parameter robustness of TR networks ( 16 ) ( 17 ) ( 18 ) where if network passes the Pearson test and zero otherwise . and are the values of input and parameter robustness of NP networks ( networks that did not pass the Pearson test ) ( 19 ) ( 20 ) ( 21 ) where if network reaches a steady state ( see Selection Criterion below ) but it does not pass the Pearson test and zero otherwise . We choose the geometric average as more suitable than the arithmetic average as a conservative approach to detecting a possible correlation , as the latter gives too much weight to much larger outliers . A trial is rejected if it takes too long to reach equilibrium , or its corresponding Jacobian with respect to the node concentrations is singular ( i . e . , is noninvertible ) . With this , we obtain matrices for the relative errors and for and . and thus are determined when , , and the fraction of successful trials , , reach equilibrium values . We reject a topology if is obviously too small to be statistically significant or takes too long to reach equilibrium ( see below ) , indicating that the parameter space leading to TR networks for that topology is too small . We sample over 50 , 000 different topologies for each and all possible 3-node topologies ( 19683 ) , and for each we randomly sample over a large number of parameter sets from a uniform distribution within the ranges and ( whenever a link exists between vertices and ) . For the topologies were randomly generated such that the value of each in the corresponding adjacency matrix can take the values −1 or 1 with probability each , and a value 0 with probability . We generated different set of topologies with . We found no significant difference in the distributions of and values depending on . The results shown here represent the collection of all the sets . We automatically reject trials whereinWe investigate the effect of the choice of the ranges above by running two separate 3-node simulation . In the first , the parameters are chosen from a uniform distribution in the ranges and , while in the second , the parameters are chosen from a uniform distribution in the ranges and . For both ranges we find a significant correlation between robustness to input and parameter perturbations ( Fig . S10 ) . For range1 and range2 we obtain the respective values 0 . 38 and 0 . 54 for the slopes and 0 . 73 and 0 . 68 for the Pearson correlation ( Fig . S10A ) . Furthermore the difference in the slopes becomes insignificant when only networks appearing in both ranges are taken into consideration ( Fig . S10B ) . As discussed above , the degree of input and parameter robustness is seen as inversely proportional to the average of the sensitivity of the steady state concentration of the output to each input and each parameter , respectively . Then , ( 22 ) ( 23 ) Inserting ( 22 ) in ( 23 ) , we obtain ( 24 ) Similarly , for parameter perturbations , and ( 25 ) In this section we analyze the parameter robustness of different types of parameters . Thus , the parameters of a topology are now divided into categories . Their corresponding measures of robustness are now defined as ( 26 ) ( 27 ) for . Thus , the measure of robustness of a topology to perturbations in its parameters of category j takes the form ( 28 ) Note that a topology does not have to have parameters belonging to all the defined categories . Next , to obtain an idea about how robustness to input and parameter perturbations correlate within the networks of each individual topology , we calculate the value which is the value of the slope obtained from the linear regression on vs for topology and category .
|
Advances in the ways that living systems can be perturbed in order to study how they function and sharp reductions in the cost of computer resources have allowed the collection of large amounts of data . The aim of biological system modeling is to analyze this data in order to pin down the precise interactions of molecules that underlie the observed functions . This is made difficult due to two features of biological systems: ( 1 ) Living things do not show an appreciable loss of function across large ranges of environmental factors . ( 2 ) Their function is inherited from parent to child more or less unchanged in spite of random mutations in genetic sequences . We find that these two features are more correlated in a specific subset of networks and show how to use this observation to find networks in which these two features appear together . Working within this smaller space of networks may make it easier to find suitable underlying models from data .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"systems",
"biology",
"biology",
"computational",
"biology"
] |
2014
|
A Network Characteristic That Correlates Environmental and Genetic Robustness
|
Pheromones are used for conspecific communication by many animals . In Drosophila , the volatile male-specific pheromone 11-cis vaccenyl acetate ( cVA ) supplies an important signal for gender recognition . Sensing of cVA by the olfactory system depends on multiple components , including an olfactory receptor ( OR67d ) , the co-receptor ORCO , and an odorant binding protein ( LUSH ) . In addition , a CD36 related protein , sensory neuron membrane protein 1 ( SNMP1 ) is also involved in cVA detection . Loss of SNMP1 has been reported to eliminate cVA responsiveness , and to greatly increase spontaneous activity of OR67d-expressing olfactory receptor neurons ( ORNs ) . Here , we found the snmp11 mutation did not abolish cVA responsiveness or cause high spontaneous activity . The cVA responses in snmp1 mutants displayed a delayed onset , and took longer to reach peak activity than wild-type . Most strikingly , loss of SNMP1 caused a dramatic delay in signal termination . The profound impairment in signal inactivation accounted for the previously reported “spontaneous activity , ” which represented continuous activation following transient exposure to environmental cVA . We introduced the silk moth receptor ( BmOR1 ) in OR67d ORNs of snmp11 flies and found that the ORNs showed slow activation and deactivation kinetics in response to the BmOR1 ligand ( bombykol ) . We expressed the bombykol receptor complex in Xenopus oocytes in the presence or absence of the silk moth SNMP1 ( BmSNMP ) and found that addition of BmSNMP accelerated receptor activation and deactivation . Our results thus clarify SNMP1 as an important player required for the rapid kinetics of the pheromone response in insects .
Pheromones are chemicals that trigger or inhibit stereotyped social behaviors , such as aggregation , courtship and mating [1] , [2] , [3] . Studies on insects have contributed enormously to our understanding of pheromones [2] , [3] . The first pheromone characterized was bombykol— a volatile 16-carbon alcohol synthesized in the female gland of the silk moth , Bombyx mori [4] , [5] . Male silk moths use bombykol as a navigation cue to find female mates , and this pheromone can be sensed over long distances [5] , [6] . Volatile pheromones are typically comprised of hydrocarbon chains [7] , and are perceived by olfactory receptor neurons ( ORNs ) in the antenna of insects . One such pheromone , 11-cis vaccenyl acetate ( cVA ) , represents the only volatile pheromone known in the fruit fly , Drosophila melanogaster . This chemical is released from the ejaculatory bulb of the males [8] and is sensed by both males and females , the latter of which receive the pheromone during copulation [9] . The ORNs that sense the volatile cVA signal are housed in one type of olfactory hair on the antenna ( trichoid sensilla ) , referred to as T1 sensilla [10] . Detection of cVA modifies a host of behaviors including male-male aggression , social aggregation , male-female and male-male courtship behavior [11] , [12] , [13] , [14] , [15] . Due to the critical roles of pheromone-induced behaviors , the mechanisms underlying insect pheromone detection have been studied extensively . The receptors for cVA ( OR67d ) and bombykol ( BmOR1 ) belong to the insect olfactory receptor ( OR ) family [16] , [17] , [18] . Another OR , referred to as ORCO , is conserved in many insects , and in Drosophila serves as a co-receptor , which is broadly required for trafficking and function of other ORs [19] , [20] , [21] . Because pheromones are hydrophobic , their solubility depends in part on association with odorant-binding proteins ( OBPs ) or pheromone-binding proteins ( PBPs ) present in the endolymph of the sensilla [22] , [23] . In Drosophila , LUSH is the OBP required for sensation of cVA [24] . Upon binding cVA , LUSH has been reported to undergo a conformational change , which in turn activates OR67d [25] . However , another study concludes that cVA directly activates the receptor [26] . SNMP1 , which is a member of the CD36-scavenger family , also contributes to the pheromone response [27] , [28] . Mutations disrupting this integral membrane protein have been reported to eliminate cVA detection [27] , [28] . SNMP1 is expressed in the antenna in the dendrites of trichoid ORNs [27] , [28] , consistent with its role in cVA detection . Loss of SNMP1 also causes a dramatic increase in spontaneous activity of T1 ORNs [27] , [28]; although , the mechanism underlying the increased spontaneous activity is unknown . Here , we found that loss of SNMP1 did not eliminate cVA-evoked activity , and was required for fast inactivation . The onset of the cVA-induced action potentials was delayed , and the activity increased slowly . Following cessation of the cVA stimulus , the activity continued for many minutes . This contrasted with the wild-type response , which terminated in less than a second . Thus , inactivation was delayed dramatically . We also demonstrated that snmp11 mutant ORNs did not exhibit an increase in spontaneous activity . Rather , the high frequency of action potentials was due to the highly persistent activity initiated by cVA in the environment . We expressed the bombykol receptor from the silk moth ( BmOR1 and BmORCO ) in Xenopus oocytes , and found that addition of the silk moth SNMP1 significantly increased the kinetics of the activation and inactivation of the receptor . Thus , we conclude that SNMP1 functions in promoting the rapid activation and inactivation of pheromone receptors to achieve fast onset and termination of pheromone sensitive ORNs .
To characterize the role of SNMP1 in the cVA response , we performed single sensillum recordings , initially using conditions similar to those described previously [27] , [28] . We recorded action potentials from trichoid sensilla ( T1 ) , which contain OR67d-expressing ORNs . Consistent with earlier studies [27] , [28] , the ORNs from snmp11 females showed high “spontaneous activity” relative to wild-type females ( Figure 1A and 1B ) . The females used in these experiments , and in the previous reports on snmp1 , were raised in groups , which included males and other female flies . Surprisingly , when we modified the rearing paradigm , and maintained the snmp11 females in isolation from the pupal stage through adulthood , the high “spontaneous activity” was eliminated , and the frequency of action potentials in the absence of cVA was similar or marginally lower ( though not significantly ) than in wild-type females ( Figure 1A and 1B ) . We also recorded background action potentials from singly housed snmp11 mutant males . Young males ( ≤2 days old ) displayed low background activity , similar to females ( Figure 1C ) . In contrast , older mutant males exhibiting higher background activity ( Figure 1C ) . This age-dependent increase did not occur with snmp11 females ( Figure 1C ) . Because males but not females produce cVA , these findings suggest that cVA released from males induce the background action potentials . To test whether exposure to environmental cVA caused the high basal activity , we reared snmp11 females under isolation , and then exposed them to cVA for 24 hours . We then measured action potentials elicited by OR67d ORNs in the absence of any cVA during the electrophysiological measurements . Pre-incubation with 10% or 100% cVA caused the snmp11 females to show significantly higher activity than the similarly treated wild-type females ( Figure 1D ) . Pretreatment of snmp11 flies either with the vehicle ( paraffin oil; 0% cVA ) or with 1% cVA had no significant effect ( Figure 1D ) . These results support the proposal that the elevated activity elicited by the grouped snmp11 mutants was caused by the environmental cVA derived from male flies . In addition to T1 , the antenna contains other trichoid sensilla that respond to fly odors [29] . To address whether SNMP1 function was required generally in ORNs for attenuating the activity of environmental fly odors , we recorded the basal activity of T3 sensilla from singly and group housed male and female flies . The action potentials from the T3 sensilla exhibited three size amplitudes ( A , B and C ) , each of which was derived from distinct ORNs ( Figure S1A ) . The frequencies of action potentials from the three different ORNs were indistinguishable between wild-type and snmp11 males and females , regardless of whether they were individually or group housed ( Figure S1B ) . Thus , all pheromone responsive ORNs in the SNMP1 mutants do not show higher basal activity in response to pheromone pre-exposure . A major problem with the hypothesis that the higher background activity in snmp11 mutants is due to environmental cVA , is that the snmp11 animals are reported to be completely insensitive to cVA [27] , [28] . One possibility was that the insensitivity to cVA was caused by the perpetual high background activity , which caused the OR67d ORNs to be unresponsive to further stimulation . To test this possibility , we stimulated the singly housed snmp11 females with cVA . However , these animals with low background activity still failed to respond to cVA , even at the highest concentration tested ( Figure 2A and 2B ) . The preceding results still do not resolve the question as to how environmental cVA could lead to elevated background activity , given that the mutant OR67d ORNs are unresponsive to cVA stimulation during single sensillum recordings . One explanation is that the cVA stimulation is inadequate , and that the snmp11 flies must be exposed to higher levels of cVA , such as those that might be achieved through close interactions with males [29] . In support of the concept that SNMP1 might not be absolutely required for activation by cVA , ectopic expression of OR67d in ab3A neurons , which lack SNMP1 , is sufficient to elicit a response to cVA if it is applied in close proximity to the sensilla [29] . Therefore , instead of using the conventional delivery method , in which cVA was diluted into air that was streamed through a tube , we puffed cVA from a pipette placed in very close proximity to the antenna ( close-range application ) . 100% cVA ( 1 second ) delivered by this close-range application evoked a robust response in wild-type flies ( Figure 2C ) . Of significance here , the snmp11 females also responded to the cVA application , although not as strongly as wild-type ( Figure 2C and 2D ) . The snmp11 females elicited responses to 10% and 100% cVA , but not to 1% or lower levels of cVA ( Figure 2D ) . In addition , there was a significant delay in production of the action potentials ( Figure 2C and 2E ) . We rescued these phenotypes by expressing a wild-type snmp1 transgene ( UAS-snmp1 ) under control of the snmp1-Gal4 ( Figure 2C , 2E and 2F ) . To provide a negative control , we tested Or67dGal4 mutant females and found that close-range application did not evoke action potentials in these animals ( Figure 2C ) . An additional and pronounced aspect of the snmp11 phenotype occurred after termination of the cVA stimulus . When we exposed wild-type flies to a transient cVA puff , the spiking activity of wild-type quickly decreased , as the firing declined by 50% in ∼1 second ( Figure 2F; t1/2; the data were binned every 0 . 5 seconds , resulting in calculations of the t1/2 to the nearest 0 . 5 second ) . In stark contrast , the weaker activity in snmp11 flies was very long-lasting and showed almost no decline 20 seconds after the cVA puff ( Figure 2F; t1/2>50 seconds ) . Strikingly , the spiking activity was still robust after 10 minutes ( Figure 3A and 3B ) . Application of the vehicle ( paraffin oil ) to the snmp11 fly had no effect ( Figure 3C ) . It was possible that the deactivation defect exhibited by snmp11 flies was a manifestation of the weak cVA response . In other words , rapid deactivation might depend on a robust response to the initial stimulus . To address this possibility , we stimulated wild-type flies with a low level of cVA ( 0 . 01% ) that evoked an initial firing rate comparable to that induced by exposing snmp11 to a 10 , 000-fold higher concentration of cVA ( 100% ) . Although the evoked firing rates were similar in these wild-type and mutant flies , only the snmp11 flies exhibited persistent action potentials following removal of cVA ( Figure 3D ) . We further investigated the slow activation of snmp11 Or67d ORNs by presenting a prolonged cVA stimulation ( 12 seconds ) . Unlike the response by wild-type flies , which reached the maximum activation within 0 . 5 seconds during the 100% cVA application , the snmp11 flies showed a gradual increase in firing activity during the stimulation ( Figure 3E; t95 = 7 . 5 seconds , time to reach 95% of the maximum response; data were binned every 0 . 5 seconds ) . Again , this was not a side effect of the weak response , as wild-type flies that displayed a similarly weak response ( evoked by 0 . 01% cVA ) also reached the maximum firing rate in the first 0 . 5 second window after stimulation ( Figure 3E ) . Elongating the stimulation to 20 seconds did not further increase the activity elicited by the 100% cVA stimulation ( Figure S2 ) . While a short close-range puff ( 1 sec ) of 1% cVA did not evoke a significant response in snmp11 flies ( Fig . 2D ) , a 20-second application of 1% cVA evoked a low level of spikes in snmp11 flies , which initiated after ∼7 . 5 seconds ( Figure S2 ) . This finding indicated that prolonged stimulation could partially compensate for the inadequacy of the low stimulation intensity in the snmp11 mutant flies . On the basis of the findings here , we conclude that the so-called “spontaneous activity” exhibited by snmp11 Or67d ORNs was not spontaneous neuronal activity . Instead , the action potentials were the result of extremely persistent cVA-induced activity , which remained long after removal of the cVA stimulus . Thus , SNMP1 was required not only for high cVA sensitivity , but also to achieve rapid on- and off-kinetics in response to cVA . To address whether SNMP1 function was specific to either cVA or its receptor ( OR67d ) , we tested whether SNMP1 affected the response to the silk moth ( Bombyx mori ) pheromone ( E , Z ) -10 , 12-hexadecadien-1-ol ( bombykol ) , after we ectopically expressed the bombykol receptor , BmOR1 in OR67d ORNs ( UAS-BmOr1 and OR67dGal4 ) . As previously reported [14] , [30] , conventional application of bombykol to these transgenic flies evoked action potentials , which quickly terminated ( Figure 4A ) . We then introduced BmOR1 in the snmp11 mutant background , and found that the ORNs still responded to bombykol applied by the conventional delivery method , but less robustly ( Figure 4A and 4B ) . In addition , loss of SNMP1 slowed the activation and deactivation of the bombykol-evoked response ( Figure 4C ) . Although the snmp11 mutation had a profound effect on deactivation , the phenotype was not as dramatic as with cVA . Consistent with this observation , pre-exposure of the transgenic flies to bombykol for 24 hours did not increase the basal firing rate ( Figure S3 ) . Nevertheless , the similar phenotypes after application of either cVA or bombykol suggested that SNMP1 functioned in the rapid activation and termination of pheromone-evoked neuronal activity . Co-expression of BmOR1 and BmORCO is sufficient to form functional ion channels in Xenopus oocytes [31] . We took advantage of this in vitro reconstitution system to address whether SNMP1 directly affected the activation and inactivation of the pheromone receptor . We expressed the bombykol receptor complex , BmOR1 and BmORCO , either with or without the silk moth SNMP1 ( BmSNMP ) in Xenopus oocytes and performing two-electrode voltage clamp recordings . To quantify the kinetics of the bombykol response , we measured the half-time of the activation during bombykol application , and the half-time of the inactivation following the wash out of the pheromone . We found that upon introduction of the BmSNMP , the activation ( t1/2 ) was nearly three-fold faster ( Figure 4D and 4E; no BmSNMP , 19 . 1±4 . 5 seconds; +BmSNMP , 6 . 8±1 . 3 seconds ) . Moreover , the inactivation ( t1/2 ) was accelerated eight-fold in the cells expressing BmOR1/BmORCO in combination with BmSNMP ( Figure 4D and 4F; no BmSNMP , 120 . 2±26 . 3 seconds; +BmSNMP , 15 . 0±1 . 7 seconds ) . These results support a role for SNMP1 in directly accelerating receptor activation and inactivation in response to pheromone stimulation . It has been reported that the OR67d ORNs from lush1 , snmp11 double mutant flies also display high “spontaneous activity” [27] , [28] . Therefore , we tested whether these action potentials were also due to very slow termination of the activity evoked by cVA . We first tested isolated lush1 mutant females using the close-range application assay . Indeed , these flies responded to the 100% cVA stimulation ( Figure 5A ) as previously reported [26] , and the response terminated within a few seconds after cessation of the stimulation . The lush1 , snmp11 double mutant female flies raised in isolation did not show high spontaneous activity and also responded to the 100% cVA stimulation ( Figure 5A ) . Similar to the snmp11 flies , the response from the lush1 , snmp11 double mutant showed very slow termination kinetics that persisted after the stimulation . ( Figure 5A and 5B ) .
Mutations that disrupt SNMP1 are reported to cause two impairments in OR67d ORNs in Drosophila [27] , [28] . The first is insensitivity to cVA , and the second is increased spontaneous activity of OR67d ORNs in the absence of cVA stimulation . This latter phenotype motivated the proposal that the presence of SNMP1 somehow suppressed the spontaneous activity of OR67d [28] . Rapid termination is critical for an appropriate pheromone response , particularly for insects that use pheromones as tracking cues such as the silk moth , which relies on pheromone trails that are composed of intermittent odor pockets separated by clean air spaces [32] . Thus , to follow this trail , the pheromone-sensitive ORNs must quickly terminate their responses . It has been suggested that rapid inactivation of the pheromone response is due to degradation mediated by pheromone-degrading enzymes [23] , [33] , [34] . However , a mathematical model proposed that a soluble scavenger is required for the fast clearance of bombykol in the sensilla lymph , as enzymatic degradation may not be fast enough [35] . In the current work , we found that in contrast to previous studies , loss of SNMP1 neither eliminated cVA responsiveness nor caused high spontaneous activity . In support of these conclusions , snmp11 mutant females raised in isolation from males did not display elevated spontaneous activity . However , the snmp11 females exhibited high frequencies of action potentials if they were raised along with males , or if the isolated females were exposed to cVA prior to performing the recordings . The snmp11 mutation also did not eliminate cVA responsiveness , since the Or67d ORNs produced cVA-induced action potentials when we puffed the pheromone in close range to the mutant females . Thus , SNMP1 was not absolutely essential for OR67d ORN activation . This conclusion is supported by the finding that when OR67d is ectopically expressed in basiconic ORNs , which lack SNMP , the ORNs can be activated by cVA , if it is applied directly to the sensilla [29] , Of primary importance here , SNMP1 was required for rapid kinetic responses to cVA—both for rapid activation and termination of the responses . The pheromone-induced action potentials were dramatically delayed as they persisted for longer than 10 minutes , as opposed to ∼1 second for wild-type . Slow termination of cVA-induced responses also occurs upon introduction of SNMP1 antibodies to the recording pipet in wild-type flies [28] . We propose that the so-called spontaneous activity displayed by snmp11 null flies , was a consequence of extremely long-lived activity of OR67 ORNs following exposure to environmental cVA . In addition to OR67d , ORCO and SNMP , a phospholipid flippase ( dATP8B ) and an OBP referred to as LUSH contribute to the sensitivity of ORNs to cVA . Loss of dATP8B affects the function of odorant receptors [36] , [37] , at least in part by decreasing the concentration of OR67d in the ORN dendrites [37] . However , the role of LUSH is controversial . While OBPs are typically thought to be carriers that transport hydrophobic odorants through the aqueous endolymph to the receptors [22] , an in vitro study indicates that the cVA-LUSH complex is the activating ligand for OR67d [25] . This conclusion has recently been questioned , in part because OR67d neurons devoid of LUSH are activated by strong cVA stimulation in vivo [26] . Consistent with this latter report , we also found that cVA evoked responses in the lush1 mutants and lush1 , snmp11 double mutants if the pheromone was applied using the close-range application assay . Therefore , we favor the proposal that OR67d ORNs are activated directly by the pheromone . SNMP1 function does not appear to be specific to cVA since the initiation and termination of the bombykol responses were also delayed in transgenic flies expressing the silk pheromone receptor , BmOR1 . However , the delayed termination in the absence of SNMP1 was not as dramatic in response to bombykol as compared to cVA . The ORNs in T3 sensilla also express SNMP1 and respond to odors from fly bodies [27] , [28] , [29] . However , the T3 ORNs from wild-type or snmp11 males or females raised in groups or in isolation displayed similar basal activities . Thus , loss of SNMP1 does not always result in extremely prolonged activities in trichoid ORNs that are exposed to their ligands . A key question is whether SNMP1 regulates the pheromone response at the level of the receptors , or whether it modulates ORN activity downstream of receptor activation . To address whether SNMP1 activity modulated the response at the level of the receptors , we expressed the bombykol receptor complex in Xenopus oocytes , since this in vitro expression system was not likely to express other downstream signaling proteins that functioned in insect ORNs . We found that introduction of SNMP1 accelerated receptor activation by bombykol , and promoted rapid inactivation during wash out of the pheromone . A simple explanation for this result is that the pheromone binds to and dissociates from the receptor faster in the presence of SNMP1 . We propose that SNMP1 facilitates the association and dissociation between ligands and receptors so that the receptor activation and inactivation are accelerated ( Figure 6 ) . On the surface , such a dual function might seem surprising , as association and dissociation are opposing processes . In this context it is noteworthy that an enzyme can increase both the forward and reverse reaction rates by lowering the activation energy of a reversible reaction . The bombykol binding site in BmOR1 is proposed to consist of a large hydrophobic cavity buried between the transmembrane domains [38] . Thus , the interface between the hydrophilic sensillum lymph and the hydrophobic cavity inside the receptor might present a barrier preventing rapid on and off of the interaction between the pheromone and receptor . We suggest that SNMP1 helps overcome this barrier by facilitating the association and dissociation between the free pheromone in the sensillum lymph , and the hydrophobic pocket in the receptor ( Figure 6 ) . The barriers may vary among different receptors and thus the energy required for overcoming a barrier without SNMP1 might also be variable , potentially explaining why the severity of impairments resulting from loss of SNMP1 differ among odorant-receptors . Finally , it is noteworthy that except for cVA , no other volatile pheromones are known in flies . In support of the existence of additional volatile pheromones , trichoid neurons other than T1 can be activated when fly cuticular extracts are released in their immediate proximity [29] . However , these neurons do not respond to the conventional odorant delivery assay . Tests for pheromone candidates other than cVA may have failed as a consequence of a lack of sensitivity provided by the conventional method for odorant delivery . Approaches that stimulate flies with high levels of pheromones may more closely replicate the situation in environments in which the animals are in close proximity , and may offer improved methods for identifying new volatile pheromones in Drosophila . These approaches include physically positioning odorants very close to the fly antenna [26] , [29] , or puffing the odorant close to the antenna as described here , which provides the additional advantage of more precise temporal control .
The mutant alleles and transgenic lines were: Or67dGal4 , snmp11 , snmp1-Gal4 , UAS-snmp1 [27] , lush1 [24] and UAS-BmOr1 [30] . cVA ( 99% purity ) and bombykol ( 95% purity ) were from Pherobank . The error bars represent SEMs . To assess statistical significance , we used the one-way ANOVA with Bonferroni-Holm post hoc analysis to compare multiple samples , and unpaired Student t-tests for comparing pairs of data .
|
Pheromones are chemicals produced and released by animals for social communication with other members of their species . For example , male fruit flies produce a volatile pheromone that is sensed by both males and females , and which functions in gender recognition . This volatile male pheromone , called 11-cis vaccenyl acetate , is detected by olfactory neurons housed in hair-like appendages on the insect antenna . To effectively sense the pheromone , especially during navigation , the olfactory neurons must respond rapidly , and then quickly inactivate after the stimulation ceases . We found that a CD36-related protein referred to as sensory neuron membrane protein 1 ( SNMP1 ) was required by olfactory neurons for the rapid on and off responses to 11-cis vaccenyl acetate . Loss of SNMP1 reduced the initial sensitivity to the pheromone , and then caused a strikingly slower termination of the response after removal of the pheromone . Our findings demonstrate that SNMP1 is a critical player that allows olfactory neurons to achieve sensitive and rapid on and off responses to a pheromone that is critical for social interactions in insects .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"and",
"life",
"sciences"
] |
2014
|
Requirement for Drosophila SNMP1 for Rapid Activation and Termination of Pheromone-Induced Activity
|
Chronic kidney disease ( CKD ) is an increasing global public health concern , particularly among populations of African ancestry . We performed an interrogation of known renal loci , genome-wide association ( GWA ) , and IBC candidate-gene SNP association analyses in African Americans from the CARe Renal Consortium . In up to 8 , 110 participants , we performed meta-analyses of GWA and IBC array data for estimated glomerular filtration rate ( eGFR ) , CKD ( eGFR <60 mL/min/1 . 73 m2 ) , urinary albumin-to-creatinine ratio ( UACR ) , and microalbuminuria ( UACR >30 mg/g ) and interrogated the 250 kb flanking region around 24 SNPs previously identified in European Ancestry renal GWAS analyses . Findings were replicated in up to 4 , 358 African Americans . To assess function , individually identified genes were knocked down in zebrafish embryos by morpholino antisense oligonucleotides . Expression of kidney-specific genes was assessed by in situ hybridization , and glomerular filtration was evaluated by dextran clearance . Overall , 23 of 24 previously identified SNPs had direction-consistent associations with eGFR in African Americans , 2 of which achieved nominal significance ( UMOD , PIP5K1B ) . Interrogation of the flanking regions uncovered 24 new index SNPs in African Americans , 12 of which were replicated ( UMOD , ANXA9 , GCKR , TFDP2 , DAB2 , VEGFA , ATXN2 , GATM , SLC22A2 , TMEM60 , SLC6A13 , and BCAS3 ) . In addition , we identified 3 suggestive loci at DOK6 ( p-value = 5 . 3×10−7 ) and FNDC1 ( p-value = 3 . 0×10−7 ) for UACR , and KCNQ1 with eGFR ( p = 3 . 6×10−6 ) . Morpholino knockdown of kcnq1 in the zebrafish resulted in abnormal kidney development and filtration capacity . We identified several SNPs in association with eGFR in African Ancestry individuals , as well as 3 suggestive loci for UACR and eGFR . Functional genetic studies support a role for kcnq1 in glomerular development in zebrafish .
Chronic kidney disease ( CKD ) affects approximately 15% of U . S adults [1] . Due in part to increasing rates of diabetes and obesity , the prevalence of CKD continues to rise [1] . Marked variability in the incidence of CKD suggests that factors other than diabetes and hypertension contribute to its etiology [2] . Recently , we identified 16 genomic loci associated with estimated glomerular filtration rate ( eGFR ) , a primary measure of CKD , using genome-wide association studies ( GWAS ) in a combined sample of 67 , 093 European ancestry individuals from the CKDGen consortium [3] , [4] . However , these loci only account for 1 . 4% of the eGFR variation , suggesting that additional loci remain to be identified [5] . African American ethnicity is a well-established risk factor for CKD , and rates of end-stage renal disease ( ESRD ) are up to 4-fold higher among African Americans as compared to European Americans [6] . Several prior studies , including the FIND consortium , have performed linkage analysis of diabetic ESRD [7]–[9] . Recent genome-wide admixture mapping studies identified genetic variation in the regions of MYH9 and APOL1 on chromosome 22 that may explain up to 70% of the differences in ESRD rates between European and African Americans [10]–[12] . While this finding has great implications for ESRD , recent evidence also suggests that African Americans progress faster from moderately decreased kidney function to ESRD , spending less time in the recognized earlier stages of CKD [13] , [14] . The identification of additional risk factors for CKD , including genetic loci in association with eGFR , may help to advance our understanding of the underpinnings of CKD in African Americans . Thus , the goal of this study was to uncover loci for kidney traits in African ancestry participants in the Candidate-gene Association Resource ( CARe ) Consortium . CARe is a consortium of 9 studies which form a combined population of approximately 40 , 000 African and European Americans genotyped on the IBC array [15] and approximately 8 , 000 African American participants genotyped using a GWAS platform . We aimed to interrogate known regions previously associated with eGFR in European Ancestry populations [3] , [4] , as well as to perform discovery analyses in African ancestry populations . In order to gain further insight into the functional implications of our associations , genes at or near loci of interest were knocked down by morpholino injection in zebrafish , and kidney gene expression and function were investigated using in situ hybridization and glomerular filtration assays .
Genome-wide association analyses were conducted for eGFR , CKD , UACR , and albuminuria . Quantile-quantile and Manhattan plots are displayed in Figure S1 and Figure S2 , respectively; lambdas ranged from 1 . 0 to 1 . 02 . We examined previously published loci in association with eGFR in participants of European Ancestry [3] , [4] , [16] in our African American GWAS ( Table 2 ) . In 23 of 24 SNPs , the directions of the beta coefficients for eGFR were identical ( p-value = 1 . 4*10−6 ) in CKDGen and CARe ( rs6420094 at SLC34A1 was the only exception ) , even though only 2 of the SNPs achieved nominal significance ( rs4293393 at the UMOD locus [p = 0 . 01] and rs4744712 at the PIP5K1B locus [p = 0 . 003] ) . We further interrogated the 250 kb flanking regions around each of these 24 SNP to identify the top SNP in CARe; statistical significance was determined based on a locus-specific Bonferroni correction ( see statistical methods for more detail ) . Of the 24 SNPs with the lowest p-value in each of these regions identified in CARe African Americans , we were able to replicate 12 loci ( UMOD , ANXA9 , GCKR , TFDP2 , DAB2 , VEGFA , ATXN2 , GATM , SLC22A2 , TMEM60 , SLC6A13 , and BCAS3 ) in independent samples of 4358 participants ( Table 3 , Figure S3 ) . We also interrogated the CARe CKD GWAS results at the chromosome 22 MYH9/APOL1 locus ( Figure S4 ) ; however , none of the G1 haplotype SNPs in APOL1 was present in our GWAS dataset . The lowest p-value was observed for rs739097 ( MAF 0 . 36 , p = 0 . 00179 ) , which was in weak LD with the previously described rs4821480 in MYH9 ( r2 0 . 03 , D′ 0 . 38 as determined by Hapmap Release 22 YRI phase 2 ) [10] . We performed discovery analyses using GWAS and the IBC chip array; p-value thresholds for discovery were p<5 . 0*10−8 for GWAS and p<2 . 0*10−6 for the IBC array [17] . We observed no genome-wide signals; instead , we carried forward 8 SNPs from GWAS that had a p-value<5*10−6 in Stage 1 for replication; results for these SNPs are presented in Table 4 . Replication was performed in 4 , 358 participants of African ancestry for eGFR and 2 , 110 for UACR . Characteristics of the replication samples are shown in Table 1 . Results from discovery , replication , and the combined GWAS analyses in African Americans are presented in Table 4 ( imputation scores can be found in Table S2 ) . Of the 8 SNPs from GWAS carried forward to replication , the combined Stage 1 + Stage 2 p-value was 5 . 3*10−7 for the association between UACR and rs4555246 in DOK6 ( Figure 1a ) and the combined p-value was 2 . 9*10−7 for association between UACR and rs2880072 in FNDC1 ( Figure 1b ) . Due to modest power of our replication set and the possibility that loci relevant for CKD may be similar across ethnic groups , we also attempted replication of the association between UACR and these two loci in the CKDGen consortium , a large consortium of participants of European ancestry ( n = 31 , 580 with UACR [18] ) . The beta coefficient for rs4555246 in DOK6 in the CKDGen data was direction-consistent , although the p-value for this SNP was non-significant ( p = 0 . 44 ) . Recognizing that regions , but not necessarily specific tagging SNPs , may replicate across ethnicities , we interrogated the 250 kb flanking region of rs4555246 ( n = 31 independent SNPs ) . The SNP with the lowest p-value was rs11151530 ( MAF 0 . 17 , p = 0 . 008 ) , which did not meet the Bonferroni-corrected threshold of p = 0 . 0016 ( 0 . 05/31 ) . For FNDC1 , the beta coefficient for the lead SNP was direction-consistent in CKDGen ( beta = −0 . 0196 ) , although the p-value was not significant ( p = 0 . 053 ) . Interrogating the 250 kb flanking region ( n = 7 independent SNPs ) revealed that rs7758822 had the lowest p-value at p = 0 . 002 ( MAF 0 . 12 ) , which met the Bonferroni-corrected threshold of p = 0 . 007 ( 0 . 05/7 ) . In European Americans , we confirmed several known loci for eGFR/CKD [3] , [4] , [16]; no novel genome-wide significant associations ( defined as p<2 . 0*10−6 ) were identified ( Table S3 ) . In African Americans , three loci for eGFR/CKD were brought forward to replication ( Table 4 ) . Of these , we observed nominal replication for rs7111394 in the KCNQ1 gene ( Figure 2a , Stage 1 + Stage 2 p-value = 3 . 6*10−6 ) . This SNP was not identified in the European American analysis as it was monomorphic in this population . Thus , we interrogated the 250 kb flanking region around this SNP ( 93 independent SNPs ) in the CARe IBC European ancestry participants . The SNP with the lowest p-value was rs81204 ( Figure 2b , MAF 0 . 16 , p-value = 0 . 00036 ) , which exceeded the corrected regional-specific threshold of 0 . 000538 ( 0 . 05/93 ) . To further understand the impact of the three new loci on kidney function and to bolster confidence in the sub-genome-wide statistical associations that we observed , we performed morpholino knockdown of kcnq1 , dok6 , and fndc1 in zebrafish embryos ( see methods ) . In situ hybridization for well-established renal markers was used to assess specific anatomic regions of the kidney during development . Kcnq1 knockdown caused abnormalities in glomerular gene expression in the majority of injected embryos , as shown by the global kidney marker pax2a at 48 hours post fertilization ( hpf ) ( see Table 5 , Figure 3F ) . Assessment of the podocyte markers wt1a at 24 hpf and nephrin at 48 hpf revealed similar , glomerular-specific effects . In contrast , the tubular markers slc20a1a and slc12a3 showed no significant changes . Analysis of glomerular architecture at 120 hpf by electron microscopy did not demonstrate significant differences between control and kcnq1 morphant embryos ( Figure S5 ) , possibly due to diminished morpholino efficacy at this later stage . Knockdown of dok6 and fndc1 did not result in generalized edema or significant developmental abnormalities of the kidney ( Table 5 , Figure S6 ) . To determine whether differences in gene expression resulted in altered kidney function , we evaluated glomerular filtration in kcnq1 morphant embryos by assessing the kidney's capacity to retain fluorescent dextran . Fluorescently labeled high-molecular weight dextran has been used in zebrafish to directly visualize functional glomerular integrity [19] . Control or kcnq1 morphant embryos were equally loaded with rhodamine-labeled 10 , 000 MW dextran by injection into the cardiac sinus venosus at 48 hpf ( Figure S7A , S7D ) . Dextran clearance was assessed by overall fluorescence in the embryo at 72 and 96 hpf ( Figure S7B , S7C , S7E , S7F ) and presence of red fluorescence in the green fluorescent tubules of cdh17:GFP reporter embryos . Kcnq1 morphant embryos exhibited decreased fluorescence by 96 hpf , indicative of increased dextran clearance compared to control embryos ( Figure 4C , 4F ) . Time course analysis confirmed equal loading and progressive loss of fluorescence over 48 hours in kcnq1 morphants . These results suggest that loss of kcnq1 causes decreased glomerular retention of macromolecules . The majority of embryos with increased loss of dextran fluorescence also exhibited generalized edema ( Figure 4A , 4B , 4D , 4E , 4G ) , which has been previously linked to kidney dysfunction in zebrafish [20] , [21] . rs4555246 in DOK6 was associated with albuminuria ( OR = 1 . 20 , p = 0 . 001 ) but not with eGFR ( p = 0 . 23 ) or CKD ( p = 0 . 92; Table S4 ) . Similarly , rs2880072 in FNDC1 was associated with albuminuria ( OR = 0 . 83 , p = 4 . 9*10−5 ) but not with eGFR or CKD . Although rs7111394 in KCNQ1 was associated with eGFR , it was not significantly associated with CKD ( direction consistent OR 1 . 07 , p = 0 . 36 ) , UACR , or albuminuria . These findings underscore the specificity of the genetic underpinnings of eGFR as compared to albuminuria . All replicating loci were stratified by hypertension and diabetes status ( Table S5 ) . We observed nominal significance for nearly all loci in the non-diabetes and non-hypertension strata , and many loci retained statistical significance in the diseased strata despite smaller sample sizes .
In more than 8 , 000 African Americans , for 24 known renal susceptibility loci identified in European ancestry consortia for eGFR and CKD , we have identified the most significant SNP in African Americans , of which several showed evidence of confirmation or replication . In addition , we performed discovery analyses using GWAS and a candidate-gene based array , and uncovered 3 suggestive loci , including KCNQ1 in association with eGFR and DOK6 and FNDC1 in association with UACR . Finally , we show that loss of function of kcnq1 leads to abnormalities in glomerular gene expression and function during zebrafish development . Despite having one of the largest GWAS datasets in African Americans , both our discovery and replication samples were of relatively modest size . Thus , we have focused primarily on interrogating regions previously identified in well-powered European ancestry meta-analyses for renal function in order to reduce our genome-wide penalty for multiple testing . While we have uncovered 3 suggestive loci , the strength of the statistical significance is only suggestive . Because of this , we have corroborated our findings with functional data from zebrafish , providing compelling evidence for the role of KCNQ1 in renal abnormalities . Several prior genome-wide studies using GWAS or admixture approaches have identified loci for renal function in participants of European [3] , [4] , [16] and African ancestry [10] , [11] . These studies have identified loci for eGFR , CKD , and non-diabetic ESRD; however , few novel discoveries and relatively limited replication have been made for renal function indices in African Americans . African ancestry populations have smaller LD blocks and hence more genetic diversity , thus offering the potential opportunity to fine map genomic regions and to identify novel genomic regions for indices of renal function . The genes located closest to our three suggestive loci are KCNQ1 in association with eGFR , as well as FNDC1 and DOK6 in association with UACR . KCNQ1 on chromosome 11 encodes for the potassium voltage-gated channel , KQT-like subfamily , member 1 ( KvLQT1 ) . The lead SNP identified among the African ancestry participants is located in an intron of KCNQ1 and is flanked by two recombination hotspots . The kcnq1 protein is abundantly expressed in the brush border membrane of renal proximal tubule cells , where it interacts with other K+ channels ( KCNE ) to mediate net K+ secretion . Loss of Kcnq1 in the mouse leads to impaired Na+ absorption with increased glucose load , however , developmental abnormalities have not been reported [22] . Variants in KCNQ1 have been associated with type 2 diabetes [23] and beta cell function [24] , due to its role as a potassium channel in the pancreatic beta cells . A recent study of individuals with diabetes from Japan identified variants in the KCNQ1 gene in association with diabetic nephropathy [25] . Our lead SNP is not in LD with this signal in either European or African ancestry individuals . Importantly , we observe a strong and significant association with KCNQ1 among participants without diabetes , and our functional work suggests an independent association of kcnq1 in renal development . Thus , it is unlikely that our observed associations are solely due to the association of KCNQ1 with diabetes . The DOK6 gene on chromosome 18 encodes a member of a family of intracellular adaptor proteins that have a role in the assembly of multimolecular signaling complexes . It is expressed in the human kidney as well as the ureteric buds of the murine developing kidney [26] . Although a direct role in renal disease has not been described so far , DOK6 interacts with Ret [26] , the ablation of which resulted in kidney agenesis in model systems [27] . The lead SNP we identified is located in an intron of a very circumscribed region of the gene . DOK6 has been recently found in a GWAS for osteoporosis [28] , although our lead SNP is not in LD with this variant . The fibronectin type III domain containing 1 ( FNDC1 ) gene on chromosome 6 is also expressed in the kidney; the lead SNP is located nearly 100 kb downstream of the gene . Little is known about the function of this gene to date; previous studies found that FNDC1 may mediate G protein signaling and have a role in hypoxia-induced apocytosis of cultured ventricular cardiomyocytes [29] , [30] . Zebrafish have been extensively used to study principal pathways of kidney development and function [31] , [32] . More recently , adult models of kidney injury have been developed [33] . For example , targeted knockdown of a prolyl 4-hydroxylase resulted in kidney dysfunction with edema and changes in podocytes and Bowman's capsule [34] . In addition , a zebrafish model of human nephrotic syndrome was generated by plce1 knockdown after positional cloning of this gene in affected siblings , similarly resulting in cardiac edema and functional abnormalities [35] . Here , we use morpholino knockdown to demonstrate that loss of kcnq1 leads to changes in global morphology and gene expression abnormalities during zebrafish kidney development . Furthermore , in vivo fluorescence-based functional analysis of zebrafish glomerular filtration capacity demonstrated decreased retention of macromolecules , as previously demonstrated for other genes affecting glomerular integrity [19] , [20] . Interestingly , KCNQ1 was identified in association with eGFR , and knockdown of kcnq1 in zebrafish predominantly causes glomerular gene expression and filtration defects . These results suggest that genes associated with polygenic chronic conditions can produce developmental phenotypes when knocked down in vivo . Our analytic approach was complemented by a large-scale candidate gene analysis using the IBC SNP Chip array [15] and replication of our findings in an additional 4358 African American individuals . By using participants from predominantly population-based cohorts , we were able to study disease initiation . An important focus of our multi-ethnic samples allowed us to explore allelic heterogeneity , and support the notion that genomic risk regions are observed across ethnicities . Our replication was derived from in silico samples , which allowed for the adjustment for principle components where necessary . Finally , we were able to corroborate our results in the zebrafish , a model organism for studying vertebrate kidney development . While the advantage of the zebrafish model is to rapidly assess gene function during development , the role of genes in aging and chronic disease cannot be modeled by transient morpholino knockdown . eGFR was estimated using the MDRD study equation , as obtaining gold-standard measures of GFR in large population-based samples is not feasible . Serum creatinine and albuminuria were measured at a single-point in time , which may misclassify certain individuals and bias our results toward the null . Even with imputation , the coverage of the genome in African Americans is not comprehensive , thus limiting our statistical power . With respect to the chromosome 22 MYH9/APOL1 locus , neither the E1 haplotype in MYH9 nor the G1 haplotype in APOL1 were in our GWAS dataset , limiting our ability to examine this region in association with more modest CKD phenotypes . Finally , our power was low for CKD to detect SNPs with odds ratios of 1 . 2 , the largest odds ratio detected in other GWAS for CKD [4] . We identified several SNPs in association with eGFR in African Ancestry individuals , as well as 3 suggestive loci for UACR and eGFR . Functional genomic studies support a role for kcnq1 in glomerular development and function in zebrafish .
Serum creatinine was measured as described in Text S1 . Serum creatinine was calibrated to NHANES in all studies ( including replication cohorts ) to account for between-laboratory variation as previously described [3] , [36] , [37] . Glomerular filtration rate was calculated based on serum creatinine ( eGFR ) with the Modification of Diet in Renal Disease ( MDRD ) equation [38] . We defined chronic kidney disease ( CKD ) as eGFR <60 ml/min/1 . 73 m2 in accordance with the National Kidney Foundation guidelines; CKD was based on a single serum creatinine measurement as described in Text S1 . Urinary albumin to creatinine ratio ( UACR , mg/g ) was computed as described in Text S1; microalbuminuria was defined as UACR >17 mg/g [men] and >25 mg/g [women] . We defined diabetes as fasting glucose ≥126 mg/dl , self-report , or pharmacologic treatment . Similarly , hypertension was defined as systolic blood pressure ≥140 mm Hg , diastolic blood pressure ≥90 mm Hg , or pharmacologic treatment . For the present study , the CARe consortium genotyped the IBC SNP chip [15] in 23767 European Americans and 8110 African Americans , as well as the Affymetrix 6 . 0 chip in 7382 African Americans . The IBC array contains nearly 50 , 000 SNPs across 2 , 000 loci . SNPs were selected using a tagging approach among populations represented in HapMap and the SeattleSNPs project . The array was designed to focus on candidate loci related to cardiovascular disease and its risk factors . More details can be found in the design paper [15] . Table S1 details the genotyping that was conducted . For the CARe study cohorts , quality control and imputation were conducted centrally using MACH 1 . 0 . 16 ( http://www . sph . umich . edu/csg/abecasis/MaCH/ ) . Imputation results were filtered using thresholds RSQ_HAT value of 0 . 3 and minor allele frequency 0 . 01 . Fractional counts between 0 and 2 were coded for the imputed genotypes in order to estimate the number of copies of a pre-specified allele . For European samples , the CEU population from HapMap 2 ( 2 . 54 million SNPs ) was used as the reference panel . For African American samples , a 1∶1 combined HapMap 2 CEU+YRI reference panel was used . This panel includes SNPs that were present in both populations , as well as SNPs segregating in one panel and monomorphic and nonmissing in the other ( 2 . 74 million altogether ) . Since the African American samples were genotyped for both the Affymetrix 6 . 0 and IBC arrays , we were able to analyze imputation performance at non-genotyped SNPs . The use of the CEU+YRI panel resulted in an allelic concordance rate of ∼95 . 6% , calculated as 1 – 1/2*|imputed_dosage – chip_dosage| for imputation on the Affymetrix chip . This is similar to rates obtained from African ancestry participants imputed using HapMap 2 YRI individuals [39] . Trait creation details are described above ( Renal Function Indices ) . Performed centrally but within each individual study , genome-wide association analyses and IBC chip analyses of natural log-transformed eGFR , UACR , CKD , and MA were conducted using linear and logistic regression with an additive genetic model . We adjusted for age , sex and study site ( when applicable ) and the first 10 principal components; relatedness was accounted for when necessary using linear mixed effect ( LME ) models for eGFR and UACR and logistic regression via generalized estimating equations ( GEE ) for CKD and MA . Additional details regarding the discovery cohorts are in Text S1 . Principal components were generated using EIGENSTRAT [40] within each study using the CARe African ancestry Affy6 . 0 genotype data . Two reference populations were included in the principal component analysis of African Americans: 1 , 178 European Americans from a multiple sclerosis GWA study ( from Dr . Phil de Jager and colleagues ) , and 756 Nigerians from the Yoruba region from a hypertension GWA ( provided by Dr . Richard Cooper and colleagues ) . Importantly , these two underwent extensively quality control procedures to remove population outliers using PCA . Ten principal components were generated for each study and used to adjust for population substructure . We performed fixed-effect meta-analyses of the IBC chip and genome-wide association data using the inverse-variance weighted approach in METAL ( http://www . sph . umich . edu/csg/abecasis/Metal/index . html ) . Genomic control correction was applied after calculating the inflation factor lambda ( λ ) within each individual study and after the genome-wide association meta-analysis was performed . The standard threshold of p<5×10−8 for genome-wide significance in the genome-wide association and p<2 . 0×10−6 in the IBC chip analyses was used . The rationale for the p-value threshold used for the IBC chip is based on an empiric test of the number of independent loci ( ∼25 , 000 ) that appear on the IBC array [17] . We selected independent SNPs ( pairwise r2<0 . 2 ) at each locus for replication . The R software ( v2 . 9 . 0 ) was used for data management , statistical analyses and graphing . We developed a set of criteria to validate the lead SNPs and interrogate regions around each of the loci that were previously reported among European ancestry ( EA ) participants [3] , [4] in our African American ( AA ) CARe samples . For each lead SNP in EA , we looked-up the respective association result with eGFR in AA . To accommodate the difference of LD structure and possible allelic heterogeneity across different ethnicities , we then interrogated the 250 kb flanking region around each lead SNP to determine whether there exist other SNPs with stronger associations with the outcome . We used the following criteria to identify the top AA SNP: 1 ) the SNP with the smallest association p-value within the region; 2 ) MAF >0 . 03; 3 ) location of the AA lead SNP within the same recombination block of the lead EA SNP , where the recombination block was defined as a 20% recombination rate . The statistical significance of each identified SNP was evaluated using a region-specific Bonferroni correction . We determined the number of independent SNPs based on the variance inflation factor ( VIF ) , which was calculated recursively within a sliding window with size 50 SNPs and pairwise r2 value of 0 . 2 using PLINK . Finally , each identified top SNP in AA was sent for replication in additional independent AA samples . Similar to the interrogation in AA for the EA lead loci , we also interrogated the newly identified loci from the CARe GWAS and the IBC chip in EA participants of the CKDGen consortium . Replication analyses were performed using imputed in silico genome-wide association data; replication studies conducted the same association analyses as the Stage 1 phase . Details regarding the replication cohorts can be found in Text S1 . Replication was performed as follows: meta-analysis was conducted in the Stage 2 studies only , and then in the Stage 1 + Stage 2 studies combined . Replication was defined as a direction-consistent Stage 2 beta coefficient; replication p-values are thus represented as one-sided tests . SNPs were declared to replicate when the p-value in the Stage 1 + Stage 2 studies combined was smaller than the p-value in the Stage 1 alone . The correlation between ln ( UACR ) and ln ( eGFR ) can range from non-significant to as high as 0 . 237 ( p<0 . 001 ) in our study; therefore , we examined the cross-trait associations across albuminuria and eGFR phenotypes . Zebrafish were maintained in accordance with established procedures and IACUC approval . At the one-cell stage , zebrafish embryos were injected with varying doses of morpholino antisense oligonucleotides ( MO , GeneTools , Philomath OR ) . MO sequences are shown in Table S6 . For in vivo observations , edema development was documented at 5-days post-fertilization . In situ hybridization was performed according to established procedures ( http://zfin . org/ZFIN/Methods/ThisseProtocol . html ) . To visualize different regions of the kidney , we used pax2a ( global kidney marker ) [41] , wt1a ( podocyte marker ) [41] , nephrin ( podocyte marker ) [31] , slc20a1a ( proximal tubule ) [42] , and slc12a3 ( distal tubule marker ) [42] . The morphology of the expression pattern was independently scored by two investigators . Electron microscopy was performed as previously described [41] . Dextran clearance was assessed as described previously [20] , 48 hours after morpholino injection , embryos were manually dechorionated , anesthetized in a 1∶20 dilution of 4 mg/ml Tricaine in egg water and positioned on their back in a 1% agarose injection mold . An equal volume of tetramethylrhodamine dextran ( 10 , 000 MW; Invitrogen ) was injected into the cardiac sinus venosus of each embryo , after which embryos were returned to fresh egg water . Embryos were imaged by fluorescence microscopy 6 hours post-injection ( 54 hpf ) to demonstrate equal loading , then subsequently imaged at 72 and 96 hpf to evaluate dextran clearance . Confocal images were obtained from agarose embedded embryos using a Zeiss LSM500 microscope .
|
Chronic kidney disease ( CKD ) is an increasing global public health problem and disproportionately affects populations of African ancestry . Many studies have shown that genetic variants are associated with the development of CKD; however , similar studies are lacking in African ancestry populations . The CARe consortium consists of more than 8 , 000 individuals of African ancestry; genome-wide association analysis for renal-related phenotypes was conducted . In cross-ethnicity analyses , we found that 23 of 24 previously identified SNPs in European ancestry populations have the same effect direction in our samples of African ancestry . We also identified 3 suggestive genetic variants associated with measurement of kidney function . We then tested these genes in zebrafish knockdown models and demonstrated that kcnq1 is involved in kidney development in zebrafish . These results highlight the similarity of genetic variants across ethnicities and show that cross-species modeling in zebrafish is feasible for genes associated with chronic human disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"chronic",
"kidney",
"disease",
"quantitative",
"traits",
"tubulointerstitial",
"disease",
"biology",
"heredity",
"genetic",
"association",
"studies",
"genetics",
"human",
"genetics",
"genetics",
"of",
"disease",
"nephrology",
"genetics",
"and",
"genomics",
"complex",
"traits"
] |
2011
|
Genetic Association for Renal Traits among Participants of African Ancestry Reveals New Loci for Renal Function
|
Non-autonomous cell-death is a cardinal feature of the disintegration of neural networks in neurodegenerative diseases , but the molecular bases of this process are poorly understood . The neural retina comprises a mosaic of rod and cone photoreceptors . Cone and rod photoreceptors degenerate upon rod-specific expression of heterogeneous mutations in functionally distinct genes , whereas cone-specific mutations are thought to cause only cone demise . Here we show that conditional ablation in cone photoreceptors of Ran-binding protein-2 ( Ranbp2 ) , a cell context-dependent pleiotropic protein linked to neuroprotection , familial necrotic encephalopathies , acute transverse myelitis and tumor-suppression , promotes early electrophysiological deficits , subcellular erosive destruction and non-apoptotic death of cones , whereas rod photoreceptors undergo cone-dependent non-autonomous apoptosis . Cone-specific Ranbp2 ablation causes the temporal activation of a cone-intrinsic molecular cascade highlighted by the early activation of metalloproteinase 11/stromelysin-3 and up-regulation of Crx and CoREST , followed by the down-modulation of cone-specific phototransduction genes , transient up-regulation of regulatory/survival genes and activation of caspase-7 without apoptosis . Conversely , PARP1+-apoptotic rods develop upon sequential activation of caspase-9 and caspase-3 and loss of membrane permeability . Rod photoreceptor demise ceases upon cone degeneration . These findings reveal novel roles of Ranbp2 in the modulation of intrinsic and extrinsic cell death mechanisms and pathways . They also unveil a novel spatiotemporal paradigm of progression of neurodegeneration upon cell-specific genetic damage whereby a cone to rod non-autonomous death pathway with intrinsically distinct cell-type death manifestations is triggered by cell-specific loss of Ranbp2 . Finally , this study casts new light onto cell-death mechanisms that may be shared by human dystrophies with distinct retinal spatial signatures as well as with other etiologically distinct neurodegenerative disorders .
The disintegration of neuronal networks owing to the non-autonomous death of neurons without primary damage is a hallmark manifestation of many neurodegenerative diseases and contributes determinately to their onset or progression [1]–[3] . Cone or rod photoreceptor neurons employ cell type-specific spectrally tuned and highly homologous phototransduction cascades . Neurodegenerative disorders affecting these neurons serve as excellent models to understand autonomous and non-autonomous cell death processes . First , early studies with chimeric mice with a mixture of healthy and unhealthy rod photoreceptors owing to the expression of rod-specific degenerative mutations by the latter showed that damaged rod photoreceptors promote the non-autonomous death of healthy rod photoreceptors [4]–[7] , but the analogous event does not appear to occur between neighboring healthy and damaged cone photoreceptors [8] . Second , rod photoreceptor-specific mutations causing the death of rod photoreceptors promote ultimately the non-autonomous death of cone photoreceptors [3] , [9]–[12] . This secondary loss of cone photoreceptors has the greatest impact on human vision , because cone photoreceptors mediate daylight and high acuity vision as well as color perception . For example , rod photoreceptor-specific mutations affecting phototransduction components of rod photoreceptors , such as the catalytic subunit of cGMP phosphodiesterase , rhodopsin and cyclic nucleotide-gated ( CNG ) channel subunits , lead to the degeneration of damaged rod and healthy cone photoreceptors [4] , [9] , [11]–[16] . By contrast , the cellular effects of cone photoreceptor-specific mutations causing cone degeneration are much less clear , but they are thought to spare the viability of rod photoreceptors . For example , cone-specific mutations impairing genes homologous to those of rod phototransduction cause the death of cone photoreceptors only [17]–[26] . A number of distinct models have been put forward to explain the secondary loss of healthy cone photoreceptors upon the primary degeneration of damaged rod photoreceptors . These include oxidative stress and metabolic imbalance caused by increased oxygen tension [27]–[29] , loss of paracrine neurotrophic [30] or vasculotrophic support [31] , microglia activation [32] , [33] and release of rod-derived toxic byproducts [34] . Distinguishing what mechanisms trigger extrinsic-elicited cell death pathways is further complicated by our limited knowledge of the intrinsic and primary cell-death pathways affecting rod photoreceptor themselves , and whether these are consistent across rod degeneration models [35]–[45] . Likewise , the knowledge about the molecular and subcellular events underlying the primary demise of cone photoreceptors is very limited [19] , [46] . Ascertaining the spatiotemporal processes causing autonomous and non-autonomous neural death is critical to our understanding of the pathogenesis of a variety of human photoreceptor dystrophies , such as retinitis pigmentosa ( RP ) and age-related macular degeneration ( AMD ) , that harbor hallmark spatiotemporal manifestations presumably driven by distinct intrinsic and extrinsic factors . The pleiotropic protein , Ran-binding protein-2 ( RanBP2 ) , is essential for organism viability and energy metabolism [47] , [48] . Prior studies on RanBP2 indicate that it plays critical cell-type-dependent physiological roles in mediating gene-environment interactions . In this regard , distinct disease stressors , such as phototoxicity [49] , [50] , Parkinsonian toxic insults [51] , and carcinogens [48] , trigger a variety of cell-context-dependent clinical and pathophysiological manifestations upon partial deficits or mutations of Ranbp2 . Further , semi-dominant mutations in human RANBP2 cause either acute necrotizing encephalopathies ( ANE1 ) or acute transverse myelitis ( ATM ) upon exposure to a variety of infectious agents [52]–[54] . In this study we set out to determine the intrinsic and extrinsic effects of lack of Ranbp2 function in the survival of cone or rod photoreceptor neurons upon selective ablation of Ranbp2 in cone photoreceptors , where Ranbp2 is highly expressed [55] . We show that cone-specific ablation of Ranbp2 promotes the autonomous non-apoptotic death of cone photoreceptors and the cone-dependent apoptotic demise of rod photoreceptors by distinct cell-type death mechanisms . Hence , a primary impairment of cone photoreceptors can promote the secondary death of healthy rod photoreceptors , a paradigm-shift observation with implications to our understanding of human neurodegenerative diseases affecting distinct photoreceptor cell types with hallmark regional distributions in the retina and other neural networks of the central nervous system .
To determine the physiological role of Ranbp2 in cone photoreceptors and uncover the effects of its genetic ablation in autonomous and non-autonomous molecular and cellular events affecting targeted cones and healthy rod photoreceptors , Ranbp2 was selectively targeted in mouse cones by Cre-mediated recombination of floxed Ranbp2 [48] , [56] , [57] ( Figure 1A ) . Hemizygous transgenic mice expressing Cre under control of the R/G cone opsin promoter ( Tg-HRGP-cre ) [56] , [57] were crossed with Ranbp2Flox/+ mice [48] to produce Tg-HRGP-cre:Ranbp2Flox/+ . These mice were then crossed to Ranbp2Flox/+ or Ranbp2Gt ( pGT0pfs ) 630Wcs/+ , which harbors a constitutively disrupted allele of Ranbp2 [47] ( Figure 1B ) , to generate the lines , Tg-HRGP-cre:Ranbp2Flox/Flox and Tg-HRGP-cre:Ranbp2Flox/Gt ( pGT0pfs ) 630Wc . These lines were morphologically and functionally indistinguishable from each other and they are hereafter designated as HRGP-cre:Ranbp2−/− . An out-of-frame Ranbp2 transcript comprising the fusion of exons 1 and 3 is produced at P7 upon Cre expression at P6 ( Figure 1C ) . Cre was specifically expressed in cell bodies of M- and S-cone photoreceptors ( Figure 1D , Figure S1 ) . We examined the temporal effects of loss of Ranbp2 expression in the morphology and survival of cones by comparing the immunostaining of retinal sections between HRGP-cre:Ranbp2−/− and HRGP-cre:Ranbp2+/− mice with anti-Cre and cone-specific anti-arrestin-4 ( Arr4 ) antibodies [58] at P9 , P13 , P20 and P27 of age ( Figure 2 ) . In both genotypes , the cell bodies of Cre-expressing cone photoreceptors migrated to the distal ( outer ) region of the outer nuclear layer ( ONL ) by P13 , where the majority of cone cell bodies are typically localized , and the outer segment ( OS ) and synaptic pedicles developed properly . However , few Cre+-cell bodies appeared displaced in the proximal ( inner ) ONL of HRGP-cre:Ranbp2−/− mice ( Figure 2D″ ) . By P20 , cones of HRGP-cre:Ranbp2−/− mice presented prominent swelling of the synaptic pedicles ( Figure 2F′ ) and retraction of some cell bodies to the proximal region of the ONL ( Figure 2F″ ) . By P27 , only a very few surviving cones were present in HRGP-cre:Ranbp2−/− ( Figure 2H′–2H″ ) ; by 6 weeks resilient cones lacking outer segments were rarely present ( Figure S2 ) , whereas no cones were present in 12-week old mice ( data not shown ) . The degeneration of cone photoreceptors was quantitatively monitored in retinal flat mounts . We compared the number of M-cones and OS length between HRGP-cre:Ranbp2−/− and HRGP-cre:Ranbp2+/− mice ( Figure 3 ) . No differences were seen in the number of M-cones or in the length of their OS at P15 . By P20 , both measures were significantly decreased in HRGP-cre:Ranbp2−/− mice , and very few M-cones remained at P27 ( Figure 3A–3C ) . Notably , the few surviving M-cones retained in P27 HRGP-cre:Ranbp2−/− mice still had OS which appeared of comparable length to those of HRGP-cre:Ranbp2+/− littermates ( Figure 3A , 3C ) . At P20 , the number of S-cones and the length of their OS were significantly decreased in HRGP-cre:Ranbp2−/− mice and prominent clumps of S-opsin were observed in the OS ( Figure S3 ) . We then examined the position of the Cre+-cell bodies within the proximal and distal ONL at peripheral and central regions of dorsal retinas ( Figure 3D , 3E ) . Unlike HRGP-cre:Ranbp2+/− littermates , P13 HRGP-cre:Ranbp2−/− mice presented displaced Cre+-cell bodies in the peripheral and central regions of the proximal ONL . At P20 , the number of Cre+-cell bodies had strongly decreased across the central and peripheral regions of the HRGP-cre:Ranbp2−/− retina , but this decrease was much more pronounced in the central retina ( Figure 3E ) . By P27 and 3-months of age , respectively , very few and no Cre+ cells were observed in any regions of the retina ( Figure 3E , data not shown ) . OS shortening is thought to precede cell death of rod and cone photoreceptors [13] , [28] . Hence , we employed multiple cell death markers to dissect out the molecular and subcellular processes underpinning the activation and progression of cell death between cone and rod photoreceptors of HRGP-cre:Ranbp2−/− mice . We used TUNEL staining to determine whether degenerating cone photoreceptors underwent apoptosis . As shown in Figures 4A and 4B , we did not identify any cell bodies that were TUNEL+ and Cre+ . Instead , we found that all TUNEL+-cell bodies were Cre− , most likely rod photoreceptors , because these neurons comprise 97% of all photoreceptor cell types [59] . Morphometric analyses showed a drastic increase of TUNEL+-cell bodies at P20 ( Figure 4B ) . Akin to the localization of Cre+ cells , this increase was significantly more pronounced also in central and peripheral regions of the distal ONL than in the counterpart proximal regions ( Figure 4B ) . To further differentiate TUNEL+-apoptotic cell bodies from necrotic cells among cone and rod photoreceptors , we carried out morphometric analysis of co-localization of TUNEL+ and cone Arr4+-cell bodies of retinal explants incubated with the membrane-impermeable and DNA-binding fluorescent dye , ethidium homodimer III ( EthD-III ) . There were no Arr4+Tunel+ or Arr4+EthD-III+ cells in either genotype ( Figure 4C ) . Instead , we found that HRGP-cre:Ranbp2−/− at P20 had about 70 and 55% of EthD-III+ Arr4− and TUNEL+Arr4− cells , respectively , with 30% of these cells being EthD-III+TUNEL+Arr4− ( Figure 4C , 4D ) . The presence of any of these apoptotic or necrotic cell bodies were negligible in HRGP-cre:Ranp2+/− ( Figure 4C , 4D ) . These data support that the TUNEL+Arr4− , EthD-III+ Arr4− and EthD-III+TUNEL+Arr4− cells represent different cell death stages of rod photoreceptors . To establish unequivocally that TUNEL+Cre−-apoptotic cell bodies are rod photoreceptors neurons , retinal sections were co-immunostained for Nr2E3 , a transcription factor specifically expressed in cell bodies of rod photoreceptors [60] . As shown in Figure 5A ( a–b″″ ) , all Cre+-cell bodies of either genotype were Nr2E3− , whereas many TUNEL+-cell bodies were Nr2E3+ in HRGP-cre:Ranbp2−/− . We also identified a small subpopulation of TUNEL+-cell bodies with Nr2E3 aggregation at discrete perinuclear foci ( Figure 5A , c–d″″ ) , suggesting that Nr2E3 localization changes and its expression decreases during rod photoreceptor death . Quantitative morphometric analysis of triple-immunostained retinas showed that about 3 . 2% of the total cells in HRGP-cre:Ranbp2−/− were TUNEL+ , while 52% and 44% of these TUNEL+ cell bodies were Cre−NR2E3+ or Cre−Nr2E3− , respectively; the remaining 4% were dying rods with discrete perinuclear aggregation of Nr2E3 ( Figure 5B ) . Further examination of the TUNEL+Nr2E3− and TUNEL+Nr2E3+ cells indicated that they were TUNEL+EthD-III+ or TUNEL+EthD-III− ( Figure S4 ) . These data support that TUNEL+Nr2E3+ and TUNEL+Nr2E3− cells also represent different stages of cell death of rod photoreceptors . We next examined whether cone and rod photoreceptor degenerations were accompanied by the activation of caspases , a cardinal feature of cell death [61] , [62] . We screened whole retinal extracts of HRGP-cre:Ranbp2−/− mice at P20 with substrates against specific caspases and found strong activation of caspases 3/7 , mild activation of caspases 8 and 9 , and no activation of caspases 1 , 2 and 6 , when compared to HRGP-cre:Ranbp2+/− mice ( Figure S5 ) . To define the spatiotemporal profile of caspase activation , we examined the activities of caspases 3 and 7 in retinal extracts and carried out morphometric analyses of retinal sections immunostained with antibodies against cleaved ( activated ) caspase 3 , 7 or 9 of age-matched mice of both genotypes . In HRGP-cre:Ranbp2−/− mice , we found that the activities of caspase 3 , caspase 7 , or both , peaked at P13 ( Figure 6A ) , well before the rise in TUNEL+-cell bodies in the ONL ( Figure 4B ) , and that these activities were negligible by P27 , when most cones had died ( Figure 6A ) . To distinguish caspase 3 and 7 activities , we immunostained retinal sections of different ages for active caspase 3 and 7 and performed morphometric analysis ( Figure 6B , 6C ) . These experiments showed that the majority of active caspase 3+-photoreceptor cell bodies were Cre−TUNEL− ( Figure 6B ) , but there was also a small but significant fraction of active caspase-3+Cre+-cell bodies at P13 in HRGP-cre:Ranbp2−/− mice ( Figure 6B , 6C ) . All types of active caspase 3+-photoreceptor cell bodies were drastically decreased in HRGP-cre:Ranbp2−/− mice and their presence were no different from HRGP-cre:Ranbp2+/− by P20 ( Figure 6B , 6C a–a′″ , d–d′″ ) . By contrast , caspase 7+-cell bodies became prominent only at P20 and most were Cre+ ( Figure 6C , b′–b′″ , e′–e′″ ) . Further , the temporal profiles of caspase 3 and 7 activities paralleled their transcriptional up-regulation ( Figure S6 ) . Collectively , the data support that activation of caspase 3 in rods and caspase 7 in cones contribute to the rise of activities of caspases 3 and 7 at P13 and P20 , respectively . Next , we examined the activation of Parp1 ( Parp1+; Figure 6C , Figure S7 ) , which is cleaved during apoptosis into 89 and 24 kDa fragments by caspase 3 or 7 [63] , [64] and has been linked to rod photoreceptor degeneration [65] . We found that Parp1+cells were never Cre+ and that their appearance was biphasic with activity peaks at P9 and P20 ( Figure 6C , Figure S7A , a–b″″ ) . These temporal peaks of Parp1+-activities in rod photoreceptors coincided with the activation of caspase 9 in rods at P9 ( Figure S7A , c–d″″ ) and caspase 7 in cones at P20 ( Figure 6C ) . No Parp1+-cell bodies could be identified at P13 ( Figure 6C ) . At P9 , caspase 9+TUNEL+ and Parp1+TUNEL+ cell bodies were always cre− rods , while TUNEL+caspase 9−cre− and TUNEL+Parp1−cre− rods could also be observed ( Figure S7A ) . These events indicate that caspase 9 and Parp1 activation in rod photoreceptors is short-lived . These manifestations were accompanied by transcriptional up-regulation of caspase 9 and Parp1 between P9 and P20 , but not of caspase 8 , which is typically activated by extrinsic cell-death mechanisms ( Figure S7B ) [61] , [62] . The formation of rod apoptotic cell bodies prompted us to examine whether apoptosis-inducing factor ( AIF ) or cytochrome c was released from the mitochondria , as these are also hallmark events of the apoptotic cascade [66]–[68] . Subcellular fractionation of retinas showed no sign of AIF or cytochrome c in the cytosol fraction of either genotype at the peak of apoptosis ( Figure S8A ) . Changes in AIF levels were also not observed in the nuclear-enriched fraction of either genotype ( Figure S8B ) . In addition , the expression levels of the apoptotic protease activating factor-1 ( Apaf-1 ) remained unchanged ( data not shown ) . Finally , we examined markers for necroptosis , such as members of the receptor-interacting proteins , RIP1 and RIP3 , and macroautophagy , such as the autophagosomal membrane marker , light chain 3B II ( LC3B II ) , since they are thought to be induced upon photoreceptor degeneration [69]–[71] . Immunoassays of retinal extracts found no differences in these markers between HRGP-cre:Ranbp+/− and HRGP-cre:Ranbp2−/− mice at P13 ( Figure S8C , S8D ) . Histological examination of semi-thin retinal sections showed that retinas of HRGP-cre:Ranbp2−/− developed prominent interstitial spaces between photoreceptors cell bodies and across the ONL by P20 , a phenotype not present in HRGP-cre:Ranbp2+/− mice ( Figure 7A ) . The interstitial spaces could always be traced to prominent euchromatic nuclei typically localized at the distal ( outer ) edge of the ONL ( Figure 7A ) , hallmark morphological and topographic features of cell bodies of cone photoreceptors . Detailed examination of the ultrastructure of the ONL showed that the interstitial spaces reflect degenerating lower fibers of cone photoreceptors that were dilated and very lucent ( Figure 7B ) . This striking phenotype led us to hypothesize that ablation of Ranbp2 in cones promotes the activation of metalloproteinase ( s ) ( MMPs ) causing the weakening and degradation of the extracellular matrix , which normally organizes photoreceptor cell bodies within the ONL [72] , and may contribute to the retraction of Cre+-cell bodies from the distal ( outer ) to the proximal ( inner ) region of the ONL ( Figure 2 ) . Hence , we screened retinal extracts for each the eleven MMP activities . In comparison to HRGP-cre:Ranbp2+/− , retinal extracts of HRGP-cre:Ranbp2−/− mice presented ∼3-fold higher activity of MMP11 , but not of any other MMPs ( Figure S9A ) . MMP11 activity was elevated at P13 and P20 and returned to control levels at P27 when most cones have degenerated ( Figure 7C ) . The increase of MMP11 activity was also accompanied by a ∼3-fold increase of the active form of MMP11 ( Figure 7D , 7E ) . The transcriptional up-regulation of Mmp11 in HRGP-cre:Ranbp2−/− mice as early as P9 followed its transient down-regulation at P7 and preceded the activation of MMP11 at P13 , whereas the transcriptional levels of Timp3 remained largely unchanged across different ages ( Figure S9B , 7C ) . To assess the cellular origin of MMP11 expression and activity , we performed immunohistochemistry of MMP-11 in retinal sections from P20 mice , and found that MMP-11 was localized prominently around cell bodies , inner segments and lower fibers of cone Arr4+-cells ( Figure 7F ) . Collectively , these data confirm that ablation of Ranbp2 in cones promotes the up-regulation of MMP11 expression and activity in these neurons , an event which likely contributes to the development of interstitial spaces in the ONL and retraction of cones cell bodies to the proximal ONL in HRGP-cre:Ranbp2−/− mice . To ascertain in greater detail the morphological changes of degenerating cone photoreceptors , we examined the ultrastructure of HRGP-cre:Ranbp2+/− and HRGP-cre:Ranbp2−/− retinal sections at P20 . The subcellular subcompartments of cone and rod photoreceptors of HRGP-cre:Ranbp2+/− mice had normal morphologies ( Figure 8A , 8A′ , 8F , 8F′ , 8I ) , whereas cone photoreceptors of HRGP-cre:Ranbp2−/− mice exhibited distinct morphological changes across multiple subcellular compartments . Across different photoreceptor cells of HRGP-cre:Ranbp2−/− , we found that cone OS membranes were extended , disorganized and collapsed into large , amorphous and lucent areas ( Figure 8B , 8B′ , 8C , 8C′ ) , that electrodense material accumulated at the connecting cilium ( Figure 8D , 8D′ ) , and that prominent electron lucent areas were present in the inner segments ( Figure 8E , 8E′ ) . Additional abnormalities were seen at the synaptic pedicles including accumulation of multilamellar bodies ( Figure 8G , 8G′ ) , mitochondria with widespread electron lucent matrix areas without cristae and formation of cytosolic lucent areas without a limiting membrane around subcellular debris ( Figure 8H , 8H′ ) . Albeit less extensive , cristae of some mitochondria in the rod synaptic spherules were also disrupted resulting in the formation of lucent areas within the matrix ( Figure 8J ) . We examined the effect of Ranbp2 ablation selectively in cones on the expression of cone photoreceptor-specific and other pertinent genes by quantitative real time-PCR ( qRT-PCR ) ( Figure 9 , Table S1 ) . We found that Ranbp2 ablation led to rapid declines of M-opsin ( Opn1mw ) and S-opsin ( Opn1sw ) mRNAs as early as P13 ( Figure 9A ) , when there are yet no prominent cellular changes in M-opsin and Cre+-neurons between genotypes ( Figure 3B , 3E ) , and such declines occurred without concomitant changes in rhodopsin ( Rho ) mRNA ( Figure 9A ) . Other cone-specific genes showed a similar pattern of down-regulation , including Pde6h , Pde6c and Gnat3 ( Figure 9B ) . By P27 , when cones have degenerated , the expression of all cone-specific genes was at or below the detection limit in HRGP-cre:Ranbp2−/− retinas ( Figure 9A , 9B ) . By contrast , the expression of the pan-photoreceptor markers , Rcvn and Osgep , increased transiently at P13 ( Figure 9C ) . We also examined the genetic variants of miR-124a ( miR-124a1/Rncr3 , miR-124a2 , miR-124a3 ) , which mediate cone survival , and its downstream target transcript , Lhx2 , whose translation is suppressed by miR-124a [73] . We found strong up-regulation of miR-124a in the HRGP-cre:Ranbp2−/− retina , which was accompanied by increased levels of Lhx2 , albeit of a lesser magnitude ( Figure 9D ) . The time course of these changes was similar , beginning at P13 , peaking at P20 and returning to HRGP-cre:Ranbp2+/− levels at P27 when cones have degenerated ( Figure 9D ) . We also assessed the expression of Trß2 , Otx2 and Crx , encoding transcription factors critical to the maturation and maintenance of cone photoreceptors ( Figure 9E ) [74]–[78] . Crx expression was decreased at P7 , when Cre-mediated excision of Ranbp2 occurs ( Figure 1C ) , and then rose steadily up to P13 , and then decreased to HRGP-cre:Ranbp2+/− levels at P27 . In comparison , expression of Trß2 and Otx2 increased later , with a sharp peak at P20 , and then declined to basal levels at P27 ( Figure 9E ) . Like Crx , CoREST ( also known as Rcor1 ) , a cofactor of REST ( repressor element 1 silencing transcription factor ) , was transiently down-regulated at P7 and then rose at P9 until P20 , when there was a transient up-regulation of neuropilin-1 ( nrp1 ) , a receptor whose transcriptional expression is modulated by miR-124 and CoREST ( Figure 9E ) [79] . The expressions of CoREST and nrp1 also returned to basal levels at P27 ( Figure 9E ) . Because Ranbp2 ablation induced the expression and activation of MMP11 , a metalloproteinase which is known to cleave selectively the α3 chain of collagen VI ( Col6α3 ) [80] , we examined the time course changes of Col6α3 expression in the HRGP-cre:Ranbp2−/− retina . A selective rise in Col6α3 expression , but not α1 chain of collagen I ( Col1α1 ) , was detected as early as P9 , shortly after a transcriptional increase in Mmp11 was also detected ( Figure 9F , S9B ) . Cola6α3 expression peaked at P20 and then became indistinguishable from HRGP-cre:Ranbp2+/− mice at P27 . We did not observe any transcriptional changes in let-7c , a miRNA reported to promote the down-regulation of Mmp11 ( Figure 9F ) [81] . Finally , we examined transcriptional changes in markers associated with neurodegenerative mechanisms including with autophagy ( e . g . cathepsin S , lysozyme and culsterin ) , glycolysis ( 6-Pfk ) , hypoxia ( Hif1α ) and inflammation ( Gfap ) . Among these , we found changes only in Hif1α and Gfap with Hif1α rising from P9 until P20 and Gfap transiently spiking at P20 when most cones are undergoing degeneration ( Figure 9G ) . Ranbp2 has multifaceted roles in biology and pathology across tissues , cell types and cell-stages . This complexity reflects the interaction of the diverse structural modules of Ranbp2 with multifunctional partners . Hence , we employed the Ingenuity pathway analysis ( IPA ) to identify and delineate connectivity maps linking Ranbp2 with the regulation of expression of genes/proteins identified by this study and genetic , protein and metabolic networks . The top network hit generated by IPA ( score 22 ) was associated to “Cellular Development , Nervous System Development and Function and Carbohydrate Metabolism” ( Figure 10 ) . This network comprised thirty-two gene products and three endogenous chemicals , D-glucose , sn-glycero-3-phosphocholine and tretinoin ( all-trans retinoic acid ) ( Figure 10 ) . Remarkably , this connectivity map revealed links between transcription factors , many of which belong to the orphan nuclear receptor family and are known to be modulated by Ranbp2 levels , and metabolites , whose levels are also affected by Ranbp2 and regulate transcriptional activities of nuclear factors [47] , [49] , [51] , [82] , [83] . Among other novel points of interest , the IPA revealed central roles of i ) huntingtin cross-talk with nuclear factors modulated by Ranbp2 , a mechanism which is thought to be disrupted in Huntington's disease ( HD ) [84] , and ii ) two secreted and extracellular signaling proteins , wingless-type MMTV integration site family , member 2 ( WNT2B ) and brain-derived neurotrophic factor ( BDNF ) , intersecting multiple nodes modulated by Ranbp2 and other factors . WNT2B and BDNF are known to play important roles in regulation of cell growth , differentiation , tumorigenesis , and to support the survival of existing neurons , respectively [85]–[87] . Genetic excision of Ranbp2 is detectable by P7 , which coincides with immediate transcriptional changes in the levels of Mmp11 , Crx and CoREST ( Figure 9E , S9B ) . Reduced levels of cone-specific transcripts , such as those encoding phototransduction proteins , were not detectable until P13 and changes in cone morphology were not observed until P20 . These observations , and the non-autonomous molecular and cellular effects of cones on rod photoreceptors , prompted us to determine the onset and progression of cone and rod physiological dysfunction caused by ablation of Ranbp2 in cones . We measured cone and rod function by light- and dark-adapted electroretinograms ( ERGs ) , respectively . Figure 11 summarizes the ERG data obtained from mice between P13 and 150 days . At P13 , dark-adapted ERGs of HRGP-cre:Ranbp+/− and HRGP-cre:Ranbp2−/− mice were comparable ( Figure 11A , left; Figure 11B ) , indicating equivalent retention of rod-mediated outer retinal activity at this age . In comparison , light-adapted ERGs , reflecting cone activity [88] , [89] were already reduced significantly at P13 ( Figure 11A , right; Figure 11C ) . By P22 , the amplitude of the light-adapted ERGs were markedly reduced in HRGP-cre:Ranbp2−/− mice ( Figures 11D ) , and cone ERGs were extinguished in HRGP-cre:Ranbp2−/− mice at P29 ( Figure 11A , 11E ) . The amplitude of the dark-adapted ERG a-wave , which reflects phototransduction in the outer segments of rod photoreceptors , was not significantly different between HRGP-cre:Ranbp2+/− and HRGP-cre:Ranbp2−/− mice at any age examined ( Figure 11A , 11B , 11F , 11G ) . In mice aged P29 and P150 , the dark-adapted ERG b-wave was reduced , but in a luminance-dependent fashion ( Figure 11F , 11G ) . At low stimulus luminances , where the b-wave reflects synaptic transmission from rod photoreceptors to rod bipolar cells , there was no significant reduction in amplitude . At higher stimulus luminances , where both rod- and cone-mediated synaptic activity contribute to the b-wave , a significant amplitude reduction was observed ( Figure 11F ) . The synaptic localization of post-synaptic density 95 ( PSD95 ) protein was comparable between genotypes at P20 ( Figure S10 ) , indicating that the b-wave reductions noted at high flash luminances reflects a loss of the cone pathway contribution to the ERG b-wave at these stimulus levels , and not to an alteration in synaptic transmission between rod photoreceptors and bipolar cells . Figure 11G plots the amplitude of HRGP-cre:Ranbp2−/− light- and dark-adapted ERG components relative to those of control littermates ( HRGP-cre:Ranbp2+/− ) ranging from P13 to 21-weeks in age . HRGP-cre:Ranbp2−/− mice have a selective reduction in the light-adapted ERG indicating that the dark-adapted ERG is not sensitive to the limited cone-induced rod loss seen by other cell biological measures previously described in this study . In agreement with the electrophysiological observations , the complete loss of cone photoreceptors and the limited cone-induced loss of rod photoreceptors did not cause significant changes in ONL thickness and cell body density when analyzed by light microscopy in mice as old as 12-weeks of age ( Figure S11 ) .
The results of this study demonstrate: i ) cone-specific ablation of Ranbp2 triggers the demise of cone photoreceptors with concomitant non-autonomous cell death of rod photoreceptors , ii ) the death of rod photoreceptors is contingent on the presence of cone photoreceptors , and iii ) the cell death mechanisms of cone and rod photoreceptors are intrinsically distinct with the former and latter undergoing atypical features of necrosis and apoptosis , respectively [61] , [62] . We found that cone photoreceptors undergo necrotic changes including massive erosive destruction of their intracellular contents and late caspase 7 activation , but without apparent changes in loss of membrane permeability . In comparison , rods undergo apoptosis , caspase 3 and Parp1 activations , and loss of membrane permeability . This study unveils early molecular events triggered by and concomitant with the ablation of Ranbp2 in cone photoreceptors , including changes in Mmp11 , Crx and CoREST expressions at P7 and of MMP11 activity at P13 , followed by the activation of caspase 7 and a compensatory burst in the expression of the MMP11 substrate , Cola6α3 , at P20 , when the degeneration of cones is well underway . These events are accompanied by the up-regulation of cone survival genes , such as genetic variants of miR-124a and its downstream target transcript , Lhx2 , at P13 , when the down-regulation of cone-specific genes , such as M- and S-opsins , becomes significant . miR-124 promotes the translational suppression of Lhx2 , a process which has been linked to the suppression of apoptosis during cone photoreceptor differentiation and aberrant sprouting of hippocampal neurons [73] . In contrast to these observations , up-regulation of miR-124a does not prevent the demise of mature cone photoreceptors and occurs concomitantly with a rise of Lhx2 levels . Further , changes in CoREST levels , a target of miR-124 [79] , preceded those of all miR-124 variants . These observations suggest that the miR-124 variants may act on other substrates , such as Foxa2 [90] , and that may modulate the survival of cones ( or rods ) independently of Lhx2 and CoREST suppression [79] . Finally , even though cones account for only ∼3% of all photoreceptor types of the mouse retina [59] , the return of cone-derived caspase 7 and MMP11 activities and rod-derived caspase 3 activity to control levels after cones have degenerated strongly support these events are triggered solely by the Ranbp2-dependent dysfunction of cones . The early and selective activation of MMP11 followed by the development of prominent interstitial spaces between photoreceptor cell bodies , swelling of the lower fibers of cones , and the loss of membrane permeability of rods supports that MMP11 plays an important role in the development of autonomous and non-autonomous photoreceptor degeneration . These roles may include autocrine , paracrine and even intracrine actions , whereby intracrine and paracrine functions contribute to intracellular erosion of cones and non-autonomous death of rods , respectively . In comparison to other MMP proenzymes , MMP11 shows important functional differences . While most MMPs are activated extracellularly , MMP11 is secreted in the active form [91] . This exposes intracellular substrates to active MMP11 , a potential intracrine signal , and its extracellular proteolytic functions may stimulate autocrine and/or paracrine signaling . Regardless of its molecular mechanisms of action ( s ) , increased levels of MMP11 expression is reported to modulate cell survival [92]–[96] , to act as a negative prognostic of cancer patient survival [97]–[99] , and to promote oncogenic homing , tumorigenesis and metastasis [92] , [96] , cardinal manifestations triggered also by haploinsufficiency and hypomorphism of Ranbp2 [48] . MMP11 up-regulation has been reported to promote the suppression of apoptotic and necrotic cell death , rather than stimulation of cell proliferation [95] , [96] . These cellular manifestations appear at odds with the physiological phenotypes of our study . It is possible that the distinct pathological outcomes produced by the induction of MMP11 upon loss of Ranbp2 reflect intrinsically distinct tissue/cell-type-dependent signaling cues produced by MMP11 substrates . Hence , etiological distinct disorders , such as cancer and neurodegeneration , may share pathomechanisms with distinct clinical outcomes . The identification of pathophysiological substrates of MMP11 will be critical to uncover the scope of its pathobiological roles and aid toward the design of tissue-selective MMP11 inhibitors . It will be interesting to explore whether paracrine factors and players within the network uncovered by the IPA presented in this work , such as Wnt2b and BDNF , their receptors or nuclear transcriptional factors , are substrates of MMP11 or exert regulatory effects on MMP11 expression or activity . Although a specific pharmacological inhibitor of MMP11 is not available [100] , [101] , the genetic interactions between Mmp11 and Ranbp2 may be revealed by assessing the effect of genetic ablation of Mmp11 on the pathophysiological manifestations observed in HRGP-cre:Ranbp2−/− ( e . g . loss of membrane permeability of rods and rod apoptosis ) and to define a potential role for MMP11 in neuroprotection against autonomous and non-autonomous cell death mechanisms affecting cone or rod photoreceptor survival . Another critical outcome of this work is the finding of non-autonomous apoptotic death of rod ( Nr2E3+ ) photoreceptors upon cone dysfunction and death . TUNEL+ cones were never identified at any age , while dying rods comprised a mixed population of TUNEL+ , EthD-III+ or both . Thus , rods undergo a programmed cell death that is triggered by the demise of cones . Important , rod cell death ceases at P27 , when cone degeneration is complete based on the lack of apoptotic bodies , the absence of Parp1+ and caspase3/7+ cells , and the preservation of the ONL and rod photoreceptor function at P27 and later ages . Our results also indicate the presence of atypical mechanisms of cell death . While classical cell death features were identified during cone or rod degeneration , the following observations do not match previously recognized canonical paradigms . First , caspase 7 activation , typically observed in apoptosis , was found in dying cones lacking cleaved Parp1 , a substrate of caspase 7 . Instead , cone death was characterized by the rampant disintegration of critical subcellular structures , such as outer segments and mitochondria , swelling of lower fibers and cone pedicles , features that are consistent with necrosis . However , EthD-III+ cones were not identified as would be expected from the typical loss of membrane permeability caused by necrosis . Second , although rod photoreceptor death is consistent with apoptosis as supported by the presence of caspase 3+ , Parp1+ and TUNEL+-cell bodies , classic necrotic features were also present , including a loss of plasma membrane permeability ( EthD-III+ cell bodies ) and limited formation of electron lucent areas in the matrix of mitochondria at synaptic spherules . These observations indicate that activation of caspase 3 and caspase 7 does not determine the mode of rod or cone cell death , respectively . This conclusion is supported by comparable rates of rod death in rd1 mice on a wild-type or caspase3−/− background [43] . Third , we did not observe a release of cytochrome C and AIF to the cytoplasm , as typified in apoptosis [66] and observed in other photoreceptor degeneration models [35] , [45] , an induction of RIP1 , RIP3 , or caspase 8 activation as observed in necroptosis [61] , [69] and lipidation of microtubule-associated protein 1 light chain 3 ( LC3/Atg8 ) that is typical of autophagy [70] , [71] . Altogether these data support the existence of complex , shared , unique and thus atypical cell death mechanisms between rod and cone photoreceptors . These manifestations are likely determined by the cell-type dependent pleiotropic molecular and metabolic activities of Ranbp2 [47]–[54] . Emerging mouse models of Ranbp2 harboring losses in selective domains and functional activities of Ranbp2 will aid in parsing the contribution of such activities to cellular functions and intrinsic or extrinsic cell death modalities unique or shared by several diseases . Finally , our data show that the non-autonomous death of rods upon loss of Ranbp2 in cones is not promoted by the absence of cones or loss of rod-cone cell contacts . Instead , a likely mechanism is the release of a diffusible factor by cone photoreceptors upon their dysfunction that is deleterious to rod photoreceptors . MMP11 activation is an excellent candidate to play a critical role in such paracrine and death signaling . Our data predict that the topographic density of cone and rod photoreceptors in the retina will play a determinant role in the level of rod photoreceptor degeneration . This issue is of crucial significance to human retinal dystrophies , because the unique and heterogeneous topographic distributions of photoreceptor types across the human retina often mirror retinal pathologies with regional tissue and clinical hallmarks , such as RP , cone-rod dystrophies and age-related macular degeneration ( AMD ) . It is likely that loss of biological activities regulated by RANBP2 will exert prominent pathological outcomes in regions of the retina , such as the macula , where the topographic density of cone and rod photoreceptors is similar .
HRGP-Cre mice ( kindly provided by Yun-Zheng Le , University of Oklahoma Health Sciences Center ) [56] , [57] were crossed to Ranbp2+/flox mice ( kindly provided by Jan M . van Deursen , Mayo Clinic College of Medicine ) [48] to produce HRGP-cre:Ranbp2+/flox . HRGP-cre:Ranbp2+/flox were then crossed to Ranbp2+/flox or Ranbp2Gt ( pGT0pfs ) 630Wcs/+ [47] to generate HRGP-cre:Ranbp2+/− ( +/− ) and HRGP-cre:Ranbp2−/− ( −/− ) . Mice were in the following mixed genetic background: 129 SvJ , C57BL/6J , FVB/N , 129olaHsd . Mice were screened also for rd1 and rd8 alleles . Mice were raised in a pathogen-free transgenic barrier facility at <70 lux and given ad libitum access to water and chow diet 5LJ5 ( Purina , Saint Louis , MO ) . Animal protocols were approved by the Institutional Animal Care and Use Committees at Duke University and Cleveland Clinic , and all procedures adhered to the ARVO guidelines for the Use of Animals in Vision Research . Primary antibodies used this study were: mouse anti-Cre ( Covance , Princeton , NJ ) , rabbit anti-Cre ( Novagen , Gibbstown , NJ ) , rabbit anti-Arr4 ( Millipore/Pel-Freez Biologicals , Billerica , MA ) , rabbit anti-L/M opsin ( #Ab 21069 ) [55] , rabbit monoclonal anti-cleaved caspase3 Asp175 ( Cell Signaling , Danvers , MA ) , rabbit polyclonal anti-cleaved-caspase 7 Asp353 ( Cell Signaling ) , rabbit anti-cleaved caspase 8 Asp391 ( Cell Signaling ) , rabbit anti-cleaved caspase 9 Asp 330 ( Cell Signaling ) , rabbit anti-cleaved Parp1 Asp214 ( Cell Signaling ) , mouse anti-full-length Parp1 ( BD Bioscience , San Jose , CA ) , mouse anti-LC3 B ( Cell Signaling , Danvers , MA ) , mouse anti-cytochrome C ( BD Bioscience , San Jose , CA ) , mouse anti-mitochondrial heat shock protein 70 ( Affinity Bioreagent , Golden , CO ) , rabbit-cytosolic heat shock protein 70 ( Assay Design , Farmingdale , NY ) , rabbit anti-GFAP ( DAKO , Carpinteria , CA ) , mouse anti-glutamine synthase ( Sigma Aldrich , Saint Louis , MO ) , mouse anti-RIP1 ( BD Bioscience , San Jose , CA ) , rabbit anti-RIP3 ( Sigma Aldrich , Saint Louis ) , rabbit anti-GAPDH and goat anti-AIF1 ( Santa Cruz Biotechnology , Santa Cruz , CA ) , rabbit anti-Apaf-1 ( LSBio , Seattle , WA ) , mouse anti PSD-95 ( Affinity BioReagent ) , mouse anti-MMP11 ( Thermo Scientific , Waltham , MA ) , rabbit anti-S-opsin ( Millipore/Pel-Freez Biologicals ) , rabbit anti-Nr2E3 ( Proteintech , Hayward , CA ) and Peanut Agglutinin TRTIC conjugate ( Sigma-Aldrich ) . Alexa-conjugated secondary antibodies ( 408 , 488 , 568 and Cy5 ) were from Invitrogen ( Carlsbad , CA ) . For total RNA and pre-miRNA isolation , retinas were homogenized with the TRIZOL Regent ( Invitrogen ) using Bullet Blender BBX24 ( Next Advance Inc . , Averill Park , NY ) in the presence of 0 . 5 mm zirconium oxide beads ( Next Advance Inc . , Averill Park , NY ) for 3 min at 8 , 000 rpm . RNA was reverse transcribed into cDNA using SuperScript II reverse transcriptase ( Invitrogen ) . For RT-PCR , the ∼380 bp amplicon encompassing fused exons 1 and 3 of recombinant Ranbp2 mRNA was amplified from 250 ng of retinal cDNA . PCR was performed using primer 1 ( Pr1:CGCCCCGAGAGTACATTTCTA ) and primer 2 ( Pr2:AAGTTTATTCCATCCATCTTCA ) with GoTaq Green Master Mix ( Promega , Madison , WI ) under the following cycling conditions: 5 min/95°C of initial denaturation , 35 cycles/94°C ( 30 s ) , 55°C ( 30 s ) and 72°C ( 30 s ) with final elongation step at 72°C for 3 mins . Same cycling conditions were applied for Cre ( CTAATCGCCATCTTCCAGCAGG , AGGTGTAGAGAAGGCACTTAGC ) and Gapdh ( GCAGTGGCAAAGTGGAGATT , GAATTTGCCGTGAGTGGAGT ) . qRT-PCR reactions were carried out with 8 ng of cDNA , 800 nM forward and reverse primers , 10 µl of 2×SYBR® Green PCR Master Mix ( Applied Bioscience , Warrington , MA ) in a 20 µl final volume in 48-well plates using the ECO™ Real-Time PCR system ( Illumina , San Diego , CA ) . The relative amount of transcripts was calculated by the ΔΔ CT method using Gapdh as reference ( n = 3–4 ) . Primer sequences and designations are provided in Table S1 . The superior region of mouse cornea was burned using Low Temperature Cautery ( Bovie Medical Corporation , St . Peterburg , FL ) immediately after mice were killed . For immunohistochemistry , eyeballs were removed and fixed with 2% paraformaldehyde/phosphate-buffer saline ( PBS ) , pH 7 . 4 for 4 hr after small incisions were made in the anterior portion . Upon removal of the lens , eyecups were infiltrated with 5% sucrose/100 mM PBS , pH 7 . 4 , for 5 hr followed by 30% sucrose/100 mM PBS for 12 hours , embedded in Tissue-Tek O . C . T . compound ( Sakura , Torrance ) and stored at −80°C . 12 µm thick retinal cryosections along the vertical meridian of the eyecup were mounted on glass slides . For flat mounts , retinas were removed from fixed eyeballs , cut in a four-quadrant cloverleaf pattern using the caruncle as an orientation landmark and fixed for an additional 15 min in a 24-well plate . Specimens were incubated in blocking buffer ( PBS , pH 7 . 4 , containing 0 . 1% Triton X-100 , 10% normal goat serum ) for 1 hr at RT ( for sections ) or 12 hr at 4°C ( flat mounts ) , followed by incubation with primary antibodies in incubation buffer ( PBS , pH 7 . 4 , containing 0 . 1% Triton X-100 , 5% normal goat serum ) overnight at RT ( for sections ) or for 3 days at 4°C ( flat mounts ) . Specimens were washed thrice with washing buffer ( PBS/0 . 1% Triton X-100 ) for 10 min , and incubated in incubation buffer for 2 hr with Alexa-conjugated secondary antibodies ( 1∶1 , 000; Invitrogen ) . For Nr2E3 immunostaining , eyeballs were fixed instead with 1% paraformaldehyde for 1 hr at room temperature . 6 µm-thick retinal cryosections were incubated in blocking buffer for 1 hr at room temperature followed by treatment with proteinase K ( 20 µg/ml , Promega , Madison , WI ) for 9 min and standard immunostaining protocols as described earlier . Specimens were washed again thrice and mounted on glass slides for visualization and image acquisition . Images were acquired with a Nikon C1+-laser scanning confocal microscope coupled with a LU4A4 launching base of 4-solid state diode lasers ( 407 nm/100 mW , 488 nm/50 mW , 561 nm/50 mW , 640 nm/40 mW ) and controlled by the Nikon EZC1 . 3 . 10 software ( v6 . 4 ) . TUNEL assays were performed with the DeadEnd Fluorometric TUNEL System ( Promega , Medison ) with the following modifications from the manufacturer's instructions . Briefly , specimens were incubated with 20 µg/ml proteinase K for 15 min at RT followed by fixation with 4% paraformaldehyde for 15 min , and incubations with primary and secondary antibodies along with DAPI ( Invitrogen , Carlsbad , CA ) . Specimens were equilibrated for 5 min in manufacturer's buffer before undergoing TdT reactions for 60 min at 37°C . Reactions were stopped with 2×SSC , washed three times with PBS and mounted . Fresh P20 retinal explants were incubated with 5 µM EthD-III ( Biotium , Hayward , CA ) in the Neurobasal−A Medium/B-27 Supplement ( Invitrogen ) for 4 hr at 37°C in a humidified 5% CO2 atmosphere , washed vigorously 3 times with 100 mM PBS for 10 min and fixed with 2% paraformaldehyde/100 mM PBS for 20 min at room temperature . Specimens were then processed as described for immunostaining and TUNEL procedures . Morphometric analyses of M-cone photoreceptors were performed from 127×127 µm image fields captured with a Nikon C1+-laser scanning confocal microscope . Optical slices were 3D-reconstructed for the whole length of outer segments ( ∼25 µm , step size of 0 . 5 µm ) from retinal flat mounts immunostained with an L/M-opsin antibody [55] . M-cone photoreceptors and the length of their outer segments were then tallied and measured from three image fields for each retina with the post-acquisition Nikon Elements AR ( ver3 . 2 ) software . Cre+ , c-casp3+ and TUNEL+-cells bodies were imaged from flat mounts and three image fields of 127×127 µm and collapsed for the whole depth of ONL ( ∼60 µm , step size of 3 µm ) from the central or peripheral retinal regions . For TUNEL+ , EthD-III+ and Arr4+-cells bodies , three image fields of 127×127×5 µm from retinal flat mounts were randomly selected and quantitatively analyzed . For tallying of DAPI+ , TUNEL+ , −Nr2E3+or Cre+-cell bodies , whole vertical meridian sections were counted and averaged to generate pie graphs ( Excel , Microsoft , Seattle , WA ) . The distal region of the ONL was arbitrarily defined as the 15 µm distal segment of the ONL up to the external limiting membrane and the remaining segment was defined as proximal . 3D-reconstruction of collapsed images and morphometric analysis was performed with Nikon Elements AR software ( ver . 3 . 2 ) . Two-tailed equal or unequal variance t-test statistical analysis was performed . p≤0 . 05 was defined as significant . Retinas were homogenized at 4°C with Bullet Blender BBX24 ( Next Advance Inc . , Averill Park , NY ) in the presence of 0 . 5 mm zirconium oxide beads ( Next Advance Inc . ) and RIPA buffer containing complete protease inhibitors ( Roche Applied Bioscience , Penzberg , Germany ) and 10 mM iodoacetamide ( Sigma Aldrich ) . Protein concentration was measured by the BCA method using BSA as a standard . Samples ( 55 µg ) were resolved on 11% SDS-polyacrylamide gel electrophoresis ( SDS-PAGE ) , immunoblotted and developed using the SuperSignal Pico West ( Thermo Scientific ) as described previously [50] . Blots were probed with mouse anti-MMP-11 antibody ( Thermo Scientific ) . Densitometry analysis of immunoblots was performed with Metamorph v7 . 0 ( Molecular Devices , Sunnyvale , CA ) . Two-tail t-test statistical analysis was performed for validation of significant change ( p<0 . 05 ) . Retinas were homogenized using BulletBlender BBX24 ( Next Advance Inc . ) with 0 . 5 mm zirconium oxide beads ( Next Advance Inc . ) and RIPA buffer . Retinal homogenates were centrifuged at 10 , 000 g for 15 min at 4°C . Supernatants were collected and protein concentrations determined by the BCA method using BSA as standard . Retinal homogenates were diluted in caspase assay buffer and caspase profiling assays were performed with the Sensolyte AFC Caspase sampler kit according to company's protocol ( AnaSpec , Fremont , CA ) . The following caspase substrates were used for screening: Ac-YVAD-AFC ( SB1 ) and Ac-WEHD-FAC ( SB2 ) for caspase 1 , Ac-VDVAD-FAC ( SB3 ) for caspase 2 , Ac-IETD-FAC , ( SB4 ) for caspase 8 , Ac-DEVD-FAC ( SB5 ) and Z-DEVD-AFC ( SB6 ) for caspases 3/7 , Ac-LEHD-FAC ( SB7 ) for caspase 9 and Ac-VEID-AFC ( SB8 ) for caspase 6 . Assays were first optimized for substrate dilution , extract concentration and reaction time . Substrate screenings were carried with three different protein concentrations . Analytical assays of caspase 3/7 ( SB5 ) were performed with diluted 50 µl of retinal homogenate ( 280 ng ) and 50 µl of substrate ( 1∶200 dilution in caspase assay buffer ) , mixed in the 96 well-plate , incubated with shaking for 1 hr under dark at room temperature . Measurements of fluorescence were performed at excitation/emission/cutoff = 380/500/495 nm with SpectraMax M5 ( Molecular Devices , Sunnyvale , CA ) . Control reactions without extracts were subtracted from the samples' readings . MMPs' screening assays were performed with Sensolyte 520 generic MMP assay kit per company's protocol ( AnaSpec , Fremont , CA ) . Retinas extracts were prepared as described previously with the exception that RIPA buffer was replaced with MMP assay buffer ( AnaSpec ) . Briefly , MMP assays are based on the dequenching of fluorescence intensity of 5-FAM upon proteolytic cleavage of 5-FRAM/QXL520 peptide substrate by MMPs . MMPs screenings , except MMP11 , were performed upon activation with 1 mM 4-aminophenylmercuric acetate ( APMA ) at different time intervals ( MMP2 , 7 , 8 and 13: 1 hr activation; MMP9 , 12 and 14: 2 hr activation; MMP1: 3 hr activation; MMP3 and 10: 24 hr activation ) at 37°C followed by 1 hr incubation with 5-FRAM/QXL520 peptide substrate . For MMP11 , retinal extract was directly mixed with substrate and incubated for 1 hr before fluorescence reading . 50 µl of activated/non-activated protein extract at 56 , 112 or 224 ng and 50 µl of substrate ( final substrate dilution 1∶200 ) were mixed in a 96 well-plate , incubated with shaking for 1 h at room temperature at dark . Measurements of fluorescence were performed at excitation/emission/cutoff = 490/520/495 nm with SpectraMax M5 ( Molecular Devices , Sunnyvale , CA ) . Control measurements without retinal extracts under the same conditions were subtracted from the sample readings . Analytical assays with MMP11 were performed with 56 ng of retinal extract . Mitochondrial and cytosolic fractions of the retina were isolated with the Mitochondria Isolation Kit for Tissue ( Abcam , Cambridge , MA ) per manufacturer's instruction . Briefly , retinas were washed with washing buffer , homogenized with a Kontes Microtube Pellet Pestle Rod with motor in isolation buffer , centrifuged at 1 , 000× g for 10 mins , the pellet was saved ( nuclear-enriched fraction ) and supernatant was centrifuged again at 12 , 000× g for 15 min . The supernatants ( cytosolic fraction ) were saved , the pellets ( mitochondrial fraction ) washed with Isolation buffer twice and re-suspended with isolation buffer with complete protein inhibitor cocktail ( Roche Applied Bioscience , Penzberg , Germany ) . Eyeballs were removed and fixed with 2% glutaraldehyde:paraformaldehyde/0 . 1% cacodylate buffer , pH 7 . 2 , overnight at 4°C . For semi-thin histological sections , 0 . 5 µm sections along the vertical meridian were mounted on glass slides and stained with 1% methylene blue . Light images of the retina sections were acquired with a Axiopan-2 light microscope controlled by Axovision Rel 4 . 6 and coupled to a AxioCam HRc digital camera ( Carl Zeiss , Germany ) . For electron microscopy , specimens were post-fixed in 2% osmium tetraoxide in 0 . 1% cacodylate buffer and embedded in Spurr resin . 60 nm-thick sections were cut with Leica Ultracut S ( Leica Microsystems , Waltzer , Germany ) , stained with 2% uranyl acetate/4% lead citrate and imaged with JEM-1400 transmission electron microscope ( JEOL , Tokyo , Japan ) coupled with an ORIUS 1000CCD camera . The Ingenuity pathway analysis ( IPA , Ingenuity Systems , Redwood City , CA ) was used to examine the gene dataset modulated by loss of Ranbp2 and to define connectivity maps and networks . Network of genes are algorithmically produced based on their connectivity . Significant network scores reflect the negative logarithm of a P value associated with the likelihood of connectivity of a set of genes in a network . The network with highest score ( score of 22 ) was chosen for further analysis . After overnight dark adaptation , mice were anesthetized ( ketamine: 80 mg/kg; xylazine: 16 mg/kg ) and eyedrops were used for pupil dilation ( 1% tropicamide; 2 . 5% phenylephrine HCl; 1% cyclopentolate HCl ) and corneal anesthesia ( 1% proparacaine HCl ) . The active electrode was a stainless-steel wire active electrode that contacted the corneal surface through 1% methylcellulose; needle electrodes placed in the cheek and tail served as reference and ground leads , respectively . Responses were differentially amplified ( 0 . 3–1 , 500 Hz ) , averaged , and stored using a UTAS E-3000 signal averaging system ( LKC Technologies , Gaithersburg , MD ) . Strobe flash stimuli were initially presented in darkness within a ganzfeld bowl . Flash luminance ranged from −3 . 6 to 2 . 1 log cd s/m2 and stimuli were presented in order of increasing luminance . A steady adapting field ( 20 cd/m2 ) was then presented in the ganzfeld . After a 7 min light adaptation period cone ERGs were evoked by strobe flash stimuli superimposed upon the adapting field . Flash luminance ranged from −0 . 8 to 1 . 9 log cd s/m2 and stimuli were presented at 2 Hz in order of increasing luminance . The amplitude of the a-wave was measured 8 ms after flash onset from the prestimulus baseline . The amplitude of the b-wave was measured from the a-wave trough to the peak of the b-wave or , if no a-wave was present , from the pre-stimulus baseline .
|
The secondary demise of healthy neurons upon the degeneration of neurons harboring primary genetic defect ( s ) is hallmark to neurodegenerative diseases . However , the factors and mechanisms driving these cell-death processes are not understood , a severe limitation which has hampered the therapeutic development of neuroprotective approaches . The neuroretina is comprised of two main types of photoreceptor neurons , rods and cones . These undergo degeneration upon heterogeneous mutations or environmental stressors and the underlying diseases present conspicuous spatiotemporal pathological signatures whose molecular bases are not understood . We employed the multifunctional protein , Ran-binding protein-2 ( Ranbp2 ) , which is implicated in cell-type and stress-dependent clinical manifestations , to examine its role ( s ) in primary and secondary photoreceptor death mechanisms upon its specific loss in cones . Contrary to prior findings , we found that dying cones can trigger the loss of healthy rods . This process arises by the immediate activation of novel Ranbp2-responsive factors and downstream cascade events in cones that promote extrinsically the demise of rods . The mechanisms of rod and cone demise are molecularly distinct . Collectively , the data uncover distinct Ranbp2 roles in intrinsic and extrinsic cell-death and will likely contribute to our understanding of the spatiotemporal onset and progression of diseases affecting photoreceptor mosaics and other neural networks .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"death",
"molecular",
"neuroscience",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"gene",
"regulation",
"neuroscience",
"animal",
"models",
"model",
"organisms",
"molecular",
"genetics",
"biology",
"mouse",
"visual",
"system",
"cellular",
"neuroscience",
"neuronal",
"morphology",
"neurons",
"genetics",
"cellular",
"types",
"sensory",
"systems",
"molecular",
"cell",
"biology",
"genetics",
"of",
"disease"
] |
2013
|
Distinct and Atypical Intrinsic and Extrinsic Cell Death Pathways between Photoreceptor Cell Types upon Specific Ablation of Ranbp2 in Cone Photoreceptors
|
Lymphatic filariasis can be associated with development of serious pathology in the form of lymphedema , hydrocele , and elephantiasis in a subset of infected patients . Dysregulated host inflammatory responses leading to systemic immune activation are thought to play a central role in filarial disease pathogenesis . We measured the plasma levels of microbial translocation markers , acute phase proteins , and inflammatory cytokines in individuals with chronic filarial pathology with ( CP Ag+ ) or without ( CP Ag− ) active infection; with clinically asymptomatic infections ( INF ) ; and in those without infection ( endemic normal [EN] ) . Comparisons between the two actively infected groups ( CP Ag+ compared to INF ) and those without active infection ( CP Ag− compared to EN ) were used preliminarily to identify markers of pathogenesis . Thereafter , we tested for group effects among all the four groups using linear models on the log transformed responses of the markers . Our data suggest that circulating levels of microbial translocation products ( lipopolysaccharide and LPS-binding protein ) , acute phase proteins ( haptoglobin and serum amyloid protein-A ) , and inflammatory cytokines ( IL-1β , IL-12 , and TNF-α ) are associated with pathogenesis of disease in lymphatic filarial infection and implicate an important role for circulating microbial products and acute phase proteins .
Although two-thirds of the 120 million people infected with Wuchereria bancrofti—the major causative agent of human lymphatic filariasis—have subclinical infections , ∼40 million have lymphedema and/or other pathologic manifestations including hydroceles ( and other forms of urogenital disease ) , episodic adenolymphangitis , tropical pulmonary eosinophilia , lymphedema , and ( in its most severe form ) elephantiasis [1] . It is assumed that repeated episodes of acute inflammation can lead to development of serious disfigurement in the face of compromised lymphatics [2] , although many other factors that contribute to the pathology associated with lymphatic filariasis are largely unknown . Typically in Wuchereria or Brugia infections , disease manifests years after exposure , while clinically asymptomatic infection is not only more common but can occur at a relatively young age [3] , [4] . Lymphatic filarial disease is felt to be a reflection of both localized and systemic immunologic and inflammatory responses mediated by pro-inflammatory cytokines and chemokines [5] , [6] . Although some of the pathological changes can likely be initiated by Wolbachia- or parasite-encoded endotoxin-like substances and/or secondary bacterial or fungal infections [1] , [7] , chronic parasite-induced immune activation is a salient feature of filarial disease . Indeed , increased frequencies of activated T cells [8] , increased parasite antigen-driven Th1 and Th17 cytokine production [6] , increased expression of Toll-like and NOD-like receptors [6] , and enhanced TLR signaling through TLR ligand stimulation [5] have all been described when comparisons are made between patients with subclinical infection and those with filarial lymphedema and/or elephantiasis . Moreover , innate immune responses also play a prominent role in development of pathology , as evidenced by the occurrence of lymphatic damage in animal models of filarial infection lacking an adaptive immune system [9] . Persistent immune activation is associated with elevations of circulating microbial products , acute phase proteins , and the so-called microbial translocation molecules [10] . Translocation of microbial products from the lumen of the intestine into the periphery is thought to contribute to induction of inflammation by stimulating immune effector cells directly through their pattern recognition receptors [11]; however , intra- and peri-lymphatic damage—an underlying feature of filarial disease [12]—might also contribute to the presence of microbial translocation products in the bloodstream . In addition , chronic immune activation that often accompanies infectious processes [13] is associated with development of an acute phase response and the presence of markers of inflammation in plasma . Moreover , increased serum levels of proinflammatory cytokines and chemokines are commonly associated with progressive immune activation . In this study , we have delineated the role of many of the known markers of inflammation and lymphatic damage that reflect the dysregulated ( or unchecked ) responses related to development of disease with the lymphatic-dwelling filariae ( Wuchereria bancrofti and Brugia malayi ) . Our data suggest that circulating ( systemic ) microbial products , acute phase proteins , and pro-inflammatory cytokines reflect the localized ( and ongoing ) chronic immune activation that underlies the pathogenesis of disease in lymphatic filariasis .
We studied a group of 91 individuals with filarial lymphedema without active filarial infection ( hereafter CP Ag− ) , 28 individuals with filarial lymphedema with active filarial infection ( hereafter CP Ag+ ) , 98 asymptomatic or subclinical , infected individuals ( hereafter INF ) , and 82 uninfected , endemic normal individuals ( hereafter EN ) in an area endemic for lymphatic filariasis in Tamil Nadu , South India ( table 1 ) . Diagnosis of active filarial infection was performed by measuring circulating filarial antigen levels by both the ICT filarial antigen test ( Binax , Portland , ME , USA ) and the TropBio Og4C3 enzyme-linked immunosorbent assay ( ELISA ) ( Trop Bio Pty . Ltd , Townsville , Queensland , Australia ) . All the CP Ag− individuals had undergone treatment with repeated doses of diethylcarbamazine ( DEC ) . None of the CP Ag+ individuals had received any DEC treatment but were administered DEC following the blood draw . All of the CP individuals had early stage lymphedema ( Grades 1 and 2 ) only and individuals with concurrent overt and active bacterial infection were excluded from the study . All individuals were examined as part of a clinical protocol approved by Institutional Review Boards of both the National Institute of Allergy and Infectious Diseases and the Tuberculosis Research Center ( NCT00375583 and NCT00001230 ) , and informed written consent was obtained from all participants . To inactivate plasma proteins , plasma samples were heated to 75°C for 5 min . Lipopolysaccharide ( LPS ) levels were measured using a limulus amebocyte lysate assay ( Cell Sciences Hycult Biotech , Canton , MA , USA ) according to the manufacturer's protocol . Commercially available ELISA kits were used to measure plasma levels of LPS- binding protein ( LBP ) , endotoxin core antibodies IgG ( EndoCAb ) ( Cell Sciences Hycult Biotech ) , and soluble CD14 ( sCD14 ) ( R&D Systems , Minneapolis , MN , USA ) . Plasma levels of C-reactive protein ( CRP ) , haptoglobin , serum amyloid protein - A ( SAA ) , and α-2 macroglobulin ( α-2M ) were measured using the Bioplex ( Bio-Rad , Hercules , CA , USA ) multiplex ELISA system according to the manufacturer's instructions . Plasma levels of cytokines , IL-1β , IL-6 , IL-12 , and TNF-α ( Bio-Rad ) were measured using the Bioplex multiplex ELISA system . Data analyses were performed using GraphPad PRISM ( GraphPad Software , Inc . , San Diego , CA , USA ) . Geometric means ( GM ) were used for measurements of central tendency . Preliminary statistical analysis was done using the non-parametric Mann-Whitney test . We then tested for group effects using linear models on the log transformed data . We used robust standard error with a recommended bias adjustment so that we need not assume that the error variance was the same for each group . We parameterized the 4 group effects using parameters for CP , infection , and CP by infection interaction . Since we tested these 3 parameters on 12 markers , we adjusted the p-values for multiple comparisons using Holm's adjustment . P-values in Table 2 are Holm's adjusted . We then built specific models using only the significant ( when Holm's adjusted p-value<0 . 05 ) effects , and we present those effects and ( unadjusted ) 95% confidence intervals as fold-change ( Table 3 ) . We repeated the models after adding an effect for age ( either as a continuous or a categorical variable ) . Linear models were done in R 2 . 14 . 0 using the sandwich R package . Correlations were calculated by the Spearman rank correlation test . The heat map was constructed in JMP v8 . 0 ( SAS , Carey , NC ) and is based on relative expression for a given analyte as a function of the geometric mean value found in the endemic normal population .
To determine the association of microbial translocation and related markers with filarial lymphedema , we measured the plasma levels of LPS , LPB , EndoCAb , and sCD14 in CP Ag+ , INF , CP Ag− , and EN . As shown in figure 1 , CP Ag+ had significantly higher levels of LPS ( GM of 4 . 24 EU/ml in CP Ag+ vs . 0 . 10 in INF; P<0 . 0001 by Mann-Whitney ) but not sCD14 or EndoCAb in comparison to INF . Conversely , CP Ag+ had significantly lower levels of LBP ( GM of 306 . 2 ng/ml in CP Ag+ vs . 21658 in INF; P<0 . 0001 ) in comparison to INF . However , no significant differences were observed in the levels of all four circulating microbial or related products between CP Ag− and EN . In addition , we consistently observed an inverse association between LPS and LBP levels in the CP Ag+ group ( r2 = 0 . 862; P<0 . 0001 ) . Thus , filarial lymphedema with active infection is characterized by elevated levels of circulating LPS . To determine the association of acute phase proteins with filarial disease , we measured the plasma levels of CRP , haptoglobin , SAA , and α-2m in the four groups . As shown in figure 2 , CP Ag+ had significantly higher levels of CRP ( GM of 30 . 9 pg/ml in CP Ag+ vs . 4 . 11 in INF; P<0 . 0001 ) , haptoglobin ( GM of 555 . 9 pg/ml in CP Ag+ vs . 140 . 1 in INF; P<0 . 0001 ) , SAA ( GM of 196 . 7 pg/ml in CP Ag+ vs . 96 . 9 in INF; P = 0 . 0037 ) , and α-2m ( GM of 4383 pg/ml in CP Ag+ vs . 1923 pg/m in INF; P = 0 . 0003 ) in comparison to INF . Similarly , among those without evidence of active filarial infection ( Ag− ) , those with CP had significantly higher levels of CRP in comparison to EN ( GM of 14 . 5 pg/ml in CP Ag− vs . 1 . 9 in EN; P<0 . 0001 ) , indicating that elevated CRP levels might be more reflective of the secondary events associated with pathology than with active infection . Thus , filarial lymphedema with active infection is characterized by elevated levels of several acute phase proteins . To determine the association of inflammatory cytokines with filarial lymphedema , we measured the plasma levels of IL-1β , IL-6 , IL-12 , and TNF-α in the four groups of subjects . As shown in figure 3 , compared with INF , those with CP Ag+ had significantly higher levels of IL-1β ( GM of 410 . 9 pg/ml in CP Ag+ vs . 210 . 3 in INF; P = 0 . 0305 ) , IL-12 ( GM of 989 . 1 pg/ml in CP Ag+ vs . 61 . 2 in INF; P<0 . 0001 ) , and TNF-α ( GM of 2455 pg/ml in CP Ag+ vs . 727 . 2 in INF; P<0 . 0001 ) but not IL-6 . However , no significant differences were observed in the levels of all four cytokines between those without active infection ( EN and CP Ag− ) irrespective of clinical status . Thus , filarial lymphedema with active infection is characterized by elevated plasma levels of inflammatory cytokines . We tested for three effects ( CP effect , infection effect , and CP-by-infection interaction effect ) on each of 12 markers using linear models on the log transformed responses . A significant ( CP-by-infection ) interaction effect meant that the geometric mean ( GM ) for the marker in the CP Ag+ group is significantly different from GM expected from the combined effects of CP and infection . As shown in Table 2 , we observed significant effects in the models for LPS , LBP , CRP , Haptoglobin , SAA , IL-1β , IL-6 , IL-12 and TNF-α . We then examined the details of the significant markers by rebuilding the linear model using only the significant ( by adjusted p-value ) effects . For LPS , Haptoglobin , IL-1β , and IL-12 , we observed that the CP Ag+ group had significantly higher responses than the other 3 groups , while for LBP we observed that the CP Ag+ group has significantly lower responses than the other groups ( see Table 3 ) . For the 4 other markers with significant effects ( CRP , SAA , IL-6 , and TNF-α ) , we observed that CRP was significantly associated with chronic pathology ( both CP Ag+ and CP Ag− ) ; SAA and TNF- α were significantly associated with infection status ( both CP Ag+ and INF ) ; and IL-6 was associated with both pathology and the infection status ( data not shown ) . Thus , by using robust statistical calculations , we have confirmed the association of LPS , acute phase proteins and inflammatory cytokines with filarial lymphedema with active infection . A more detailed examination of the associations is presented in the Text S1 . The relationships between the levels of LPS and/or LBP levels and plasma cytokines were next assessed ( figure 4 ) . As shown in figure 4A , levels of LPS exhibited a highly significant positive correlation with the plasma levels of IL-1β ( r = 0 . 4942; P<0 . 001 ) , IL-12 ( r = 0 . 4802; P<0 . 0001 ) , and TNF-α ( r = 0 . 4494; P<0 . 0001 ) in all actively infected individuals . Conversely , LBP levels were significantly negatively correlated with the plasma levels of IL-12 ( r = −0 . 3255; P = 0 . 0005 ) ( figure 4B ) . Thus , the process by which microbial translocation occurs appears to be significantly associated with the pro-inflammatory cytokine levels in filarial infection . We also compiled the comparative analysis of all the 12 parameters in the 4 groups of individuals as a heat map , depicting the log transformed data on a scale relative to EN . As shown in figure 5 , CP Ag+ individuals exhibit a distinct biomarker signature characterized by elevated levels of LPS , acute phase proteins , and certain inflammatory cytokines compared with the other 3 groups ( EN , INF , and CP Ag− ) , again reiterating the important association of these factors with pathogenesis of filarial pathology .
Studies in experimental animal models suggest that intestinal injury and systemic endotoxemia are two major factors leading to morbidity in helminth infections [14] , [15] . Disruption of the integrity of the intestinal epithelium and translocation of microbial products into the circulation is thought to occur in intestinal helminth infections [16] . Thus , infection with intestinal helminths is characterized by enhanced leakiness of the intestinal epithelium , mediated by activated mast cells , which can lead to the movement of bacterial LPS into the portal circulation [17] , [18] . Even in non-intestinal helminth infections , such as schistosomes that reside in the mesenteric veins , damage caused by worm eggs traversing the gastrointestinal epithelium can result in systemic translocation of bacteria [14] , [19]; however , the role of microbial translocation in the pathogenesis of disease in systemic helminth infections is not clear . Lymphatic filariasis is a disease characterized by the dysfunction of lymphatics leading to severe ( and often ) irreversible lymphedema and elephantiasis . It has been shown that residence of adult parasites in the lymphatics leads to a cascade of events that ultimately results in tissue scarring and fibrosis [20] . Studies addressing the mechanisms underlying parasite-induced lymphatic dilatation suggest that parasite-mediated lymphatic dilatation and lymphangiogenesis might be important features in the development of pathology [7] , [12] . In addition , the more severe forms of lymphedema are often associated with secondary bacterial and/or fungal infections leading to dermatolymphangioadenitis , which also contribute to the pathogenesis of disease [2] . Finally , filarial lymphedema has been shown to be associated with increased bacterial loads in the lymphatics [21] , [22] , [23] , [24]; these damaged lymphatics could then serve as a potential nidus for bacterial translocation through the lymphatic endothelium . Thus , the predominant feature of lymphatic filarial disease is the establishment of a systemic inflammatory milieu due to both parasite-derived and host-induced inflammation . We examined four important circulating microbial or related products in our study . LPS ( a key indicator of microbial translocation ) was found to be significantly elevated in CP Ag+ compared with INF but not in CP Ag− compared with EN . In addition , LPS levels were also found to be significantly elevated in the CP Ag+ compared to all the other 3 groups combined or individually . Strikingly , we observed the exactly opposite profile with LBP , the LPS binding protein commonly produced by gastrointestinal and hepatic epithelial cells in response to LPS stimulation [25] . LBP is also known to bind and transfer LPS to high-density lipoproteins to decrease the bioactivity of LPS [25] and therefore , the lower levels of LBP in CP Ag+ individuals might reflect an inability to clear LPS in circulation . Although we examined the levels of sCD14 , which binds LPS and is produced by monocytes/macrophages [25] , and the naturally occurring IgG antibody to the LPS core oligosaccharide ( EndoCAb ) [26] in all groups of subjects , we found no differences in these particular molecules . Our study therefore suggests that circulating LPS and LBP ( but not sCD14 or EndoCAb ) are potentially associated either with the development of pathology or function as markers for pathogenesis . While elevated levels of LPS in CP Ag+ compared with INF could potentially be attributed to presence of secondary bacterial infection , the elevated immune activation observed in chronic pathology patients with active infection suggests that the interaction between filarial infection and pathology is a major contributor to microbial translocation , fueling systemic immune activation . Interestingly , our findings are similar to findings reported in other infectious diseases also characterized by systemic immune activation , including HIV [27] , [28] , hepatitis B and C [29] , and schistosomiasis [19] , [30] . Acute phase proteins derive primarily from the liver , and plasma concentrations are felt to be a reflection of the response to pro-inflammatory cytokines [31] . Measurement of acute phase proteins is of clinical importance in determining the presence and extent of inflammatory tissue damage as well as in providing diagnostic and prognostic information [32] , [33] . Moreover , circulating microbial products are well known inducers of acute phase proteins , with SAA and haptoglobin known to be markedly elevated following challenge with LPS [34] . Elevated levels of CRP have been reported in lymphatic filarial disease [35] , but other acute phase proteins have not been examined . In the present study , we confirmed that CRP levels are indeed elevated in actively infected patients with chronic lymphedema compared with the asymptomatic group , but we also found that haptoglobin , SAA , and α-2m are also elevated . Upon , further analysis , only haptoglobin was observed to be significantly associated with filarial-infection with pathology , while SAA was significantly associated with filarial infection per se ( both CP Ag+ and INF ) . Interestingly , CRP levels were significantly elevated even in those patients with chronic lymphedema without active infection , indicating that CRP is probably a nonspecific marker of inflammation in filarial disease , whereas haptoglobin might serve as a more accurate biomarker of filarial infection-driven pathology . Although persistent and progressive inflammation is postulated to be a hallmark of lymphatic filarial disease , very few studies have actually examined the levels of inflammatory cytokines or chemokines in the circulation of infected or diseased individuals . Previous reports have suggested that IL-6 and IL-8 are morbidity markers in acute and chronic disease [36] , while IL-6 and TNF-α are involved in the pathogenesis of adverse reactions following treatment [37] . Our examination of cytokine expression levels in the four groups of individuals reveals that IL-1β and IL-12 are significantly associated with overt pathology in actively infected individuals . Conversely , TNF-α was associated significantly with groups having active infection ( CP Ag+ and INF ) indicating a possible association with filarial infection rather than pathology alone . Because inflammatory cytokines are intricately linked to induction of both circulating microbial products and acute phase proteins , we also examined their interrelationship in the CP Ag+ population . Detection of microbial invasion by cells of the innate immune system usually results in increased production of pro-inflammatory cytokines such as IL-1β , IL-6 , IL-12 , and TNF-α [11] . Studies in HIV infection reveal a direct association between levels of microbial translocation markers such as LPS and the inflammatory cytokines [27] , [28] . In agreement with such studies , our examination of filaria-infected individuals also reveals a significantly positive association between LPS and pro-inflammatory cytokines in filaria-infected-individuals . Our study clearly implicates an association for LPS , the acute phase proteins , and several of the pro-inflammatory cytokines with filaria-induced lymphatic pathology . Investigations into filarial disease pathogenesis have implicated host pathways in disease progression . In particular , dysregulated inflammatory responses and lymphatic dysfunction are thought to be central processes in severe filarial pathology [7] , [12] . Our study reveals novel insights into the pathogenesis of lymphatic filarial dysfunction , despite some minor limitations . Since DEC had been administered only to the CP Ag− group and the presence of other parasitic infections not examined , the effect of treatment with DEC as well as the influence of other parasitic infections could not be ascertained in this study . Another minor limitation of the study was that plasma levels of inflammatory markers—such as circulating microbial products , acute phase proteins , and cytokines—are relatively nonspecific and may be influenced by short half-life , nonspecific induction , and plasma levels not reflecting biologic activity . Notwithstanding these limitations , plasma levels of some of these same biomarkers have yielded important insights in the diagnosis and/or prognosis of various infectious diseases and cancers [10] , [38] , [39] . Our study clearly identifies a signature set of biomarkers that serves to indicate filarial infection-driven morbidity associated with a persistent and progressive inflammatory milieu . While requiring validation in future studies , these results point to potential prognostic indicators of severe filarial disease .
|
Lymphatic filariasis afflicts over 120 million people worldwide . While the infection is mostly clinically asymptomatic , approximately 40 million people suffer from overt , morbid clinical pathology , characterized by swelling of the scrotal area and lower limbs ( hydrocele and lymphedema ) . Host immunologic factors that influence the pathogenesis of disease in these individuals are not completely understood . Circulating microbial products such as LPS and markers associated with microbial translocation have been shown to play an important role in disease pathogenesis of certain infections like HIV . Similarly , proteins associated with the acute phase response and related cytokines also play an important role in pathogenesis . We have attempted to elucidate the role of the above mentioned factors in disease pathogenesis by comparing the plasma levels of the various markers in four groups of individuals: chronic pathology individuals with or without active filarial infection , asymptomatic , filarial infected individuals and uninfected , endemic normal individuals . We show that circulating levels of LPS , acute phase proteins and certain cytokines are significantly elevated in filarial disease with active infection but not in the other groups indicating that filarial infection induced increased production of these factors correlated with the development of filarial lymphatic pathology .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"clinical",
"immunology",
"immunology"
] |
2012
|
Circulating Microbial Products and Acute Phase Proteins as Markers of Pathogenesis in Lymphatic Filarial Disease
|
We have developed a manufacturing strategy that can improve the safety and genetic stability of recombinant live-attenuated chimeric dengue vaccine ( DENVax ) viruses . These viruses , containing the pre-membrane ( prM ) and envelope ( E ) genes of dengue serotypes 1–4 in the replicative background of the attenuated dengue-2 PDK-53 vaccine virus candidate , were manufactured under cGMP . After deriving vaccine viruses from RNA-transfected Vero cells , six plaque-purified viruses for each serotype were produced . The plaque-purified strains were then analyzed to select one stock for generation of the master seed . Full genetic and phenotypic characterizations of the master virus seeds were conducted to ensure these viruses retained the previously identified attenuating determinants and phenotypes of the vaccine viruses . We also assessed vector competence of the vaccine viruses in sympatric ( Thai ) Aedes aegypti mosquito vectors . All four serotypes of master vaccine seeds retained the previously defined safety features , including all three major genetic loci of attenuation , small plaques , temperature sensitivity in mammalian cells , reduced replication in mosquito cell cultures , and reduced neurovirulence in new-born mice . In addition , the candidate vaccine viruses demonstrated greatly reduced infection and dissemination in Aedes aegypti mosquitoes , and are not likely to be transmissible by these mosquitoes . This manufacturing strategy has successfully been used to produce the candidate tetravalent vaccine , which is currently being tested in human clinical trials in the United States , Central and South America , and Asia .
The dengue virus ( DENV ) complex , genus Flavivirus , comprises four distinct serotypes of viruses ( DENV-1 to -4 ) . DENV causes the most rapidly spreading mosquito-borne human viral disease in the world . About 50–100 million new dengue infections occur each year in more than 100 endemic countries [1] , with an estimated 500 , 000 cases exhibiting hemorrhagic manifestations [1] , [2] . DENV infection can cause subclinical disease or overt illness ranging from mild dengue fever to life threatening dengue hemorrhagic fever ( DHF ) or dengue shock syndrome ( DSS ) . Infection with one DENV serotype usually provides life-long immunity to the same serotype , but it does not elicit long-term cross-protective immunity against other serotypes . People with a previous DENV infection have a higher risk of developing more severe disease in a subsequent infection with a different serotype of DENV [3] . It has been shown that non-neutralizing cross-reactive antibodies bound to DENV enhances viral entry into Fcγ-receptor-bearing cells , resulting in increased virus load and/or production of certain cytokines that can increase viral infection and pathogenesis [4] . Therefore , the ideal DENV vaccine should provide simultaneous protection against all four serotypes of DENV . We have developed a tetravalent recombinant live-attenuated dengue vaccine ( DENVax ) that is currently being tested in Phase 1 and Phase 2 human clinical trials . This vaccine consists of 4 serotypes of recombinant vaccine viruses , designated as DENVax-1 , -2 , -3 , and -4 . The preclinical research-grade viruses were developed using reverse genetic technology at the Division of Vector-Borne Diseases ( DVBD ) , Centers for Disease Control and Prevention ( CDC ) [5]–[7] , while the commercial-grade DENVax vaccine viruses were re-derived by Inviragen Inc . in collaboration with the CDC [7] , in a cGMP facility at Shantha Biotechnics Ltd . , Hyderabad . The infectious clone-derived DENVax-2 is based upon an attenuated DENV-2 strain , DEN-2 PDK-53 [8]–[11] . The DENVax-1 , -3 and -4 are chimeric , recombinant viruses that were originally designated as D2/1-V , D2/3-V , and D2/4-V ( research-grade ) [6] and express DENV-1 , -3 , and -4 serotype-specific prM and E viral proteins , respectively , in the common genetic background of the attenuated DENVax-2 . The major attenuation loci , nucleotide 5′NCR-57-T , NS1-53-Asp , and NS3-250-Val , of the DENV-2 PDK-53 vaccine have been previously determined , and all of them are shared by the common PDK-53 virus-specific genetic background of the four DENVax viruses [6] , [12] . The genetic sequence of the three attenuation loci as well as the previously established in vitro and in vivo attenuation phenotypes of these vaccine candidates were carefully monitored for the cGMP-manufactured DENVax seeds . This report describes the strategies used to generate the master virus seeds ( MVS ) as well as their genetic and phenotypic characterization . These MVS can be used to manufacture clinical and ultimately commercial vaccine materials .
All animal experiments were conducted in accordance with the “Public Health Service Policy on Humane Care and Use of Laboratory Animals” by NIH , “Animal Welfare Act and Amendments” by USDA , “Guide for the Care and Use of Laboratory Animals” by National Research Council ( NRC ) , “Occupational Health and Safety in Care and Use of Research Animals” by NRC , and “Biosafety in Microbiology and Biomedical Laboratories” by CDC . The animal experimental protocol was approved by the DVBD/CDC Institutional Animal Care and Use Committee . DENV-1 16007 , DENV-2 16681 , DENV-3 16562 , and DENV-4 1034 served as wild type ( wt ) DENV controls , and they were the parental genotype viruses for the DENVax viruses . DENVax progenitor research-grade viruses , designated as D2/1-V , D2 PDK-53-VV45R , D2/3-V , and D2/4-V , were prepared previously [6] , [12] . Vero ( African green monkey kidney ) cells used for making the cell banks for vaccine production originated from the American Type Culture Collection CCL81 cell line that has been characterized by the World Health Organization ( WHO ) for manufacturing vaccines ( WCB-Vero cells ) . The engineered DENV infectious cDNA clones , pD2-PDK-53-VV45R , pD2/1-V , pD2/4-V , and in vitro-ligated pD2/3-V , containing the full genome-length viral cDNAs , were used to make fresh viral RNA transcripts by in vitro transcription as described previously [6] , [12] . The RNA transcripts were treated with DNase I followed by low-pH phenol/chloroform extraction and ethanol precipitation to remove the template cDNA and proteins . Each sample was estimated to yield 2–4 ug of genome-length viral RNA that could produce over 5 log10 pfu/ml of the viruses after transfecting into 4×107 Vero cells by electroporation using the Gene Pulser Xcell total system ( BioRad Laboratories ) . Transfected cells were cultured in 30 ml of cell growth medium ( MEM with 10% FBS ) , and incubated at 36°C±1°C , 5% CO2 for 6 to 11 days . These passage 1 ( P1 ) virus seeds were harvested , clarified by centrifugation , stabilized , and stored in small aliquots below −60°C . The P1 virus seeds were used to propagate DENVax pre-master and master virus seed ( MVS ) lots through a strategy designed to ensure the optimal genetic stability and safety of the manufactured vaccines . This strategy included three serial plaque purifications , as well as genetic analyses of viruses to select the optimal clonal virus for continued seed production ( Table 1 ) . Briefly , the P1 seeds were amplified once in Vero cells at a MOI of 0 . 001 to generate the P2 seeds . The P2 seed stocks were evaluated by plaque morphology and complete viral genomic sequencing . The genetically confirmed P2 stocks were plated on Vero cells with overlay medium as described in the plaque assay below to generate well-isolated plaques , and six individual plaques from each serotype of DENVax were isolated ( plaque clone A–F ) and mixed into 0 . 5 ml of culture medium ( P3 ) . Each plaque suspension was subjected to two additional rounds of plaque purification , resulting in twice- and thrice-plaque purified virus seeds at passages P4 and P5 , respectively . The P5 viruses were amplified through two sequential Vero passages to produce P7 seed stocks . Genetic analysis of the three major DENVax attenuation loci [6] , [12] and plaque phenotype analysis were conducted to screen all 24 P7 seeds . Seeds possessing appropriate initial characteristics were further characterized by full genomic sequencing . Based on the presence of the 3 major DENV-2 PDK-53 attenuating loci , minimal genomic sequence alterations , and expected plaque phenotype , one P7 clone of each serotype was selected to be the pre-master seed . The MVS ( P8 ) was then generated by a one-passage amplification of the pre-master seed at MOI of 0 . 001 in multiple 175 cm2 flasks of Vero cells . The MVS stocks were harvested at 6–10 days post infection ( pi ) , clarified by centrifugation , stabilized by the addition of serum or F127 ( a pluronic block copolymer ) , trehalose and human albumin [13] , and stored as 1-ml aliquots below −60°C . Virus titers and plaque sizes were measured by plaque assay using Vero cells as previously described [5] . For accurate comparison , plaque sizes of all viruses were measured and compared in the same experiment . After visualization with neutral red on day 9 pi , up to 10 well isolated plaques for each virus were measured for mean plaque size calculation . Fewer plaques were measured for wt DENV-1 , -3 , and -4 , whose larger plaque sizes often did not permit measurement of 10 well-separated plaques . Viral RNA was extracted from DENVax seeds and used for cDNA amplification as previously described [5] , [6] . Automatic sequencing of the cDNA was conducted on the 3130xl Genetic analyzer ( Applied Biosystems ) , and results were analyzed using Lasergene SeqMan software ( DNAStar , Inc ) . Taqman-based mismatch amplification mutation assay ( TaqMAMA ) , a quantitative single nucleotide polymorphism assay , was previously developed to permit finer assessment of the level of reversion at the 5′NC-57 locus of attenuation [14] , and was further optimized for this study . The specific forward primers used to detect DENV-2 wt and vaccine sequences were D2-41-GC and D2-40-TT , respectively . Triplicate reactions for each wt- and vaccine-specific assay were conducted for each sample , and specificity of the assay was confirmed by testing each RNA standard with the heterologous genotype primer/probe sets to ensure minimum cross-reactivity in every experiment . The real time RT-PCR was performed with the iQ5 or CFX-95 system ( BioRad ) , using a BioRad iScript RT-PCR ( for probes ) kit . The results were reported as the percentage of viral genomes showing reversion . Previously , due to higher cross-reactive backgrounds that limited the input RNA levels for this assay , the detection sensitivity was at 0 . 1% reversion ( discrimination power ) [14] . In this study , the assay has been further optimized using improved equipment and reaction kits , and the cross-reactive background was decreased considerably at much higher levels ( 7–8 log10 copies ) of RNA template input . This optimization resulted in significant improvement of the detection sensitivity , down to 0 . 01–0 . 07% reversion . The replication phenotypes of the DENVax MVS stocks were compared with their parental wt DENVs in C6/36 mosquito cells ( Aedes albopictus ) and Vero cells grown in 6-well plates . Cells were infected at a MOI of 0 . 001 and incubated with DMEM medium containing 2% FBS in a 5% CO2 incubator at 28°C ( C6/36 ) , or 37°C and 39°C ( Vero ) . Aliquots of the culture supernatant were collected for each virus on day 5 ( Vero ) or 6 ( C6/36 ) pi , mixed with an equal volume of medium containing 40% FBS , and stored at −80°C until processed by plaque titration . All samples were tested in duplicate for each experiment . Aedes aegypti mosquitoes used for this study were from a colony established in 2002 from a village near Mae Sot ( 16′N , 33′E ) , Thailand . Five-to-seven day old female mosquitoes were used for infectious blood meal feeding or intrathoracic ( IT ) inoculations . Aliquots of freshly cultured DENVax and wt DENV were used immediately upon harvest ( without any freeze-thaw cycle ) to make virus blood meals , and the remaining virus suspensions were supplemented with 20% FBS and stored at −80°C for plaque titration and IT inoculation . The DENVax seeds for these experiments were prepared from one passage of the pre-master seeds in Vero cells , and were considered DENVax MVS equivalents . Infectious blood meals were prepared by mixing fresh virus at a ratio of 1∶1 with defribrinated chicken blood ( Colorado Serum Company ) . Mosquitoes were sugar-starved overnight and then offered the blood meal for 1 hour using a Hemotek membrane feeding system ( Discovery Workshops ) . A 50-µl aliquot of the blood meal was retained at −80°C for back-titration of virus doses . Fully-engorged females were sorted under cold anesthesia and placed into cartons with 10% sucrose solution provided ad libitum at 28°C with a photoperiod of 16∶8 h ( light∶dark ) . After 14 days , 25–30 mosquitoes from each group were anesthetized by triethylamine ( Carolina Biological Supply Company ) exposure , and one hind leg was removed and placed in 0 . 5 ml of DMEM with 10% FBS and penicillin/streptomycin ( 100 U/ml and 100 µg/ml respectively ) . Saliva was collected by inserting the proboscis of the mosquito into a capillary tube containing 2 . 5% FBS and 25% sucrose solution . Mosquitoes were allowed to salivate for at least 15 minutes and then capillary tubes and bodies were placed into separate tubes containing DMEM . For IT inoculation , mosquitoes were cold-anesthetized and inoculated with approximately 50 pfu of virus in 0 . 34 µl inoculum . Inoculated mosquitoes were kept for 7 days in the same conditions as described above . Mosquitoes were then anesthetized , and their saliva and bodies were collected as described above . All collected samples were stored at −80°C until they were further processed by plaque titration . Body and leg samples were homogenized with copper coated BBs ( Crossman Corporation , NY ) at 24 cycles/second for 4 min using a mixer mill , and then clarified by centrifugation . Saliva tubes were centrifuged at 3 , 000× g for 3 minutes to expel fluid from capillary tubes . Virus plaque titration results from bodies , legs , and saliva were used for determining the infection , dissemination , and transmission rates , respectively . Timed pregnant female ICR mice were obtained from Taconic Labs , and monitored for birth of pup litters . Approximately 12–24 hours after birth , two litters of eight pups per virus ( n = 16 ) in each experiment , were challenged with 103 to 104 pfu of virus in 20 µl of diluent by intracranial ( ic ) inoculation using a 30-gauge needle . Animals were monitored at least 3 times daily for at least 32 days following challenge . At the first sign of illness ( rough fur , hunched back , weight loss , abnormal movement , paralysis , or lethargy ) animals were euthanized .
DENVax viruses were re-derived under cGMP condition by transfection of viral RNA transcribed from the full-length recombinant cDNA into production-certified Vero cells , resulting in P1 virus seeds . The four P1 viruses were then amplified to obtain P2 seeds , which were subjected to full-length genome sequencing . Results showed that each of the four serotypes of P2 viruses was genetically identical to its homologous progenitor , research-grade candidate vaccine virus [6] . The original plaque phenotypes were also retained in the P2 viruses . Six plaque-purified viruses ( A–F clones ) were isolated for each serotype of DENVax from the P2 seeds , and each isolated plaque was directly plaque purified two more times . The third plaque purification ( P5 ) of each virus was amplified twice in Vero cells to produce the potential pre-master P7 DENVax seeds ( Table 1 ) . Genome sequences and plaque phenotypes of the P7 seeds were analyzed and compared to the P2 seeds ( Table 2 ) . Plaque phenotypes of the P7 viruses were generally similar to those of the P2 seeds . One virus ( DENVax-2 C ) had somewhat larger than expected plaques , 2 viruses were smaller ( DEVax-4 E and F ) , and 4 viruses ( DENVax-1 C and D , DENVax-4 B and D ) had slightly clearer plaques ( Table 2 ) . Virus titers reached over 6 . 0 log10 pfu/ml for most of the P7 seeds , except for 5 viruses . We determined previously that two ( NS1-53 and NS3-250 ) of the three major attenuation determinants of the DENV-2 PDK-53 genetic vector are extremely stable . Genome sequencing of more than 60 candidate vaccine virus seeds after 10 or more serial passages in Vero cells identified no reversion event at these 2 loci [[14] and unpublished data] . All sequence chromatograms generated from both forward and reverse sequencing for these two sites were very clean without any minor nucleotide populations evident at the NS1-53 and NS3-250 genetic loci . In contrast to the NS1 and NS3 sites , different levels of reversions at the 5′NCR-57 attenuation locus were previously identified from multiple serially passaged research-grade vaccine viruses [14] , suggesting this locus might not be as stable as NS1 and NS3 after multiple passages in cell culture . Therefore , a sensitive TaqMAMA was developed to accurately measure the reversion rate at the 5′NCR-57 locus by real-time RT-PCR [14] . Depending on the concentration of the input viral RNA for each virus in the assay , the sensitivity of the assay ranged between 0 . 01% and 0 . 07% reversion , which is much more sensitive than the 10–30% reversion sensitivity limit detectable by consensus genome sequence analysis . We found 15 of the 24 viruses had minimal or undetectable reversion ( <0 . 07% ) , 1 virus ( DENVax-3 D ) had almost 100% reversion , and 8 viruses ( 1 DENVax-1 , 1 DENVax-2 , 2 DENVax-3 , and 4 DENVax-4 ) had partial reversion ranging from 0 . 08% to 12 . 85% ( Table 2 ) . Full-length genome sequencing was conducted for 16 of the 24 P7 viruses with low levels of 5′NCR57 reversion as measured by TaqMAMA . All of the sequenced viruses maintained the other 2 DENVax attenuation determinants at NS1-53 and NS3-250 , and all had acquired additional mutations that were not present in the original , engineered recombinant cDNA clones ( Table 2 ) . DENVax-1 A , DENVax-2 F , DENVax-3 F , or DENVax-4 F was selected as the most optimal pre-master seed for each serotype because their genotypes and plaque phenotypes most closely resembled those of the originally designed vaccine viruses [6] . These pre-master seeds were further amplified to generate the MVS ( Table 1 ) . Full-length genome sequencing revealed that the MVS for DENVax-1 , DENVax-2 , and DENVax-3 were identical to their respective pre-master seeds ( Table 2 and 3 ) . The DENVax-4 MVS acquired an additional amino acid mutation , at locus NS2A-K99K/R ( from K to K/R mixed genotype ) during production of the MVS . TaqMAMA results demonstrated that the 5′NCR-57 reversion rate was minimal or undetectable in all 4 MVS lots ( Table 3 ) . Plaque phenotypes of the DENVax MVS were compared with wt DENVs and their homologous research-grade chimeric viruses in Vero cells ( Fig . 1 ) . All of the MVS of DENVax-1 , -2 , and -3 produced plaques that were significantly smaller than their wt homologs and very similar ( within 0 . 4-mm differences ) to their homologous research-grade viruses in Vero cells . DENVax-4 MVS was also significantly smaller than the wt DENV-4 , but was slightly larger ( 0 . 9 mm difference ) than the original research-grade D2/4-V chimera . Temperature sensitivity was tested in Vero cells for the DENVax MVS and compared with their homologous wt and the original research-grade chimeric vaccine virus . The wt DENV-3 16562 was not temperature sensitive . The wt DENV-1 and DENV-4 were moderately temperature sensitive at 39°C ( titers were approximately 1 . 0 log10 pfu/ml lower at 39°C than at 37°C , Fig . 2 ) . Wt DENV-2 16681 was the most temperature sensitive of the wt DENVs tested , and resulted in a 100-fold titer drop at 39°C . DENVax-1 , -2 , and -3 were as temperature sensitive as their original homologous research-grade chimeric vaccine viruses ( Fig . 2 ) . Titers at 39°C dropped between 2 . 0 and 3 . 0 log10 pfu/ml for these DENVax strains . DENVax-4 was also temperature sensitive , demonstrating a 5-fold reduction in titer . However , the original research-grade D2/4-V showed about 10-fold reduction in titer . The final stabilized DENVax-4 MVS contained pluronic block copolymer F127 , and we have previously shown that this copolymer can enhance thermal stability of the DENVs [13] . The presence of the copolymer F127 in DENVax-4 MVS likely contributed to the less pronounced temperature sensitivity of the virus in the Vero culture assay . In a separate experiment , we further evaluated the temperature sensitivity of an MVS-derived DENVax-4 strain in the absence of F127 . To remove the F127 from the strain , viral RNA was isolated from a DENVax-4 bulk virus preparation and transfected into Vero cells . This DENVax-4 virus appeared to be as temperature sensitive as the D2/4 V ( titer reduced 1 . 5 log10 pfu/ml ) on day 3 pi in the absence of F127 ( Fig . 2 ) . Previous studies showed that the research-grade chimeric vaccine viruses retained the attenuation phenotype of the backbone DENV-2 PDK53 virus in C6/36 cells [6] , [15] . The DENVax MVS were grown in C6/36 cells to verify their retention of this in vitro attenuation phenotype . Compared to the wt DENVs , DENVax-1 , DENVax-2 , and DENVax-4 MVS showed marked growth reduction ( at least 3 log10 pfu/ml reduction ) in C6/36 cells on day 6 pi ( Fig . 3 ) . The DENVax-3 MVS also exhibited reduced growth compared to the wt DENV-3 16562 , but the reduction was not as marked ( 1–2 log10 pfu/ml reduction ) . However , the titer of the DENVax-3 was similar ( within 1 log10 pfu/ml difference ) to the titer of the research-grade chimeric D2/3-V vaccine virus . Oral infection experiments were conducted in Ae . aegypti mosquitoes to evaluate the ability of DENVax to infect midgut ( overcome midgut infection barrier; MIB ) , disseminate to secondary tissues ( overcome midgut escape barrier; MEB ) , and be transmittable ( release virus to saliva ) . Infectious blood meals were back-titrated to measure the virus titers and only the experiments with similar virus titers in the blood meal ( less than 1 log10 pfu/ml differences ) between parental DENV and DENVax for each serotype were included for comparisons in Table 4 . DENVax-1 , DENVax-2 , and research-grade D2 PDK-53-VV45R did not infect mosquitoes through oral feeding , which was significantly different ( p<0 . 0001 ) from their parental viruses , DENV-1 16007 ( 44% infection ) and DENV-2 16681 ( 43 . 3% infection ) . Since no mosquito was infected by DENVax-1 and -2 , there was no dissemination concern for these two vaccine viruses . While DENVax-4 did infect 2 of the 55 mosquitoes through oral feeding , the infection rate was significantly lower ( p<0 . 05 ) than the wt DENV-4 1036 . DENVax-3 did not infect any mosquitoes in two experiments with blood meal viral titers of 5 . 2±0 . 02 log10 pfu/ml ( Table 4 ) , and in a separate experiment with blood meal viral titer of 6 . 0 log10 pfu/ml , only 1 out of 30 mosquitoes became infected ( data not shown in Table 4 ) . However , wt DENV-3 16562 also had a very low infection rate ( 8% ) at 5 . 2 log10 pfu/ml , and the rate did not increase in a separate experiment with a higher blood meal viral titer at 6 . 2 log10 pfu/ml ( 3% , 1 positive out of 30 mosquitoes ) . Although the wt DENV-3 and DENV-4 had significantly lower infection rates than the wt DENV-1 and DENV-2 , the mean virus titers in the infected mosquitoes were similar ( 3 . 1 to 3 . 9 log10 pfu/mosquito ) . In contrast , the DENVax-4 titers from the two infected mosquitoes were both minimal ( 0 . 7 log10 pfu/mosquito ) , which was 1 , 000-fold lower than the titer from the mosquitoes infected by wt DENV-4 1036 ( 3 . 9±1 . 5 log10 pfu/mosquito ) . For those mosquitoes that were infected , dissemination out of the midgut could be assessed by determining whether virus was present in the legs . The four wt DENVs resulted in dissemination rates ranging between 36 . 3% and 62 . 5% , and their mean virus titers ( in log10 pfu ) from the legs were between 0 . 9±0 . 3 and 2 . 2±0 . 7 ( excluding negative samples ) . Neither of the two DENVax-4 infected mosquitoes resulted in virus dissemination to the legs ( Table 4 ) . Since the oral feeding results showed minimal or no detectable midgut infection and body dissemination by DENVax , we could not evaluate their transmission potentials from this experiment . In addition , while disseminated virus was detectable in the legs , none of the four wt DENVs was detectable in saliva of orally infected mosquitoes ( not shown in Table 4 ) which suggested that the oral feeding experiment may not be sufficiently sensitive to measure the transmission rate of these DENVs . Therefore , highly stringent artificial infections by IT inoculation were performed to directly infect the mosquito body bypassing both MIB and MEB . Except for DENVax-4 , all viruses ( wt and DENVax ) achieved 100% infection of the IT inoculated Ae . aegypti . The DENVax-4 inoculum had a slightly lower viral titer than the other three viral inocula , but it still successfully infected 70% of the inoculated mosquitoes ( Table 4 ) . Despite the high body infection rates achieved by IT inoculation , all four DENVax viruses exhibited significantly lower ( p<0 . 005 ) or non-detectable transmission rates ( 0–10% ) compared to the wt DENVs ( 43–87% , Table 4 ) . The mean expectorate virus titers calculated from saliva ( positive samples only ) were slightly higher from wt DENV infected mosquitoes ( 0 . 92±0 . 75 log10 pfu , 0 . 88±0 . 31 , 0 . 96±0 . 60 , and 0 . 86±0 . 63 , for wt DENV-1 to -4 , respectively ) than those infected with DENVax ( 0 . 40±0 . 00 , 0 . 55±0 . 21 , 0 . 88 from only 1 mosquito , and none detectable , for DENVax-1 to -4 , respectively ) . Overall , the DENVax viruses showed no or limited infection and dissemination after oral feeding , and the highly stringent IT results affirmed the minimal transmission capacity of these DENVax viruses in Ae . aegypti . The research-grade vaccine viruses were previously shown to be highly attenuated for neurovirulence in newborn ICR mice maintained in-house at DVBD/CDC [5] , [6] . All of these mice survived ic challenge with 104 pfu of each vaccine virus . The wt DENV-2 16681 virus , on the other hand , resulted in 62 . 5%–100% mortality in these CDC-ICR mice in various experiments . Prior to the current study the in-house ICR mouse colony was eliminated at DVBD/CDC . Therefore , commercial ICR mice obtained from Taconic Labs ( Taconic-ICR ) were used for this study . We observed that newborn Taconic-ICR mice were significantly more susceptible to DENV-2 infection than the previous CDC-ICR mice . Figure 4A summarizes the neurovirulence of wt DENV-2 16681 virus in CDC-ICR colony ( previous published and unpublished experiments ) and Taconic-ICR newborn mice challenged ic with 104 pfu of virus . Clearly , the Taconic-ICR mice ( 100% mortality in 32 mice , average survival time of 8 . 3±0 . 5 days ) were much more susceptible to ic DENV-2 16681 challenge than the previous CDC-ICR mice ( 91% mortality in 72 mice , average survival time of 14 . 6±2 . 3 days ) . To evaluate neurovirulence of the DENVax MVS , the Taconic-ICR mice initially were challenged ic with a dose of approximately 104 pfu of wt DENV-2 16681 , D2 PDK-53 VV45R , D2/3-V , or DENVax 1–4 virus in one ( n = 16 ) or two ( n = 31–32 ) experiments ( Fig . 4B ) . At this dose , D2/3-V research-grade virus , as well as DENVax-1 , and DENVax-3 MVS exhibited fully attenuated neurovirulence phenotypes ( no illness or mortality ) . As expected , wt DENV-2 was uniformly fatal , with average mouse survival time ( AST ) of 8 . 8±0 . 4 days . In these DENV-2-sensitive Taconic-ICR mice , the D2 PDK-53-VV45R research-grade virus resulted in 81 . 3% mortality with AST of 11 . 1±1 . 1 days . The DENVax-2 MVS and DENVax-4 MVS were uniformly fatal in the Taconic-ICR , showing AST values of 11 . 7±1 . 1 and 11 . 5±0 . 9 days , respectively . The mortality and AST between the D2 PDK-53-VV45R and DENVax-2 at this dose were not significantly different ( p>0 . 05 , student t-test ) . The research-grade D2/4-V was not tested at this dose for comparison with DENVax-4 . In an attempt to identify a more discriminating ic challenge dose for those viruses that exhibited neurovirulence in the Taconic-ICR mice , we compared the neurovirulence of wt DENV-2 16681 virus with that of D2 PDK-53-VV45R , DENVax-2 MVS and DENVax-4 MVS , as well as D2/4-V research-grade virus , at a 10-fold lower dose ( 103 pfu , Fig . 4C ) . The wt DENV-2 retained a uniformly fatal neurovirulent phenotype , with AST of 9 . 0±1 . 4 days , at this lower challenge dose . The other 4 viruses exhibited intermediate neurovirulence phenotypes , and the degree of neurovirulence was serotype-specific . The D2 PDK-53-VV45R virus and its DENVax-2 MVS cognate showed significant attenuation ( 32 . 3% survival with AST of 13 . 8±3 . 6 days and 31 . 2% survival with AST of 11 . 5±3 . 4 days , respectively ) and the differences were not significant ( p = 0 . 4 ) . Both the DENVax-4 MVS and the research-grade D2/4-V virus were highly attenuated for neurovirulence ( 81 . 3% survival with AST of 19 . 4±5 . 8 days and 100% survival , respectively ) . The slightly higher virulence of the DENVax-4 relative to D2/4-V was likely due to the higher virus dose that was inoculated into the mice receiving DENVax-4 ( 3 . 4 log10 pfu ) versus the dose in the D2/4-V group ( 2 . 7 log10 pfu ) , based on the back titration results of the inocula . Overall , our results indicated that MVS of DENVax-1 and -3 exhibited complete attenuation of neurovirulence , while DENVax-2 and -4 MVS lots retained attenuation phenotypes that closely resembled their homologous research-grade vaccine candidates .
This report describes the process of re-deriving and preparing the DENVax manufacturing seed strains under cGMP conditions , and the genetic and phenotypic characterizations of the MVS . Serial plaque purifications and full-genome sequence analyses were incorporated into the manufacturing process to ensure production of vaccine seeds with optimal safety and genetic stability . All four of the chosen pre-master seeds had undetectable reversion at any of the 3 attenuation loci , contained one or two new amino acid substitutions in the viral translated proteins , and retained the small plaque phenotypes of the previous research-grade vaccine viruses . All of the DENVax viruses also were tested for identity , sterility , and lack of detectable adventitious agents as part of the manufacturing product release ( data not shown ) . Full-genome sequence analysis revealed that an additional amino acid mutation evolved in the DENVax-4 MVS versus its pre-master seed , while the other three DENVax MVS lots retained the consensus genome sequence of their pre-master seeds . Overall , from derivation of the P1 seeds to the pre-master ( P7 ) seeds , only 1 or 2 non-synonymous mutations occurred in a given seed . From P1 to MVS ( P8 ) seeds , 2 to7 nucleotide substitutions were identified in any given DENVax seed and only 2 to 3 of these substitutions resulted in amino acid changes . RNA viruses are error-prone in their genome replication , so genetic substitutions in the flavivirus genome during cell passages are not unexpected . None of the silent mutations in the MVS were within the 5′ or 3′NCR that may affect virus replication . Most of the amino acid substitutions were very conservative ( similar in residue size , pKa , and chemistry structure ) . Only the change in prM-K52E of the DENVax-2 , and the substitution in NS2A-D66G of DENVax-4 were non-conservative changes . The NS2A-D66G mutation of the DENVax-4 resides in the nonstructural gene region of the shared DENV-2 PDK-53 genetic background . Although the NS2A-66 locus is usually D among various strains of DENV-2 , interestingly it is usually G for DENV-4 . It is possible that the D to G change in the DENVax-4 is important for fitness of the DENVax-4 in Vero cells . The DENVax-2 prM-K52E mutation resides in the C-terminal portion of the prM that is cleaved out from the mature virus particles . Overall , our phenotypic characterization results confirmed that none of the mutations in the MVS seeds significantly altered the attenuation phenotypes of the candidate vaccine viruses . Our results suggested that the DENVax viruses are genetically stable during the manufacturing process . The highly sensitive TaqMAMA of the 5′NCR-57 locus demonstrated minimal or undetectable reversion in the MVS of each DENVax serotype . The 5′NCR-57 reversion rates of the DENVax MVS ( P8 ) were significantly lower than the 5′NCR-57 reversion rates that evolved in research-grade vaccine candidates after 8-serial passages in Vero cells ( 6–45% reversion ) [14] . The retention of the 3 attenuation loci were also confirmed in the P10 seeds that were passaged twice from the MVS ( data not shown ) . Thus , the strategy for large-scale manufacturing of the DENVax seeds presented in this report resulted in genetically stable vaccine seeds which retained the expected markers of attenuation . Complete evaluation of the phenotypic markers of viral attenuation , including small plaque phenotype , temperature sensitivity , reduced replication in mosquito cells , reduced infection/dissemination/transmission by mosquitoes , and attenuation in newborn ICR mice , were assessed for the MVS stocks . Although we cannot totally rule out that the slight phenotype variations between the DENVax and original research vaccine viruses were caused by the additional mutations that evolved in DENVax MVS stocks , none of the variations was very significant and all of the DENVax were still considerably attenuated relative to their wt counterparts . It will be important to analyze the clinical study outcomes in the context of genome sequences of clinical vaccine lots , and perhaps , viruses isolated from vaccinees to fully evaluate the safety and genetic stability of DENVax in humans . Vector competence is an important safety component for live-attenuated flavivirus vaccine viruses [16]–[21] . We have previously tested the research-grade DENV-2 PDK-53-VV45R virus and wt reversion mutant derivatives in Ae . aegypti , and found that the NS1-53-Asp attenuating mutation was the dominant determinant for impaired replication in the mosquito [22] . The other two major attenuation loci of the DENV-2 PDK-53 vaccine , nucleotide 5′NCR-57-T and NS3-250-Val , also exhibited some inhibiting effect on replication in mosquitoes , thus providing additional , redundant restrictions for mosquito vector competence . This study is the first to report the mosquito oral and IT infection and replication for all four DENVax strains . DENVax-1 , -2 , and -3 did not infect any Ae . aegypti mosquitoes through oral infection ( Table 4 ) . The DENVax-4 infected only 3 . 6% of orally exposed mosquitoes , a level significantly lower than that of the wt DENV-4 with a replicative mean titer in the mosquito bodies lower than that of wt DENV-4 infected mosquitoes . Most importantly , no DENVax-4 was detected in the legs of the infected mosquitoes , suggesting that DENVax-4 was not able to escape from the midgut barriers following oral infection . The infection rates for the DENVax-1 , -2 , and -4 were all significantly less than their wt counterparts , but the difference was not significant between DENVax-3 and wt DENV-3 16562 due to the very low infection rates for both viruses . Compared to other wt strains of DENV assessed in Ae . aegypti collected from the same Mae Sot Province , Thailand [21] , the parental wt DENV strains used for engineering our DENVax appeared to have lower infection and dissemination rates following oral exposure . The wt DENV-1 PUO359 , DENV-2 PUO218 , DENV-3 PaH881/88 , and DENV-4 1288 used for engineering the yellow fever virus ( YFV ) 17D vaccine-based ChimeriVax-DEN vaccines had infection rates ranging 47–77% . In contrast , the YFV 17D vaccine cannot infect Ae . aegypti [21] . Although the ChimeriVax strains contained the prM-E from these highly infectious wt DENV , the ChimeriVax retain the mosquito attenuation phenotype of their YFV 17D replicative backbone [21] . Our results also indicated that the mosquito attenuation phenotype of the DENV-2 PDK-53 backbone was maintained in the DENVax strains . In addition , using the wt DENV strains with lower mosquito-infectivity in our constructs provides an additional safety feature for DENVax . The oral infection results showed clearly that the DENVax had minimum mosquito infectivity and dissemination capacity . In addition , we performed the more sensitive IT inoculation to infect the mosquito body directly . This allowed us to establish a model to evaluate the potential of transmission if , by rare chance , any of the DENVax were able to overcome both MIB and MEB . Our results demonstrated that all four DENVax viruses had non-detectable or minimal mosquito transmission potential compared to their wt counterparts even with such an artificially high infection rate . DENVax transmission could only theoretically occur if ( 1 ) a vector feeds on a vaccinee having a sufficient viremic titer to infect mosquito midgut epithelium; ( 2 ) the virus is capable of replicating in these cells and subsequent dissemination out of the midgut; and ( 3 ) the disseminated virus can replicate in salivary gland and expectorate sufficient virus in saliva for transmission . The threshold of human viremia required to infect mosquitoes has not been established adequately , but human viremia can be 106–108 mosquito infectious dose50 ( MID50 ) /ml after natural wt DENV infection [23] , [24] . This MID50 was based on direct IT inoculation of mosquitoes with diluted human plasma . A previous study analyzing DENVax in nonhuman primates ( NHP ) indicated that viremic titers following DENVax immunization were very low ( less than 2 . 4 log10 pfu/ml ) and lasted for 2–7 days [25] . Given the low viremia levels and the low mosquito infection , dissemination , and transmission capacity of DENVax , it is unlikely that these vaccine viruses could be transmitted by mosquitoes in nature . Unlike some other flaviviruses , such as Japanese encephalitis virus and YFV that can cause encephalitis in humans and show neuropathic effects in the NHP neurovirulent test , wt DENVs are generally not neurotropic in humans and the NHP [26] , [27] . Previous neurovirulence studies of DENVs , including the wt DENV-1 16007 , DENV-2 16681 , DENV-3 16562 , and DENV-4 1036 that were the parental viruses for engineering the 4 DENVax strains , showed that they caused only minimal neuropathophysiology after intracerebral injection into NHP [26] , [28]–[31] . In addition , the DENVax seeds were not generated in a manner that could enhance neurovirulence; thereby they would not be expected to be neuropathologic in a NHP test . Infant mice have been demonstrated to be an accurate surrogate for the neurovirulence test of DENV vaccines in NHP previously [32] , and the newborn mouse neurovirulence model reported in this study provided further support that certain young suckling mice can be a sensitive model for DENV vaccine evaluation . In summary , we have successfully developed a unique manufacturing strategy to optimize the genetic stability and safety of the manufactured DENVax MVS . Since the main attenuation loci of the DEN-2 PDK-53-based DENVax have been well characterized previously [6] , [12] , we were able to integrate genome sequence and the TaqMAMA to identify optimal pre-master seeds for preparation of the MVS . Our results highlight the advantage of employing strategically designed live-attenuated vaccines with well-defined molecular attenuation determinants in predicting vaccine safety . In addition to the attenuation characteristics analyzed and described in this report , we have tested and confirmed the safety and immunogenicity of the DENVax MVS in AG129 mice [33] and NHP [25] . The present study described here and the ongoing clinical trial studies of DENVax will provide critical information to further evaluate the safety and efficacy of tetravalent DENVax as a vaccine to protect humans against dengue .
|
Transmitted by Aedes spp . mosquitoes found worldwide , dengue is the most important mosquito-borne viral disease in the world . The incidence of dengue has increased 30-fold over the past 50 years , and is now endemic in over 100 countries . Vaccination is believed to be one of the most effective strategies in dengue prevention . However , no vaccine is currently available , and prevention strategies to control mosquitoes in endemic areas have been insufficient in controlling dengue . We have developed a recombinant live-attenuated tetravalent vaccine against all four serotypes of dengue virus . This candidate vaccine is currently under human clinical evaluation . In this report , we provide information regarding our manufacturing strategy , and present details of the genetic and biological characterization of the master seed virus for each vaccine serotype . The study described here , our previously reported and ongoing pre-clinical studies , and current clinical trials will provide critical information to evaluate the safety and efficacy of the vaccine to protect humans against dengue .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"virology",
"viral",
"vaccines",
"biology",
"microbiology"
] |
2013
|
Genetic and Phenotypic Characterization of Manufacturing Seeds for a Tetravalent Dengue Vaccine (DENVax)
|
Iron is essential for a wide range of cellular processes . Here we show that the bZIP-type regulator HapX is indispensable for the transcriptional remodeling required for adaption to iron starvation in the opportunistic fungal pathogen Aspergillus fumigatus . HapX represses iron-dependent and mitochondrial-localized activities including respiration , TCA cycle , amino acid metabolism , iron-sulfur-cluster and heme biosynthesis . In agreement with the impact on mitochondrial metabolism , HapX-deficiency decreases resistance to tetracycline and increases mitochondrial DNA content . Pathways positively affected by HapX include production of the ribotoxin AspF1 and siderophores , which are known virulence determinants . Iron starvation causes a massive remodeling of the amino acid pool and HapX is essential for the coordination of the production of siderophores and their precursor ornithine . Consistent with HapX-function being limited to iron depleted conditions and A . fumigatus facing iron starvation in the host , HapX-deficiency causes significant attenuation of virulence in a murine model of aspergillosis . Taken together , this study demonstrates that HapX-dependent adaption to conditions of iron starvation is crucial for virulence of A . fumigatus .
Iron is an essential nutrient for virtually every organism . The ability to exist in two redox states makes this metal an essential cofactor of proteins involved in numerous major cellular processes including respiration , amino acid metabolism and DNA metabolism . However , excess iron has the ability to generate toxic reactive species that can damage cellular components [1] . Despite its general abundance , the bioavailability of iron is very limited owing to its oxidation into insoluble ferric hydroxides by atmospheric oxygen . Consequently , all organisms have developed tightly regulated homeostatic mechanisms in order to balance uptake , storage and consumption of iron . Moreover , the mammalian immune system utilizes iron-withholding mechanisms to deny invading microorganism's access to free iron [2] , [3] . Consequently , the control over access to iron is one of the central battlefields deciding the fate of an infection . Furthermore , iron starvation activates not only iron uptake but also virulence determinants in many prokaryotic and eukaryotic pathogens . Aspergillus fumigatus is a typical ubiquitous saprophytic mold . Nevertheless , it causes life-threatening invasive disease especially in immuno-compromised patients and has become the most common airborne fungal pathogen of humans [4] . A . fumigatus lacks specific uptake systems for host iron sources as heme , ferritin , or transferrin [5] . However , it employs two high-affinity iron uptake systems , siderophore-assisted iron uptake and reductive iron assimilation , both of which are induced upon iron starvation . Siderophores are low molecular mass , ferric iron-specific chelators [6] . A . fumigatus excretes the siderophores fusarinine C ( FsC ) and triacetylfusarinine C ( TAFC ) to mobilize extracellular iron . Subsequent to chelation of iron , the ferric forms of FsC and TAFC are taken up by specific transporters [7] . For release of iron , the siderophores are intracellularly hydrolyzed [8] and the iron is transferred to the metabolic machinery or stored . A . fumigatus employs also intracellular siderophores: ferricrocin ( FC ) for hyphal storage and distribution of iron , and hydroxyferricrocin ( HFC ) for conidial iron storage [9] , [10] . FsC is a cyclic tripeptide consisting of three N5-cis-anhydromevalonyl-N5-hydroxyornithine residues linked by ester bonds , TAFC is the N2-acetylated FsC , FC is a cyclic hexapeptide with the structure Gly-Ser-Gly- ( N5-acetyl-N5-hydroxyornithine ) 3 and HFC is the hydroxylated FC [6] . The siderophore biosynthetic pathway is shown in Fig . S1 . The first committed step in the biosynthesis of all four siderophores is hydroxylation of ornithine ( Orn ) . Subsequently , the pathways for biosynthesis of TAFC and FC split involving acylation of N5-hydroxyornithine , assembly of siderophore-back bones by nonribosomal peptide synthetases ( NRPS ) , and derivatization by acetylation or hydroxylation . Five A . fumigatus genes encoding respective enzyme activities have been identified [5] , [10]: sidA ( N5-ornithine-monooxygenase ) , sidF ( N5-hydroxyornithine:cis-anhydromevalonyl coenzyme A-N5-transacylase ) , sidC ( FC NRPS ) , sidD ( fusarinine C NRPS ) and sidG ( fusarinine C:acetyl coenzyme A-N2-transacetylase ) . Elimination of both intra- and extracellular siderophores ( ΔsidA mutants ) results in absolute avirulence of A . fumigatus in a mouse model of pulmonary aspergillosis [5] . Deficiency in either extracellular ( ΔsidF or ΔsidD mutants ) or intracellular siderophores ( ΔsidC mutants ) causes partial attenuation of virulence [10] . Recently , siderophores have also been implicated in virulence of Histoplasma capsulatum and various phytopathogenic ascomycetes [11] , [12] , [13] . Consequently , the siderophore system represents an attractive target for antifungal therapy . However , not all fungi produce siderophores; notable examples are Saccharomyces cerevisiae , Candida albicans and Cryptococcus neoformans [6] . In agreement with iron playing an important role in the pathophysiology of A . fumigatus , increased bone marrow iron stores represent an independent risk factor for invasive aspergillosis [14] . Moreover , polymorphonuclear leukocytes inhibit growth of A . fumigatus conidia by lactoferrin-mediated iron depletion [15] , and the human body produces proteins able to sequester fungal siderophores [16] . Consistently , the chelators EDTA and deferasirox enhance the efficacy of amphotericine B in animal models for invasive pulmonary aspergillosis [17] , [18] . In Aspergillus nidulans , maintainance of iron homeostasis is mediated by two transcription factors , SreA and HapX , which are interconnected in a negative feed-back loop: SreA represses expression of hapX during iron sufficiency and HapX represses sreA during iron starvation [19] , [20] . SreA is a DNA-binding GATA-factor whereas HapX functions by protein-protein interaction with the heterotrimeric CCAAT-binding factor . SreA represses iron uptake during iron sufficiency to avoid toxic effects and HapX represses iron-dependent pathways during iron starvation to spare iron . This regulatory circuit is largely conserved in Schizosaccharomyces pombe and orthologs to SreA and HapX are found in most fungal species; a notable exception is the fungal prototype S . cerevisiae , which employs entirely different regulators [6] , [21] . We have previously demonstrated the role of SreA in repression of iron acquisition in A . fumigatus [22] . In this study we characterized the function of HapX and its interplay with SreA . We demonstrate that HapX function is crucial for the metabolic reprogramming required for adaption to iron starvation and for virulence of A . fumigatus .
In A . nidulans , HapX has been shown to repress iron-dependent pathways during iron starvation [20] . The A . fumigatus HapX ortholog displays 70% overall identity and contains all typical features common to this class of transcription factors: an N-terminal 17 amino acid motif , which is essential for interaction with the CCAAT-binding complex , a bZip domain , and three cysteine-rich regions , which are potentially involved in iron-sensing . Genome-wide transcriptional profiling revealed that the transcript level of the A . fumigatus hapX ortholog ( Afu5g03920 ) is SreA-dependently down-regulated in a shift from iron depleted to iron-replete conditions [22] . In agreement , Northern analysis demonstrated up-regulation of the hapX transcript level under steady-state iron depleted compared to iron-replete conditions and partial derepression during iron-replete conditions in a ΔsreA mutant ( Fig . 1A ) . This expression pattern matches that of the A . nidulans ortholog [20] . In order to analyze the function of HapX in A . fumigatus , a deletion mutant ( ΔhapX ) was generated as described in Methods . Consistent with undetectable expression of hapX during iron sufficiency in wt ( Fig . 1A ) , ΔhapX displayed no significant difference to the wt with respect to conidiation and growth rate on solid or liquid media during iron sufficiency ( Fig . 2 ) . In contrast , ΔhapX showed mildly reduced radial growth on solid media ( Fig . 2A ) and was not able to form colonies from single conidia in the presence of the iron chelator bathophenanthroline disulfonate ( BPS ) ( Fig . 2B ) . Furthermore , hapX deletion decreased conidiation to 62% of the wt during iron starvation and 4% during iron starvation in the presence of BPS ( Fig . 2C ) . In iron-starved liquid culture , hapX deletion decreased the biomass production to 58% of the wt ( Fig . 2E ) and caused a reddish pigmentation of the mycelia ( Fig . 2D ) . Reintegration of a functional hapX copy at the hapX locus in the ΔhapX strain , yielding strain ΔhapXC , cured these and all other defects ( Fig 2 and data not shown ) , which demonstrates that the ΔhapX phenotype is a direct result of the loss of HapX activity . Notably , germination of ΔhapX was wt-like under iron-replete and depleted conditions ( data not shown ) demonstrating that the phenotypes of ΔhapX are caused by growth defects . Limitation of nitrogen , carbon , copper , or zinc decreased biomass production of ΔhapX and wt to similar extents ( Fig . 2E ) , which indicates that inactivation of HapX does not result in general sensitivity to starvation but in particular to iron starvation . In line with the growth defect of ΔhapX under iron starvation but not iron sufficiency , expression of hapX was repressed by iron at the transcriptional level , partly dependent on SreA ( Fig . 1A ) . In turn , HapX repressed sreA during iron starvation ( Fig . 1A ) . A similar expression pattern was described previously for the hapX and sreA orthologs of A . nidulans and S . pombe [20] , [21] . Genome-wide transcriptional profiling revealed that expression of hapX is repressed within ≤30 minutes in a shift from iron depleted to iron-replete conditions [22] , which predicts that HapX targets also respond quickly to the availability of iron . This HapX feature allowed analysis of short-term effects of hapX deletion . In order to identify the genes that are negatively affected by HapX at the transcript level , we therefore searched by genome-wide transcriptional profiling for genes fulfilling three criteria: ( i ) up-regulation in a 1h-shift from iron starvation to iron sufficiency in wt ( identification of genes repressed by iron starvation ) , ( ii ) decreased up-regulation in a 1h-shift from iron starvation to iron sufficiency in ΔhapX compared to wt ( identification of genes showing a short-term response to HapX inactivation ) , and ( iii ) up-regulation during steady-state iron starved growth in ΔhapX compared to wt ( identification of genes showing a long-term response to HapX inactivation ) . This strategy is supposed to select for rather direct effects of HapX inactivation . Among the 131 genes negatively affected by HapX ( Table S1 in Supporting Information S1 and Table 1A ) , 34% can be directly assigned to iron-dependent pathways including respiration , TCA cycle , amino acid metabolism , iron-sulfur-cluster biosynthesis , heme biosynthesis , oxidative stress detoxification , biotin synthesis ( Afu6g03670 ) , vacuolar iron storage ( CccA , Afu4g12530 ) , and iron regulation ( SreA , Afu5g11260 ) . This gene set included the orthologs of all five previously identified A . nidulans HapX-repressed genes [20]: cycA ( cytochrome C , respiration , Afu2g13110 ) , acoA ( aconitase , TCA cycle , Afu6g12930 ) , hemA ( α-amino-levulinic acid synthase; heme biosynthesis , Afu4g11400 ) , lysF ( homoaconitase , lysine biosynthesis , Afu5g08890 ) , and sreA ( repressor of iron uptake ) . A representative Northern analysis of cycA is displayed in Fig . 1B . The majority of the cellular iron-consuming pathways , e . g . heme biosynthesis , iron-sulfur cluster biosynthesis , respiration , TCA cycle , is localized in mitochondria , which might explain the co-regulation of mitochondrial components that are not directly iron-dependent , e . g . the mitochondrial processing peptidase ( Afu1g14200 ) , which is essential for import of all mitochondrial matrix proteins . Strikingly , 31% ( n = 41 ) of the genes negatively affected by HapX encode proteins that are localized in mitochondria ( Table S1 in Supporting Information S1 and Table 1A ) , which indicates a significant impact of HapX on mitochondrial metabolism . 23% ( n = 30 ) of the identified genes repressed during iron starvation in a HapX-dependent manner are involved in ribosomal biogenesis and translation ( Table S1 in Supporting Information S1 and Table 1A ) . These data might reflect the iron-dependence of the translation machinery due to the essentiality of iron-sulfur clusters for function of Rli1 ( RNase L inhibitor , Afu1g10310 ) . Because of its fundamental role in translation initiation and ribosome biogenesis , RLI1 is one of the most conserved proteins present in all organisms except eubacteria and it is essential in all organisms tested [23] . Consistent with its iron-dependence , Rli1 expression is repressed during iron starvation in a HapX-dependent manner ( Table S1 in Supporting Information S1 ) . The down-regulation of translation during iron starvation indicates a slow-down of the entire metabolism , which might serve extended cellular survival . Among the 131 A . fumigatus genes negatively affected by HapX , 21 have orthologs in S . pombe ( Table S1 in Supporting Information S1 ) , which are negatively affected by the HapX ortholog Php4 [24] . S . cerevisiae lacks an HapX ortholog and down-regulation of iron-dependent pathways during iron starvation is mediated by the paralogous proteins Cth1 and Cth2 , which promote decay of target mRNA's during iron starvation [25] . A total of 21 of HapX-repressed genes have orthologs in S . cerevisiae , which are repressed during iron starvation via Cth1/2 ( Table S1 in Supporting Information S1 ) . Taken together , all three fungal species repress 15 orthologous genes during iron starvation ( Table S1 in Supporting Information S1 ) . All 15 deduced gene products are involved in iron-dependent pathways including respiration , iron sulfur cluster biosynthesis , TCA cycle , amino acid metabolism and translation and all are localized in mitochondria with exception of Rli1 and the leucine biosynthetic enzyme Leu1 ( Table S1 in Supporting Information S1 ) . These data underscore the evolutionary conservation of iron-sparing in different fungal species . Comparison of the genes negatively affected by HapX with the previously identified SreA regulon [22] displayed no overlap ( Table S1 in Supporting Information S1 ) . However , 38% of these genes were previously found to be up-regulated indirectly by SreA-deficiency; i . e . , in a shift from iron starvation to iron sufficiency these genes were up-regulated in ΔsreA only at late time points . As iron represses expression of hapX at the transcriptional level ( see above ) and most likely also post-translationally , as shown for its orthologs in A . nidulans and S . pombe [20] , [26] , these data suggest that the up-regulation of these genes in ΔsreA cells is caused by inactivation of HapX through the iron overload in ΔsreA . To identify the genes affected positively by HapX , the inverse criteria compared to the screening for HapX-repressed genes by transcriptional profiling were applied ( see above ) : ( i ) down-regulation in a shift from iron starvation to iron sufficiency in wt , ( ii ) decreased down-regulation in a shift from iron starvation to iron sufficiency in ΔhapX compared to wt , and ( iii ) down-regulation during steady-state iron starved growth in ΔhapX compared to wt . Genes affected positively by HapX or its ortholog have been described for neither A . nidulans nor S . pombe yet . However , the transcriptional profiling identified 139 such genes in A . fumigatus , which are mainly involved in siderophore metabolism , amino acid metabolism , protein degradation and uptake , carbohydrate metabolism , and lipid metabolism ( Table S2 in Supporting Information S1 and Table 1B ) . Strikingly , 27% of these genes were previously found to be SreA targets , i . e . repressed during iron sufficiency by SreA [22] , e . g . , genes involved in siderophore metabolism ( Table S2 in Supporting Information S1 and Table 1B ) . As hapX deletion derepressed expression of sreA during iron starvation ( see above ) , hapX deletion might repress SreA-targets indirectly via its transcriptional derepression of sreA . However , hapX deletion affected expression of various SreA-target genes differently ( Table S2 in Supporting Information S1 , Fig . 1C ) . HapX-deficiency drastically reduced the transcript levels of the putative siderophore transporter-encoding mirB and the siderophore-biosynthetic sidG but had only minor effects on the siderophore transporter-encoding mirD and siderophore-biosynthetic sidA and sidF , which indicates SreA-independent effects . In line , 73% of the genes negatively affected by hapX deletion do not appear to be SreA targets ( Table S2 in Supporting Information S1 and Table 1B ) . A prominent example is one of the major allergens of A . fumigatus , the ribotoxin Aspf1 ( Afu5g02330 ) [27] . The microarray data ( Table S2 in Supporting Information S1 ) and Northern analysis revealed that the transcriptional up-regulation of AspF1 during iron starvation is strictly dependent on HapX ( Fig . 1B ) and not affected by SreA as shown previously [22] . AspF1 is cytotoxic and was shown to induce apoptosis of human immature dendritic cells , which indicates that it is involved in immune evasion of A . fumigatus [28] . However , AspF1 was previously shown to be dispensable for virulence of A . fumigatus in a murine model of aspergillosis [29] . A possible explanation for this discrepancy is that the immunosuppressive regimen used in the murine model interferes with the ability of the immune system to preferentially identify the mutant strains . As AspF1-activity is neither iron-dependent nor directly involved in iron acquisition , iron starvation might serve in this case as a signal for expression of a general virulence determinant not related to iron uptake . On the other hand , AspF1 might indirectly increase iron supply of A . fumigatus during the interaction with predators and hosts via cellular iron release due its cytotoxicity . To further investigate the link between HapX and SreA activities we aimed to generate an A . fumigatus mutant lacking both regulators . However , several approaches to generate a ΔsreAΔhapX double mutant failed indicating that deletion of sreA and hapX is synthetically lethal as shown previously in A . nidulans ( Hortschansky et al . , 2007 ) , which underlines the importance of iron regulation . Genes involved in common pathways tend to be genomically clustered in filamentous fungi . Therefore it is interesting to note that among the genes affected by HapX , 41 are organized in gene clusters ( Tables S3 and S4 in Supporting Information S1 ) . Interestingly , the AspF1-encoding gene is neighboured by a co-regulated gene encoding a hypothetical protein ( Afu5g02320 ) and this gene organization is conserved in various fungal species , e . g . , Neosartorya fischeri , Aspergillus clavatus , Microsporum canis , and Arthroderma benhamiae ( data not shown ) . The acetyl transferase-encoding gene Afu5g00720 , one of the clustered genes , was subjected to deletion analysis . Due to its expression pattern it appeared to be a good candidate for the still unidentified acetyl transferase required for FC biosynthesis ( Fig . S1 ) . However , the deletion did not reveal any phenotype ( data not shown ) . A . fumigatus excretes the siderophores FsC and TAFC in roughly equal amounts ( Fig S2 ) . Inactivation of HapX did not substantially alter FsC production but reduced TAFC production to 18% of the wt ( Fig . 3A ) . TAFC is derived from FsC by SidG-catalyzed N2-acetylation [22] . ( Fig . S1 ) . Consistent with the reduction of TAFC production , the sidG ( Afu3g03650 ) transcript level was drastically reduced in ΔhapX as shown by Northern and microarray analyses ( Fig . 1 , Table S2 in Supporting Information S1 ) . Blocking TAFC synthesis by inactivation of SidG has previously been shown to result in increased FsC production [22] . As FsC production was not increased in ΔhapX , it appears unlikely that SidG is the only siderophore biosynthetic activity affected in ΔhapX . In agreement , the microarray analyses ( Table S2 in Supporting Information S1 and Table 2 ) revealed transcriptional down-regulation of other FsC biosynthetic enzymes such as SidF ( Afu3g03400 ) and SidI ( Afu1g17190 ) . Moreover , supply of the siderophore precursor Orn might play a role in siderophore production ( see below ) . The transcriptional profiling ( Table S2 in Supporting Information S1 ) also revealed down-regulation in ΔhapX of the NRPS SidC ( Afu1g17200 ) , which is essential for FC biosynthesis . Consistently , the FC content of ΔhapX was decreased to 68% of the wt . In A . nidulans , HapX inactivation also decreased TAFC production but increased FC production [20] . Despite the general similarity of iron homeostasis-maintaining mechanisms of these two Aspergillus species , these data reveal differences . In contrast to wt , ΔhapX mycelia displayed a reddish pigmentation concomitant with red autofluorescence during iron depleted but not iron-replete conditions ( Fig . 2D and data not shown ) , which is characteristic for accumulation of PpIX , the iron free precursor of heme [20] . Accordingly , the PpIX content of ΔhapX resembled the wt during iron-replete conditions but was 17-fold increased during iron starvation ( Fig . 3B ) . These data indicate derepression of heme biosynthesis during iron starvation in ΔhapX , consistent with the expression profile of genes encoding 5-aminolevulinate synthase ( Afu5g07750 ) , ferrochelatase ( Afu5g07750 ) and a putative heme transporter ( Afu4g11400 ) revealed by the microarray analysis Table S1 in Supporting Information S1 and Table 1A ) . The transcription profiling indicated changes in iron-dependent and -independent steps of the amino acid metabolism in response to HapX inactivation ( Tables S1 and S2 in Supporting Information S1 , Table 1 ) . To gain further insight , we measured the relative composition of the free amino acid pool in wt , ΔhapX , ΔsreA , and ΔsidA during iron sufficiency and starvation ( Table 2 and Table S5 in Supporting Information S1 ) . In wt , iron starvation caused a dramatic remodeling of the composition of free amino acid pool: the relative amounts of nine amino acids ( Arg , Asn , Gln , His , Lys , Met , Orn , Phe , and Trp ) increased whereas that of three amino acids ( Ala , Glu and Val ) decreased >1 . 5-fold . ΔhapX and ΔsidA displayed differences compared to wt mainly during iron starvation , whereas ΔsreA showed differences mainly during iron sufficiency . This is in line with the expression pattern of the deleted genes: hapX and sidA are repressed while sreA is induced by iron ( Fig . 1 ) . During iron starvation , siderophore production reaches up to 10% of the biomass and the major amino acid precursor for siderophore biosynthesis is Orn . The 6 . 9-fold increase of the Orn pool during iron starvation compared to iron sufficiency in wt indicates that the enormous Orn demand for siderophore biosynthesis is matched by active up-regulation of Orn biosynthesis during iron starvation and not by de-repression of Orn biosynthesis via its consumption , which could be expected to decrease the Orn pool . Consistently , blocking Orn consumption for siderophore biosynthesis by inactivation of the Orn hydroxylase SidA ( ΔsidA ) caused a further 2 . 9-fold increase of the Orn pool during iron starvation compared to wt . Orn is synthesized from glutamate or from Orn-derived Arg ( Fig . 4 ) . Consistent with the amino acid analysis , Northern analysis confirmed transcriptional up-regulation of several key enzymes of the Orn/Arg biosynthetic pathway not only in wt but also in ΔsidA , which does not consume Orn for siderophore biosynthesis ( Fig . 4 ) . Strikingly , the Orn pool was 12 . 5-fold decreased in ΔhapX , while Arg was 2 . 0-fold increased ( Table 2 ) . Consequently , the Arg∶Orn ratio changed from 1 . 5 in wt to 49 . 9 in ΔhapX . Northern analysis demonstrated wt-like transcriptional up-regulation of most key enzymes of the Orn/Arg pathway in ΔhapX ( Fig . 4 ) . In perfect agreement with the microarray data ( Tables S1 and S2 in Supporting Information S1 ) , however , transcript levels of four involved enzymes were changed in ΔhapX during iron starvation ( Fig . 4 ) . Consistent with the increased Arg∶Orn ratio in ΔhapX , transcriptional up-regulation of the carbamoyl-phosphate-synthetase ( Afu5g06780 ) and transcriptional down-regulation of the mitochondrial ornithine exporter AmcA ( Afu8g02760 ) in ΔhapX during iron starvation is expected to promote production of Arg relative to Orn; up-regulation of ornithine aminotransferase: ( Afu4g09140 ) , ornithine decarboxylase ( Afu4g08010 ) , and proline oxidase ( Afu6g98760 ) indicates increased consumption of ornithine for purposes other than biosynthesis of siderophores . Taken together , these data indicate that HapX is required for the up-regulation of the Orn pool to fuel siderophore biosynthesis . Therefore , the largely decreased Orn pool of ΔhapX might be in part responsible for the reduced production of TAFC and FC in addition to the transcriptional down-regulation of siderophore biosynthetic enzymes ( see above ) . Consistently , derepression of siderophore biosynthesis during iron sufficiency by deletion of sreA ( ΔsreA ) , when HapX is inactive , decreased the Orn pool to 38% of wt ( Table 2 ) . The 4 . 9-fold increased lysine pool in ΔhapX compared to wt during iron starvation is consistent with transcriptional up-regulation of the lysine biosynthetic enzymes homoaconitase LysF ( Afu5g08890 ) and homocitrate synthase ( Afu4g10460 ) ( Table S1 in Supporting Information S1 ) . The iron-dependence of LysF might explain the 0 . 7-fold decrease of the lysine pool in ΔsidA ( Table 2 ) because lack of siderophore biosynthesis in ΔsidA causes increased iron starvation , which in turn down-regulates and inactivates iron-dependent pathways . In the first commited step of heme biosynthesis , 5-aminolevulinate is synthesized from glycine and succinyl-CoA by HemA . As HemA expression and the heme biosynthetic pathway is derepressed during iron starvation in ΔhapX ( see above ) , the 7 . 5-fold increase in the glycine pool might indicate synchronization of heme biosynthesis and supply of its precursor glycine by HapX ( Table 2 ) . Here , HapX would formally function as a repressor , whereas it acts as an activator for biosynthesis of siderophores and their precursor ornithine . We have previously shown that iron starvation down-regulates heme biosynthesis [22] . Therefore , the possibility of a regulatory link of glycine and heme biosynthesis is underlined by the 0 . 7-fold decrease of the glycine pool during iron starvation compared to iron sufficiency in wt and the further 0 . 5-fold decrease in ΔsidA . Recently , iron starvation was found to influence the composition of the free amino acid pool in S . cerevisiae only mildly [30] with low concordance to A . fumigatus ( Table S6 in Supporting Information S1 ) . The difference might be due to the different life styles of A . fumigatus and S . cerevisiae and of course the inability of S . cerevisiae to synthesize siderophores . As mentioned above , 31% of the genes de-repressed during iron starvation in ΔhapX encode mitochondrial-localized proteins and 66% ( 27 genes ) of those are up-regulated in ΔsreA during iron sufficiency ( Table 1A and Table S1 in Supporting Information S1 ) , which indicates a major impact of iron de-regulation on mitochondrial metabolism . Live cell imaging by laser scanning confocal microscopy of the mitotracker-stained mitochondrial network revealed no differences between wt , ΔhapX , and ΔsreA neither during iron-replete nor iron-depleted conditions ( data not shown ) . Next we analyzed the mtDNA content of wt , ΔhapX and ΔsreA by qPCR normalized against the content of nuclear DNA ( Table S7 in Supporting Information S1 ) . Concomitant with derepression of genes encoding mitochondrial proteins , HapX deficiency increased the mtDNA content during iron starvation 1 . 9-fold but had no effect during iron sufficiency . Vice versa , SreA-deficiency increased the mtDNA content during iron sufficiency 2 . 3-fold but had no effect during iron starvation . Little is known about the molecular mechanisms coordinating replication of nuclear DNA and mtDNA in Aspergilli . Inactivation of SreA and HapX , respectively , may disturb this coordination by deregulation of either the general mitochondrial metabolism ( proteins and/or metabolites ) and/or of a specific regulator . Notably , ΔhapX and ΔsreA display decreased growth rates under the conditions , in which they have increased mtDNA contents ( see above and [22] ) . Formally , it is therefore also possible that toxic effects caused by deficiency in SreA and HapX slow down nuclear DNA replication , whereby the deregulation of nuclear-encoded mitochondrial proteins disturbs the coordination with mitochondrial replication . HapX-deficiency also decreased resistance to tetracycline , an inhibitor of bacterial and mitochondrial protein synthesis [31] , during iron-depleted but not iron-replete conditions ( Fig . 2A ) , which underlines that HapX-deficiency affects mitochondrial metabolism . We have previously shown that there is a close connection between zinc and iron metabolism [32] . In order to avoid zinc excess and zinc toxicity , iron starvation down-regulates expression of genes encoding plasma membrane zinc transporters such as zrfB ( Afu2g03860 ) and the respective transcription activator zafA ( Afu1g10080 ) and concomitantly up-regulates the vacuolar zinc/cadmium transporters zrcA ( Afu7g06570 ) and cotA ( Afu2g14570 ) . The expression profiling indicated increased expression of zrfB and zafA ( Fig . S1 ) and decreased expression of zrcA and cotA ( Fig . S2 ) in ΔhapX during iron starvation suggesting increased zinc uptake and decreased vacuolar zinc storage . In agreement , ΔhapX displayed increased sensitivity to zinc ( Fig 2E ) , which indicates a role of HapX in coordination of iron and zinc homeostasis . To determine whether HapX-mediated regulation is relevant for growth of A . fumigatus in the environment of the host , we compared the virulence of the ΔhapX strain with that of the complemented ΔhapXc strain and the wt strain in two different mouse models of pulmonary invasive aspergillosis: ( i ) a leucopenic mouse model using immunosuppression with both cortisone acetate and cyclophosphamide [33] , [34] , and ( ii ) a non-leucopenic model with immunosuppression by cortisone acetate [35] , [36] . In the leucopenic host , a cellular immune response is virtually absent and development of invasive aspergillosis is characterized by extensive invasive growth of the fungus [37] . Thus , this model allows assessing whether fungal factors are required for survival and growth on lung tissue in general . In contrast , the cortisone acetate model allows recruitment of neutrophils and monocytes , which , despite partially impaired phagocytosis , attack fungal cells and prevent rapid fungal dissemination [38] . Mice were infected with 1×105 conidia in the leucopenic mouse model and 1×106 conidia in the cortisone acetate mouse model to account for the decreased killing rate; survival was monitored over a period of 14 days , followed by histological analyses of the lungs . As shown in the survival curves in Fig . 5A both wt and ΔhapXc caused high mortality rates in the leucopenic mouse model , which were statistically not significantly different ( p = 0 . 29 ) by Kaplan-Meyer estimation and log rank tests . The ΔhapX mutant displayed attenuation in virulence , which was however statistically significant only compared to ΔhapXc ( p = 0 . 033 ) but not compared to wt ( p = 0 . 28 ) . At necropsy , the reduced virulence of ΔhapX was reflected in the incidence of macroscopic lung alterations in comparison to both ΔhapXc and wt ( Fig . 5C ) : eight of ten mice infected with ΔhapXc , seven of ten mice infected with wt , but only one of ten mice infected with ΔhapX displayed lung alterations . The presence of invasive mycelia could be confirmed in the majority of mice infected with ΔhapXc and wt but no mycelium could be found in any mouse infected with ΔhapX ( Fig . 5C ) . In the cortison acetate mouse model , no statistically significant difference in survival were detected between mice infected with either wt or ΔhapXc ( p = 0 . 67 ) . In contrast , ΔhapX was completely attenuated compared to both wt and ΔhapXc ( p = 0 . 004 ) . Consistently , the lungs of all 10 mice infected with ΔhapX were unaltered whereas the lungs of six of ten mice infected with ΔhapXc showed clear symptoms of inflammation ( data not shown ) . The expression of hapX is repressed by iron ( see above ) , and , consistently , deleterious effects of hapX-inactivation are limited to iron-starved conditions ( see above ) . Therefore , the attenuated virulence of ΔhapX is in agreement with A . fumigatus facing iron-limited conditions in the host and the requirement of HapX for virulence . This is also in accordance with the importance of the iron-repressed siderophore system and the dispensability of the iron-induced iron regulator SreA for pathogenicity [5] , [10] , [22] . Notably , supplementation with iron-free TAFC or FsC to a final concentration of 10 mM did neither cure the growth defect nor inhibit the PpIX accumulation of ΔhapX during iron starvation in liquid flask cultures ( data not shown ) indicating that the reduced TAFC production does not account for the full extent of the ΔhapX phenotype . Together with the fact that HapX-deficiency causes decreased production of TAFC but not FsC ( see above ) and the previous finding that the A . fumigatus ΔsidG mutant strain , which produces FsC but not TAFC , displays unaltered virulence in a mouse model for pulmonary aspergillosis [10] , these data suggest that the reduced virulence of ΔhapX is not caused , at least not solely , by the decreased TAFC production . Therefore , the attenuated virulence of ΔhapX might be caused by the general deregulation of gene expression ( i . e . the missing metabolic adaption to iron starvation ) , the accumulation of toxic metabolites such as PpIX , and/or the down-regulation of possible virulence determinants such as AspF1 ( see above ) . The ΔhapX mutant appeared to be slightly more virulent in the leucopenic mouse model compared to the cortisone-acetate model ( Fig . 5A and B ) . As HapX-deficiency results in sensitivity to iron starvation , these data indicate that the attack of neutrophils and monocytes , which is absent in the leucopenic model , increases extracellular iron starvation or imposes iron starvation by internalization . In this respect it is interesting to note that the siderophore system was recently shown to play a crucial role in intracellular growth and survival in murine alveolar macrophages demonstrating that A . fumigatus faces iron starvation after phagocytosis [39] . In agreement , the siderophore system was shown to be essential to alter immune effector pathways and iron homeostasis of murine macrophages [40] . Apart from A . fumigatus HapX , only one fungal iron regulator has been shown to be required for virulence so far: C . neoformans Cir1 , the ortholog of A . fumigatus SreA [41] . Similar to SreA-deficiency in A . fumigatus , Cir1-deficiency impairs growth during iron-replete but not depleted conditions , which does not implicate a crucial role in virulence at first sight . But in contrast to A . fumigatus SreA , which is not required for virulence [22] , C . neoformans Cir1 functions also as an activator for growth at 37°C ( host temperature ) and capsule formation , which are both important virulence traits . This study demonstrates that the metabolic reprogramming required for adaption to iron starvation depends on HapX and that this adaption is essential for virulence of A . fumigatus . The identification of numerous HapX-affected genes with yet uncharacterized link to iron or starvation will aid in the further characterization of the metabolic pathways required for adaption to iron starvation and consequently virulence traits of A . fumigatus . This study appears to be exemplary for the iron metabolism and virulence of most fungal species as HapX is widely conserved with exception of species closely related to S . cerevisiae .
Fungal strains used were A . fumigatus wild-type ATCC46645 ( American Type Culture Collection ) , ΔsreA ( ATCC46645 , ΔsreA::hph ) , ΔsidA ( ATCC46645 , ΔsidA::hph ) , ΔhapX ( ATCC46645 , ΔhapX::hph ) , and ΔhapXC ( ΔhapX , ( p ) ::hapX ) . ΔsreA and ΔsidA were described previously [5] , [22]; generation of ΔhapX and ΔhapXC is described below . Generally , A . fumigatus strains were grown at 37°C in Aspergillus minimal medium according to Pontecorvo et al . [42] containing 1% glucose as the carbon source and 20 mM glutamine as the nitrogen source . Iron-replete media contained 30 mM FeSO4 . For iron depleted conditions , iron was omitted . The BPS and tetracycline concentrations used were 200 µM and 2 mg ml−1 respectively . For growth assays , 104 and 108 conidia were used for point-inoculation on plates or inoculation of 100 ml liquid media , respectivly . RNA was isolated using TRI Reagent ( Sigma ) . For Northern analysis , 10 µg of total RNA was analyzed as described previously [43] . Hybridization probes and Primers used are listed in Table S7 in Supporting Information S1 . For extraction of genomic DNA , mycelia were homogenized and DNA was isolated according to Sambrook et al . [44] . For general DNA propagations Escherichia coli DH5α strain was used as a host . For generating the ΔhapX mutant strain , the bipartite marker technique was used [45] . Briefly , A . fumigatus was co-transformed with two DNA fragments , each containing overlapping but incomplete fragments of the pyrithiamine resistance-conferring ptrA gene fused to 1 . 2-kb hapX 5′- and 3′-flanking sequences , respectively . The hapX 5′-flanking region ( 1207bp ) was PCR-amplified from genomic DNA using primers ohapX-1 and ohapX-4 . For amplification of the 3′flanking region ( 1156bp ) primers ohapX-2 and ohapX-3 were employed . Subsequent to gel-purification , these fragments were digested with SacI ( 5′flanking region ) and XhoI ( 3′flanking region ) , respectively . The ptrA selection marker was released from plasmid pSK275 by digestion with SmaI and XhoI , respectively , and ligated with the 5′- and 3′-flanking region , respectively . The transformation construct A ( 2558bp , fusion of the hapX 5′-flanking region and the prtA split marker ) was amplified from the ligation product using primers ohapX-5 and optrA-2 . For amplification of the transformation construct B ( 2166bp , fusion of the hapX 3′-flanking region and the supplementary prtA split marker ) primers ohapX-6 and optrA-1 were employed . For transformation of A . fumigatus ATCC46645 both constructs A and B were simultaneously used . This strategy deleted the sequence −228 to 1383 bp relative to the translation start site in hapX . For the reconstitution of the ΔhapX strain with a functional hapX copy , a 3615bp PCR fragment generated with primers ohapX-5 and ohapX-6 was subcloned into pGEM-T ( Promega ) according to the supplier's manual , resulting in pHapX . A 2410bp SphI fragment from pAN7-1 containing the hygromycine B resistance-conferring hph gene was inserted into the SphI site of pHapX resulting in pHapXhph . The resulting 9 . 0-kb plasmid pHapXhph was linearized with EcoRV and used to transform A . fumigatus ΔhapX . Transformation of A . fumigatus was carried out as described previously [10] . For selection of transformants 0 . 1 µg ml−1 pyrithiamine ( Sigma ) or 0 . 2 mg ml−1 hygromycin B ( Calbiochem ) was used . Screening of transformants was performed by PCR and confirmed by Southern blot analysis . The hybridization probes for Southern blot analysis of ΔhapX and ΔhapXc strains were generated by PCR using the primers ohapX-5 and ohapX-4 ( Table S8 in Supporting Information S1 ) . Analysis of siderophore , PpIX and free amino acids was carried out by reversed phase HPLC as described previously [20] , [43] , [46] . To quantify extracellular or intracellular siderophores , culture supernatants or cellular extracts were saturated with FeSO4 and siderophores were extracted with 0 . 2 volumes of phenol . The phenol phase was separated and subsequent to addition of 5 volumes of diethylether and 1 volume of water , the siderophore concentration of the aqueous phase was measured photometrically using a molar extinction factor of 2996/440nm ( M−1cm−1 ) . A . fumigatus total DNA was isolated with the QIAamp kit ( Qiagen ) . MtDNA compared to nuclear DNA content was determined by quantitative real-time PCR ( qPCR ) with CYBR green I ( ABI; ABI 2400 Applied Biosystems , USA ) by PCR amplification of a fragment of the mitochondrial apocytochrome B gene ( BAA34151 , 73 . t00020 ) using primers oAfcytB-1 and oAfcytB-2 , and a fragment of the nuclear mirD gene ( Afu3G03440 ) gene , using primers oAfmirD-1 and oAfmirD-2 . The PCR reactions cycle used ( Applied Biosystems standard conditions ) was 40 cycles at 95°C 15″ , 60°C 1′ . PCR assays were performed in biological triplicates and technical duplicates for each DNA sample . The expression of mtDNA copy number relative to nuclear DNA was determined using the 2−ΔCT method . The A . fumigatus Af293 DNA amplicon microarray containing 9 , 516 genes [47] was used in this study . To profile the genome-wide expression responses to the shift from iron depleted to iron-replete conditions and to identify the genes influenced by HapX , we conducted microarray analysis with the wt and ΔhapX strains grown for 16 h at 37°C in iron-depleted ( −Fe ) medium ( 0 h time point ) . Subsequently iron was added to a final concentration of 30 mM and growth was continued for 1 hour ( sFe ) . Labelling reactions with RNA , and hybridization were conducted as described in the PFGRC standard operating procedures ( PFGRC SOP's ) found at http://pfgrc . tigr . org/protocols/protocols . shtml . The sample from 0 h served as reference in all hybridizations with the 1 h iron shift samples in order to identify genes exhibiting altered transcription after the shift from iron depleted to iron replete conditions in ΔhapX compared to wt . Additionally , the 16 h iron starvation cultures of ΔhapX and wt were directly compared with the wt serving as reference . ( Fig . S1 and S2 ) Hybridized slides were scanned using the Axon GenePix 4000B microarray scanner and the TIFF images generated were analyzed using the TM4 suite of microarray analysis tools ( http://www . tm4 . org ) . Spotfinder was used to obtain relative transcript levels . Data from Spotfinder were stored in MAD , a relational database designed to effectively capture and store microarray data . Data normalization was accomplished using a local regression technique LOWESS ( LOcally WEighted Scatterplot Smoothing ) for hybridizations using the TM4 MIDAS tool . The resulting data was averaged from triplicate gene spots on each array and from duplicate flip-dye arrays for each experiment , taking a total of 6 intensity data points for each gene . Differentially expressed genes at the 95% confidence level were determined using intensity-dependent Z-scores ( with Z = 1 . 96 ) as implemented in MIDAS and the union of all genes identified at each time point from the wild-type were considered significant in this experiment . Microarray data are deposited in the GEO database ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? token=jhqbdcwiigqyypa&acc=GSE22052 ) . Virulence assays in two murine models for pulmonary aspergillosis were performed as described previously [33] , [34] . Infections were performed with two groups of five mice for each tested strain . A control group remained uninfected ( inhalation of PBS ) to monitor the influence of the immunosuppressive regime . Survival data were plotted as Kaplan-Meyer curves and were analyzed statistically by a log rank test using Graph Pad Prism version 5 . 00 for Windows ( GraphPad Software , San Diego , CA ) . Lungs from euthanized animals were removed , fixed in formalin and paraffin-embedded for histopathologic analyses according to standard protocols . Sections were stained with Periodic acid-Schiff ( PAS ) according to standard protocols and analyzed by bright field microscopy using a Zeiss AxioImager . M1 microscope equipped with a SPOT Flex Shifting Pixel Color Mosaic camera ( Diagnostic Instruments , Inc . , Sterling Heights , USA ) . Mice were cared for in accordance with the principles outlined by the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes ( European Treaty Series , no . 123; http://conventions . coe . int/Treaty/en/Treaties/Html/123 . htm ) . All animal experiments were in compliance with the German animal protection law and were approved ( permit no . 03-001/08 ) by the responsible Federal State authority ( Thüringer Landesamt für Lebensmittelsicherheit und Verbraucherschutz ) and ethics committee ( beratende Komission nach § 15 Abs . 1 Tierschutzgesetz ) .
|
Due to its requirement for a wide range of cellular processes , iron is an essential nutrient for virtually every organism . The mammalian immune system utilizes iron-withholding mechanisms to deny access to free iron . Therefore , pathogens must overcome extreme iron limitation . Patients with suppressed immune systems due to cancer treatments , organ transplantation , or genetic disorders are at high risk of infection with the ubiquitously present fungal pathogen Aspergillus fumigatus . In this study we found that in Aspergillus fumigatus iron starvation results in drastic metabolic changes depending on the transcription factor HapX . During iron starvation , HapX functions include the repression of iron-consuming pathways to spare iron and activation of iron uptake by siderophores . Siderophores are small molecules able to “steal” iron from host molecules and have previously been shown to play a crucial role in the virulence of Aspergillus fumigatus . Genetic inactivation of HapX attenuates virulence of Aspergillus fumigatus in a murine model of aspergillosis , demonstrating that adaption to iron limitation is a crucial virulence determinant . The identification of numerous HapX-affected genes with a yet uncharacterized link to iron will aid in the further characterization of the metabolic pathways required for fungal adaption to iron starvation and virulence traits .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"biochemistry",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/functional",
"genomics",
"infectious",
"diseases/fungal",
"infections",
"microbiology/microbial",
"physiology",
"and",
"metabolism",
"microbiology/microbial",
"growth",
"and",
"development",
"cell",
"biology/chemical",
"biology",
"of",
"the",
"cell",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
] |
2010
|
HapX-Mediated Adaption to Iron Starvation Is Crucial for Virulence of Aspergillus fumigatus
|
The fusion of two distinct prominences into one continuous structure is common during development and typically requires integration of two epithelia and subsequent removal of that intervening epithelium . Using confocal live imaging , we directly observed the cellular processes underlying tissue fusion , using the secondary palatal shelves as a model . We find that convergence of a multi-layered epithelium into a single-layer epithelium is an essential early step , driven by cell intercalation , and is concurrent to orthogonal cell displacement and epithelial cell extrusion . Functional studies in mice indicate that this process requires an actomyosin contractility pathway involving Rho kinase ( ROCK ) and myosin light chain kinase ( MLCK ) , culminating in the activation of non-muscle myosin IIA ( NMIIA ) . Together , these data indicate that actomyosin contractility drives cell intercalation and cell extrusion during palate fusion and suggest a general mechanism for tissue fusion in development .
Tissue fusion is required for the morphogenesis of numerous vertebrate organ systems , including neural tube closure , heart morphogenesis , urogenital development , and craniofacial development , and failure of tissue fusion leads to birth defects in these contexts [1] . In craniofacial development , tissue fusion is required during the formation of the primary and secondary palates , with deficits in these processes resulting in cleft lip and cleft palate , respectively [2 , 3] . The secondary palate arises from bilateral outgrowths of the maxillary processes called palatal shelves , which undergo a highly coordinated morphogenesis involving vertical outgrowth , elevation , horizontal growth , and ultimately fusion with one another to form the intact roof of the mouth [4] . The external surface medial edge epithelium ( MEE ) of the palatal shelves is composed of an outer layer of flat periderm cells covering an inner layer of basal cuboidal cells on a basement membrane [5 , 6] . Periderm cells have been proposed to provide temporal and spatial regulation of adhesion competence and are thought to undergo apoptosis and slough off immediately prior to palatal shelf contact [5] . Electron microscopy studies of unpaired palatal shelves have shown that cells of the MEE extend filopodial and lamellipodial projections prior to the palatal shelves touching [7–9] . Whether projections persist until the shelves meet and whether they have functional significance in the initiation of fusion is not clear . Additionally , relatively little is known about the dynamic cellular behaviors that occur immediately upon contact of the independent palatal shelves . Static histological observations have indicated that apposing palatal shelf epithelial cells combine to form a common medial epithelial seam ( MES ) , possibly by a convergent extension-like mechanism , but no direct evidence of convergence has been documented , and how these epithelia integrate is not known [10–12] . After the palatal shelves meet , the MES must be removed to achieve confluence of the underlying mesenchyme . The cellular mechanisms by which this occurs have been the subject of considerable investigation , and three mechanisms have been proposed: ( 1 ) epithelial to mesenchymal transition ( EMT ) , ( 2 ) apoptotic cell death , and ( 3 ) cell migration [7 , 13] . Initial support for EMT , based on histological observation and ex vivo lineage tracing with vital dyes [5 , 14–16] , has since been refuted by genetic lineage tracing experiments showing that the palatal epithelium does not give rise to mesenchymal cells that are maintained in the secondary palate [17 , 18] . Instead , it has been proposed that apoptotic cell death may be solely responsible for disappearance of the MES . Indeed , there have been multiple reports of apoptosis in the MES during fusion stages , and apoptosis is reduced in some mutants that fail to undergo proper palatal fusion [2] . Whether apoptosis is sufficient for removal of the MES is uncertain , however , because pharmacological or genetic inhibition of apoptosis has been inconclusive [19–24] . Finally , based on studies involving epithelial labeling and static observation at progressive time points , MES cells have also been proposed to actively migrate in the oronasal and anteroposterior dimensions to allow confluence of the underlying mesenchyme [10 , 23] . No direct observation of cell migration in the palatal epithelium has been reported , however , and relevance of migration in palate fusion remains unknown . Cell migration and other morphogenetic cell behaviors are mediated by non-muscle myosin II ( NMII ) , a subfamily of actin-based molecular motors that generate actomyosin contractility [25 , 26] . Each NMII unit is a hexamer composed of a pair of heavy chains ( NMHC ) , a pair of essential light chains ( ELC ) and a pair of regulatory light chains ( RLC ) . Three different isoforms of NMII ( NMIIA , NMIIB , and NMIIC ) are named according to the identity of their heavy chains , NMHCIIA , NMHCIIB , and NMHCIIC , which are encoded by the genes Myh9 , Myh10 , and Myh14 , respectively . Whereas the ELC stabilizes the heavy chain structure , phosphorylation of the RLC positively regulates NMII-mediated actin contractility . Two major upstream kinases , rho-kinase ( ROCK ) and myosin light chain kinase ( MLCK ) , phosphorylate RLCs to activate NMII [26 , 27] . Actomyosin contractility mediated by NMII has been demonstrated to be critical for a wide variety of morphogenetic events . During tissue closure events , such as Drosophila dorsal closure , C . elegans ventral enclosure , and Zebrafish epiboly , the formation of a supracellular actin cable at the leading edge of the epithelium maintains a uniform epithelial advance for which NMII provides contractile force to progressively close the opening in a purse-string or ratchet-like action [28–33] . During Drosophila dorsal closure , filopodia sample the closely apposed epithelium and begin to interdigitate , zipping the epithelia together [31 , 34] . Actomyosin contractility also drives intercalation and convergent extension behaviors during development . For example , NMII-driven actomyosin contractility drives cell neighbor exchange required for intercalation and germband extension in Drosophila and is required for polarized motility and convergent extension in Xenopus [35 , 36] . Here , we analyze the cellular mechanisms of tissue fusion in the mammalian secondary palate using live imaging to reveal the dynamic and multistep nature of this tissue fusion . We find that the formation and subsequent resolution of a multilayered epithelium to a shared , single-cell layer MES involves cell intercalation behaviors and convergence of the MES to the midline . Concomitant to this convergence , MES cells undergo coordinated cell movement in the oronasal axis and frequent cell extrusion events , indicating that convergent displacement and extrusion contributes to the removal of the MES . Actomyosin contractility is required for normal intercalation , displacement and extrusion of the MES cells , and disruption of actomyosin contractility by pharmacologic or genetic methods results in a failure of proper palatal shelf fusion . Perturbation of upstream regulators of actomyosin contractility had similar consequences on palatal shelf fusion , allowing the initial assembly of a pathway controlling these cell behaviors . These studies reveal a novel cellular mechanism for tissue fusion and provide a basis for studying the involvement of known regulators of tissue fusion in these cellular processes .
To directly observe epithelial cell behavior during the course of tissue fusion , we used confocal live imaging of recently adhered E14 . 5 palatal shelves in explant culture ( S1A Fig ) . Because our conditions for explant culture were different from those previously reported , we first confirmed complete fusion of the secondary palate by histology ( S1B Fig ) . For resolution of the epithelium in the context of the secondary palate mesenchyme , we took advantage of the ROSA26mTmG reporter mouse , which expresses membrane-targeted tdTomato prior to Cre-mediated excision , and membrane-targeted GFP following recombination mediated by an epithelial-specific Cre ( K14-Cre ) ( S1C Fig ) [37 , 38] . At the initiation of the culture period , the palatal shelves had made contact and adhered in the middle palate , but not yet in the anterior region . We first focused on cell behaviors occurring immediately before and during initiation of secondary palate fusion by imaging a region of the anterior palate that had not made contact . In this region , we observed active epithelial projections prior to fusion ( Fig 1A–1H , arrowheads; S1 Movie ) ; these appeared as larger cellular protrusions into the oral cavity rather than as thin filopodial projections . Protrusions were observed to reach across and form junctions with MEE cells from the opposite palatal shelf ( Fig 1I , arrow; S2 Movie ) . Upon contact of these extensions with cells from the apposing palatal shelf , cellular bridges formed connecting the two shelves ( Fig 1A , circles; S1 Movie ) . These bridges , which initially appeared at spatial intervals , filled in with epithelium moving from deeper optical sections to form a multilayered epithelium shared between the two palatal shelves ( Fig 1A–1D; S1 Movie ) . Over a period of 12 h , the multilayered epithelium converged toward the midline , ultimately resolving in a single-layered MES that was shared between the two palatal shelves ( Fig 1E–1H; S1 Movie ) . Cell tracking of epithelial cells from either side of the multilayered MES indicated a net displacement of cells from lateral to medial , whereas the mesenchyme showed less directed movement toward the midline , supporting the overt observation of epithelial convergence ( Fig 1J , 1K , 1L and 1M ) . As neighbor exchange is a defining characteristic of cell intercalation , we examined whether similar rearrangements occurred during the integration of the MES . Cells initially sharing junctions with three or four cells rearranged to align along the midline , ultimately sharing junctions only with two neighboring cells ( Fig 1O–1R , S3 Movie ) . Such junctional rearrangements were observed 56 times across seven adjacent optical sections over the course of imaging . We further quantified the extent of junctional rearrangement by counting all horizontal ( -30°-30° ) and vertical ( 60°–120° ) junctions relative to the midline . At the beginning of the culture period there were similar numbers of horizontal and vertical junctions; the proportion of junctions that were horizontal decreased with time while the proportion of vertical junctions increased , supporting the widespread occurrence of junctional rearrangement in the MES during convergence ( Fig 1S ) . A hallmark of intercalation is an increase in the distance between neighboring cells , as cells intercalate between them and drive them apart . We found that the distance between the cell centers of pairs of neighboring epithelial cells ( n = 19 pairs ) consistently increased , whereas the distance between pairs of neighboring mesenchymal cells ( n = 16 pairs ) did not tend to increase ( Fig 1T ) . Together , these data demonstrate that palate fusion is initiated by formation of a multicellular epithelium that resolves to a shared single-layer epithelium by an active cellular convergence process that involves widespread cell intercalation of MES cells . We next sought to determine whether cell migration might play a role in the removal of the MES . Four-dimensional cell tracking and displacement analysis revealed collective displacement of deeper ( i . e . , more nasal ) MES cells toward the oral surface ( Fig 1J and 1N ) relative to the lateral palatal epithelium , which did not show marked movement . This oronasal displacement was a consistent behavior that was shared by the bulk of the MES cells analyzed ( Fig 1J and 1N ) . In a few cases in the middle palate , we also observed a coherent migration of MES cells on the oral surface toward the posterior unfused part of the secondary palate ( S4 Movie ) . Together , these data provide direct evidence for involvement of cell displacement in the removal of the MES during palatal fusion . Cell intercalation and cell migration require actomyosin contractility provided by non-muscle myosin activity [26 , 39] . The three non-muscle myosin isoforms are defined by their heavy chain subunits , which exhibit tissue-specific and developmentally regulated expression and function [26 , 40 , 41] . We therefore sought to evaluate the involvement of each of the non-muscle myosin isoforms in palate fusion by examining expression of the three heavy chain genes . Myh9 exhibited elevated mRNA expression in the MEE and presumptive point of fusion between the nasal septum and palatal shelves immediately prior to and during fusion stages; immunostaining for NMHCIIA confirmed this expression ( S2A and S2D–S2I Fig ) [42] . Myh10 was expressed broadly throughout the mesenchyme and epithelium , though at apparently lower levels in the epithelium than Myh9 , and Myh14 was not detected by in-situ hybridization ( S2B and S2C Fig ) . To determine whether NMII might be involved in palate fusion , we first treated E13 . 5 palatal explant cultures with blebbistatin , a selective inhibitor of NMII ATPase activity [43] . After 72 h of culture , the MES had disappeared in nearly all control DMSO-treated palatal explants ( mean fusion score = 3 . 92 ) , whereas treatment with 10 μM blebbistatin resulted in failed fusion and maintenance of an intact MES in most sections analyzed ( mean fusion score = 2 . 36 ) ( S2J–S2L Fig ) . This was not a consequence of effects on cell proliferation , as there was no change in the percentage of Ki67+ cells in the epithelium or mesenchyme as assayed by immunostaining ( S2M Fig ) . Based on the relative expression of the three isoforms , we hypothesized that NMHCIIA might play a dominant role in this context; siRNA knockdown of Myh9 in explant culture mimicked the effects of blebbistatin , supporting this hypothesis ( S2N–S2S Fig ) . We next tested whether genetic ablation of NMHCIIA function affected palatal fusion in vivo . Because generalized loss of function of NMHCIIA results in early embryonic lethality as a consequence of a failure to form a polarized visceral endoderm [44] , we generated Tgfβ3Cre/+; Myh9lox/lox embryos to mediate loss of NMHCIIA in the palate epithelium [45] . Whereas Tgfβ3Cre/+ control embryos displayed a complete removal of the MES by E15 . 5 , Tgfβ3Cre/+; Myh9lox/lox embryos retained the MES at this stage ( Fig 2A–2G ) . Because Tgfβ3Cre activity is not strictly restricted to the epithelium [45] , we also employed K14-Cre to genetically ablate NMHCIIA from the fusing epithelium . Similar to Tgfβ3Cre/+; Myh9lox/lox embryos , the MES at E15 . 5 failed to disappear in K14-Cre; Myh9lox/lox embryos ( S3A Fig ) , supporting a specific requirement for NMHCIIA in the epithelium during palatal fusion . Greater dissolution of the MES occurred in the K14-Cre; Myh9lox/lox embryos compared with Tgfβ3Cre/+; Myh9lox/lox embryos ( S3B Fig ) . Though Tgfβ3Cre is known to mediate some recombination in the palate mesenchyme , cell proliferation index was not significantly different between Tgfβ3Cre/+; Myh9lox/lox and control embryos , indicating that perturbed fusion was not attributable to effects of loss of NMHCIIA on mesenchymal cell proliferation ( S3C Fig ) . Instead , this less severe phenotype corresponded to persistent higher levels of NMHCIIA expression in K14-Cre; Myh9lox/lox compared with Tgfβ3Cre/+; Myh9lox/lox embryos , indicating that K14-Cre did not mediate complete loss of NMHCIIA from the MES ( Fig 2L and 2M , S3D Fig ) . Examination of 100 histological sections across three embryos indicated that the MES defect in NMHCIIA-deficient embryos was consistent , and comparison of quantified fusion scores revealed significant differences between K14-Cre; Myh9lox/lox or Tgfβ3Cre/+; Myh9lox/lox embryos and controls ( Fig 2G; S3B Fig ) . When we examined the E14 . 5 MES at higher magnification we found that whereas control embryos already exhibited an integrated single-layered MES , Tgfβ3Cre/+; Myh9lox/lox embryos maintained a multilayered epithelium that failed to converge ( Fig 2H and 2I ) . Furthermore , at E15 . 5 , when the MES was completely lost from control embryos , Tgfβ3Cre/+; Myh9lox/lox embryos still retained a multilayered MES ( Fig 2J and 2K ) . This multilayered epithelium , which could be visualized by E-cadherin immunostaining , was not composed solely of inappropriately retained periderm cells , because all cells appeared to express p63 , a marker of the basal epithelium that is not expressed in periderm cells ( Fig 2N and 2O ) [46] . Failure of the normal removal of the MES did not lead to a cleft palate phenotype in Tgfβ3Cre/+; Myh9lox/lox embryos , but the trapped epithelium was broken into smaller islands that were still apparent even at E17 . 5 ( S3E Fig ) . Because Myh10 exhibited broad expression that included the palate epithelium , we next asked whether the NMHCIIB isoform might also contribute to palatal fusion . Whereas disruption of NMHCIIB in Tgfβ3Cre/+; Myh10lox/lox embryos did not lead to defects in MES removal ( S3F Fig ) , Tgfβ3Cre/+; Myh9lox/lox; Myh10lox/lox mutant embryos did exhibit poorer MES removal and a submucous-type cleft of the posterior palate , supporting additive roles for NMHCIIA and NMHCIIB during palate fusion ( S3G Fig ) . These results indicate that non-muscle myosin is critical for the organization of a multilayered epithelium into a shared MES and progression to normal fusion of the mammalian secondary palate . NMII is regulated by phosphorylation of the regulatory light chain ( RLC ) by multiple kinases . For the most part , the context-specific expression and function of each of the six RLC genes has not been reported , though recently Myl9 was shown to be involved in Xenopus convergent extension [47] . We used in-situ hybridization with probes against each of the RLC genes to determine that only one gene , the Myl9 regulatory light chain gene , exhibited elevated expression in the fusing MES ( S2T–S2V Fig ) . siRNA knockdown of Myl9 led to significantly reduced palatal fusion after 72 h of explant culture ( mean fusion score = 2 . 26 ) compared with scrambled siRNA control ( mean fusion score = 4 . 38 ) ( Fig 3A–3D ) . Rho kinase , ROCK , phosphorylates the RLC , and inhibitory myosin phosphatase , MYPT1 to activate actomyosin contractility during morphogenesis [48] . We therefore tested whether ROCK is involved in palatal fusion by pharmacologically inhibiting its function during palatal fusion . After 72 h in culture , DMSO-treated control palatal explants had fused ( mean fusion score = 3 . 8 ) , whereas treatment with the ROCK inhibitor Y27632 almost completely blocked palatal fusion ( mean fusion score = 1 . 4 ) , leading to the maintenance of a multi-layered epithelium ( Fig 3E–3G ) . After 48 h of culture , DMSO-treated control explant cultures had established a single-layered MES , whereas Y27632-treated explant cultures failed to resolve the multilayered epithelium ( Fig 3H and 3I ) , similar to what we observed in embryos lacking NMII function . Whereas ROCK has substrates independent of this pathway , myosin light chain kinase ( MLCK ) is relatively specific in its phosphorylation of the RLC to activate actomyosin contractility [26] . We therefore asked whether pharmacological inhibition of MLCK might also affect palatal fusion . Similar to ROCK inhibitor , treatment with the MLCK inhibitor ML-7 in 72 h explant cultures resulted in a significant reduction of palatal fusion ( mean fusion score = 1 . 82 ) compared with control ( mean fusion score = 4 . 26 ) ( Fig 3J–3L ) ; 48 h of culture in ML-7 resulted in a failure of the MES to integrate ( Fig 3M and 3N ) , suggesting that MLCK may also be involved in this pathway . Given the dramatic effect of ROCK inhibition on palatal fusion , we next asked how ROCK influences cell behaviors during the fusion process . Confocal live imaging of explant cultures treated with Y27632 revealed that although the epithelia of the palatal shelves became apposed , cell intercalation and neighbor exchange did not occur ( Fig 4A–4L; S5 Movie ) . Cell tracking indicated that failure of cell intercalation was concomitant with a failure of the apposed epithelia to integrate toward the midline , and two layers of epithelium were still apparent even after 12 h of culture ( Fig 4M–4P ) . Similarly , Y27632 prevented oronasal displacement of MES cells to the oral surface ( Fig 4Q ) . Together these results indicate that an actomyosin contractility pathway including ROCK , MLCK , and NMII is required for the convergence and oronasal displacement behaviors underlying normal secondary palate fusion . Given the requirement for NMII in normal palate fusion , we next examined how filamentous actin was distributed over the course of palatal fusion in relation to the cell behaviors we observed . To observe actin filament dynamics in real time , we imaged explant cultures of Lifeact-mRFPruby transgenic mice , which ubiquitously express an RFP-conjugated Lifeact peptide that binds filamentous actin [49 , 50] . We imaged Lifeact fluorescence over an 8-h period to show actin filament remodeling over the course of initial apposition , formation of a multicellular epithelium , convergence into a single-layered MES , and ultimately , MES breakage . As the palatal shelves approached the midline , multicellular actin cables were observed along the apical edge of the advancing palatal shelves ( Fig 5A red arrowheads , S6 Movie ) . As the palatal shelves moved closer to one another , an abundance of epithelial cells that were rich in filamentous actin were displaced from deeper sections between the edges of the MEE to form a multicellular intermediate structure ( Fig 5A , 5B , 5Q and 5R S6 Movie ) , followed by resolution to an organized MES that was bound laterally by the multicellular actin cable ( Fig 5A–5H; S7 Movie ) . In deeper sections that were undergoing seam breakage , the actin cable on either side of the newly formed MES appeared to contract while the MES pulled apart into separate islands of epithelium and began to dissolve ( Fig 5F–5H , S7 Movie ) . Treating explants with blebbistatin considerably inhibited convergence and oronasal displacement of the MES ( Fig 5I–5L , 5S and 5T; S8 Movie ) , and though the MES was thinner at deeper positions , we did not observe the accumulation of lateral actin cables or breakage of the seam into epithelial islands ( Fig 5M–5P; S9 Movie ) . Treatment of cultures with actin polymerization inhibitors cytochalasin D or latrunculin A had similar effects on the MES to blebbistatin , including loss of a clear leading edge and lateral actin cable structures ( S4 Fig ) . Live imaging with Lifeact-mRFPruby transgenic mice also revealed striking cell extrusion-like events , in which cellular rosettes surrounded cells with a circular cable of elevated filamentous actin and appeared to squeeze them out of the MES ( Fig 6A–6D , 6A’–6D’ , S10 Movie ) These events greatly resembled cell extrusion that has been reported in other systems , in which an epithelial cell destined for extrusion is surrounded by other epithelial cells which contract an actomyosin ring that squeezes the cell out [51–54] . During normal palatal fusion , these events were frequent , with 101 observed across 16 optical sections of the anterior palate over an 8-h period in two independent live imaging experiments . Treatment with blebbistatin or cytochalasin D reduced the formation of multi-cellular rosette structures , and extrusion events were rarely observed , with 18 seen across 14 optical sections over an 8-h period in two independent live imaging experiments treated with blebbistatin and 20 seen across 12 optical sections over an 8-h period when treated with cytochalasin D . The few extrusion-like events that were observed in blebbistatin-treated palate explant cultures did not exhibit well-formed rosettes ( Fig 6E–6H , 6E’–6H’ , S11 Movie ) . To determine whether cell extrusion occurred in intact embryos , and whether extruding cells were undergoing apoptosis , we performed immunostaining of transverse sections of E14 . 5 embryos for cleaved caspase-3 and E-cadherin ( Fig 6I–6K’ ) . Across six embryos , we found that approximately 8 . 9 ± 1 . 12% ( mean ± SEM ) of MES cells that were apoptotic were also part of a rosette . We found approximately the same proportion of apoptotic cells not part of rosettes ( 8 . 6 ± 2 . 54% ) , suggesting that apoptotic cell extrusion occurs with similar frequency to extrusion-independent apoptosis ( Fig 6L ) . Although static imaging was not able to determine whether these cells underwent apoptosis during or after extrusion , it does indicate a relationship between cleaved caspase-3 positive cells and multicellular rosettes during palatal fusion in intact embryos . In addition to apoptotic cell extrusion , live cell extrusion can be induced by increased density and crowding force dependent on the activity of stretch-activated ion channels such as Piezo1 [51] . To test whether such a mechanism is at play during secondary palate fusion , we performed explant culture in the presence of gadolinium ( Gd3+ ) , an inhibitor of stretch-activated ion channels [51 , 55 , 56] . Treatment with 50 μM Gd3+ blocked palate fusion ( mean fusion score = 1 . 78 ) compared with control ( mean fusion score = 4 . 26 ) , suggesting that live cell extrusion may play a key role in this process ( Fig 6M–6O ) . Together , these data indicate that supracellular actin cable structures act concomitantly with cell convergence and displacement behaviors during secondary palate fusion and suggest that cellular extrusion-like behavior may contribute to removal of MES cells .
Here , we identify dynamic cell behaviors underlying mammalian tissue fusion by combining direct observation by live imaging with functional ex vivo and genetic in vivo experiments . Our findings point to a new model for tissue fusion in the secondary palate ( Fig 7 ) . Initiation of palatal fusion involves cellular protrusions that establish contacts with the opposite shelf and cellular bridges that eventually give rise to complete contact of the palatal shelves . Recently , live imaging of neural tube closure in mouse embryos revealed that the midbrain neural tube does not close by simple zippering , as in the hindbrain , but rather by a “buttoning” action , where initial contacts lead to intermediate closure points followed by zippering to close the gap between these contacts [57] . Our data indicate that initiation of palatal shelf fusion may be similar to what occurs in the midbrain , with the establishment of transient epithelial bridges at intervals along the fusion front that subsequently zipper closed . In the secondary palate , these bridges form a transient , multi-layered MES structure in a process that involves the displacement of MES cells from deeper positions in the fusing palatal shelves . This multi-layered structure is then integrated toward the midline by intercalation involving cell neighbor exchange . Intercalation is concurrent with displacement of MES cells in the oronasal axis . Blocking convergence by disrupting actomyosin contractility also resulted in failed oronasal MES displacement , suggesting a convergent displacement mechanism for the initiation of MES clearance . Convergence of the MES requires force generated by NMHCIIA , as well as upstream regulators of actomyosin contractility including ROCK , MLCK , and the RLC encoded by the Myl9 gene . We conclude that convergence force is intrinsic to the epithelium , because loss of NMHCIIA function specifically in this cell type resulted in a failure of MES integration . Our current data do not , however , allow us to determine whether a contribution is made to convergence by non-MES epithelium; a recent report indicates that secondary palate elongation is driven in part by vertical intercalation in the non-MES palate epithelium , suggesting the possibility that convergence forces could in part be generated outside of the MES [58] . As the shared MES begins to dissolve , we observed the lateral accumulation of filamentous actin into multicellular cables , which surrounded and appeared to contract during MES breakage , and NMII-generated contractility was required for MES breakage into epithelial islands . The involvement of NMHCIIA in palate development takes on particular significance in light of multiple reports that the MYH9 gene locus is associated with non-syndromic cleft lip and palate in humans [42 , 59–62] . Although loss of NMHCIIA alone in the palatal epithelium did not lead to a cleft secondary palate in mice , compound disruption of NMHCIIA and NMHCIIB did result in a submucous cleft in the posterior palate . Our analysis clearly supports the involvement of non-muscle myosins in the fusion stage of secondary palate development , which may contribute to its role in human clefting . Live imaging revealed abundant cell extrusion events characterized by the formation of multi-cellular rosettes surrounding an actin ring and subsequent constriction and squeezing-out of epithelial cells . Cell extrusion can occur by apoptotic or live cell mechanisms [63] . In apoptotic cell extrusion , a cell undergoing apoptosis signals to its neighbors early in the apoptotic process , leading to formation and contraction of an actomyosin ring and squeezing out of the apoptotic cell [53] . In live cell extrusion , epithelial overcrowding induces extrusion of cells that are viable but may then undergo anoikis due to loss of attachment [63] . Our results inhibiting stretch-activated channels indicate that crowding-induced cell extrusion is required for palate fusion . Whether cell extrusion in the MES occurs mainly by apoptotic or live-cell mechanisms remains a question for future study . Regardless , the involvement of cell extrusion in palate fusion might explain the apparent lack of requirement for caspase-mediated apoptosis in removal of the MES because blocking caspase activity had no effect on apoptotic or live cell extrusion [21–23 , 51 , 53] . We propose that epithelial cells are ultimately extruded to the oral and nasal surfaces , giving rise to the “epithelial triangles” that have been previously described to contain an elevated number of dying cells ( Fig 7C ) [11] . Interestingly , early ultrastructural studies are consistent with this model . In transmission electron microscopy studies of rat palatogenesis , Shüpbach et al . observed a process termed “cell exfoliation” and documented the appearance of cell blebs “apparently erupting” from the epithelial triangles of the MES into the oral or nasal cavity [64 , 65] . More recently , the anterior palate has been described to exhibit a cobblestone-like appearance that depends on Runx1 expression and correlates with fusion competence , and it will be interesting to learn whether this appearance is related to cell extrusion [66] . Though cell extrusion has been typically observed to regulate cellular homeostasis in a proliferating epithelium , this mechanism has also been shown to be required for epithelial morphogenesis during Drosophila dorsal closure [54] . In this case , apoptotic cell extrusion and resultant contraction of the amnioserosa epithelial sheet promotes dorsal closure by pulling epithelial sheets together [54] . The fact that inhibition of apoptosis in explant culture does not interfere significantly with the thinning of the MES suggests it is not likely that apoptosis is the only force driving convergence [20] . Though we are not able to definitively determine the relative contributions of cell extrusion , extrusion-independent apoptosis , and epithelial displacement by convergence to removal of the MES , we did observe that apoptotic cells were at least as commonly part of rosette-like structures as not , indicating that cell extrusion-driven cell death is at least as significant a contributor as extrusion-independent apoptosis . This is likely to be an underestimate of the role of extrusion in fusion , because static quantification may misidentify some cleaved caspase-3 positive cells as not associated with extrusion because ( 1 ) the rosette may already have resolved or ( 2 ) our inability to observe a rosette not aligned with the plane of optical section . In addition , this calculation does not consider live cell extrusion , which we have shown is also likely to be involved . It is notable that in many systems cellular overcrowding induces cell extrusion; during fusion of the secondary palate , the contact and integration of two palatal shelves results in a doubling of cell density . Furthermore , when cultured in amniotic fluid , palatal shelf epithelium was not removed unless contact was made between the shelves [67] , and the anterior secondary palate required contact for initiation of cell death [20] . Consistent with the idea that MES removal is coupled to cell density , culturing with an inhibitor of stretch-activated channels resulted in failed palate fusion , supporting a requirement for crowding-induced extrusion in palate fusion . Taken together , our findings support a novel mechanism of tissue fusion that involves actomyosin-driven convergence , oronasal cell displacement , and cell extrusion .
The animal experiments were performed in accordance with the protocols of the University of California at San Francisco Institutional Animal Care and Use Committee . Myh9lox/lox ( Myh9tm5Rsad , MGI ID: 4838521 ) and Myh10lox/lox ( Myh10tm7Rsad , MGI ID: 4443039 ) mice have been reported previously [68 , 69] and were maintained in a 129/Sv and C57BL/6J mixed genetic background . K14-Cre ( Tg ( KRT14-Cre ) 1Amc , MGI ID: 2445832 ) [37] , Tgfβ3cre ( Tgfb3tm1 ( cre ) Vk , MGI ID: 3768673 ) [45] , ROSA26mTmG ( Gt ( ROSA ) 26Sortm4 ( ActB-TdTomato , -EGFP ) Luo , MGI ID: 3716464 ) [70] , and Lifeact-mRFPruby ( Tg ( CAG-mRuby ) #Rows , MGI ID 4831038 ) [50] mice were each maintained on a C57Bl/6J coisogenic genetic background . Confocal live imaging of explant cultures was used to visualize palate fusion . Culture media ( DMEM/F12 + 20% fetal bovine serum [FBS] + 2mM L-glutamine + 100U/ml Penicillin/100μg/ml Streptomycin + 200μg/ml L-ascorbic acid ) was pre-warmed at 37°C . Recently-adhered E14 . 5 secondary palatal shelves were dissected in culture media and then placed oral surface down in a glass bottom dish ( MatTek ) with media containing 0 . 6% low-melting agarose . The dish cover was sealed with petroleum jelly to prevent evaporation of the media . After the culture media was semi-solidified , the culture dish was mounted on a Leica TCS White light SP5 confocal microscope equipped with a 37°C chamber ( Experiments Figs 1 and 4; S1–S5 Movie ) or a Zeiss Cell Observer spinning disk confocal microscope ( Experiments Figs 5 and 6 , S4 Fig; S6–S11 Movie ) . Time-lapse images were captured with 488 nm and 561 nm laser excitation at the indicated intervals . 10 μM Y27632 ( Cayman chemical ) , 50 μM blebbistatin ( Sigma ) , 6 μM cytochalasin D ( Sigma ) , or 2 μM latrunculin A ( Cayman chemical ) was added to media before imaging . Multiple Z-stack images were obtained using Leica LAS AF software ( for the Leica microscope; Figs 1 and 4; S1–S5 Movie ) or Zeiss Zen software ( for the Zeiss microscope; Figs 5 and 6 , S4 Fig; S6–11 Movie ) . The original live imaging files ( . lif or . czi ) were opened in Imaris image processing software ( Bitplane ) for analysis . Bleaching and attenuation correction were performed and signal levels were adjusted in entire movies . Three-dimensional and time crop functions were used to narrow down the region and time window of interest . For cell tracking ( Figs 1 , 4 and 5; S1 Fig ) , eGFP , tdTomato , and mRFPruby signals were segmented to generate 3-D membrane surfaces . To identify the center points of the cells , surface signals were masked and inverted images were created . Spot detection found each cell based on these signals , and movements of the spots were traced with autoregressive motion algorithms ( S1 Fig ) [71] . To visualize cell intercalation in a single plane , images and traces were projected to the XY dimension . Quantification graphs for anteroposterior ( X ) , mediolateral ( Y ) , and oronasal ( Z ) cell movements were created as Imaris vantage plots . For S3 Movie , surfaces were generated manually in a series of images , and displacement of the center points was traced . In 3-D tracking and 2-D ( XY ) projections , color-coded spots were inserted to indicate the time points when the spot and trace disappeared . For the middle palate images in S4 Movie , the MtrackJ ( ImageJ ) plugin was used to track cell migration [72] . To generate rose diagrams in Figs 1 and 4 , the displacement angles of traced epithelial and mesenchymal cells were calculated relative to the midline using Imaris software and were plotted using Rose . Net software . For the junctional rearrangements , horizontal and vertical cell junction numbers were counted in seven optical Z sections . A junctional line was drawn over the image and the angle relative to the midline was measured using ImageJ . Junctions between -30º and 30º were considered as horizontal junctions . Junctions between 60º and 120º were counted as vertical junctions . Cell distances between randomly selected pairs of neighboring cells in the same Z section were measured at each time point . Cell intercalation events by neighbor cell exchange were counted in seven Z sections . In Fig 6 , cell extrusion events were measured in both live imaging and static immunostained MES sections . Caspase3-positive cells were counted to examine how many cells are undergoing apoptosis in the rosette structure . Nitrocellulose filters ( Millipore ) were placed on triangular metal grids mounted in 6 cm center well culture dishes ( Corning ) . A pair of secondary palatal shelves was dissected at E13 . 5 and placed in close proximity on the filter paper , which was submerged in culture media ( BGjb + 100 U/ml Penicillin/100 μg/ml Streptomycin + 200 μg/ml L-ascorbic acid ) with the oral side upward . The center well dish was placed inside of a 10 cm dish on top of gauze soaked with water to maintain humidity . DMSO , 10 μM blebbistatin ( Sigma ) , 10 μM Y27632 ( Cayman chemical ) , 50 μM ML-7 ( Calbiochem ) , or 50 μM GdCl3 was added to the media . Explants were cultured at 37°C with 5% CO2 for 72 h for fusion analysis or 48 h for immunostaining . Culture media were replaced every day . Explants were transfected with pools of four siRNA sequences at 500 nM for Myh9 , Myl9 , or a pool of four scrambled RNAs ( ON-TARGETplus SMARTpool , Thermo Scientific ) using Lipofectamine 2 , 000 ( Life Technology ) diluted in BGjb media to knockdown gene expression . Target sequences were as follows Myh9-05: AAAUUCAUUCGUAUCAACU , Myh9-06: GAGGCACGAGAUGCCACCC , Myh9-07: UUUGGAAACGCCAAGAGGU , Myh9-08: GUAUCAAUGUGACCGACUU and Myl9-09: GCGACCGAUUCACGGAUGA , Myl9-10: CGAGAUGUACCGCGAGGCA , Myl9-11: CCCAAAGGCAAGAUGUCGA , Myl9-12: AUAAGGAGGACCUGCACGA . Palate explants were cultured with siRNA for 72 h and media was changed every day . Fusion scores were analyzed as previously described [73] . Palate fusion score was scored from 1 ( incomplete fusion ) to 5 ( complete fusion ) . The score was determined based on how much of the MES cell layer was retained after 72 h in culture as follows: Score 1 = A complete multi-cell layer of MES cells remained between two palatal shelves; Score 2 = A mostly continuous MES layer is persistent in the midline of the seam with some regions of a single-cell layer and few breaks; Score 3 = Sections exhibit around 50% removal of the MES; Score 4 = Few MES islands are present; Score 5 = Fusion is complete , with no MES cells remaining . SiRNA knockdown was verified by quantitative real time ( qRT ) -PCR using the BioRad CFX 96 Real-Time PCR detection system . Gapdh was used as a reference gene to normalize the level of expression . QRT-PCR primer sequences were as follows: Myh9 , Forward 5’-GGCCCTGCTAGATGAGGAGT-3’ and Reverse 5’-CCGGCATAGTGGATAATGCAGA-3’; Myl9 , Forward 5’-ACAGCGCCGAGGACTTTTC-3’ and Reverse 5’-AGACATTGGACGTAGCCCTCT-3’; Gapdh , Forward 5’-CACTGAGCATCTCCCTCACA-3’ and Reverse 5’-TGGGTGCAGCGAACTTTATT-3’ . Embryos were dissected at specified stages and fixed in Bouin’s solution for at least 24 h . Embryos were dehydrated through a graded series of ethanol , embedded in paraffin and sectioned at 7 μm thickness prior to staining with hematoxylin and eosin . Fusion score counting was performed using the same criteria as in explant fusion assay [73] . Embryos were fixed with 4% paraformaldehyde in phosphate-buffered saline ( PBS , pH 7 . 4 ) overnight and submerged in 12 . 5% , 25% sucrose , and 25% sucrose/OCT solutions sequentially . Fixed embryos were embedded in OCT compound ( Tissue-Tek ) and sectioned at 12 μm thickness using a cryostat ( Fisher Scientific ) . Tissue slides were kept at -20°C . In situ hybridization was performed with digoxygenin-labeled Myh9 , Myh10 , Myh14 , and Myl9 antisense or sense probes according to standard methods . Immunostaining was performed using the following primary antibodies: anti-rabbit NMHCIIA ( Covance , 1:500 ) , Phalloidin-conjugated Alexa 488 ( 1:200 ) , anti-rat E-cadherin ( Invitrogen , 1:300 ) , anti-rabbit p63 ( Abcam , 1:200 ) , anti-rabbit Ki67 ( Thermo Scientific , 1:200 ) , and anti-rabbit cleaved caspase-3 ( Asp175 ) ( Cell signaling , 1:200 ) . Anti-rabbit IgG-Cy3 , anti-rat Cy2 , or Dylight 649 ( Jackson Immunoresearch Lab ) were used as secondary antibodies to visualize the signals . Images were captured using a Zeiss Axio Imager Z2 microscope or a Zeiss Cell Observer spinning disk confocal microscope .
|
Tissue fusion , the process by which two independent prominences become united to form one continuous structure , is common during development , and its failure leads to multiple structural birth defects . In this study , we directly examine the cellular and molecular mechanisms by which tissue fusion occurs using the mouse secondary palate as a model . Using live imaging , we find that fusion of the secondary palatal shelves proceeds by a progression of previously undescribed cell behaviors . Cellular protrusions and establishment of contacts between palatal shelves leads to the formation of a transient multicellular epithelial structure that then converges toward the midline , driven by cell intercalation . This convergence occurs together with displacement of the epithelium and epithelial cell extrusions that squeeze epithelial cells out from between the palatal shelves and mediate continuity of the structure . We show that in mice this morphogenesis requires an actomyosin contractility pathway culminating in non-muscle myosin IIA activation . Altogether , these data support a new model for tissue fusion during mouse embryogenesis in which convergence , displacement , and cell extrusion drive the union of independent structures .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Convergence and Extrusion Are Required for Normal Fusion of the Mammalian Secondary Palate
|
Venezuelan equine encephalitis ( VEE ) is a re-emerging , mosquito-borne viral disease with the potential to cause fatal encephalitis in both humans and equids . Recently , detection of endemic VEE caused by enzootic strains has escalated in Mexico , Peru , Bolivia , Colombia and Ecuador , emphasizing the importance of understanding the enzootic transmission cycle of the etiologic agent , VEE virus ( VEEV ) . The majority of work examining the viral determinants of vector infection has been performed in the epizootic mosquito vector , Aedes ( Ochlerotatus ) taeniorhynchus . Based on the fundamental differences between the epizootic and enzootic cycles , we hypothesized that the virus-vector interaction of the enzootic cycle is fundamentally different from that of the epizootic model . We therefore examined the determinants for VEEV IE infection in the enzootic vector , Culex ( Melanoconion ) taeniopus , and determined the number and susceptibility of midgut epithelial cells initially infected and their distribution compared to the epizootic virus-vector interaction . Using chimeric viruses , we demonstrated that the determinants of infection for the enzootic vector are different than those observed for the epizootic vector . Similarly , we showed that , unlike A . taeniorhynchus infection with subtype IC VEEV , C . taeniopus does not have a limited subpopulation of midgut cells susceptible to subtype IE VEEV . These findings support the hypothesis that the enzootic VEEV relationship with C . taeniopus differs from the epizootic virus-vector interaction in that the determinants appear to be found in both the nonstructural and structural regions , and initial midgut infection is not limited to a small population of susceptible cells .
Venezuelan equine encephalitis virus ( VEEV ) has been recognized as an etiologic agent of neurologic disease in humans and equids for nearly 80 years . Closely related to eastern ( EEEV ) and western equine encephalitis viruses ( WEEV ) , VEEV belongs to the family Togaviridae , genus Alphavirus . First recognized in the 1920s , Venezuelan equine encephalitis ( VEE ) outbreaks are typically episodic with several years elapsing between outbreaks . However , when outbreaks do occur , they can cause severe and sometimes fatal disease in hundreds-of-thousands of equids and humans . For instance , after an interval of 19 years with no documented cases between 1973 and 1992 , clusters of cases emerged in Venezuela [1] and Chiapas , Mexico [2] prior to a major outbreak involving ca . 100 , 000 people in 1995 [3] . In general , disease manifestations of VEE range from flu-like illness to fatal encephalitis . It is estimated that central nervous system ( CNS ) involvement occurs in 4–14% of human cases , and children are at the greatest risk to develop encephalitis and to die from infection [4] . Of the four subtypes of VEEV , IC and IAB are considered epizootic as they are known to cause disease in horses , to use these hosts for amplification , and are also capable of utilizing a variety of epizootic mosquito vectors , such as Aedes ( Ochlerotatus ) taeniorhynchus , A . ( Och . ) sollicitans , Psorophora confinnis , Culex ( Deinocerites ) pseudes , Mansonia indubitans , and M . titillans , among others [5]–[10] . Many of these mosquitoes thrive near coastal brackish water , can fly long distances from larval sites , prefer to feed on humans or other large mammals , and can tolerate feeding in sunny areas , although they may rest in shaded sites . In contrast , enzootic VEEV subtypes IE and ID generally cause little or no viremia or disease in equids , but like the epizootic strains , can cause fatal disease in humans [11]–[14] . Mosquito vectors that maintain these enzootic viruses in nature include a variety of species within the Spissipes section of the subgenus Culex ( Melanoconion ) , and subtype IE strains specifically utilize C . ( Mel . ) taeniopus . The enzootic cycle typically occurs in shaded , intact forests with stable pools of water that are available for larval development . Some larvae also require the presence of a specific aquatic plant ( i . e . , Pistia spp . ) for respiration [15] . Recent identification of extensive endemic disease in Peru , Bolivia , Ecuador , Colombia and Mexico , caused by spillover of enzootic strains in subtypes ID and IE , indicates the importance of VEEV as a continuous public health threat in Central and South America [16] , [17] . The recent documentation of widespread endemic disease is likely associated with increased surveillance as well as the clearing of sylvatic forest habitats to accommodate the expansion of agricultural land types in areas of Latin America where enzootic VEEV persists [18]–[20] . The resulting fragmentation of sylvatic habitats results in an increase in ecotones that can support the life cycle of enzootic VEEV mosquito vectors [21] , which also increases the likelihood of an enzootic VEEV strain adapting to epizootic transmission [22] . Enzootic ID strains are known to be a source for the emergence of epizootic IC strains and this emergence has occurred on multiple occasions [1] , [23] . While IE strains had not been associated with the emergence of epizootic strains before 1993 , recent outbreaks of epizootic-like IE strains were found to infect epizootic mosquito vectors and cause disease in equids [2] , [24] . Historically , IE VEEV strains have been found in isolated sylvatic transmission cycles between C . taeniopus mosquitoes and rodent hosts , such as cotton rats ( Sigmodon spp . ) , spiny rats ( Proechimys spp . ) and other rodent species , including Liomys salvini and Oligoryzomys fulvescens [25]–[27] . Phylogenetic studies of IE strains show that they diverged from other subtype I VEEV viruses , including enzootic ID strains [28] , indicating that IE strains have long been established and most likely isolated within their enzootic habitats for at least centuries . Examination of the low threshold for infection and specificity of IE strains for C . taeniopus vectors suggests that IE stains have co-adapted to be highly fit for replication in and transmission by this vector [29]–[31] . The stable , enzootic VEEV IE-C . taeniopus relationship is in sharp contrast to the transient interaction that occurs between epizootic virus strains and mosquito vectors during sporadic outbreaks . However , the majority of experimental studies examining VEEV-vector interactions have utilized epizootic vectors as models . We hypothesize that IE viruses are highly adapted to their enzootic vector through a long-term evolutionary relationship such that the dynamics of infection of IE viruses within their vector differ inherently from those observed in epizootic virus-vector interactions . Reverse genetic studies of epizootic IC VEEV indicate that infection determinants reside within the E2 glycoprotein gene [21] , [24] , [28] , [32] . We hypothesized that the transient nature of the epizootic virus limits its infection determinants to a localized region of the genome to allow for rapid adaptation to a competent vector , whereas the enzootic infection determinants are not limited to a single region in the structural portions of the genome due to the long adaptation of the genome to infection and replication within C . taeniopus . To test this hypothesis , we generated four chimeric VEEVs ( Fig . 1 ) , using a strain with a known high susceptibility to C . taeniopus ( i . e . , subtype IE strain 68U201 ) and a strain known to be poorly infectious for C . taeniopus [i . e . , subtype IAB Trinidad donkey ( TrD ) strain] . These chimeras allowed us to discern the contributions of the structural and nonstructural protein regions as well as the 3′ untranslated region ( UTR ) in infection and dissemination in C . taeniopus . We also examined the initial midgut infection dynamics of the enzootic mosquito model as compared to what has been previously observed in the epizootic model with IC VEEV and A . taeniorhynchus . There is only a small population of VEEV-susceptible midgut cells in A . taeniorhynchus , and thus the midgut infection is initiated by a very small number of infected cells and presumably virions [32] . Evolutionary theory would suggest that a bottleneck in the population of replicating viral genomes might deleteriously affect viral fitness through Muller's ratchet [33]–[35] . However , epizootic strains might regain fitness through recombination [32] . While this is a plausible strategy for an epizootic virus , which only interacts transiently with its mosquito vector during an outbreak , the enzootic virus must maintain a certain level of fitness to persist in nature over centuries or longer and repeated bottlenecks would likely be highly detrimental . We therefore hypothesized that most or all midgut epithelial cells in C . taeniopus are susceptible and , therefore , the population of enzootic VEEV that infect the midgut epithelium does not undergo a severe bottleneck during the infection of the midgut . To examine this hypothesis , we utilized viral-like particles ( VLP ) to establish the number , distribution , and susceptibility of midgut epithelial cells initially infected in the IE enzootic model .
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Research Council . The protocol was approved by the Institutional Animal Care and Use Committee of the University of Texas Medical Branch ( IACUC Protocol # 0209068 , approved July 13 , 2010 ) . Plaque , cytopathic effect ( CPE ) assays , and replication curves were performed on Vero ( African green monkey kidney ) , and BHK-21 ( baby hamster kidney ) cells were used for electroporation to rescue parental and recombinant viruses as well as replicon particles from transcribed RNA . Both cell types were propagated in Dulbecco's modified eagle medium ( DMEM ) supplemented with fetal bovine serum ( FBS ) and penicillin/streptomycin . For CPE assays of mosquito samples , amphotericin B ( 50 µg/mL ) ( Sigma-Aldrich , St . Louis , MO ) was added to the DMEM . Cells from an A . albopictus mosquito cell line , C6/36 , maintained in DMEM media supplemented with 10% FBS , penicillin/streptomycin , and 1% tryptose phosphate broth ( Sigma-Aldrich , St . Louis , MO ) were utilized for in vitro replicon co-infection experiments and replication curves . Viruses used for this study were derived from infectious cDNA clones V3000 IAB Trinidad Donkey ( TrD ) ( kindly provided by Nancy Davis and Robert Johnston ) [36] and IE 68U201 [37] . Prior to the generation of the V3000 clone , this TrD strain had been passaged once in guinea pig brains and 14 times in embryonated eggs . The 68U201 isolate had been passaged once in newborn mice and two times in BHK-21 cells prior to the construction of the clone . From these clones , four chimeric variants were developed: two with matching cis-acting elements and two with mismatched elements ( Fig . 1B and C ) . Two IE replicons , 68UGFP and 68UCFP , were derived from a full length IE 68U201 clone as previously described ( Fig . 2 ) [32] . Replicons are replication deficient VLPs that can be utilized to analyze the initial sites of infection without the complication of cell-to-cell spread . These particles were generated by electroporating two RNA species simultaneously . The replicon , consists of the nonstructural open reading frame expressing a fluorescent reporter and associated cis-acting elements; the helper contains the structural portions of the genome . Co-electroporation of these two RNAs generates deficient particles that are unable to package the structural genes , but continue to express only the nonstructural genes from the replicon packaged into the particle . Replicons and helpers were transcribed using a T7 mMessage mMachine ( Ambion , Austin , Texas ) , electroporated into BHK-21 cells , and harvested after 24 hours . The first two chimeric clones were derived with mismatched cis-acting elements to directly compare the roles of the structural and nonstructural protein cassettes in mosquito infection and dissemination . Specifically , IAB/IE had the 5′ UTR and nonstructural protein gene region derived from IAB TrD and the structural protein gene region and 3′ UTR derived from IE 68U201 . The reciprocal version , IE/IAB , had the 5′ UTR and nonstructural protein gene region of IE 68U201 and structural protein gene region and 3′ UTR derived from IAB TrD . Fusion PCR utilizing Phusion High Fidelity Polymerase ( Finnzymes , Lafayette , CO ) and designed around the Tth111I restriction enzyme site ( Fig . 1 ) in the 26S UTR for each parental virus was used to generate a PCR fragment joining the two different viral cDNAs . Initially , for each reciprocal chimera , two overlapping fragments that encompassed the fusion site of the two genomes were generated by PCR using a forward primer from within the nsP4 region ( 7041 F IAB AND 6509 F IE ) with a reverse fusion primer ( IAB/IE R and IE/IAB R ) and reverse primer downstream of the junction site ( 8007 R IAB and 8312 R ) paired with a forward fusion primer for each chimera ( IAB/IE F and IE/IAB R ) ( Table 1 ) . The two individual fragments were joined by a PCR reaction on both templates utilizing the outermost primer sets . The fusion PCR fragment was cleaved with respective restriction enzymes ( BssHII and PspOMI for IAB/IE and Bsu36I and NheI for IE/IAB ) and ligated to the two other cDNA fragments with T4 DNA ligase ( New England Biolabs , Beverly , MA ) . Ligated fragments were transformed into One Shot OmniMAX cells ( Invitrogen , Carlsbad , CA ) , and resulting colonies were screened and sequenced prior to cesium chloride ( CsCl ) plasmid DNA purification . The IAB/IE and IE/IAB constructs were then utilized to generate the infectious clones with matching cis-acting elements: IAB/IE/IAB and IE/IAB/IE . For both IAB/IE/IAB and IE/IAB/IE chimeras , a fusion PCR was designed at the junction at the end of the structural protein gene region and the start of the 3′ UTR . As described above , two PCR amplicons were generated using primers 10191 F IE , IAB/IE 3′ UTR R , IAB/IE 3′ UTR F , and 12030 R for IAB/IE/IAB and 9528 F IAB , IE/IAB 3′ UTR R , IE/IAB 3′ UTR F , and 12030 R for IE/IAB/IE ( Table 1 ) . The two fragments were ligated and then cleaved with restriction enzymes ( SpeI and SacII for IAB/IE/IAB and SgrAI and EcoRI for IE/IAB/IE ) to generate a single cloning fragment . These clones were ligated in 3 fragments , transformed , purified , and sequenced as described above . Prior to transcription , plasmids were linearized with either NotI ( V3000 backbone ) or MluI ( 68U201 backbone ) restriction enzymes [37] , [38] . RNA was generated using the mMessage T7 RNA Polymerase Kit in the presence of an analog cap ( Ambion , Austin , TX ) . The yield and integrity of transcripts were evaluated by agarose gel electrophoresis directly prior to electroporation . BHK-21 cells were electroporated using previously described conditions [39] . Virus was harvested at 48 hours post-electroporation when CPE was observed in greater than 80% of the cells . Virus titers were determined by plaque assay on Vero cells . Replication kinetics of each of the two parental strains and four chimera strains were compared on Vero and C6/36 mosquito cells to identify any deficiencies and compare to in vivo infection and dissemination in C . taeniopus . Cells were seeded at a concentration of 106 cells/well in six well plates and allowed to attach and settle for 4 hours . Monolayers were infected in triplicate at a multiplicity of infection ( MOI ) of 5 PFU/cell and allowed to incubate for one hour at 37°C . Following incubation , cells were washed 3 times with phosphate-buffered saline ( PBS ) , and overlaid with complete DMEM . Media were collected and stored from each well and replaced with the same volume of fresh media at predetermined time points , followed by plaque assays to measure viral yield . To compare the viral replication curves , a two-way ANOVA test and post-hoc multiple comparisons test with a Bonferroni correction was performed using JMP software , version 8 . 0 . 2 ( SAS Institute Inc . , Cary , NC ) . P-values≤0 . 05 were considered significant . Two parental viruses ( IAB , IE ) and four chimeras ( IAB/IE/IAB , IE/IAB/IE , IE/IAB , IAB/IE ) were evaluated for their ability to infect and disseminate in C . taeniopus . The C . taeniopus colony was established from mosquitoes collected from Chiapas , Mexico in 2007 as described previously [40] . For all studies , 10-week old female CD1 mice ( Charles River Laboratories ) were used as viral hosts . To develop natural viremia , mice were infected with 3 log10 PFU of each virus [41] by subcutaneous ( SC ) inoculation , held for 24 hours , anesthetized by intraperitoneal ( IP ) administration of sodium pentobarbital ( 50 mg/kg ) , and bled via the retro-orbital sinus to determine viremia levels . Since the replicon particles utilized for this study do not replicate beyond the initial cell infected , and C . taeniopus will not feed on artificial bloodmeals , we utilized an artificial system in which we inoculated CD1 mice intravenously ( IV ) allowing for an immediate nonreplicative viremia . Mice were anesthetized by IP inoculation of sodium pentobarbital and 200 µl of a stock replicon or a 1∶1 mix of replicons was inoculated into the tail vein . Particles were allowed to circulate for 1–2 minutes before blood was collected from the retro-orbital sinus to estimate the artificial viremia level achieved; the animal was then exposed to mosquitoes for ca . one hour , after which blood was collected again from the retro-orbital sinus to detect any changes in the circulating replicon concentration . Following exposure , engorged mosquitoes were sorted and incubated for 14 days at 28°C with 75–80% humidity . A 10% sucrose solution was provided ad libitum . Statistical analysis of rates of infection and dissemination were broadly examined using a contingency analysis , and specific 2×2 comparisons were evaluated using Fisher's exact test with JMP software ( SAS Institute Inc . , Cary , NC ) . P-values≤0 . 05 were considered significant . Viral titers of rescued viruses and animal sera were determined by plaque assay on Vero cells . Following the 14-day extrinsic incubation period ( eip ) , legs and wings were removed from mosquitoes and stored at −80°C . Samples were triturated , centrifuged at 9500× G for 5 minutes , and used to infect monolayers of Vero cells in CPE assays . Triturated body samples that generated CPE were indicative of an infected mosquito , while legs and wings were used to detect a disseminated infection . Replicon titration was done in a similar manner to the plaque assay; ten-fold serial-dilutions were plated on a monolayer of Vero cells and allowed to incubate for one hour prior to an overlay with DMEM supplemented with FBS and penicillin/streptomycin . After 24 hours , the media were removed and the monolayer was fixed with 4% paraformaldehyde ( PFA ) ( Affymetrix , Santa Clara , CA ) for one hour . The number of fluorescent cells per well was counted using an Olympus Is71 inverted fluorescent microscope and reported as fluorescence units ( FU ) . Mosquito samples were processed 72 hours after blood feeding to minimize chances of damaging the midgut while distended with blood and to allow for clear images of the midgut epithelia . Mosquitoes were cold anesthetized and submerged for 30 seconds to 1 minute in 70% EtOH prior to being transferred to a PBS solution . Midguts were extracted and covered with a drop of 4% PFA on a glass slide for 30 minutes , then was rinsed twice with PBS before the addition of ProLong Gold Antifade with DAPI ( Invitrogen ) . Mosquito midguts were imaged on an Olympus BX61 fluorescent microscope and high-resolution images were taken on an Olympus FluoView FV1000MPE confocal microscope . In vitro dual infection experiments were visualized on an Olympus DSU-IX81 spinning disk confocal microscope and analyzed with MetaMorph Software ( Molecular Devices , Sunnyvale , CA ) .
Figure 1 shows the genetic composition of the viruses utilized in this study . The two parental viruses included subtype IE strain 68U201 and subtype IAB strain TrD , which share 77% nucleotide and 89 . 8% amino acid identity . Four chimeric strains were derived from these parental strains: IAB/IE/IAB and IE/IAB/IE were designed with matching 5′ and 3′ cis-acting sequence elements . IAB/IE and IE/IAB were designed with mismatched cis-acting elements , where the 3′ UTR matched the strain used for the structural protein region of the chimera and the 5′ UTR matched the strain used for the nonstructural protein region of the chimera . It has been shown repeatedly that conserved regions in both the 5′ and 3′ UTRs of alphaviruses are essential for proper synthesis of both negative and positive strand RNA species [42]–[45] . Therefore , chimeras with both matching and mismatching cis-acting elements were utilized to compare these regions of interest and their effect on replicative efficiency . One-step replication curves of the parental and chimeric strains were performed on Vero cell monolayers at an MOI of five PFU/cell to identify any replication deficiencies that could bias experimental findings in the mosquito model . All of the chimeras showed similar replication , with no major deficiencies when compared to the parental strains ( Fig . 3A ) . However , analysis of variance indicated that the replication of the viruses was significantly different ( p<0 . 0001 ) . Multiple comparison tests showed that strain TrD exhibited higher replication levels at multiple time points ( not shown ) , which does not likely correlate to the in vivo mosquito model because this strain is unable to infect and disseminate in C . taeniopus [40] , [46] . One-step replication analyses were also performed on monolayers of C6/36 A . albopictus cells to compare to the in vivo mosquito model ( Fig . 3B ) . No major replication deficiencies were observed; however , analysis of variance did indicate differences among the viruses ( p<0 . 0001 ) . Unlike the Vero cell replication curves where just the TrD strain differed from all other viruses , differences were seen between nearly all viruses upon pairwise comparisons ( not shown ) . The only three pairs out of the total 15 comparisons that did not show any statistical differences from one another were between IE versus IE/IAB/IE , IAB/IE/IAB versus IE/IAB , and IE/IAB/IE versus IE/IAB . Adult female C . taeniopus were exposed to a range of oral doses for each of the parental and chimeric strains of VEEV and tested for infection and dissemination into the hemocoel following a 14-day eip ( Table 2 ) . Two pairs of chimeras with matched and mismatched cis-acting elements were utilized to independently evaluate the roles of the nonstructural and structural polyprotein open reading frames as well as the 3′ UTR in mosquito infection and dissemination . As expected , the parental IAB TrD virus was unable to infect C . taeniopus at blood meal titers as high as 6 . 2 log10 PFU/ml , which is in agreement with previous work [46] , [47] . Similarly , as predicted based on previous studies [30] , [40] , [46] , C . taeniopus mosquitoes were highly susceptible to infection with the parental subtype IE 68U201 strain at oral doses as low as 4 . 2 log10 PFU/ml . All four chimeras showed an intermediate ability to infect and disseminate in C . taeniopus when compared to the parental IAB and IE strains ( Table 2; Fig . 4 ) . The effect of the exposure dose on infection rate was evaluated by contingency analysis for each of the chimeric viruses ( IAB/IE/IAB , IE/IAB/IE , IAB/IE , and IE/IAB ) and found to be significant for each ( p<0 . 05; p<0 . 001; p<0 . 05; p<0 . 0001 , respectively ) ( Fig . 4A ) . In order to compare individual virus strains , a Fisher's exact test was utilized to determine differences in infection rates ( Table 3 ) . As expected , comparisons between the parental viruses and the chimeric viruses were all highly significant ( p<0 . 0001 ) , with the exception of the comparison between the IE parental virus and IE/IAB chimera , for which the IE strain had a less notable infectious advantage than the chimera ( p<0 . 0071 ) . Interestingly , infection rates did not differ significantly among three of the four chimeras: IAB/IE , IAB/IE/IAB , and IE/IAB/IE . However , IE/IAB showed a significantly higher infection rate when compared to each of the other three chimeras ( p<0 . 0001; p<0 . 0037; p<0 . 0025 , respectively ) . Although each chimera showed the ability to disseminate into the hemocoel after midgut infection , the dissemination rates among the chimeras were low overall ( Figs . 4B and 4C ) ; therefore , no transmission experiments were performed . Previous studies of alphaviruses as well as other arboviruses have shown that infected mosquitoes often have virus restricted to the midgut , which is likely explained by a commonly recognized but poorly understood barrier to viral escape of the mosquito midgut [46] , [48]–[50] . While dissemination rates increased as the exposure dose was increased , the overall rates of dissemination were too low to perform reliable statistical analysis . Fisher's exact tests comparing the rates of infected mosquitoes with dissemination ( Fig . 4C ) , showed no differences between the four chimeras . To observe the number of epithelial cells initially infected , the location of the infected cells , and to determine whether there is a subpopulation of cells within the midgut that is more susceptible than other epithelial cells , C . taeniopus mosquitoes were exposed to a range of doses of 68U201 replicon particles expressing fluorescent proteins ( Fig . 2 ) . For the single replicon infections , a clear dose-response was observed such that the lowest oral dose ( average of pre- and post- exposure titers ) of 3 . 0 log10 FU/ml infected only 11% of examined midguts with only 2 infected cells/midgut , whereas the highest dose of 7 . 2 log10 FU/ml infected 100% of examined midguts with a range 535–1757 infected cells/midgut ( Table 4 ) . Infected cells were not limited to any particular region of the abdominal midgut ( Fig . 5 ) , and only a minority of the midguts ( 9% ) were found to have infection focused in the posterior portion , whereas 25% showed a focused infection in the anterior portion of the abdominal midgut . The remaining 66% of infected midguts showed a mixed infection with concentrated infection within the middle portion of abdominal midgut . Infection of the midgut/foregut junction was not observed . To determine if there was differential susceptibility of midgut cells , C . taeniopus mosquitoes were orally infected with a 1∶1 mixture of 68UGFP and 68UCFP . A total of fifteen mosquitoes was examined for co-infection at two different doses . The low exposure dose achieved by artificial viremia was a mixture of 5 . 4 log FU/ml 68UGFP and 5 . 0 log FU/ml of 68UCFP , and the high dose achieved was 6 . 5 log FU/ml of each replicon . At the low dose , an average of 70 midgut epithelial cells were infected with 68UGFP and an average of 52 cells was infected with 68UCFP . At the high dose , the average number of cells infected with 68UGFP was 896 , whereas the average number of 68UCFP infected cells was 866 . At the low dose and of the five co-exposed mosquitoes examined , no co-infected cells were observed . At the high dose where 15 midguts were examined , there appeared to be 2–3 cells with co-localization; however , it was determined that these areas of co-localization were a result of either signal bleed-through or overlap . Even so , there was still an average of less than one observed co-infected cell per midgut in the highest dose group .
As human populations continue to expand into rural environments , the incidence of emerging and re-emerging zoonotic pathogens will continue to climb . This has already been observed with other arboviral viruses , such as chikungunya , dengue , yellow fever , and Japanese encephalitis viruses [51] . Similar trends have also been observed with enzootic strains of VEEV that have caused endemic disease as well as outbreaks in Peru , Central America , and Mexico [16] , [17] . Historically , studies of VEEV emergence have focused on epidemic strains within subtypes IAB and IC; however , enzootic ID and IE strains can also cause a large burden of endemic disease , which can often be misdiagnosed as dengue fever [52] . Recent studies have also shown that the primary mosquito vector of enzootic IE , C . taeniopus , can be an efficient vector of newly emerged epizootic IE strains in Mexico [40] , [53] . Considering the growing risk of enzootic VEEV strains in causing human disease , it is important to understand the determinants and dynamics for viral infection of the primary enzootic vector , which we hypothesize to be different from what is known about the epizootic virus-vector interaction . We first examined the genetic determinants of infection and dissemination utilizing chimeric viruses to analyze the molecular determinants for VEEV specificity to the enzootic mosquito vector , C . taeniopus . We used two viruses with distinct phenotypes for these chimeras , to help identify the major genome regions that contribute to specific infection of the enzootic vector . However , because these viruses have such a wide genetic divergence ( 10 . 2% at the amino acid level ) , extrapolation of this method to clarify the roles of each gene during enzootic mosquito infection could be prone to bias from incompatibilities between open reading frames within each chimera ( Table 5 ) . Therefore , to ensure that chimerization did not result in general attenuation of virus replication , the parental and chimeric strains were evaluated using in vitro replication curves on Vero and C6/36 cell monolayers; no replication deficiencies were observed . It was noted that the replication in mosquito cells was different from what was observed in the in vivo mosquito model in that the parental IAB virus , which showed no deficiencies in vitro , was unable to infect the in vivo model . Similarly , the differences observed between the IE parental and the four chimeras in the in vivo model were not demonstrated in the in vitro model . These results emphasize the importance of using an in vivo mosquito model to detect differences in viral replication , which may not be detected in a mosquito cell line . We orally exposed C . taeniopus mosquitoes to varying doses of two parental strains , subtype IAB TrD and subtype IE 68U201 , as well as four chimeric strains , IAB/IE/IAB , IE/IAB/IE , IAB/IE , and IE/IAB , and evaluated the role of the nonstructural and structural protein genes and the 3′ UTR as determinants of infection and dissemination in C . taeniopus . We hypothesized that , unlike the epizootic virus strains and their vectors , the genetic determinants for enzootic infection include multiple genes and they are not limited to a single region in the structural portion of the genome . Our data supported this hypothesis , as all four chimeras were able to infect and disseminate in C . taeniopus , albeit at rates lower than the wild-type IE parental strain . We included multiple replicates within each viral group to compensate for variations within the mosquito colony . If the E2 or the structural regions were the primary determinants of infection and dissemination , those chimeras with IE-derived structural regions would have infected mosquitoes at a higher rate than those chimeras with IAB-derived structural regions , and this was not observed . We anticipated that the chimeras with mismatched 3′UTR regions would show diminished infection and dissemination rates based on previous alphavirus studies examining the effects of mismatched cis-acting elements [54] , [55]; however , our IE/IAB chimera showed a significantly higher rate of infection than that of the other three chimeras . This suggests that the 3′ UTR plays some role in infection of the enzootic vector . A closer examination of the effect of the 3′ UTR on infection showed that the chimera with IAB structural genes and IAB 3′ UTR ( IE/IAB ) had the highest infection rate , while the chimeras with a mixed structural-3′ UTR makeup ( IE/IAB/IE or IAB/IE/IAB ) had intermediate infection abilities , and the chimera with IE in the structural and the 3′ UTR ( IAB/IE ) actually had the lowest rate of infection . This suggests that 3′ UTR acts in concert with other portions of the genome , although it is unclear which specific regions are important for this cooperative effect . These potential interactions should be further explored with 3′UTR-specific chimeras , such as a IAB virus backbone with a IE derived 3′UTR and a IE backbone with a IAB-derived 3′ UTR . While the role of these regions was not mirrored by in vitro mosquito infections , our C6/36 data were based on cells from A . albopictus , which in laboratory experiments has been shown to be equally susceptible to epizootic IC and enzootic ID VEEV strains [56] . Previous studies examining chimeras between Ross River virus ( RRV ) and Sindbis virus ( SINV ) , two genetically distant alphaviruses , have shown that mismatched 3′ UTR regions can result in depressed RNA synthesis in vitro , although the effects on replication in vivo have not been examined [55] . However , studies of chimeras between more closely related alphaviruses , such as o'nyong-nyong ( ONNV ) and chikungunya ( CHIKV ) viruses , indicate that chimerization does not have a deleterious affect on the infection of the CHIKV mosquito vector , A . aegypti . There was also no indication that mismatched 3′ UTRs altered infection rates [57] . There were no statistical differences in the infection rates between chimeras IAB/IE , IAB/IE/IAB , and IE/IAB/IE , indicating that both the structural and nonstructural protein regions of the enzootic virus play a role in vector infection , as none of the chimeras displayed infection rates as high as the parental IE strain . However , we observed a trend in which the two chimeras with IE-derived nonstructural protein genes showed higher rates of infection at higher doses . Specifically , the chimeras with IE-derived nonstructural protein genes reached 100% infection at the highest doses , while the chimeras with IAB-derived nonstructural protein gene regions never reached 100% infection even at the highest doses . The diminished infection of all chimeras implies that there are multiple determinants of infection that reside in different genome regions and may act synergistically . Our results show that the determinants for infection of the enzootic vector do not reside solely in the structural protein genes , specifically not only in the E2 glycoprotein of the genome , which supports our hypothesis that infection determinants for VEEV in the enzootic mosquito vector relies on both structural or nonstructural protein regions of the genome . Interestingly , our findings suggest that in the enzootic model , the nonstructural elements are stronger determinants of vector infection . We also hypothesized that the characteristics of initial midgut infection of the enzootic mosquito vector would be inherently different than those used by the epizootic virus in A . taeniorhynchus . To test this hypothesis , we exposed the enzootic vector , C . taeniopus , to replicon particles generated from a subtype IE enzootic strain . Examination of the initial sites of infection in the midgut indicated multiple locations in the abdominal portion with no predilection for either the anterior or the posterior region; we detected no infection of the cardial epithelium at the midgut/foregut junction . Similar to what was observed in A . taeniorhynchus , a clear response was observed between the oral dose and the number of infected midgut cells , although the ID50 for C . taeniopus was lower and the maximum number of infected cells was higher than the 100 susceptible A . taeniorhynchus cells previously estimated [32] . The greater number of infected cells ( >1700 ) in C . taeniopus following high oral doses indicates that a larger number of its midgut epithelial cells is susceptible to VEEV IE infection compared to A . taeniorhynchus and VEEV IC . This observation , in conjunction with our observation of no co-infected midgut cells in the mixed replicon experiments , supports the hypothesis that the population size of enzootic VEEV virions during initial infection of the midgut is not severely restricted by a limited number of susceptible C . taeniopus epithelial cells . The average population of cells infected by the 68UGFP replicon at the highest dose did not differ from the average number of cells infected by the 68UCFP replicon . This suggests that there is no effect of co-exposure on individual particle infection rates . Utilizing the same methods , previous studies in the epizootic VEEV/mosquito model found an average of 26 midgut cells co-infected with two replicons , which was greater than what we observed in the enzootic model . This indicates that the initial infection of the enzootic vector differs from that of the epizootic VEEV strain . Using the Poisson distribution and given the epizootic data ( a model with a small population of susceptible cells ) , we determined the probability of observing less than a single co-infected cell out of our five C . taeniopus midgut replicates to be 5 . 1×10−12 , indicating an extremely low likelihood that there is a subpopulation of midgut epithelial cells with enhanced susceptibility . Our studies illustrate the contrast in the virus-vector interactions between the enzootic and epizootic VEEV cycles . Not only do these interactions persist in different ecological cycles and infect different species of mosquitoes , but they also behave differently within their respective vectors . This difference may be explained by the dissimilar selective pressures that are exerted on each subtype during transmission . For example , epizootic viruses produce a very high level of viremia in equids , which facilitates VEEV transmission by epizootic vectors even if only a small population of their midgut cells are initially infected . However , the highly susceptible enzootic vector , which appears to have a greater number of susceptible midget cells that can be initially infected even after small oral doses , can transmit efficiently among populations of rodents that develop only moderate viremia titers [25]–[27] . As the growing impact of enzootic VEEV on human health is becoming more apparent , especially after the recent emergence of epizootic-like IE strains , understanding how these viruses interact with vectors is critical to estimating their threat to human health and for refining public health prevention strategies as well as developing vaccines . For instance , the design strategy of a vaccine that is protective against epizootic and enzootic strains that are currently causing human disease must also consider mosquito vectors that could potentially acquire and transmit should a vaccinee become viremic . If the epizootic vector only has a few susceptible midgut cells and is examined for competence for a given vaccine strain , it may appear to be incompetent . However , the same vaccine may be able to establish an infection in the enzootic vector . Considering that the determinants for infection appear to differ between the two vector types , vaccine strains that are derived from epizootic VEEV and depend on the elimination of mosquito infection may not necessarily reflect how infectious these vaccine candidates would be for enzootic vectors . As enzootic habitats are encroached upon and enzootic cycles gain close proximity to epizootic habitats , it is essential to consider the contribution of enzootic vectors to viral emergence and the potential introduction of vaccine strains into natural cycles .
|
Venezuelan equine encephalitis virus ( VEEV ) is transmitted to humans and horses by mosquitoes in Mexico , Central and South America . These infections can lead to fatal encephalitis in humans as well as horses , donkeys and mules , and there are no licensed vaccines or treatments available for humans . VEEV circulates in two distinct transmission cycles ( epizootic and enzootic ) , which are differentiated by the ecological niche that each virus inhabits . Epizootic strains , those that cause major outbreaks in humans and equids , have been studied extensively and have been used primarily to develop and test several vaccine candidates . In this study , we demonstrate some important differences in the roles of different viral genes between enzootic/endemic versus epizootic VEEV strains that affect mosquito infection as well as differences in the way that enzootic VEEV more efficiently infects the mosquito initially . Our findings have important implications for designing vaccines and for understanding the evolution of VEEV-mosquito interactions .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"ecology",
"viral",
"transmission",
"and",
"infection",
"vector",
"biology",
"virology",
"biology",
"microbiology"
] |
2012
|
Genetic and Anatomic Determinants of Enzootic Venezuelan Equine Encephalitis Virus Infection of Culex (Melanoconion) taeniopus
|
How the molecular mechanisms of stress response are integrated at the cellular level remains obscure . Here we show that the cellular polarity machinery in the fission yeast Schizosaccharomyces pombe undergoes dynamic adaptation to thermal stress resulting in a period of decreased Cdc42 activity and altered , monopolar growth . Cells where the heat stress-associated transcription was genetically upregulated exhibit similar growth patterning in the absence of temperature insults . We identify the Ssa2-Mas5/Hsp70-Hsp40 chaperone complex as repressor of the heat shock transcription factor Hsf1 . Cells lacking this chaperone activity constitutively activate the heat-stress-associated transcriptional program . Interestingly , they also exhibit intermittent monopolar growth within a physiological temperature range and are unable to adapt to heat stress . We propose that by negatively regulating the heat stress-associated transcription , the Ssa2-Mas5 chaperone system could optimize cellular growth under different temperature regiments .
Sensing ecological parameters and mounting an appropriate adaptive response allows organisms to thrive in changing habitats . Free-living microorganisms , such as yeast , can maintain regular patterns of growth over a relatively wide range of environmental temperatures . Cellular adaptation to increased temperature is one of the most robust and evolutionarily conserved mechanisms ( reviewed in [1] ) . This response involves temperature-induced activation of genes encoding so-called heat shock proteins ( HSPs ) [1]–[4] . Molecular chaperones constitute a predominant class of HSPs and their up-regulation is essential for maintaining protein homeostasis at elevated temperatures ( reviewed in [1] ) . Although much of heat shock-associated transcription is transient , recent analyses indicate deviation from this dynamic pattern and existence of genes with expression constitutively regulated by temperature [1] , [5] , [6] . One of the early events in the heat stress pathway ultimately setting off an extensive downstream gene expression program involves activation of critical transcriptional regulators , the heat shock factors ( HSFs ) [7] , [8] . HSFs are evolutionary conserved winged helix-turn-helix proteins that bind to cis-acting DNA promoter sequences called heat shock elements ( HSEs ) [9] . HSFs are essential for viability in many fungi and control important developmental processes in higher eukaryotes suggesting that they may regulate basal transcription in addition to their function in stress response [10]–[14] . The ability of HSFs to respond to cellular stresses is under negative regulation by chaperones , modulation of nucleocytoplasmic shuttling , post-translational modifications and , in higher eukaryotes , transition from monomeric to trimeric state ( reviewed in [15] ) . HSF repression by the Hsp90 and the Hsp70s-Hsp40 chaperone complexes [16]–[18] could provide a negative feedback loop titrating production of chaperones to allow for optimal protein folding [19] . Although essential for orchestrating an acute response to changing environment , constitutive Hsf1 transcriptional induction is known to retard cellular growth [20] . However , molecular mechanisms reducing HSF activity following the original heat-induced surge are poorly understood . Although fascinating , the cellular biology of adaptation to growth at elevated temperatures is largely unchartered . The fission yeast Schizosaccharomyces pombe ( S . pombe ) are rod-shaped cells that grow at cell tips and divide by placing a medial septum . Upon division , each daughter cell initiates growth at the cell tip inherited from the mother , the so-called old-end ( Fig . 1A ) . During the G2 phase of the cell cycle , cells transition to bipolar growth in a process termed New-End-Take-Off ( NETO ) [21] . NETO occurs when growth sets a sufficient distance between the cell tips [21] , [22] and requires the function of a large protein complex called polarisome and active actin cytoskeleton remodeling [23]–[25] . The polarisome , built around the kelch-repeat protein Tea1 and its binding partner Tea4 , ensures the fidelity of polarized growth through recruitment of downstream polarity factors ( Fig . 1B and [23] , [24] , [26] , [27] ) . Polarized growth requires local plasma membrane ( PM ) and cell wall remodeling . The targeted secretion of cell wall enzymes , such as the β-glucan synthase Cps1 [28] relies on long-range exocytic vesicle transport along actin cables and the tethering of the secretory vesicles to the PM by the exocyst complex [29] . Polarized assembly of both actin cables and the exocyst depends on the Rho family GTPase Cdc42 [30]–[34] . The guanine-nucleotide-exchange-factors ( GEFs ) promote Cdc42 activity while the GTPase-activating proteins ( GAPs ) repress it [35] , [36] . The fission yeast GEFs , Scd1 and Gef1 , are enriched at the cell tips [37] , [38] , while the Cdc42 GAP Rga4 localizes to the lateral cortex [39] . This results in predominant activation of Cdc42 at the cell poles . Recent high-end imaging techniques have uncovered oscillations in Cdc42 activity at the PM suggesting that spatiotemporal activation of Cdc42 is governed by a series of positive and negative feedback loops [40]–[43] . The polarisome and the Cdc42 circuitries appear to intersect at various levels [26] , [39] , [44] . The mechanisms regulating polarized growth exhibit a high degree of plasticity [45] , [46] . Yeast cells readily orient growth towards their mating partners [47] , [48] and relocalize the growth machinery to the site of cell injury [49] . In Saccharomyces cerevisiae ( S . cerevisiae ) , heat stress activates the cell wall integrity pathway and promotes nucleotide exchange of the Rho-family GTPase Rho1 , which in turn initiates F-actin dependent redistribution of β-glucan synthase throughout the entire cellular cortex [50] , [51] . Repolarization of actin as cells adapt to elevated temperature appears to depend on a negative feedback mechanism where the protein kinase C Pkc1 activates MAPK cascade to inhibit the Rho1-GEF [52] . In fission yeast several types of stress , including heat and osmolytes , also cause F-actin redistribution throughout the cellular cortex [53]–[56] and lead to activation of MAP kinases [57]–[59] . Here we utilize a multi-disciplinary approach to explore how the heat stress and associated transcription impact polarized growth in fission yeast . We identify the Mas5-Ssa2 chaperone complex as a negative regulator of the heat shock transcription factor Hsf1 . Using cells lacking the chaperone function we probe the physiological outcome of constitutively upregulated heat stress response .
Lectins bind polysaccharides of the yeast cell wall with high affinity but without impairing cell growth [60] . Thus , sites of cell wall deposition subsequent to fluorophore-coupled lectin staining can be readily visualized as a dip in fluorescence intensity [61] . To monitor cell growth upon heat stress we stained log-phase wild type cells growing at 24°C with FITC-Lectin , washed-out the excess dye and immediately shifted cells to 36°C or allowed them to grow at 24°C ( Fig . 1C ) . Samples were collected at 30-minute intervals , fixed and subjected to epifluorescence microscopy . Cells continuously growing at 24°C elongated at a rate of 2 . 51±0 . 77 mm per hour after the dye washout . Conversely , cells shifted to 36°C for one hour fully arrested longitudinal growth ( Fig . 1C and 1D , n>60 cells per sample , pWelch's t-test = 4 . 10⋅10−17 ) . Interestingly the diameter of heat-stressed cells increased ( Fig . 1C and 1E , width increased by 9±7% , n>60 cells per sample , pWelch's t-test = 7 . 45⋅10−15 ) and the overall FITC-Lectin staining of the cell visibly decreased ( data not shown ) suggesting that the cell wall underwent remodeling throughout the cell periphery rather than at the cell tips . After a period of depolarized cell wall deposition , cells resumed polarized growth and subsequently elongated at a rate of ∼28% greater than cells grown at 24°C ( Fig . 1D ) . Most control cells in the G2 phase of the cell cycle exhibited bipolar growth pattern ( Fig . 1C and 1F ) . In line with previous observations [21] , the majority of heat-stressed cells elongated only at one cell tip ( Fig . 1C and 1F ) . When propagated overnight at 36°C , most wild type cells became bipolar ( 83%; Fig . S1A ) . Thus , the heat stress leads to a transient arrest of longitudinal growth followed by regaining cellular polarity through an intermediate monopolar growth phase . Temperature had no effect on the distribution of the polarisome components Tea1-YFP , Tea4-GFP and Pom1-GFP as these markers exhibited bipolar localization in late G2 phase cells at both 24°C and 36°C within the observed timeframe ( Fig . 1G , S1B , S1C , see Fig . S1F for quantifications ) . In contrast , β-glucan synthase GFP-Cps1 , which predominantly localized to both cell tips in late G2 cells at 24°C , could be observed in intracellular vesicle-like structures and at the lateral cell cortex 45 minutes after cells were shifted to 36°C ( Fig . 1H ) . Longer incubation at 36°C ( 105 minutes ) allowed all cells to repolarize GFP-Cps1 to the cell tips , albeit in a monopolar fashion ( Fig . 1H and S1F ) . Similarly , the exocyst component Sec6-GFP localized to both cell tips in unstressed late G2 phase cells but was observed throughout the cell cortex in cells shifted to 36°C for 45 minutes ( Fig . 1I ) . Cells growing at 36°C for 105 minutes did repolarize Sec6-GFP but approximately half of late G2 phase cells confined it to only one tip ( Fig . 1I and S1F ) . Polarization of actin structures typical for yeast cells grown at normal temperatures was completely abolished following incubation at 36°C for 45 minutes . Cells were able to re-polarize F-actin cytoskeleton after 105 minutes at 36°C but it was unequally distributed between the two tips of most late G2 phase cells ( Fig . S1D , S1F and S1G ) . Similarly , both fluorophore-tagged formin For3 and the type-V myosin Myo52 relocalized from the tips to the entire cortex upon heat stress but then re-established polarity after 105 minutes at 36°C , although in a monopolar fashion ( Fig . 1J , S1E and S1F ) . Taken together our results suggest that processes confining the growth machinery to the polarisome-demarcated cell tips are rapidly perturbed by temperature increase . Cells become monopolar in the adaptive phase following the heat stress but eventually resume normal bipolar growth . Both F-actin distribution and exocyst localization are governed by the activity of a Rho-family GTPase Cdc42 [29] . To monitor the active Cdc42 in vivo during heat-stress we used the GFP-tagged CRIB ( Cdc42/Rac-interactive binding ) domain reported to interact with the GTP-bound Cdc42 [39] . G2 phase cells grown at 24°C focus the Cdc42 activity to the cell tips with moderate oscillations in its levels ( [40] and Fig . 2A , top panel ) . Minutes after cells were shifted from 24°C to 36°C , CRIB-GFP was seen spreading from the cell tips towards the cell equator establishing zones of Cdc42-activity at the cell sides ( Fig . 2A , middle panel ) . After a period of depolarization , cells restricted CRIB-GFP to just one tip and initiated monopolar growth ( Fig . 2A , bottom panel ) . In contrast to the bipolar localization in almost all of G2 phase cells grown at 24°C , a third of G2 phase cells grown at 36°C for 105 minutes exhibited a strictly monopolar CRIB-GFP pattern ( Fig . 2B and S2A ) . While total fluorescence signal of CRIB-GFP was not perturbed by temperature , its cortical association was decreased as compared to unstressed cells ( Fig . 2C ) . Both Cdc42-GEFs exhibited a bipolar localization pattern in unstressed late G2 cells ( Fig . 2D , 2E and S2A ) . Forty-five minutes upon temperature shift from 24°C to 36°C , Gef1-3YFP localized to patches distributed throughout the plasma membrane whereas Scd1-GFP was no longer detectable at the cell cortex ( Fig . 2D and 2E ) . Both Gef1-3YFP and Scd1-GFP also localized to intracellular structures upon heat stress . After 105 minutes at 36°C , both GEFs became confined to cell tips but in a monopolar fashion ( Fig . 2D , 2E and S2A ) . Interestingly , Gef1-3YFP and Scd1-GFP levels were severely diminished at the cortex of cells repolarizing growth at elevated temperature . The average intensity of Scd1-GFP at a single cell tip decreased by 49±23% ( pWelch's t-test = 1 . 43⋅10−7 , n>25 cells per sample ) , whereas Gef1-3YFP signal dipped by 65±17% ( pWelch's t-test = 7 . 34⋅10−12 , n>25 cells per sample ) . The only known fission yeast Cdc42-GAP , Rga4 , localizes to the lateral cell cortex and to the new cell end in pre-NETO cells [39] . Upon heat stress , the Rga4-GFP localization remained restricted to the cell sides throughout the course of the experiment although its absolute amount at the cortex progressively decreased ( Fig . 2F; after 105 minutes at 36°C the cortex-associated Rga4-GFP signal fell by 34±22% , pWelch's t-test = 7 . 35⋅10−8 , n>25 cells per sample ) . The diminished cortical levels of Cdc42 regulators in cells shifted to the elevated temperature were not a consequence of a gross decrease in total amounts of these proteins ( Fig . S2B ) . Since heat-stressed fission yeast cells eventually resume bipolar growth ( Fig . S1A ) , we analyzed cortical levels of Cdc42-GAP and GEFs in cells growing overnight at 36°C . Gef1-3YFP and Scd1-GFP did assume bipolar localization in late G2 cells ( Fig . S2E–F ) . Surprisingly , the average signal intensity per cell tip remained relatively low , as did Rga4-GFP levels at the lateral cortex . In fact , even though levels of active Cdc42 at the cortex were similar in cells growing at 18°C , 24°C and 30°C ( Fig . S2C and S2D ) , the cortical levels of the three Cdc42 regulators inversely correlated with the growth temperature ( Fig . S2E–G ) . Immunoblot analyses of cells grown to log-phase at 18°C , 24°C , 30°C and 36°C showed a similar trend in total protein levels of Scd1-GFP , Gef1-3YFP and Rga4-GFP ( Fig . S2H ) . Thus , cortical Cdc42 activity appears to be extensively regulated at both initial and adaptive phases of the heat-stress response and during steady state growth at different temperatures . Cells respond to stress by adjusting their transcriptional output [1] , [62] . Heat-stress associated transcription is primarily mediated by a temporary activation of the HSF family of transcription factors ( reviewed in [63] ) , Hsf1 is the only HSF homologue present in the fission yeast genome and it was reported to be essential not only for heat shock response but also for vegetative growth [11] , [64] . We confirmed that cells depleted of Hsf1 ceased growth ( Fig . S3A ) . We wondered if cell polarity could be affected by the Hsf1 transcriptional activity alone , without the temperature change . To this aim , we replaced the fission yeast hsf1 genomic promoter with a strong , thiamine responsive nmt1 promoter . Cells with high levels of Hsf1 eventually arrested growth ( Fig . S3B ) . Overexpression of Hsf1 was sufficient to promote its transcriptional activity since we observed vastly increased levels of the heat stress-induced protein Hsp104 ( Fig . S3C–D ) and the GFP reporter expressed under hsp104 regulatory elements ( Fig . 3A and S3E , see Materials and Methods for details on reporter construction ) . Strikingly , Calcofluor staining revealed that Hsf1-overexpressing cells became increasingly monopolar ( Fig . 3B and 3B; on average the new-end growth decreased by 59±22% in nmt1:hsf1 cells as compared to wild type cells , pWelch's t-test = 4 . 58⋅10−10 , n>50 cells per sample ) . Furthermore , cells overexpressing Hsf1 exhibited decreased tip association of CRIB-GFP that predominantly localized to only one cell tip ( Fig . 3D and 3E ) . Taken together , it appears that high Hsf1 activity prevents normal growth patterning and eventually leads to growth arrest in fission yeast . Hsp90 and Hsp70-Hsp40 chaperones are known repressors of Hsf1 activity but both Hsp90 and Hsp70 proteins are known to engage a large number of client proteins [65] . To isolate specificity factor ( s ) for Hsp70 that could function in this process , we screened the deletion library of sixteen non-essential genes encoding nucleocytoplasmic Hsp40 chaperones ( Supplementary Table S2 ) . Two screening strategies were employed . One approach was based on our finding that Hsf1 overexpression leads to decreased growth rates and eventually a growth arrest . We hypothesized that mutant strains with elevated levels of Hsf1-associated transcription would display decreased growth rates already at 24°C and would not be able to grow at 36°C due to high Hsf1 activity . Within the deletion library of nonessential Hsp70-binding proteins only the strain lacking the DnaJ chaperone Mas5 exhibited such growth pattern ( Fig . 4A ) . The second screening strategy exploited the fact that a mild heat stress increases survival of cells exposed to a subsequent severe transient heat shock , likely due to higher expression of protective factors in pre-stressed cells [66] . As outlined in Fig . 4B , wild type cells and deletion mutants were grown to log-phase at 24°C and culture aliquots were either allowed to continue growth at 24°C or shifted to 50°C for 15 minutes before returning them to 24°C . Additionally , a sample of wild type cells was shifted to 36°C for 45 minutes before being exposed to 50°C . Survival was measured by assessing the number of colony forming units in each culture ( Fig . 4B ) and confirmed by a dilution spotting assay ( Fig . 4C ) . Wild type and most tested mutant strains exhibited a survival rate below 0 . 5% when shifted from 24°C to 50°C for 15 minutes ( n>200 CFU per strain ) . Conversely , 29% of wild type cells survived if a mild heat stress preceded the severe heat shock . Interestingly , cells lacking Mas5 showed an 8% survival rate when shifted directly from 24°C to 50°C for 15 minutes ( Fig . 4B and 4C ) . Both major nucleocytoplasmic Hsp70 chaperones , Ssa1 and Ssa2 , physically interacted with Mas5 as shown by co-immunoprecipitation experiments ( Fig . 4D and S4A ) . Cells lacking Ssa2 but not Ssa1 exhibited increased survival following a transient heat shock , similar to mas5Δ cells ( Fig . 4B and 4C ) . In line with this , the fluorophore-tagged marker proteins Hsp104 , Swo1 and Psi1 that are known to be strongly upregulated upon heat stress were highly abundant in mas5Δ and ssa2Δ but not in ssa1Δ cells , again suggesting that Ssa2-Mas5 complex may have a specific function in heat stress response ( Fig . S4B and data not shown ) . Excluding the possibility that elevated fluorescence levels were due to selective stabilization of heat-responsive proteins in mutant cells , the UTRsHsp104:GFP reporter was also strongly elevated in both ssa2Δ and mas5Δ but not in ssa1Δ cells ( Fig . 4E ) . Thus , mas5Δ and ssa2Δ cells express high levels of heat stress markers already at 24°C , which may increase their chances of survival following transient heat shock . Doubling time of log-phase mas5Δ and ssa2Δ cultures growing at 24°C increased approximately 1 . 3-fold and 2-fold as compared to the wild type ( Fig . S5A left panel , pWelch's t-test = 1 . 61⋅10−2 and 7 . 16⋅10−3 , n = 3 ) . Moreover , cells lacking Mas5 and Ssa2 divided at a consistently shorter length as compared to the wild type ( Fig . S5A , right panel ) . Calcofluor staining suggested that both ssa2Δ and mas5Δ cells exhibited significantly decreased new-cell-end growth ( Fig . 5A and 5B , on average , mas5Δ and ssa2Δ cells reduced the new-end growth by 50±24% and 78±93% respectively , pWelch's t-test = 2 . 28⋅10−28 and 2 . 46⋅10−11 , n>60 per genotype ) . Progression through the cell cycle was not affected in mas5Δ and ssa2Δ mutants , as evaluated by FACS ( Fluorescence Activated Cell Sorting ) analysis ( Fig . S5B ) , suggesting that the monopolar growth in these mutants could not be explained by delaying entry into the G2 phase of the cell cycle . Subcellular distribution of the polarisome components Tea1-YFP , Tea4-GFP and Pom1-GFP was unaffected by mas5 gene deletion ( Fig . S5C–S5F ) . In contrast , localization of the β-glucan synthase GFP-Cps1 was predominantly monopolar late into the G2 phase of the cell cycle ( Fig . S5C and S5G ) . Similarly , exocyst subunit Sec6 localized to only one cell tip in a majority of mas5Δ cells ( Fig . S5G and S5K ) . Imaging of LifeAct-GFP [67] showed that the actin cytoskeletal structures were also enriched at just one cell tip of late G2 phase mas5Δ cells ( Fig . S5I , S5C , S5J ) . Consistently , For3-GFP and Myo52-GFP were also monopolar in most mas5Δ cells longer than 9 µm ( Fig . S5C , S5K and S5L ) . Next , we performed live imaging of wild type , tea1Δ , ssa2Δ and mas5Δ cells grown in a perfusion chamber at room temperature ( Supplemental Movie 1 , see Materials and Methods ) . After an initial phase of monopolar growth , the newly born wild type cells underwent NETO and assumed bipolar growth pattern clearly seen in the kymograph ( Fig . 5C ) . As expected [27] tea1Δ cells maintained a constitutively monopolar growth pattern and were rarely seen undergoing NETO ( Fig . 5C ) . Consistent with the data presented earlier , over 90% of ssa2Δ and mas5Δ cells grew only at the old cell tip . Surprisingly , all mutant cells exhibited an intermittent growth pattern ( Fig . 5C , n>10 cells ) with periods of complete growth arrest lasting from 5 to over 30 minutes followed by growth bursts of 5 to over 30 minutes . Active Cdc42 , as visualized by CRIB-GFP , also localized to one cell tip in cells lacking Mas5 or Ssa2 ( Fig . 5D , 3% of wild type , 93% of ssa2Δ and 58% of mas5Δ late G2 cells were monopolar ) . Furthermore , the absolute levels of CRIB-GFP signal at the polar cortex were diminished in ssa2Δ and mas5Δ cells ( Fig . 5E ) . The time-lapse spinning-disk confocal microscopy revealed that the CRIB-GFP recruitment to the cell tips was unstable in mas5Δ and ssa2Δ cells ( Fig . 5F ) . In wild type cells , CRIB-GFP signal at the continuously growing old cell end displayed only minor intensity oscillations during the course of imaging . On the other hand , CRIB-GFP levels at the new cell end underwent major on-off oscillations prior to NETO but eventually stabilized exhibiting only minor oscillations ( Fig . 5F , left panel , also see [40] ) . However , CRIB-GFP levels underwent on-off oscillations at both tips throughout the cell cycle in mas5Δ and ssa2Δ cells . We observed that high CRIB-GFP levels at the old cell end coincided with periods of tip growth ( Fig . 5F , middle and right panels ) . CRIB-GFP levels at the new cell end of ssa2Δ and mas5Δ cells underwent on-off oscillations that would rarely stabilize to allow for noticeable tip elongation during the course of imaging . These results suggest that the intermittent growth pattern of ssa2Δ and mas5Δ cells is a consequence of the unstable Cdc42 activity at the cell tips . Furthermore , the decreased Cdc42 activity at the new cell end could explain the monopolar growth pattern of mutants lacking Ssa2-Mas5 chaperone complex . The low levels of active Cdc42 in mas5Δ cells prompted us to explore the distribution of the Cdc42-GAP and GEFs in the mutant . Both Gef1-3YFP and Scd1-GFP exhibited bipolar localization in majority of late wild type cells but were predominantly monopolar in cells lacking Mas5 ( Fig . S5M , S5N and S5C ) . The Cdc42-GAP , Rga4-GFP , remained restricted to the lateral cell cortex in ∼90% in mas5Δ cells ( Fig . S5O ) . Levels of cortically associated Cdc42 regulators appeared decreased in mas5Δ cells as compared to the wild type . The average intensity of Gef1-3YFP at a cell tip was decreased by 38±28% whereas Scd1-GFP signal diminished by 53±27% ( pWelch's t-test = 2 . 51⋅10−6 and 7 . 94⋅10−7 respectively , n>25 cells per sample ) . A 20±22% decrease in cortically associated levels was also observed for the Rga4-GFP in cells lacking Mas5 ( pWelch's t-test = 2 . 71⋅10−4 , n>25 cells per sample ) . The total protein levels of the three examined Cdc42 regulators were also diminished in mas5Δ cells ( Fig . S5P ) . Gef1-3YFP and Scd1-GFP levels decreased by 18% and 48% respectively and a 21% decrease was observed for Rga4-GFP . These results suggested that the lack of Ssa2-Mas5 chaperone complex resulted in monopolar growth due to insufficient function of the Cdc42 GTPase module . Indeed , mild overexpression of the constitutively active Cdc42-G12V mutant but not wild type Cdc42 in mas5Δ cells led to its enrichment at the cell tips ( Fig . S5Q ) and a significant increase in bipolar growth ( Fig . 5G and 5H , p = 4 . 65⋅10−7 ) . Interestingly , upregulation of Cdc42 activity also largely rescued the intermittent growth phenotype of mas5Δ cells , as observed by time-lapse microscopy ( Fig . 5I , continuous growth was sustained in 11/20 cells overexpressing Cdc42-G12V and only 3/20 cells overexpressing the wild type Cdc42 ) . We wondered if the abnormal monopolar growth observed in cells lacking Mas5-Ssa2 chaperone complex could be a manifestation of an upregulated heat stress response . We investigated the gene expression profiles of mas5Δ , ssa2Δ and heat-stressed wild type S . pombe cells by RNAseq analysis ( See Materials and Methods for experimental details ) . Interestingly , we observed a striking similarity between their gene expression programs - majority of genes that >2-fold up- or down-regulated in the chaperone mutants exhibited similar regulation in heat-stressed cells ( Fig . 6A ) . Our RNAseq-based profiling of heat stress exhibited remarkable overlap with the microarray data on different abiotic stresses [62] allowing direct comparison between two technology platforms ( Fig . S6A ) . The correlation between expression profiles in mas5Δ and ssa2Δ strains and wild type cells under distinct stress conditions confirmed similarities between expression profiles of the chaperone mutants and wild type cells exposed to heat and also suggested some similarities to cadmium-treated cultures ( Fig . 6B and Fig . S6B–C ) . Gene expression linkage analysis identified a cluster enriched in genes reported as exclusively induced or super-induced in response to heat stress ( Fig . 6B , red cluster , p = 2 . 65⋅10−7 ) [62] . Moreover , when we considered genes distinctly induced by different abiotic stresses , only the heat stress-specific genes were consistently upregulated in mas5Δ and ssa2Δ cells ( Fig . 6C , pmas5Δ = 1 . 44⋅10−2 , pssa2Δ = 5 . 25⋅10−6 for 1 . 5-fold cut-off ) . No significant overlaps were detected for genes specifically up-regulated by other types of stress ( Fig . 6C; pbinomial test>0 . 25 for oxidative , osmotic stress and cadmium treatment ) . The microarray and qPCR analysis confirmed that genes specifically induced in response to heat stress were upregulated in mas5Δ cells and genes downregulated during heat-stress exhibited similar behavior in mas5Δ cells ( Fig . S6D; See Supplementary Table S5 for detailed microarray analysis ) . We concluded that the transcriptional profiles of the chaperone mutants qualitatively matched the heat stress gene expression signature . Both mas5Δ and ssa2Δ mutants exhibited differential regulation of most common environmental stress response ( CESR ) genes ( Fig . 6D and [62] ) . Stress response is activated in part by MAP kinase ( MAPK ) signaling through the MAP kinase Sty1 and Atf1 transcription factor [59] , [68] , [69] . Interestingly , the most consistently up- or down-regulated CESR gene clusters were those independent of either Sty1 and/or Atf1 ( Fig . 6D , clusters 4 and 6; Fig . S6E ) . While many Sty1- and Atf1-dependent CESR genes were also differentially regulated in the chaperone mutants ( Fig . 6D ) , we failed to observe synthetic genetic interactions between mas5Δ and cells lacking Sty1 [70] or between mas5Δ and mutants carrying the constitutively activated MAPK kinase allele wis1 . DD [59] . The phenotype of mas5Δ cells was also unchanged in the presence of Polo kinase mutant alleles plo1S402A and plo1S402E known to be defective in some aspects of the MAPK signaling [55] . We next wondered how the gene expression programs of heat-stressed cells and the chaperone mutants compared to that of cells overexpressing Hsf1 . To this end , we performed RNAseq analysis of nmt1:hsf1 cells grown in the absence of thiamine for 20 hours , at a time point when most Hsf1-overexpressing cells exhibited monopolar growth ( Fig . 3 ) . We observed a significant overlap between the sets of genes regulated by Hsf1 overexpression and the heat stress ( Fig . 6E ) . Importantly , almost half of the differentially expressed genes in mas5Δ and/or ssa2Δ cells were similarly regulated in cells overexpressing Hsf1 ( Fig . 6E ) . Many of those genes appeared the heat stress-responsive Hsf1-targets ( Fig . S6F ) . It is possible that genes up-/down-regulated in the chaperone mutants that were not affected by Hsf1 overexpression could be regulated through other pathways . Alternatively , overexpression of Hsf1 could fail to fully recapitulate its native post-transcriptional activation . Taken together , our gene expression analyses imply that cells lacking the Mas5-Ssa2 chaperone complex exhibit elevated levels of heat stress-associated transcription , at least in part due to abnormal activation of Hsf1 . We wondered if polarized growth defects observed in the chaperone mutants were in fact related to high Hsf1 activity . To test this hypothesis , we sought to down-regulate Hsf1 expression in mas5Δ cells by promoter replacement . The nmt81-hsf1 transcriptional shut-off allows full depletion of Hsf1 but the strains carrying this construct do not germinate and therefore are not suitable for genetic analyses ( Fig . S3A and data not shown ) . Thus we replaced the native hsf1 promoter with a recently reported uracil-regulatable purg1 promoter [71] . Repression of the purg1-driven Hsf1 in wild type cells did not lead to growth arrest , suggesting that even in its repressed , basal state , purg1 promoter allows Hsf1 expression sufficient for supporting vegetative growth . In line with this , we did not observe differences in the basal levels of Hsp104-GFP reporter upon repression of purg1-hsf1 in unstressed wild type cells ( Fig . 7A–B ) . However , Hsp104-GFP levels were strongly reduced in mas5Δ cells when the purg1-driven Hsf1 expression was repressed ( Fig . 7A–B ) . Importantly , down-regulation of Hsf1 largely corrected the growth patterning defects of the chaperone mutant cells . Calcofluor staining revealed a significant increase in bipolar growth in mas5Δ cells with repressed purg1-hsf1 expression as compared to those carrying the wild type hsf1 allele ( Fig . 7C–D ) . Curiously , down-regulation of Hsf1 also rescued the intermittent growth phenotype , with mas5Δ purg1-hsf1 cells demonstrating continuous tip extension when the promoter was repressed ( Fig . 7E; 12/20 mas5Δ purg1-hsf1 cells sustained continuous tip extension vs 5/20 mas5Δ cells ) . Taken together , our results indicate that high Hsf1 activity leads to downregulation of Cdc42 module resulting in inability to support normal bipolar growth . Following stress-triggered Hsf1 activation , the Hsp70-Hsp40 chaperone pair Ssa2 and Mas5 could function to repress it , thus allowing cells to reset their transcriptional output and return to normal growth .
Organisms rapidly adjust their physiology in response to environmental stimuli such as changes in nutrient and mate availability , injury and abiotic stresses . A variety of stresses , including heat and exposure to osmolytes , trigger transient depolarization of the actin cytoskeleton and the cell wall remodeling machinery [25] , [50] , [53] , [55] . After a period of adaptation to new conditions , the polarized growth is restored ( Fig . 1 and S1 ) . Our work focuses on this latter phase of stress response when cells resume polarized growth in an altered environment . Following heat stress , the GTP-bound , active Cdc42 is no longer confined to the cell poles and redistributes throughout the entire cortex ( Fig . 2 ) . Cdc42 functions at the vertex of cellular growth ( reviewed in [23] , [72] ) so its mis-localization likely propagates down to the cell wall remodeling machinery by delocalizing exocyst and the actin cytoskeleton ( Fig . 1 and S1 ) . Redistribution of the growth machinery throughout the cortex has been proposed to function in strengthening the cell wall and possibly to fix injuries caused by a temperature increase [49] , [73] , [74] , although we did not readily detect major changes in cell wall appearance in heat-stressed S . pombe cells by electron microscopy ( data not shown ) . The reported diffusion rates of the membrane-bound Cdc42 are relatively low [75] suggesting that mis-localization is likely due to Cdc42 activation occurring throughout the cortex rather than the lateral movement of active GTPase from the cell tips . Overexpression of Gef1 is sufficient to promote Cdc42 activation throughout the entire plasma membrane [76] suggesting that modulation of Gef1 activity and/or membrane recruitment upon heat stress could allow for ectopic activation of Cdc42 . Interestingly , inhibition of the evolutionary conserved NDR kinase Orb6 promotes invasion of Cdc42 into the lateral cortical domain possibly through direct phosphorylation of Gef1 [37] . Activity of GEFs could be also modulated by interaction with plasma membrane proteins sensing the environment or the structural integrity of the cell wall . For instance , in budding yeast the cell wall sensors Wsc1 and Mid2 have been shown to interact with the GEF for Rho1 , Rom2 , and stimulate the Rho1 nucleotide exchange [51] . Since all landmark proteins including Pom1 , the direct modulator of Rga4/Cdc42-GAP activity [39] , remain properly polarized during heat stress response ( Fig . 1 ) , the mechanisms regulating delocalization of active Cdc42 are likely to function independently of the polarisome . Once the initial phase of stress response is over , cells must return to normal growth in new conditions . Curiously , transient depolarization of the growth machinery upon mild heat stress is followed by a prolonged period of monopolar growth ( [21] , Fig . 1 ) . Since the entire cell population becomes predominantly monopolar it means that heat stress could delay both the onset of NETO in newly born cells and re-initiation of persistent bipolar growth in late G2 phase of the cell cycle . The predominantly monopolar growth pattern in cells adapting to heat stress could be possibly explained by the recently developed model correlating the cell length to the bipolar Cdc42 activation [40] . The salient prediction of the model is that reduction of the positive relative to the negative Cdc42 feedback would result in an increased incidence of monopolar cells . Reduction in cortical levels of Cdc42-GEFs in heat-stressed cells could indeed weaken the positive feedback [42] , [77] , [78] . It is known that monopolar growth does result from the perturbed regulation of Cdc42 activity . S . pombe cells lacking Cdc42-GEF Gef1 have been shown to delay NETO [76] . Similarly , mutants in the Cdc42 effector PAK kinase that likely functions as a part of Cdc42 activity feedback [40] are monopolar [79] . The nature of negative feedback is less clear but it is unlikely to be limited to Cdc42-GAP [40] , [41] . Alternatively , the heat-induced chaperones could directly modulate the cytoskeletal dynamics [80] , [81] and in turn , affect the Cdc42 module activity . Of note , cells with high levels of heat stress associated transcription either due to overexpression of the heat shock transcription factor Hsf1 or lacking the chaperone complex Ssa2-Mas5 , also grow in a monopolar fashion and exhibit decreased cortical levels of active Cdc42 ( Fig . 3 and Fig . 5 ) . Unlike in a polarisome-deficient monopolar mutant tea1Δ , the active Cdc42 can be recruited to both tips of mas5Δ cells ( Fig . S5R ) . Yet , only the old cell end that was inherited from the mother is capable of maintaining Cdc42 activity levels sufficient for growth ( Fig . 5F ) . Consistent with an idea that low Cdc42 activity does not allow bipolar growth in cells where the heat stress response is on , mild overexpression of the constitutively active Cdc42 mutant rescued the growth patterning defects of mas5Δ cells ( Fig . 5G–I ) . Once wild type cells attain the new steady state growth at elevated temperature , the cortical activity of Cdc42 returns to normal and the bipolar growth is resumed ( Fig . S2 ) . Interestingly , the abundance of Cdc42 regulators is inversely correlated with environmental temperature ( Fig . S2H ) , which may suggest a degree of temperature compensation in the Cdc42 module . The eukaryotic circadian clock is a typical example of a complex feedback network that buffers the period of oscillations against temperature ( reviewed in [82] ) . The temperature compensation of the Neurospora clock relies on phosphorylation of the key regulator Frq by the casein kinase 2 [83] , an ortholog of the gene isolated as orb5 in the fission yeast morphogenesis screen [84] . Furthermore , Hsf1 itself is known to regulate temperature compensation of the mammalian circadian clock [85] . The decrease in the Cdc42 activity in cells lacking the Ssa2-Mas5 complex could be potentially explained by over-compensation of Cdc42 module in the absence of the actual temperature increase . Our work suggests that in fission yeast , the Hsp40 protein Mas5 functions in repressing Hsf1 activity as a specificity co-factor for the Hsp70 chaperone Ssa2 . Heat shock transcription factors are extensively regulated [63] and both Hsp90 and Hsp70 chaperones have been proposed to function as their repressors [18] , [19] , [86] . We show that transcriptional depletion of Hsf1 indeed suppresses the heat-stress associated transcription in mas5Δ cells and also rescues the polarized growth defects in this chaperone mutant ( Fig . 7 ) . In line with this , the slow growth of budding yeast lacking Ssa1 and Ssa2 is rescued by a partial loss-of-function mutation in Hsf1 [87] . An interesting twist to the chaperone-mediated regulation is that mRNA abundance of these chaperones increases upon heat stress ( 1 . 9-fold and 3 . 3-fold increase , respectively ) , possibly placing Hsf1 activity within a negative feedback loop . The cytosolic Hsp70 chaperones could also act as primary sensors of stress at least in some conditions , e . g . Ssa1 in budding yeast has been shown to release Hsf1 from the inhibitory complex upon binding to the thiol-reactive compounds [88] . The human Mas5 and Ssa2 orthologs have been shown to bind Hsf1 in vitro [89] , an interaction that we were unable to detect in fission yeast , possibly due to its transient nature . Several reports suggested that subcellular distribution of Hsf1 could affect its activity ( surveyed in [90] ) . Interestingly , we found that during the steady state growth in S . pombe , Hsf1 exhibited a predominantly nuclear localization but it was fully exported into the cytoplasm following heat stress ( Fig . S7 ) . However , in the absence of Mas5 and Ssa2 Hsf1 failed to exit the nuclei of heat-stressed cells , suggesting that the chaperone complex could also regulate Hsf1 through controlling its subcellular distribution ( Fig . S7 ) . The temporally resolved studies of Hsf1-driven transcription triggered by heat stress suggest that the initial burst of Hsf1 activity is rapidly followed by its repression [6] , [8] , [62] , [91] . Removing Hsf1 from the nucleus could be one of the mechanisms to attenuate its activity . The Mas5-like Hsp40 chaperones are highly promiscuous in their interactions with client proteins [65] . This property could potentially allow them to serve as sensors of global folding status linking the machineries handling environmental stress response and the normal cellular growth . Indeed , Ydj1 , the budding yeast ortholog of Mas5 , has been proposed to function as a growth rate sensor regulating G1/S entry [92] , [93] . Curiously , bursts of fairly normal tip extension in individual mas5Δ or ssa2Δ cells are interspersed with periods of rest ( Fig . 5 ) , suggesting that one or more factors could become limiting for growth when availability of these chaperones is decreased . Given that the instantaneous growth rate in yeast has been negatively correlated to upregulation of heat-responsive genes [94] , the Hsf1-chaperone interaction could function in a feedback loop sustaining the normal cellular growth . It is worth noting that while both upregulation of Cdc42 activity and partial downregulation of Hsf1 in the chaperone mutant cells rescued the polarity-related phenotypes , neither genetic manipulation ameliorated the slow doubling time of mas5Δ cells . Placing the central transcriptional activator of the heat shock response under negative regulation by chaperones with a wide client base may ensure efficient and highly sensitive means of adapting to frequently changing environment . Moreover , this circuitry appears to be coopted in regulating the normal cell cycle progression as a function of growth rate . By regulating Hsf1 activity through the Ssa2-Mas5 chaperone system , fission yeast cells could optimize Cdc42-driven polarized growth under different temperature regiments . Given that the heat stress associated transcription modulates diverse cellular functions [1] , similar adaptive dynamics are likely to operate in other physiological processes .
S . pombe strains used in this study and their genotypes are listed in Supplementary Table S1 . Growth media and genetic methods were as described in [95] . The actin marker Lifeact-GFP was a kind gift from M . Balasubramanian . The cell wall dyes Calcofluor White and FITC-Lectin were obtained from Sigma-Aldrich ( St . Louis , MO , USA ) . We placed GFP under the control of the hsp104 promoter region ( 1 . 5 kb upstream of the start codon ) and the 3′ UTR ( 0 . 7 kb downstream of the stop codone ) to construct a reporter of heat-activated expression . The construct was integrated into hsp104 chromosomal locus without disrupting the Hsp104 ORF . For Hsf1 overexpression and thiamine-responsive transcriptional shut-off we replaced the hsf1 promoter with nmt1 and nmt81 promoters respectively at the native chromosomal locus . Similarly , for uracil-responsive modulation of Hsf1 expression we replaced the hsf1 promoter with urg1 promoter ( 675 bp upstream of the start codon ) at the native chromosomal locus . FACS was performed on MACSQuant Analyzer ( Miltenyi Biotec GmbH , Bergisch Gladbach , Germany ) using a protocol described by [96] with RNase obtained from Roche ( Basel , Switzerland ) and propidium-iodide from Sigma-Aldrich ( St . Louis , MO , USA ) . Epifluorescence images were collected using mercury lamp as an illumination source with appropriate sets of filters on a Zeiss Axiovert 200M microscope ( Carl Zeiss , Inc . ) using an Plan Apochromat 100X , 1 . 4 N . A . objective lens , CoolSnap camera ( Photometrics , Tucson , AZ ) and Uniblitz shutter driver ( Photonics , Rochester , NY ) under the control of Metamorph software package ( Universal Imaging , Sunnyvale , CA ) . We acquired either single z-planes or whole cell image stacks that consisted of 9 z-sections with 0 . 5 µm spacing . The z-stack maximum or average projection images were obtained by ImageJ 1 . 46b software package ( http://rsb . info . nih . gov/ij/; National Institutes of Health , Bethesda , MD , USA ) . The same setup was also used in DIC microscopy on fission yeast cells grown in liquid YES medium using ONIX perfusion chambers ( CellASIC , Hayward , CA , USA ) under the control of the proprietor software and the flow of 4 psi . Scanning confocal microscopy was performed on a LSM510 microscope equipped with a Plan Apochromat 100X , 1 . 4 N . A . objective lens , a 488-nm argon laser and a 543-nm HeNe laser . We acquired either single z-planes or whole cell image stacks that consisted of 9 z-sections with 0 . 5 µm spacing . The z-stack maximum or average projection images were obtained by ImageJ 1 . 46b software package . Time-lapse fluorescence microscopy images were generated on Nikon TiE system ( CFI Plan Apochromat VC 100XH 1 . 4 N . A . objective ) equipped with Yokogawa CSU-X1-A1 spinning disk unit , the Photometrics CoolSNAP HQ2 camera and a DPSS 491 nm 100 mW and DPSS561 nm 50 mW laser illumination under the control of MetaMorph Premier Ver . 7 . 7 . 5 . We acquired single z-plane images . Imaging was performed on S . pombe cells placed in sealed growth chambers containing 2% agarose YES medium . Longitudinal growth in lectin stained cells was assessed manually as a distance from the cell tip to the lectin signal along the long cell axis . New-cell-end growth was quantified from calcofluor-stained cells as the distance from the birth scar to the proximal cell tip and expressed as a function of total cell length . We employed ANCOVA analysis to assess the statistical significance of variation in new-end growth between cells with different genotypes ( n>30 cells per genotype ) . For simplified presentation we also estimated average new-end length in cells 12–15 µm long while making sure that the average cell length between samples was comparable , p-values were obtained using Welch's t-test . ImageJ in-built line-scan module was used to create an intensity profile along the long axis of the cells expressing CRIB-GFP . The positional information was normalized as the percentage of the cell length . Cellular fluorescence profiles were calculated from background-subtracted images and the nuclear signal was set as one arbitrary unit since it did not vary significantly between samples . In Fig . 2C unscaled fluorescence intensities are reported and were obtained from cells with the same levels of total CRIB-GFP fluorescence . Fluorescence levels of Cdc42 regulators were analyzed from background subtracted images by manually outlining the whole cells and cell tips . Reported relative changes in fluorescence intensities are based on the average calculated values . ImageJ 1 . 46b software package was used for all image quantifications . For co-immunoprecipitation , yeast cells were grown to log phase , harvested and washed with Buffer A ( 6 mM Na2HPO4 , 4 mM NaH2PO4 . H2O , 150 mM NaCl , 2 mM EDTA , 50 mM NaF , 0 . 1 mM Na3VO4 , protease inhibitor cocktail ( Roche ) ) . Cell pellets were resuspended in 200 µL of buffer A , mixed with glass beads and homogenized in a Mini Bead Beater ( Biospec , Bartlesville , OK , USA ) at 4°C . Total cell lysate were harvested and centrifuged ( 16 , 000× g , 10 min ) to remove cell debris . Soluble fractions were adjusted to same total protein concentration using Buffer A and 350 µl were incubated with either mouse anti-myc ( Milipore , Billerica , MA , USA ) antibodies and Protein A beads ( Invitrogen , Carlsbad , CA , USA ) or just GFP-Trap beads ( ChromoTek , Munich , Germany ) for 1 hour . Beads were washed 6 times with 1 mL of buffer A and resuspended in 50 µL of SDS-Loading buffer . For total protein quantifications , cells were pelleted and resuspended in 200 µl of BufferA ( 1 . 85M NaOH , 1M β-mercaptoethanol , 5× EDTA-free Complete Protease Inhibitor Cocktail ( Roche ) ) and kept on ice for 10 min . We added 450 µl of water and 350 µl of 6 . 1N trichloroacetic acid . The samples were incubated in ice for 10 min and centrifuged for 10 min at 4°C . The obtained pellet was washed with 0 . 5 M Tris-base , dissolved in gel loading buffer and kept at 65°C for 10 min and 95°C for 3 min . Protein samples were subjected SDS-PAGE and standard Western Blotting . Proteins of interest were probed by mouse anti-GFP ( Roche ) , rabbit anti-HA ( Roche ) and mouse anti-myc ( Milipore ) . Rabbit anti-HistoneH3 ( Abcam , Cambridge , MA , USA ) probing served to monitor sample loading . IRDye800 conjugated anti-mouse and IRDye700 conjugated anti-rabbit antibodies were used prior to analysis on the Odyssey Infrared Imaging system , all purchased from LI-COR Biosciences ( Lincoln , NE , USA ) . Cells were grown in standard YE ( wild-type , mas5Δ , ssa2Δ ) or MM media ( wild type , nmt1:hsf1 without thiamine , nmt81:hsf1 with thiamine ) at 24°C . A sample of wild type cells was also shifted from 24°C to 36°C for 15 minutes . RNA was extracted using Qiagen RNeasy Mini Kit ( Qiagen , Venlo , Netherlands ) following manufacturer's protocol . Samples were sequenced by single end 50 bp sequencing at the University of Utah's Bioinformatics and Microarray Next Generation Sequencing Shared Resources ( Salt Lake City , Utah , USA ) after Illumina TruSeq RNA Sample Prep with oligo-dT selection . The RNA sequencing results were analyzed using Galaxy software package [97]–[99] . Briefly , upon quality control using the FastQ Groomer [100] module , reads were mapped to the fission yeast genome using the Tophat module [101] and differential gene expression was assessed by the Cufflinks module [102] . Analyses of differential gene expression are presented in Supplementary Table S4 . The relevant raw RNAseq data have been deposited in NCBI's Gene Expression Omnibus [103] and are accessible through GEO Series accession number GSE50156 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE50156 ) . The RNA samples from mas5Δ and wild type cells grown at 24°C or shifted to 36°C for 45 min were prepared using phenol-chlorophorm extraction described by [104] and reverse transcribed using SuperScript III ( Invitrogen ) and random decamer primers ( N10 ) at 50°C for 60 min . The reverse transcriptase was inactivated at 70°C for 10 min . cDNA samples were analyzed by quantitative real-time PCR ( qPCR ) using a StepOnePlus real-time PCR system ( Applied Biosystems ) and Fast SYBR Green Master Mix ( Applied Biosystems ) with 0 . 15 µM of each forward and reverse primers . The following cycling program was used: 95°C for 20 s , followed by 45 cycles of a three-step reaction , denaturation at 95°C for 5 s , annealing and extension at 60°C for 45 s . The primers used are listed in Supplementary Table S3 . Data shown were normalized to the expression levels of actin ( act1 ) and similar results were obtained when using GAPDH ( gpd1 , data not shown ) to normalize the data . The gene expression profile of mas5Δ cells was obtained as previously described by [104] . The significance of differential gene regulation between wild type and mas5Δ cell was assessed by the Welch's t-test ( Supplementary Table S5 ) . Only genes with p<0 . 05 and fold change >1 . 5 were included in subsequent analysis . The expression profiles for stress conditions were taken from [62] . Clustering analysis was performed using Gene Cluster [5] using uncentered correlation similarity metrics average-linkage method . The visualization was done using TreeView software [105] . Correspondence at the top plot was derived from [106] . Briefly , the genes were sorted into a list according to how pronounced their up/down-regulation was in mas5Δ as compared to wild type cells . The binomial test was then employed to generate p-values as a function of an increasing number of genes included into the comparison between mas5Δ expression profile and that of stressed cells .
|
Heat stress , caused by fluctuations in ambient temperature , occurs frequently in nature . How organisms adapt and maintain regular patterns of growth over a range of environmental conditions remain poorly understood . Our work in the simple unicellular yeast Schizosaccharomyces pombe suggests that the heat stress-associated transcription must be repressed by the evolutionary conserved Hsp70-Hsp40 chaperone complex to allow cells to adapt the polarized growth machinery to elevated temperature .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Hsp70-Hsp40 Chaperone Complex Functions in Controlling Polarized Growth by Repressing Hsf1-Driven Heat Stress-Associated Transcription
|
Neurotrophic interactions occur in Drosophila , but to date , no neurotrophic factor had been found . Neurotrophins are the main vertebrate secreted signalling molecules that link nervous system structure and function: they regulate neuronal survival , targeting , synaptic plasticity , memory and cognition . We have identified a neurotrophic factor in flies , Drosophila Neurotrophin ( DNT1 ) , structurally related to all known neurotrophins and highly conserved in insects . By investigating with genetics the consequences of removing DNT1 or adding it in excess , we show that DNT1 maintains neuronal survival , as more neurons die in DNT1 mutants and expression of DNT1 rescues naturally occurring cell death , and it enables targeting by motor neurons . We show that Spätzle and a further fly neurotrophin superfamily member , DNT2 , also have neurotrophic functions in flies . Our findings imply that most likely a neurotrophin was present in the common ancestor of all bilateral organisms , giving rise to invertebrate and vertebrate neurotrophins through gene or whole-genome duplications . This work provides a missing link between aspects of neuronal function in flies and vertebrates , and it opens the opportunity to use Drosophila to investigate further aspects of neurotrophin function and to model related diseases .
In vertebrate brain development , neurons are produced in excess , and surplus neurons are eliminated through apoptosis ( cell death ) , adjusting innervation , targeting , and connectivity to target size [1] . Neurotrophins ( NTs ) are the major class of molecules promoting neuronal survival in vertebrates . They also control cell proliferation and neuronal differentiation , and they are required for axonal and dendritic elaborations , synaptic plasticity , excitability , and long-term potentiation ( LTP , the basis of memory and learning ) [2–5] . NTs underlie most aspects of vertebrate nervous system development and function , and abnormal NT function is linked to psychiatric disorders [6–9] . NTs are the key molecules linking nervous system structure and function in vertebrates [2 , 3] . Despite such fundamental roles , NTs have been missing from invertebrates . There is compelling evidence that neurotrophic factors exist in Drosophila . As in vertebrates , about half the neurons die in the fruit fly central nervous system ( CNS ) during embryogenesis [10] . Apoptosis occurs in most neuroblast lineages [11 , 12] , and there is dramatic hyperplasia in mutant embryos lacking programmed cell death [13] . In multiple CNS contexts , the survival of subsets of neurons and glia requires long-range , nonautonomous support . For instance , there are no glial cells of retinal origin; glia enter the retina through the optic stalk , and if they are defective , such as in repo mutants , retinal neurons die in excess [14] . In disconnected mutants , the optic lobes ( where the retinal photoreceptor neurons project to in the brain ) degenerate . When mosaic clones of disconnected mutant cells are generated in the brain optic lobes in otherwise normal flies , retinal neurons die [15] . Lack of connectivity at the optic lobe also results in massive optic lobe neuronal death [16 , 17] due to abnormal function originating from the retina rather than the brain [16 , 17] . A trophic factor for retinal neurons is predicted to emanate from the brain optic lobe glia [16 , 17] . In the embryonic CNS , upon glial ablation or mutations in glial cells missing , there is excess neuronal apoptosis [18] . Glia are also produced in excess: most dramatically , in the embryo , 75% midline glia and a small subset of longitudinal glia die during axon guidance ( prior to homeostatic functions of glia ) [19–24] . Identified gliatrophic factors include the neuregulin homolog Vein [24] and the TGFα homolog Spitz [19 , 25 , 26] , both ligands of EGFR , and the ligands of the PDGR homolog PVR [27] . Other properties commonly assigned to complex brains and to NT function , such as synaptic plasticity , LTP , and complex behaviour , all occur in flies . However , no neurotrophic factor has been identified in Drosophila . The NTs comprise brain-derived neurotrophic factor ( BDNF ) , nerve growth factor ( NGF ) , NT3 , and NT4/5 ( plus NT6/7 in fish ) and bind the Receptor Tyrosine Kinases TrkA , -B , -C , the atypical TNFR superfamily member p75 , and Integrin α9β1 [28–30] . Pro-NTs bind p75 to promote cell death , and mature NTs bind Trk and p75 receptors , or p75 alone , to promote cell survival [3 , 28 , 30] . Vertebrate NTs bind Trks to activate the MAPKinase/ERK and AKT pathways ( promoting cell survival ) , PLC-γ ( regulating calcium levels ) , and NFκB ( promoting cell survival ) [3 , 30] . Binding of NTs to p75 independently of Trks results in cell death or cell survival , through JNK and NFκB , respectively [30] . In an evolutionary context , p75 is more ancient than the Trks [30] . The most conserved NT among vertebrates is BDNF , and BDNF mutations correlate with epilepsy , anxiety , depression , attention deficit disorder , autism , and other cognitive and psychiatric disorders ( e . g . , [6–9] ) . NTs underlie an endogenous mechanism of CNS repair [31] , and disregulation of NGF underlies chronic pain ( e . g . , in cancer ) [32] . Drosophila is a very powerful model organism used to understand gene networks and model disease; however , a surprising void has been the lack of NT studies in flies . NT ligands and receptors have been identified throughout the invertebrate deuterostomes ( Figure S1 ) . There are functional Trk receptors in the lancelet Amphioxus [33] , and p75 and Trk orthologs have been identified in sea urchin and acorn worm [34–36] . Searches of sequenced genomes have revealed NTs in all deuterostome groups , represented by Amphioxus NT ( Bf-NT ) , acorn worm NT ( Sk-NT ) , and sea urchin NT ( Sp-NT ) [34 , 37 , 38] ( Figure S1 and Table S1 ) . In protostome invertebrates , a bona fide Trk ( in the snail Lymnea ) and an atypical Trk ( in the snail Aplysia ) , have also been identified in molluscs [35 , 36 , 39] . The function of these ancient NTs and receptors is unknown . These findings indicate that NTs are more ancient in evolution than previously thought , although no NT has been identified in protostomes . The presence of NTs in flies has been controversial . Structural and biochemical features of the Drosophila protein Spätzle ( Spz ) revealed an NGF domain [40 , 41] . However , a parallel similarity to horseshoe crab coagulogen , involved in the blood-clotting cascade [41] , overshadowed that earlier finding . An initial computational analysis of the sequenced genomes based on BLAST searches declared lack of NTs in flies [42] . However , this simple BLAST search missed 30% of Drosophila genes and would have missed any proteins with structural conservation despite sequence divergence . Structural predictions have confirmed that Spz belongs to the NT superfamily [43] . There are to date no functional studies of Spz in the CNS , so whether it plays neurotrophic roles is unknown . To investigate whether a NT may underlie some of the structural and functional aspects of the insect nervous system , we searched the sequenced Drosophila genome for NT sequences . We show here that Drosophila Neurotrophin 1 ( DNT1 ) is a NT superfamily member that promotes neuronal survival and targeting , and that there is a NT family in Drosophila formed by DNT1 , DNT2 , and Spz .
We used 28 known full-length and Cystine-knot domain ( Cysknot , characteristic of NTs ) vertebrate NT sequences to query release 2 of the Drosophila genome with TBLASTN and PSI-BLAST , which is specific to detect distantly related sequences ( Figure 1A and Text S1 ) . When using carp BDNF as query , both searches identified CG18318 . In turn , CG18318 identified BDNF from multiple species , from fish to human . After isolating the full-length cDNA3 from CG18318 ( Figures 1C and S2; GenBank accession number: FJ172423 ) , we used the protein sequence to carry out a structure-based search using FUGUE ( Figure 1A ) [44] . FUGUE identifies distantly related proteins , the sequence of which may have diverged through evolution while retaining structural conservation [44] . FUGUE compares the query protein sequence with the HOMSTRAD database of proteins of known structure , and it assigns amino-acid substitutions a score depending on how this affects protein structure [44] . FUGUE identified the human NTs with over 99% certainty as probable homologs of cDNA3 from CG18318 ( Table S2 ) , above similarity to coagulogen . Search of the ENSEMBL human database using cDNA3 protein sequence as query also identified human BDNF ( Figure 1A ) . Thus , we named the protein encoded by cDNA3 Drosophila Neurotrophin1 ( DNT1 ) . PSI-BLAST searches using Spz as query to the Drosophila genome had identified distant spz paralogs [45]: DNT1 is spätzle 2 ( spz2 ) . To verify the structural features of DNT1 , we carried out a structural alignment of DNT1 to known NT sequences from human , Xenopus , and the ancient NTs from lamprey ( Lf-NT ) , Amphioxus ( Bf-NT ) , sea urchin ( Sp-NT ) , and acorn worm ( Sk-NT ) ( Figure 1B ) . All the essential residues that form the NT Cysknot ( positions 499–601 in DNT1 ) are conserved in all these sequences , i . e . , the six cysteines , the glutamine ( position 539 ) , and conservative substitutions of all the residues of the hydrophobic core . Interestingly , DNT1 shares more conserved residues with acorn worm Sk-NT than with other NTs ( Figure 1B ) . The DNT1 Cysknot is highly conserved in all sequenced insects , such as fruit fly ( Drosophila ) , mosquito ( Anopheles ) , and bee ( Apis ) ( Figure 2 ) , and conservation outside the Cysknot is also high among all Drosophila species . There is high sequence divergence , particularly outside , but also within the Cysknot among all ancient NTs ( Lf-NT , Bf-NT , Sp-NT , Sk-NT , and DNT1 ) . We have attempted phylogenetic analyses of DNT1 and spz compared to all known NTs , including all ancient NTs , as above , using three standard methods ( Figure S3 ) . Sequence divergence precludes direct proof of orthology between DNT1 and vertebrate NTs . An ancestral NT gene was quite likely the predecessor of DNT1 in protostomes and NTs in deuterostomes . NTs are secreted proteins with a Cysknot domain , cleaved from pre-pro-precursors , and which dimerise to become functional [46] . Similarly , Spz becomes functional following cleavage to form a Cysknot dimer [40 , 47–49] . Instead , coagulogen does not dimerise to be functional [41 , 50] . DNT1 is a 886-amino acid ( aa ) protein with a 102-aa Cysknot followed by a 286-aa COOH tail , predicted to be secreted and cleaved , possibly at position 498 ( Figures 1D , 1E , and S2 , predicted cleavage site: FSLSKKR RE; see Text S1 ) . DNT1 is cleaved upstream of the Cysknot in cell culture ( Figure 1F ) , although the putative protease cleaving DNT1 in vivo may be absent in S2 cells . Both recombinant Cysknot and the Cysknot with the COOH extension ( Cysknot3-tail ) form dimers ( Figure 1G ) , hence they fold correctly upon expression . Thus , DNT1 presents structural , cleavage , and dimerising features of canonical NTs . The functional characteristics of mature vertebrate NTs are: ( 1 ) they are expressed by target cells in limiting amounts; ( 2 ) they maintain neuronal survival; and ( 3 ) they enable targeting and connectivity . Thus , we asked whether DNT1 satisfies any or all of these criteria . DNT1 is expressed in target cells throughout development . In the embryo , DNT1 is expressed at the CNS midline ( Figure 3A–3C ) , the intermediate target for interneurons ( the vertebrate floorplate is also an intermediate target that supports neuronal survival [51] ) , in two lateral CNS cells per hemisegment at stage 17 and in the epidermis ( unpublished data ) , and in the muscles ( Figure 3D–3F ) , the target for motor neurons . In the larva , it is expressed in the lamina ( Figure 3G–3I ) of the optic lobe , the target for photoreceptor axons . In the adult , it is expressed in the optic lobes and central brain ( Figure 3J and 3K ) , in the site of learning and memory . In order to analyse the incidence of apoptosis upon loss or gain of function for DNT1 , apoptotic cells were visualised in vivo with anti-cleaved Caspase-3 ( Caspase-3 ) antibodies , and we developed a computer software programme for the automatic quantification of Caspase-3 stained cells , called DeadEasy ( Figure S4 , Text S1 , and M . G . Forero , J . A . Pennack , A . R . Learte , K . Kato , R . L . Griffiths , et al , unpublished data ) . Caspase-3 antibodies stain specifically apoptotic cells ( Figure S4A ) ; they have the advantage over other methods of not staining necrotic cells , and are extensively used to visualise apoptotic cells in multiple model organisms and in human paradigms ( e . g . , [13 , 52–56] ) . Six or seven trunk segments ( depending on stage ) of stained embryos are scanned at the confocal microscope throughout the whole thickness of the CNS ventral nerve cord ( VNC ) . A region of interest ( ROI ) is selected over the VNC for quantification to exclude the epidermis . DeadEasy identifies stained cells in each individual section throughout the VNC and subsequently in 3-D , based on minimum volume and pixel intensity , and produces the total number of cells per VNC . The programme has been verified and validated ( see Text S1 ) . To ask whether DNT1 can rescue naturally occurring cell death ( NOCD ) , we expressed in all neurons ( with elavGAL4 ) four forms of the protein: ( 1 ) full-length; ( 2 ) pro-domain ( i . e . , cDNA1 , lacking the Cysknot , see Text S1 ) ; ( 3 ) Cysknot; and ( 4 ) Cysknot3-tail comprising the Cysknot plus the COOH extension ( Figures 1E and 4B ) . We stained embryos with Caspase-3 ( Figure 4A and 4B ) and quantified CNS apoptosis in the VNC automatically with DeadEasy software . Expression of either the full-length protein or the pro-domain does not reduce apoptosis levels compared to wild type ( Figure 4C ) . However , expression of either the Cysknot or the Cysknot3-tail results in a significant reduction in apoptosis compared to wild type ( Figure 4C ) . The disparity between the full-length and the cleaved forms is reminiscent of vertebrate NTs [3 , 46] and of the fact that the cleaved Spz Cysknot is functional when expressed in transgenic flies , whereas full-length Spz is not [47] . Expression of either DNT1 Cysknot or the Cysknot 3-tail at the midline ( with simGAL4 ) also reduces significantly apoptosis compared to wild type ( Figure 4C ) , implying that DNT1 is normally present in limiting amounts at this target . These data show that DNT1 can promote cell survival in the embryonic CNS . To ask whether DNT1 is required to promote cell survival , we generated by homologous recombination genetic null mutant alleles , DNT141 and DNT155 , as verified by PCR , Southern blot , and reverse transcriptase PCR ( Figures 1C and 5A–5C and Text S1 ) . DNT141 homozygous mutant flies are viable . In the CNS of homozygous DNT141 null mutant embryos , apoptosis levels are comparable to wild type at stages 13/14 , and there are no axon guidance defects ( unpublished data ) . At stage 17 , DNT141 , DNT141 /DNT155 , and DNT141/Df ( 3L ) ED4342 null mutant embryos show a significant increase in apoptosis in the CNS ( Figure 4E ) . To verify that the increase in apoptosis is a direct consequence of loss of DNT1 function , we expressed the DNT1 Cysknot in all neurons in embryos mutant for DNT1 ( Figure 4E , rescue ) . This leads to a significant reduction in apoptosis compared to DNT1 mutants , confirming that loss of DNT1 function causes the increase in apoptosis in the mutants . To verify whether the cells dying in excess in the mutants are neurons , we labelled DNT1 mutant embryos with the neuronal markers anti-HB9 and anti-Eve ( as well as Caspase-3 ) , which label complementary sets of motor neurons and interneurons . HB9 stains the majority of the ventrally and laterally projecting motor neurons [57 , 58] . This corresponds to motor neurons that project via intersegmental nerve ( ISN ) , ISN branch b/d ( ISNb/d ) , segmental nerve branch a ( SNa ) , and SNc , four RP neurons , and a ventral motor neuron , but it does not stain the Eve dorsally projecting motor neurons [57 , 58] . HB9 is expressed in many interneurons , including serotonergic neurons and a subset of FasII-negative interneurons [58] . Eve-expressing neurons are pCC , fpCC , and EL interneurons and dorsally projecting motor neurons , including aCC , RP2 , and the Us/CQs [59] . Colocalisation of Caspase-3 with HB9 increases significantly in DNT141/DNT155 and DNT141/Df ( 3L ) ED4342 mutant embryos compared to wild type ( Figure 6A ) . Colocalisation of Caspase-3 and Eve in the EL interneurons and in the U/CQ motor neurons also increases significantly in DNT141 mutants ( Figure 6B ) . Apoptosis causes cell loss , and there is a significant reduction in the number of Eve-positive neurons in DNT141 mutants compared to wild type ( Figure 6C ) . These data show that neurons die in excess in the absence of DNT1 . To ask whether DNT1 maintains CNS cell survival nonautonomously from the midline intermediate target , we reduced levels of DNT1 transcripts containing the Cysknot ( CK ) by expressing three different DNT1-RNAi ( RNA interference ) sequences in target cells ( Figure 4D ) : CK-RNAi , pwCK-RNAi , and pro-domain-RNAi . The pro-domain-RNAi knocks down all DNT1 transcripts , whereas CK-RNAi knocks down only cDNA3 . Three different and partially overlapping constructs were used to rule out the contribution of off-target effects to the phenotype . To enhance the specificity and penetrance of RNAi , experiments were carried out in embryos heterozygous for the null allele DNT141 or for Df ( 3L ) ED4342 that uncovers the DNT1 locus . Targeted DNT1 pro-RNAi , pwCK-RNAi , and CK-RNAi restricted to a narrow strip of cells at the CNS midline ( with simGAL4 ) increase apoptosis significantly throughout the CNS cortex compared to controls at stage 17 ( Figure 4E ) . Since the shorter cDNA1 does not promote neuronal survival , the increase in apoptosis is due to the knocking down of cDNA3 in all cases . Thus , reducing DNT1 levels at the midline is sufficient to induce apoptosis throughout the VNC . These data , together with the facts that DNT1 mutants have excess apoptosis throughout the VNC despite being expressed in a very small group of cells and DNT1 rescues NOCD when overexpressed at the midline only ( Figure 4C ) , show that DNT1 promotes cell survival nonautonomously in the CNS . To investigate whether DNT1 is required for axonal targeting , we analysed the axonal projections of FasII-positive motor neurons in DNT141 , DNT141/DNT155 , and DNT141/Df ( 3L ) ED4342 mutant embryos and upon DNT1-CKRNAi and DNT1-pro-RNAi targeted to the muscle ( with 24BGAL4 ) in embryos heterozygous for Df ( 3L ) ED4342 . In all cases , there is a significant increase in the incidence of misrouting phenotypes in ISNb/d and SN fascicles compared to wild type , including effects in more than one projection per hemisegment ( e . g . , misrouting plus loss; Figure 7D–7G , 7J , 7K , 7M , and 7N ) . To verify the target-dependent origin of these phenotypes , we targeted DNT1-RNAi to all neurons as a control . The incidence of axonal phenotypes upon RNAi targeted to all neurons is not significantly different from wild type , whereas it is significantly different from the incidence upon RNAi targeted to the muscle ( Figure 7M and 7N ) . This shows that the phenotypes caused by RNAi targeted to the muscle are due to the loss of DNT1 function in this target . To verify whether targeting to the muscle requires a limited source of DNT1 , we overexpressed DNT1 Cysknot 3-tail in the muscle . Excess of DNT1 Cysknot3-tail prevents targeting by motor neuron terminals at muscle 6/7 and 12/13 ( Figure 7H and 7L–7N ) . Thus , DNT1 produced by the muscle enables correct motor neuron targeting . Given the proposal of an NGF domain in Spz [40 , 41 , 43] , we asked how Spz relates to the vertebrate NTs . PSI-BLAST search using BDNF and all other vertebrate NTs as query against the Drosophila genome fails to identify spz . Following the same approach as for DNT1 , a PSI-BLAST search using spz sequence as a query against the SWISSPROT database produces no significant hit to any NT . DNT1 and spz are paralogs [45] , but under the same PSI-BLAST search conditions , DNT1 can be linked to some members of NTs ( e . g . , fish BDNF ) , whereas no such link can be established between Spz and any NT . Percentage identity within the Cysknot is higher for DNT1 and BDNF ( 26 . 4% ) than for Spz and NGF ( 19% ) or than for any other spz paralog when compared with NTs . Conservation of spz in insects is lower ( or absent , e . g . , A . gambiae ) than that of DNT1 , but although DNT1 is not conserved in beetles ( Tribolium ) , spz is ( Figure S5A ) . These observations suggest that DNT1 retains the sequence features of an ancient neurotrophin ancestor better than spz does . Nevertheless , Spz still forms a Cysknot [40 , 41 , 43] that can be aligned to the NT Cysknot superfamily ( Figure 1B ) . Thus , we next asked whether Spz may have NT function . Spz is expressed at the embryonic CNS midline ( Figure 8A ) , and its receptor , Toll , is in all CNS axons ( Figure 8B and 8B′ ) . Expression of activated Toll in all neurons rescues NOCD at stage 17 ( but not at stages 13 and 14 ) ( Figure 8D ) , showing that it can maintain neuronal survival . To ask whether Spz and Toll are required for neuronal survival , we looked at stage 17 embryos where maternal product enables normal early development , as confirmed by the eclosion of homozygous spz2 mutant flies . Apoptosis increases in the CNS of spz2 and Tollr3/Df ( 3R ) ro80b mutant embryos , indicating that both Spz and Toll are required for neuronal survival ( Figure 8D ) . Altogether , these data show that spz also has neurotrophic function , but it is weaker than DNT1 . DNT1 and Spz do not seem to play fully redundant roles . Apoptosis does not increase further in spz2 DNT141 double mutants , and expression of activated spz in DNT141 mutant embryos rescues apoptosis slightly , but it does not rescue the DNT141 mutant phenotype ( Figure 8D ) . We cannot rule out the possibility that DNT1 may rescue the spz mutant phenotype . This indicates that DNT1 and Spz may promote the survival of distinct subsets of neurons . spz is also expressed in bands along the location of embryonic lateral muscles ( Figure 9A ) , in a complementary pattern to that of DNT1 in muscles ( Figures 3E and 9B ) , suggesting that they may aid targeting by different axonal projections . Loss of Spz function affects predominantly targeting by the SNa motor axons ( Figure 9C ) . Loss of DNT1 affects ISNb/d projections more severely that SNa projections ( Figure 7M and 7N ) . The SNa axonal phenotype of double-mutant embryos lacking both DNT1 and Spz functions ( spz2 DNT141 ) is epistatic to spz ( Figure 9C ) . Altogether , these observations suggest that the targeting functions of DNT1 and Spz depend on neuronal modality . DNT1 is closer to spz and spz5 ( CG9972 ) than to the other paralogs [45] ( see Table S2 ) . Structure-based alignment using FUGUE reveals that DNT1/Spz2 , Spz , and Spz5 are more closely related to each other and to human NTs , whereas Spz3 ( CG7104 ) and Spz6 ( CG9196 ) are less closely related to vertebrates NTs ( Table S2 ) . Spz5 is very highly conserved amongst insects ( Figure S5B ) . The Cysknots of Spz3 , Spz4 ( GC14928 ) , and Spz6 differ from the canonical NT Cysknot: Spz3 and Spz4 have two extra cysteines and the Spz6 Cysknot lacks two of the conserved cysteines and has three extra ones in unusual locations . The Cysknots of Spz6 and Spz4 also differ from the rest in that they lack a conserved intron [45] . Furthermore , Spz4 is closest to coagulogen ( 29% identity ) , while also being closer to Spz3 ( 51% identity ) than to other paralogs , and its expression is up-regulated upon immune challenge [45] . Thus , the six spz paralogs fall into two groups: one formed by DNT1/spz2 , spz , and spz5 , and the other formed by spz3 , spz4 , and spz6 . Nevertheless , at least five of the six paralogs are expressed in the nervous system . There are no mutants available for spz3 , -4 , and -6 . Structural considerations suggest that Spz5 could also have neurotrophic function . To investigate whether spz5 has neurotrophic function we first visualised its expression pattern . spz5 is expressed at the embryonic CNS midline ( Figure 8C ) , in muscles ( Figure 9A ) , in the epidermis ( unpublished data ) , and in the embryonic head peripheral nervous system ( PNS ) ( unpublished data ) . To ask whether Spz5 can maintain neuronal survival , we expressed the Cysknot domain of spz5 ( UASDNT2 CK ) in all neurons ( with ElavGAL4 ) , which rescues NOCD ( at stage 17 ) ( Figure 8D ) . To ask whether loss of spz5 function affects neuronal survival , we used the only available mutant allele—spz5e03444—and deficiency Df ( 3L ) Exel6092 uncovering the spz5 locus . There is an increase in apoptosis in spz5e03444/Df ( 3L ) Exel6092 transheterozygous mutant embryos ( Figure 8D ) . Altogether , these data mean that Spz5 maintains neuronal survival in the embryonic CNS . Loss of both DNT1 and spz5 in double-mutant embryos results in an increase in apoptosis compared to each of the single mutants ( Figure 9D ) , revealing redundant or complementary functions in the control of neuronal survival . Thus , we rename spz5 ( CG9972 ) as DNT2 . In the muscle , the expression of DNT2 overlaps with that of both DNT1 ( in muscles 6 , 7 , 12 , and 13 ) and spz ( in SBM , LT lateral muscles ) . Both ISNb/d and SNa projections are mildly affected in DNT2e03444/Df ( 3L ) exel6092 mutant embryos . The penetrance of ISNb/d targeting defects increases in DNT1 DNT2 double-mutant embryos , although not significantly ( genotype: DNT2e03444 DNT141/Df ( 3L ) Exel6092 DNT141 ) ( Figure 9D ) . The penetrance of both SNa and ISNb/d targeting defects increases in DNT2e03444 DNT141spz2 triple-mutant embryos , compared to the double or single mutants ( Figure 9D ) . In the triple mutants , misrouting phenotypes can be very dramatic , and there are cases of loss of all ISNb/d motor axons ( not seen in single mutants ) ( Figure 9D , far left ) . Misrouting of the transverse nerve ( TN ) can be very dramatic in triple mutants ( Figure 9E ) , although milder effects in this nerve occur with comparable penetrance in all genotypes ( ∼10% ) . Misroutings of ISN are negligible in single and double mutants , but they increase and can be dramatic in triple-mutant embryos ( 12 . 7% , Figure 9F ) . These findings indicate that there is a synergistic interaction between DNT1 , Spz , and DNT2 in targeting , suggestive of redundant functions in this context . Synergism between the DNTs is further revealed by the effects of these mutations in viability . Whereas both DNT141 and DNT2e03444 mutants are viable and fertile , viability is somewhat affected in DNT141 DNT2 e03444 double mutants ( Table S3 ) : in homozygosis , DNT141 DNT2 e03444 flies are viable ( although some larval lethality , as well as when in trans over DNT141 Df ( 3L ) 6092 , was observed ) , but DNT141 DNT2 e03444/TM6B flies do not produce homozygous progeny at 18 °C , suggesting unsuccessful larval competition . Whereas homozygous spz2 flies can eclose as adults , the DNT141 spz2 double and triple mutants are completely lethal ( 100% penetrance ) . This suggests that DNT1 , Spz , and DNT2 play redundant functions for viability . Neurotrophin mutant mice display abnormal locomotion [60–63] . To ask whether DNT mutant flies move normally , we tested the ability of adult flies to climb over the rim of a Petri dish and walk along it—something wild-type flies do without difficulty and without falling off ( Video S1 ) . DNT141 DNTe03444 and DNT141 DNT2e03444/DNT141 Df ( 3L ) Exel6092 double-mutant flies display a range of phenotypes ( Table S4 ) including: inability to estimate the location of the rim ( Videos S2 , S3 , and S6 ) , falling off ( Video S3 and S5 ) , sluggishness ( Video S4 ) , inability to climb ( Video S6 ) , slow , uncoordinated movements ( Video S7 ) , and wobbling ( Video S8 ) ; Spz2 mutant flies can barely walk ( Video S9 ) . These phenotypes may be due to abnormal targeting or muscle structure or function or synaptic activity . They suggest that an involvement of DNTs in higher neuronal functions is a possibility .
Previous reports had revealed an NGF domain in Spz and biochemical evidence supports a similar mechanism of activation for Spz and the vertebrate NTs [40 , 47 , 49] . A theoretical structural analysis of Spz had shown that Spz forms a NT Cysknot [41 , 43] . These are features also found in DNT1 . However , when we carried out bioinformatic searches for Spz , a relationship between Spz and the NTs could not be established . Sequence identity between Spz and NGF is lower than for DNT1 and BDNF . Spz is also less conserved in insects than DNT1 is . The sequence of Spz is more diverged from the vertebrate NTs than DNT1 is . Nevertheless , Spz , together with Toll , also plays neurotrophic functions . Structural analysis of the spz paralogs indicates that DNT1 , Spz , and Spz5 are more closely related to each other and to the NTs , whereas Spz3 , Spz4 , and Spz6 are highly diverged . We cannot at this stage rule out the possibility that Spz3 , Spz4 , and Spz6 may also play functions in the nervous system . Spz5 is structurally close to the NTs and very highly conserved in insects . We have shown that Spz5/DNT2 has neurotrophic functions , as it rescues NOCD , and loss of Spz5/DNT2 function results in increased CNS apoptosis and axon targeting errors . We have renamed spz5 as DNT2 . Thus , there is a NT family in Drosophila formed of at least DNT1/Spz2 , DNT2/Spz5 , and Spz . Orthologs are genes related by ancestry . The identification of DNT1 by sequence homology to BDNF does not mean that DNT1 is a BDNF ortholog . BDNF resulted from the duplication of an ancestral vertebrate NT , thus a relationship between DNT1 and vertebrate NTs goes back to an ancestral NT ( Figure 10A ) . Consistently , DNT1 and Spz are more closely related to Sk-NT from acorn worm . The sequence relatedness between DNT1 and the NTs is unlikely to be due to convergence since it was found using three independent types of searches , including a structure-based search , and confirmed with two types of reverse searches , and biochemical features and function are also conserved . Direct proof that DNT1 , DNT2 , and spz are general NT orthologs cannot be obtained . High sequence divergence amongst all invertebrate NTs precludes the phylogenies to resolve . The same conclusion had been reached for the analysis of ancient deuterostomian NTs [38] . Our phylogenetic analyses of DNT1 and spz compared to all known NTs , revealed interesting features: first , the invertebrate deuterostomian NTs are closer to DNT1 and Spz than the vertebrate NT . Second , amongst those , acorn worm NT ( Sk-NT ) is the closest to DNT1 and Spz . Third , two other protein families contain Cysknots , TGFβ and PDGF , but these Cysknots differ from that of NTs . The Cysknot in DNT1 and Spz [41] is unequivocally closer to the NT Cysknot . The most parsimonious explanation ( Figure 10A ) is that an ancestral NT gene present in Urbilateria ( the presumed common ancestor of all bilateral organisms ) gave rise to the NTs in deuterostomes and in protostomes . The deuterostome NT duplicated twice to give rise to BDNF , NGF , NT3 , and NT4 in vertebrates , and the protostome ancestor duplicated more than once to generate at least DNT1 , DNT2 , and spz , while sequences diverged , retaining the structural features of the NT Cysknot that enabled function . A similar scenario is encountered in the tumour necrosis factor ( TNF ) superfamily , in which sequence similarity and identity between TNF members is restricted to the TNF homology domain where it is also low ( 19%–30% ) , but they are nevertheless considered members of a protein superfamily based on structural and functional conservation [64] . Thus , deuterostomian invertebrate NTs ( Bf-NT , Sp-NT , and Sk-NT ) belong to the NT superfamily based on sequence similarity in the Cysknot [34 , 38] , and we show that DNT1 , DNT2 , and Spz belong to the NT superfamily based on sequence , structural , and functional criteria . It had long been thought that NTs were missing from the Drosophila genome [42 , 65–68] . A similarity between Spz and NGF had been previously proposed [40 , 41] but remained controversial . First , structural considerations had also revealed a similarity between Spz and horseshoe crab coagulogen [41] , involved in the blood-clotting cascade . However , this study [41] did not use FUGUE , which was developed later to infer structural relationships between distantly related proteins [44] . A later study confirmed that Spz belongs to the NT superfamily [43] . Our phylogenetic analysis does not resolve coagulogen as sufficiently distinct from DNT1 , Spz , or the NTs . The Toll signalling cassette is conserved in horseshoe crab , including a Toll receptor and the downstream target NFκB [37 , 69] . Although it is unknown whether coagulogen may also have NT function in the horseshoe crab CNS , it is an intriguing possibility . We show here that FUGUE analysis comparing DNT1 to all proteins of known structure reveals a closer relationship of DNT1 to vertebrate NTs than to coagulogen . Second , an initial comparison of the sequenced human and Drosophila genomes with BLAST reported that there were no NTs in Drosophila [42 , 68] . However , this simple BLAST missed 30% of the Drosophila genes and would have missed any proteins with structural conservation despite sequence divergence . In fact , a recent report has reiterated the relationship of Spz to the NT superfamily [43] . We identified DNT1 using searches optimised for distantly related sequences , PSI-BLAST and FUGUE . In PSI-BLAST sequence searches , carp BDNF reveals sequence relatedness of DNT1 to NTs . Reverse BLAST and PSI-BLAST reveal similarity of DNT1 to BDNF from multiple fish species and humans . Structure-based searches with FUGUE demonstrate that DNT1 is structurally related to human BDNF , NGF , NT3 , and NT4 . Thus , DNT1 retains the features of all four human NTs . Thus , there is high sequence divergence among the NTs that nevertheless retain the functional Cysknot . The neurotrophic theory originally proposed that NTs promote neuronal survival in a target-dependent manner [1] , although NTs can also promote neuronal survival prior to innervation and in autocrine and paracrine manners [4 , 70] . Important evidence that vertebrate NTs promote neuronal ( and glial ) survival was the finding that exogenous application of NTs rescues neurons ( and glia ) from NOCD , both in cell culture and in vivo [71–82] . We find that expressing DNT1 either in all CNS neurons or at the midline can rescue NOCD in vivo . Expressing DNT2 or activated Toll in all CNS neurons also rescues NOCD . These findings indicate that , like in vertebrates , the DNTs can promote cell survival . The prosurvival functions of the DNTs are nonautonomous as the three DNTs are expressed virtually only at the CNS midline , but in the mutants , apoptosis is induced throughout the VNC; DNT1-RNAi targeted to the midline induces apoptosis throughout the VNC , and overexpression of DNT1 only at the midline rescues NOCD throughout the VNC . Loss of vertebrate NTs in individual mouse NT knockouts or their receptors affect the CNS very weakly , and do not generally cause an increase in CNS apoptosis [60–63 , 83–90] . Loss of DNT1 , spz , Toll , or DNT2 function does not cause massive CNS neuronal death either . Nevertheless , apoptosis increases significantly in the embryonic CNS in all DNT mutants . The dying cells are at least partly HB9 and Eve neurons . We did not find significant apoptosis phenotypes in DNT1 mutants or upon gain of function in the developing retina ( unpublished data ) . Vertebrate NTs play partially redundant functions [60 , 61 , 63 , 72 , 83–86]: some can substitute for one another to rescue apoptosis in mutants , and in multiple knock-out combinations , e . g . , BDNF−/−NT3−/−NT4−/− or TrkB−/−TrkC−/− , a 20% reduction in motor neurons and a dramatic increase in brain apoptosis , respectively , were observed compared to single mutants . The DNTs play redundant roles in the embryonic CNS in some , but not all , contexts . Expression of activated spz in DNT141 mutant embryos is not sufficient to fully rescue apoptosis ( however , we have not tested the reciprocal experiment ) , but apoptosis increases in DNT1−/− DNT2−/− double mutants , indicating redundancy between DNT1 and DNT2 for cell survival . Vertebrate NT function depends on neuronal modality: different neurons require different NTs for survival , and increases in apoptosis in the brain were observed when looking at specific neuronal types ( e . g . , parvalbumin-positive neurons in BDNF knock-out mice ) [60 , 72 , 84 , 91] . In DNT1 mutants , we observe an increase in apoptosis of HB9- and Eve-positive neurons , and loss of Eve neurons . Neuronal modality differences are revealed in the targeting by motor axons ( see below ) . Alterations in DNT1 function affect primarily ISNb/d motor axons , whereas loss of Spz function affects SNa motor axons , correlating with complementary domains of spz and DNT1 expression in different subsets of muscles . Locomotion deficits and/or lethality are a further feature of NT knock-out mice [60–63] . In fruit flies , some double-mutant combinations of the DNTs and triple mutants die during embryogenesis . DNT1 DNT2 double-mutant and spz2 mutant viable adult flies have distinct locomotion and/or behavioural deficits . Locomotion defects can reflect proprioception or muscle or synaptic problems . NTs play roles in synaptic plasticity , LTP , and behaviour , and altered NT function causes psychiatric and cognitive disorders in humans [2 , 3] . At least DNT1 is expressed in the adult central brain in the centres controlling learning and memory . Perhaps the DNTs are involved in higher neuronal functions . DNT1 produces two types of transcripts: the longer contain the Cysknot domain ( cDNA3 ) , and shorter ones ( cDNA 1 , cDNA2 , and cDNA4 ) comprise only most of the pro-domain . We have shown that expression of the shorter isoform does not rescue apoptosis , rather it ( and the full-length protein ) may increase it ( see Figure 4C ) . This is reminiscent of the opposite functions of the mature and full-length vertebrate NTs in the control of neuronal survival and death , respectively [3] , and of the fact that in transgenic flies , full-length Spz is not functional in immunity , whereas the cleaved Cysknot is [47 , 48] . We do not know whether the shorter DNT1 isoforms play other roles , but conceivably they may modulate the function of mature DNT1 , as the pro-domain of spz can inhibit signalling by the Spz-Cysknot [43 , 92] . Loss of vertebrate NTs severely affects the PNS , and rather weakly affects the motor neurons [60–63 , 83–88 , 93] . Virtually all vertebrate PNS neurons require NTs for survival . In Drosophila , the effect of DNT1 mutations in the embryonic PNS is milder than in the CNS ( unpublished data ) . Exogenous application of NTs can rescue vertebrate motor neuron survival [76–78] , but loss of individual vertebrate NTs does not induce motor neuron apoptosis [61–63 , 88] . Only 20%–30% of motor neurons die in triple knock-out mice lacking multiple NTs or all Trk receptors [85 , 93] . In fact , the main trophic factor maintaining vertebrate motor neuron survival is GDNF , which does not belong to the NT superfamily ( e . g . , [94] ) . Motor neurons are not produced in vast excess in Drosophila , but there is motor neuron apoptosis in normal embryos , as detected with the motor neuron markers HB9 and Eve , although the underlying cause is not known [13] . We observe a significant increase in HB9 neuronal apoptosis in DNT1 mutant embryos compared to wild type ( although HB9 also labels interneurons ) . Loss of Eve motor neurons is also observed in DNT1 mutants , as well as loss of all the FasII-positive ISNb/d axons in triple-mutant embryos . It has previously been reported that RP motor neurons can be missing in Toll mutant embryos , although this could reflect an autocrine function [95] . We have not been able to conclusively determine whether motor neuron death in DNT1 and triple mutants is due to the target-derived function of DNTs in the muscle , or an autocrine/paracrine requirement in the motor neurons . Expression of DNTs at the midline could influence the motor neurons within the CNS . Abundant evidence indicates that motor neurons live and function well in the absence of the muscle target in Drosophila [96] . For instance , upon genetic elimination or surgical ablation of the muscle [97–99] and in the absence of muscle-derived signals [100] , motor neurons grow towards the muscle but fail to target or target to ectopic sites . In normal embryos and larvae , the projection patterns of motor neurons is very stereotypic [96 , 101 , 102] . Accordingly , it would appear that motor neuron survival may not depend on the target muscle in Drosophila embryos and larvae . Vertebrate NTs influence muscle innervation by motor neurons [103] . In Drosophila , the existence of a muscle-derived sprout-promoting factor to which Toll-expressing motor neurons would respond had been anticipated [95] . We show that a target-derived function of DNTs in the muscle is required for guidance and targeting by motor axons . Loss of function for all three DNTs , as well as gain of DNT1 function , disrupts axon guidance and targeting by motor axons . The domains of expression of DNT1 and spz in the muscles are complementary , and both overlap that of DNT2 . Consistently , DNT1 and spz , together with DNT2 , affect targeting by complementary sets of motor axons , and the triple mutants have dramatic defects in all motor neuron projections ( see above ) . The larval neuromuscular junction ( NMJ ) offers the most amenable synapse in Drosophila . There is abundant evidence of synaptic plasticity at the NMJ [104 , 105] . However , so far , the identification of the responsible retrograde signals has been rather scarce [105–108] . The identification of the muscle-derived secreted DNTs is promising in this context . All three DNTs are expressed at the CNS midline and in the muscles . At least the Spz receptor Toll is expressed transiently in the muscle; Toll and spz mutants have muscle defects , and Toll is involved in motor neuron synaptogenesis [95 , 109] , although some of the Toll mutant muscle defects may be due to nonautonomous effects [110] . We have also observed muscle defects in spz mutants and most severely in the triple mutants . However , targeting errors were also observed in the presence of normal muscle patterns ( see Figure S6 ) , indicating that targeting and putative muscle functions can be dissociated . We cannot rule out the possibility that DNTs may play roles in midline-derived glia or neurons , including motor neurons , or in the muscles . Interestingly , vertebrate NTs also have functions in the muscle [111] . Signalling by DNT1 and DNT2 may not necessarily proceed by binding canonical vertebrate-like Trk and p75 receptors . Ligand and receptor pairs do not necessarily coevolve [34 , 112] . For instance , Toll-like receptors are highly conserved , but bind very different ligand types in flies and vertebrates [113] . DNT1 and DNT2 may bind yet-unidentified Trk and p75 homologs in Drosophila or other receptors that activate equivalent signalling pathways and result in equivalent cellular , neurotrophic responses . Trk homologs were originally reported in Drosophila and subsequently showed not to belong to the Trk family [39] . However , a Trk homolog has been found in the protostome mollusc Lymnea [36 , 39] , suggesting that either Trks may have been lost in Drosophila or not found . Trk receptors are modular , thus exon shuffling during evolution could have led to the separation of domains into different proteins while retaining function [34 , 112] . Consistently , an intracellular Trk-like tyrosine kinase domain has been found in Aplysia in a receptor , ApTrk , with an extracellular domain unrelated to the Trks [35] . The converse situation is conceivable . DNT1 may bind a receptor tyrosine kinase , or a TNFR-like receptor ( as p75 is ) , or resembling Spz , a Toll-like receptor , or , as with vertebrate NTs , DNT1 may be a promiscuous ligand binding multiple receptor types . As with vertebrate NT receptors , binding to one receptor type may result also in interactions with other receptors that alter cellular outcomes depending on context . There is a TNF receptor and multiple Toll-like receptors in Drosophila [114] . Signalling by Toll and mammalian Toll-like receptors underlies innate immunity [115] , and it is an ancient pathway present also in the cnidarian Nematostella and in Caenorhabditis elegans . Vertebrate NTs are also involved in immunity . Perhaps Toll signalling is an ancient mechanism underlying the functions of both the nervous and immune systems . Interestingly , the extracellular domain of Toll resembles that of Trk receptors ( with the unusual combination of Leu-rich repeats and cysteine repeats ) , and intracellularly , Toll activates a downstream signalling pathway very similar to that of p75 , resulting in the activation of NFκB [34] . Our data indicate that the evolutionary trajectory of neurotrophin signalling in arthropods travelled through—although may not be restricted to—Toll . DNTs may also bind other receptor types . Toll , p75 and the TNFR family are more ancient than the Trks [30] . Drosophila Spz/Toll , and vertebrate Toll-related , p75 and TNFR receptors signal through NFκB ( promoting cell survival ) and c-Jun ( promoting cell death ) [115] . Vertebrate Toll-like–related receptors also activate MAPKinases [115] , and p75 also activates AKT [30] . These pathways are compatible with the neurotrophic functions of DNT1 , DNT2 , and Spz . NFκB is also involved in synapse formation , synaptic plasticity , learning , and memory , and alterations in NFκB function also lead to psychiatric conditions [116 , 117] . Inhibition of NFκB signalling in crabs ( protostome arthropods like flies ) leads to deficits in learning and memory , functions traditionally assigned to NTs [118] . Conceivably , also higher functions of DNTs may be controlled by NFκB . Our findings and those of others [33–36 , 38 , 41 , 45] suggest that the evolution of neurotrophin signalling may have resulted in diversification of receptors and/or downstream signalling pathways . We have not found DNT1 sequences in the snail Aplysia ( see Text S1 ) . This could mean that NTs appeared independently in deuterostomes and insects , and their similarity is due to convergence . However , it is equally possible that structure and function were conserved despite high sequence divergence , that the sequences have not been found yet , or that NT were lost from some or many animals . A Trk-like tyrosine kinase domain has been found in Aplysia , ApTrk [35] , and a bona fide Trk ortholog in another snail , Lymnea , suggesting that the NT signalling pathway is present in molluscs . Our unsuccessful search in Aplysia is likely due to incomplete genome sequence and expressed sequence tag ( EST ) collection [119] . If a NT was present in Urbilateria ( Figure 10A ) , then NTs may be important in the nervous system development and function of all animals with a centralised nervous system or brain . What about simpler animals such as anemones and corals , which do not have a centralised nervous system , but a diffuse , nerve net ( Figure 10A and 10B ) ? To ask this , we searched for NTs in a cnidarian , Nematostella , but we did not find a DNT1 homolog . Sequence divergence and/or incomplete EST database may have also prevented the identification of NT sequences in Nematostella . Orthologs of Toll and downstream targets of Toll , p75 , and Trk receptors , such as NFκB , MAPKinase , and ERK , are all present in Nematostella [120] . Alternatively , NTs may have originated in Urbilateria and are absent from simpler animals , or perhaps a preexisting NT may have been lost in Nematostella and other cnidarians ( just as NTs were lost in the deuterostome Ciona [38] ) , as extensive gene loss is known to have occurred in cnidarians [121] . Consistently with the view that elaborations of neurotrophin signalling underlie brain complexity , perhaps the diffuse net structure of the cnidarian nervous system does not require neurotrophin signalling , resulting in their loss . However , the acorn worm also has a diffuse , nerve net nervous system , and it has a NT and p75 receptor . This suggests that NTs may also be present in other animals with a nerve net , where they may have a subset of functions ( e . g . , axon guidance , connectivity , or synaptic functions ) . Our data suggest that a NT was most likely present in Urbilateria , the common ancestors of all bilateral organisms—protostomes and deuterostomes ( Figure 10A ) —it duplicated independently in vertebrates and invertebrates , and NTs were retained in organisms with a centralised nervous system and/or brain . NTs may be more ancient and have been either retained or lost in animals with diffuse neuronal nets ( Figure 10B ) . Our findings imply that the control of cell survival and targeting by the NT superfamily is an ancient mechanism of nervous system development . Further functions of the DNTs could also include synaptic and neuronal activity , learning , and memory . Our findings support the notion of a common origin for nervous system centralisation in evolution [122 , 123] . They suggest that in the course of evolution “elaborations of what went before” [124]—an available molecular mechanism involving the ancestral NTs—and “tinkering” [125] with NT signalling accompanied the diversification of nervous systems and behaviours . The identification of DNTs bridges a void in neuronal studies using Drosophila as a model for understanding the brain . Conserved molecular mechanisms involving the NT superfamily may underlie aspects of retrograde transport , dendrite formation , axonal remodelling , synaptic plasticity , LTP , and learning and memory also in flies—all functions for which NTs are responsible in vertebrates . This work opens a wide range of opportunities to further the understanding of brain formation and evolution and to model human brain diseases using Drosophila .
Full-length and Cystine-knot sequences from 28 known vertebrate NTs were used in PSI-BLAST searches ( http://www . ncbi . nlm . nih . gov/blast/blast . cgi ) against release 2 of the Drosophila genome . Carp ( Cyprinus carpio ) BDNF showed homology with CG18318 both in BLAST and PSI-BLAST searches as the only hit in Drosophila . This hit was verified by reverse-BLAST . When DNT1 is used as a query in structure-based searches using FUGUE , it identifies with over 99% certainty the human neurotrophins , comprising BDNF , NGF , NT3 , and NT4 as probable homologs . To verify the homology of DNT1 to NTs , we carried out structural alignments . The sequences for the NT Cystine-knot domains were aligned against the HOMSTRAD [126] ( http://www-cryst . bioc . cam . ac . uk/homstrad/ ) entry of the nerve growth factor ( NGF ) family using FUGUE [44] . Using this alignment , a model of DNT1 was built with MODELLER [127] . Phylogenetic analysis was attempted using sequences comprised with the Cysknot domain only , as sequences diverge considerably outside the Cysknot . Methods used were Maximum Parsimony , Neighbour Joining , and Maximum Likelihood . The BLAST server at FlyBase ( http://www . flybase . org/blast/ ) was used to identify orthologs of DmNT1 and DmSpz in other insect species ( see also: http://rana . lbl . gov/drosophila/ ) . Cleavage prediction analysis using the ProP server ( http://www . cbs . dtu . dk/services/ProP ) reveals two high scores at positions 283 and 294 . However , the sequence most likely to match the cleavage site of Spz by Easter is FSLSKKR RE at position 498 . We searched for sequence homologs of DNT1 in the sequenced genome of Nematostella vectensis and the EST collections of N . vectensis and Aplysia californica . For details of the mutants , alleles , transgenic lines of flies , and GAL4 driver lines of flies used , see Text S1 . Null alleles for DNT1 were generated by homologous recombination using the ends-out protocol . The coding region of DNT1 , including the ATG and the whole Cysknot domain , was replaced by the coding region of the white gene . The DNT1 locus corresponds to CG18318 from release 2 and CG32244 plus CG32242 from release 3 of the sequenced genome ( http://www . flybase . org ) . Full-length cDNA3 was amplified by PCR from cDNA libraries . DNT1 ( cDNA3 ) was sequenced and presents the following characteristics: MW 100 , 315 kDa , PI: 6 . 17 . DNT1 is 886 aa long , with a Signal Peptide ( 1–30 aa ) , a pro-domain ( 31–498 aa ) , a 102-aa Cysknot domain ( 499–601 aa ) , and an extended , disordered 285-aa COOH tail ( 602–886 aa ) . For further details on this and on the generation of gain-of-function and RNAi constructs for transgenesis , see Text S1 . RT-PCR was used to verify that targeted RNAi in a heterozgygous mutant background resulted in a down-regulation of DNT1 transcripts encoding the Cysknot . Under the same conditions , the null DNT141 mutants do not produce transcripts , whereas heterozygote embryos produce transcripts in normal levels . Cell culture and western blotting were used to verify cleavage and dimerisation of DNT1 . These methods were carried out following standard protocols , except that for Toll stainings , embryos were fixed for 10 min . Wide-field microscopy was carried out with Nomarski optics with a Zeiss Axioplan 2 and confocal microscopy with Leica SP2 and Radiance 2000 laser scanning confocal microscopes . We purposely wrote DeadEasy software as an ImageJ plug-in , to quantify automatically cells stained with the apoptotic marker anti-cleaved Caspase-3 ( M . G . Forero , J . A . Pennack , A . R . Learte , K . Kato , R . L . Griffiths , and A . Hidalgo , unpublished data ) . For details , see Text S1 . Statistical analyses of all experiments , with rationale , tests applied , confidence intervals , and p-values , are given in Text S1 . Filming was carried out with a Motic camera mounted on a Leica MZ8 microscope and using Motic Images Plus 2 . 0 software . The DNT1 cDNA sequences have been deposited in GenBank; for accession numbers , see text and Text S1 .
|
Neurotrophins are secreted proteins that link nervous system structure and function in vertebrates . They regulate neuronal survival , thus adjusting cell populations , and connectivity , enabling the formation of neuronal circuits . They also regulate patterns of dendrites and axons , synaptic function , memory , learning , and cognition; and abnormal neurotrophin function underlies psychiatric disorders . Despite such relevance for nervous system structure and function , neurotrophins have been missing from invertebrates . We show here the identification and functional demonstration of a neurotrophin family in the fruit fly , Drosophila . Our findings imply that the neurotrophins may be present in all animals with a centralised nervous system ( motor and sensory systems ) or brain , supporting the notion of a common origin for the brain in evolution . This work bridges a void in the understanding of the Drosophila and human nervous systems , and it opens the opportunity to use the powerful fruit fly for neurotrophin related studies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"evolutionary",
"biology",
"neuroscience",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] |
2008
|
Drosophila Neurotrophins Reveal a Common Mechanism for Nervous System Formation
|
Hepadnavirus at very low doses establishes in woodchucks asymptomatic , serologically undetectable but molecularly evident persistent infection . This primary occult infection ( POI ) preferentially engages the immune system and initiates virus-specific T cell response in the absence of antiviral antibody induction . The current study aimed to determine whether POI with time may culminate in serologically identifiable infection and hepatitis , and what are , if any , its pathological consequences . Juvenile woodchucks were intravenously injected with inocula containing 10 or 100 virions of woodchuck hepatitis virus ( WHV ) to induce POI and followed for life or up to 5 . 5 years thereafter . All 10 animals established molecularly detectable infection with virus DNA in serum ( <100–200 copies/mL ) and in circulating lymphoid cells , but serum WHV surface antigen and antibodies to WHV core antigen remained undetectable for life . By approximately 2 . 5–3 . 5 years post-infection , circulating virus transiently increased to 103 copies/mL and virus replication became detectable in the livers , but serological markers of infection and biochemical or histological evidence of hepatitis remained undetectable . Nonetheless , typical hepatocellular carcinoma ( HCC ) developed in 2/10 animals . WHV DNA integration into hepatic and lymphatic system genomes was identified in 9/10 animals . Virus recovered from the liver virus-negative or virus-positive phases of POI displayed the wild-type sequence and transmitted infection to healthy woodchucks causing hepatitis and HCC . In summary , for the first time , our data demonstrate that an asymptomatic hepadnaviral persistence initiated by very small amounts of otherwise pathogenic virus , advancing in the absence of traditional serological markers of infection and hepatitis , coincides with virus DNA integration into the host's hepatic and immune system genomes , retains liver pro-oncogenic potency and is capable of transmitting liver pathogenic infection . This emphasizes the role for primary occult hepatitis B virus infection in the development of seemingly cyptogenic HCC in seronegative but virus DNA reactive patients .
It is estimated that 370 million people have serologically evident chronic hepatitis B virus ( HBV ) infection and over 2 billion have been exposed to this virus [1] . Chronic hepatitis B ( CHB ) frequently ( 20–25% ) advances to cirrhosis and liver failure , while hepatocellular carcinoma ( HCC ) develops in approximately 5% of the cases [1] . HBV is considered to be primarily hepatotropic; however , it also infects cells of the immune system where it persists for decades even when hepatitis resolves [2]–[4] . These events are closely mimicked in the natural animal model of HBV infection , the eastern North American woodchuck infected with woodchuck hepatitis virus ( WHV ) [4] , [5] . WHV invariably invades the immune system and persists there for life irrespective of whether infection is symptomatic and serologically evident , i . e . , serum WHV surface antigen ( WHsAg ) and antibody to WHV core antigen ( anti-WHc ) positive , or asymptomatic and serologically silent , i . e . , serum WHsAg and anti-WHc nonreactive [6]–[9] . Based on the findings in naturally and experimentally WHV-infected woodchucks , two forms of occult hepadnaviral persistence were uncovered . Secondary occult infection ( SOI ) continuing after resolution of acute hepatitis ( AH ) and apparent clearance of serum WHsAg is accompanied by lifelong persistence of anti-WHc and WHV DNA in serum , liver and immune system [5]–[7] . The liver in SOI can display moderate inflammation with periods of normal morphology; nonetheless HCC develops in up to 20% of animals [7] . This form of infection appears to commonly underlie reactivation of hepatitis B in immunocompromised patients and those on cytotoxic therapies [10] and development of HCC in individuals with past exposure to HBV [11] , [12] . Another form of hepadnaviral persistence , primary occult infection ( POI ) , was uncovered in offspring of woodchuck dams convalescent from AH and in animals inoculated with WHV doses ≤103 virions [6]–[9] . POI progresses in the absence of identifiable serum WHsAg , anti-WHc and antibodies to WHsAg ( anti-WHs ) , and hepatitis , but WHV and its replication are detectable at low levels in the immune system and sporadically in the liver [6] , [9] , [13] . Virus-specific T cell , but not B cell , responses are induced and , unlike SOI , protective immunity is not established . Our recent study showed that repeated exposures to liver nonpathogenic WHV amounts , i . e . , ≤103 virions , do not culminate in serologically detectable infection and hepatitis or generate immune protection [13] . This type of infection can be suspected in HBV DNA reactive individuals who are seronegative for HBV surface antigen ( HBsAg ) and antibodies to HBV core antigen ( anti-HBc ) [14] . The main objective of the current study was to identify lifelong liver pathological consequences of POI , given the known high oncogenic potency of WHV and HBV [4] , [15] and the notion that occult HBV infection undetectable by clinical testing might be responsible for HCC of unknown etiology in some cases [11] , [12] . We also aimed to identify characteristics of POI regarding virus transmissibility and pathogenic potency in virus-naïve hosts , the status of WHV DNA-host genome integration during POI , and compatibility between virus sequences occurring in the liver virus-negative and the liver virus-positive phases of POI .
All animal experiments and the animal maintenance protocols were performed in compliance with the Institutional Animal Care Committee at Memorial University , St . John's , Newfoundland and Labrador , Canada ( protocol identification number 13–159-M ) that follows the guidelines and is accredited by the Canadian Council on Animal Care in Science . Infection experiments were carried out in 1–2 year old healthy woodchucks housed in the Woodchuck Viral Hepatitis Research Facility at Memorial University , St . John's , Newfoundland and Labrador , Canada . All the animals were captured from a pristine region of Northern Canada . Normal liver function and morphology was ascertained by testing serum biochemical markers of hepatic performance , including sorbitol dehydrogenase ( SDH ) and γ-glutamyl transferase ( GGT ) , macroscopic inspection of the liver during laparotomy , and by histological examination of liver biopsy taken prior to initiation of the study . Prior exposure to WHV was excluded by negative testing for serum WHsAg and anti-WHc , and by the absence of WHV DNA as determined by highly sensitive polymerase chain reaction/nucleic acid hybridization assays ( PCR/NAH ) ( sensitivity ≤10 copies or virus genome equivalents [vge]/mL or ≤10 vge/µg total DNA ) [7]–[9] . WHV/tm3 inoculum ( GenBank accession number AY334075 for 3 identical clones ) induced serum WHsAg-positive hepatitis in >90% of woodchucks after intravenous ( i . v . ) administration of doses ≥103 DNase-digestion protected vge , i . e . , virions [7]–[9] . WHV/tm5 inoculum ( GenBank accession numbers KF874491-3 for 3 clones ) was derived from a woodchuck with chronic hepatitis and HCC . WHV/tm5 whole genome sequencing showed 99 . 7% ( 3298/3308 ) and 99 . 58% ( 1668/1675 ) identity in the nucleotide ( nt ) and amino acid sequences , respectively , when compared to WHV/tm3 . Prior to infection , WHV/tm5 was fractionated on cesium chloride gradient to separate virions from free WHsAg , essentially as reported [8] . The recovery of intact virions was ascertained by a DNase-digestion protection assay [2] . To induce POI , 3 animals were i . v . injected with 10 virions of WHV/tm5 and 7 others with 100 virions of WHV/tm3 . In addition , 2 animals inoculated with 106 WHV/tm5 virions , 2 inoculated with 1010 WHV/tm3 virions , and 2 healthy woodchucks not exposed to WHV , all followed for duration of their lifespan , served as controls . The animals were bled biweekly until 16 weeks post-infection ( w . p . i . ) and then bimonthly . They were followed for life until senility ( animals 1/F , 3/F , 7/M , 10/M and 12/F ) , HCC development ( 5/M , 9/M and 11/M ) , a WHV-unrelated severe health issue requiring termination of follow-up ( 2/F , 8/M and 14/F ) or challenge with 1010 virions WHV/tm3 at 66 months post-infection ( m . p . i . ) ( 4/F , 6/M and 13/F ) . The last group of animals was observed for an additional 5 . 2 months , and bled biweekly until 14 weeks post-challenge ( w . p . c . ) and then monthly . Liver biopsies were obtained before WHV inoculation and at 6 w . p . i . , 8 m . p . i . and then at approximately yearly intervals until autopsy . At autopsy , serum , peripheral blood mononuclear cells ( PBMC ) , liver , bone marrow , spleen , lymph nodes and other organ samples were collected . WHV inocula were prepared from the liver virus DNA-negative and the liver virus DNA-positive phases of POI by pooling 24 mL of serum and plasma from 6/M and 7/M , and from 8/M that was liver WHV DNA nonreactive for life . Pellets recovered by ultracentrifugation at 200 , 000× g for 20 hours at 4°C were suspended in 1 . 3 mL of sterile phosphate buffered saline , pH 7 . 4 . One mL of suspension was i . v . injected into a virus-naïve woodchuck and 0 . 3 mL used for WHV quantification and sequencing . Thus , A/F animal was injected with 1370 vge from 6/M , C/F with 1460 vge from 7/M , and E/F with 1460 vge from 8/M , all obtained from the liver virus-negative phase of POI . Also , B/F was infected with 2070 vge from 6/M and D/F with 1460 vge from 7/M collected from the liver virus-positive phase of POI . As a control , F/F was injected with 1010 virions of WHV/tm3 . Plasma and PBMC samples were collected weekly until 8 w . p . i . , biweekly until 6 m . p . i , and then bi-monthly . Liver samples were obtained before inoculation , 6–7 w . p . i . , 6 m . p . i . , then yearly , and at autopsy . PBMC and plasma were harvested from sodium EDTA-treated blood after density gradient centrifugation [7]–[9] . PBMC were cryopreserved , and serum and plasma samples stored at −20°C . Liver samples obtained at biopsy or autopsy were washed , snap frozen and stored at −80°C . For histological examination , liver samples were processed to paraffin , stained and hepatic inflammatory alterations enumerated [7] , [8] . Liver neoplastic changes were assessed following morphological criteria reported before [7] , [16] . WHsAg , anti-WHc were evaluated by enzyme-linked immunosorbent assays ( ELISA ) reported previously [7]–[9] , [13] , with sensitivities comparable to or greater than those of clinical assays for detection of equivalent HBV infection markers . The sensitivity of WHsAg ELISA was 3 . 25 ng/mL while anti-WHc were detectable up to end-point dilution of 1∶64 , 000 . Serum SDH served as a biochemical measure of liver injury and serum GGT as an indicator of HCC development [7] , [13] . DNA from 100–400 µL of serum or plasma and from PBMC , liver , bone marrow and lymph nodes was extracted by the proteinase K-phenol-chloroform method [6] , [7] . WHV DNA was assessed by direct and , if negative , nested PCR/NAH using primers and conditions reported [7]–[9] . Each sample was tested with 3 primer sets specific for WHV core ( C ) , envelope ( S ) and X genes [7]–[9] . For nested PCR/NAH detecting WHV covalently closed circular DNA ( cccDNA ) ( sensitivity , ∼102 copies/mL ) , enzymatic treatment , primers and conditions previously established were applied [8] , [13] . Detection of WHV cccDNA was verified by sequencing . WHV RNA was detected by reverse transcription-PCR ( RT-PCR; sensitivity , <10 copies/mL ) using RNA extracted with Trizol ( Invitrogen Life Technologies , Burlington , Canada ) , treated with DNase ( Sigma-Aldrich , Oakville , Canada ) , and reversely transcribed to cDNA [8] , [13] . Each test RNA sample without reverse transcriptase added served as a DNA contamination control [8] , [13] . In selected cases , WHV DNA was quantified by real-time PCR ( sensitivity , 10–100 vge/mL ) using DNA equivalent to 25 µL of plasma or 400 ng of total DNA from cells or tissues , and WHV C and X gene primers . For all assays testing WHV DNA or RNA presence , mock extractions and respective nucleic acid preparations from WHV-positive and WHV-negative woodchuck livers or PBMC were routinely included as controls [6]–[8] . NAH analysis of PCR products was always performed to verify the specificity of virus detection and the validity of controls [6]–[8] . Low levels of WHV DNA in POI made full virus genome amplifications unfeasible , therefore fragments amplified with C , S and X gene-specific primers and regions spanning WHV polymerase ( P ) gene between nucleotides ( nt ) 2948-407 and 1080–1755 , X/preC region nt 1503–2122 and preS nt 2948-407 were sequenced ( nt positions according to WHV/tm3 AY334075 in GenBank ) . These regions were selected because they were found to have the most variable sequence based on analysis of full-length WHV genomes using Sequencher v5 ( Gene Codes Corporation , Ann Arbor , MI ) . Amplicons were cloned using the TOPO-TA system ( Invitrogen ) . Ten clones per amplicon were sequenced bidirectionally [17] . The same variants found in at least 2 clones were reported . Liver and bone marrow DNA of 10–20 kbp purified from agarose served as a template for inverse-PCR ( invPCR ) , as reported [18] . To identify WHV X region-host genome junctions , DNA was digested with Nsi-I that cuts WHV at nt 1915 ( nt positions according to WHV/tm3 AY334075 in GenBank ) and the woodchuck's sequence at unknown sites . To detect WHV preS region-host genome junctions , DNA was treated with EcoR-I that cuts WHV/tm3 at nt 3308/1 . Diluted digests were circularized with T4 DNA ligase and linearized with Sph-I ( for X invPCR ) or Pst-I ( for preS invPCR ) . The possibility of self-ligated virus double-stranded DNA was excluded by Psi-I or Pml-I digestion . Primers were designed based on consensus sequence of WHV isolates identified in this laboratory ( GenBank accession numbers: AY334075 , AY6280 and GU734791 ) . For the X region , direct and nested primer pairs were located at nt 1782–1808 and 1718–1737 , and 1853–1876 and 1654–1673 , respectively . For the preS region , direct primers were located at nt 3231–3253 and 3009–3028 , and nested primers at 3202–3222 and 2964–2985 . The bands carrying WHV sequences were identified by NAH . DNA was purified by excision from agarose and either directly sequenced bidirectionally or cloned and sequenced . Non-WHV sequences were analyzed with NCBI BLAST and Refseq ( National Center for Biotechnology Information , Bethesda , MD ) . WHV sequences were mapped by aligning with the full-length WHV/tm3 using BioEdit ( Ibis Biosciences , Carlsbad , CA ) . WHV sequences derived from the liver WHV-negative and liver WHV-positive phases of POI reported in this study were submitted to GenBank under accession numbers KJ755421 for woodchuck 6/M and KJ755420 for 7/M . WHV sequences identified in plasma and spleen of 8/M animal with POI have GenBank accession numbers KJ755405 , KJ755406 , KJ755410 , KJ755411 , KJ755415 and KJ755416 , while those in E/F woodchuck injected with plasma inoculum derived from 8/M animal have GenBank accession numbers KJ755407-KJ755409 , KJ755412-KJ755414 , and KJ755417-KJ755419 . WHV genome-woodchuck DNA integration sites identified in livers and bone marrows of animals with POI which developed HCC have accession numbers KG817076-85 , KG817088 , KG817089 , KG817091 and KG817092 , and those found in livers , PBMC and lymphoid tissues in woodchucks with POI without HCC have accession numbers KG817074 , KG817075 , KG817086 , KG817087 , KG817090 , and KG817093-99 . The sequence of woodchuck HCC H19 gene fragment identified in this study has GenBank accession number KG8117082 .
Animals inoculated with 10 or 100 virions of WHV/tm5 or WHV/tm3 , respectively , showed no serological evidence of WHV infection for up to 5 . 5 years p . i . , as revealed by undetectable serum WHsAg and anti-WHc ( Figs . 1A and 1B ) . Nonetheless , WHV DNA was detected in serum/plasma and PBMC throughout the entire follow-up at levels of 100–200 vge/mL or <103 vge/µg cell DNA , respectively ( Figs . 1A and 1B ) . In contrast , woodchucks injected with 106 or 1010 virions developed transient serum WHsAg positivity , anti-WHc for life , and biochemical ( not shown ) and histological evidence of self-limited AH ( SLAH ) ( Fig . 1C ) . WHV cccDNA and/or WHV RNA were identified in PBMC ( Fig . 2 and Fig . 3A ) throughout the lifespan and in lymphoid organs at autopsy in woodchucks with POI ( Fig . 3B ) , similarly as in animals with lifelong SOI continuing after SLAH and as reported [7]–[9] . Sequential plasma or serum , liver and PBMC samples from healthy WHV-naïve woodchucks serving as controls remained WHV DNA negative when tested by nested PCR/NAH , while the animals liver and PBMC samples were WHV RNA nonreactive by nested RT-PCR/NAH during the entire observation period ( data not shown ) . It was not until 32 to 40 m . p . i . that all animals injected with 10 or 100 virions became consistently liver WHV DNA reactive ( Fig . 1 ) and showed evidence of hepatic WHV replication , i . e . , detection of WHV cccDNA or WHV RNA or both ( Fig . 2 ) . However , 9/M showed transiently a low level of virus DNA in the liver at 6 w . p . i . , while 1/F and 5/M were reactive from 26 m . p . i and 6 w . p . i . onwards , respectively ( Figs . 1A and 1B ) . Liver samples from 2/F and 8/M were consistently WHV DNA negative , even up to autopsy performed at 22 and 34 m . p . i . , respectively . Liver histology and serum SDH levels remained entirely normal during the whole follow-up , except minimal inflammatory lesions limited to a few portal areas found at 32 m . p . i . in 5/M and at autopsy in 1/F and 10/M ( Fig . 1 ) . Despite this infection pattern , typical multinodular HCC has developed in 5/M and 9/M at 55 m . p . i . ( Fig . 1B ) . The diameter of tumor nodules ranged between 2–3 mm ( numerous ) to 1 . 5–2 cm ( singular ) and they were spread throughout the entire livers . Histological examination revealed foci of well-differentiated HCC with hepatocytes arranged in trabeculea ( Fig . 4 ) and , occasionally , with regions of compact cancer tissue . The HCC appearance coincided with moderately elevated serum GGT levels ( data not shown ) . Control woodchucks injected with liver pathogenic doses of WHV showed transiently elevated serum SDH ( data not shown ) and SLAH followed by SOI accompanied by persistent low-level WHV replication in both liver and PBMC , and intermittent minimal to mild liver inflammation ( Fig . 1C ) , as reported [7] , [9] . One of the woodchucks ( 11/F ) inoculated with 106 virions of WHV/tm5 developed HCC at 70 m . p . i . ( Fig . 1C ) . Healthy controls not exposed to WHV had normal serum SDH and GGT levels during follow-up . Their livers remained normal during their lifespan when inspected macroscopically during laparotomies and by histological examination of serial biopsies obtained at approximately yearly intervals ( data not shown ) . To determine whether virus persisting as POI retained its infective and pathogenic properties , WHV recovered by ultracentrifugation from pooled serum/plasma collected from the liver WHV-negative or the liver WHV-positive POI phases was administered at doses between 1370 and 2070 virions to virus-naïve woodchucks . All animals developed transiently serum WHsAg-positive infection from 57–84 d . p . i . lasting for up to 113 d . p . i ( Fig . 5 ) . The WHsAg appearance was delayed by 22–49 days when compared to F/F control injected with 1010 virions . Anti-WHc became detectable at 70–113 d . p . i . ( at 57 d . p . i . in F/F ) and persisted to the end of follow-up . In animals inoculated with WHV from the liver virus-negative phase of POI , hepatic WHV load at 7 w . p . i . ranged from 30 to 100 vge/µg DNA , whereas in those with WHV from the liver virus-positive phase between 2 . 5×103 and 1 . 1×106 vge/µg DNA ( 9 . 5×106 vge/µg DNA for F/F ) . However , subsequent liver biopsies showed comparable WHV DNA levels ranging between 2×102 and 2×103 vge/µg DNA . Similar WHV DNA loads were detected in sera ( 10-102 vge/mL ) and PBMC ( 10-102 vge/µg DNA ) from the beginning of infection regardless of the inoculum source , but these levels subsequently increased by 10–100-fold . All animals ( n = 5 ) developed mild to minimal hepatitis that persisted through the observation period ( Fig . 5 ) . Interestingly , 3 of them , including two inoculated with WHV from the liver virus-negative POI phase , developed HCC within 4 . 5 to 35 m . p . i . accompanied by a variable degree of hepatitis ( Fig . 5 ) . To recognize whether initiation of the liver virus-positive phase of POI might be related to the emergence of a specific WHV variant , 2060-bp of WHV sequences derived from the liver virus-negative and liver virus-positive phases of POI from 6/M and 7/M were compared . The results showed that the WHV sequence from the liver-virus negative phase of 6/M differed only by one non-synonymous mutation in the preC region when compared to that of the virus from the liver virus-positive phase ( Table 1 ) . When WHV sequences from the equivalent phases from 7/M were compared , WHV from the liver virus-negative period showed 5 non-synonymous mutations not encountered in the virus from the liver WHV-positive phase ( Table 1 ) . However , none of the mutations were compatible with that identified in the preC region of 6/M WHV sequence , suggesting that unlikely a unique hepatotropic variant initiated the liver virus-positive phase of POI . To determine whether WHV derived from the liver-virus negative POI phase retained its sequence after administration to a virus-naïve host , WHV sequences in inoculum and spleen from 8/M , which remained liver virus-negative until autopsy ( Fig . 1B ) , and WHV from plasma , PBMC and liver from E/F , which was injected with 8/M inoculum ( Fig . 5 ) , were compared to each other and to WHV/tm3 . This analysis showed that 2060-bp of the WHV sequence from 8/M inoculum and spleen displayed very few point mutations when compared to WHV/tm3 ( 11/2060 ) , 8 conferred amino acid changes and 6 occurrrd in both samples ( Table 2 ) . WHV sequences from serum and liver of E/F were highly compatible to that of 8/M inoculum , while E/F PBMC showed a number of non-synonymous variants which were unaccounted for in WHV/tm3 inoculum ( n = 31 ) , 8/M inoculum ( n = 25 ) or E/F serum or liver ( n = 21 ) ( Table 2 ) , suggesting that the virus after transmission propagated most actively in the lymphoid cells . Multiple WHV DNA-host genome junctions were identified in animals with POI which developed HCC ( 5/M and 9/M ) or not ( 1/F , 2/F , 7/M and 10/M ) ( Table 3 ) . Among virus-host integrants detected in liver biopsy and autopsy samples from 5/M and 9/M , various host sequences were joined predominantly with WHV X gene and less often with the polymerase ( P ) gene , and preS region sequences ( Table 3 ) . None of the virus-host integration sites was identified more than once in the material investigated; however , particular junctions were frequently found in more than one clone ( Table 3 ) . Notably , in HCC tissue from 9/M , the 264-bp host sequence flanked by the virus X gene sequence showed 80% homology with mouse H19 cDNA ( GenBank accession number AF214115 . 1 ) . H19 is a tumor suppressor gene and its knockdown may play a role in HCC development [19] . Further to virus-host junctions , multiple virus DNA rearrangements were identified in liver samples from animals with POI-associated HCC , but less frequently in those without cancer ( Table 3 ) . Viral-host junctions were also detected in autopsy bone marrow , lymph node and PBMC samples in all 6 animals analyzed ( Table 3 ) . 4/F and 6/M with POI lasting for 5 . 5 years were challenged with 1010 virions of WHVtm3 to determine whether the animals might be protected from reinfection . Both woodchucks became serum WHsAg positive at 2 w . p . c . and remained positive until 18 w . p . c . ( Fig . 6 ) . Anti-WHc became detectable from 13–14 w . p . c . Serum SDH levels increased and peaked at 8–14 w . p . c . , while liver histology displayed moderate to severe AH ( Fig . 6 ) . WHV DNA levels in serum and PBMC were similar to those detected in control animals over the course of SLAH ( Fig . 1C ) . Additionally , samples collected from 4/F and 6/M at autopsy , when serum WHsAg was undetectable , displayed low levels of WHV DNA in serum , liver and lymphatic organs , implying existence of SOI . Thus , both 4/F and 6/M developed acute hepatitis despite being persistently infected with WHV at a low-level . This was in contrast to 13/F with established SOI , which was protected from challenge ( Fig . 6 ) , similarly as previously reported [7] , [9] , [13] .
We uncovered that minute amounts of hepadnavirus establish infection that persists indefinitely in the woodchuck model of hepatitis B in the absence of conventional serological markers of infection and hepatitis , but is detectable molecularly when sensitive virus nucleic acid-specific amplification assays are applied . We also documented that this form of asymptomatically hepadnaviral carriage , designated previously as POI [5] , [9] , [13] , has both pathogenic and epidemiological relevance since it can lead to the development of HCC and , under certain conditions , transmit infection and cause hepatitis and HCC in virus-naïve hosts . Another important finding , albeit expected , was that POI is associated with hepadnavirus DNA integration into the hepatic and immune system DNA , which likely underpins liver oncogenic potency of the virus persisting during the course of this asymptomatic form of hepadnaviral carriage . The results from the current investigations also showed that during POI , WHV replication expands to the liver with time , but the level and/or type of cells infected appear to be inadequate to trigger hepatitis . In previous studies , woodchucks with experimental POI were followed for up to 25 m . p . i . without detection of WHV in the liver or evidence of HCC , while WHV replication was detectable in circulating and organ lymphoid cells [9] , [13] . Also , offspring born to woodchuck dams with SOI , which acquired lymphatic system-restricted POI , did not show liver engagement and the development of HCC during the 42-month observation period [6] . Although there might be several factors contributing to the development of HCC during POI , the virus spreading to the liver and the extended period of POI follow-up appear to be critical . WHV genome fragments from the liver virus-negative and the liver virus-positive POI phases showed essentially the same predicted amino acid sequences ( Table 1 ) , which also were highly compatible to that of wild-type WHV inocula used to induce POI in this study . We analyzed more than 62% of the total WHV sequence , including virus regions identified as having the highest sequence variability based on our preceding analysis of the complete WHV sequences reported in GenBank . We used this approach because the trace quantities of WHV found during POI and the inherently lower sensitivity of the extended PCR amplifying long WHV sequences made full virus genome amplification not feasible . We also identified that the WHV/tm3 sequence , as far as we were able to determine , was conserved in the animals injected with WHV prepared from the liver WHV-negative or the liver WHV-positive phases of POI , which developed WHV infection engaging both the liver and the lymphatic system . These findings imply that the dual tropism of WHV towards hepatocytes and immune cells is unlikely due to the existence of cell type-specific viral variants but is an intrinsic propensity of the naturally occurring virus . This is consistent with data from in vitro infection experiments in which the same wild-type WHV was serially passaged in cultured woodchuck hepatocytes and lymphoid cells [17] . This issue has not yet been investigated in HBV infection . Our previous studies showed that serologically overt WHV infection coinciding with hepatitis is resultant from i . v . administration of WHV doses greater than 1×103 virions ( liver pathogenic doses ) , while lower doses of the same wild-type virus ( liver non-pathogenic doses ) consistently induced POI in woodchucks [9] , [13] . In the current study , concentration by ultracentrifugation of virus from animals with POI to levels above the previously identified liver pathogenic threshold was accomplished and , as documented , the recovered virus readily induced serologically overt infection and hepatitis upon transmission to virus-naïve animals ( see Fig . 5 ) . It appears that the inability of WHV to engage the liver during the initial phase of POI was related to the very low quantities of the produced virus which , however , can be temporally augmented to the level sufficient to invade the liver . In this regard , we detected a transient increase in plasma WHV load to approximately or above 1×103 vge/mL that preceded detection of WHV DNA and its replication intermediates in hepatic tissue . It can be assumed that this temporal increase in circulating WHV was adequate to engage the liver during later phase of POI , which prior to that was restricted to the lymphatic system . This appears to be consistent with identification of a 100 to 1000-fold greater affinity of synthetic analogues of WHV cell binding site for activated woodchuck lymphoid cells than woodchuck hepatocytes , suggesting that very low quantities of virus may preferentially invade the immune system [20] , [21] . The mechanism of liver carcinogenesis in hepadnaviral infection is not well understood , but it is likely a multistep process in which persistent virus infection and virus genome integration into host DNA are among the principal contributors [22] . Random HBV DNA integration into the liver genome was found in up to 22% of patients with CHB and is a typical finding in HBV-related HCC ( >80% patients ) [22]–[24] . On the other hand , the status of HBV DNA integration into HCC DNA coinciding with occult HBV infection was only occasionally investigated and mainly in cirrhotic patients [11] , [12] . Nonetheless , the data convincingly showed that HBV DNA integrates into both HCC and non-HCC liver DNA in serum HBsAg-negative patients , with or without detectable anti-HBc [11] , [12] . In WHV-related HCC , virus DNA insertions were identified in tumors developing during chronic hepatitis and SOI continuing after SLAH [25] . WHV DNA integration was frequently found near the myc pro-oncogenes in HCC coinciding with chronic WHV hepatitis [26] , [27] . We did not find this relation in woodchucks developing HCC during POI . However , this might become more apparent when a greater number of relevant cases are analyzed . About two-thirds of the virus-host genome junctions detected in this study encompassed the WHV X gene sequence ( Table 3 ) . This resembles the predisposition of HBV X gene to integrate into the host genome reported in serum HBsAg-negative patients with HCC [28] . In our study , HCC had developed in the absence of hepatitis and cirrhosis . In contrast to CHB , chronic WHV hepatitis never leads to cirrhosis and very rarely to fibrosis ( <1% ) [4] . However , the occurrence of HBV-related HCC in the absence of cirrhosis has been reported [28] . The present finding of the POI-associated HCC mimics the human disease situation where HBV-related HCC develops in the absence of apparent chronic liver disease and serological evidence of HBV infection . Notably , the finding of WHV DNA sequence insertions within bone marrow DNA in POI parallels HBV DNA and WHV DNA integration into lymphoid cells and lymphatic organ genomes in CHB and in woodchucks with chronic hepatitis and SOI [29] , [30] . This study also revealed that POI during lifelong follow-up did not culminate in serologically apparent infection or hepatitis , and did not induce protective immunity . These findings add new dimensions to the previous investigations on POI [6] , [9] , [13] . Among others , our previous study showed that repeated i . v . injections ( 12 in total ) with 100 WHV virions did not initiate serologically detectable infection or hepatitis , but molecularly evident POI was established and continued until challenge with a liver pathogenic dose ( >103 virions ) of the same virus inoculum [13] . In the current study ( data not shown ) and in the previous investigations [9] , [13] , [31] , [32] , WHV-specific T cell reactivity occurring in the absence of virus-specific antibody response did not protect from challenge with liver pathogenic doses of WHV ( >103 virions ) . This is in marked contrast to WHV-specific T cell responses coinciding with virus-specific antibodies in SOI continuing after recovery from symptomatic WHV infection and hepatitis which yield total protection against challenge with even massive doses of WHV ( >1010 virions ) [7] , [31] . The former may parallel a situation in unvaccinated individuals having repeated contacts with infected persons and intravenous drug users repetitively exposed to small amounts of HBV . Our data implies that these individuals would unlikely become serum HBsAg and anti-HBc reactive or immune to HBV infection , but the development of HCC in such persons cannot be excluded . Nonetheless , the current findings are in contrast to data indicating that one virion of HBV derived from a HBV transgenic mouse was able to induce serologically evident chronic hepatitis in chimpanzee [33] . Differences in the liver pathogenic potency between a single HBV isolate from a transgenic mice and intact , naturally occurring WHV might explain this discrepancy . Although the existence of POI in humans has not yet been thoroughly investigated , the prevalence of HBV DNA-reactive infection seronegative for HBsAg and anti-HBc has been reported between 0 . 07 and 7 . 6% of subjects in different areas of HBV endemicity [12] , [34] . It can be expected that HBV POI is much more frequent because the assays available for HBV DNA detection are approximately 10–100-fold less sensitive than these utilized in this study . Further , HBV-specific T cell responses in the absence of serum HBsAg and anti-HBc have been identified in HBV DNA-reactive patients , further supporting that this silent form of HBV infection naturally occurs [14] . It is of note that WHV-specific T cell responses were also examined in the current study and they persisted at borderline levels after a period of heightened reactivity lasting between 6 and 20 w . p . i . ( data not shown ) . In conclusion , this study revealed the oncogenic capacity and potential epidemiological significance of asymptomatic hepadnaviral carriage initiated by very small amounts of otherwise pathogenic virus that advances in the absence of traditional serological markers of infection and hepatitis . The data emphasize the role for primary occult HBV infection in the development of seemingly cyptogenic HCC in HBV seronegative patients .
|
Introduction of highly sensitive molecular assays for detection of hepatitis B virus ( HBV ) identified the existence of persistent occult HBV infection years after recovery from an episode of hepatitis B and in individuals exposed to HBV but without symptoms and classical markers of infection . Because HBV integrates into human DNA and is a potent human carcinogen , it is postulated that occult HBV infection can be a cause of hepatic cancer in many individuals in which the tumor origin remains currently unknown . A causative relation between occult HBV infection and hepatocarcinoma is highly challenging to investigate in humans since occult HBV persistence is rarely diagnosed with current clinical assays and cancer development takes 15–30 years . However , we have established excellent models of occult HBV infection in the eastern North American woodchucks which are naturally susceptible to a virus closely related to HBV and in which chronic infection advances to liver cancer . In the current study , exploring experimental primary occult infection in woodchucks , we proved that the silently progressing infection , which is not detectable by serological markers , can culminate in hepatocellular carcinoma and that the persisting virus remains infectious , and causes hepatitis and liver cancer when transmitted to virus-naïve hosts .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases",
"infectious",
"disease",
"immunology",
"medicine",
"and",
"health",
"sciences",
"infectious",
"hepatitis",
"hepatitis",
"b",
"hepatitis",
"clinical",
"immunology",
"gastroenterology",
"and",
"hepatology",
"biology",
"and",
"life",
"sciences",
"immunology",
"viral",
"diseases",
"liver",
"diseases",
"hepatocellular",
"carcinoma"
] |
2014
|
Primary Seronegative but Molecularly Evident Hepadnaviral Infection Engages Liver and Induces Hepatocarcinoma in the Woodchuck Model of Hepatitis B
|
PARN is one of several deadenylase enzymes present in mammalian cells , and as such the contribution it makes to the regulation of gene expression is unclear . To address this , we performed global mRNA expression and half-life analysis on mouse myoblasts depleted of PARN . PARN knockdown resulted in the stabilization of 40 mRNAs , including that encoding the mRNA decay factor ZFP36L2 . Additional experiments demonstrated that PARN knockdown induced an increase in Zfp36l2 poly ( A ) tail length as well as increased translation . The elements responsible for PARN-dependent regulation lie within the 3′ UTR of the mRNA . Surprisingly , changes in mRNA stability showed an inverse correlation with mRNA abundance; stabilized transcripts showed either no change or a decrease in mRNA abundance . Moreover , we found that stabilized mRNAs had reduced accumulation of pre–mRNA , consistent with lower transcription rates . This presents compelling evidence for the coupling of mRNA decay and transcription to buffer mRNA abundances . Although PARN knockdown altered decay of relatively few mRNAs , there was a much larger effect on global gene expression . Many of the mRNAs whose abundance was reduced by PARN knockdown encode factors required for cell migration and adhesion . The biological relevance of this observation was demonstrated by the fact that PARN KD cells migrate faster in wound-healing assays . Collectively , these data indicate that PARN modulates decay of a defined set of mRNAs in mammalian cells and implicate this deadenylase in coordinating control of genes required for cell movement .
The poly ( A ) tail added to mRNAs during processing in the nucleus stimulates mRNA export and translation through its association with poly ( A ) -binding proteins . In contrast , the removal of the poly ( A ) tail renders transcripts translationally silent and is also the first step in decay of the majority of transcripts in eukaryotic cells [1] . As such , the process of deadenylation has the ability to profoundly influence cellular gene expression on multiple levels . Numerous mammalian deadenylases have been identified and characterized to varying extents . They fall into two enzymatic groups; the DEDD-type ( including PARN , CAF1/CNOT7 and PAN2 ) which bear an Asp-Glu-Asp-Asp motif in their active site , and the Exonuclease/Endonuclease/Phosphatase ( EEP ) type ( including CCR4/CNOT6 , Nocturnin ( CCRN4L ) and Angel proteins ( ANGEL1 , ANGEL2 ) ) [2] . Of the many known poly ( A ) shortening enzymes , the CCR4/NOT complex and PARN are by far the best studied . CCR4/NOT represents the major cytoplasmic deadenylase in yeast where it initiates decay of the majority of mRNAs [3] . The yeast CCR4/NOT deadenylase is a large complex and contains two subunits with deadenylase activity ( Ccr4p and Caf1p ) as well as several other factors including the NOT proteins ( Not1p-Not5p ) . In mammals , there are five CCR4-like proteins ( CNOT6 , CNOT6L , CCR4N4L/Nocturnin , ANGEL1 and ANGEL2 ) and three CAF1-like proteins ( CNOT7 , CNOT8 and CAF1Z/TOE1 ) . Of these , CNOT6 , CNOT6L , CNOT7 and CNOT8 associate with the mammalian NOT proteins to form various CCR4/NOT complexes [4] . In mammalian cells , CCR4/NOT complexes have been implicated in both miRNA-mediated and AU-rich element ( ARE ) mediated mRNA decay mechanisms . CCR4/NOT is recruited to mRNAs by the miRNA-associated GW182 protein [5] , and by the ARE-binding protein tristetraprolin ( TTP/ZFP36 ) [6] . Thus , for mRNAs bearing certain sequence determinants , the CCR4/NOT class of related deadenylases has an important role to play in initiating mRNA decay . On a biochemical level , PARN is perhaps the best understood deadenylase , in part because it is the predominant activity in mammalian cell extracts [7] . PARN is unique in being able to interact with both the cap and poly ( A ) tail [7]–[9] and has been linked with mRNA decay mechanisms in both the nucleus [10] and the cytoplasm [7] , [11] . In the cytoplasm , PARN plays an important role in controlling gene expression during the maternal-zygotic transition in Xenopus [12] and during the DNA damage response in mammalian cells [11] . In the nucleus , PARN has been linked with decay of transcripts undergoing 3′ end formation following DNA damage [10] and is important for trimming the 3′ ends of snoRNAs during their maturation [13] . In addition , PARN interacts directly with RNA-associated factors including CELF1/CUGBP1 [14] , PUM2 [15] and CPEB [16] . Very few bona fide mRNA substrates of PARN in mammals have been identified to date . The Fos and Myc mRNAs exhibit increased abundance following PARN KD in HeLa cells [10] and Tnfa transcripts are deadenylated in a PARN-dependent manner in vitro [17] but to our knowledge there is no published evidence for a direct effect of PARN on mRNA stability in living mammalian cells . Because of the large number and diversity of deadenylase activities in mammalian cells it has been challenging to discern their individual roles and their global impact on cell function . It remains unknown whether each mRNA must be targeted by a specific deadenylase to achieve appropriate control of gene expression . The impact of deadenylase activity on mRNA decay rates , mRNA abundance and translation efficiency is also not clear . Previous attempts to address these questions using RNA interference approaches have suggested partially overlapping roles for the CCR4-like ( CNOT6 , CNOT6L ) and CAF1-like ( CNOT7 , CNOT8 ) deadenylases [18] . Surprisingly , less than 2% of all mRNAs showed changes in abundance following depletion of either CNOT6/CNOT6L or CNOT7/CNOT8 [18] , implying that there is either redundancy in function between the many different deadenylase enzymes and/or that changes in mRNA abundance are not a good measure of deadenylase impact . Global measurements of mRNA decay rates following knockdown of deadenylases are necessary in order to distinguish these possibilities . In this study we aimed to shed light on the role of PARN deadenylase in C2C12 myoblasts by directly assaying global mRNA decay rates and mRNA abundances following knockdown of PARN . We identified a relatively small set of 40 mRNAs whose decay was reduced following PARN KD and independently verified this observation for four of these transcripts . For Zfp36l2 mRNA we also showed that PARN knockdown induces elongation of the poly ( A ) tail and increased protein abundance . Enhanced translation efficiency in PARN KD cells was also observed for a reporter bearing the Zfp36l2 3′UTR . We conclude that PARN is directly required for deadenylation of Zfp36l2 and almost certainly other mRNAs within the stabilized set . Interestingly , slower mRNA decay did not result in the expected increases in abundance for the majority of stabilized mRNAs . We attribute this to reduced transcription rates , supporting the recently established idea of coupling between mRNA decay and transcription [19]–[22] . We also investigated the effects of PARN depletion on cellular function . We determined that loss of PARN activity decreased the abundance of transcripts encoding factors linked with cell adhesion and cell movement; processes that require extracellular matrix ( ECM ) interactions . This led to the discovery that depletion of PARN enhances wound healing in C2C12 myoblasts .
We used a lentiviral vector encoding an shRNA targeting the 3′UTR of murine Parn to generate a stable clonal C2C12 myoblast line with reduced expression of PARN ( PARN KD ) . Parn mRNA and protein abundance were evaluated by qRT-PCR ( Figure 1A ) and western blotting ( Figure 1B ) in the PARN KD cell line and in a cell line generated with a control lentiviral vector lacking shRNA sequences ( CTRL ) . The PARN KD cell line showed a robust reduction in PARN expression ( Figure 1A and 1B ) . We first wanted to assess mRNA decay rates in the PARN KD cells and compare them to those we obtained previously in the CTRL cell line [23] . Briefly , both cell lines were treated with Actinomycin D ( Act-D ) for 30 minutes and samples were collected at 0 , 10 , 50 , 110 and 230 minutes after transcription inhibition . Total RNA was isolated from each sample and used to generate cDNA probes for hybridization to microarrays . The experiment was repeated in triplicate and three independent half-lives were generated for each transcript in each cell line by plotting the abundance at each time point and fitting to an exponential decay curve . As an example the half-lives for the Gpsm1 mRNA in the two cell lines are shown in Figure 1C . Each half-life was considered reliable if the data fit well to the curve ( p<0 . 05 ) and the 95% confidence interval was less than twice the half-life . We required that the half-life met these criteria for at least two of the three replicates . Both PARN KD and CTRL cells were assayed at the same time but analysis of the results from the CTRL cells was published previously [23] . Reliable half-lives were generated for 1581 mRNAs in the PARN KD cells ( Dataset S1A; GSE35944 ) . Although this dataset is somewhat smaller than that previously obtained for the CTRL cell line [7398 mRNAs; 23] , it is nevertheless large enough to be informative . Overall , we obtained half-lives in both cell lines for 1389 mRNAs ( Dataset S1B; GSE35944 ) . Comparison of half-lives in CTRL and PARN KD cells allowed us to identify 64 transcripts that showed a statistically significant difference in decay rate between the two cell lines with 40 transcripts showing stabilization and the remaining 24 being destabilized ( Table 1 and Table S1 , respectively ) . To ascertain that the microarray analysis reflected true changes in mRNA decay rates , we assayed half-lives following Act-D treatment for four of the stabilized transcripts ( Adora2b , Zfp36l2 , Gpsm1 and Ankrd54 ) by qRT-PCR ( Figure 2A–2D ) . These transcripts were selected because they have relatively short half-lives ( less than 2 hours ) allowing us to assess their decay over a time frame that minimizes the toxic effects of Act-D on the cell . All four transcripts were significantly more stable following PARN knockdown , as predicted by the microarray analysis . Moreover , instability of the Zfp36l2 mRNA was restored by transfection of an expression vector encoding shRNA-resistant human PARN demonstrating that stabilization was not caused by off-target effects of the shRNA on expression of unrelated genes ( Figure 2E and 2F ) . Thus , we conclude that the PARN deadenylase influences decay rates of a subset of mRNAs in mammalian cells . Given that PARN is a deadenylase , we predicted that mRNAs stabilized by PARN KD would show effects on the length of their poly ( A ) tail . We investigated this possibility for the Zfp36l2 mRNA using an RNase H/northern blotting approach . Briefly , total RNA isolated from CTRL and PARN KD cells was treated with an oligonucleotide and RNase H to induce cleavage ∼120 nt upstream of the poly ( A ) tail . After separation on a polyacrylamide gel followed by electroblotting , the 3′ fragment was detected using a radiolabeled probe complementary to the 3′UTR . As shown in Figure 3A and 3B , the poly ( A ) tail of Zfp36l2 mRNA was clearly elongated in PARN KD cells compared to the CTRL cells . In fact , in the CTRL cells the vast majority of Zfp36l2 mRNA had a surprisingly short poly ( A ) tail of just 20–30 nt . In the PARN KD cells the amount of Zfp36l2 mRNA with a long poly ( A ) tail of up to ∼190 nt was two to three fold more than in the CTRL cells . This was not a general effect on all mRNAs as the β-Actin ( Actb ) mRNA showed no difference in poly ( A ) tail length between the two cell lines ( Figure S1 ) . Although abundance of Zfp36l2 mRNA was similar in CTRL and PARN KD cells ( Figure 3C ) , western blotting ( Figure 3D ) demonstrated a small increase in abundance of ZFP36L2 protein which would be consistent with enhanced translation resulting from the elongation of the poly ( A ) tail . We also saw evidence for increased abundance of ZFP36L2 protein by immunofluorescence ( Figure S2 ) . In order to determine whether the effects of PARN on the Zfp36l2 mRNA are mediated by sequences in the 3′UTR we cloned the 3′UTR into a luciferase reporter construct ( Luc-36L2 ) and measured luciferase activity following transfection into CTRL and PARN KD cells . The empty vector ( Luc ) was used as a control and gave very similar activity regardless of whether expressed in the CTRL or PARN KD cells ( Figure 4A ) . Interestingly , the Luc-36L2 reporter produced significantly less luciferase activity than the Luc reporter in the control cell line suggesting that the sequences contained therein either repress translation or promote decay of the reporter mRNA . Importantly , PARN KD cells reproducibly exhibited a two-fold higher luciferase activity than the control cells ( Figure 4A ) and this was also seen when PARN was knocked down with a different shRNA ( Figure S3 ) showing that this effect is PARN-specific . Interestingly , the clear increase in luciferase activity following PARN depletion is mediated predominantly by enhanced translation as there was little effect on abundance of either reporter mRNA in PARN KD cells ( Figure 4B ) . The increase in luciferase expression is in the same range as the increase in abundance of endogenous ZFP36L2 protein in PARN KD cells ( Figure 3D ) . Together these results indicate that the action of PARN on the Luc-36L2 reporter results in translation repression presumably through poly ( A ) shortening . Moreover , factors associated specifically with the 3′UTR of Zfp36l2 mRNA are likely responsible for the effects of PARN on Zfp36l2 gene expression . At this time we do not know what factor might be responsible for recruiting PARN , but the Zfp36l2 3′UTR does have AU-rich elements like those reported to bind proteins such as TTP/ZFP36; a protein that induces PARN-mediated deadenylation in vitro [17] . The relatively small number of mRNAs affected by PARN at the level of mRNA stability precluded a meaningful analysis of Gene Ontology ( GO ) terms or sequences that might impacted by reduced PARN activity . Still , we did note that several of the stabilized transcripts encode proteins with roles in mRNA metabolism ( Toe1/Caf1z , Edc3 , Zfp36l2 , Dgcr14 , Nufip1 ) and transcription ( Gata2 , Zfp219 , Klf14 ) indicating that PARN may influence gene expression at multiple levels and impact a wider range of genes . To investigate this possibility we used the 0 minute time point from the array experiments to estimate global mRNA abundances in CTRL and PARN KD cells . We found that of the 18 , 201 transcripts detected , 1199 showed a 1 . 5-fold or greater change in mRNA abundance in PARN KD cells ( Dataset S2 ) . Surprisingly , given that PARN KD was expected to increase expression of its target mRNAs , the majority ( 63 . 7% ) of the affected mRNAs were down-regulated . We verified the abundance changes for several transcripts by qRT-PCR and found that of 14 mRNAs examined , all but one ( Lama2 ) showed changes similar to those predicted by the array ( Figure 5A ) . Moreover , there was generally a good correlation between the change predicted by the microarray and that observed by qRT-PCR in untreated cells although the qRT-PCR indicated changes of a greater magnitude than the array ( Figure S4 ) . This confirms that Act-D treatment did not globally affect our mRNA abundance measurements and that the 0 minute time point mRNA abundances are generally an acceptable indicator of relative differences in mRNA abundance between PARN KD and CTRL cell lines . We next took advantage of the availability of both mRNA abundance and decay data to analyze the impact of changes in mRNA stability on overall mRNA levels . We were surprised to discover that for the 40 transcripts showing clear evidence for stabilization following PARN knockdown , there was generally only a small effect on mRNA abundance and in many cases abundance was reduced rather than increased ( Figure 5B ) . There was a similar inverse correlation for the mRNAs that were destabilized ( Figure 5B ) . In order to verify this observation , we measured the abundance of three transcripts that were stabilized by PARN depletion in proliferating myoblasts ( Figure 5C ) . Interestingly , Adora2b mRNA ( 1 . 4-fold stabilized ( Figure 5D and Figure 2A ) ) showed ∼2-fold reduced abundance in PARN KD cells , while Ankrd54d mRNA ( 1 . 85-fold stabilized ) showed no statistically significant change in abundance ( Figure 5C ) . In contrast , Gpsm1 mRNA ( 1 . 96-fold stabilized ) did show a small increase in abundance by this assay . As described earlier ( Figure 3C ) , there was no significant change in abundance of the Zfp36l2 transcript despite a ∼2 . 4-fold increase in stability . Taken together , these results strongly suggest the existence of coupling between transcription and decay for many transcripts such that changes in mRNA decay rate are compensated for by opposing effects on transcription [20]–[22] . In order to further support this idea we assessed the abundance of newly transcribed pre-mRNAs for each of the four stabilized transcripts . Briefly , C2C12 cells were labeled for a short time with 4-thiouridine ( 4sU ) and total RNA was prepared . Newly transcribed 4sU-labeled RNAs were biotinylated and isolated on streptavidin beads . Pre-mRNAs were detected and quantified by qRT-PCR using one intronic primer and one exonic primer . As shown in Figure 5E , all four pre-mRNAs exhibited significantly reduced abundance in the PARN KD cells , consistent with slower transcription rates for these transcripts in this cell line . To summarize , each of the four mRNAs we evaluated showed increased stability following PARN KD ( Figure 5D ) but reduced levels of pre-mRNAs indicating reduced transcription ( Figure 5E ) . This change in the relative balance of decay and transcription results in only small changes in mRNA abundance ( Figure 5C ) . GO analysis using DAVID [24] revealed that the transcripts whose expression was most affected by PARN shared some interesting features ( Tables S2 and S3 ) . In particular , amongst the down-regulated genes there was a significant enrichment of mRNAs encoding proteins required for blood vessel development , cell adhesion , cell motion and axon guidance ( Table S2 ) . This is supported by the observation that a large proportion ( ∼15% ) of the down-regulated mRNAs encoded extracellular proteins including several collagens ( Col1a1 , Col1a2 , Col6a1 , Col6a2 , Col3a1 , Col12a1 ) , biglycan ( Bgn ) and matrix metalloproteases ( Mmp19 , Mmp2 ) . In contrast , the up-regulated mRNAs were more likely to encode components of large ribonucleoprotein complexes such as the ribosome and spliceosome ( Table S3 ) . Our GO analysis suggested that PARN knockdown might influence cell motility as cell movement requires extensive interactions with the extracellular matrix and is required for processes such as axon guidance and blood vessel development . We used a wound healing assay to investigate the ability of CTRL and PARN KD cells to migrate . Briefly , CTRL and PARN KD cells were grown to near confluence and then deprived of serum to prevent cell division . The monolayer was scratched to remove cells and incubated for eight hours to permit cells to migrate into the wound . Wound healing was assessed by counting the number of cells present within the boundaries of the wound . There was a clear increase in the wound healing capacity of PARN KD cells compared to the CTRL cells ( Figure 6A and 6B ) indicating that PARN KD cells migrate more rapidly . Moreover wound healing was restored to near normal levels following transfection of a plasmid encoding human PARN ( Figure 6C ) . We conclude that PARN modulates processes required for cell motility in C2C12 myoblasts .
In this study we identified a set of mRNAs whose decay is dependent on PARN deadenylase . For one stabilized mRNA , Zfp36l2 , we demonstrated that PARN-dependent regulation is mediated through sequences in the 3′UTR and that poly ( A ) tail length is increased following PARN KD . Depletion of PARN leads to increased ZFP36L2 protein abundance , but has negligible effects on mRNA abundance . Unexpectedly , we found that for the majority of affected transcripts mRNA stabilization slightly reduces mRNA abundance suggesting that mRNA decay rates are coupled to transcription . Finally , abundance of mRNAs encoding extracellular factors required for cell motility and adhesion was decreased by PARN knockdown and this observation led to the discovery that PARN KD cells migrate significantly faster than control cells in wound healing assays . To our knowledge , ours is the first study to examine the role of the PARN deadenylase in mammalian cells , and the first to examine the global impact of depletion of an mRNA decay enzyme on mRNA decay rates . Our results suggest that while PARN directly impacts decay of relatively few transcripts , it has surprisingly wide-ranging effects on expression of over 1000 genes . This could reflect that some of the genes directly regulated by PARN have important roles in regulating transcription and other cellular processes , generating a knock-on effect . In addition , PARN-mediated deadenylation also clearly regulates translation efficiency ( Figure 4 ) , perhaps in some cases without altering mRNA decay rates . Any mRNAs whose poly ( A ) tail length is increased without a dramatic change in mRNA decay rate in PARN KD cells would not be detected by our analysis , although downstream effects of such regulation could be picked up as mRNA abundance changes . PARN is known to induce reversible deadenylation as a means to silence translation [16] , however further experimentation will be required to distinguish targets regulated in this manner . Despite the fact that poly ( A ) shortening is thought to enhance decay of mRNAs , we detected 24 mRNAs that were actually less stable following PARN KD . Some of these may be direct targets; perhaps when PARN is depleted a more aggressive decay pathway substitutes . However , given the wide-ranging effects of PARN KD on gene expression , we feel it is more likely that destabilization is an indirect effect of the PARN KD mediated by a factor ( s ) encoded by one of the stabilized transcripts ( such as ZFP36L2 , CAF1Z/TOE1 or EDC3 ) . It also remains possible that some of these mRNAs are destabilized through off-target effects of the shRNA used to deplete PARN . Future experiments will aim to distinguish between these possibilities . PARN KD stabilizes 40 of the 1389 mRNAs ( 2 . 9% ) that we generated half-lives for in both cell lines . Remarkably , stabilization resulted in a decrease in abundance , or no significant change in abundance for the majority of these affected mRNAs ( Table 1 , Figure 3C , Figure 5B and 5C ) . This was seen in both the microarray and the qRT-PCR analyses . Although this seems counterintuitive , our results are actually very similar to recent observations on mRNA stability and abundance made in two closely related yeast strains [22] . These authors determined that as many as half of the evolutionary changes in mRNA degradation rates between S . cerevisiae and S . paradoxus were coupled to opposing changes in transcription rates . It was suggested that such coupling facilitates transient responses to environmental stimuli by enabling a more rapid return to basal expression levels . Additional studies , also in yeast , have established that mRNA decay rates are dependent on events that occur at the promoter [19]–[21] demonstrating that communication between transcription and mRNA turnover pathways exists . In yeast , some of this coupling has been attributed to the Rpb4/7 subunits of RNA polymerase II and to the CCR4-NOT complex , each of which have roles in both transcription and mRNA decay [19] , [22] . Our data strongly imply that the cell attempts to compensate for loss of PARN by reducing transcription to maintain appropriate mRNA abundance . Interestingly , many stabilized transcripts appear to have slightly decreased abundance in PARN KD cells ( Figure 5B ) . This overshoot suggests that the feedback mechanism is perhaps not highly accurate , or that it may additionally compensate for increased translation efficiency . Further investigation will be required to understand the mechanisms behind this feedback as so far there is no evidence that PARN modulates transcription directly , although it has been linked with mRNA 3′end processing events [10] . It is important to note that if extensive coupling of transcription and decay exists then mRNA abundance should be considered a poor indicator of both the magnitude and direction of effects on mRNA stability . This may explain why an earlier study found that depletion of CNOT6 and CNOT7 deadenylases had a relatively minor impact on overall gene expression [18] . Cell movement requires tightly controlled interactions between the cell and the ECM coupled with dynamic changes in the cytoskeleton . It can be described in three basic phases: Protrusion of the leading edge , adherence of the leading edge to the substrate and detachment of the cell body and trailing edge from the substrate [25] . Increased migration can be achieved by increased protrusion rate , by more efficient adherence to the substrate or by more rapid detachment from the substrate . While protrusion rate is primarily dependent on cytoskeletal dynamics , adherence and attachment can be influenced by cellular proteases or by the composition of the ECM . In general , those cell types that migrate most rapidly , such as leukocytes , have weaker interactions with the substrate whereas fibroblasts and myoblasts have stronger contacts and move more slowly [26] . Thus , the increased motility of PARN KD cells could be a direct result of alterations in the ECM caused by down-regulation of collagens , biglycan and other ECM components . Alternatively , increased motility could be due to altered expression of intracellular factors , such as TRIP10 and CDK5R1 . The Trip10/Cip4 mRNA ( stabilized 1 . 44-fold by PARN KD ) encodes Cdc42-interacting protein 4 which is localized to the leading edge of migrating cells and directly enhances cell motility through regulation of the actin cytoskeleton [27] . The Cdk5r1 mRNA ( stabilized 1 . 42-fold by PARN KD ) encodes p35 , an activator of the CDK5 kinase required for cell migration in neuroblastoma cells [28] and for myogenic differentiation [29] . Interestingly Cdk5r1 mRNA is subject to extensive post-transcriptional control through both miRNA- and ARE-mediated mechanisms [28] , [30] . Thus , increased expression of either CDK5R1 or TRIP10 proteins might directly enhance cell migration in PARN KD cells . The fact that PARN KD enhances wound healing and cellular motility is interesting in light of previous observations that depletion of several RNA-binding proteins can affect wound healing in C2C12 cells; depletion of hnRNPD/AUF1 , ELAVL1/HuR or IGF2BP2 impaired wound healing capacity and motility [31] . Another RBP , the zipcode binding protein IGF2BP1 , has been implicated in regulating localized expression of mRNAs involved in cell adhesion in breast cancer cells [32] . Finally , the Drosophila 5′-3′ exoribonuclease Pacman ( XRN1 in mammals ) is required for normal wound healing [33] . These results suggest that factors important for cell motility may be subject to extensive post-transcriptional control . Future studies will aim to further characterize the phenotype of PARN KD cells in order to decipher the mechanism by which cell motility is affected . It will also be interesting to determine whether PARN acts primarily on nuclear or cytoplasmic mRNAs . Finally , a high priority for future research is to uncover the mechanism by which changes in mRNA decay rates are signaled to the nucleus to influence transcription and to determine whether this is a global phenomenon in mammalian cells .
The mouse C2C12 myoblast cell line was obtained from the American Type Culture Collection ( CRL1772 ) . Two derivatives of the C2C12 cell line , CTRL and PARN KD , were cultured in Dulbecco's Modified Eagle's Medium ( DMEM ) containing 10% Fetal Bovine Serum ( FBS ) , 1 µg/ml puromycin , 10 U/ml penicillin and 10 µg/ml streptomycin in 5% CO2 at 37°C . Cells were maintained at or below 70% confluency except during wound healing assays . Transfections were performed using Lipofectamine 2000 as described previously [34] and transfection efficiency was routinely in the 50–70% range . The CTRL cell line stably transfected with LKO1 vector was described previously [23] . The PARN KD clonal cell line was generated by puromycin selection following transduction with an shRNA-encoding lentivirus derived from the LKO1 vector [35] . This vector is described in more detail below . The half-life experiment , microarray hybridization and analysis were all performed as described previously [23] . Briefly , CTRL and PARN KD cells were treated with Act-D ( 8 µg/ml ) for 30 minutes prior to the start of the time course . Total RNA was isolated at several time points using TRIzol ( Invitrogen ) according to the manufacturer's directions . 300 ng of total RNA were used to generate labeled cDNA fragments for hybridization to Mouse Affymetrix Gene 1 . 0 ST Arrays following the manufacturer's protocol ( GeneChip WT cDNA Synthesis Kit #900652 and #900720 ) . Production of probes and hybridization was performed by the Colorado State University Genomics and Proteomics Core Facility . Half-life experiments were conducted in triplicate , with each time point hybridized to a single array . For normalization of probe sets , we utilized the GC-bin method for background correction and applied median normalization by Affymetrix Power Tools ( APT ) with the ‘no adjustment’ option . Then , all probe set values were normalized to the 5th percentile value of all probe sets on the same array . Transcripts whose probe sets gave detection above background ( DABG ) p-value<0 . 05 in at least two out of three replicates at the 0 minute time point were considered expressed and used for subsequent analyses . A nonlinear least squares model [36] was used to calculate half-lives using the microarray data . A half-life measurement was considered reliable if it met both the following criteria: ( i ) the microarray data had a good fit to the nonlinear least squares model ( p-value<0 . 05 ) and ( ii ) the 95% confidence interval for the half-life was less than two times the half-life . Transcripts with reliable half-lives in at least two of three replicates were selected for further analyses . The mRNAs whose half-lives were significantly different in PARN KD compared to CTRL C2C12 cells were selected based on the t-test ( p-value<0 . 05 ) . The datasets were deposited in the GEO database ( GSE35944 ) . For functional analysis , lists of Gene IDs for those transcripts affected >1 . 5-fold were uploaded to the Database for Annotation , Visualization and Integrated Discovery ( DAVID; [24] ) along with the list of Gene IDs for all detected transcripts as Background . Functional clustering analysis was used to identify enriched groups of Gene Ontology ( GO ) terms . Clusters with enrichment scores of less than 1 . 3 ( equivalent to a p-value greater than 0 . 05 ) were excluded . Total RNA was isolated using the TRIzol ( Invitrogen ) method as recommended by the manufacturer . All samples were treated with DNase 1 to remove genomic DNA . In experiments using samples from cells transfected with luciferase plasmids an additional step was employed to ensure effective removal of plasmid DNA . After the initial DNase treatment , RNA was treated with EcoRI and EcoRV to digest residual plasmid DNA and treated a second time with DNase 1 . 1 µg of total RNA was reverse transcribed in the following conditions , according to the manufacturer's instructions: 35 mM Tris-Cl pH 8 . 3 , 50 mM NaCl , 5 mM MgCl2 , 5 mM DTT , 500 ng random hexamers , 10 U RNase Inhibitor , 1 µl Improm II Reverse Transcriptase ( Promega ) . The resulting cDNA was used for qPCR with BioRad SYBR green supermix according to the manufacturer's instructions . A two-step amplification protocol was used in either a BioRad MyIQ , or a BioRad CFX96 instrument with annealing at 60°C for 30 seconds and extension at 95°C for 30 seconds for 40 cycles . mRNA abundances were normalized to the abundance of Gapdh mRNA except for experiments using 4-sU where 7SL RNA was used as a reference . Primer sequences are listed in Table S4 . The pLightSwitch_3UTR vector was purchased from SwitchGear Genomics . The 3′UTR of Zfp36l2 was PCR amplified from C2C12 myoblast cDNA using the following oligos ( 5′-GCTAGCCTCTCCATCTCCGACGACTG-3′ and 5′-CTCGAGTTGGGGGAAACTACAAAAC-3′ ) . The resulting product was digested with Xho1 and Nhe1 and ligated into pLightSwitch_3UTR digested with the same enzymes to generate pLuc-36L2 . The PARN expression clone bears the open reading frame of human PARN which was amplified using primers PARN1F ( 5′-CATGTCGACATGGAGATAATCAGGAGCAATTTT-3′ and PARN1R ( 5′-CATGGTACCTTACCATGTGTCAGGAACTTCAA-3′ ) and cloned between the Xho1 and Kpn1 sites of pcDNA3 . 1 ( - ) ( Invitrogen ) . It is not targeted by the murine PARN shRNA . The PARN targeting shRNA vector was generated by cloning annealed and kinased oligonucleotides ( 5′-CCGGGCGTGTGTGTTATTAACTAATCTCGAGATTAGTTAATAACACACACGCTTTTTG-3′ and 5′-AATTCAAAAAGCGTGTGTGTTATTAACTAATCTCGAGATTAGTTAATAACACACACGC-3′ ) into the Age1 and EcoR1 sites of the pLKO . 1puro plasmid ( a gift from R . Schneider; [35] ) . Oligonucleotide sequences were chosen from the Broad Institute's RNAi Consortium database ( http://www . broadinstitute . org/rnai/trc ) . This particular shRNA targets the 3′UTR of murine PARN . The template used to generate the 5S rRNA probe was previously described [37] . In order to generate templates for probes against Zfp36l2 and Actb mRNAs , total RNA was isolated from proliferating C2C12 cells , and the poly ( A ) tails were removed by RNase H treatment in the presence of oligo ( dT ) 18 . An RNA linker ( Integrated DNA Technologies , Linker 3 ) was ligated to the 3′ ends of the RNAs using T4 RNA ligase treatment as described previously [38] . Ligated RNAs were subjected to reverse transcription using a specific primer complementary to the RNA linker ( for details see [38] ) . The resulting cDNA which corresponded to the 3′ ends of the Actb and Zfp36l2 mRNAs were then PCR amplified using the a primer complementary to the linker and an upstream oligo ( ActB PAT 5′-CACTCCTAAGAGGAGGATGGTCGCGTC-3′ for actin and Zfp PAT 5′-CAGTTGGAGCACCGCGTGTG-3′ for Zfp36l2 ) and ligated into the pGemT-Easy vector ( Promega ) . This process generated the pGemT-Zfp36l2 and pGemT-Actin plasmids which encode the 3′-terminal 300 nt of the Actb mRNA and the 3′-terminal 183 nt of the Zfp36l2 mRNA . Whole cell lysate was prepared by lysis of cells in RIPA buffer ( 50 mM Tris-HCl ( pH 7 . 4 ) , 150 mM NaCl , 1 . 0% deoxycholate , 1% Triton X-100 , 1 mM EDTA , and 0 . 1% SDS ) . 40 µg of each lysate was boiled in 6× protein loading buffer , resolved on a 10% SDS-polyacrylamide gel and blotted to PVDF membrane . PARN was detected using rabbit anti-sera ( 1∶20 , 000 ) [14] . ZFP36L2 was detected using rabbit polyclonal antibodies ( Genway GWB-C5FC76 ) . GAPDH ( Chemicon mAB374 ) or Tubulin ( Sigma Aldrich T5168 ) were used as loading controls ( 1∶20 , 000 ) . Results were visualized using a BioRad Chemidoc system and quantified using QuantityOne software ( BioRad ) . Reported values are a measure of the pixel density of the band of interest relative to the pixel density of the loading control ( GAPDH or Tubulin ) . These ratios were normalized relative to control samples . Reported uncertainties are standard deviations . 10 µg of total RNA was incubated with 2 µM DNA oligo ( ActB RNH 5′-AAGCAATGCTGTCACCTTCC-3′ for actin and Zfp RNH 5′-CGCGGTGCTCCAACTGTACCTA-3′ for Zfp36l2 ) , heated to 95°C for three minutes and slow cooled to 4°C over a period of 30 minutes . RNaseH ( 7 units ) and RNase Inhibitor ( 20 units ) were added in the supplied reaction buffer ( Fermentas Cat# EN0201 ) . For generating poly ( A ) tail minus ( A0 ) controls 100 ng/µl of oligo ( dT ) 18 was included . Reactions were incubated at 37°C for 30 minutes . RNAs were then resolved on a 5% denaturing polyacrylamide gel ( 7 M urea , 1× TBE ) , and electroblotted to nylon membrane ( Hybond-XL GE Healthcare ) at 700 mA for 30 minutes in 1× TBE . Nucleic acids were immobilized by UV-crosslinking ( Stratalinker ) . Membranes were pre-hybridized for 1 hour at 60°C in 25 ml hybridization buffer ( 50% formamide , 750 mM NaCl , 75 mM sodium citrate , 1% SDS , 0 . 1 mg/ml salmon sperm DNA , 1 mg/mL polyvinylpyrrolidone , 1 mg/mL ficoll , 1 mg/mL bovine serum albumin ( BSA ) ) . Membranes were then hybridized to radio-labeled RNA probe overnight at 60°C also in hybridization buffer . Blots were washed two times in 25 ml non-stringent wash buffer ( 0 . 1% SDS , 300 mM NaCl , 30 mM sodium citrate ) and two times in 25 ml stringent wash buffer ( 0 . 1% SDS , 30 mM NaCl , 3 mM sodium citrate ) for 20 minutes each time at 60°C . Membranes were exposed to storage phosphor screens and imaged on the Typhoon Trio Imager ( GE Healthcare ) . Results were analyzed using ImageQuant software ( GE Healthcare ) . α32P-labeled RNA probes were generated by in vitro transcription reactions as described below . Internally radio-labeled RNAs were generated by in vitro transcription reactions ( 20 U T7 or SP-6 RNA polymerase , 10 U RNase inhibitor , 40 mM Tris pH 7 . 9 , 6 mM MgCl2 , 10 mM DTT , 10 mM NaCl , 2 mM spermidine , 500 µM ATP , GTP , CTP , 50 µM UTP and [α-32P]-UTP ( 4 . 5 µCi/µl ) , 716 Ci/mmol ) were carried out for 3 hours at 37°C using 1 µg of linearized plasmid DNA as template . For the RNase H/northern blot probes , the pGemT-Zfp36l2 construct was linearized with SpeI and transcribed with T7 RNA polymerase . The pGemT-Actin construct was linearized with SacII and transcribed with SP6 RNA polymerase . Transcription products were separated on a 5% polyacrylamide gel containing 7 M urea , excised and eluted overnight in 400 mM NaCl , 50 mM Tris-Cl pH 7 . 5 , and 0 . 1% SDS at 22°C . RNA was precipitated and resuspended in H2O . C2C12 cells or PARN KD cells were transfected with a mixture of pEGFP-N1 ( Clontech ) and either Luc , or Luc36L2 plasmids . After 24 hours , the cells were trypsinized and collected in PBS . Coelenterazine ( Promega ) was added to a final concentration of 3 µM . Luciferase activities were measured in a Turner TD-20e Luminometer . Error bars represent pooled standard deviations derived from at least three independent experiments . Proliferating cultures of C2C12 myoblasts were treated with 4-thiouridine ( 200 µM; SIGMA ) for 15 minutes . Following this labeling period , cells were collected in TRIzol and RNA was isolated according to the manufacturer's recommendation . Biotinylation and fractionation of RNAs was performed as described previously [39] . Briefly , this involved incubating 50 µg of total RNA with 100 µg of Biotin-HPDP in 100 mM Tris-Cl ( pH 7 . 4 ) in the presence of 1 mM EDTA for 2 hours in the dark . An equal volume of chloroform and isoamyl alcohol ( 24∶1 ) was added to the biotinylation reaction and transferred to a Phase-Lock Gel tube ( 5 Prime ) , mixed by inversion , and centrifuged at full speed for 10 minutes at 4°C . RNA was precipitated in an equal volume of isopropanol in the presence of 0 . 5 M NaCl . The pellet was washed in 70% ethanol , resuspended in T . E . ( 10 mM Tris-Cl ( pH 7 . 4 ) 1 mM EDTA ) , warmed to 65°C for 10 minutes and snap chilled on ice . RNA was mixed with an equal volume of streptavidin magnetic beads for 15 minutes at room temperature and loaded on an equilibrated MACS® Separation Column ( Miltenyi Biotechnology ) . The beads were washed four times with 200 µl of wash buffer ( 100 mM Tris-Cl ( pH 7 . 4 ) 10 mM EDTA , 1 M NaCl , and 0 . 1% Tween-20 ) at 65°C and twice more at room temperature . Labeled RNA was eluted with 100 mM DTT . Eluted RNA was precipitated as described above , washed and resuspended in 20 µl of ddH2O . 40 ng of eluted RNA was reversed transcribed using random hexamers . The resulting cDNA was used in qPCR to assess levels of pre-mRNA using primer pairs in which the reverse primer was complementary to an exon and the forward primer matched a region within the upstream intron . The pre-mRNA abundance was normalized to that of 7SL RNA . C2C12 myoblasts ( 1 . 5×105 cells ) were seeded in 12-well dishes in growth media . Once cultures approached confluency , the monolayer was scratched with a 200 µl pipette tip . Cultures were washed with PBS , switched to low serum growth media ( 0 . 1% FBS ) , and imaged . After eight hours , cultures were imaged again . Margins of the scratch area were determined , and the number of cells migrating to the vacated area counted and graphed . Error bars represent the standard deviation from three experiments .
|
Almost all cellular mRNAs terminate in a 3′ poly ( A ) tail , the removal of which can induce both translational silencing and mRNA decay . Mammalian cells encode many poly ( A ) -specific exoribonucleases , but their individual roles are poorly understood . Here , we undertook an analysis of the role of PARN deadenylase in mouse myoblasts using global measurements of mRNA decay rates . Our results reveal that a discrete set of mRNAs exhibit altered mRNA decay as a result of PARN depletion and that stabilization is associated with increased poly ( A ) tail length and translation efficiency . We determined that stabilization of mRNAs does not generally result in their increased abundance , supporting the idea that mRNA decay is coupled to transcription . Importantly , knockdown of PARN has wide ranging effects on gene expression that specifically impact the extracellular matrix and cell migration .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"genome",
"expression",
"analysis",
"functional",
"genomics",
"gene",
"regulation",
"anatomy",
"and",
"physiology",
"rna",
"stability",
"muscle",
"molecular",
"genetics",
"musculoskeletal",
"system",
"cell",
"adhesion",
"extracellular",
"matrix",
"gene",
"expression",
"biology",
"molecular",
"biology",
"muscle",
"cells",
"microarrays",
"rna",
"systems",
"biology",
"cell",
"biology",
"nucleic",
"acids",
"physiology",
"cellular",
"types",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
The PARN Deadenylase Targets a Discrete Set of mRNAs for Decay and Regulates Cell Motility in Mouse Myoblasts
|
There is limited information on antivenom pharmacokinetics . This study aimed to investigate the pharmacokinetics of an Indian snake antivenom in humans with Russell’s viper bites . Patient data and serial blood samples were collected from patients with Russell’s viper ( Daboia russelii ) envenoming in Sri Lanka . All patients received Indian F ( ab’ ) 2 snake antivenom manufactured by VINS Bioproducts Ltd . Antivenom concentrations were measured with sandwich enzyme immunoassays . Timed antivenom concentrations were analysed using MONOLIXvs4 . 2 . One , two and three compartment models with zero order input and first order elimination kinetics were assessed . Models were parameterized with clearance ( CL ) , intercompartmental clearance ( Q ) , central compartment volume ( V ) and peripheral compartment volume ( VP ) . Between-subject-variability ( BSV ) on relative bioavailability ( F ) was included to account for dose variations . Covariates effects ( age , sex , weight , antivenom batch , pre-antivenom concentrations ) were explored by visual inspection and in model building . There were 75 patients , median age 57 years ( 40-70y ) and 64 ( 85% ) were male . 411 antivenom concentration data points were analysed . A two compartment model with zero order input , linear elimination kinetics and a combined error model best described the data . Inclusion of BSV on F and weight as a covariate on V improved the model . Inclusion of pre-antivenom concentrations or different batches on BSV of F did not . Final model parameter estimates were CL , 0 . 078 Lh-1 , V , 2 . 2L , Q , 0 . 178Lh-1 and VP , 8 . 33L . The median half-life of distribution was 4 . 6h ( 10-90%iles:2 . 6-7 . 1h ) and half-life of elimination , 140h ( 10th-90th percentilesx:95-223h ) . Indian F ( ab’ ) 2 snake antivenom displayed biexponential disposition pharmacokinetics , with a rapid distribution half-life and more prolonged elimination half-life .
Snake envenoming is a major health issue in South and South-eastern Asia [1] . Although antivenom is the most important treatment for snake envenoming , it can cause early systemic hypersensitivity reactions [2 , 3] , and there is limited evidence to support currently practiced dosing schedules . Dosing and assessment of the effectiveness of antivenom in human envenoming remains controversial and treatment protocols are not based on the kinetics of venom or antivenom . There are few studies of the pharmacokinetics of antivenom , and most of these are in animals [4] . Snake envenoming is a common problem in Sri Lanka and large amounts of antivenom are used throughout the country each year . A number of different Indian antivenoms are currently used and the initial dose ranges from 10 to 20 vials [5–7] . The initial dose is based on ED50 studies and clinical experience by titrating dose against the resolution of coagulopathy and neurotoxicity . However , the clinical effects of envenoming in these species are generally irreversible so determining if enough antivenom has been given and deciding to re-dose is often arbitrary and not based on whether all venom has been bound , or on the pharmacokinetics of antivenom . Measurement of venom and antivenom concentrations in patients with snake bite is required to improve effective initial and repeat dosing [8] . The pharmacokinetics of antivenom are expected to be similar to other intravenous drugs being delivered to the central compartment with zero order input kinetics ( constant rate of infusion ) . Antivenom is then distributed throughout the body and is eliminated by the kidneys and/or the reticuloendothelial system [4] . Decreasing antivenom concentrations in the central compartment are therefore due to both distribution and elimination . Different types of antivenom have different pharmacokinetics due to the difference in their molecular masses [4] . Fab antivenoms have much larger volumes of distribution ( VD ) than F ( ab’ ) 2 or whole IgG [5 , 9] . Most studies of antivenom pharmacokinetics show a biphasic ( two-compartment ) decline after intravenous administration of whole IgG and F ( ab’ ) 2 antivenoms , as a result of an initial rapid decline ( distribution phase ) and a slower decline ( terminal elimination phase ) [4 , 9] . Most studies of the pharmacokinetics of antivenom are in animals [4 , 10] , and the pharmacokinetics appear to differ between animals making animal models problematic for defining the pharmacokinetics of antivenom in humans [10] . Although there have been several publications of antivenom concentrations in snake envenoming , there are only a few studies of the pharmacokinetics of antivenom in human snake envenoming [4 , 5 , 9 , 11–14] . These studies were all in small numbers of patients using a classic two phase approach , without including input processes ( i . e . delivery of the antivenom , usually via an infusion to the central compartment as a zero order process ) and providing limited information on the pharmacokinetics and variation between patients . A population approach to pharmacokinetic analysis is increasingly being used to define the pharmacokinetics of drugs in humans because it provides information about population variability and the need for individualisation of drug treatment . The traditional approach to pharmacokinetic analysis ( two stage analysis ) estimates the pharmacokinetic parameters for each individual patient and then provides summary statistics which only give a population average and standard error . In contrast the population approach estimates the typical value of each parameter for the population and the variability of the parameters simultaneously . This provides an estimate of unexplained random variation and allows the effects of covariates to be accounted for in the model ( e . g . weight , renal function ) . There are no previously published population pharmacokinetic analyses of antivenom in humans or animals . The aim of this study was to investigate the pharmacokinetics of antivenom in patients with snake bites using a population based analysis , including an investigation of the covariates that may influence the pharmacokinetics of antivenom .
The study was approved by the Ethical Review Committee , Faculty of Medicine , University of Peradeniya , Sri Lanka . All patients gave written and informed consent for the collection of clinical data and blood samples . All patients ( >15 years old ) from October 2010 to March 2012 with a suspected snake bite who presented to the Base Hospital Polonnaruwa were recruited to a prospective cohort study . Those with coagulopathy were then entered in a dose finding randomised clinical trial of fresh frozen plasma . The entry criteria for the trial was a suspected Russell’s viper ( Daboia russelii ) bite with coagulopathy defined as an abnormal 20 minute whole blood clotting test ( 20WBCT ) . This resulted in a small number of patients being recruited where Russell’s viper ( D . russelii ) venom was not detected and on further testing , hump-nosed viper ( Hypnale spp . ) venom was detected ( in some Hypnale bites the 20WBCT and coagulation studies may be abnormal [15 , 16] ) . In this pharmacokinetic study , patients were only recruited from the clinical trial and were included if they had serial serum collection for antivenom measurement and complete demographic details ( including weight ) . All patients received the Indian polyvalent snake antivenom intravenously manufactured by VINS Bioproducts Limited ( batch numbers: 1060 [MFD 2008] , 1096 [MFD 2009] , 1102 [MFD 2009] , 01015/10-11 [MFD 2010] , 01AS11112 [MFD 2011] ) . For a dose of antivenom , each of 10 vials of antivenom are reconstituted in 10ml of normal saline for a total of 100ml of antivenom . From a 500ml bag of normal saline 100ml volume is removed and replaced by the 100ml of antivenom so the 10 vials are administered in a total of 500ml of normal saline . This is given over 1 hour . The following data were collected prospectively in all cases: demographics ( age , sex and weight ) , time of the snake bite , clinical effects ( local envenoming , coagulopathy , bleeding and neurotoxicity ) and antivenom treatment ( dose , time of administration and antivenom batch number ) . Blood samples were collected for research on admission and regularly throughout each patient admission . Blood was collected in serum tubes for venom-specific enzyme immunoassay ( EIA ) and antivenom EIA . All blood samples were immediately centrifuged , and then the serum aliquoted and frozen initially at -20°C , and then transferred to -80°C within 2 weeks of collection . A sandwich enzyme immunoassay was used to measure antivenom in serum samples as previously described [8 , 17] . The plate was first coated with Russell’s viper venom and then stored and blocked overnight . Serum was then added to the plates . The detecting antibodies were conjugated with horseradish peroxidase . Russell’s viper ( D . russelii ) and hump-nosed ( Hypnale spp . ) viper venoms were measured in samples with a venom specific enzyme immunoassay as previously described [6 , 8 , 17] . Briefly , polyclonal IgG antibodies were raised in rabbits against Russell’s viper ( D . russelii ) and hump-nosed viper ( Hypnale spp . ) venom . The antibodies were then bound to microplates and also conjugated to biotin for a sandwich enzyme immunoassay using streptavidin-horseradish peroxidase as the detecting agent . All samples were measured in triplicate , and the averaged absorbance converted to a concentration using a standard curve made up with serial dilutions of antivenom and using a sigmoidal curve . The limit of quantification for the antivenom enzyme immunoassay assay was 40μg/ml and for the venom enzyme immunoassay was 2ng/mL for Russell’s viper and 0 . 2ng/ml for hump-nosed viper . Patient data was analysed using MONOLIX version 4 . 2 ( Lixoft , Orsay , France . www . lixoft . com ) . MONOLIX uses the Stochastic Approximation Expectation Maximization algorithm ( SAEM ) and a Markov chain Monte-Carlo ( MCMC ) procedure for computing the maximum likelihood estimates of the population means and between-subject variances for all parameters [18] . One , two and three compartment models with zero order input and first order elimination kinetics were assessed and compared to determine the best structural model . Proportional and combined models were evaluated for the residual unexplained variability . Method M3 was used to deal with antivenom concentrations below the limit of quantification ( BLQ ) [19] . Between-subject variability ( BSV ) was included in the model and assumed to have log-normal distribution . Models were parameterized in terms of volume of distribution ( VD; V , VP , VP2 ) , clearance ( CL ) , inter-compartmental clearance ( Q; Q1 , Q2 ) and relative bioavailability ( F ) for either 1- , 2- or 3-compartment models . Initial estimates of parameters were taken from a previous pharmacokinetic study of anti-venom [9] . Uncertainty in antivenom dose was included in the model by allowing BSV on F to account for batch to batch variation in antivenom ( five different batches ) and for variation within batches . F was fixed to 1 and the BSV was estimated for each patient similar to including uncertainty on dose as previously described [18] . The BSV on F was plotted for each batch to determine if there was a difference between batches . The effect of covariates , including age , sex , weight , and pre-antivenom concentrations in patients with detectable venom , were explored by visual inspection of the individual parameter estimates versus the covariate of interest . Age , sex and pre-antivenom concentrations were not included in the final model evaluation due to the absence of an association visually . The influence of weight ( wt ) on volume was included in the modelling process . Weight was assumed to be related to V by a power function . The covariate was centred to the average weight . Thus in the model the estimation of the effect of weight on volume is: V = θV x ( wt/wtav ) ∧fwt Where θV is the typical value of volume of distribution , wt is the individual patient weight , wtav is the average weight and fwt accounts for the influence of wt on volume . Model selection decisions were based on a decrease in the objective function ( OFV ) , a decrease in residual error , clinical relevance of the pharmacokinetic parameters and goodness of fit plots . The log likelihood was computed for each model and used to discriminate through the difference in log likelihood ( −2LL ) . A p-value of 0 . 05 was considered statistically significant , equivalent to a drop in OFV by 3 . 84 . From the final model we simulated 1000 patients using the individual predicted patient parameters from the final model with MatLab to explore different initial doses and repeat doses . The following scenarios were explored: One dose ( 10 vials ) of antivenom given with infusions rates of 20 minutes , 1 hour and 2 hours . Two doses of antivenom given , each over 1 hour and 6 hours apart . Two doses of antivenom given , each over 1 hour and 12 hours apart . The median antivenom concentration versus time was plotted with 10% and 90% percentiles .
There were 75 patients with a median age of 38 years ( 16 to 64y ) and 64 were male . Seventy one were Russell’s viper envenoming cases and 52 of these had detectable venom prior to the administration of antivenom . Four patients had hump-nosed viper envenoming ( confirmed by detectable hump-nosed viper venom ) . In all four patients with hump-nosed viper envenoming there was a steady decline of venom concentrations despite the administration of antivenom consistent with the antivenom not being raised against this snake venom . In nineteen patients meeting the inclusion criteria venom was not detected prior to antivenom , most likely because the blood was collected prior to envenoming . The demographics of the patients are listed in Table 1 . There were 510 antivenom concentration data points but only 411 had detectable antivenom , the other 99 were serial samples after the disappearance of antivenom . There were 54 patients who had a single dose of antivenom who had 265 antivenom concentration measurements with a median of five antivenom concentrations in each patient ( Range: 2 to 10 ) , and a median antivenom concentration of 1607μg/ml ( Range: 40 to 13673μg/ml ) . There were 21 patients who had multiple doses of antivenom who had 146 antivenom concentrations with a median of seven antivenom concentrations in each patient ( Range: 3 to 11 ) and a median antivenom concentration of 2293μg/ml ( Range: 40 to 12599μg/ml ) . The observed concentration versus time data is shown in Fig 1 . A two compartment model with zero order absorption and linear elimination kinetics and a combined error model best described the data . The final model incorporated BSV on F , which was fixed to 1 to allow variability between patients in dose . The model also incorporated weight as a covariate with a power effect on central volume , V . The inclusion of pre-antivenom concentrations on BSV of F did not improve the model . Plots of the BSV on F versus the batch number showed no relationship between the batch and BSV on F ( S1 Fig ) . The final model parameter estimates were CL , 0 . 078 Lh-1 , V , 2 . 2L , Q , 0 . 178Lh-1 and VP , 8 . 33L . The median half-life of distribution was 4 . 6h ( 10th-90th percentiles: 2 . 6 to 7 . 1h ) and the half-life of elimination , 140h ( 10th-90th percentiles: 95 to 223h ) . There was no difference in the parameter estimates between those with Russell’s viper envenoming with detectable venom prior to antivenom ( 52 ) , those with Russell’s viper envenoming and no detectable venom prior to antivenom ( 19 ) and those with hump-nosed viper envenoming ( S2 Fig ) . S3 and S4 Figs shows the goodness-of-fit plots for the final model . The individual PK parameter estimates from the base models with modelling decisions and final model parameters are described in Table 2 . There was also no difference in parameter estimates between patients given 1 dose of antivenom and those given 2 doses , or between patients with different initial venom concentrations ( S5 Fig ) . Simulations for one dose ( 10 vials ) of antivenom given over 20 minutes , 1 hour and 2 hours shows there is a slightly lower and later peak antivenom concentration with slower infusions ( Fig 2 ) . Simulations for two doses of antivenom shows that antivenom concentrations decrease rapidly after each dose and there are low but persistent levels of antivenom after one dose and both two doses regimens ( Fig 3 ) .
The study adds to the limited information available on the pharmacokinetics of antivenom in humans supporting previous studies [4 , 9] . Indian F ( ab’ ) 2 snake antivenom displayed biexponential disposition pharmacokinetics , with a rapid half-life of distribution and a much longer half-life of elimination . Weight accounted for some of the variability in the central volume , and the volumes of the central and peripheral compartment were consistent with a large molecule which does not have a large volume of distribution . Including variability on F improved the model showing that there was significant random variability in dose . The plots in Figs 2 and 3 show the expected antivenom concentration profiles in the first 24 hours after administration . Previous human and most animal studies have also shown a biexponential decay in antivenom concentrations [4 , 9 , 20 , 21] , with similar values for the distribution half-life of 2 to 4 hours and much longer elimination half-life of 90 to 230h . Previous studies have been small with 10 or less patients in each analysis ( for different antivenoms ) and a classic two phase approach has been undertaken . Such an approach will over-estimate the error and not account for true random variability or covariate effects . In this study we have undertaken a population approach , which provides information on the variability of the pharmacokinetics in the population and an improved model by including weight and variability in dose . Previous studies have not shown why they chose particular models ( 2-compartment versus 3-compartment ) , with no statistical criteria or goodness of fit plots . Some previous animal models and one human study have described the pharmacokinetics with a tri-exponential decay in animals [14 , 22 , 23] . These analyses have not included an input process in the analysis which will bias the estimation of the disposition parameters , particularly with three or more compartments when the initial very short half-life is similar to the time of the input phase . Ismail et al . estimated the initial rapid half-life in animals to be 0 . 2h and Vazquez et al estimated it to be 0 . 25h , which are both similar to the usual infusion rate of antivenom over 10 to 30 minutes . It is possible that there is only 2-compartmental disposition kinetics in these studies , and future pharmacokinetic analyses need to include an input phase in the model . A possible limitation of our study was that there may have been insufficient sampling in the initial period after antivenom administration to detect a third compartment . In contrast to this , Vazquez et al were likely to have taken samples in the input phase , since the first sample was taken 5min after antivenom administration , although they do not report the infusion time or rate [14] . One animal study of a F ( ab’ ) 2 has shown that the pharmacokinetics of antivenom are the same in envenomed and non-envenomed rabbits [24] . This is consistent with this study demonstrating that pre-antivenom venom concentrations did not influence the pharmacokinetics of antivenom , including different initial venom concentrations ( S5 Fig ) . However , this may be different for Fab antivenoms where high molecular weight toxins may change the route of elimination from renal ( for free Fab antivenom ) to phagocytosis/reticulo-endothelial system for Fab-toxin molecules . The latter has been shown in rabbits with anti-Vipera Fab antivenom [25] . There has always been concerns about the variability between different batches of antivenom leading to potential differences in the dose administered between batches . The study did not support this concern and found that there was no difference in F on average between different batches ( S1 Fig ) . However , the study found that including between subject variability on relative bioavailability did improve the model . This suggests there was random variability in the dose administered which is likely to be due to variable losses occurring during reconstitution of the individual freeze dried vials of antivenom . So , although there may be variability between batches , the variability in dosing errors appears to be larger than the differences between batches . There are a number of limitations to the study including the fact that the sample collection was not optimally designed and sample times ( windows ) were based on timing of clinical samples and other research assays required for the clinical trial . This is unlikely to have a major influence on the analysis because a population approach will allow for both sparse and rich sampling in patients . Another issue is that antivenom is not a pure substance and consists of varying amounts of polyclonal antibodies to multiple toxins in the venom with varying affinities . However , the assay uses a single detecting antibody ( anti-horse antibody ) , so will detect all antibodies against the snake toxins irrespective of their toxin target or affinity . Finally , the assay will only detect antibodies that bind to the snake toxins . In most antivenoms , specific antibodies to snake toxins make up only 10 to 20% of the total protein/immunoglobulin content . This is unlikely to have affected the pharmacokinetic analysis because only immunoglobulins binding to snake toxins are relevant to the analysis . This population pharmacokinetic analysis demonstrates that Indian F ( ab’ ) 2 antivenom has biexponential disposition kinetics and following an initial decline in antivenom concentrations in the first 12 hours , low concentrations are present for days after administration . The study demonstrates that the antivenom concentrations were not affected by the initial venom concentrations suggesting that sufficient antivenom in excess of the venom was being administered . Understanding the pharmacokinetics of antivenom may assist in improving antivenom dosing by matching antivenom pharmacokinetics to the neutralisation of venom ( pharmacodynamics ) , as well as clinical effects .
|
Snake envenoming is a neglected tropical disease that affects hundreds of thousands of people in the rural tropics . Antivenom is the main treatment for snake bites but there is limited information on the pharmacokinetics and appropriate dosing regimen . Most studies have been done in animals and dosing guidelines are based on arbitrary and often irreversible clinical signs . In this study we measured serial antivenom concentrations in patients with Russell’s viper envenoming given antivenom . Using this data we modelled the pharmacokinetics of antivenom in the population and showed that antivenom concentrations had a bi-exponential decay with an initial decrease over 12 hours and then a slow decrease over days . There was significant variability in the dose given which was not affected by the particular antivenom batch given . The presence of venom did not appear to modify the pharmacokinetics of antivenom . Understanding the time course of antivenom in patients with snake envenoming will provide a better basis for antivenom dosing .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Population Pharmacokinetics of an Indian F(ab')2 Snake Antivenom in Patients with Russell's Viper (Daboia russelii) Bites
|
Human movement is likely an important risk factor for environmentally-transmitted pathogens . While epidemiologic studies have traditionally focused on household risk factors , individual movement data could provide critical additional information about risk of exposure to such pathogens . We conducted global positioning system ( GPS ) tracking of urban slum residents to quantify their fine-scale movement patterns and evaluate their exposures to environmental sources of leptospirosis transmission . We recruited participants from an ongoing cohort study in an urban slum in Brazil and tracked them for 24 hours at 30-second intervals . Among 172 subjects asked to participate in this cross-sectional study , 130 agreed to participate and 109 had good quality data and were included in analyses . The majority of recorded locations were near participant residences ( 87 . 7% within 50 meters of the house ) , regardless of age or gender . Similarly , exposure to environmental sources of leptospirosis transmission did not vary by age or gender . However , males , who have higher infection rates , visited a significantly larger area during the 24-hour period than did females ( 34 , 549m2 versus 22 , 733m2 , p = 0 . 005 ) . Four male participants had serologic evidence of Leptospira infection during the study period . These individuals had significantly larger activity spaces than uninfected males ( 61 , 310m2 vs 31 , 575m2 , p = 0 . 006 ) and elevated exposure to rodent activity ( p = 0 . 046 ) and trash deposits ( p = 0 . 031 ) . GPS tracking was an effective tool for quantifying individual mobility in the complex urban slum environment and identifying risk exposures associated with that movement . This study suggests that in addition to source reduction , barrier interventions that reduce contact with transmission sources as slum residents move within their communities may be a useful prevention strategy for leptospirosis .
Household characteristics are frequently used as a measure of the risk of pathogen exposure . However , these household risk factors are static proxies for exposure , which is a dynamic function of movement through a contaminated environment . Many neglected tropical diseases—including schistosomiasis , helminthiasis , and leptospirosis , among others—result from contact with environmental pathogens . Individual movement data can provide a more complete picture of exposure to such pathogens than can household-based proxies , and so add value to epidemiological studies [1] . The availability of ever cheaper , smaller , and more accurate location-aware devices has made it feasible to quantify movement [2] . Mobile phone data are increasingly used to study large-scale population movements [3 , 4] and GPS data to study fine-scale individual activities [5–9] . Tracking technology has been applied to diseases such as malaria [4 , 10] , dengue fever [2 , 11 , 12] , schistosomiasis [13 , 14] , and non-communicable disease [9 , 15] , and has yielded valuable insights into exposure and transmission . But despite these examples of effective use of movement data , knowledge of human movement patterns and their influence on exposure remains limited . The study of environmentally transmitted disease will benefit from an improved understanding of human movement . For some diseases , pathogen sources are generally known and movement can help understand interaction with them . For others , the specific source causing human infection is unknown . Risk factor associations between infection and household characteristics can indicate that exposure occurs in the peridomestic environment . However , an important question is what mechanism these associations capture: do they indicate that the pathogen is present in the household environment or that the individual comes into contact with the pathogen at a source near the home ? Specific information on how human-pathogen contacts occur is important for planning public health interventions , and movement may provide this information . Leptospirosis is an environmentally transmitted disease [16] for which movement may play a critical role in understanding exposure risk . This spirochetal zoonosis is transmitted to humans when cut or abraded skin comes into contact with water or soil contaminated with the bacteria , which are released into the environment via the urine of an infected mammalian host [17] . While many cases are asymptomatic or mild , leptospirosis causes approximately one million hospitalized cases and nearly 60 , 000 deaths per year worldwide [18] . Severe cases can manifest as Weil’s disease with jaundice and renal failure or a pulmonary hemorrhagic syndrome [19] , which have case fatality rates of >10% or >50% respectively [20] . Leptospirosis occurs worldwide , but its burden is highest in subsistence farming and urban slum populations in the developing world [18 , 20] . Individual movement data may be particularly informative for studies of leptospirosis in the urban slum environment . Urban slums are home to a large and growing population , expected to total two billion by 2030 [21] . Leptospirosis has emerged as a major public health problem in slums worldwide because their poor sanitation and housing infrastructure promote rat-borne transmission of the pathogen [22–26] . Detailed longitudinal studies of urban slum leptospirosis in Salvador , Brazil and other locations have revealed that it manifests as seasonal , rainfall-associated epidemics [27–29] . Infection rates for males are nearly twice those for females in an urban slum in Salvador ( 48 . 3 vs 25 . 9 infections per 1000 follow-up events [30] , see also [27 , 31–33] ) , consistent with findings from other urban locations [34–36] . Risk factors for infection are concentrated in the peridomestic environment , where leptospirosis disproportionately affects individuals living in rat-infested households in close proximity to transmission sources such as sewers , open trash deposits , mud , and floodwaters [24 , 32 , 37 , 38] . However , the risk of pathogen transmission varies widely over small distances [30] and there is substantial unexplained variation in risk , even within households , where only some residents are infected . Urban slums are complex , compact environments that can vary widely at small spatial scales and individual movement through this heterogeneous environment may play a key role in generating the observed risk heterogeneity . There may also be differences in movement between groups , such as males and females , with different infection rates . Movement could also help answer the question of whether leptospires are present in the peridomestic environment or whether slum residents contact the pathogen at sources near the home . We used GPS tracking to quantify the movement patterns of urban slum residents and their resulting exposure to environmental sources of leptospires . Because urban slums are highly heterogeneous environments , movement studies must be conducted at fine spatial scales . We densely sampled participant movement , recording their location every 30 seconds for 24 hours , to obtain high spatial resolution data . We recruited participants from four demographic groups: males aged 15–34 years , females aged 15–34 years , males aged ≥35 years , and females aged ≥35 years . This recruitment scheme allowed us to describe age- and gender-specific patterns and evaluate the hypothesis that males’ increased risk of leptospirosis is due to their movement patterns and resulting exposure to environmental features associated with leptospirosis infection .
Pau da Lima is an urban slum in Salvador , the third largest city in Brazil . This slum , at the periphery of the city , occupies the floor and sides of several connected valleys . Sewers—most of which are open—flow through the valley floors , and drainage tributaries run down the sides of the valleys . Housing ranges from formal concrete block structures to dwellings constructed of found materials . Informal employment is frequent , but regular employment is available at a nearby commercial area and throughout the city , to which the slum is connected via bus . A 2012 census of the site identified 12 , 651 residents who were mostly squatters ( 88% ) , did not complete primary school ( 66% ) and subsisted on a median per capita daily income of US$2 . 60 . This site is a high-transmission setting for leptospirosis: outpatient surveillance identified a mean annual incidence for clinical disease of 143 cases per 100 , 000 population since 2010 , and hospital-based surveillance initiated in 2001 identified 19 . 8 hospitalized cases per 100 , 000 population per year . Participants in this GPS study are drawn from a cohort of Pau da Lima residents being followed prospectively to determine the seroincidence of leptospirosis . This cohort study , initiated in 2013 , enrolls residents who sleep at least three nights a week in Pau da Lima , are at least 5 years old , and consent to participate . Researchers visit participants every six months to take blood for serological analysis and conduct interviews about demographic , socioeconomic , and behavioral features . Household exposures are determined through these interviews and geographic information systems ( GIS ) surveys . Samples are evaluated using the microscopic agglutination test , the gold standard serologic diagnostic test for leptospirosis [17] . Infections are defined by a seroconversion ( from a titer of <1:50 to ≥1:50 ) or a four-fold rise in titer between sequential samples . The samples bracketing the GPS study were taken in March and September 2014 . This study , conducted from June-September 2014 , focused on a high-risk sub-cohort living at low elevation in a single valley . Residents at low elevation in this site have low socioeconomic status and live near open sewers and flood-prone areas , all features associated with leptospirosis infection . GPS study participants were drawn from cohort participants at least 15 years old due to the low incidence of leptospirosis in children . We recruited participants from four groups: young males ( aged 15–34 years ) , young females ( 15–34 years ) , older males ( aged ≥35 years ) , and older females ( aged ≥35 years ) . The study was approved by Institutional Review Boards at Yale University and in Brazil . All participants provided written informed consent . Consent was also obtained from parents or guardians of participants who were minors . We considered three commercially available GPS models ( the Mobile Action Technology iGot-U 120 and 600 , and the SleuthGear iTrail ) and chose the Mobile Action Technology iGot-U 120 for its balance of cost and features . This model or its predecessor the iGot-U 100 was used in previous studies [7 , 8 , 14] . The mean battery life of our units at a 30-second interval was 46 . 7 hours , insufficient for 48-hour recording , leading us to focus our sampling effort on 24-hour periods . We conducted accuracy tests using three randomly selected units . These were turned on , given five minutes to establish satellite connection , then placed together on a flat surface and left to collect data every 30 seconds for 10 minutes . We repeated this procedure in locations around the study neighborhood and city and under varying meteorological conditions to produce an accuracy measure reflective of the range of conditions experienced by participants . To the GPS points taken under each testing condition , we fitted a bivariate Normal distribution with zero mean , independent components and a pooled standard deviation . We used the mean pooled standard deviation across testing conditions , 5 . 67 meters , hereafter referred to as SDACC , in our analyses . Movement is sensitive information , and we took steps to ensure participants were comfortable with our GPS study . We benefitted from the extensive work conducted by Paz-Soldan et al . [6] in Iquitos , Peru , where they implemented a GPS study to study the role of movement in dengue virus infection . We created an information sheet modeled after theirs that explained why we were studying movement , how this would help us understand leptospirosis , and what participation entailed . It also answered common questions including what the GPS could and could not record , that there are no known health risks associated with the technology , and what would happen if the unit was lost , stolen , or broken ( participants had no responsibility for these incidents ) . Sheets contained both written and visual versions of the information due to the high illiteracy rate in the study site . We used this information sheet during recruitment and gave participants a copy so they could explain the study to others . We conducted several pilot studies of the GPS enrollment and wearing procedure , updating the protocol and information sheet to address participant questions . We recruited participants through door-to-door sampling ( the method used to enroll participants in the source cohort study ) . To avoid biasing our sample by employment status , we recruited throughout the day and on all days of the week . We recruited one individual per household at a time to avoid accidental switching of GPS units . Each recruitment week , we took 20 GPS units into the field with the goal of recruiting five people from each of the four demographic groups . We evaluated recruitment numbers weekly , and if any groups were under-represented , we preferentially recruited members of that group when encountered . When we met an eligible individual , we gave a verbal introduction to the project complemented by the written and illustrated information sheet . We allowed them to examine the GPS unit and ask questions about the study and technology , then asked if they were willing to participate . If not , we recorded the reason ( s ) for refusal . If they consented , we completed a consent form and entry interview to collect demographic data , then gave them the GPS unit and instructions . Sixty-one participants were also enrolled in an activity diary study . The diary , conducted retrospectively when we returned to pick up the GPS unit , asked about the participant’s activities , behaviors , and exposures each hour of their 24-hour tracking period . GPS units were delivered to participants after being programmed to turn on one hour before the desired start time of data collection ( to ensure that the unit was on and connected to satellites by that time ) and off at the end of collection . Units were set to record the GPS location every 30 seconds . Settings were not modifiable by participants . We instructed participants to wear the units around the neck on the attached ribbon , a method found to minimize clothing-related interference and maximize acceptability to study participants [2] . Participants were told when to start and stop wearing the GPS unit . We arranged a time to come back to collect it and asked that participants leave it with someone in their house if they could not be there at the set time . At the end of the assigned 24-hour period , we returned to the house , retrieved the GPS unit , and conducted an exit interview to gather data on compliance with study protocols and answer questions that arose during participation . If the participant was enrolled in the activity diary study , we also completed the diary at this time . We downloaded the GPS data using the unit’s software ( @TripPC , http://www . a-trip . com ) to a secure server ( REDCap ) and erased the unit’s memory . Data were formatted and analyzed in R version 3 . 2 [39] and visualized in QGIS version 2 [40] . We used exit interviews to categorize participants as fully , partially , or non-adherent to the study protocol . We then evaluated the length of data collection ( affected when units either suffered early battery exhaustion or started recording data after the programmed start time ) and density of data collection ( a function of the actual inter-point interval , which was programmed to be 30 seconds ) . We tested whether the amount of missing data varied by age , gender , household elevation and per-capita income , GPS unit , and where the unit was worn ( the neck , as instructed , or other locations as reported by participants during their exit interview ) . We trimmed data to the desired start and end times , and excluded self-reported time when the participant did not wear the unit . Each GPS point included the time , date , latitude , longitude , and distance from previous point . We calculated the time interval and velocity ( distance divided by interval ) between points . We also implemented a one meter per second velocity filter , excluding points with velocities above this threshold . This filter removes points with obviously incorrect recorded locations while discarding minimal data [13] . It serves an additional purpose in our study by excluding points when participants were moving too fast to be traveling by foot . Because leptospirosis is transmitted by contact with contaminated soil or water , we did not consider time in motorized transport to be at-risk . We first measured the concentration of participant activity near the house and within the neighborhood . We retrieved the GPS coordinates of participant households from our cohort study data and calculated the distance of each GPS point from the wearer’s house . We defined the neighborhood as the valley in which the study was conducted , and used a geographic information system shapefile of that valley’s boundary to determine whether each GPS point was in or outside the neighborhood . We then calculated the area visited in 24 hours ( activity space ) using the Daily Path Area ( DPA ) method from Zenk et al . [9] . The activity space is quantified by buffering an individual’s GPS points then calculating the total area contained within the buffer . This buffering step can account for positional inaccuracy as well as movement between data points . This is important for diseases like leptospirosis where exposure can occur anywhere an individual contacts the environment , not just at their GPS points . The distance between the recorded GPS point and the wearer’s actual location is a function of variation along both the north-south and east-west axes , and we modeled this using a bivariate Normal distribution . This induces a Rayleigh distribution for the distance between recorded and actual locations , which requires a radius of 2 . 45 times the standard deviation to capture with 95% probability the actual location of the GPS point . We calculated the activity space using two different buffer radii to evaluate the sensitivity of the measure . We first used a radius of 2 . 45*SDACC that accounts largely for positional uncertainty of the GPS point . We then tested a 2 . 45 SDACC + 20 meter buffer to incorporate both positional uncertainty and space potentially visited while the participant was walking in the interval between points . We next measured exposure to leptospirosis transmission sources resulting from participant movement . Previous studies in our site and others have identified a number of household environmental features associated with leptospirosis infection [30–33 , 37 , 38] . In this study we evaluated exposure to open sewers and trash deposits , low elevation ( a proxy measure for flood risk ) , land cover ( vegetated , exposed soil , or impervious surface ) , and rodent activity . Open sewers and public trash deposits were delineated through environmental surveys . Elevation was extracted from a one-meter elevation contour of the site . We used remote sensing to classify each two-meter pixel on a high-resolution satellite image of our site as vegetation , exposed soil , or impervious surface [41] . Tracking plates , which have been shown in this site to be a sensitive measure for quantifying rodent activity [42] , were used to generate a fine-scale map of predicted rodent activity [41] . Several of these exposure measures require intensive field surveys and so were only available within the study site , not the entire city . Exposure analyses were thus restricted to movement within the site . To quantify exposure , we generated a grid of points with 2 . 5 meter spacing that covered the study neighborhood . We then assigned each grid-point a value for each exposure . These values were defined as: the reciprocal of the squared distance to sewer or trash , the closest 1-meter elevation contour , binary values for presence of each land use class at the grid point , and predicted rodent activity at the grid point . We then overlaid each GPS point on the grid and used a Gaussian kernel smoother with formula–exp[ ( distance between grid point and GPS point ) 2 / ( 2* SDACC2 ) ] to calculate that GPS point’s weighted mean for each exposure based on nearby grid-points . We calculated the mean exposure to each environmental feature across a participant’s full set of GPS points . Because of this spatially smoothed estimation , exposure values are best interpreted as a relative measure , not a directly interpretable quantity .
The study area contained 402 eligible individuals , of whom we contacted 172 during the study period and enrolled 130 ( 75 . 6% , Table 1 ) . Contact rates were slightly higher for females but enrollment rates are similar across age and gender . Participants are generally similar to the total slum population except that they are significantly more likely to live within 10 meters of an open sewer ( p < 0 . 001 ) . This difference is expected due to our focus on high-risk households along the valley bottom . The most common reason for refusal to participate was being too busy . Of 130 participants , 100 ( 76 . 9% ) wore the GPS unit for the entire assigned time , 11 ( 8 . 5% ) complied partially with instructions and could report times when they were non-adherent , and 19 ( 14 . 6% ) were considered entirely non-adherent . Forgetting to wear the unit was the main reason for both full and partial non-adherence . For two fully adherent participants , no data were recorded by the GPS unit . Analyses were thus conducted on 109 participants ( 98 fully and 11 partially adherent participants whose data was censored during their self-reported non-adherent time ) . The length of data collection was generally high , but the density was lower than expected . Early battery exhaustion or late recording start were both relatively rare , and 87 participants ( 79 . 8% ) had GPS points spanning more than 90% of the assigned 24 hours . The median interval between GPS points was 35 seconds , with 90 . 8% under 1 minute and 99 . 6% under 5 minutes . The GPS records of eleven individuals ( 10 . 1% ) included at least 75% of the expected 2880 points , and 65 ( 59 . 6% ) included at least 50% . Females had more missing data than males ( p = 0 . 041 ) . The amount of missing data did not vary by age of the wearer ( p = 0 . 742 ) , GPS unit ( p = 0 . 198 ) , where individuals wore the GPS unit ( neck versus other location , p = 0 . 602 ) , or per-capita income ( p = 0 . 840 ) or elevation ( p = 0 . 191 ) of the wearer’s household . Certain movement characteristics are similar across age and gender , but others differ by demographic group . Regardless of age or gender , participants spent most of their time near their residence . A median of 94 . 9% of points occurred within the study site and 87 . 7% within 50 meters of the home ( Fig 1 ) . In contrast , activity space , the entire area encompassed by an individual’s movement , does vary significantly by age and gender . Males visit a much larger space on a daily basis than females ( mean 34 , 549m2 versus 22 , 733m2 , p = 0 . 005 ) , as do older participants compared to younger ( mean 31 , 739m2 versus 23 , 111m2 , p = 0 . 033 ) . Results are reported for the 2 . 45 SDACC radius , but are qualitatively similar for the 2 . 45 SDACC + 20 meter radius . While most time is spent in the peridomestic environment , males and older adults move more within that space . In addition , behavioral data recorded in activity diaries for 61 ( 56 . 0% ) of the GPS participants showed that males reported spending nearly three more hours per day outdoors than females ( mean 4 . 95 versus 2 . 28 hours , p = 0 . 002 ) . This is consistent with males visiting a larger area in a 24-hour period . We also estimated movement-induced exposure to environmental features associated with leptospirosis infection . Interestingly , we did not identify any features for which exposure varied by age or gender ( Table 2 ) , despite variation in activity space . Though males move through a larger area and spend more time outside , their exposures are similar to those of other groups . Four participants had serologic evidence of leptospirosis infection during the 6-month interval containing the GPS study . This infection prevalence of 3 . 7% ( 4/109 ) is similar to the 2 . 9% ( 47/1600 ) in the full cohort during the study period ( p = 0 . 563 ) . All infected individuals were male , with two in each age group ( 15–34 and ≥35 years ) . Infected participants are generally similar to all male participants except on ethnicity—infected individuals all self-identified as Black , compared to 47 . 5% of all male participants ( p = 0 . 035 ) . Both the movement and exposure of infected individuals showed unique features . Because males have larger activity spaces than females , we compared the activity spaces of infected males ( n = 4 ) to those of uninfected males ( n = 36 ) . Infected individuals had three of the ten largest activity spaces ( Fig 2 ) and a significantly larger mean activity space than uninfected males ( 61 , 310m2 vs 31 , 575m2 , p = 0 . 006 ) . While we did not identify associations between exposure and specific risk groups at the population level , we did in infected individuals . Exposure to both trash and rodent activity were significantly higher in infected individuals than in other study participants ( see Table 3 ) .
Steps taken to optimize our GPS protocol were successful , resulting in few refusals or problems with non-adherence . Our selected GPS model , the MobileAction iGot-U 120 , was reasonably priced yet programmable , durable , and password protected . High participant acceptance rates indicated that privacy concerns do not preclude GPS studies in this setting . This high acceptance rate may be due in part to our group’s familiarity to the community , but residents found the technology interesting and were eager to wear the units , an attitude which may transfer to other locations . Additionally , other groups have had success implementing GPS studies in diverse settings [8 , 9 , 13] . GPS tracking has limitations that must be considered when determining the method’s suitability for specific research questions and settings . First , GPS locations are not exact . The standard deviation of our units’ recorded locations was 5 . 67 meters in each coordinate direction , and even more expensive higher-accuracy units have errors on the order of meters . In complex environments with fine-scale environmental heterogeneity , exposure can vary at smaller distances than this . Second , GPS points are not taken exactly as scheduled , and data can contain gaps of several hours ( often associated with participants going indoors ) . Third , the GPS unit will record data regardless of whether it is being worn as instructed . Finally , recorded movement patterns may be atypical , either due to unusual circumstances or a conscious effort to modify one’s behavior due to observation . In the absence of directly observing GPS participants , researchers must rely on self-reported adherence to study instructions of wearing the GPS during a day of normal activity . We were able to avoid or adjust for these limitations in the current study . By buffering GPS points in the activity space analyses we took into account both participant movement between points and point location error . Similarly , in our exposure analyses , we used a spatial kernel to take a weighted average of exposure values near the recorded GPS point to account for positional uncertainty . Exposure to leptospires likely occurs outdoors , so GPS points lost while participants were indoors should not affect relevant exposure measures . Finally , by conducting careful exit interviews , we were able to elicit information about non-adherence to study protocols . Population-level analysis revealed that participants spent most of their time near their home and that exposure to environmental features associated with leptospirosis infection was similar across age and gender groups . People living in households near trash deposits , open sewers , vegetation and exposed soil , areas of high rodent activity , and flood-prone areas have an elevated risk of leptospirosis infection [32 , 37 , 38] . However , when we examined movement-induced contact with these environmental features , we did not identify a specific exposure associated with males , the group at highest risk of infection . It seems that within the peridomestic environment where residents spend most of their time , exposure to these features is ubiquitous and does not differentiate between risk groups . It is possible that our focus on high-risk individuals living in the valley bottom obscured differences that would be observed in the total slum population , but our study population was representative of the total population on all metrics except household proximity to an open sewer . Though we did not find associations between specific environmental exposures and leptospirosis risk , these results highlight the importance of the peridomestic environment , an area with dense exposure to transmission sources where people spend most of their time , as a critical location for public health action . Interventions in this setting will improve conditions experienced during a high proportion of slum residents’ time . While males spend as much time near the home as females , over 24 hours males visit a significantly larger area within the peridomestic environment . Thus , activity space identified the population group at high risk ( males ) . Males have higher employment rates than females , but within each gender activity space does not significantly differ by employment status ( p = 0 . 182 for males and 0 . 184 for females ) . That males have larger activity spaces seems therefore to be a difference between genders which is unrelated to their employment status . This daily travel through a larger area may place males at higher risk of infection by exposing them to more potential transmission sources , some of which by chance impart a sufficient inoculum dose . Analysis of the four infected individuals revealed unique movement and exposure characteristics . Infected males had significantly larger activity spaces than the average male and some of the largest activity spaces among all participants , lending further support to the hypothesis that movement is a risk factor for infection . Additionally , infected individuals had significantly elevated exposure to rodent activity and to trash deposits , which provide rodent habitat and food . These associations hint that movement generates exposure to bacteria excreted in the environment by rodents , the reservoir of urban leptospirosis . Our results indicate that household-based risk factors capture movement of individuals to bacteria at sources near the household , instead of general peridomestic contamination . If household proxies captured risk due to pathogens in the household environment , infection rates should be equal across age and gender groups according to our time-near-home and exposure analyses . But instead , movement behaviors captured by the activity space are associated with infection . Public health measures that construct barriers to movement-induced contact with environmental bacterial sources may therefore be an important additional intervention against leptospirosis . This study tracked a small number of high-risk urban slum residents over a single 24-hour period per person . Because we tracked a subset of participants on each sampling day , we could not robustly examine temporal interactions between individuals . We therefore focused on general movement and exposure patterns instead of shared interactions or specific locations visited . The study population lives closer to sewers than the total urban slum population but is otherwise representative , indicating that our findings about where slum residents spend time and how much they move are generalizable across this site . Despite our relatively small sample size of 109 , we identified four infections and detected associations in the population as a whole as well as among the four infected individuals . Our finding that males have larger activity spaces may be influenced by the fact that females have more missing data . However , males also report spending nearly three more hours per day outdoors on average than do females . This is consistent with males having both a higher number of GPS points and a larger activity space , so we feel that our study captured a true difference in behavior by gender . Finally , the battery life of our units limited us to 24 hours of tracking at a time . While we tracked each individual once to maximize population coverage , future studies could track individuals over several days to estimate inter-day variation in movement and exposure . Slum resident movement patterns likely vary according to specific features of the urban context , but the finding that individuals spend much of their time near home may be generalizable to other populations with low levels of formal employment . The association between movement and infection risk may apply to other diseases for which the causative agent is present in the environment . In conclusion , GPS tracking is an effective tool for understanding movement and exposure in urban slum populations . By tracking urban slum residents for a 24-hour period , we identified features of their movement patterns and resulting exposures that will inform further studies of leptospirosis as well as preventive measures . While it is important to consider methodological limitations of GPS tracking during study design , individual movement could provide critical additional information for studies of the broad range of environmentally-transmitted pathogens , both in urban slums and other settings .
|
Environmental features of urban slums including inadequate sanitation , substandard housing , and population crowding predispose residents to numerous infections . Despite this shared environment , not all slum residents , even within households , have equal risk of infection with specific pathogens and we do not know why . Individual movement data will help us better understand how slum residents interact with their environment . We conducted GPS tracking of 109 urban slum residents in Brazil to quantify their movement patterns and how these influence exposure to leptospirosis , an environmentally transmitted infection common in urban slums . Slum inhabitants , regardless of age and gender , spent most of their time close to home and had similar exposures to environmental features associated with leptospirosis infection . However , males visited a larger area on a daily basis , which may explain their higher leptospirosis risk . Based on screening of the slum population conducted at six-month intervals , four individuals ( all male ) became infected with Leptospira during our study . These individuals visited a significantly larger area than other males and had higher exposure to rodents and trash deposits than did other participants . GPS tracking allowed us to identify movement and movement-induced exposure as risk factors for leptospirosis infection and could provide similarly important information for other environmentally-transmitted pathogens .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"global",
"positioning",
"system",
"engineering",
"and",
"technology",
"pathogens",
"tropical",
"diseases",
"geographical",
"locations",
"social",
"sciences",
"vertebrates",
"human",
"mobility",
"animals",
"mammals",
"bacterial",
"diseases",
"research",
"design",
"urban",
"environments",
"cohort",
"studies",
"neglected",
"tropical",
"diseases",
"human",
"geography",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"geography",
"zoonoses",
"south",
"america",
"navigation",
"brazil",
"leptospirosis",
"people",
"and",
"places",
"rodents",
"eukaryota",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms",
"terrestrial",
"environments"
] |
2018
|
Fine-scale GPS tracking to quantify human movement patterns and exposure to leptospires in the urban slum environment
|
Environmentally mediated infectious disease transmission models provide a mechanistic approach to examining environmental interventions for outbreaks , such as water treatment or surface decontamination . The shift from the classical SIR framework to one incorporating the environment requires codifying the relationship between exposure to environmental pathogens and infection , i . e . the dose–response relationship . Much of the work characterizing the functional forms of dose–response relationships has used statistical fit to experimental data . However , there has been little research examining the consequences of the choice of functional form in the context of transmission dynamics . To this end , we identify four properties of dose–response functions that should be considered when selecting a functional form: low-dose linearity , scalability , concavity , and whether it is a single-hit model . We find that i ) middle- and high-dose data do not constrain the low-dose response , and different dose–response forms that are equally plausible given the data can lead to significant differences in simulated outbreak dynamics; ii ) the choice of how to aggregate continuous exposure into discrete doses can impact the modeled force of infection; iii ) low-dose linear , concave functions allow the basic reproduction number to control global dynamics; and iv ) identifiability analysis offers a way to manage multiple sources of uncertainty and leverage environmental monitoring to make inference about infectivity . By applying an environmentally mediated infectious disease model to the 1993 Milwaukee Cryptosporidium outbreak , we demonstrate that environmental monitoring allows for inference regarding the infectivity of the pathogen and thus improves our ability to identify outbreak characteristics such as pathogen strain .
Modeling infectious disease transmission by person-to-person contact has a long history in the scientific community . Environmentally mediated transmission modeling , on the other hand , has only recently been explored , notably including i ) the Codeço model [1] ( based on an older model by Capasso and Paveri-Fontana [2] ) that has strongly influenced the field of cholera modeling ( e . g . [3–9] ) ; ii ) a series of enteric pathogen transmission models that investigated multiple transmission pathways [10–13]; iii ) the Environmental Infection Transmission System ( EITS ) model [14] , which suggested that properties of the environment could mediate between frequency- and density-dependent transmission; and iv ) the SIWR ( Susceptible , Infectious , Water , Recovered ) model [15–19] , which has been used to develop analytic results for models of waterborne disease with multiple transmission pathways , especially on networks . Explicit modeling of pathogens in the environment can generate additional insight into how environmental processes affect infectious disease dynamics and allow modelers to incorporate knowledge from experimental studies into their models . In particular , it allows for consideration of pathogen fate and transport [20] and the functional relationship , called the dose–response relationship , between the amount of pathogen a person is exposed to ( dose ) and the probability of infection , illness , or death ( response ) . Ultimately , because many interventions work through environmental media ( e . g . water treatment , surface decontamination , etc . ) , environmental modeling can improve the predictions arising from disease transmission models , enhancing its applicability to outbreak control and mitigation planning . Microbial dose–response modeling has largely grown out of the field of quantitative microbial risk assessment ( QMRA ) and was established , for gastrointestinal pathogens in particular , by seminal work by Haas [21 , 22] and Teunis [23] . Work in this area has emphasized the biological plausibility of the exponential and beta–Poisson single-hit models , which semi-mechanistically model pathogen distribution in doses and survival in the host . Empirical models , which come from the field of chemical toxicology and are based on the theory of tolerance distributions [24] , have also been used , particularly for foodborne diseases [25 , 26] . In practice , those seeking to develop a dose–response relationship for QMRA must find data for an appropriate host organism that aligns with the exposure route and desired endpoint ( e . g . symptoms or clinical infection ) [27] . Once appropriate data are found , the choice of functional form from among a plausible set is usually a statistical one ( goodness-of-fit or best-fit ) . In applied work , transmission modelers have thus far considered only the most mathematically tractable of dose–response relationships in transmission models that explicitly consider pathogen dynamics: the Codeço model uses the Hill-1 function and both the EITS and SIWR model use a linear relationship between pathogen levels in the environment and the probability of infection . However , the relationship between exposure and infection risk could be more complex , and the consequences of the misspecification of this relationship on model dynamics and predictions have not been explored in detail . Theoretical work , such as by Wang and Liao [28] , has generally been agnostic to the functional form of the dose–response relationships and does not compare the forms used in experimental science or consider which are most appropriate for practical applications . Thus , there remains a clear need to formalize which dose–response forms should be used in transmission modeling and what considerations are necessary for evaluating competing dose–response models . To address this need , here we Finally , we apply our methodology to the 1993 Milwaukee Cryptosporidium outbreak , using the identifiability results to demonstrate the added power of environmental monitoring .
A dose–response function f connects a number of pathogens ( dose x ) , with a probability of infection ( response f ( x ) ) . Biologically justified dose–response functions have the following properties: i ) zero probability of infection when there are no pathogens ( f ( 0 ) = 0 ) ; ii ) larger probabilities of infection at larger doses ( f is increasing ) ; and iii ) saturation at 100% probability of infection for an infinite dose ( limx → ∞ f ( x ) = 1 ) . Hence , any cumulative distribution function can , in principle , be a dose–response function . Here , we consider eight dose–response functions that are used in QMRA , mathematical modeling , or experimental science: the linear , exponential , exact beta–Poisson , approximate beta–Poisson ( also known as Lomax ) , Hill-1 ( also known as Michaelis–Menten or Langmuir ) , Hill-n ( also known as log-logistic ) , log-normal ( also known as log-probit ) , and Weibull functions . The equations and selected properties for these functions are given in Table 1 . Although several other empirical functions are also fit to dose–response data , we restrict ourselves to the most common examples . We consider four properties of these dose–response functions , which we subsequently define in greater detail: i ) whether they are single-hit models , ii ) whether they low-dose linear , iii ) whether they are scalable , and iv ) whether they are concave . Single-hit models are considered by many to be the most biologically plausible of the common dose–response relationships [29] and are derived from two main assumptions , namely that a single organism can cause an infection and that pathogens act independently [30] . We discuss the form and derivation of single-hit models in greater detail below . The behavior of dose–response functions in the low dose regime , not only for pathogens but also for radiation , chemical exposure , etc . , is difficult to assess experimentally . Several theoretical behaviors have been posited , including thresholding ( no effect below a certain exposure ) , linearity ( effect proportional to dose even at the lowest doses ) , and hormesis ( small exposures have a net benefit ) . Arguments have been made for low-dose linearity for radiation and chemical exposures [31 , 32] , and for pathogens , experimental data are not consistent with thresholds: threshold models have a steeper slope at the median dose than the exponential model , but nearly all experimental data indicate slopes equal to , or shallower than , the exponential model [29] . Moreover , it is generally well-accepted that a single organism is capable of causing disease [29] . Here , we distinguish low-dose linear model behavior by the technical property 0 < f′ ( 0 ) < ∞ . A dose–response model is scalable if there is a parameterization of the function in which the dose x and some parameter appear only as a product with the other . If a model is scalable , the shape of the dose–response function will not depend on the units of the dose . Scalability is most important when measurements of pathogen concentration are not necessarily in units of individually infectious organisms , which is particularly common for viruses , where doses may be measured in TCID50 ( median tissue culture infectious dose ) , pfu ( plaque forming units ) , or ffu ( focus-forming units ) , etc . , rather than number of individual viruses . Convexity ( f″ ( x ) ≥ 0 ) in a dose–response function means that an additional pathogen in a dose increases the probability of infection beyond what it would independently and implies synergy , or cooperation , between pathogens . Pathogen independence , on the other hand , assumes no cooperation and implies concavity ( f″ ( x ) ≤ 0 ) , due to saturation effects . Review of experimental evidence points toward independent action [33] , and , thus , argues for the use of concave dose–response functions . Please note that functions that are concave may not appear so when plotted with dose on the log-scale . We consider an environmentally mediated infectious disease transmission model—based on the EITS and SIWR models [14 , 15]—that includes a dose–response relationship and a latent period . All pathogens we consider are more realistically modeled with a latent period , incorporated by including one or more compartments for an exposed class of individuals who have become infected but are not yet infectious . Although one compartment is often sufficient , additional compartments reduce the variance of the modeled latent period ( see S1 Appendix for model extensions ) . A schematic of the model is shown in Fig 1 , the variables and parameters of the model are given in Table 2 , and the equations are as follows . S ˙ = - κ f ( ρ W ) S E ˙ = κ f ( ρ W ) S - σ E I ˙ = σ E - γ I R ˙ = γ I W ˙ = α I - ξ W ( 3 ) Although the original EITS model counted the number of pathogens in the environment , it is more straightforward for our purposes to track the concentration of pathogens . We also parameterize the sum of pathogen pathogen pick-up κ ( ρ/V ) N and pathogen decay μ rates as a single pathogen removal rate parameter ξ = κ ( ρ/V ) N + μ . The corresponding model tracking numbers of pathogens may be found in the supplment . We also analyze the corresponding model with a linear dose–response relationship , where a linear pathogen infectivity parameter π , which corresponds to the low-dose linear slope π of the functions given in Table 1 , replaces the dose–response function . S ˙ = - π κ ρ W S E ˙ = π κ ρ W S - σ E I ˙ = σ E - γ I R ˙ = γ I W ˙ = α I - ξ W ( 4 ) The models in Eqs ( 3 ) and ( 4 ) may differ from other environmentally mediated transmission models in a number of ways , including that we do not consider human birth and death ( that is , we have a constant population N = S + E + I + R ) as we are considering epidemic , rather than endemic , timescales , and we do not include a distinct person-to-person transmission pathway . While variations to these assumptions are important in some contexts , we focus here on a simple model to highlight the impact of including a dose–response relationship on environmentally mediated transmission . Equations for versions of the model incorporating person-to-person transmission or birth–death dynamics are included in S1 Appendix for reference , and we discuss the robustness of the basic reproduction number results to changes in these assumptions in a later section .
Here we present a number of examples to highlight the impact the choice of a dose–response function has on model dynamics . As an initial example , we fit seven dose–response functions to data for the Iowa strain of Cryptosporidium parvum [45] by maximum likelihood estimation . Best-fit parameters and negative log-likelihoods are given in Table 3 . There is good agreement , qualitatively , among the seven functions ( Fig 2a ) ; in particular , the exact and approximate beta–Poisson models are indistinguishable . These seven functions are used as the dose–response relationship f in the model given in Eq ( 3 ) , parameterized to reasonably approximate Cryptosporidium ( note that parameters with significant uncertainty , V and α in particular , were chosen so that the exponential model and beta–Poisson models , which are the models most commonly used in practice , give reasonable outbreaks ) . Despite the seeming agreement among the dose–response functions , the corresponding dynamics differ significantly in the total size of the outbreak and the timing and size of the outbreak peak ( Fig 2b , Table 4 ) . The differences in dynamics are solely a result of the shape of the dose–response function; each simulation uses the same model , parameters , and initial conditions and differs only in the choice of dose–response function . The following observations help to understand this phenomenon . First , the dynamics of these simulations are driven by low-dose exposure: the average number of pathogens per exposure does not exceed one pathogen for any simulation at any time ( Fig 2c ) . Second , when we consider only the low-dose range of the dose response functions ( Fig 2d ) , there is a substantial spread in the dose–response behavior , resulting in markedly different reproductions numbers ( Table 4 ) . Choosing a dose–response functional form based on statistical fit alone is problematic; the probabilities of infection in the low-dose range , which control the dynamics , differ widely over the possible functional forms despite nearly equivalent statistical fits to the experimental data . There is much less uncertainty in the low-dose regime for any one functional form ( especially the one-parameter forms ) than is seen across the gamut of functional forms ( Fig 3 ) , and thus the choice of a functional form by statistical test can artificially shrink the confidence bounds of the results . Hence , it is important for modelers to conduct sensitivity analyses when selecting a dose–response function . To further explore the issue incorporating dose–response models into transmission models , we next consider dose–response functions and corresponding dynamics for Vibrio cholerae and Shigella flexneri . Analogous analyses for influenza , rotavirus , and Salmonella typhi may be found in S1 Appendix . For Vibrio cholerae , there is significantly less agreement among the dose–response functions than there was for Cryptosporidium ( Fig 4 ) . The Weibull , Hill-n and log-normal functions have unrealistically high modeled single-pathogen infection probabilities ( around 0 . 2–0 . 4 ) . Moreover , for the Weibull and Hill-n functions , R 0 takes the uninterpretable value of ∞ and will have an outbreak for every initial condition . For the log–normal function , R 0 = 0; even though R 0 < 1 , because the log-normal function is convex at the origin and the initial conditions are not sufficiently close to the disease-free equilibrium , this is a scenario in which an outbreak is nonetheless observed . The same is true for Shigella ( Fig 5 ) . While the exact and approximate beta–Poisson models gave equivalent outbreaks for Vibrio cholerae , there is a slight difference in their corresponding outbreak dynamics for Shigella , even though there is no visible discrepancy ( on this scale ) in the dose–response functions . The estimated value of β is higher for Vibrio cholerae than for Shigella ( β ≈ 16 vs . 10 ) , and , more importantly , the difference between the low-dose slope for the exact and approximate functions is an order of less for Vibrio cholerae than for Shigella ( πaBP − πeBP ≈ 4E-5 vs . 5E-4 ) . Our results suggest that the approximate beta–Poisson function is indeed an acceptable approximation in most cases but that care should be taken to consider the possibility of discrepancy . In summary , middle- and high-dose data for fitting dose–response functions do not satisfactorily constrain the behavior of the dose–response model at low doses . Transmission dynamics , especially in non-outbreak conditions , are likely characterized by low-dose conditions [29 , 57] , and , indeed , our results show that typical outbreak curves can result from low-dose conditions . The choice of an appropriate function for a transmission model , therefore , should not be solely based on either statistical fit or simple mathematical tractability , as misspecification at this level will propagate through the model . By rejecting the empirical functional forms , we can constrain the uncertainty in the low-dose response to some degree , and , in a later section , we demonstrate how we can better manage this uncertainty with identifiability analysis . There are many sources of uncertainty beyond the choice of dose–response function fit to experimental data . For example , the experimental data is generally obtained from healthy members of the population—or a surrogate host population—using attenuated pathogen strains . Uncertainty , therefore , is introduced because the data may not adequately represent the infection probabilities for a given outbreak . Further , there is uncertainty in several other transmission model parameters , including contact rate and shedding rate , which we discuss in greater detail in upcoming sections . One unresolved issue for environmentally mediated infectious disease transmission models is how to aggregate environmental exposure into discrete doses correctly . Consider airborne pathogens , for which contact with the environment is the act of breathing . It is difficult to define the contact rate in a meaningful way: Does one breath constitute an independent dose ? Or is it one hour of exposure , or one day , that is aggregated to an independent dose ? Although enteric pathogens may seem simpler at first ( one can define contact as a specific act of ingesting food or water or of touching a contaminated surface ) , we face the same problem: is a dose a swallow , a glass of water , or all of the water imbibed in a day ? For example , if a person ingests 100 infectious cells on three separate occasions in a day , the person’s probability of infection is modeled as 1 − ( 1 − f ( 100 ) ) 3 if each ingestion event is assumed to be independent , and it is modeled as f ( 300 ) if we assume that all exposure in a single day can be aggregated into a single dose . In the context of the transmission model , which is focused on the population scale ( going from Reed–Frost to Kermack–McKendrick mass-action [58] ) , the question is whether the force of infection , that is the rate at which susceptible people become infected , is 3f ( 100 ) or f ( 300 ) . To use dose–response functions in an environmentally mediated transmission model , we must specify the time-scale on which exposures can be considered independent , that is , we must define the contact rate with environment κ ( where each contact is an independent exposure ) and the per-exposure pathogen pick-up volume ρ . In doing so , we must decide whether the total number of pathogens contacted in a day comes in many , small or few , large independent doses . In this section , we show that whether the force of infection is greater or lesser for many , small doses versus few , large doses will depend on the form of the dose–response function . To demonstrate the potential effects of how exposure is aggregated on model dynamics , we compare the ingestion of many , small doses with fewer , large doses while keeping the total dose ( κρW ) constant under high , medium , and low total dose conditions . In particular , we consider the daily force of infection κf ( ρW ) ( assuming W is approximately constant on this time scale ) relative to the force of infection when κ = 8 contacts per day , i . e . we consider κf ( ρW ) / ( 8f ( κρW/8 ) ) . The choice of κ = 8 here corresponds to independent doses each hour over an eight-hour exposure . [14] . We then vary the contact rate κ , keeping the total number of pathogens picked up by an individual in a day ( κρW ) constant ( Fig 6 ) . These examples demonstrate the κ–ρ trade-off across a range of doses . We find that , for most dose–response functions and doses , increasing the contact rate but keeping the total pathogens picked up constant increases the force of infection . That is , modeling an exposure as more , smaller doses is more likely to cause infection than modeling it as fewer , larger doses . The one exception that we see is for the log-normal function at low doses . In fact , this property is controlled by the concavity/convexity of the function at the reference point , with force of infection increasing with contact rate for concave functions ( the response function saturates for low contact rates ) and decreasing for convex ones . Further , we see that the κ–ρ trade-off makes the most difference at higher doses , which may cause misestimation of the dynamics near the peak of the epidemic . Finally , the effects of the κ–ρ trade-off is minimized , as one would expect , at the lower doses for those functions that are low-dose linear . The above work assumes that each dose has an independent probability of causing infection , but previous work has demonstrated that dose timing may have an impact on the probability of infection [59 , 60] . In fact , Pujol et al . [59] found that the same dose spread over a longer period of time significantly reduced the modeled risk because the explicitly modeled immune system was overwhelmed when the dose was administered in a short time window . That result considers the real , biological effects of dose timing , whereas ours considers model misspecification . It is unclear whether , in practice , the assumption that there is a time threshold below which doses are additive and above which they are independent is realistic . Few experimental studies have included dose-timing in the assessment of exposure effects . One notable example is Brachman et al . [61] , in which cynomologus monkeys were exposed to anthrax . These authors and other subsequent analysis [60] found evidence that exposure did not need to accumulate to cause infection . This analysis suggests that modelers using dose–response functions should carefully consider the implications of their choice of contact and pick-up rates in the context of the pathogen and setting . If aggregated doses stay within the linear range of the dose–response function , then the choice of how doses are aggregated becomes less important . In general , more experimental research is needed to better characterize the impact of dose timing . We first give the basic reproduction number for our environmentally mediated infectious disease transmission model with a dose–response relationship . The mathematics is well-established; indeed the basic reproduction number has been previously calculated for many similar environmentally mediated infectious disease transmission models , e . g . [14 , 15 , 28 , 62 , 63] . Nevertheless , we give the proof here because it facilitates the proof of Theorem 1 below . Proposition 1 . The basic reproduction number of the model in Eq ( 3 ) is R 0 = α κ ρ N γ ξ · f ′ ( 0 ) . ( 12 ) Proof . In the notation of the Next Generation Method , let x = ( E , I , W ) T be the disease compartments and y = ( S , R ) T the non-disease compartments . Then , F = κ f ( ρ W ) S 0 0 , ( 13 ) V = σ E - σ E + γ I - α I + ξ W , ( 14 ) and we have new-infection and compartment transfer matrices F = 0 0 κ ρ N f ′ ( 0 ) 0 0 0 0 0 0 , ( 15 ) V = σ 0 0 - σ γ 0 0 - α ξ . ( 16 ) Finally , K = F V - 1 = α κ ρ N f ′ ( 0 ) γ ξ α κ ρ N f ′ ( 0 ) γ ξ κ ρ N f ′ ( 0 ) ξ 0 0 0 0 0 0 ( 17 ) Because the matrix is upper triangular , the spectral radius ρ ( K ) is the largest diagonal entry . □ Remark . As a consequence of neglecting birth and death rates , the models in Eqs 3 and 4 have not one but many disease-free equilibria . They can be described as { ( S , 0 , 0 , R , 0 ) : S + R = N} . Above , we have computed R 0 at the equilibrium with a fully susceptible population , ( N , 0 , 0 , 0 , 0 ) . However , it is straightforward to write the effective reproductive number R—that is , the average number of secondary cases arising from a typical primary case in a population that is not fully susceptible—when the initial condition is S0 ≠ N: R = α κ ρ S 0 γ ξ · f ′ ( 0 ) . ( 18 ) We see that the basic reproduction number is controlled by the derivative of the dose–response function at zero , demonstrating the importance of the low-dose range . The value of f′ ( 0 ) corresponds to the linear infectivity parameter π in the formulation of R 0 in for the original EITS model [14] . However , when replacing π by a generic dose–response function f , the value of f′ ( 0 ) may cause R 0 to be zero or infinite , which , although valid from the local-stability perspective , cannot reasonably be interpreted in the sense of the expected number of new infections . Hence , to better facilitate comparison among the dose–response models , we also include the R 0 for the analogous stochastic model , denoted R 0 * , as derived using branching theory [64] . The proof is left to S1 Appendix . Here , f′ ( 0 ) is replaced by f ( 1 ) . Proposition 2 . The basic reproduction number for the stochastic analog of the model given in Eq ( 3 ) is R 0 * = · α κ ρ N γ ξ · f ( 1 ) . ( 19 ) Although the basic reproduction number controls the local stability of the disease-free equilibrium for all biologically reasonable models , global stability results are much stronger and more useful because they give an epidemic threshold for all initial conditions , not just those sufficiently close to the disease-free equilibrium . When the basic reproduction number controls the global stability of the disease-free equilibrium , counter-intuitive scenarios , such as when an outbreak occurs despite R 0 < 1 , cannot occur . Here , we prove that concave-down dose–response functions have the desired global stability properties . As described above , these models do not have a single disease-free equilibrium but rather a set of equilibria with different numbers of susceptible and recovered people summing to N . Laukó [65] previously used Lyapunov methods to find criteria that determine the stability of disease-free sets , but the proof we provide here , which we believe will be more intuitive to readers , uses an extension of the method described by Shuai and van den Driessche [66] . Theorem 1 . Let Θ = { ( S , 0 , 0 , R , 0 ) T: S + R = N} , a one-dimensional subset of the state space { ( S , E , I , R , W ) T} . If f is concave in Eq ( 3 ) , then , if R 0 < 1 , Θ is globally asymptotically stable . Proof . The aim is to construct a Lyapunov function and thus demonstrate the global asymptotic stability of Θ . First , we note that Ω = ( S , E , I , R , W ) T : 0 ≤ S ≤ N , 0 ≤ E + I + R ≤ N - S , 0 ≤ W ≤ α N ξ ( 20 ) is a compact , invariant set for trajectories of Eq 3 . Let x , y , F , and V remain as defined in the proof of Proposition 1 . Define h ( x , y ) = ( F - V ) x - F ( x , y ) + V ( x , y ) = κ ρ W f ′ ( 0 ) N - κ f ( ρ W ) S 0 0 . ( 21 ) We will need h ( x , y ) to be non-negative to construct our Lyapunov function . To this end , assume that f is concave . Then ρWf′ ( 0 ) ≥ f ( ρW ) , and , since S ≤ N , h ( x , y ) ≥ 0 . We now construct our Lyapunov function Q . Let ωT be the left eigenvector of V−1 F corresponding to the R 0 eigenvalue , namely , ωT = ( 0 , 0 , 1 ) . Define Q ≔ ω T V - 1 x = α ξ γ ( E + I ) + 1 ξ W ( 22 ) Then , when R 0 < 1 , Q is a Lyapunov function since Q ˙ = ( R 0 - 1 ) ω T x - ω T V - 1 h ( x , y ) < 0 , ( 23 ) as h ( x , y ) is non-negative . Now , Q ˙ simplifies to Q ˙ = α κ f ( ρ W ) ξ γ · S - W , ( 24 ) and so Q ˙ ≡ 0 if and only if W = 0 . Now , Θ is the largest invariant set in { ( S , E , I , R , W ) T: W = 0} . Hence by Theorem 2 of Lasalle [67] , Θ is globally asymptotically stable . □ Remark . Although the proof was written as if the initial conditions of the system were S0 = N , it is straightforward to see that the proof holds for other initial conditions . One uses R ( Eq 18 ) and notes S ≤ S0 . Corresponding results can be obtained if person-to-person transmission is included in the model; the proof is given in S1 Appendix . A corresponding result when birth–death dynamics are included in the model—which allows an endemic equilibrium to exist and reduces the set of disease free equilibria to a point—is an extension of a result previously shown by Wang and Liao [28] and can also be proved using Theorem 2 . 2 of Shuai and van den Driessche [66] . Corollary 1 . If human birth–death dynamics and person-to-person transmission are included in Eq ( 3 ) , then , if f is concave , we have that Stability results for the endemic equilbrium are likely possible in this scenario as well . If we do not constrain to concave dose–response functions , it can also be shown that the choice of dose–response function can affect the kinds of dynamics that can arise in an infectious disease system , such as multiple equilibria and or periodic orbits [68] . Our results strongly encourage the use of concave , low-dose linear functions . Low-dose linear functions give biologically reasonable ( i . e . finite , non-zero ) values for the basic reproduction number , and concavity ensures that the epidemic dynamics actually correspond to the value of R 0 for all initial conditions . Except in cases where pathogen cooperation or thresholds are specifically being considered , the choice of a concave , low-dose linear dose–response function is biologically sensible and ensures that the basic reproductive number is a useful and relevant measure of the global dynamics of the system . Choosing one of the single-hit models in Table 1 satisfies these properties . In addition to the uncertainty introduced by using medium- and high-dose data to estimate low-dose infectivity , as discussed in a previous section , other uncertainties associated with environmental measures should be considered . In particular , the volume of the environment ( implicitly appearing in the shedding rate α ) is difficult to measure , and the shedding rate can be highly variable depending on pathogen strain or host immune status . One solution is to identify combinations of parameters that control the dynamics , such as R 0 , that we can estimate with confidence from outbreak data . In this section , we show that when we are interested in describing the time series of prevalence , it is not necessary to know f or even uniquely identify the low-dose per-pathogen infectivity π , as long as the average dose remains within the linear regime of the dose–response function . We demonstrate this through identifiability analysis of the linear model ( Eq ( 4 ) ) ; identifiability analysis is the assessment of which parameters , or parameter combinations , can be uniquely determined from the data . The following theorem states that , given time-series prevalence data , we can identity values of parameter combinations that uniquely describe the model fit to the data . We do not need information on the specific parameter values that constitute the combinations . This result follows the work of Eisenberg et al . [16] on the SIWR model , which developed the first identifiability results for an environmentally mediated transmission model and emphasized the role of environmental monitoring in inference of the shedding rate . ( In the notation of Eisenberg et al . [16] , βI = 0 , βW = κρπ , k = 1 ) . Theorem 2 . The identifiable combinations of the model given in Eq ( 4 ) given time series data of prevalence of infected individuals I are απκρ , ξ , γ + σ , and γσ . If the time series of the environmental compartment W is also observed , then α is separately identifiable . Remark . Parameters γ and σ are locally identifiable because there are only two solutions given γ + σ and γσ . In most cases , external information will resolve any ambiguity . The proof is left to S1 Appendix . This result means that , if we have prevalence data but no environmental monitoring , the individual values of the shedding rate α and the pathogen infectivity π can vary without changing the outbreak dynamics as long as the product απκρ does not change . For example , because the product απκρ appears in the numerator of R 0 , the model can be parameterized to match R 0 by balancing the low-dose slope of the dose–response function fit to data ( i . e . π ) with the shedding rate α . ( Because κ and ρ implicitly appear in another identifiable combination , ξ = μ + κ ( ρ/V ) N , they cannot be adjusted without changing this combination if μ is well known from experimental data , although if κ ( ρ/V ) N ≪ μ , the difference may smaller than the measurement error in the data , meaning that the change is not practically identifiable ) . This allows the modeler to mitigate misspecification of the dose–response model . With environmental monitoring , α can be estimated from the data , which allows us in turn to gain more information about π . We illustrate the identifiability results with three examples . First , we expand upon the Cryptosporidium example above by maintaining all of the same parameters given for Fig 2 except for the shedding rate α , which is set so that each simulation has R 0 * = 2 . ( We use the stochastic basic reproduction number instead of the deterministic one to facilitate comparison with the models that have necessarily zero or infinite deterministic R 0 . ) The four low-dose linear functions ( exponential , exact and approximate Beta–Poisson , and Hill-1 ) now give identical dynamics ( Fig 7 ) . The differences in outbreak dynamics given the differences in the fit of the dose–response functions disappear if we privilege our information about R 0 over the shedding estimates . Because of their curvature , the Hill-n ( no outbreak ) , log-normal ( no outbreak ) , and Weibull ( outbreak too large ) still behave badly , further highlighting the importance of non-zero , finite low-dose linearity . In the second example , we use simulated data of an extended , low-prevalence outbreak of Cryptosporidium in a village of 1 , 000 situated by a small body of water . Details for the simulation of data are given in S1 Appendix . The number of cases of cryptosporidosis is recorded monthly . The model ( Eq ( 3 ) ) is fit to the data using each of functional forms found in Fig 2a except for the Weibull , which did not converge , and the exact beta–Poisson . Parameter combinations ακρ , ξ , γ , and σ are estimated . All of the models are able to fit the data reasonably well ( Fig 8a ) . However , the models give very different estimates of concentration of pathogens in the water ( Fig 8b ) under basic assumptions of the frequency and volume of water consumption and an initially fully susceptible population . If we additionally monitor the environment , then the shedding rate α is separately identifiable , and we can see from the lack of fit in Fig 8b that the models with dose–response functions do not capture this parameter or the environmental data correctly . The differences in low-dose infectivity among the dose–response forms can be offset by other parameters when the model is fit the case data alone , but this is not possible when the environment is also measured and daily volume of water ingested is reasonably well known . In both Fig 8a and 8b , we plot the best-fit ( fit to both data sets ) linear model ( Eq ( 4 ) ) , showing that the model with linear infectivity suffices to capture the dynamics . This example demonstrates two important points . First , the additional information available in environmental monitoring can be a powerful tool for parameter estimation , and , second , fixing a dose–response function is essentially a constraint on the identifiable parameter combinations that may lead to spurious estimation of other parameters . When using environmentally mediated infectious disease transmission models , a linear infectivity parameter in lieu of a dose–response function is sufficient when transmission dynamics occur in a low-average-dose setting . In the case that a dose–response function is in fact needed , we should only consider low-dose linear , concave , single-hit ( i . e . biologically plausible ) functions ( e . g . the exponential , beta–Poisson , or , with caveats , Hill-1 ) and be cognizant of the fact that medium- and high-dose exposures are used to fit the dose–response models that are used to examine the impact of low-dose environmental exposures . We should not automatically accede to the best-fit model presented in the literature , especially when multiple functions fit well , but rather acknowledge the uncertainty across functions in the low-dose regime and conduct sensitivity analyses . Using a dose–response function also requires us to consider a biological basis for separating the rate of contact with the environment from the per-exposure pick-up rate . The uncertainty in the low-dose infectivity parameter and in aspects of the environment itself can be better managed by considering the identifiable combinations of the model . Because the shedding rate and the infectivity , along with the contact rate and pick-up volume , occur in an identifiable product , their individual values do not affect the model dynamics , as long as the value of the parameter combination is preserved . This product can be estimated from case data or possibly from the basic reproductive number R 0 . The a priori choice of a dose–response function amounts to a constraint on the value of the infectivity , which , if not appropriate for the particular outbreak , will lead to spurious estimates of other parameters . Alternatively , environmental monitoring provides additional information that can be used to identify shedding rates and , via this identifiable product , pathogen infectivity .
|
Many infectious disease interventions , including water treatment , hand hygiene , and surface decontamination , target pathogens in the environment . Explicitly modeling the concentration of pathogens in the environment within transmission models can be a useful way to consider not only the impact of such mitigation efforts but also the spatial spread of pathogens and sampling strategies for environmental monitoring . However , we need to understand the dose–response relationship , that is , how exposure to pathogens translates into a probability of infection . The field of quantitative microbial risk assessment has developed dose–response models from experimental data , but little work has been done to assess the impact the choice of dose–response model has on transmission model dynamics . We show that dynamics of simulated transmission models incorporating a dose–response model that has been fit to experimental data can vary widely despite little statistical difference in the fit to the experimental dose–response data . This and other results allow us to give specific guidance for the use of dose–response functions in a transmission modeling context . We also underscore the usefulness of environmentally mediated transmission models by demonstrating how environmental monitoring data can be used to provide new information about pathogen strain .
|
[
"Abstract",
"Introduction",
"Models",
"and",
"methods",
"Results",
"and",
"discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"vibrio",
"microbiology",
"cryptosporidium",
"parasitic",
"protozoans",
"physiological",
"processes",
"statistical",
"data",
"vibrio",
"cholerae",
"protozoans",
"mathematics",
"statistics",
"(mathematics)",
"materials",
"science",
"population",
"modeling",
"materials",
"physics",
"population",
"biology",
"infectious",
"disease",
"control",
"bacteria",
"bacterial",
"pathogens",
"infectious",
"diseases",
"molting",
"medical",
"microbiology",
"turbidity",
"microbial",
"pathogens",
"infectious",
"disease",
"modeling",
"physics",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"organisms"
] |
2017
|
Dose-response relationships for environmentally mediated infectious disease transmission models
|
Rabbit Hemorrhagic disease virus ( RHDV ) , a calicivirus of the Lagovirus genus , and responsible for rabbit hemorrhagic disease ( RHD ) , kills rabbits between 48 to 72 hours post infection with mortality rates as high as 50–90% . Caliciviruses , including noroviruses and RHDV , have been shown to bind histo-blood group antigens ( HBGA ) and human non-secretor individuals lacking ABH antigens in epithelia have been found to be resistant to norovirus infection . RHDV virus-like particles have previously been shown to bind the H type 2 and A antigens . In this study we present a comprehensive assessment of the strain-specific binding patterns of different RHDV isolates to HBGAs . We characterized the HBGA expression in the duodenum of wild and domestic rabbits by mass spectrometry and relative quantification of A , B and H type 2 expression . A detailed binding analysis of a range of RHDV strains , to synthetic sugars and human red blood cells , as well as to rabbit duodenum , a likely gastrointestinal site for viral entrance was performed . Enzymatic cleavage of HBGA epitopes confirmed binding specificity . Binding was observed to blood group B , A and H type 2 epitopes in a strain-dependent manner with slight differences in specificity for A , B or H epitopes allowing RHDV strains to preferentially recognize different subgroups of animals . Strains related to the earliest described RHDV outbreak were not able to bind A , whereas all other genotypes have acquired A binding . In an experimental infection study , rabbits lacking the correct HBGA ligands were resistant to lethal RHDV infection at low challenge doses . Similarly , survivors of outbreaks in wild populations showed increased frequency of weak binding phenotypes , indicating selection for host resistance depending on the strain circulating in the population . HBGAs thus act as attachment factors facilitating infection , while their polymorphism of expression could contribute to generate genetic resistance to RHDV at the population level .
Rabbit hemorrhagic disease virus ( RHDV ) , a single stranded positive-sense RNA virus belonging to the Lagovirus genus of the Caliciviridae family , is the cause of rabbit hemorrhagic disease ( RHD ) , a disease affecting wild and domestic rabbits of the Oryctolagus cuniculus species . RHD was first described in Angora rabbits in China in 1984 . By 1987 RHD was detected in Czechoslovakia and Italy , and rapidly expanded to most European countries [1] . RHDV usually kills rabbits within 48 to 72 hours of infection . The disease is characterized by acute necrotizing hepatitis and haemorrhages , sometimes preceded by tracheitis and generally associated with disseminated intravascular coagulation in many organs , particularly the lungs , heart and kidneys . There are three different clinical courses of RHD , the peracute form is distinguished by sudden death with no previous clinical signs . The acute form of RHD involves depression , anorexia , apathy , rapid respiration , anemia and some animals show signs of abdominal distress . Animals perish after one to three days . The sub acute form involves slight clinical symptoms and the animals recover within 2–3 days [2] , [3] . Mortality rates are as high as 50–90% although rates are lower in young animals less than 6–8 weeks-old , and no mortality occurs in animals less than 4 weeks-old . Kittens can become infected and shed virus but do not show clinical signs of the disease . The most common routes of infection are the oral and upper respiratory routes , mainly through direct contact between animals or through contact with water or contaminated food . The virus is present in the blood , organs , secretions and skin or fur of infected animals . It is excreted in large amounts through urine and feces and can also be spread by insects [4] . In addition RHDV is resistant in the environment particularly in dry conditions and to date there is no evidence that RHDV can infect other species [5] . RHDV has become endemic in the original distribution range of the rabbit , Spain , Portugal and France , where it has caused severe long term decline of rabbit population size [6] , [7] . The drastic decline also threatens species dependent on rabbits such as the Iberian lynx ( Lynx pardinus ) and the Spanish imperial eagle ( Aquila adalberti ) , which are specialist predators , and to a lesser extent the Bonelli eagle ( Hieraaetus fasciatus ) . The decline of wild rabbit populations has also had an impact on lizard populations , which use rabbit warrens during hot summer periods [8] . In addition , RHDV can cause devastating losses for rabbit producers , although efficient vaccines are commercially available allowing protection of farmed rabbits [9] . Anti-RHDV antibodies have been detected in wild rabbit serum sampled prior to the reported emergence of RHDV [10] . This has led to speculation that non-pathogenic RHDV strains may have been circulating in rabbit populations prior to the first detected RHDV outbreak . Such strains were indeed discovered later in countries such as Italy , France and Australia [11] , [12] , [13] , [14] . Infection by these non-pathogenic strains has been reported to confer complete protection [11] , partial protection [15] or no protection [12] , [16] against RHDV in rabbits through cross-recognizing antibodies . The genetic diversity between RHDV isolates is quite low even between isolates that are not geographically correlated . Indeed , the nucleotide and amino acid differences between strains range between 1–10% and 1–6% , respectively , which is far lower than the differences observed for other caliciviruses [12] , [16] . Nevertheless , it has been suggested that French RHDV isolates can be assigned into six genetic groups , G1 to G6 following spatio-temporal distribution [17] . In France , G1 has almost completely disappeared and is found exclusively in the south-west near the Spanish border . G2 , in which had been included the strain isolated in the first reported outbreak in China in 1984 , has not been isolated recently . The genetic group G4 emerged from G3 , while G5 and G6 appeared as new independent groups [17] , the latter corresponding to the first antigenic variant identified , RHDVa [18] . More recently , the pathogenic forms of RHDV were shown to cluster into four major groups [19] , [20] with the genetic groups G3 , G4 and G5 as an artificial subdivision of Group 4 identified by Kerr and co-workers , which clusters Western Europe and Bahrain strains collected from 1989 onward . RHDV has previously been shown to bind the oligosaccharide H type 2 and A type 2 ( Fig 1A ) , histo-blood group antigens ( HBGAs ) expressed on the duodenal surface and trachea of rabbits , two possible doors of entry for the virus [21] . HBGAs are polymorphic carbohydrate structures representing terminal exposed portions of larger glycans O- or N-linked to proteins or to glycolipids . In many vertebrate species they are mainly expressed on epithelial cells and only a few primate species , including humans , express them on vascular endothelial cells and erythrocytes . They are synthesized by stepwise addition of monosaccharide units from several precursors by specific glycosyltransferases ( Fig . 1 ) . The recent occurrence of the highly pathogenic RHDV with a documented HBGA binding ability can be expected to provide a useful model to study the impact of the virus on the host's HBGA diversity and reciprocally of the host diversity on the virus HBGA-binding properties . We have here determined host diversity regarding HBGA expression on the duodenum surface , a likely point of entry for the virus , and strain-specific binding of RHDV by examining the binding to synthetic sugars , haemagglutination and binding to the duodenal mucosa of both wild and domestic rabbits . The role of each glycan for binding was determined through specific enzymatic removal of the monosaccharide comprising the A , B or H epitopes . Host variation was determined through semi-quantitative A , B and H phenotyping of rabbit duodenums and structural characterization of glycans by mass spectrometry . The role of HBGA binding in RHDV infection was further tested by challenging AB negative and AB positive rabbits with a strain largely dependent on AB binding , revealing an important role of HBGA binding at lower dose infections . In addition , preliminary evidence for selection of weak-binding ABH phenotypes were detected in wild rabbit populations following RHDV outbreaks . Understanding the selection of weak-binding RHDV phenotypes is important as resistance to infection generates problems for controlling the large rabbit population in Australia and provides possibilities of selecting RHDV resistant animals in areas where the rabbit populations are threatened .
The strains used in this study were chosen to represent each of the six chronologically established genetic groups ( G1–G6 ) previously identified by Le Gall-Reculé and co-workers [17] . To analyze the phylogenetic position of these strains , a neighbor-joining tree was constructed using complete nucleotide sequences of the capsid gene . Similar to the finding of Kerr et al . [20] , the resulting phylogenetic diagram shows a division of the analyzed strains into four rather than six clades with G3 , G4 and G5 forming a major group while G1 , G2 and G6 are clearly distinct highly supported groups ( Fig 2 ) . G1 corresponds to strains that circulated in Western Europe during the first RHDV outbreaks [17] and now exclusively circulate in the Iberian Peninsula and sporadically in the South of France [19] , [22] . G6 corresponds to the antigenic variant strains ( RHDVa ) first described by Capucci et al [18] . To address the question of carbohydrate binding of RHDV , six different strains designated G1–G6 were used . RHDV liver extracts with high virus titres , at least 1×1010 viral RNA copies , were used to screen a panel of 38 polyacrylamide ( PAA ) -conjugated oligosaccharides and 19 human serum albumin ( HSA ) - conjugated oligosaccharides ( see Table S1 ) . The carbohydrates with any capacity to bind RHDV were then used to determine binding over a range of RHDV dilutions ( Fig 3 ) . The dilution corresponding to equivalent amounts of viral RNA , determined through real time RT-PCR , are shown in the figure with a vertical line . The antigenic variant G6 was detected with a mouse monoclonal antibody 2G3 previously determined to bind G6 strains as well as all other pathogenic RHDV strains ( kindly provided by Lorenzo Capucci ) . G1–G5 were on the other hand detected with a high-titered anti-RHDV rabbit serum and the amount of G6 may therefore not be completely comparable to that of the other strains . All strains showed strongest binding to B type 2 . G1 was the only strain showing strong binding to Ley and the binding to H type 2 , A type 2 and B trisaccharide varied between the strains tested . Therefore H type 2 is not the only HBGA which may be of relevance for RHDV binding and individual strains show distinct specificities for synthetic oligosaccharides . Human red blood cells ( RBC ) carry A , B and H type 2 on their surface , all of which may be ligands of the G1–G6 strains tested on synthetic sugars . Therefore the ABH expression in this non-synthetic system was used to test for RHDV strain binding . Moreover , RHDV strains have previously been described to be either haemagglutinating or non-haemagglutinating [23] . We tested all 6 strains of RHDV on human A , B and O blood ( Table 1 ) . All RHDV strains were able to agglutinate B blood consistent with B type 2 recognition . The G2 and G3 strains also showed strong binding to H type 2 on O RBC . G1 and G5 showed weak binding to O RBC , while G4 and G6 did not agglutinate O blood . G1 was able to agglutinate A blood to the same extent as B , whilst all other strains showed weak agglutination of A , indicating B type 2 as a ligand for G1–G6 , H type 2 as a ligand for G2 and G3 and A as a ligand for G1 . To study HBGAs expression and distribution in the rabbit duodenum and trachea , the proposed sites of viral entry , as well as in the liver , a major site of replication , monoclonal antibodies against A and B as well as Ulex europaeus lectin ( UEA-I ) were used to detect A , B and H type 2 ( Ley ) on wild rabbit tissue sections , respectively ( Fig 4 ) . HBGA expression was always restricted to epithelial cells . The nine French wild rabbits tested were found to be either A and B positive ( A+B+ ) at the duodenum surface or A and B negative ( A−B− ) . Staining with UEA-I was much stronger and homogenous on sections from A−B− rabbits ( data not shown ) , suggesting partial masking of H type 2 by the A and B epitopes . In addition , expression of the B antigen appeared to differ from those of the A and H antigens . A , B and H were expressed on the crypts of Lieberkühn ( surface layer of the mucosa ) and not on the Brünners' glands ( deep layer ) . Yet , staining by the anti-B appeared more patchy and irregular than staining by the anti-A and it was always weaker . In addition , in A+B+ animals , A antigen , but not B antigen was detected on the surface of the trachea and on the biliary ducts of portal spaces in the liver . Neither A , B or H antigen could be detected on the liver parenchyma . In order to get insights into the diversity of HBGA expression in rabbits , a more quantitative assay was needed . Therefore , fresh rabbit duodenums were collected from 84 rabbits of both wild and laboratory origin for more detailed studies of ABH expression and RHDV binding . A semi-quantitative ELISA system for rabbit duodenum scrapings determined that all animals do express either H type 2 , A or B . No animal was found to be of a clear non-secretor phenotype as described for humans , and all rabbits expressed detectable levels of either H type 2 or A and/or B , albeit with great individual variations and where A expression generally was stronger than B . Ranking animals by increasing H type 2 detection clearly showed an inverse relationship with A expression and to a lesser degree with B expression , indicating that A and B epitopes mask the H motif ( Fig 5 and Table S2 ) . Based on the detection of the A or B antigens , animals were phenotyped A+B+ , A−B− or A+B− with the respective frequencies 0 . 52 , 0 . 38 and 0 . 1 respectively . The carbohydrates of the rabbit duodenum were further characterized through mass spectrometry ( MS ) . Both N-linked and O-linked glycans were analyzed and blood group antigens have been mainly found on O-glycans , while N-glycans are largely terminated with galactose or sialic acid with virtually no fucose on their antennae ( Fig S1 ) . Many of the O-glycan's peaks identified through MALDI-TOF analysis ( Table S4 ) have been shown to be a mixture of different structures after MS/MS analysis ( Fig S2 ) . In O-glycans , we observed a predominance of core 2 structures , followed by core 3 ( Fig 1C ) . The smallest fucosylated glycan was observed at m/z 708 , corresponding to a core 1 fucosylated trisaccharide bearing a blood group H epitope , while the largest fucosylated glycan had a composition of 8 residues arranged on a biantennary core 2 structure ( m/z 1822 ) with one antenna carrying an A blood group epitope ( Fig 6 ) . Most of the samples analyzed showed high abundance peaks at m/z 708 , and/or at mass 954 , a trisaccharide with composition HexNAc2 , Fuc , Hex , corresponding to different structures in different samples: after MS/MS sequencing , a mixture of core 1 , 2 and 3 structures bearing blood group H , A and Lewis epitopes have been observed ( Fig 6 and Fig S2 ) . A peak at m/z 1199 , with composition HexNAc3 , Fuc , Hex , has also been observed in most of the samples analyzed , showing after MS/MS sequencing a core 3 structure carrying an H or Lewis X epitope in some of the rabbits , while in other rabbits it was recognized as a core 2 , 3 and 4 with terminal A blood group on one antenna ( Fig 6 and Fig S2 ) . In 3 of the 10 samples analyzed , a peak at m/z 1373 was found , with composition HexNAc3 , Fuc2 , Hex , and after further analysis it was found to be a core 3 structure carrying a blood group A epitope ( Fig S3 ) . Detailed compositions of O-glycans of the 10 samples analyzed are reported in Table S3 . In summary , we observed a high variety of O-glycan structures carrying blood group antigens H , A and B ( Table S3 ) . Blood group antigen B has been detected in samples 1 , 4 , 5 , 6 , 8 , and 9 for a total of six B positive rabbits out of ten analyzed , in accordance with data from antibody binding analysis , while blood group antigen H has been detected in nine out of ten samples analyzed and only sample 9 did not show any glycan bearing H blood group epitopes . Mass spectrometry analysis could not distinguish between type 1 or type 2 based antigens but we observed poor reactivity on rabbit duodenal extracts of an anti-H type 1 specific antibody compared to the UEA-I reactivity , recognizing H type 2 . On human saliva from O secretors , both reagents reacted equally well , suggesting that in rabbit duodenum , there is little , if any , type 1 based histo-blood group antigens ( data not shown ) . In all of the 10 rabbits , both wild and domestic rabbits , detection of the B type 3 structure by mass spectrometry matched the B phenotype obtained with an anti-B antibody showing broad specificity toward all types of B antigens ( Table 2 ) . In contrast , detection of A histo-blood group structures by mass spectrometry and using a broadly reactive anti-A did not match . Thus , although three animals were unequivocally phenotyped as A- , the A type 3 structure was found in all 10 rabbit duodenum samples analyzed by mass spectrometry . A type 2 was only found in the rabbits that through antibody detection were determined to be A+ , though two of the A+ phenotyped rabbits did not express A type 2 . In order to determine if the discrepancy between the MS and the phenotyping results for the A type was not due to the specific anti-A that was used in the first place , the A- rabbits were then phenotyped again with several other broad binding , anti-A specific antibodies with confirmed recognition of the PAA-conjugated A trisaccharide . Yet , despite attempts to amplify the signal , the A- phenotyped rabbits remained A- , indicating that in some animals , although present , A type 3 epitopes are not detected on duodenum extracts by ELISA or immunohistochemistry , and the phenotypes of the 10 rabbit duodenum samples as determined by ELISA are noted in Table 2 . Despite this discrepancy , it remains that both the A+ versus A− and B+ versus B− phenotypic dichotomies are as clear-cut as the A or B versus O distinction between humans . To determine the sites of RHDV attachment to the duodenum , A+B+ rabbit duodenum sections were used and all six strains were found to bind to the duodenum surface but not the Brünner's gland , in accordance to A , B and H type 2 expression as discussed above . Binding was also detected on the surface epithelium of the trachea that expresses detectable amounts of H type 2 epitopes despite expression of A antigen , but not on the biliary ducts in the liver where B and H type 2 epitopes are not detectable in A+B+ animals ( Fig 4 ) . The six strains of RHDV were then analyzed for binding to duodenum extracts using the same semi-quantitative system as for ABH phenotyping . Virus binding and ABH phenotypes were normalized , to account for variation of duodenum scraping , against Concanavalin A binding , a lectin which binds mannose of N-glycans . To analyze relationships between A , B or H expression and virus binding , animals were separated into three equal groups of weak , medium and strong binders and correlated to A , B and H presence or absence . G2 binding was significantly correlated to H type 2 expression . G3 binding did not correlate with the presence of either A , B or H antigen . G4 , G5 , G6 and G1 binding significantly correlated to A and B expression and inversely correlated to H type 2 expression , with the exception of G6 binding which did not correlate to B expression ( Table 3 , Table S5 ) . It should be noted that most animals that express A also express B ( A+B+ or A−B− phenotypes ) . Therefore in this association study , A and B antigens are linked . Nevertheless , a small subgroup of rabbits expresses A independently of B ( A+B− phenotype ) ( Table S2 ) . This subgroup was significantly associated with low G2 binding ( p = 0 . 006 ) and inversely with strong G4 , G5 , G6 and G1 binding ( p<0 . 05 ) . In order to visualize differences in binding of these strains to individual animals , rabbits were ranked according to G4 binding in increasing order ( Fig 7 ) . Keeping the same order of relative binding for the five other strains clearly showed important individual differences with animals strongly recognized by some strains but poorly by others . Each strain showing a unique binding pattern , despite similar binding characteristics of G4 , G5 and G6 as described above . To confirm that binding of each strain to rabbit duodenum extracts required the presence of A , B or H epitopes and to get a more complete picture of the strains specificities , A , B or H antigens of duodenal extracts were removed with specific glycosidases prior to virus binding assay . Extracts from three rabbits of three different phenotypes ( A−B− , A+B+ and A+B− ) were treated with either a galactosidase or an N-acetyl galactosaminidase , removing the galactose ( Gal ) of B or the N-acetylgalactosamine ( GalNAc ) of A respectively , to determine the role of A and B binding in the duodenum , followed by an α1-2 fucosidase to remove the fucose of H type 2 ( Fig 8 ) . Efficacy of each enzyme was determined by testing the binding of either an anti-A , an anti-B or UEA-I before and after treatment . The results shown in Fig 8D indicate that the fucosidase removed H type 2 almost completely , whereas treatments with either the N-acetyl galactosaminidase or the galactosidase only resulted in a partial removal of the A and B antigens , respectively . Removing the fucose of H type 2 confirmed that binding of all RHDV strains to A−B− rabbits was dependent on H ( Fig 8A ) . In an A+B− rabbit , cleavage of the GalNAc sugar of A followed by removal of the fucose of H type 2 confirmed the significance of RHDV binding and the importance of A antigen expression ( Fig 8B ) . Indeed , for G2 which was the only strain unable to recognize the A synthetic oligosaccharide and to show no relationship between RHDV binding and A expression , cleavage of the GalNAc residue of A allowed binding since it resulted in appearance of the underlying H type 2 , which became accessible for binding . Inversely , the G1 strain preferred binding to A over H type 2 , as seen in the decreased binding after removal of the A epitope , consistent with the results of agglutination . The other strains were not affected by the removal of A but showed a clear decrease of binding following removal of both A and H , indicating similar binding to A as to H type 2 on the duodenum extracts . Cleavage of the Gal and GalNAc followed by removal of H type 2 of an A+B+ rabbit indicated a major importance of B in the presence of A , as removal of the Gal of B followed by removal of the fucose of the underlying H type 2 abolished binding for G1 , G2 , G3 and G5 ( Fig 8C ) . G4 and G6 binding also preferably bound B over A as removal of B decreased binding , whereas removal of A did not affect binding , or even slightly increased RHDV binding . However , neither G4 nor G6 binding was further decreased after removal of H as A was still expressed in the duodenal scrapings , allowing for binding of G4 and G6 . Thus , the results of RHDV strain binding so far indicates binding of G2 to H type 2 and B , while the other strains are able to bind A , B and H type 2 with variable strength . Human norovirus binding to HBGA's has been determined to correlate with symptomatic infection in human volunteer studies and in outbreak studies . To determine the importance of carbohydrate binding regarding RHDV infection , domestic rabbits were challenged with the G4 strain ( GenBank accession number AJ535094 ) . This genetic group was chosen because it proved to be strongly dependent on binding to A and B antigens and because a breed of rabbits with a previously determined high A−B− frequency ( 63% ) was available at the animal facility . Pre-challenge serum was collected from 6 of the animals , and no RHDV antibodies were detected . In addition , the rabbits from this animal facility proved negative during routine screening for the non-pathogenic calicivirus strain . Furthermore , any protection from cross-reacting antibodies can be excluded as non-pathogenic calicivirus strains circulating in France have been shown to provide no protection against RHD [12] . The 12 week old rabbits were then infected orally with 105 , 107 or 109 genome copies of a G4 liver extract . At necropsy of dead rabbits , typical RHD lesions were observed . Surviving animals were sacrificed 11 days after infection , however one rabbit within the highest infectious dose group was sacrificed 7 days post infection due to ethical considerations as it seemed severely ill . Post-mortem examination confirmed the presence of RHD lesions . Duodenum and liver samples were collected from all of the rabbits post-mortem . 7/10 , 6/11 and 3/10 rabbits died from RHDV with an average survival time of 3 . 5 days , 3 . 5 days and 5 days in the groups infected with 109 , 107 and 105 genome equivalents , respectively ( Fig S5 ) . Post-mortem ABH duodenum phenotyping of the rabbits determined 3 , 3 and 4 A+B+ rabbits in the 105 , 107 and 109 infectious dose groups , respectively , the remaining animals being A−B− ( Table 4 ) . Analysis of the ABH duodenum phenotype and G4 binding to the duodenum of all infected animals resulted in a B expression well correlated with virus binding of the rabbit duodenum of the infected animals ( r2 = 0 . 78 ) ( Fig 9A ) . Nevertheless , it should be noted that all animals were recognized by the G4 strain , albeit with great individual variation . Real time RT-PCR of RNA isolated from the duodenum and liver of the rabbits revealed viral replication in the liver of all animals ( Fig 9B ) . The RNA levels in both liver and duodenum of rabbits succumbed to infection were significantly higher than viral RNA levels of rabbits sacrificed at 11 days post infection ( Mann-Whitney , p<0 . 0001 ) . It should be noted that this difference might be partly due to the difference in sampling time between dead and surviving rabbits ( 3–5 days vs 7–11 days ) . Within the rabbits infected with the lowest dose of 105 genome copies all A+B+ rabbits ( n = 3 ) died , while all A−B− rabbits ( n = 7 ) survived the infection ( p = 0 . 008 , Fischer's exact test ) ( Fig 9C , D , Table 4 ) . This was however not the case for the two higher-dose challenges ( Fig S4 , Table 4 ) , indicating that A and B binding facilitates infection , though the lack of A and B antigens can be overcome by a high viral dose . In order to study possible selection of ABH phenotypes after RHDV outbreaks , rabbit duodenums were collected from two populations located 15 km apart near Perpignan , southern France , where detailed information was available regarding RHDV . Rabbits were sampled by hunters from the two different populations , Claira and Canohès . The Claira population had never been affected by an RHDV outbreak . The Canohès population was heavily reduced during September 2006 by RHDV , where G5 was the primary RHDV- circulating in the area , though Iberian G1 strains were also detected during this period , and the population size strongly decreased ( Stéphane Marchandeau , personal communication ) . Twenty two rabbit duodenums were collected in 2009 from the Claira population and only 5 from Canohès due to the low density of rabbits in the population . All of the 5 rabbits sampled from the Canohès population were B- , and all of them , either of the A+B− and A−B− subtypes , were significantly lower binders of G5 , and similar binding results were seen with G1 ( Table 5 ) . In contrast , the Claira population where RHDV had never been detected had a high frequency of A+B+ ( 82% ) and therefore only few B− animals ( 18% ) . Since the G5 strain binds preferentially to the B antigen , these results suggest that the B− phenotype could have been selected at Canohès following the devastating 2006 outbreak . In Australia , rabbits have been repeatedly infected with the G2 Czech strain of RHDV to control the rabbit population . Infecting a rabbit population with the same strain gives an interesting perspective to study selection from a single RHDV strain . Rabbits were sampled at three different locations , Hattah , Bendigo and Bacchus Marsh . Experimental challenges with RHDV have shown the Hattah and Bacchus Marsh populations to have developed partial resistance to infection ( Brian Cooke , personal communication ) and the non-pathogenic , partially protecting virus RCV-A1 has been detected in the Bendigo and Bacchus Marsh populations , but not in the Hattah population [15] . Rabbit duodenum extracts were analyzed for ABH phenotype and G2 strain binding . Similar to the above described French G2 strain , the Czech G2 strain bound Australian A+B− rabbits poorly ( Table 6 ) . The G2 Czech strain also showed binding to synthetic B and H , but not A , similar to the French G2 strain used above . In addition , both strains showed poor binding to the A+B− individuals regardless of the animals' origin ( data not shown ) . Hattah was the population of the significantly highest frequency of A+B− rabbits and inversely with the lowest frequency of A−B− animals , which are most frequently strongly recognized by the G2 strains , suggesting selection for a subgroup of rabbits with potential of protection against infection with a G2 strain .
A recent study of the phylodynamics of RHDV indicated that France has been the most important source population for RHDV [20] . Although this may be due to sampling bias , a chronological relationship matching their phylogenetic positions has been established in France for the G2 to G5 strains [17] . The G1 strain used in the present study is a recent strain of Iberian origin and G6 strains showed no apparent chronological link with other strains . In France , G1 and G2 , which includes the strain isolated in the first reported outbreak in China in 1984 , have not been isolated since 1990 , though G1 currently circulates almost exclusively on the Iberian Peninsula . In addition , since 2000 a few Iberian G1 strains have been identified in the South of France , along the Spanish border [19] . This may be the result of virus spread across the Pyrenean Mountains via insects or the wind [22] . The Iberian strains , suspected to originate from a single introduction of G1 , have evolved separately from the other RHDV strains and cluster into 6 Iberian clades ( IB1 to IB6 ) [22] . G3 has been isolated in France between 1990–1997 and G4 has been isolated from 1993–1999 . G5 , originally first detected in 1994 and G6 first detected in 1999 are both currently circulating in France . G6 corresponds to the first antigenic variant identified , RHDVa [18] . Although the neighbor-joining tree constructed using nucleotide sequences allowed the allocation of the strains tested into each of the six previously identified genetic groups , it is comparable to the topology presented by Kerr et al . [20] . That the G3 , G4 and G5 genetic groups did not appear as clearly independent as previously reported [17] might be due to the use of complete nucleotide sequences of the capsid in the more recent studies [20] rather than just partial sequences used in the earlier study [17] . Alternatively , this can also be a result of the inclusion of strains that cover most of the worldwide genetic diversity . Nevertheless , the topology of the tree is highly supported by the bootstrap values which are all above 90% for the major nodes . Neighbour joining trees were constructed using amino acids sequences of the entire capsid or of the P2 subdomain with the aim to infer a correlation between the HBGA binding profiles and the evolution of RHDV . No such correlation was observed ( data not shown ) . Regardless , the trees constructed using either nucleotide or amino acid sequences showed that that the six selected virus strains G1 to G6 represent a good cross section of the antigenic diversity amongst the known pathogenic forms of RHDV . The gastrointestinal tract is protected by a thick layer of O-glycans constituting the glycocalix of epithelial cells or presented as soluble mucins . It is therefore not uncommon for pathogens of the gastrointestinal tract to interact with such carbohydrates to be able to access the underlying epithelial cell membrane . A major route of transmission of RHDV is the fecal-oral route and so far viral RNA of the closely related non-pathogenic viruses have been exclusively recovered from the small intestine [11] , [12] , [13] , suggesting that these viruses are primarily enteric viruses but that the pathogenic strains do not remain confined in the gut [24] , [25] . Since an RHDV strain was previously shown to bind to a carbohydrate structure expressed in the gastrointestinal and upper respiratory tracts of rabbits [21] , in the present study we first aimed at determining if this characteristic was shared by other pathogenic strains belonging to different clusters of the RHDV phylogeny and this was analyzed by several methods . Analysis of the binding to a set of HBGA related synthetic neoglycoconjugates revealed distinct binding patterns between strains , although strong binding to the B type 2 motif was common to all strains . The strong H type 2 binding previously observed for a G2 strain was confirmed but the magnitude of H type 2 binding was quite variable among strains . All strains except G2 and G3 were able to bind the A epitope and binding to Ley was observed only for G1 and G6 . The uniquely strong binding of G1 , a strain restricted to the Iberian Peninsula , to Ley suggests an important role of the host genetic background . G1 is the only genetic group present in that region , a fact that might be correlated with the Pyrenees acting as a barrier to the dispersal of both the virus and rabbits , but also by the fact that the European rabbit originated in that area and that populations of other parts of the world consist of only a subset of that original gene pool [26] , [27] , [28] , . The higher genetic diversity of the Iberian rabbits may be reflected in the need of the virus to explore other carbohydrates for binding . Binding to synthetic sugars provides information on the specificity for isolated carbohydrate motifs but the presentation and density of the sugar on the scaffold may not mimic expression in the tissue or on cell surfaces . In order to confirm that all RHDV strains bind to HBGAs , albeit with some differences in specificity , binding to human red blood cells ( RBC ) that express type 2 based HBGAs ( H type 2 on O RBC , A type 2 on A RBC and B type 2 on B RBC ) was examined through agglutination . All strains tested agglutinated human erythrocytes . They showed distinct blood group specificities , although here also a common feature was B blood group recognition . The ability of RHDV to agglutinate human RBCs has been observed soon after the virus discovery [2] , although the existence of non-agglutinating strains was later reported [23] , [31] . As the six strains that we tested spanned the RHDV phylogenetic diversity , our results suggest that all RHDV strains could be agglutinating . There was no indication of the RBCs blood groups used in the reports of non-agglutinating strains [23] . The use of O or A RBCs could thus well explain why some strains appeared non-agglutinating . Regardless , our results show that the ability of RHDV to recognize HBGAs has been maintained throughout RHDV evolution but with somewhat distinct strains specificities . We therefore investigated whether rabbits showed HBGA diversity . HBGA expression in rabbits has previously been reported through analysis of glycolipids of the small intestine or by immunohistochemistry in the small and large intestines , respectively [21] , [32] , [33] , [34] . However , the diversity of expression in the small intestine of individual animals has not been studied previously . We observed here that rabbits are able to express either A , B or H ( Ley ) , mainly based on type 2 and type 3 precursors . With immunohistochemistry , nine wild rabbits tested were found to be either A+B+ or A−B− . A and B were expressed on the duodenum surface mucosa , whereas the underlying Brünners' glands were negative . Expression of A appeared stronger than B expression , and in tissues as trachea and bilary ducts where A and H type 2 expression was weak , B was negative , which was further confirmed with the semi-quantitative phenotype assay . In both assays , the concentration of anti-B necessary to detect the presence of B epitopes was at least 10 times higher than that required to detect B antigen either on human epithelial tissues or saliva . In contrast optimal dilutions of anti-A or UEA-I were similar to detect A or H epitopes in both species . This clearly indicated that in rabbits , B antigen was present in smaller amounts than A and H , or that it was less accessible . Mass spectrometry analysis of the duodenum determined that HBGA motifs were mainly present on O-linked rather than on N-linked glycans and revealed extensive individual variation . This great individual variation was also found through a semi-quantitative analysis of A , B or H expression with specific reagents . Strikingly however , no rabbits , out of over 200 screened individual duodenums , were found to completely lack expression of A , B or H , indicating that a clear-cut non-secretor phenotype does not exist in rabbits , unlike in humans . These data are in line with the findings of Guillon et al . [35] who observed that despite a large polymorphism of the Fut2 and Sec1 coding sequences of wild rabbits , involved in synthesis of H type 2 and H type 3 , all of the detected Fut2 enzyme variants were functional , indicating that the genetic diversity of H expression in rabbits is controlled by the level of expression of the α1 , 2fucosyltransferases or by other as yet unknown mechanisms , but not due to null alleles of an α1 , 2fucosyltransferase gene like in humans . Antibody binding data and mass spectrometry were in accordance for detection of B antigen amongst the 10 rabbits analyzed with mass spectrometry , revealing B+ and B- rabbits . As for detection of A antigen , there was divergence between the mass spectrometry data and the antibody binding . Three rabbits with detectable A epitopes in mass spectrometry appeared A negative using antibodies , despite the use of 4 different well-characterized antibodies able to bind all types of A . For reasons still unclear , all A epitopes are not accessible to antibody binding . A possible explanation is that the A antigen of these three rabbits is exclusively of type 3 , i . e . very short O-glycans where larger surrounding glycans may sterically hinder binding of the antibodies . A second complication may be that glycans may be present not only on proteins but also on glycolipids . Earlier work on glycolipids from rabbit small intestine showed expression of A type 2 and B type 2 in the small intestine of rabbits [33] , [34] . A differential expression of A on O-glycans and glycolipids along with difficult detection of the short A type 3 may account for the difference between mass spectrometry analysis and antibody assay . A complete understanding of the ABH polymorphism in rabbits will require a full genetic description of the system , which appears quite different from that in humans . Thus the frequencies of the different A and B phenotypes ( A+B+ , A+B− and A−B− ) in the various populations that we studied cannot be explained by the polymorphism of a single ABO gene as in humans and our preliminary genetic analysis indicates that there are at least 6 Abo genes in rabbits located in tandem in the genome . A similar situation has already been described in rats which have been reported to have a variable number of Abo genes ( up to 5 ) , with some genes encoding A enzymes and others encoding B enzymes [36] , [37] . Therefore , despite a generally conserved expression of ABH antigens across mammalian species , the genetic mechanisms leading to diversity of expression and intraspecies polymorphism are quite variable . Here the major observation regarding rabbits is that they present extensive individual variability of A , B and H expression in the duodenum which is a primary site of attachment for RHDV leading us to analyze the relationships between ABH expression and the binding properties of different strains . As rabbit duodenums express complex patterns of HBGAs , statistical calculations were used to reveal relationships between individual RHDV strains binding and HBGA expression . Strong relationships were established with A+B+ , A+B− and A−B− in a strain-specific manner , though the relative importance of each epitope was not completely clear . Enzymatic removal of each A , B or H epitope allowed a better assessment of the role of each of these ligands . The results showed that the strains are neither A , B nor H specific but more or less dependent on the level of expression of each of these antigens . Nevertheless and most importantly , for each RHDV strain , binding to individual duodenum extracts ranged from very weak to very strong , with the various strains clearly showing differential recognition of individual animals . Considering the results of the binding assay to synthetic sugars , of human RBCs agglutination , of the association with rabbits ABH phenotypes and of the enzymatic removal of ligands , G2 binds to B antigen as well as to H type 2 but is not able to recognize A . As a result , due to the masking of H type 2 by A , G2 only poorly binds to A+B− animals . Interestingly , strains which later displaced G2 in France , G3–G6 , as well as G1 are all able to bind A . G3 was able to bind A , B and H type 2 , which explains why no significant association was found between duodenum binding and rabbit ABH phenotypes . G4 , G5 , G6 and G1 preferentially recognized A+B+ or A+B− animals over A−B− animals . For these strains , binding to duodenum samples was more strongly associated with the presence of A antigen than of B antigen and inversely associated with the presence of H antigen . However , when A and B were both present , enzymatic removal indicated a greater importance of B over A despite an apparent higher expression of A antigen compared to B antigen in the duodenum . This apparent discrepancy can be explained by their stronger binding to B antigen than to A antigen along with the fact that B antigen is always present when A antigen is also expressed whereas A antigen can be expressed in absence of B in A+B− animals . The ability to bind the A+B− animals therefore explains the stronger association with A than with B expression . Nevertheless , despite similar binding patterns between G4 , G5 , G6 and G1 , these strains showed slight differences resulting in different recognition patterns for individual rabbits . A schematic diagram summarizing these observations is presented in Fig 10 . Although HBGA binding has been determined as crucial for symptomatic infection by norovirus in humans ( see below ) , the role of HBGA binding has not been directly established for RHDV infection of rabbits . For this purpose G4 , a strain with a marked B-specific and to a lesser extent A-specific binding , was used in a challenge experiment in combination with a rabbit breed of domestic rabbits with high A−B− frequency . When rabbits were challenged with 105 genome copies there was a clear relationship between survival and the A−B− phenotype ( or low virus binding ) but the relationship vanished at higher virus challenge doses . It should be noted that duodenum extracts of all animals were recognized by the G4 strain , although binding to some individuals was weak . Thus , infection with a high viral dose can compensate for weak viral binding to HBGAs . The increased survival rate at low infectious doses was not achieved by avoiding infection . Regardless of their survival , all rabbits in the experimental challenge study had detectable levels of RHDV-RNA in the duodenum and liver , indicating that all rabbits became infected . In addition , the groups infected with high doses of virus ( 109 and 107 ) still showed untypically high survival rates ( 30% and 54% , respectively ) , and lower virus loads in the tissues of the survivors . These findings suggest additional mechanisms of controlling the RHDV infection , leading to a less severe course of the disease and overall lower virus titers . In addition , G4 , a strain no longer detected in France , may have been less virulent than other circulating strains . Noteworthy , hepatocytes are prone to potent viral propagation despite being completely devoid of HBGA [38] . Thus , our results suggest that HBGAs function as an attachment factor rather than the main cellular receptor . Previous studies show that the amount of virus in fly spots from flies feeding on RHDV infected livers were sufficient to infect rabbits [39] . Similarly , it was shown that as few as 1 to 10 viral particles may be sufficient for infection with human noroviruses [40] . Thus , it is conceivable that 107 or 109 genome copies correspond to very high amounts of viral infectious particles , much higher than the doses involved in natural transmission during RHDV outbreaks . It is therefore likely that mechanisms allowing protection at low infectious doses will have noticeable effects on rabbit survival rates during natural outbreaks . In order to investigate if selection pressure from RHDV leads to increased frequencies of low binding HBGA phenotypes in natural populations , wild rabbits from several locations in France and Australia were investigated . As rabbit populations are closely monitored in France , this allows for studying populations with a known RHDV-infection history that may have been under selection pressure for weak binding HBGA phenotypes . We had access to rabbits from two different French wild rabbit populations located 15 km apart . The first site , Claira , carries a high density population of rabbits in an area otherwise heavily hit by RHDV outbreaks , and has no evidence of previous RHDV infection . The second site , Canohès , has a population recovering from a major RHD outbreak due to a G5 ( or possibly Iberian G1 ) strain in September 2006 that left few survivors , and small additional mortalities were again recorded in 2007 and 2008 . Despite the limited sample size from the recovering population , we observed a highly significant weaker G5 and G1 binding associated with B- phenotypes among the descendants of RHDV survivor rabbits when compared to the neighboring control population , suggesting that the virus contributed to select animals with a weak binding HBGA phenotype . In Australia , RHDV has been purposely introduced in 1996 to limit the damage caused by rabbits . RHDV quickly proved to be efficient at drastically limiting the population size [41] , however in recent years rabbit survival has again increased due to several factors , such as development of immunity in the populations [42] , [43] , facilitated by partially cross-protecting antibodies from infection by non-pathogenic caliciviruses [15] . In addition , the appearance of genetic resistance to RHDV has recently been observed at some locations where the virus had previously proven very efficient at eliminating rabbits ( Brian Cooke , personal communication ) . Rabbits were sampled from three different populations in Victoria , Hattah Kulkyne National Park , a site with indication of genetic resistance to RHDV and no protection from the non-pathogenic RCV-A1 , Bacchus Marsh , a population with resistance to RHDV and protection from circulating RCV-A1 and Bendigo with little resistance to RHDV and with circulation of RCV-A1 ( Tanja Strive , unpublished results ) [14] . We found the A+B− frequency to be significantly increased in the Hattah population . This phenotype is correlated with low virus binding of both the French G2 strain as well as the G2 Czech , the only strain currently circulating in Australia . The presence of a partially protective benign calicivirus in the two other populations likely resulted in increased survival rates during initial RHDV outbreaks and therefore likely promoted the establishment of immunity in the populations which protected them against subsequent outbreaks . It is feasible that this reduces the selective pressure towards weaker binding HBGA phenotypes as a means to avoid lethal RHDV infection . At present we cannot formally exclude that the observed associations between low HBGA binding and survival could be due to genetic drift rather than from selection by the virus . Nevertheless , it is quite remarkable that the observed associations occurred in a strain-specific pattern , where in the French population affected by a G5 strain with largely B-dependant binding a large increase in B- rabbit frequency occurred . On the other hand , the G2 strain steadily introduced in Australia is able to bind B as well as H type 2 with no ability to bind A , therefore poorly recognizing rabbits with no B expression and where A masks the otherwise available H type 2 ( A+B− ) . Thus , the observed associations between survival or genetic resistance and HBGA binding matched the strain carbohydrate specificity . In addition , a prior study by Guillon et al . showed an association between survival to another devastating RHDV outbreak in France and a haplotype at the α1 , 2fucosyltransferases locus [35] . Although in the latter case the binding of the responsible RHDV strain could not be analyzed , collectively these analyses of wild animals indicate that RHDV outbreaks may select animals depending on their HBGA binding characteristics in a strain-dependent manner . Characterization of the binding profile of two G2 strains showed that they did not bind to the A antigen , unlike the five other RHDV strains tested . Since G2 strains appear to be the first pathogenic strains that circulated when emergence of RHDV was observed , acquisition of A antigen recognition by subsequent strains could have allowed a better coverage of the ABH diversity and targeting of the less susceptible and positively selected A+B− animals . Collectively , the data suggest that HBGA diversity tends to restrict the virus transmission and therefore to protect the host population , which is in accordance with the results of the models developed by Fouchet et al . [44] . The initial observation that RHDV binds to HBGAs led to the discovery that human noroviruses of the Norovirus genus ( NoV ) , which cause gastrointestinal infection commonly known as “winter-vomiting disease” , also bind HBGAs [45] , [46] , [47] , [48] . Most human norovirus strains bind α1 , 2-fucosylated HBGA structures such as H type 1 , Lewis b , Lewis y , A or B antigens ( Fig 1 ) , which are expressed on epithelial cells and mucins of the gastrointestinal tract and saliva of individuals of the secretor phenotype [49] . Secretor individuals possess a functional FUT2 gene ( Fig 1A ) , encoding an α1 , 2 fucosyltransferase expressed on many epithelial cell types . Inactivating mutations of the FUT2 gene results in a non-secretor phenotype where none of the above mentioned structures are expressed on most epithelial cells [50] . Human noroviruses cause a relatively mild , transient infection in healthy individuals although they can be responsible for more severe gastroenteritis in immunocompromised individuals , the elderly and in young children , particularly in developing countries [51] . In human volunteer studies , strain-specific binding of HBGAs correlated to infection and norovirus antibody detection [52] , [53] . Thus , nonsecretor individuals proved completely resistant to infection by the Norwalk strain that binds to H type 1 , Lewis b and the A antigen . Moreover B-type individuals who are poorly recognized by the strain were more likely to be either non-infected or to remain asymptomatic . Likewise , analysis of outbreaks showed a strong impact of either the secretor or the ABO phenotypes on infection [54] , [55] , [56] , [57] , [58] . Analyses of the carbohydrate specificity of many strains showed great variation depending on the combined ABO and FUT2 polymorphisms . Thus , each strain recognizes a subset of the population only , although collectively NoVs can recognize all individuals except the very small subgroup of nonsecretors/Lewis negative ( FUT2-/FUT3- ) individuals who lack both the α2-linked fucose added by the FUT2 enzyme and the α4-linked fucose added by the FUT3 enzyme [45] , [59] , [60] . These data indicate that HBGAs act as attachment factors required for infection by noroviruses but also that their polymorphism contributes to restriction of the transmission of any given strain . Such a relationship between the host and the pathogen diversity suggests co-evolutionary processes or an adaptation of the virus to pre-existing host diversity . Histo-blood group antigens polymorphism is maintained in human populations for reasons that are still unclear . Nevertheless , it has been shown that the ABO and FUT2 genes are under balancing frequency-dependent selection [61] , [62] , [63] , [64] , [65] . In addition , the FUT2 gene has been subjected to intense gene conversion with its paralogues FUT1 and Sec1during mammalian evolution [66] . In various species , including humans , many primate species , rat , mouse , and pig Sec1 is a pseudogene . Yet , some of the alleles encode weakly functional enzymes in rabbits [35] . The ABO gene family underwent a birth-and-death mode [67] of evolution during vertebrate evolution [4] . These elements strongly suggest that through their polymorphism HBGAs are involved in interactions with environmental factors , most likely pathogens and the present observations strongly support of this view . In conclusion , our work showed that similar to human noroviruses and despite a lower genetic diversity , RHDV strains bind to carbohydrates of the HBGA family with distinct specificities allowing them to preferentially recognize some subgroups of the host population . Due to variable cross-recognition of the available carbohydrate epitopes , strain binding patterns do not fit precisely with the ABH phenotypes of individual animals , generating complex patterns of recognition within populations . These carbohydrates act as attachment factors that facilitate infection , or at least symptomatic expression of the disease . Survivors to outbreaks are selected among animals showing the lowest binding and most protected from lethal disease . The polymorphism of ABH expression would thus act to generate genetic resistance to RHDV at the population level .
Work involving the acquisition and sampling of French wild rabbits was carried out in strict accordance with the bylaw ( Arrêté N° 2009-014 ) issued by the Paris Prefecture . The acquisition and sampling of Australian wild rabbits was carried out in strict accordance with the guidelines of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes . All procedures were approved by the Commonwealth Scientific and Industrial Research Organisation ( CSIRO ) Sustainable Ecosystems Animal Ethics Committee ( Licence # SEAEC 06-31 ) . Rabbits were caught alive in cage traps or with ferrets . The cages were set on or around active rabbit warrens and baited with diced carrots . When ferrets were used , warren entrances were covered with purse nets and a ferret was released into the warren to flush out the rabbits . Caught animals were killed by cervical dislocation . The challenge study performed on laboratory animals was carried out in strict accordance with the recommendations of the French National Guide for the Ethics of Animal Experiments . The protocol was approved ( Permit Number: 10/03/09-05B ) by the ethics committee “ComEth Anses/ENVA/UPEC” registered under number 16 to the National Committee of Ethics on Animal Experiments of the Ministry of Higher Education and Research . All efforts were made to minimize suffering ( observation of animals twice a day , euthanasia of animals showing signs of suffering ) and euthanasia was performed under xylacine and ketamine anesthesia . Rabbit duodenums were collected from either domestic New Zealand White rabbits , from 22 wild rabbits of a dense population in Claira , France and from 15 wild rabbits from small populations , including Canohès , with previously documented RHDV infection in the area of Perpignan , France , not far from the Spanish border and from 63 animals from 3 populations in Australia . The first 5 cm posterior to the gastroduodenal junction was removed after clearing the section from intestinal contents , the sample was vigorously rinsed in phosphate buffered saline ( PBS ) and stored in RNAlater ( Sigma-Aldrich , St . Louis , MO ) . Sections of the duodenum were then rinsed in PBS , opened and scraped into RTL lysis buffer ( Qiagen , Hilden , Germany ) containing β-mercaptoethanol . The tissue scrapings were homogenized and split into three parts for ELISA assays , RNA and DNA extraction . The ELISA scrapings were boiled for 10 minutes . Rabbit duodenum scrapings were phenotyped using ELISA . Briefly , the duodenum scrapings were diluted in duplicates in eleven two-fold dilutions with final dilutions ranging from 1/100 to 1/102 , 400 in 0 . 1 M sodium carbonate buffer pH 9 . 5 on a maxisorb plate ( ThermoFischer scientific , Waltham , MA ) . Antibody dependent assays were blocked with 5% non-fat dry milk ( Régilait , Saint-Martin-Belle-Roche , France ) diluted in PBS while lectin assays were blocked with synblock ( AbD serotec , Oxford UK ) . A antigen was detected using mouse monoclonal anti-A antibodies , 2A12 , 2A21 , and 2A15 , all previously characterized to specifically bind A antigen based on all types of precursor structures [68] . B antigen was detected using a specific mouse monoclonal B49 , a B-specific broadly reacting antibody [69] . H type 2 ( Ley ) expression was determined using HRP conjugated Ulex europaeus-I ( Sigma-Aldrich , St . Louis , MO ) . Secondary horse radish peroxidase ( HRP ) conjugated anti-mouse ( Uptima/Interchim , Montlucon FR ) was used for A antigen detection and a biotin conjugated anti-mouse ( vector laboratories , Burlingame , CA ) followed by HRP conjugated avidin ( vector laboratories , Burlingame , CA ) for B detection due to relatively low amounts of B antigen in rabbit tissues . TMB ( BD Bioscience , San Jose CA ) was used as a substrate for all assays and O . D . values were measured at 450 nm . RHDV was prepared from infected rabbit livers . Livers were cut into small pieces and homogenized with PBS at 0 . 25 g/mL . Liver/PBS mixtures were centrifuged 20 minutes to remove cellular debris . RNA was isolated from 200 µL of each liver preparation with the RNeasy mini kit ( Qiagen , Hilden , Germany ) according to manufacturer's instructions and 5 µL of isolated RNA was reverse transcribed from each liver . RNA was reverse transcribed with random primers and Superscript II reverse transcriptase ( Invitrogen , Carlsbad CA ) according to instructions for first-strand cDNA synthesis . 2 µL cDNA/well was further analyzed in real-time PCR with primers and probes designed according to Taqman chemistry , forward primer: 5′ TCTGTCGTCAGGCGCACC 3′ , reverse primer: 5′ GACGAGTAGTTGTTGAGCGAAAG 3′ and probe: 5′ FAM-CAGTACGGCACAGGCTCCCAACCA-TAMRA 3′ . A plasmid containing the amplicon was used as a standard of 4 different concentrations for each run on an Mx3005P ( Agilent technologies , Santa Clara CA ) . All real-time PCR tests were run in triplicates and together with several previously quantified RHDV strain cDNA . In all further assays RHDV from each strain was used in similar concentrations . RHDV binding to rabbit duodenum scrapings was analyzed in the same manner as the duodenum phenotype ELISA . Here rabbit duodenum scrapings were diluted in a range of 11 dilutions and coated in 0 . 1 M sodium carbonate buffer . Plates were blocked with 5% non-fat dry milk diluted in PBS or distillated water . RHDV from six strains belonging to genetic groups G1 to G6 ( Fig 2 ) was prepared from infected livers as described above . High titered rabbit sera Lp4 or an RHDV monoclonal antibody 2G3 was used for G1–G5 and G6 detection , respectively . Secondary antibodies anti-rabbit conjugated with HRP were used against Lp4 and anti-mouse biotin followed by avidin-HRP was used for RHDV detection with 2G3 anti-RHDVa ( G6 ) monoclonal antibody [70] . Addition of substrate and measurement of the plates were performed as described above . RHDV binding to synthetic sugars was tested by screening a panel of PAA-conjugated and BSA-conjugated sugars ( Table S1 ) . 1 µg of synthetic sugars were coated . For the first screening a high concentration of RHDV was used . For the detailed analysis of binding to the positive synthetic sugars a range of dilutions was used . RHDV binding ELISA was performed as described above . RHDV binding step was performed at 4°C for PAA-conjugated sugars and at 37°C for BSA-conjugated sugars , though no major difference was visible at the alternate temperatures . A threshold was set at 3 times the background for each phenotype or RHDV binding assay and dilution value of each sample for crossing the threshold were analyzed . All values were normalized against values obtained with the mannose-binding lectin Concavalin A to control for differences in the amount of material scraped , as protein quantification proved to be difficult due to the addition of β-mercaptoethanol to remove any potential anti-RHDV antibodies in the duodenum that would interfere with the analysis . Different sets of rabbits regarding ABH phenotypes and RHDV binding were analyzed with a chi-squared test or Fischer's exact test . Blood from A , B and O individuals was washed 3 times in PBS and diluted to 2% RBC . RBC were added to virus dilutions in V-shaped wells , 1×109 genomic copies of virus was used as determined through real time RT-PCR , described above . Agglutination titers were determined after 3 h incubation at room temperature . Tissue sections of nine French wild-rabbit duodenums , liver and trachea were de-parafinated through baths in LMR and ethanol . Endogenous peroxidase activity was blocked with 0 . 3% hydrogen peroxide . Non-specific binding was blocked with 5% goat serum in PBS . HRP conjugated Ulex europaeus-I ( Sigma-Aldrich , St . Louis , MO ) at 0 . 8 µg/mL , anti A monoclonal antibody 2A21 and anti B monoclonal antibody B49 were used for binding to H type 2 , A and B phenotyping respectively . A rabbit expressing both A and B antigens was used for RHDV binding . All 6 strains described above were used at 2×109 genome copies/ml ( see RHDV preparation ) and detected with the mouse monoclonal anti-RHDV antibody 2G3 [70] . Dilutions of lectin , antibodies and virus were done in 1% BSA in PBS and binding at 4°C overnight . A biotinylated anti-mouse antibody ( vector laboratories , Burlingame , CA ) diluted in 1% BSA in PBS was bound to all of the assays with primary mouse antibodies . Binding of the biotinylated anti-mouse antibody was followed up with HRP-conjugated avidin vector laboratories , Burlingame , CA ) also this diluted in 1% BSA in PBS . Substrate was added to the slides ( AEC kit , vector laboratories , Burlingame , CA ) followed by Mayer's hemalum solution ( Merck , Whitehouse Station , NJ ) for contrast staining . 31 New Zealand white rabbits of 12 weeks were grouped into three groups of 10 or 11 rabbits and placed in 3 rooms ( 2 cages containing 5 rabbits per room ) at BSL2 experimental facilities with filtered air according to biosafety and bioethical procedures . The experimental study was performed under the authorization of animal experimentation number 10/03/09-05B delivered by the ethics committee ComEth Anses/ENVA/UPEC . Six rabbits were sampled for pre-challenged serum just prior to infection . The rabbits were orally infected with 105 , 107 or 109 genome copies of the G4 virus 95–10 [17] in a total of 1 mL PBS . At the time of infection the rabbits were 12 weeks of age . As rabbits succumbed to infection , post-mortem examinations were realized and liver and duodenum were sampled in RNAlater ( Sigma-Aldrich , St . Louis , MO ) . Surviving animals were killed humanely after 11 days , examined for macroscopic lesions , and tissue samples were collected . Liver and duodenum scrapings were analyzed for RHDV quantification as described above . Duodenum was also phenotyped for ABH expression as described above and tested for G4 binding , also described above . Fut2 XM_00273737 Sec1 X80225 Fut1 NM_001082403 G1 JF438967 G2 FR823355 G3 FR823354 G4 AJ535094 G5 AM085133 G6 AJ969628
|
Rabbit hemorrhagic disease virus ( RHDV ) , detected as late as 1984 , has spread to large parts of the world , threatening rabbit populations and other species dependent on rabbits in many European countries . Mortality has been shown to be as high as 90% and rabbits are killed 48 to 72 hours after infection . Related viruses called noroviruses , infect humans in a manner dependent on the expression of histo-blood group antigens ( HBGAs ) , which are not only expressed on red blood cells , but also on epithelial cells , in saliva and on mucins of the intestinal tract . RHDV also binds to HBGA and in this report we characterize binding of strains of all genetic groups of RHDV to different HBGAs . We also demonstrate HBGAs to function as attachment factors in a challenge experiment . As polymorphisms of genes involved in HBGA synthesis divide the rabbit population into different subgroups , we find selection of low-binding subgroups of wild rabbits in populations recovering from devastating outbreaks of RHDV . This is the first demonstration of differential HBGA specificities of RHDV strains , description of function in infection and demonstration of host selection due to RHDV infection based on HBGA phenotype .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"veterinary",
"diseases",
"emerging",
"infectious",
"diseases",
"veterinary",
"microbiology",
"virology",
"microbial",
"pathogens",
"biology",
"evolutionary",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"evolutionary",
"genetics",
"animal",
"management",
"veterinary",
"science"
] |
2011
|
Histo-Blood Group Antigens Act as Attachment Factors of Rabbit Hemorrhagic Disease Virus Infection in a Virus Strain-Dependent Manner
|
The sole report of annual leptospirosis incidence in continental Africa of 75–102 cases per 100 , 000 population is from a study performed in August 2007 through September 2008 in the Kilimanjaro Region of Tanzania . To evaluate the stability of this estimate over time , we estimated the incidence of acute leptospirosis in Kilimanjaro Region , northern Tanzania for the time period 2012–2014 . Leptospirosis cases were identified among febrile patients at two sentinel hospitals in the Kilimanjaro Region . Leptospirosis was diagnosed by serum microscopic agglutination testing using a panel of 20 Leptospira serovars belonging to 17 separate serogroups . Serum was taken at enrolment and patients were asked to return 4–6 weeks later to provide convalescent serum . Confirmed cases required a 4-fold rise in titre and probable cases required a single titre of ≥800 . Findings from a healthcare utilisation survey were used to estimate multipliers to adjust for cases not seen at sentinel hospitals . We identified 19 ( 1 . 7% ) confirmed or probable cases among 1 , 115 patients who presented with a febrile illness . Of cases , the predominant reactive serogroups were Australis 8 ( 42 . 1% ) , Sejroe 3 ( 15 . 8% ) , Grippotyphosa 2 ( 10 . 5% ) , Icterohaemorrhagiae 2 ( 10 . 5% ) , Pyrogenes 2 ( 10 . 5% ) , Djasiman 1 ( 5 . 3% ) , Tarassovi 1 ( 5 . 3% ) . We estimated that the annual incidence of leptospirosis was 11–18 cases per 100 , 000 population . This was a significantly lower incidence than 2007–08 ( p<0 . 001 ) . We estimated a much lower incidence of acute leptospirosis than previously , with a notable absence of cases due to the previously predominant serogroup Mini . Our findings indicate a dynamic epidemiology of leptospirosis in this area and highlight the value of multi-year surveillance to understand leptospirosis epidemiology .
Leptospirosis is a major cause of illness worldwide with an estimated 1 . 03 million cases , 59 , 000 deaths , and 2 . 90 million disability adjusted life years lost annually [1 , 2] . The burden of disease is thought to be greatest in tropical countries , although reported estimates of incidence in continental Africa are scarce [3 , 4] . Accurate estimates of incidence are important for estimation of disease burden and consequently , appropriate allocation of resources for diagnosis , treatment , and prevention . Challenges in estimating incidence that may account for the scarcity of reports of incidence in Africa include lack of availability of diagnostic tests [5] , low clinician awareness [6] , and non-specific presentation . Although active , population-based surveillance is an ideal method for accurately determining incidence , resource and logistic challenges often preclude its use . Multiplier methods have been used successfully to estimate the incidence of acute infectious diseases in resource-limited settings by extrapolating from hospital based data [7 , 8] . Specifically , multiplier methods were used to determine the incidence of acute leptospirosis in the Kilimanjaro Region during 2007–08 [9] . Using hospital based prevalence data and multipliers from a linked health-care seeking behaviour survey [10] , the annual incidence of acute leptospirosis was estimated as 75–102 cases per 100 , 000 [9] . This estimate of incidence based on empirical data was substantially higher than an estimate ( 7–38 cases per 100 , 000 population ) for Tanzania based on a modelling approach using incorporated data from a systematic review of risk factors [2] . Leptospirosis may cause endemic disease , but is also capable of causing epidemics during flooding or other extreme weather events [11] . As such , data gathered from the same location from multiple time periods can provide insights into the dynamics of disease incidence over time , distinguish periods of endemic and epidemic transmission , and help determine more representative burden of disease estimates . We sought to estimate the incidence of acute leptospirosis in northern Tanzania from 2012 until 2014 using a similar methodology to the previous estimate in the same region in order to describe trends over multiple year periods .
Study Site: We studied patients at two referral hospitals in Moshi , Tanzania . Moshi is the administrative centre for the Kilimanjaro Region that has a population of 1 . 6 million . Moshi is situated at an elevation of approximately 890m and has a tropical climate with rainy seasons from October through December and March through May . Aside from urban Moshi , the region is rural with inhabitants practicing cultivation and small-holder farming . Kilimanjaro Christian Medical Centre ( KCMC ) is a 450 bed hospital and the zonal referral centre for several regions in Northern Tanzania . Mawenzi Regional Referral Hospital ( MRRH ) is a 300 bed hospital and the referral centre for the Kilimanjaro region . Enrolment procedures: From 20 February 2012 through 28 May 2014 the study team approached all adult patients who were admitted to KCMC with a febrile illness as well as all adult or paediatric patients who were admitted at MRRH . In addition we approached every second patient who presented with fever to the outpatient department at MRRH . Hospitalized participants were eligible for enrolment if they had a history of fever within the previous 72 hours or an axillary temperature of >37 . 5°C or a tympanic , oral or rectal temperature of ≥38 . 0°C at admission . Non-hospitalized patients were eligible if they had an axillary temperature of >37 . 5°C or a tympanic , oral or rectal temperature of ≥38 . 0°C . All adult study participants provided written informed consent . For those under 18 years , a parent or guardian provided written informed consent . In addition , written assent was provided for those aged 12 to 18 years . This study differed from the previous Kilimanjaro Region incidence study in its enrolment of outpatients and the enrolment of children at MRRH rather than at KCMC . Enrolment occurred only on weekdays . Enrolled patients underwent phlebotomy , with blood allocated for acute leptospirosis serology only if there was sample available after blood parasite microscopy and blood culture . Participants were requested to return for collection of convalescent serum 4–6 weeks after enrolment . For those who did not attend the scheduled follow up , we attempted to contact them and encourage attendance . Additionally , we recorded inpatient death . Unlike the previous study estimating leptospirosis incidence in the Kilimanjaro Region , we did not record inter-hospital transfer . Laboratory methods: Serology for leptospirosis was performed on acute and convalescent serum samples using the standard microscopic agglutination test ( MAT ) with a panel of 20 Leptospira serovars belonging to 17 serogroups at the United States Centers for Disease Control and Prevention . These included serogroups: Australis ( represented by L . interrogans serovar Australis , L . interrogans serovar Bratislava ) , Autumnalis ( L . interrogans serovar Autumnalis ) , Ballum ( L . borgpetersenii serovar Ballum ) , Bataviae ( L . interrogans serovar Bataviae ) , Canicola ( L . interrogans serovar Canicola ) , Celledoni ( L . weilii serovar Celledoni ) , Cynopteri ( L . kirschneri serovar Cynopteri ) , Djasiman ( L . interrogans serovar Djasiman ) , Grippotyphosa ( L . interrogans serovar Grippotyphosa ) , Hebdomadis ( L . santarosai serovar Borincana ) , Icterohaemorrhagiae ( L . interrogans serovar Mankarso , L . interrogans Icterohaemorrhagiae ) , Javanica ( L . borgpetersenii serovar Javanica ) , Mini ( L . santarosai serovar Georgia ) , Pomona ( L . interrogans serovar Pomona ) , Pyrogenes ( L . interrogans serovar Pyrogenes , L . santarosai serovar Alexi ) , Sejroe ( L . interrogans serovar Wolffi ) , and Tarassovi ( L . borgpetersenii serovar Tarassovi ) . Case definitions: We defined confirmed acute leptospirosis as participants who demonstrated a four-fold rise in agglutinating antibody titres between acute and convalescent serum samples . Cases were defined as probable if a participant’s serum had a single agglutinating titre of at least 1:800 . These definitions were identical to those used to obtain the previous incidence estimate [9 , 12] . The predominant reactive serogroup , for confirmed cases was defined as the serogroup containing the serovar with the largest rise in titres between acute and convalescent sera . For probable cases , we used the serovar with the highest titre to define the serogroup . A time multiplier of 1 . 40 was used to account for enrolment occurring only on weekdays ( 5 of every 7 days ) . Additionally a study duration multiplier of 0 . 44 was included to calculate annual incidence from a study that enrolled for 27 months ( 20 February 2012 through 28 May 2014 ) . We applied enrolment and blood draw multipliers to account for eligible patients who either did not enrol or for whom blood was not available for leptospirosis serology . Calculations of these multipliers are presented in the results . We were unable to include a transfer multiplier in the current study as details of inter-hospital transfer of participants were not recorded . For the estimation of incidence based solely on confirmed cases , a paired sera multiplier was applied to account for those patients who did not have paired sera drawn . Diagnostic test multipliers were used to account for the sensitivity and specificity of MAT serology . The sensitivity was estimated at 100% for paired sera , 48 . 7% for participants with solely acute sera and 93 . 8% for those with solely convalescent sera . The specificity was estimated at 93 . 8% . The estimates are based on a published evaluation of diagnostic tests [9 , 13] and matched those used in the 2007–08 study . A healthcare utilisation survey was carried out in the Moshi Urban ( population 184 , 292 ) and Moshi Rural ( population 466 , 737 ) Districts of Kilimanjaro Region between 13 June and 22 July 2011 as previously reported [9 , 14] . Briefly , 30 of the 45 wards were selected randomly using a population-weighted approach . A study member collected data from the heads of the first 27 households encountered within the ward . A total of 810 households were sampled , comprising 3 , 919 household members . All households had at least one member >15 years of age , 361 had at least one member aged between 5 and 15 years of age , and 198 households had at least one member aged below 5 years . The demographic characteristics from the healthcare utilisation survey have been previously compared to the 2002 Tanzanian Census [9] . Age specific population data has not yet been released from the 2012 Census [14] . Questions relating to health-care seeking behaviour in the event of febrile illness were used to identify participants likely to present to KCMC or MRRH . These questions included , ‘what is the name of the health care facility with an inpatient ward where you/your family would go if you/your family had fever ? ’ and ‘what will you do if a [household member subdivided by age bracket] has a fever for ≧ 3 days ? ’ . The hospital multipliers are presented in Table 1 . Each multiplier is the reciprocal of the proportion of survey participants who responded that they would attend KCMC or MRH as their first or second choice healthcare provider . We used population totals from the 2012 census [14] . As age specific population data were not available from the 2012 census , we multiplied age specific proportions from the 2002 census by the 2012 population total to estimate age-specific populations . The 2007–2008 Kilimanjaro Region incidence estimate used population totals from the 2002 census . We compared incidence by using the estimate of incidence derived from confirmed and probable cases from each of the study periods and the estimated population sampled as the denominator . As shown in Table 2 , the estimated population sampled was calculated by multiplying the total population by the proportion of participants in the healthcare utilization survey that identified KCMC or MRRH as hospitals they would attend in the event of febrile illness . We compared the highest estimates of incidence in each of the study periods . We repeated all calculations using both probable and confirmed cases and then using confirmed cases only . Additionally we performed a one-way sensitivity analysis by varying hospital multipliers according to answers to alternative relevant questions in the healthcare utilisation survey that might also reflect the behaviour of participants and diagnostic test multipliers by using a range of alternative plausible sensitivity values for MAT [15–17] . Data was entered using the Cardiff Teleform system ( Cardiff , Inc . , Vista , CA , USA ) into an Access database ( Microsoft Corporation , Redmond , WA , USA ) . Incidence calculations were carried out using Microsoft Excel 2010 ( Microsoft Corporation . Redmond , WA , USA ) spreadsheets . Other analyses were performed using STATA , version 13 . 1 ( STATA-Corp , College Station , TX , USA ) . We used a test of proportions to compare the incidence between 2007–08 and 2012–14 . All p values are 2 sided and statistical significance was set at p<0 . 05 . This study was approved by the KCMC Research Ethics Committee ( #295 ) , the Tanzania National Institutes for Medical Research National Ethics Co-ordinating Committee ( NIMR1HQ/R . 8cNo1 . 11/283 ) , the Institutional Review Board of Duke University Medical Center ( IRB#Pro00016134 ) and the University of Otago Human Ethics Committee ( Health ) ( H15/055 ) .
Of 1 , 115 participants enrolled from within the study districts , 1 , 017 ( 91 . 2% ) had blood drawn for leptospirosis testing . Of the 1 , 115 participants , 409 ( 37 . 7% ) <5 years , 111 ( 10 . 0% ) 5–14 years , and 595 ( 53 . 4% ) were aged ≥15 years . A total of 593 ( 46 . 9% ) participants were male . A total of 758 ( 74 . 6% ) participants reported having a fever for at least 3 days . Of 1 , 017 participants tested for leptospirosis , 12 ( 1 . 2% ) met the case definitions for confirmed leptospirosis and an additional 7 ( 0 . 7% ) met the case definitions for probable acute leptospirosis . The predominant reactive serogroups among confirmed and probable cases of leptospirosis are summarised in Table 3 . Of both confirmed and probable cases , there were seven ( 1 . 7% , 95% confidence interval [CI] 0 . 4–2 . 9% ) cases among 416 outpatients and 12 ( 1 . 9% , 95% CI 0 . 8–3 . 1% ) cases among 601 inpatients . There was no statistically significant difference in the prevalence of leptospirosis between inpatients and outpatients ( p = 0 . 72 ) . The annual incidence of acute leptospirosis in the Moshi Urban and Rural Districts ( 2012–2014 ) was 11–18 cases per 100 , 000 population using hospital multipliers derived from the question ‘To which facility would you go if you were unwell with a fever lasting ≧3 days ? ’ . When using respnses to the question , ‘What is the name of the health care facility with an inpatient ward where you/your family would go if you/your family had fever ? ’ the incidence was 9–18 cases per 100 , 000 population . The annual incidence was highest in adults , ranging from 13 to 29 cases per 100 , 000 population . Details of the calculation and age-specific incidences are included in Table 5 . The estimated incidence for confirmed and probable cases was ( 18 cases per 100 , 000 population ) was statistically significantly lower than the estimate of 102 cases per 100 , 000 population from 2007–08 ( p<0 . 0001 ) . Estimates of annual incidence in six monthly time blocks is summarised in Table 6 . These data show a higher incidence during the first few months of the study . The results of the one-way sensitivity analysis are presented in Table 7 . When we derived hospital multipliers from alternative questions from the healthcare utilisation survey that might also reflect the behaviour of participants , the estimated annual incidence ranged from 8–37 cases per 100 , 000 population . When we varied the estimated sensitivity of MAT from the lowest to highest plausible values [5 , 13 , 16 , 18 , 19] , the estimated annual incidence varied from 10–25 cases per 100 , 000 population .
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention . Use of trade names and commercial sources is for identification only and does not imply endorsement by the US Department of Health and Human Services or the Centers for Disease Control and Prevention .
|
Leptospirosis is an infectious disease that causes a fever . It can be severe or fatal . Understanding how many people get leptospirosis helps to determine priorities in allocating resources for disease diagnosis , treatment , and prevention . There are few data about leptospirosis incidence in sub-Saharan African countries . The only mainland estimate is from northern Tanzania for the years 2007–08 . To see if leptospirosis incidence had changed since 2007–08 , we measured leptospirosis incidence in the same location in 2012–2014 . To do this , we systematically approached people at two hospitals in the Kilimanjaro Region and tested them for leptospirosis . We adjusted the number of identified cases of leptospirosis found at the hospitals to account for people with fever who did not come to hospital for testing and care . We also adjusted for imperfect testing methods . We found that the number of people who developed leptospirosis annually had dropped from 75–102 cases per 100 , 000 people during 2007–08 to 11–18 cases per 100 , 000 people during 2012–14 . Also , the subtype of leptospirosis responsible for the most cases during 2007–08 was not present during 2012–14 . The number of people developing leptospirosis was not stable , highlighting the value of measuring how commonly leptospirosis occurs over several years .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"leptospira",
"pathology",
"and",
"laboratory",
"medicine",
"infectious",
"disease",
"epidemiology",
"pathogens",
"tropical",
"diseases",
"geographical",
"locations",
"microbiology",
"census",
"bacterial",
"diseases",
"research",
"design",
"signs",
"and",
"symptoms",
"tanzania",
"neglected",
"tropical",
"diseases",
"bacteria",
"africa",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"serology",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"leptospirosis",
"people",
"and",
"places",
"diagnostic",
"medicine",
"survey",
"research",
"fevers",
"biology",
"and",
"life",
"sciences",
"leptospira",
"interrogans",
"organisms"
] |
2016
|
Comparison of the Estimated Incidence of Acute Leptospirosis in the Kilimanjaro Region of Tanzania between 2007–08 and 2012–14
|
Woolly mammoths inhabited Eurasia and North America from late Middle Pleistocene ( 300 ky BP [300 , 000 years before present] ) , surviving through different climatic cycles until they vanished in the Holocene ( 3 . 6 ky BP ) . The debate about why the Late Quaternary extinctions occurred has centred upon environmental and human-induced effects , or a combination of both . However , testing these two hypotheses—climatic and anthropogenic—has been hampered by the difficulty of generating quantitative estimates of the relationship between the contraction of the mammoth's geographical range and each of the two hypotheses . We combined climate envelope models and a population model with explicit treatment of woolly mammoth–human interactions to measure the extent to which a combination of climate changes and increased human pressures might have led to the extinction of the species in Eurasia . Climate conditions for woolly mammoths were measured across different time periods: 126 ky BP , 42 ky BP , 30 ky BP , 21 ky BP , and 6 ky BP . We show that suitable climate conditions for the mammoth reduced drastically between the Late Pleistocene and the Holocene , and 90% of its geographical range disappeared between 42 ky BP and 6 ky BP , with the remaining suitable areas in the mid-Holocene being mainly restricted to Arctic Siberia , which is where the latest records of woolly mammoths in continental Asia have been found . Results of the population models also show that the collapse of the climatic niche of the mammoth caused a significant drop in their population size , making woolly mammoths more vulnerable to the increasing hunting pressure from human populations . The coincidence of the disappearance of climatically suitable areas for woolly mammoths and the increase in anthropogenic impacts in the Holocene , the coup de grâce , likely set the place and time for the extinction of the woolly mammoth .
The woolly mammoth , Mammuthus primigenius , was an herbivorous mammal that lived in the cool and dry open steppe-tundras of the Northern Hemisphere from late Middle Pleistocene ( 300 thousand years before presend [ky BP] ) , or even earlier [1] . They are thought to have finally become extinct 3 . 7 ky ago , on Wrangel Island , Arctic Siberia , [2] . The climate became progressively cooler and drier from the last interglacial period ( 126 ky BP ) , to the Last Glacial Maximum ( 21 ky BP ) , and then became warmer and wetter toward the mid-Holocene ( 6 ky BP ) . These profound climatic oscillations produced a transformation of the vegetation and a reduction in the geographical range of open steppe-tundra habitats , where the last woolly mammoths lived during the mid-Holocene [3] . At the same time , human populations started dispersing across northern Eurasia around 40 ky BP [4] . While data confirm the coexistence of woolly mammoths and humans [5] , some authors suggest that direct evidence of woolly mammoth hunting is scarce [6] . Previous analyses have related the contraction of the mammoth's geographical range and other Late Quaternary Extinctions to both environmental [7–9] and anthropogenic factors [10 , 11] , or a combination of both [12] , but they have often been based upon qualitative or descriptive approaches ( but see [13] and [14] ) . Although the pattern of contraction of their geographical range is known [3 , 15–17] , progress concerning the contribution of environmental factors [18] to explain the extinction of woolly mammoths requires a more quantitative assessment of the contraction of their geographical range and the collapse of suitable climate conditions ( Figure S1 ) . We combined a climate envelope model and a dynamic population model to investigate the extent to which the extinction of the woolly mammoth might have been driven by the collapse of its suitable climate conditions and the intensification of human hunting . The climate envelope of the woolly mammoth was characterised based on statistical associations between the fossil record and palaeo-climate simulations [19 , 20] . We compiled the 14C-dated distribution of fossil records of woolly mammoths in Eurasia for four time periods ( ∼42 ky BP , 30 ky BP , 21 ky BP , and 6 ky BP ) and palaeo-climate simulations for 126 ky BP , 42 ky BP , 30 ky BP , 21 ky BP , and 6 ky BP periods to characterise and project the mammoth's climatic envelope ( Figure S1 ) . We assumed that the climate envelope of the mammoth can be reasonably described using three variables: mean temperature of the coldest month , mean temperature of the warmest month , and annual precipitation . We modelled this envelope combining data for three periods: 42 ky BP , 30 ky BP , and 21 ky BP , and we projected the distribution of the climatic conditions suitable for woolly mammoth for the 126 ky BP and the 6 ky BP periods . We used these results to estimate the decrease in number of woolly mammoths , and we modelled the hunting intensity needed to extinguish the species in four periods: 42 ky BP , 30 ky BP , 21 ky BP , and 6 ky BP . Our results show that the extent of suitable conditions for woolly mammoths in Eurasia progressively collapsed after 42 ky BP: 89% of the species' geographical range disappeared between 42 ky BP and 6 ky BP , probably causing a drop in population size and making the species vulnerable to hunting pressures from a growing human population .
We first evaluated whether the measured climatic conditions in which the woolly mammoths were living changed during the Late Pleistocene . We found that their climatic preferences did not differ significantly during the 42 ky BP , 30 ky BP , and 21 ky BP periods ( Kruskal-Wallis test , n = 54 , n is the number of locations with a fossil presence of woolly mammoths , p = 0 . 186 for mean temperature of the warmest month , p = 0 . 504 for mean temperature of the coldest month , p = 0 . 536 for annual precipitation ) . We also found that their climatic preferences did not differ statistically when we replicated the analysis with n = 141 . On the contrary , we found that they differ when we replicated the analysis with all the records of woolly mammoths , n = 270 ( See Table S1 . This disagreement could be the result of the effect of spatial autocorrelation in the p-values of the replication with all the cases , or could be because of incomplete/biased fossil records , or because of slight inaccuracies in the climate simulations ) . During the three periods analysed , woolly mammoths occupied areas with , on average , 240 mm precipitation per year , and temperatures ranging from −30 . 3 °C to 14 . 5 °C as the coldest and warmest months , respectively ( Figure S2 ) . We split all measured climatic suitability scores into quartiles ( Figure 1 ) to describe different degrees of climate suitability for the mammoth . Deviation from the most suitable conditions is associated with higher Mahalanobis distance ( MD ) scores ( see Material and Methods ) . Therefore , the most suitable conditions are represented by the first quartile of suitability scores ( Q1 ) , corresponding to MD scores below 1 . 02 , and the less suitable conditions within the modelled niche ( Q4 ) correspond to suitability scores above 3 . 27 . Our results show that the most suitable geographic area available to woolly mammoths ( Figure 2 ) , Q1 , increased by 7 . 7 million km2 from the last interglacial , 126 ky BP , to 42 ky BP ( from 0 . 3 to 8 . 1 million km2 ) . There was a 0 . 5 million km2 decrease in the most suitable area between 42 ky BP and 30 ky BP periods , and then a 3 . 7 million km2 decrease between 30 ky BP and 21 ky BP ( from 7 . 5 to 3 . 8 million km2 ) . Finally , between 21 ky BP and 6 ky BP , there was a 2 . 9 million km2 decrease . By the 6 ky BP period , only 0 . 8 million km2 of the most suitable climatic conditions , Q1 , remained ( Figure 2 and Figure S3 ) . A large reduction in the available suitable climate conditions for the species is expected to cause a reduction in its distributional range , thus contributing ( Figure 3 ) to a reduction in woolly mammoth population sizes and therefore a potential increase in the extinction risk [21] . This hypothesis is supported , firstly , by our estimation of reduced range area and hence reduced population sizes of woolly mammoths through time ( see Materials and Methods ) . A marked reduction in population size of the woolly mammoths is evidenced for the Holocene ( 6 ky BP ) , whatever the woolly mammoth population density value selected ( Figure 4A ) . Secondly , the assumption is supported by the results of a model of hunting intensity ( HI; the number of woolly mammoths required to be killed per person per year in order to drive the species to extinction; see Materials and Methods ) . Irrespective of the cull rate used ( CR; the percentage of the mammoth population that must be killed to drive the species to extinction ) , HI clearly varies through time; the number of woolly mammoths that need to be killed per person per year in order to drive the species into extinction is fairly similar for the 42 ky BP and the 30 ky BP periods , starts to decrease by the 21 ky BP period , and becomes very low in the 6 ky BP period ( Figure 4B–4E and Table S2 ) . According to our analyses , even a high density ( 4 individuals/km2 ) and vigorous ( CR = 2 . 7%; see Materials and Methods ) woolly mammoth population would have been driven to extinction with an HI of 0 . 37 individuals killed per person per year in the 6 ky period . In other words , for these optimistic parameters , one woolly mammoth killed every three years by each human being inhabiting its distribution range would be sufficient to lead the species to extinction . With a low density ( 0 . 1 individuals/km2 ) and suboptimal woolly mammoth population ( CR = 0 . 35%; see Materials and Methods ) , the HI value drops down to 0 . 0049 woolly mammoths killed per person per year; this is roughly one mammoth killed by each person every 200 years . These results support the view that the synergy between the collapse of suitable climatic conditions for the woolly mammoths and northward increase in human population densities during the Holocene set the place and time of the woolly mammoth's extinction . The last nonislands records of the woolly mammoth in the Holocene [22] ( dated after 11 ky BP ) were found around the Tamyr Peninsula , Bikada and Nizhnaya Taimyra rivers , and the Pronchishchev Coast ( Figure 3 ) , coinciding with areas classified by our models as highly suitable for woolly mammoths at 6 ky BP ( scores of MD below 0 . 52; Figure S4 ) . The youngest remains of woolly mammoth found on Wrangel Island are located within the less suitable ( Q4 ) regions ( MD score of 6 . 03 ) . The quality of our projections is further supported by the high spatial agreement between our climate suitability model for the woolly mammoth and the line delimiting the forest [23] and open tundra habitats in the 6 ky BP period ( Figure S4 and Protocol S1 ) . This correspondence provides an independent evaluation [24] of the accuracy with which our climate envelope models infer the environmental conditions that would have affected the survival of woolly mammoths in Eurasia . Theories about species extinctions rely on two different paradigms [25] , which consider either the factors contributing to the general decline of species before their populations become rare—the declining-species paradigm [26 , 27]—or the genetic and demographic factors promoting the extinction of small populations—the small-population paradigm [28] . Most debate about the extinction of the woolly mammoth has focused on trying to separate the contributions of humans [29 , 30] and environmental changes [7–9] toward the extinction of the species . Our results support both perspectives . We suggest that the final extinction of the mammoth might have been the result of the combined effects of climate change and human impacts involving both extinction paradigms within the common framework of metapopulation dynamics [31] . By quantifying the magnitude of the impacts of climate change on woolly mammoth distributions for different periods of time , we show that climate change posed serious challenges for the survival of the species and those areas with suitable climate conditions for the woolly mammoth became severely reduced at 6 ky BP . In the absence of human hunting , however , mammoth populations might have been able to survive in small pockets of suitable habitat and use suboptimal habitats outside the core of their climate envelope , as must have happened at 126 ky BP ( Figure 3 ) . Our analyses suggest that the humans applied the coup de grâce and that size of the suitable climatic area available in the mid-Holocene was too small to host populations able to withstand increased human hunting pressure . Our envelope model for the 6 ky BP period also projected the existence of highly suitable conditions for the occurrence of woolly mammoths outside the High Arctic Siberia ( Figure S4 ) in places where no records have been found , such as in the Ob River basin ( 60 °N – 75 °E ) or southward just within Mongolia ( 49 °N – 95 °E ) . The predicted suitable conditions for woolly mammoths in Mongolia , for example , coincide with the Uvs Nuur Basin , a UNESCO World Heritage Centre , that currently represents one of the best-preserved natural steppe landscapes of Eurasia . If these areas had the potential to host core populations , understanding what happened there would enlighten our knowledge about the last days of the mammoth . To contribute to this debate , new surveys in these areas should be undertaken to ( a ) determine whether populations remained there during mid-Holocene and ( b ) examine why woolly mammoths disappeared and were excluded from these regions . Our results suggest that climate change and human impacts progressively cornered the mammoth in the northernmost land masses of Arctic Siberia and some arctic islands , leaving them with nowhere to run away from extinction .
Records of presence for woolly mammoths were obtained from printed sources and public online databases ( Dataset S1 ) . Two different types of radiometrically dated occurrences were accepted: directly dated mammoth fossils and dates obtained from other materials in the mammoth bearing layer . In direct datings , it is assumed that different ages for the same locality represent different individuals that died at different times ( unless evidence exists to the contrary ) . In indirect datings , when several dates were available for a single layer , we computed an age interval for the layer , taking the upper and lower confidence limits of the oldest and youngest dates , respectively , and eliminating occurrences with incoherent or widely varying age estimates . Radiocarbon dates ( uncalibrate 14C dates ) were calibrated into calendar years ( including 95% confidence intervals ) using the CalPal 2005 SFCP calibration curve [32] . We assumed a 6 ky time interval ( i . e . , 3 ky above and below interval date ) as an arbitrary temporal window . Woolly mammoth occurrences for each time interval were defined as those having their calibrated 95% confidence intervals within these time intervals ( 42 ± 3 ky BP , 30 ± 3 ky BP , 21 ± 3 ky BP , and 6 ± 3 ky BP , respectively ) , resulting in 270 records being included . We estimated the climatic conditions for the locations with woolly mammoth records from the global climate models ( GCMs ) outputs . We estimated the climatic conditions for the locations with woolly mammoth records from GCM simulations . Palaeoclimatic simulations were performed with the GENESIS 2 GCM [33] . Five simulations were used: one for the Eemian ( ∼126 ky BP ) , two for Oxygen Isotope Stage 3 ( OIS 3 ) , one for the Last Glacial Maximum ( LGM; ∼21 ky BP ) , and one for the mid-Holocene ( ∼6 ky BP ) . The OIS 3 simulations represent the warmer middle part ( ∼42 ky BP ) , and colder later part ( ∼30 ky BP ) of Stage 3 . Modelled annual averaged temperatures for the OIS 3 warm simulation are approximately 1–2 °C warmer than the OIS 3 cold simulation over Europe and 0 . 5–1 °C warmer over most of Asia . Carbon dioxide levels were specified at 345 ppm for the Eemian simulation [34] , 200 ppm for the OIS 3 and LGM simulations [35] , and 280 ppm for the mid-Holocene simulation [36] . Sea surface temperatures ( SSTs ) for the OIS 3 and Last Glacial Maximum simulations were taken primarily from CLIMAP [37] , with modifications from GLAMAP-2000 and other sources [38] . SSTs for the mid-Holocene simulation were prescribed at present-day values [39] . Ice sheets for the OIS 3 and LGM simulations followed the ICE-4G [40] and other reconstructions [38 , 41] , while present-day ice sheets were used for the mid-Holocene simulation [36 , 38] . The Eemian interglacial simulation uses a mixed-layer slab ocean with dynamic sea ice [34] . Simulated Eemian temperatures in Eurasia are warmer than present-day and show reasonable agreement with temperatures inferred from pollen and plant macrofossils [42] . In all cases , insolation was calculated using orbital parameters [43 , 44] . The Eemian simulation used prescribed vegetation , the OIS 3 and LGM simulations were interactively coupled to the BIOME4 vegetation model [45] , and the mid-Holocene simulation was interactively coupled to the EVE vegetation model [46] . All simulations were spun up to equilibrium . Results are 10-y averages . Temperatures are in °C and precipitation is in mm per year . We used a Kruskal-Wallis test , a nonparametric alternative to one-way ANOVA , to test for differences between the climate conditions occupied by the species at 42 ky BP , 30 ky BP , and 21 ky BP periods; p > 0 . 05 was taken to indicate that climate conditions do not differ significantly between time periods . To avoid inflation of p-values due to spatial autocorrelation in the Kruskal-Wallis test , we randomly filtered out 80% of the cases ( 216 of 270 cases were removed ) . This filtering process was only applied to the Kruskal-Wallis test . To define the climatic niche of the mammoth , we used all of the available 270 records . We used MD to model the ecological niche of the woolly mammoths . We repeated the same analysis filtering out 129 of 270 cases . The MD technique relies on a multivariate mean and a covariance matrix , and performs an oblique positioning of an elliptical envelope within a multidimensional climatic space . Such an envelope is defined by combinations of climatic variables with equal MD to a vector of average climatic conditions , defined as the mean of all the observations available for the target species . MD scores should be interpreted as a similarity index to sites where the species has been recorded . Mathematically , the MD is defined as: where m is the mean vector and C is the covariance matrix of S . The rows ( vectors ) of S stand for observations of fossil presence of woolly mammoths and the columns for climatic indices . S , therefore , represents the climatic conditions from grid cells with a fossil presence of woolly mammoths . The T superscript denotes the transpose operator . The vector m represents the average climatic conditions from grid cells with a fossil presence of woolly mammoths , and x is a vector indicating climatic conditions of a particular grid cell with a fossil presence of woolly mammoth . We performed a bootstrap with n = 1 , 000 resampling runs to assess the stability of model projections using the boot library in R . Through resampling with replacement of the rows of observations , the bootstrap allows us to estimate m and C , and then subtract the estimate of accuracy from the initial real measure to obtain a corrected estimate . Bootstrapping shows that corrected and estimated values of m and C were similar ( Figure S5 and Figure S6 ) . Finally , we divided MD scores into quartiles ( Q1 , Q2 , Q3 and Q4 , with increasing MD scores , i . e . , decreasing climatic suitability ) , and mapped the potential range of woolly mammoths for each quartile during each period , projecting it also to the 126 ky BP and 6 ky BP intervals . Previous studies [47 , 48] , on 192 plant species in Israel and 71 plant species in southern Africa , using the MD approach , only qualified as potentially suitable those areas with MD scores below 4 and 2 . 5 , respectively . The aim of the HI model is to estimate the hunting intensity by anatomically modern humans ( AMH ) necessary to drive the whole woolly mammoth population of Eurasia to extinction . HI is thus defined as the number of woolly mammoths killed per person per year . Let HIt be the hunting intensity necessary to drive woolly mammoths to extinction at time interval t: where CR is the cull rate , defined as the percentage of the woolly mammoth population that must be killed to drive it to extinction . CR was based on the computer-based simulation of mammoth population dynamics and exploitation in constant , fluctuating and deteriorating environments developed by Mithen [14] . According to Mithen's simulations , a cull greater than 2 . 7% of the total number of individuals may drive to extinction a vigorous mammoth population in a constant environment , and a proportion as low as 0 . 35% will do for a less vigorous , suboptimal , population . Thus , we computed our model using both values for CR . It is important to note that in Mithen's model , this CR represents the killing of animals which would have otherwise survived until the following year . Thus , it does not include the death of old or weak animals which would have died from other causes . Nmt and Nht are the total population size ( number of individuals ) of , respectively , woolly mammoths and humans at each time interval t ( 42 ky BP , 30 ky BP , 21 ky BP , and 6 ky BP periods ) , and are obtained from: where Dmt is mammoth population density ( individuals/km2 ) , Dht is human population density ( individuals/km2 ) , and Amt is the area of coexistence of woolly mammoths and AMH for time interval t , since AMH should coexist with mammoths to hunt them ( therefore , Amt is removed from the equation because it represents the same area for woolly mammoths and humans ) . Thus , equation ( 2 ) may be written as: and HIt represents the hunting intensity necessary to drive the woolly mammoth population to extinction inside their area of coexistence with AMH for time interval t . A range of woolly mammoth population densities ( Dm ) were estimated based on population densities of modern elephants and allometric body mass relationships . African elephants ( Loxodonta africana ) are considered a good analogue for woolly mammoths [49] , and their reported average population density is 1 . 09 individuals/km2 [50] , although it varies within the range of 0 . 25 to 5 individuals/km2 [51] . Population density for Asian elephants , Elephas maximus , the other extant proboscidean species , ranges from 0 . 12 to 1 . 0 individuals/km2 [51 , 52] . Since both are tropical species and population density is known to depend on primary productivity , which decreases with latitude [53 , 54] , these values likely overestimate actual mammoth densities . Thus , we estimated Dm from woolly mammoth body mass using the allometric equation computed by [55] for temperate herbivorous mammals , obtaining maximum and minimum values of 0 . 74 and 1 . 79 individuals/km2 respectively . However , [14] estimated Siberian woolly mammoth population densities in the range 0 . 038 to 0 . 23 individuals/km2 . Due to this high variation of estimates , we decided to use 0 . 1 and 4 individuals/km2 as a conservative and an optimistic estimates of the possible range of mammoth population densities . For simplicity , these estimates of Dm were considered to be time-independent and homogeneous across the entire area that is environmentally suitable for woolly mammoths . AMH population densities ( Dht ) for the four time intervals considered in our analysis were obtained from [56] . We used three different estimates of Dht ( Table S3 ) . Their minimum Dht for each cultural period , for example—assuming that the Aurignacian estimate represents our 42 ky BP period , the Gravettian estimate represents our 30 ky BP interval , the Glacial maximum represents our 21 ky BP period , and the Late Glacial represents our 6 ky BP period—were 0 . 066 individuals/km2 , 0 . 072 individuals/km2 , 0 . 101 individuals/km2 , and 0 . 285 individuals/km2 , respectively ( see Table S3 for average and maximum population densities ) . The likely effect of the temporal discrepancies between our data and the periods defined in [56] would be the underestimation of human population density for the 6 ky BP period and the slight overestimation for 42 ky BP and 30 ky BP . Also , we consider human population density to be homogeneous across all the area that was environmentally suitable for the woolly mammoth , which is an optimistic estimate of the abilities of ancient human populations to survive at high latitudes during the upper Pleistocene . Indeed , the first recorded human presence above 60 °N dated from 11 ky BP [57] . As a result , our model will tend to underestimate the HIt value for 42 ky BP , 30 ky BP , and 21 ky BP , and to overestimate its value for 6 ky BP . Total Eurasian woolly mammoth population sizes for the five time intervals ( 126 ky BP , 42 ky BP , 30 ky BP , 21 ky BP , and 6 ky BP ) have been estimated assuming ( a ) that the entire environmentally suitable area is occupied ( Q1 + Q2 + Q3 + Q4 ) ; and ( b ) that woolly mammoth population density Dmt is homogeneous throughout the area and is comprised between 0 . 1 and 4 individuals/km2 . Since both assumptions are simplistic , the obtained numbers might overestimate the actual metapopulation size , although they are useful to represent the general trend through time . A marked reduction in population size is seen for the Holocene ( 6 ky BP ) , whatever the woolly mammoth population density value selected .
|
What caused the woolly mammoth's extinction ? Climate warming in the Holocene might have driven the extinction of this cold-adapted species , yet the species had survived previous warming periods , suggesting that the more-plausible cause was human expansion . Testing these competing hypotheses has been hampered by the difficulty in generating quantitative estimates of the relationship between the mammoth's contraction and the climatic and/or human-induced drivers of extinction . In this study , we combined paleo-climate simulations , climate envelope models ( which describe the climate associated with the known distribution of a species—its envelope—and estimate that envelope's position under different climate change scenarios ) , and a population model that includes an explicit treatment of woolly mammoth–human interactions to measure the extent to which climate changes , increased human pressures , or a combination of both factors might have been responsible . Results show a dramatic decline in suitable climate conditions for the mammoth between the Late Pleistocene and the Holocene , with hospitable areas in the mid-Holocene being restricted mainly to Arctic Siberia , where the latest records of woolly mammoths in continental Asia have been found . The population model results also support the view that the collapse of the climatically suitable area caused a significant drop in mammoth population size , making the animals more vulnerable to increasing hunting pressure from expanding human populations . The coincidence of the collapse of climatically suitable areas and the increase in anthropogenic impacts in the Holocene are most likely to have been the “coup de grâce , ” which set the place and time for the extinction of the woolly mammoth .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"ecology"
] |
2008
|
Climate Change, Humans, and the Extinction of the Woolly Mammoth
|
The formation of virus movement protein ( MP ) -containing punctate structures on the cortical endoplasmic reticulum is required for efficient intercellular movement of Red clover necrotic mosaic virus ( RCNMV ) , a bipartite positive-strand RNA plant virus . We found that these cortical punctate structures constitute a viral replication complex ( VRC ) in addition to the previously reported aggregate structures that formed adjacent to the nucleus . We identified host proteins that interacted with RCNMV MP in virus-infected Nicotiana benthamiana leaves using a tandem affinity purification method followed by mass spectrometry . One of these host proteins was glyceraldehyde 3-phosphate dehydrogenase-A ( NbGAPDH-A ) , which is a component of the Calvin-Benson cycle in chloroplasts . Virus-induced gene silencing of NbGAPDH-A reduced RCNMV multiplication in the inoculated leaves , but not in the single cells , thereby suggesting that GAPDH-A plays a positive role in cell-to-cell movement of RCNMV . The fusion protein of NbGAPDH-A and green fluorescent protein localized exclusively to the chloroplasts . In the presence of RCNMV RNA1 , however , the protein localized to the cortical VRC as well as the chloroplasts . Bimolecular fluorescence complementation assay and GST pulldown assay confirmed in vivo and in vitro interactions , respectively , between the MP and NbGAPDH-A . Furthermore , gene silencing of NbGAPDH-A inhibited MP localization to the cortical VRC . We discuss the possible roles of NbGAPDH-A in the RCNMV movement process .
Eukaryotic positive-strand RNA viruses replicate their genomes using membrane-bound virus replication complexes ( VRC ) , which contain viral replicase proteins , viral RNA templates , and host-factor proteins [1]–[5] . Viral replicase proteins modify the host intracellular membrane morphology , including swelling , invagination , and the formation of spherules . Thus , VRC formation is accompanied by the remodeling of intracellular membranes [3] . Studies of the subcellular localization of viral proteins using immunoelectron microscopy or confocal laser scanning microscopy ( CLSM ) have shown that the movement proteins ( MPs ) of several plant viruses colocalize with the viral replicase protein [6]–[10] , thereby suggesting that MPs are also components of VRCs . However , the MPs that localize to VRCs are not likely to be involved in the replication of viral genomic RNA because mutant viruses that do not encode functional MP can accumulate viral genomic RNA similar to that of the wild type virus with functional MP in infected protoplasts [11] , [12] . MPs play central roles in the cell-to-cell and systemic movement of plant viruses , and they have been investigated intensively to determine their biochemical characteristics; their subcellular localization , including the cellular pathways that target them to the plasmodesmata ( PD ) , the cytoplasmic channels connecting plant cells; and their interactions with host membranes or proteins [13]–[17] . The early process of VRC formation has been well characterized in Tobacco mosaic virus ( TMV ) , which is the type member of the well-studied genus Tobamovirus [18] . TMV virions that enter host plant cells are partially uncoated , and the exposed 5′ cap structures are assumed to be recruited by unknown host factors to form small particles that contain viral RNA on the cortical endoplasmic reticulum ( ER ) , before moving along the ER and actin filaments . The replication cycles starts after translation of the replicase proteins and the VRC increases in size . The punctate-like TMV VRC forms on the cortical ER and moves along actin filaments [18]–[20] or microtubules [21] , [22] . The intracellular movement of VRC on actin filaments requires motor proteins such as myosin , and it is considered to be necessary for the targeting of viral genomic RNA to the PD [23]–[25] . The roles of TMV MP in the VRC are unknown . The “126-bodies” , which comprise the fusion protein of TMV 126-kDa replicase component protein and green fluorescent protein ( GFP ) , can move along actin filaments without MP [20] . Red clover necrotic mosaic virus ( RCNMV ) is a positive-strand RNA virus with a bipartite genome that belongs to the genus Dianthovirus , in the family Tombusviridae [26] . Genomic RNA1 encodes p27 auxiliary replication protein , p88 RNA-dependent RNA polymerase ( RdRp ) , and coat protein ( CP ) , while RNA2 encodes MP ( Figure S1A ) . p27 and p88 induce the production of an aggregate structure from ER membrane and they form the 480-kDa replication complex , which is a key enzyme complex for virus replication , via interactions with host chaperone proteins such as heat shock protein ( HSP ) 70 and HSP90 , and membrane traffic-associated proteins such as Arf1 and Sar1 [27]–[31] . Using a bimolecular fluorescence complementation ( BiFC ) assay , these four host factors were shown to interact directly with p27 in the large aggregate adjacent to the nucleus in Nicotiana benthamiana cells [30] , [31] . However , when the p27-GFP fusion protein was expressed with p88 and RNA2 , it formed small punctate structures on the cortical ER and later formed large aggregates adjacent to the nucleus [32] . These results suggest that RCNMV VRC forms small punctate structures on the cortical ER , which then change their subcellular localization to form a large aggregate adjacent to the nucleus . RCNMV MP belongs to the 30K superfamily and it is required for viral cell-to-cell and systemic movement [33] , [34] . RCNMV is considered to pass through PD in the form of a viral RNA-MP complex because CP is dispensable for viral cell-to-cell movement [34] and MP also has the ability to bind single-stranded nucleic acids [35] . Microinjected RCNMV MP can increase the size-exclusion limit of PD and enable the transport of coinjected viral RNA into neighbor cells [36] . Alanine-scanning mutant analysis was used to determine the functional domains of the MP that bind RNA and that target it to PD , both of which are required for viral cell-to-cell movement [37] , [38] . However , the cellular pathway that allows MP and/or MP-viral RNA complexes to target PD is unknown . Previously , we reported the subcellular localization of the fusion protein of RCNMV MP and GFP ( MP-GFP ) in N . benthamiana [10] . In addition to PD localization , MP-GFP expressed by a recombinant virus formed punctate structures with p27 on the cortical ER . Transiently expressed MP-GFP also localized to punctate structures on the cortical ER , which was associated with the replication of RNA1 , but not with that of RNA2 . These results suggest that MP is recruited to the cortical ER by the viral replicase complexes formed with RNA1 . To demonstrate the importance of cortical punctate structures containing MP , we conducted a deletion analysis of MP and showed that 70 C-terminus amino acids are required for both cortical punctate structure formation and viral cell-to-cell movement [39] . Based on these results , we hypothesized that the recruitment of MP by the viral replication complex might help MP to acquire viral genomic RNA1 that does not encode MP , thereby leading to the efficient cell-to-cell movement of RNA1 . To further investigate the mechanism that facilitates the movement of RCNMV , we performed tandem affinity purification of MP from virus-infected N . benthamiana leaves and analyzed the co-purified host proteins , by mass spectrometry . One of these host proteins was glyceraldehyde 3-phosphate dehydrogenase subunit A ( GAPDH-A ) . GAPDHs are ubiquitous enzymes involved in glycolysis and gluconeogenesis , and GAPDH-A is a component of the Calvin-Benson cycle of photosynthetic organisms [40] . GAPDH-A and another subunit , GAPDH-B , are both located in the chloroplast in plants and algae [41] . Thus , we isolated the full-length cDNA of N . benthamiana GAPDH-A ( NbGAPDH-A ) and investigated its involvement in RCNMV multiplication . Our results demonstrate that NbGAPDH-A is involved in virus cell-to-cell movement by influencing MP localization to the VRC . We discuss the possible mechanism that underlies this process .
Previously , we reported that RCNMV MP colocalized with the viral replicase protein p27 to the punctate structures on the cortical ER in virus-infected N . benthamiana cells during the early stage of infection . Later , most of these cortical punctates disappeared and a large aggregate was formed adjacent to the nucleus in epidermal cells [10] , [32] . These aggregates contained newly synthesized viral RNAs and the host-factor proteins essential for replication , and they were shown to be the sites of RCNMV RNA replication [30] , [31] . However , no evidence of viral RNA replication in the cortical punctates has been reported . Thus , we detected double-stranded RNA ( dsRNA ) , the replication intermediates of positive-stranded RNA viruses , by immunostaining using the antibody against double-stranded RNA ( J2 antibody ) in N . benthamiana protoplasts . J2 antibody has been widely used to detect the replication sites of animal and plant RNA viruses , and cellular RNAs such as ribosomal RNA are below the limit of detection [43]–[45] . N . benthamiana protoplasts were inoculated with in vitro transcripts of recombinant RCNMV , which expressed the fusion protein of MP and a red FP , mCherry ( MP-mCherry , Figure S1B ) . Using CLSM , MP-mCherry and dsRNA were detected as overlapping small punctate signals near the surfaces of protoplasts at 16 h post inoculation ( hpi ) ( Figure 1 , left 2 rows of panels ) . At 24 hpi , most of these small punctates disappeared and large aggregates were detected adjacent to the nucleus , which contained both MP and dsRNA ( Figure 1 , center 2 rows of panels ) , thereby confirming the results reported previously [30] , [31] . No fluorescent signals for dsRNA were detected in mock-inoculated protoplasts ( Figure 1 , right panels ) . These results indicate that both the cortical punctates formed during an early stage of RCNMV infection and the aggregates formed adjacent to the nucleus during the later stage of infection are the sites of RCNMV RNA replication . Subsequently , we refer to the small punctate-like structures that contain the MP in the cortical region as ‘cortical VRC . ’ To identify the host proteins that interact with RCNMV MP , we performed two-step affinity purification of MP fused to a tandem affinity purification tag sequence . The tagged MP was functional because it supported virus cell-to-cell and systemic movement with the same efficiency as the native MP in N . benthamiana plants ( Figure S2 ) . Binary vector plasmid pBICR12/MP-TAP ( Figure S1D ) , and pBICR12 ( Figure S1C ) as the negative control , were infiltrated via Agrobacterium into N . benthamiana . The tandem affinity purified fraction prepared from pBICR12/MP-TAP-infiltrated leaves contained several silver-stained bands , which were not detected in the negative control ( Figure 2A ) . The clear silver-stained band that represented the MP-FLAG was not detected in the MP-TAP lane for unknown reason . Considering its size ( 35 . 6 kDa ) , the band is probably masked in the broad range of the stained area below the 42 kDa marker . Actually nano-liquid chromatography-tandem mass spectrometry ( LC/MS/MS ) analysis demonstrated that a piece of wide gel cut out from MP-TAP lane ( Figure 2A , red arrow ) contained the MP ( Table S1 ) . MP-FLAG was also detected by Western blotting analysis in the tandem affinity purified fraction prepared from pBICR12/MP-TAP-infiltrated leaves but not from the negative control leaves ( Figure 2B ) . These silver-stained bands in the MP-TAP lane , and the similar regions of the gel for the negative control lane were excised and subjected to in-gel trypsin digestions and LC/MS/MS analyses . We identified RCNMV MP and several host proteins from the stained bands , and these proteins were not detected from the negative control gels . Among these , we focused on GAPDH-A . A partial GAPDH-A sequence was amplified by RT-PCR using the total RNA of N . benthamiana , where the primer designs were based on N . tabacum GAPDH-A . The full-length cDNA of GAPDH-A was cloned according to 5′ and 3′ RACE methods , which we refer to as NbGAPDH-A ( accession number AB937979 ) . The deduced amino acid sequence of NbGAPDH-A was almost identical to the reported partial GAPDH-A of N . tabacum ( 96 . 9% shared identity , except for nine N terminal amino acids ) and very similar to that of Arabidopsis thaliana , except for 60 N terminal amino acids ( Figure 2C ) . To investigate the possible involvement of NbGAPDH-A in RCNMV multiplication , we downregulated the gene using the Apple latent spherical virus ( ALSV ) vector [46] . A plasmid that expressed wild type ALSV , or that expressed the recombinant ALSV containing 294 nucleotides of NbGAPDH-A ( ALSV/gsGAP vector ) , was mobilized into Agrobacterium and the bacterium was used to inoculate young N . benthamiana plants . The accumulation level of NbGAPDH-A mRNA in the newly developed leaves was determined 2–3 weeks later by real time RT-PCR . NbGAPDH-A was silenced effectively in ALSV/gsGAP vector-infected plants; the mRNA level of NbGAPDH-A was reduced to 3% of that in the wild type ALSV-infected plants ( Figure 3A ) . This result coincided with that by semi-quantitative RT-PCR in which mRNA level in ALSV/gsGAP vector-infected plants was about 1/32 of that in wild type ALSV-infected plants ( Figure S3 ) . No symptoms or growth inhibition were detected in the NbGAPDH-A-silenced plants and wild type ALSV-infected plants ( Figure S4 , see Discussion ) . Hereafter , all of the ALSV/gsGAP-infected plants and the protoplasts prepared from those plants were tested by real time or semi-quantitative RT-PCR to confirm the NbGAPDH-A gene was silenced . We then subjected the ALSV- and ALSV/gsGAP-infected plants to challenge via the mechanical inoculation of in vitro transcripts of the recombinant RCNMV containing the GFP gene ( RCNMV-GFP; Figure S1E ) . The percentage of fluorescent foci with multiple cells in the ALSV/gsGAP-infected plants was about 1/3 of that in the ALSV-infected plants at 20 hpi ( Figure 3B ) . The result suggests that RCNMV multiplication was negatively affected by the silencing of NbGAPDH-A . In order to evaluate the effect of the gene silencing on RCNMV multiplication more objectively , we further performed challenge inoculation with pBICR1sG2 ( Figure S1F ) , which expressed RCNMV-GFP , via Agrobacterium infiltration , and the multiplication level of the recombinant virus was estimated by western blot analysis for GFP . The level of GFP accumulation at 35 hpi in the leaves of ALSV/gsGAP-infected plants was approximately 20% of that in the leaves of the ALSV-infected plants ( Figure 3C ) . The majority of the fluorescent foci were comprised of more than 10 cells in the latter plants , whereas such a wide spread of fluorescence was barely detected in the former plants ( Figure 3C , lower panels ) . At 48 hpi , most of the fluorescent foci in the ALSV/gsGAP-infected plants became larger and the level of GFP accumulation was about 80% of that in the wild type ALSV-infected plants ( Figure S5 ) . Thus , RCNMV multiplication was impaired in the NbGAPDH-A-silenced N . benthamiana leaves , at an early stage of infection . To investigate whether downregulation of the NbGAPDH-A gene could affect the multiplication of viruses other than RCNMV , we inoculated ALSV- and ALSV/gsGAP-infected N . benthamiana plants with a recombinant Tomato mosaic virus ( ToMV ) , where the CP gene was replaced with the GFP gene . The spread of GFP fluorescence was indistinguishable at 40 and 48 hpi by epifluorescence microscopy and the GFP accumulation level was also similar in both plants ( Figure S6 ) . These results indicate that the NbGAPDH-A gene is not involved in the multiplication of ToMV . To investigate the effect of NbGAPDH-A silencing on RCNMV accumulation at the single cell level , we infiltrated ALSV- , or ALSV/gsGAP-infected plants with Agrobacterium that contained pBICR12fsMP , which expressed movement-deficient RCNMV RNAs ( Fig . S1G ) [10] . At 26 and 43 hpi , similar amounts of positive-stranded viral RNAs accumulated ( Figure 4A ) , thereby suggesting that RCNMV multiplied at similar levels in the initially infected cells . To further investigate the multiplication levels of RCNMV in single cells , protoplasts were prepared from ALSV- and ALSV/gsGAP-infected plants and inoculated with in vitro transcripts of the recombinant RCNMV , which expressed GFP and MP tagged with HA ( MP-HA ) ( Figure S1H ) . Similar amounts of GFP accumulated in both protoplasts ( Figure 4B , upper panel ) , thereby indicating that gene silencing of NbGAPDH-A did not affect the accumulation of the recombinant virus at the single cell level . We also analyzed the accumulation of MP-HA . As shown in the middle panel of Figure 4B , the levels of MP-HA were similar with either inoculation , which suggests that NbGAPDH-A is not involved in the translational control or stability of MP . Overall , these results suggest that NbGAPDH-A is unlikely to be involved in the replication of RCNMV RNAs and that it is involved in the cell-to-cell movement of RCNMV via its interaction with MP . NbGAPDH-A is assumed to localize to the chloroplasts . However , RCNMV replication occurs in association with the ER membrane and no relationship with the chloroplasts has been reported previously . To investigate the possible interaction between NbGAPDH-A and RCNMV proteins in vivo , we examined the subcellular localization of NbGAPDH-A in the absence or presence of RCNMV factors . When NbGAPDH-A tagged with GFP ( NbGAPDH-A-GFP ) alone was expressed transiently in N . benthamiana leaves via agroinfiltration , the protein localized exclusively to chloroplasts ( Figure 5A , left two panels and Figure 5B , panels 1 and 2 ) . The localization pattern of NbGAPDH-A-GFP was not altered by coexpression with RCNMV MP-mCherry . NbGAPDH-A-GFP signals were detected in the chloroplasts and were never detected in PD ( Figure 5A , right four panels ) . This suggests that the transiently expressed MP does not interact with NbGAPDH-A in vivo . Next to investigate whether the subcellular localization of NbGAPDH-A-GFP could be affected by RCNMV RNA replication , NbGAPDH-A-GFP was coexpressed with RNA1 . The GFP signals were detected in punctate structures that formed near the surface regions of epidermal cells , as well as in chloroplasts ( Figure 5B , panel 3 ) . These cortical signals colocalized with ER marker signals ( Figure 5B , panel 4 ) . Similar cortical punctate signals of NbGAPDH-A-GFP were also detected when it was coexpressed with both RNA1 and RNA2 ( Figure S7 ) , but not with the viral replicase component proteins , p27 and p88 ( Figure 5B , panels 5 and 6 ) . Lack of the cortical punctate signals of NbGAPDH-A-GFP in the latter leaves does not seem to be due to the low level of viral replicase proteins . p27 accumulated efficiently in the latter leaves ( Figure S8 ) . p88 was below the limit of detection in these agroinfiltrated leaves , as described previously [28] , [29] . These results suggested the association between the localization of NbGAPDH-A-GFP to punctates on the cortical ER and the replication of RNA1 . To examine this association is specific to NbGAPDH-A-GFP , we investigated the localization of free GFP or the GFP with chloroplast-targeting signal peptide in the presence of RNA1 . No cortical punctate signals were detected in the leaves expressing these GFP proteins with RNA1 ( Figure 6 and Figure S9 ) . These results suggested that NbGAPDH-A was recruited to the cortical punctate structures in association with the replication of RNA1 . The interaction between NbGAPDH-A and RCNMV MP in vivo was confirmed by BiFC assays in N . benthamiana epidermal cells . NbGAPDH-A was fused to the C-terminal half of yellow fluorescent protein ( YFP ) at the C terminus ( NbGAPDH-A-cYFP ) and was expressed with Tomato bushy stunt virus ( TBSV ) silencing suppressor p19 in N . benthamiana via agroinfiltration . Recombinant RCNMV transcripts that expressed the MP fused to the N-terminal half of YFP at the C terminus ( MP-nYFP , Figure S1I ) was mechanically inoculated at 16 h post infiltration . At 28 hpi with the recombinant virus , fluorescence was observed using CLSM . YFP fluorescence was reconstituted in the presence of NbGAPDH-A-cYFP and the MP-nYFP ( Figure 7A , left panel ) . No YFP fluorescence was detected in control experiments ( Figure 7A , center and right panels; Figure S1J ) . With higher magnification , reconstituted YFP signals were observed as punctate structures in the cortical region ( Figure 7B , left panel ) and were also detected in the cell wall ( Figure 7B , right panel , see Discussion ) . Reconstituted YFP signals in the cortical punctates were confirmed to overlap with ER marker signals ( Figure 7C ) . These results , together with the localization results of the MP expressed from recombinant virus to the cortical VRC ( [10] and Figure 1 ) show that the reconstituted YFP signals are on the cortical VRC . Subcellular localization results ( Figure 5 ) and BiFC results ( Figure 7 ) suggest that NbGAPDH-A interacts with both viral replicase protein ( s ) and MP in association with the replication of viral RNA . To confirm the direct interaction between NbGAPDH-A and RCNMV MP , or NbGAPDH-A and p27 , we performed GST pulldown assays in vitro . Bacterially expressed and purified NbGAPDH-A with an N-terminal 6× His tag and C-terminal myc tag ( His-GAP-myc ) was incubated with N-terminally GST- and C-terminally HA-tagged MP ( GST-MP-HA ) , N-terminally GST-fused p27 ( GST-p27 ) , or GST , which were captured on glutathione-bound beads . Immunoblot analyses using an anti-myc antibody demonstrated that His-GAP-myc was pulled down by GST-MP-HA and GST-p27 , but not by GST ( Figure 8 ) , thereby indicating that His-GAP-myc binds to both MP and p27 in vitro . To rule out the possibility that coprecipitation in the GST-pulldown experiment was mediated by interaction with any unspecific RNA that bound to MP or p27 , we included RNaseA to the reaction . Addition of 50 µg/ml of RNaseA did not affect the result ( Figure S10 ) , suggesting that NbGAPDH-A interacted with the MP and p27 directly . To address the possible effects of NbGAPDH-A on the subcellular localization of MP , we investigated whether MP targeting to the PD or to the cortical VRC was affected by the silencing of NbGAPDH-A . Our previous results showed that the transient expression of MP-GFP in N . benthamiana cells resulted in its localization exclusively to the PD , while infection with recombinant RCNMV RNAs that encoded MP-GFP resulted in the formation of cortical VRC and localization to the VRC as well as to the PD [10] . Agroinfiltration of pBICRMsG that expressed MP-GFP fusion protein [10] into ALSV/gsGAP-infected N . benthamiana plants resulted in the same localization to PD that was found in ALSV-infected plants ( Figure S11 ) . This showed that NbGAPDH-A had no effect on the intracellular transportation of RCNMV MP to the PD . Next , we investigated the effects of NbGAPDH-A-silencing on the localization of MP to the cortical VRC . The pBICR1/MsG2fsMP plasmid , which expressed recombinant RCNMV RNAs that encoded MP-GFP ( Figure S1K ) [10] , was agroinfiltrated into ALSV- or ALSV/gsGAP-infected plants . During the early stage of infection at 38 h post infiltration , cortical fluorescent punctates were detected in most of fluorescent mesophyll and epidermal cells in ALSV-infected plants ( Figure 9A , left planels ) , whereas the majority of the fluorescence exhibited a dispersed cytoplasmic pattern in mesophyll cells of ALSV/gsGAP-infected plants ( Figure 9A , upper right panels ) . In epidermal cells , cortical punctates were detected rarely and the PD localization of MP-GFP was detected in ALSV/gsGAP-infected plants ( Figure 9A , lower right panel ) . The ratio of fluorescent cells with cortical punctates was 7 . 4 times higher in ALSV-infected plants compared with ALSV/gsGAP-infected plants ( Figure 9B ) . At 44 h post infiltration with the recombinant RCNMV , the ratio of fluorescent cells with cortical punctates increased to 41 . 0% in ALSV/gsGAP-infected plants , although the number of cortical punctates in a single fluorescent cell was lower compared to that in ALSV-infected plants ( Figure S12 ) . The negative effect of NbGAPDH-A-silencing on the localization of MP to the cortical VRC was confirmed using protoplasts . At 12 h post infection with the transcripts of pUCR1-MsG and pRNA2fsMP ( Figure S1L ) [10] , cortical fluorescent punctates with MP-GFP were detected in the protoplasts prepared from ALSV-infected plants , whereas they were barely detectable in the protoplasts prepared from ALSV/gsGAP-infected plants ( Figure 9C ) . Probably MP-GFP molecules that were diffused in the cytoplasm could not be detected by CLSM . Despite the reduced fluorescence , MP-GFP accumulated at similar levels in both protoplasts ( Figure 9D ) , thereby showing that NbGAPDH-A silencing did not affect the expression , or stability of MP-GFP . These results suggest that NbGAPDH-A is involved in the recruitment of RCNMV MP to the cortical VRC , or that it may stabilize the interaction between MP and VRC . Finally , these protoplasts were subjected to immunofluorescent staining of dsRNA for the detection of VRC . In the protoplasts prepared from ALSV/gsGAP-infected plants , MP-GFP was rarely detected . However , cortical punctate-like structures of dsRNA were detected in these cells , as well as in the protoplasts prepared from ALSV-infected plants ( Figure 10 ) . The accumulation level of p27 protein was also similar in both protoplasts ( Figure S13 ) . These results suggest that NbGAPDH-A does not affect the formation of cortical VRC and that it is associated with the recruitment of RCNMV MP to the cortical VRC .
Replication and movement processes are assumed to be linked to facilitate successful infection by plant viruses . In addition to the temporal regulation of MP expression [47]–[49] , spatial regulation is required to allow MPs to encounter the viral genomes . Thus , localized MP synthesis at the VRC , or specific MP recruitment to the VRC , would facilitate efficient and specific virus cell-to-cell movement [50] . In the present study , we showed that MP-containing cortical punctate structures formed during the early stage of RCNMV infection and large aggregates assembled adjacent to the nucleus during the late stage of infection in N . benthamiana cells , and both were sites of viral RNA replication ( Figure 1 ) . These results suggest that RCNMV VRC changes its location from cortical to perinuclear ER-containing structures , while the VRC also increases in size , as the infection stage proceeds . A division of roles between the VRCs formed during the early and late stages of infection has recently been proposed for Potato virus X ( PVX ) . The early VRCs of PVX are formed in a membranous structure called the ‘cap’ at the orifice of PD , and triple gene block ( TGB ) -type MPs that accumulate at the cap and PD pore play roles in trafficking the replicated viral genomic RNA via the PD [51] . Furthermore , the X-body formed during the late stage of infection compartmentalizes the TGB1 protein and prevents it from having roles in translational activation , which could lead to the destabilization of PVX virions , while the VRCs that surround the TGB1 core maximize the replication of the viral RNA and the production of virions [52] . The cap structure at the PD orifice and MP compartmentalization in the X-body were not detected in RCNMV-infected cells [10] , but it is likely that the two types of RCNMV VRCs have distinct roles . Given that RCNMV cell-to-cell movement occurs before the large aggregate-type VRCs form in virus-infected N . benthamiana epidermal cells [10] , [39 , unpublished results] , it is probable that only the cortical VRC contributes to virus cell-to-cell movement whereas the large aggregate-type VRC ( X-body ) might maximize the production of progeny virions . Our previous studies showed that the host proteins that contribute to the replication of RCNMV RNAs colocalized with p27 in the perinuclear large aggregates [30] , [31] , rather than the cortical VRC . It is possible that the modes of VRC formation differ between the cortical VRC and the perinuclear large aggregates . Further studies using specific antibodies against the host factors associated with the VRC are required to answer this question . We identified NbGAPDH-A as an interacting partner for RCNMV MP ( Figure 2 ) . Although VIGS of NbGAPDH-A using ALSV vector that contained 294 bases of the gene fragment reduced the accumulation of the mRNA to 3% of that in the empty ALSV infected plants ( Figure 3A ) , the silencing had no effect on plant growth ( Figure S4 ) . This result contradicted a previous report where transgenic tobacco ( Nicotiana tabacum ) plants with silenced GAPDH-A exhibited severe growth inhibition compared with the wild type plants [53] . Over 1 , 000 bases of the GAPDH-A coding region had been introduced into these transgenic plants to express an antisense RNA that was complementary to GAPDH-A mRNA . The induction of gene silencing using such a long sequence might have affected the expression levels of unidentified GAPDH orthologs , which could have led to growth inhibition . Alternatively , N . tabacum GAPDH-A might have a greater impact on growth than that of N . benthamiana , or transgenic plants in which the gene was silenced had a greater effect on the phenotype than VIGS . Our preliminary results showed that induction of VIGS of NbGAPDH-A by the other widely-used VIGS vector based on Tobacco rattle virus ( TRV ) caused the same symptoms as those by the empty TRV vector ( Figure S14 ) . This supports that VIGS of NbGAPDH-A in N . benthamiana does not cause severe symptoms . We showed that NbGAPDH-A is a host protein that is involved in the cell-to-cell movement of RCNMV ( Figures 3 and 4 ) . In addition to chloroplast localization , the NbGAPDH-A-GFP fusion protein also localized to cortical VRCs . The localization to VRC was associated with viral RNA replication , not the replicase component proteins alone ( Figure 5B ) . Gene silencing of NbGAPDH-A inhibited the targeting of RCNMV MP to cortical VRCs ( Figure 9 ) , but it did not affect the targeting of MP-GFP to the PD ( Figure S11 ) , or the stability of MP ( Figures 4 and 9 ) . Based on the overall results obtained in the present study , we propose that NbGAPDH-A is contained in the VRC without influencing viral RNA replication and is an interstitial agent between RCNMV MP and the VRC . NbGAPDH-A probably plays a role in recruiting MP to the VRC , or stabilizing the interaction between MP and VRC . BiFC assays confirmed the in vivo interaction between NbGAPDH-A and the MP ( Figure 7 ) . BiFC assays also showed that the interaction occurred not only in the cortical VRC , but also in the large aggregates and in the cell wall ( Figure 7B ) . The distribution pattern of the reconstituted YFP signal was quite similar to that of the MP-GFP expressed from the recombinant virus [10] . The signals of the reconstituted YFP observed in the cell wall could be due to the NbGAPDH-A-YFP-MP complexes that had been transported to PD by the function of MP . However , significance of the observed colocalization of NbGAPDH-A and MP in the cell wall is ambiguous . This is because the reconstitution of YFP is irreversible [54] , and because NbGAPDH-A-GFP did not localize in the cell wall when coexpressed with RCNMV RNA1 and RNA2 ( Figure S7 ) . Further study is needed to elucidate the role , if any , of NbGAPDH-A in the cell wall in virus infection . GST pulldown assays confirmed that NbGAPDH-A interacted with both p27 and MP in vitro ( Figure 8 ) . This suggests that NbGAPDH-A may be a bridge between MP and p27 that is a constituent of the VRC . In vivo , however , relocalization of NbGAPDH-A-GFP to the cortical VRC and interaction of NbGAPDH-A-cYFP and MP-nYFP occurred only in association with the viral RNA replication ( Figures 5B and 7 ) . By contrast , coexpression of NbGAPDH-A-GFP and p27 fused with DsRed-monomer in N . benthamiana cells using agroinfiltration resulted in different localization patterns , with the former in chloroplasts and the latter in the ER-containing large aggregate ( [32] and Figure S15 ) . Furthermore , coexpression with MP-mCherry did not affect the localization of NbGAPDH-A-GFP to the chloroplasts ( Figure 5A ) . These results suggest that unidentified factors might be involved in the in vivo interaction between NbGAPDH-A and viral proteins . Three possibilities can be considered , 1 ) Enhancement of the local concentration of viral proteins: RNA1 replicates autonomously , and the replication coupled with the translation of replicase proteins p27 and p88 , followed by the formation of 480 kDa replication complex [30] , [55] , [56] . This replication cycle might increase the local concentration of p27 in or near the VRC on the cortical ER membrane to higher levels than transiently expressed p27 . Such a process might improve the probability that p27 and NbGAPDH-A will encounter in the cortical VRC . In association with this assumption , transiently expressed p27-GFP alone forms a large aggregate ( [32] and Figure S15 ) . Formation of such aggregates might sequester p27 and prevent the interaction with NbGAPDH-A . 2 ) Involvement of unknown host proteins associated with the VRC: The formation of the 480 kDa replication complex of RCNMV requires not only p27 and p88 but also viral RNAs in host cells [30] . The 480 kDa replication complex contains many host proteins that have not been identified yet . It is possible that such unknown proteins are involved in the recruitment of NbGAPDH-A to the VRC or in the stabilization of the interaction in vivo . 3 ) Involvement of viral RNA: Cytoplasmic GAPDH ( GAPDH-C ) has been reported to interact with the cis-acting elements of many RNA viruses , some of which affect the multiplication of viruses ( [57] , [58] and references therein ) . Although chloroplastic and cytoplasmic GAPDHs are assumed to have evolved from different lineages [59] , their amino acid sequence identity is as high as ca 45% in Arabidopsis thaliana ( NCBI Gene ID: 819567 and 822277 ) and Zea mays ( NCBI Gene ID: 542367 and 542368 ) . It is possible that NbGAPDH-A also has an RNA-binding ability and that it is recruited to VRCs by interacting with RCNMV RNAs in vivo , as shown in Hepatitis delta virus-infected cultured cells [60] . These three possibilities may not be mutually exclusive . Further studies on the molecular mechanisms of the VRC formation are awaited . In addition to glycolysis , cytoplasmic GAPDH-C has a variety of functions , such as membrane fusion and vesicular transport ( reviewed in [61] , [62] ) . In contrast , only the classical functions associated with carbon fixation have been reported previously for GAPDH-A . Thus , further experimental evidence is required to explain the involvement of GAPDH-A in the intracellular transport mechanism of RCNMV RNA . Alternatively , RCNMV might have evolved to utilize highly expressed and ubiquitous GAPDH-A by diverting it from its natural functions . Several plant RNA viruses use host metabolic enzymes and housekeeping proteins in ways that are unrelated to their original functions [63] . Several chloroplast-localizing proteins have recently been shown to regulate virus multiplication . Among these , chloroplastic phosphoglycerate kinase ( PGK ) was isolated from RNA-dependent RNA polymerase ( RdRp ) fraction prepared from Bamboo mosaic virus-infected N . benthamiana . PGK positively regulates multiplication of the virus through the interaction with the 3′ untranslated region of the viral genomic RNA and transportation to the chloroplasts where the viral RNA replication occurs [64] , [65] . ATP synthase-γ subunit ( AtpC ) and Rubisco activase ( RCA ) were also isolated from the RdRp fraction prepared from the TMV-infected N . tabacum . AtpC and RCA negatively regulate the movement and accumulation of the virus , respectively [66] . Interestingly , gene silencing of these genes led to the increased number and the smaller size of VRC . These results are in contrast to our results that VIGS of NbGAPDH-A did not affect the number and the size of the cortical VRC in RCNMV infected cells and that it interfered with the recruitment of the MP to the cortical VRC ( Figures 9 and 10 ) . Plant RNA viruses might have evolved to utilize abundant chloroplast-localizing proteins as the positive or negative regulators through the interaction with viral proteins .
mCherry gene was amplified from pmCherry-N1 ( Clontech ) using primers 1 and 2 . The amplified PCR product was digested with ClaI/MluI and inserted into the same sites of pUCR1-MsG [39] , producing pUCR1-MmC ( Figure S1B ) that expresses MP-mCherry fusion protein from the subgenomic RNA . AscI/SacI fragment of pUC118RA1 ( AscI ) [10] that contains the expression cassette of RCNMV RNA1 and AscI/SmaI fragment of pUC118RA2 ( AscI ) [10] that contains the expression cassette of RCNMV RNA2 was inserted into SacI/SmaI site of pBIC18 [67] binary vector , producing pBICR12 ( Figure S1C ) that expresses full-genome of RCNMV . pBICp27-iFTH is a binary vector plasmid expressing p27 tagged with FLAG-TEV protease recognition peptide-HA [28] . The TEV protease recognition peptide was replaced by 3C protease recognition peptide by recombinant PCR to produce pBICp27TEP ( Mine and Okuno , unpublished ) . The tag sequence for tandem affinity purification was amplified from pBICp27TEP using primers 3 and 4 . The cauliflower mosaic virus 35S promoter and 5′ half of RNA2 was amplified from pBICR12 using primers 5 and 6 . A DNA fragment containing the 3′ half of RNA2 and the 35S terminator sequence was amplified from pBICR12 using primers 7 and 8 . These three fragments were mixed and used as the template for recombinant PCR using primers 6 and 8 . The generated PCR product was digested with AscI/SmaI and inserted into the same sites of pBICR12 , producing pBICR12/MP-TAP ( Figure S1D ) . R1-MP:GFP plasmid [38] was digested with ClaI and MP coding sequence was removed . The larger fragment was self-ligated , producing pR1-sGFP ( Figure S1E ) . pR1-sGFP was digested with BglII/MluI and the 0 . 8 kb fragment containing GFP gene was inserted into the same sites of pUC118RA1 ( AscI ) , producing pUC118RA1sGC ( AscI ) . pUC118RA1sGC ( AscI ) was digested with AscI/SacI and the 4 . 4 kb fragment was inserted to the same sites of pBICR12 , producing pBICR1sG2 ( Figure S1F ) . A DNA fragment containing HA and the 3′ non-coding region of RNA2 was amplified from pRC|2G using primers 9 and 10 . A DNA fragment containing T7 promoter and the 5′ half of RNA2 and HA was amplified from pRC2|G using primers 11 and 12 . These fragments were mixed and used as the template for recombinant PCR using primers 10 and 12 . The generated PCR product was digested with EcoRI/SmaI and inserted into the same site of pUC119 , producing pUCR2MP-HA ( Figure S1H ) . EcoRI/HindIII fragment of pBE2113 [69] that contains a 35S promoter-∧ sequence-nos terminator cassette was inserted to the same sites of pUC19 ( Takara Bio Inc . ) producing pUC2113 . The XbaI site downstream of ∧ sequence in pUC2113 was digested and filled in with T4-polymerase , and the linker sequence containing SacI site was ligated , producing pUC2113 ( SacI ) . EcoRV fragment of piL:G3 ( 0 . 4 kb ) [70] containing the 35S promoter and the 5′ sequence of Tomato mosaic virus ( ToMV ) was inserted to the same site of pUC2113 ( SacI ) , producing pUC:ToMVrec . EcoRI/HindIII fragment of pBE2113 was replaced by the EcoRI/HindIII fragment of pUC:ToMVrec , producing pBE:ToMVrec . The KpnI/MluI fragment of piL:G3 containing ToMV MP and GFP sequence was inserted to the same sites of pTLW3 [71] , producing pTLWdCP-GFP . The ribozyme sequence of Tobacco ringspot virus satellite RNA was PCR-amplified from pUCBR1R plasmid [72] and introduced to MluI site downstream of ToMV 3′ noncoding sequence in pTLWdCP-GFP , producing pTLWdCP-GFP-rib . Finally , StuI/SacI fragment of pTLWdCP-GFP-rib , containing most of the recombinant ToMV and the ribozyme sequences was inserted into the same sites of pBE:ToMVrec , producing pToMVdCP-GFP . N . benthamiana plants were grown on commercial soil ( Tsuchi-Taro , Sumirin-Nosan-Kogyo Co . Ltd . ) at 25±2°C and 16 hours illumination per day . N . benthamiana protoplasts were prepared according to Li et al . ( 2013 ) [73] and Navas-Castillo et al . ( 1997 ) [74] with minor modifications . Briefly , young expanded leaves from 5 week-old plants were cut into 1-mm strips with a razor blade and digested in 15 ml of enzyme solution ( 1% cellulase RS [Yakult Pharmaceutical Ind . Co . Ltd . ] , 0 . 5% macerozyme R-10 [Yakult] , 0 . 5 M mannitol , 10 mM CaCl2 , 5 mM MES , pH 5 . 7 ) within a petri dish at 25°C in the dark with gentle shaking ( 40 rpm ) for 4 to 5 h . After being filtered through 4 layers of cheesecloth , protoplasts were precipitated by centrifugation at 80× g for 2 min and were suspended with 10 ml of MMC solution ( 0 . 5 M mannitol , 10 mM CaCl2 , 5 mM MES , pH 5 . 7 ) . Concentration of cells were counted using hemacytometer . Protoplasts ( 1×105 cells in 100 µl ) were mixed with 5 µg of viral RNAs and 200 µl of PEG solution ( 1 g of PEG4000 [Sigma-Aldrich #81240] , 125 µl of sterile distilled water , 1 . 25 ml of 0 . 8 M mannitol , 250 µl of 1 M Ca ( NO3 ) 2 ) and mixed completely by gently tapping the tube . Then 2 ml of MMC solution was added and mixed . After 15 min of incubation on ice , protoplasts were precipitated by centrifugation at 80× g for 2 min , resuspended in 4 ml of MMC solution and precipitated by centrifugation again . Protoplasts were resuspended in 0 . 5 ml of W5 solution ( 154 mM NaCl , 125 mM CaCl2 , 5 mM KCl , 2 mM MES , pH 5 . 7 ) and incubated . Preparation of protoplasts from NbGAPDH-A-silenced or ALSV-infected N benthamiana plants was essentially described above , except that the plants were 6–8 weeks old . Fixation of N . benthamiana protoplasts and immunolabeling procedure were as described by Liu et al . ( 2005 ) [20] . For the detection of double-stranded RNA , formaldehyde-fixed protoplasts were incubated with mouse monoclonal antibody J2 ( diluted 1∶200; Scicons ) for 16 h in a moisturized chamber at 4°C . The samples were washed three times and then incubated with Alexa Fluor 488-conjugated goat anti-mouse IgG antibody ( diluted 1∶200; Invitrogen ) for 2 h at room temperature . After washing three times , the samples were subjected to CLSM . N . benthamiana plants and Agrobacterium tumefaciens GV3101 ( pMP90 ) were used for infiltration experiments as described previously [67] . A . tumefaciens transformed by pBICR12/MP-TAP , or negative control pBICR12 was used for expression of MP-HA from viral context . 1 . 67 g of Agrobacterium-infiltrated leaves at 48 h post infiltration were ground in liquid nitrogen and homogenized in 5 ml of extraction buffer A ( 50 mM Tris-HCl [pH 8 . 0] , 150 mM NaCl , 5% glycerol , 0 . 5% Triton X-100 , 1 tablet of Complete Mini protease inhibitor cocktail [EDTA-free , Roche Diagnostics]/10 ml ) , followed by centrifugation at 21 , 000× g for 10 min at 4°C to remove cell debris . The supernatant ( 4 . 0 ml ) was divided into 5 tubes ( 800 µl each ) , and each incubated with 20 µl of Anti-HA Affinity Matrix ( Roche #11815016001 ) for 4 h at 4°C with gentle rotation . The resin was washed three times with 1 ml of washing buffer 1 ( 50 mM Tris-HCl [pH 7 . 4] , 150 mM NaCl , 5% glycerol , 0 . 1% Triton X-100 ) , and equilibrated with 3C buffer ( washing buffer 1 containing 1 mM DTT ) . Then the resin was incubated with 20 units of PreScission protease ( GE Healthcare ) in 1 ml of 3C buffer for 16 h at 4°C with gentle rotation . The resins were centrifuged at 500× g for 1 min , and the supernatant ( 1 ml ) was immunoprecipitated again with 50 µl of ANTI-FLAG M2 Affinity Gel ( Sigma-Aldrich #A2220 ) for 4 h at 4°C with gentle rotation . The gel was then washed three times with washing buffer 1 . The bound proteins were eluted by 125 µl of elution solution ( washing buffer 1 containing 150 ng/µl FLAG peptide [Sigma-Aldrich #F3290] ) for 30 min at 4°C with gentle rotation . This elution process was repeated once again , and the total of 250 µl was precipitated with trichloroacetic acid . The affinity-purified preparation and its control preparation were subjected to SDS-PAGE , and the several bands that were not detected in the negative control lane ( Figure 2A ) were cut out and subjected to liquid chromatography-tandem mass spectrometry analysis , as described previously [28] . RNA extraction from N . benthamiana leaves was performed using PureLink Plant RNA Reagent ( Invitrogen ) and treated with DNase ( RQ1 RNase-free DNase; Promega ) . Reverse transcription was carried out using PrimeScript RT reagent Kit ( Takara ) using oligo-dT . Based on the GAPDH-A sequence of N . tabacum ( gi|120661 ) , primers 13 and 14 were designed . An 1176-bp nearly full-length cDNA fragment of GAPDH-A gene was amplified from cDNA derived from N . benthamiana RNA using primers 13 and 14 . The 5′ and 3′ sequences of NbGAPDH-A were amplified by SMARTer RACE cDNA amplification Kit ( Clontech ) using the gene-specific primers 15 and 16 , respectively , and cloned into pGEM-T Easy ( Promega ) . From each of 8 clones the 5′ and 3′ ends of NbGAPDH-A gene were determined . Nucleotide sequence data of NbGAPDH-A gene is available in the DDBJ/EMBL/Genebank databases under accession number AB937979 . Based on the GAPDH-B sequence of N . tabacum ( gi|120665 ) , primers 17 and 18 were designed . A 1270-bp partial fragment of GAPDH-B cDNA was amplified from cDNA derived from N . benthamiana RNA using primers 17 and 18 , and cloned into pGEM-T Easy . Partial sequence of NbGAPDH-B gene was determined . Full-length cDNA of NbGAPDH-A was amplified from cDNA derived from N . benthamiana RNA using primers 19 and 20 . The generated PCR product was then cloned into the BamHI site of pBICP35 , producing pBICNbGA-myc . Full-length cDNA of NbGAPDH-A was amplified from pBICNbGA-myc using primers 14 and 19 . The generated PCR product was digested with BamHI/ClaI and cloned into the same sites of pUB/RMsG [39] , producing pUBNbGA-sG . BamHI/HindIII fragment of pUBNbGA-sG , containing NbGAPDH-A-GFP and 35S terminator , was cloned into the same sites of pBICP35 , producing pBICNbGA-sG . This was used for transient expression of NbGAPDH-A-GFP fusion protein by agroinfiltration . Chloroplast targeting sequence of RbcS was amplified from cDNA derived from A . thaliana RNA using primers 21 and 22 . sGFP sequence was amplified from pUBsGFP [39] using primers 23 and 24 . Recombinant PCR fragment was amplified using primers 21 and 24 . The generated PCR product was then cloned into the BamHI and KpnI sites of pUBP35 [67] , producing pUBTPRbcS-sGFP . HindIII/SalI fragment of pUBTPRbcS-sGFP was cloned into the same sites of pBICP35 , producing pBICRbcSTP-sGFP . MP-mCherry sequence was amplified from pUCR1-MmC using primers 25 and 26 . The generated PCR product was digested with BamHI/EcoRI and cloned into the same sites of pBICP35 , producing pBICRMmC . pBICRbcSTP-sGFP and pBICRMmC was introduced into A . tumefaciens and used for the transient expression of RbcSTP-GFP and MP-mCherry , respectively . Construction of ALSV vector pBICAL1 , pBICAL2 and pBICAL2gsPDS was described previously [75] . Partial fragment of NbGAPDH-A ( 294 nucleotides ) was amplified from cDNA derived from N . benthamiana RNA using primers 27 and 28 . The fragment was digested with BamHI/XhoI and inserted into the same sites of pBICAL2 , producing pBICAL2gsNbGAP-A . The plasmids containing the ALSV expression cassette were introduced into A . tumefaciens GV3101 ( pMP90 ) . Similar amount of fresh colonies of Agrobacterium containing pBICAL1 and each of pBICAL2 , pBICAL2gsPDS , or pBICAL2gsNbGAP-A were collected using sterile toothpicks and suspended in 0 . 2 ml of Agro Incubation Buffer ( 10 mM MgCl2 , 10 mM MES-KOH , pH 5 . 7 , 0 . 15 mM Acetosyringone ) at OD600 of 2 . 0–3 . 0 , and were incubated at 20°C for more than 3 h in the dark . Sterile toothpick was soaked in the Agrobacterium suspension and stuck 4 times into 1st , 2nd and 3rd true leaves of 17–21 days old N . benthamiana plants . Three to 4 days later , toothpick-inoculation was repeated to a newly developed leaf . After inoculation , the plants were incubated in a moist chamber at 22°C overnight and transferred to a plant growth room at 25°C . Two to 3 weeks later , silencing of PDS or NbGAPDH-A was induced in the non-inoculated upper leaves . Total RNA extracted from N . benthamina leaves or protoplasts were subjected to reverse transcription using PrimeScript RT reagent Kit ( Takara ) using oligo-dT according to manufacturer's protocol . Real-time PCR was carried out using SYBR Premix Ex Taq ( RR420A , Takara ) using primers 29 and 30 for EF1 and primers 16 and 31 for NbGAPDH-A . Quantitative analysis of each mRNA was performed using a Thermal cycler Dice Real Time System TP800 ( Takara ) . Semi-quantitative RT-PCR was performed using the same cDNA and primers and amplified by Ex Taq polymerase ( Takara ) . Protein extraction and western blot analyses were performed as described previously [76] . Total RNA extraction from N . benthamiana leaves or protoplasts and northern blot analysis were performed as described previously [76] . Probes used for detection of positive-strand RCNMV RNA1 and RNA2 were as described previously [49] . The signals were detected with a luminescent image analyzer ( LAS 1000 plus , Fuji Film Co . Ltd . ) and the signal intensities were quantified using the Image Gauge program version 3 . 1 ( Fuji Film ) . The spread of GFP fluorescence was observed using an Olympus BX53 fluorescence microscope equipped with an Olympus DP72 camera using the imaging program Olympus cellSens . Subcellular localizations of proteins tagged with FPs and dsRNA that was detected with fluorescent antibodies were observed using an Olympus FluoView FV500 confocal microscope . Both a Nikon 60× Plan Apo oil immersion objective lens ( numerical aperture 1 . 4 ) and a Nikon 40× UPlan Apo oil immersion objective lens ( numerical aperture 1 . 0 ) were used . The sets of dichroic mirror , beam splitter , and emission filter used were DM488/543 , SDM560 , and BA505-525 for GFP , and DM488/543/633 , SDM630 , and BA560-600 for mCherry For the detection of mCherry signal and chloroplast autofluorescence simultaneously , emission filter BA610IF was used . In experiments for detecting dual localization , scanning was performed in sequential mode to minimize signal bleed-through . All images shown are from optical sections taken at 1 or 2 µm intervals and were processed using Adobe Photoshop CS6 software . BamHI/EcoRI fragment of pBICRMP-HA [39] that contains MP-HA gene was inserted into pCold-1 ( Takara ) , producing pCold-MP-HA . EcoRI/KpnI fragment of pCold-MP-HA that contains MP-HA gene was inserted into the same sites of pColdGST , producing pColdGST/MP-HA . pBICNbGAP-myc was digested with BamHI and the smaller fragment containing NbGAPDH-A-myc was cloned into the same site of pCold-1 in the correct orientation , producing pColdNbGAP-myc . E . coli BL21 ( DE3 ) strain was transformed with plasmids containing the prefix pCold and used for the expression of GST and GST-fused viral proteins and NbGAPDH-A tagged with a myc . All the conditions and procedures are described previously [31] . The sequence of the C-terminal half of YFP was amplified from pBICHA:cYFP [30] using primers 32 and 33 [30] . The amplified DNA was digested with StuI and cloned into StuI-digested pBICAsc2 [30] , producing pBICHA-cYFPAsc2 . Full-length cDNA of NbGAPDH-A was amplified from pUBNbGA-sG using primers 34 and 35 . The generated PCR product was digested with BamHI and cloned into pBICHA-cYFPAsc2 , producing pBICGAP-HA-cYFP . myc-nYFP sequence was amplified from pBICMP-myc-nYFP using primers 36 and 37 . The amplified PCR products were digested with ClaI/MluI and cloned into the same sites of pUCR1-MsG , producing pUCR1-MnY ( Figure S1I ) . p88-myc-nYFP sequence was amplified from pUCR1-MnY using primers 37 and 38 . p88 sequence was amplified using primers 39 and 40 The recombinant PCR products were generated from mixture of these products using primers 37 and 39 , and were digested with MluI/XhoI and cloned into the same sites of pUCR1-MsG , producing pUCR1-nYFP ( Figure S1J ) . Twenty-five to 28 days old N . benthamiana plants were used for BiFC assays . pBICGAP-HA-cYFP , or control pBICHA-cYFPAsc2 plasmid , together with pBICp19 that expresses TBSV silencing suppressor protein p19 was infiltrated via A . tumefaciens GV3101 ( pMP90 ) as described above . The plants were incubated in a moist chamber at 22°C for 18 h . Then in vitro transcripts ( 1 µg/µl ) of the recombinant RCNMV that expresses MP-nYFP ( Figure S1I ) , or negative control virus that expresses nYFP ( Figure S1J ) were mechanically inoculated onto the leaves . The plants were incubated in a moist chamber at 17°C for 27–30 h and were subjected to CLSM . NbGAPDH-A was registered through DDBJ and accession number AB937979 was given on May 27 2014 .
|
Intercellular movement of plant viruses is the crucial step during systemic viral infections . Red clover necrotic mosaic virus ( RCNMV ) , a bipartite positive-strand RNA plant virus , forms movement protein ( MP ) -containing punctate structures on the cortical endoplasmic reticulum in infected cells , which are required for efficient intercellular movement of the virus . We provide evidence that these cortical punctate structures constitute the viral replication complex ( VRC ) , which forms during the early stage of virus infection . Furthermore , we show that a host protein of Nicotiana benthamiana , glyceraldehyde 3-phosphate dehydrogenase-A ( NbGAPDH-A ) , possibly intercalates between the cortical VRC and MP . Knockdown of NbGAPDH-A diffused subcellular localization of MP and reduced intercellular movement of the virus . Chloroplastic NbGAPDH-A relocalized to the cortical VRC after infection with the virus . Our results suggest that the cortical VRC serves not only as the replication factory of viral RNA but also as a transportation hub , which transports viral RNA to neighboring uninfected cells via plasmodesmata .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology",
"virology",
"biology",
"and",
"life",
"sciences",
"microbiology",
"molecular",
"cell",
"biology",
"cellular",
"structures",
"and",
"organelles"
] |
2014
|
GAPDH-A Recruits a Plant Virus Movement Protein to Cortical Virus Replication Complexes to Facilitate Viral Cell-to-Cell Movement
|
DNA–protein interactions are involved in many essential biological activities . Because there is no simple mapping code between DNA base pairs and protein amino acids , the prediction of DNA–protein interactions is a challenging problem . Here , we present a novel computational approach for predicting DNA-binding protein residues and DNA–protein interaction modes without knowing its specific DNA target sequence . Given the structure of a DNA-binding protein , the method first generates an ensemble of complex structures obtained by rigid-body docking with a nonspecific canonical B-DNA . Representative models are subsequently selected through clustering and ranking by their DNA–protein interfacial energy . Analysis of these encounter complex models suggests that the recognition sites for specific DNA binding are usually favorable interaction sites for the nonspecific DNA probe and that nonspecific DNA–protein interaction modes exhibit some similarity to specific DNA–protein binding modes . Although the method requires as input the knowledge that the protein binds DNA , in benchmark tests , it achieves better performance in identifying DNA-binding sites than three previously established methods , which are based on sophisticated machine-learning techniques . We further apply our method to protein structures predicted through modeling and demonstrate that our method performs satisfactorily on protein models whose root-mean-square Cα deviation from native is up to 5 Å from their native structures . This study provides valuable structural insights into how a specific DNA-binding protein interacts with a nonspecific DNA sequence . The similarity between the specific DNA–protein interaction mode and nonspecific interaction modes may reflect an important sampling step in search of its specific DNA targets by a DNA-binding protein .
DNA-binding proteins play an essential role in many fundamental biological activities , including DNA transcription , replication , packaging , repair and rearrangement . Interactions relevant to these activities typically involve specific binding sites on both proteins and DNA . Over the past several decades , many efforts have been made in order to understand basic principles that determine the specific DNA-protein interactions . It is well-known that there does not exist a simple recognition code between protein amino acids and DNA base pairs [1]–[4] . This poses a great challenge for the prediction of DNA-protein interactions . The daunting task of elucidating DNA-protein interactions can be addressed with the assistance of computational modeling . Methods for docking the complex from separated protein/DNA structures have been developed [5]–[7] . As an early example , the Monte Carlo program MONTY has been applied to sample configurations of a single DNA-protein complex in the vicinity of its native state [6] . The development of an efficient geometric recognition algorithm [8] , which allows a global search for optimal surface complementarity though rigid body rotation and translation , greatly advanced the molecular docking field . An implementation of the algorithm , FTDOCK , was applied to DNA-protein docking [5] , with encouraging benchmark results reported on modeling eight DNA/repressor complexes starting from unbound protein structures and canonical B-DNA . A more recent approach , HADDOCK , starts with a similar rigid body docking procedure , followed by semi-flexible refinement [7] . Excellent docking models were obtained for three examples by HADDOCK . The docking methods assume the availability of both protein and DNA structures . Given only the structure of a DNA-binding protein , it is of interest to determine the DNA-binding protein residues without the knowledge of the associated specific DNA sequence and structure with which the protein interacts . In the last few years , several methods have been developed to address this problem [9]–[15] . Most focus on analyzing characteristic patterns of DNA-binding residues from the solved structures of complexes . Standard machine-learning techniques , such as Support Vector Machine [10] , [13] and neural networks [9] , [14] , have been adopted to differentiate DNA-binding residues from non-DNA-binding residues , using features like sequence composition , evolutionary profile , solvent accessibility , and electrostatic potential . Recently , a knowledge-based method DBD-Hunter that combines structural comparison and evaluation of a statistical pair potential was proposed for predicting DNA-binding proteins and associated binding residues [11] . The method yields an accuracy of 87% on DNA-binding site prediction in comprehensive benchmarks . However , the method is limited by the availability of appropriate DNA-protein complex structures to be used as templates . In this study , we present a novel approach for predicting the protein residues that bind DNA and DNA-protein interaction modes , given the structure of a DNA-binding protein as the input . We systematically docked 44 specific DNA-binding proteins in both holo ( DNA-bound ) and apo ( DNA-free ) forms to a nonspecific canonical B-DNA molecule . Using energy evaluation and model clustering , we obtained representative complex models that provide structural insights into how DNA-binding proteins interact with a nonspecific DNA sequence . For about 80% of the proteins , the sites for specific DNA recognition are among the favorable interaction sites for nonspecific DNA binding . Furthermore , the interaction modes observed in the top ranked , nonspecific DNA-protein encounter complexes bear a certain similarity to the specific DNA-protein binding mode in the experimental structure . The biological implications of this similarity are discussed . Moreover , we demonstrate that our approach achieves better performance than three established methods based on machine-learning techniques . In addition to experimental structures , we show that our method can be applied to predicted protein models , generated by the state-of-the-art modeling program TASSER [16] . Satisfactory results were obtained for protein models with a root-mean square deviation , RMSD , ≤5 Å of their Cα atoms from their native holo-structures . We also show that our method can be further improved by considering conformational changes of DNA .
The apo- and holo-structures of 44 non-redundant specific DNA-binding proteins ( Table S1 ) are docked separately to a nonspecific B-DNA composed of 16 dA·dT base pairs , following the modeling procedure illustrated in Figure 1A . For each structure , we keep the top 2500 docking complex models ranked by their DNA-protein interfacial energy . We first compare DNA-interacting protein residues observed in top ranked encounter complexes with those observed in the native ( experimental ) complex structures . For this purpose , the Matthews Correlation Coefficient ( MCC ) is used to quantify the similarity between interaction sites for specific and nonspecific DNA on the protein's surface . A complex model is considered near-native if the associated MCC is higher than 0 . 5 , which is the mid-point between perfect overlap ( MCC = 1 . 0 ) and a random model ( MCC = 0 . 0 ) . As a representative example , Figure 1B and 1C show the energy and MCC for the top 2500 docked structures of Epstein-Barr nuclear antigen-1 , whose top energy ranked model is a near-native model with a high MCC of 0 . 76 . Analysis of docking solutions suggests that specific DNA-binding sites on proteins are typically among the energetically favorable sites for sampling the nonspecific DNA . As shown in Figure 2 , the MCC between specific and nonspecific DNA-binding sites is anti-correlated with the DNA-protein interfacial energy . A representative example is provided for the Epstein-Barr nuclear antigen-1 , which has a Pearson Correlation Coefficient ( PCC ) of −0 . 46 between MCC and the interfacial energy ( Figure 2A ) . On average , the PCCs are −0 . 40/−0 . 43 for the APO/HOLO sets , respectively ( Figure 2B ) . Although the correlation is not very strong , the analysis does indicate that the specific DNA-binding sites on the protein are more likely involved in forming encounter complexes with a nonspecific DNA , as compared to the other regions of the protein . These nonspecific encounter complexes provide a structural basis for understanding the process known as facilitated diffusion [17] , [18] , during which a DNA-binding protein diffuses along nonspecific DNA in search of its specific DNA target sequence ( see Discussion ) . For the purpose of sampling DNA sequence , the DNA-binding sites on the protein surface are energetically favorable to both specific and nonspecific DNA , resulting in the observed overlap between these sites . One can utilize this observation to predict specific DNA-binding sites on protein through analyzing nonspecific DNA-protein docking solutions . Figure 3A and 3B show the number of proteins with at least one near-native complex model under various rank thresholds . According to the DNA-protein interfacial energy , we obtained a near-native top one model for 17 ( 39% ) and 23 ( 52% ) proteins , using apo and holo protein structures for docking , respectively . By comparison , shape complementarity ranking merely provides 2 ( 5% ) and 9 ( 20% ) proteins with a near-native top one ranked model based on apo- and holo-structures . Among the top ten energy ranked models , one can find at least one near-native model for 34 ( 77% ) and 37 ( 84% ) proteins from the APO and HOLO sets , while only 12 ( 27% ) and 30 ( 68% ) proteins from the same sets have a near-native model on the top ten list based on shape complementarity ranking . To further improve model selection , we introduced a clustering procedure and compared various model selection schemes shown in Figure 3C and 3D . As expected , a randomly chosen model from the 2500 docking solutions gives a mean MCC very close to zero , 0 . 005/0 . 036 on the APO/HOLO sets , respectively . The mean MCC values of the top one shape complementarity ranked models , 0 . 06/0 . 11 on APO/HOLO sets , are slightly better than the means of random models . A significant jump to a mean MCC of 0 . 39/0 . 44 ( APO/HOLO ) is seen by selecting the top one energy ranked model , EN1 , and these increase to 0 . 51/0 . 59 using the best of top five energy-ranked models , EN5 . Clustering further improves model selection , with the best of top five clustering representative models , CL top5 , yielding mean MCCs of 0 . 54/0 . 62 , accuracies of 87%/89% , sensitivities of 57%/62% , specificities of 94%/95% , and precisions of 69%/77% , for the APO/HOLO sets ( see Table 1 ) . Interestingly , the top ranked cluster model , CL top1 , has a MCC of 0 . 40/0 . 44 , which is only slightly better than the EN1 model . Our method can readily take advantage of known information about DNA-binding sites , such as data collected from mutagenesis studies , NMR experiments , or sequence conservation analysis . The information can be used to derive contact restraints for model filtration [5] , [7] . To illustrate this point , we randomly picked native DNA-binding protein residues and filtered all models in which these residues do not contact DNA . When applying more than one such restraint , we obtained significantly better top one models ( Figure 4 ) . The mean MCC values for the CL top1 models of apo-structures , for example , systematically increases from 0 . 40 without any restraint , to 0 . 45 , 0 . 52 , and 0 . 59 with two , three , and five restraints , respectively . Next , we compare interaction modes between representative nonspecific DNA-protein encounter complexes and the native ( experimental ) specific DNA-protein complexes . For this comparison , we need a mapping between the nonspecific DNA and the specific DNA complexed with the protein in the native structure . The mapping was obtained by gaplessly threading the nonspecific DNA along the native DNA with a scoring function that maximizes the overlap of the DNA-protein residue contacts . Then , the native DNA-protein contacts observed in the model were counted , and the RMSD of native interfacial residues relative to their positions in the model was calculated by optimally superposing these interfacial residues . For each protein , the best result of top five clustering models is shown in Figure 5A . In these models , the optimal alignment typically covers 85% of the length of the shorter DNA , and more than 95% of the native interfacial residues . On average , the fractions of native contacts ( denoted as Fnat ) observed in the model are 33%/41% for the APO/HOLO sets , respectively , and the corresponding DNA-protein interfacial RMSDs ( denoted as RMSDint ) are 4 . 6/3 . 4 Å . The results indicate some resemblance between nonspecific DNA-protein interaction modes and the specific-DNA-protein binding mode , though consistent specific base recognition cannot be expected due to the different DNA sequence employed and the possible conformational changes involved . About 70% of contacts involving specific base recognition in the specific complex are either lost or converted to backbone contacts in the corresponding nonspecific contacts . From the prediction prospective , we may define a DNA-protein complex model as acceptable if the model satisfies one of the following two conditions: ( i ) Fnat≥30% , or ( ii ) Fnat≥10% and RMSDint≤4 Å , the criteria adopted from the Critical Assessment of PRedicted Interactions ( CAPRI ) [19] . Using these criteria , the predicted DNA-binding modes for 71%/86% of APO/HOLO proteins can be classified as acceptable , resulting in a mean RMSDint of 3 . 9/3 . 1 Å and a mean Fnat of 37%/44% . Three examples of predicted nonspecific DNA-protein complex models based on apo-structures are compared with the corresponding native specific DNA-protein complex structures in Figure 5B–D . The Antennapedia homeodomain from a Drosophila melanogaster transcription factor represents a classic DNA-binding domain that recognizes DNA through a helix-turn-helix motif [20] , [21] . Using an apo protein structure [20] , the best clustering model contains 14 DNA-interacting protein residues; all are among the 19 DNA-binding residues bound to the specific DNA sequence . The native-like binding mode of the predicted model is reflected by a RMSDint of 1 . 9 Å and a Fnat of 54% ( Figure 5B ) . The model , promoted from the sixth place on the energy ranking list to the second place through clustering , is the closest to the native structure among all 2500 docking solutions . The second example from Saccharomyces cerevisiae Ndt80 is a DNA-binding domain belonging to the immunoglobulin-fold family of transcription factors [22] , [23] ( Figure 5C ) . The native DNA-protein interface exhibits a unique binding mode involving mainly loop residues . The top energy-ranked model correlates well with the native structure , having a MCC of 0 . 71 , which is only slightly lower than the best value of 0 . 72 found among all docking solutions . The interfacial RMSD of 3 . 0 Å and Fnat of 55% suggest close similarity between the predicted and native binding mode . The third example is a type II restriction endonuclease , EcoRV ( Figure 5D ) , whose structures have been solved in DNA-free [24] and DNA-bound forms with either a cognate or a non-cognate DNA sequence [24] , [25] . In the top energy-ranked model obtained with the unbound structure , residues involving DNA-protein interactions include about half of the protein residues contacting the cognate sequence in the experimental structure , yielding a moderate MCC of 0 . 51 . The result is expected since the cognate DNA significantly deviates from the canonical B-DNA form by a bending angle of ∼50° , as shown in the native complex structure . As a result , the nonspecific DNA can only be partially aligned to the cognate DNA . In fact , the interaction mode presented by our model more closely resembles the binding mode of the non-cognate DNA-protein complex structure ( Figure 5D ) . All ten DNA-binding residues involving non-cognate DNA recognition are predicted as DNA-binding according to our model . Note that EcoRV functions as a homodimer , and only the monomer was employed for docking . Our approach was further validated on predicted protein models . First , the sequences of these 44 DNA-binding proteins were input into the threading algorithm PROSPECTOR_3 . 0 [26] . Depending on the confidence levels of the structural templates identified , proteins were classified into two groups: 30 Easy targets , which typically have good quality templates , and 14 Hard targets , which usually do not have a reliable template hit . Note that we excluded from the template library any structure that shares>30% global sequence identity with a given target . The best template , ranked by the TM-score structural similarity metric [27] , has a mean RMSD of 7 . 9 Å with respect to the native holo-structure over about 92% alignment coverage , and the mean sequence identity of these templates is 19% . After TASSER runs for model assembly and refinement [16] , the mean RMSDs of the top TASSER model and of the best of top five models were improved to 6 . 9 Å and 6 . 4 Å over the regions aligned with the templates . Overall , the mean TM-scores of the top and the best of five top models compared against the native holo structure are 0 . 61 and 0 . 63; the latter is ∼9% higher than the average TM-score of the best threading templates . Systematic model improvement over the best templates is evident , as an improved structural model was obtained in 37 of 44 cases . For reach protein , the top TASSER model was employed for docking and subsequent analysis . The number of proteins whose top TASSER model has a RMSD≤5 . 0 Å from the native holo-structures is 24 ( 55% ) ; all but one are from the easy set ( Figure 6A ) . Among these 24 proteins , the best of top five DNA-protein complex models yields an average MCC/accuracy of 0 . 51/84% for DNA-binding site prediction . For the Easy/Hard sets , the best of top five models gives mean MCC of 0 . 50/0 . 23 , a RMSDint of 5 . 9/11 . 2 Å , and Fnat of 29%/16% , respectively ( Figure 6B ) . While we obtained acceptable binding mode predictions for 12 ( 40% ) of the targets from the Easy set , the predicted binding mode for Hard targets is generally incorrect , which is expected due to poor protein model quality . Overall , the binding site and mode predictions are satisfactory for the Easy set . One example , the DNA-binding domain from an E . coli group IV σ factor , is illustrated in Figure 6C . The protein initiates transcription by binding to a specific promoter region and recruiting an RNA polymerase [28] . The closest template , an Aquifex aeolicus group I σ factor structure resolved in its DNA-free form , shares a sequence identity of 24% with the target . The top ranked TASSER model has an RMSD of 2 . 5 Å from the crystal protein structure ( Figure 6C ) . The high quality model permitted us to build reliable docking complex models . The best of top five docking models predicts 11 of 15 DNA-binding amino acids at 92% precision; and the predicted interaction mode closely mimics the native binding mode exhibited by the crystal structure with an interfacial RMSD of 2 . 5 Å and Fnat of 53% . In addition to the DNA-protein energy function described above , we also tested the performance of three other statistical pair potentials proposed previously , including two quasichemical potentials , one at the residue , QCRes [5] and two others at the all-atom level , QCAA [29] and RAPDF [30] ( see Methods ) . While the residue-level quasichemical potential uses a single distance cutoff of 4 . 5 Å , the all-atom potentials are distance dependent up to 10 Å . Since in previous studies , the potentials were derived from relatively small data sets , we re-parameterized these three potentials with the same set of 179 crystal complex structures used for our functional-group level quasichemical potential derivation [11] . Then , for each target from the APO/HOLO sets , the top 2500 docking solutions described above were re-ranked according to the energies calculated with the new potentials . Table 2 shows the results of binding site and mode predictions for the best of the top five models . On average , our energy function outperforms these three potentials . The mean MCC for the binding site prediction is 0 . 59/0 . 51 for the APO/HOLO sets using our energy function without clustering , compared with 0 . 55/0 . 47 , from both the residue and all-atom quasichemical potentials , and 0 . 40/0 . 24 from the conditional probability scoring function RAPDF . Correspondingly , our energy function selected acceptable binding complex models in 77%/59% of the cases , whereas the residue-based and the two all-atom potentials selected acceptable models in 71%/50% , 71%/55% , and 40%/32% of the cases , respectively . These results suggest that detailed all-atom representations do not necessarily have an advantage over simplified residue or functional-group level potentials when applied to rank docking solutions from a non-specific DNA sequence . We also note that the clustering models , which have a mean MCC of 0 . 62/0 . 54 , are significantly better than models selected by the three potentials ( Wilcoxon signed-rank tests P<0 . 04 ) . Our approach was compared with three established methods [9] , [13] , [14] that predict DNA-binding sites based on protein structures . Note that none of these three methods is capable of predicting the DNA-protein interaction mode . For the purpose of comparison , all calculations were carried out on the same set , AS62 [9] , composed of DNA-binding protein structures in their holo-forms . As shown in Table 3 , the top model from our approach already yields better results than previous methods on average . The mean MCC of our top model is 0 . 53 , compared to 0 . 49 obtained independently by the Kuznetsov group [13] and by Tjong and Zhou's method named DISPLAR [14] . Moreover , the best of our top five models significantly improves the DNA-binding site prediction with a mean MCC of 0 . 62 and a mean accuracy of 87% , leading the results from the Kuzentsov method or DISPLAR by about one standard deviation unit . The latter two methods perform better than that proposed by Ahmad et al . [9] . This reason can be partially attributed to the fact that the Ahmad et al . did not use position-specific sequence profiles in their method . We further compared the performance of our method on apo structures with DISPLAR . The predictions of DNA-binding sites of 44 proteins structures from the APO set were performed using the DISPLAR webserver . The averages of MCC/accuracy by DISPLAR are 0 . 39/82 . 5% , which are slightly lower than 0 . 40/82 . 7% from the results by the first ranked model of our method . The difference is statistically insignificant . However , the performance of the best of the top five models by our method , 0 . 54/86 . 7% , is significantly better than that of DISPLAR ( Wilcoxon signed-rank test P<0 . 001 ) . In practice , the multiple ( but limited number of ) models generated by our method can be filtered through incorporation of existing experimental studies on binding-sites , thereby further improving the prediction . The difference between the predicted docking model and the native complex structure may be explained by two main reasons: First , nonspecific instead of specific DNA was used for docking . Second , rigid-body docking does not consider the conformational changes of either the DNA or the protein . The effects of conformational changes in protein are clear as holo-structures consistently produce models closer to the native state than those using apo-structures . In principle , by also taking DNA conformational changes into account , one should be able to obtain improved models . The flexibility problem can be partially addressed through docking the protein to a library of DNA in various conformations [7] . To explore this idea , we constructed a DNA library composed of three poly dA·dT B-DNA structures , whose backbone RMSDs range from 1 to 3 Å with respect to the canonical B-DNA used above , and the canonical B-DNA itself ( see Table S2 ) . For convenience , we name the canonical B-DNA as D0 , and the DNA library as Dlib . Using Dlib , we obtained complex models generated by docking the protein to each DNA in the library . For each of the four protein-DNA combinations , the same docking procedure described above was followed , and the top five clustering models were selected and pooled together . From this pool of twenty clustering models we selected top five models according to their interfacial energy . As shown in Figure 7 , the mean MCCs of DNA-binding residues predictions are improved from 0 . 54/0 . 62 ( D0 docking ) to 0 . 57/0 . 68 ( Dlib docking ) for the APO/HOLO sets , respectively . One can further estimate the upper limit of such improvement by docking holo protein structures to nonspecific DNA that adopts the native specific-DNA conformation , though in general one cannot assume that the nonspecific DNA associates with the protein in exactly the same conformation as the specific DNA . In this estimation , we took the native DNA structures from the 44 complex structures and mutated all base pairs into dA·dT with the program 3DNA [31] . We name this set of DNA structures Dnat . Each protein structure from the HOLO set was then docked to the corresponding DNA structure in Dnat . The resulting average MCC for binding site prediction from the best of top five clustering models is 0 . 71 ( Figure 7 ) , which is slightly higher than 0 . 68 from docking holo protein forms to Dlib . While we expect to see further improvement with fully flexible docking , it poses a challenging problem in practice [7] , [32] . So far , successful examples are limited to local refinement , which requires that the initial rigid body models subjected to flexible refinement are sufficiently close to their native conformation . A thorough study on flexible docking , however , is beyond the scope of the current study .
How a DNA-binding protein locates its specific DNA target sequence is a fundamental , unsolved problem in biology . It has been proposed that association with nonspecific DNA sequences and subsequent travel along the sequence facilitates the search for the specific DNA target sequence [17] , [18] . In this regard , it has been shown that specific DNA-binding proteins , such as transcription factors and restriction endonucleases , can locate target sites at rates several orders of magnitude faster than that estimated by random three-dimensional diffusion , through mechanisms known collectively as facilitated diffusion [17] , [18] . A crucial step of the facilitated diffusion processes involves the association of the protein with a nonspecific DNA sequence; this is followed by one-dimensional sliding along the DNA or hopping over short distances to accelerate the search for a specific DNA target sequence . Despite recent advances that provide visualizations of protein sliding along DNA [33] , the structural details of how a DNA-binding protein associates with a nonspecific DNA remain elusive , primarily due to weak interactions between nonspecific DNA and the protein . Indeed , due to the fact that the interactions are nonspecific , there exist only a few solved atomic structures for nonspecific DNA-protein complexes [24] , [34] . Our study provides useful structural insights into how a specific DNA-binding protein interacts with a nonspecific DNA sequence during the facilitated diffusion process . The similarity between the specific DNA-protein interaction mode and nonspecific interaction modes may reflect an important sampling step in search of its specific DNA targets by a DNA-binding protein . By systematically studying encounter complexes of 44 specific DNA-binding proteins with a nonspecific DNA molecule , we found that the vast majority of these DNA-binding proteins favorably interact with nonspecific DNA at the same binding sites for their specific DNA targets . Using APO/HOLO-structures for docking and a pair potential for energy ranking , we obtained at least one near-native model among the top ten models for 77%/84% of APO/HOLO proteins . In these models , protein residues that contact the nonspecific DNA coincide with those that contact the specific DNA with a MCC>0 . 5 . By introducing a clustering procedure , the most native-like model among the top five cluster representatives has an average MCC of 0 . 54/0 . 62 when APO/HOLO structures are used . Moreover , the DNA-protein interaction modes observed in these models resemble the corresponding native binding modes with specific DNA . The average interfacial RMSD is 4 . 6/3 . 4 Å , and the fraction of native contacts observed is 33%/41% for APO/HOLO proteins , respectively . Our results therefore suggest that a DNA-binding protein frequently samples nonspecific DNA using the same binding sites as used for specific DNA recognition . The results are consistent with a recent Langevin dynamics study on the diffusion of three DNA-binding proteins along nonspecific DNA [35] , and are also consistent with the few available atomic structures of DNA-binding proteins in complex with both specific and nonspecific DNA [24] , [34] . One interesting example is the endonuclease EcoRV , which locates a specific cleavage site through a combination of 1D sliding along nonspecific sequence and 3D jumping [36] , [37] . The nonspecific DNA recognition observed in our top model and in a crystal structure of the nonspecific DNA-EcoRV complex involves the same set of protein residues which also participate in specific DNA recognition [24] . However , the majority of native contacts formed in the cognate DNA-protein complex structure are lost in our model , largely due to the absence of the dramatic bending exhibited by the cognate DNA . The overlap of nonspecific and specific DNA interaction sites on the protein surface allows us to predict DNA-binding residues . The best of top five models generated with holo-structures have an average MCC of 0 . 62 , which is 15% higher than the average MCC of 0 . 54 obtained with apo-structures . Despite the notable difference , the performance of our method is satisfactory for apo-structures . This validation on apo-structures has important practical applications . Going beyond the DNA-binding site prediction , our method also provides models for the DNA-protein interaction modes . For 86%/71% of HOLO/APO structures , at least one of the top five models exhibits an interaction mode somewhat similar to the native binding mode , with a mean RMSDint of 3 . 1/3 . 9 Å and a Fnat of 44%/37% . These complex models are acceptable using CAPRI criteria [19] . The performance of our method in DNA-binding site prediction has been compared with three machine-learning based methods . We note that the top model by our method already performs better than the other methods in terms of MCC and overall accuracy . While machine learning based methods typically provide only one model for assessment , our method generates a limited number of representative models for selection . This can be a great advantage for practical application , since incorporation of existing experimental studies on binding-sites may greatly improve model selection . On average , the best of our top five models by our method achieves a MCC of 0 . 62 and accuracy of 87% , which is significantly better than the MCC of 0 . 49 and accuracy of 81% of DISPLAR [14] , the best among other methods . In addition , our method has the advantage of predicting the binding mode , an ability that the machine-learning methods lack . A downside of our method , however , is that it is computationally more demanding than machine-learning methods , typically requiring hours versus minutes of computation time for one target . Nevertheless , given the widespread availability of computational resources , this is not a significant limitation . Despite these successes , the method is not designed for predicting the specific DNA sequence recognized by a DNA-binding protein; this is a related , yet very challenging problem . Knowledge-based distance-dependent contact potentials at the residue [4] or the all-atom level [29] , [30] , [38] , and physics-based all-atom potentials [39] , [40] , have been applied to predict DNA specificity . While these studies have reported success on a few cases , they are limited to known atomic complex structures or models from closely related complex structures with almost identical DNA-binding interface . Nevertheless , they suggest that a successful approach must address structural flexibility and cooperativity among partners that form a DNA-protein complex . Another interesting question is whether one can use the current approach to determine DNA-binding function given a protein structure . To explore this issue , we applied the method to ∼3 , 000 non-DNA-binding proteins collected previously [11] . Unfortunately , we were not able to derive a practical interfacial energy threshold to differentiate DNA-binding proteins from non-DNA-binding proteins , despite the notable difference of average interfacial energy . For DNA-binding function prediction , the knowledge based approach DBD-Hunter [11] , which requires that the structure of a target protein be related to that of a known DNA binding protein , seems more appropriate . Future efforts may involve expanding the template library for DBD-Hunter by adding complex structure models obtained from the current approach . In the post-genomic era , the rapid progress of structural genomics projects has greatly advanced our knowledge about structural biology . Each year thousands of new protein structures have been determined and deposited to the PDB . In principle , the accumulation of protein structures enables a practical solution to the folding problem through template based modeling [16] . Using the well-established modeling method , TASSER , we have obtained a top ranked protein model within 5 Å from their native structures for over half of the 44 DNA-binding proteins . These models were constructed and refined from homologous/analogues templates with less than 30% sequence identity . We have demonstrated that one can satisfactorily predict DNA-binding sites using these good models . The average MCC and accuracy are 0 . 51 and 84% for the best of top five complex models . This is roughly comparable to the performance when experimentally solved apo-structures are used . Ultimately , the combination of modeling and DNA-protein docking may lead the way to the high throughput prediction of DNA-protein interactions .
A flowchart of the modeling protocol is provided in Figure 1 . In the first step , a DNA-binding protein was docked to a poly ( dA·dT ) 16 B-DNA with the FFT-based rigid-body docking program FTDOCK [5] . A grid size of 0 . 7 Å , a rotation angle step of 12° , and surface thickness of 1 . 2 Å were employed for docking . The B-DNA structure was built with the program 3DNA [31] , using a canonical B-DNA fiber model . The top 10 , 000 docking models ranked by the shape complementarity score were retained . These models were subsequently filtered by the requirement that the protein must contact at least one heavy atom from the two central DNA base pairs . This helps to reduce the redundancy of the models due to the helical symmetry of the DNA and also to remove models in which the protein clashes with DNA termini . The remaining complex models were re-ranked according to their DNA-protein interfacial energy given by ( 1 ) where is a statistical pair potential at the functional group level [11] , and is a surface burial term given by −0 . 02 kT/Å2 × Buried Surface Area ( BSA ) . BSA was calculated with the program NACCESS [41] . The statistical pair potential was developed from an analysis in 179 DNA-protein complex structures [11] . For each target , we derive a corresponding potential by excluding any homologous protein with >35% sequence identity from the 179 complex set and repeat the analysis . The top 2500 energy-ranked models were retained for clustering , which uses the coordinates of the COM of DNA-binding protein residues . The clustering procedure starts by selecting the top energy-ranked model as a clustering seed . All models within a COM distance of 6 Å from the seed are assigned to this cluster , and removed from subsequent clustering . We then repeat this procedure until no model is left . Finally , the clusters were ranked using the average energy of all members in each cluster . From each cluster , we select the lowest energy model as the representative model . A protein residue is assigned to be DNA-binding ( or DNA-interacting ) if at least one heavy atom from the protein residue is within 4 . 5 Å of at least one heavy atom from the DNA . Using this definition , about 18% of protein residues can classified as true DNA-binding in the analysis of the HOLO set . Given the imbalanced nature of the DNA-binding residues and non-DNA-binding residues , the Matthews correlation coefficient is a suitable metric for assessing overlap or prediction of DNA-binding residues between an encounter complex and the native complex . The MCC is defined by [42]where TP , FP , TN , and FN are true positives , false positives , true negatives and false negatives , respectively . A true positive refers to a DNA-binding protein residue observed in the native specific complex . Other performance measures calculated are the following: In the DNA-binding mode analysis , we mapped the nonspecific DNA to the specific DNA by maximizing DNA-protein contact overlap . A DNA-protein contact is defined at the residue level . The RMSD between two structures was calculated using the coordinates of backbone Cα and/or DNA C1′ atoms . The interfacial RMSD was calculated for interfacial protein/DNA residues observed in the native specific-DNA-protein complex structure . The structures of the 44 proteins from the APO/HOLO sets were predicted following the TASSER methodology [16] . Briefly , a target sequence was threaded against a non-redundant protein structure library by the program PROSPECTOR_3 [26] , and the resulting structure templates are used for subsequent model assembly and refinement by the program TASSER , which uses a Monte Carlo replica exchange algorithm for sampling . Note that we excluded any template that shares>30% global sequence identity with the target . The replica trajectories were clustered and representative models generated from these clusters . We built all-atom protein models from the reduced-atom TASSER models with the program PULCHRA [43] . In this study , the top ranked TASSER model is employed for DNA-docking . Four knowledge-based statistical DNA-protein pair potentials were developed from an analysis of 179 non-redundant DNA-protein complex crystal structures [11] . These include three quasichemical potentials at the residue [5] , functional-group [11] , and all-atom [29] levels , and another all-atom potential ( termed RAPDF , residue-specific all-atom conditional probability discriminatory function ) using a different reference state [30] . RAPDF was originally derived using the Bayesian probability formalism [30] , [44]; it can be expressed equivalently under the Boltzmann distribution formalism . Here , we introduce all these potentials using the Boltzmann formalism , which assumes that the frequencies of observed pair interaction states follow a Boltzmann distribution [45] . Consequently , the pair interaction energy E can be deduced from the inverse of Boltzmann's law ( 2 ) where α and β are protein/DNA residues , functional-group , or heavy-atom types for the corresponding potentials , respectively , and , and are the observed and expected frequencies of the αβ pair at the distance d , respectively . For residue and functional-group level potentials , the distance d is defined as the minimum distance between a pair of heavy atoms from the corresponding the αβ pair; and a single distance cutoff of 4 . 5 Å was used . Multiple distance bins from 3 Å to 10 Å with a bin width of 1 Å were employed for the two all-atom potentials . The observed frequency can be obtained by ( 3 ) where denotes the number of observed αβ contact pairs at the distance d . For quasichemical potentials , the expected frequency is given by ( 4 ) where and are the mole fractions of type α and β . The mole fraction for each type is the overall mole fraction in the entire template library , following a scheme known as the composition-independent scale [46] . For RAPDF , the expected frequency is estimated by ( 5 ) For a DNA-protein complex structure , the corresponding DNA-protein interfacial energy is the summation of all observed pair interactions in the structure . The RAPDF parameterization was performed using the program implemented previously [30] . In a benchmark test on the DNA-protein docking decoy set compiled by Robertson and Varani [30] , our new set of RAPDF parameters yield an average Z-score of −11 . 0 for the native complex structures , slightly better than the previous average Z-score of −9 . 6 obtained by parameters determined on a smaller set composed of 52 DNA-protein complex structures . A web-server implementation of the method described here is available at http://cssb . biology . gatech . edu/skolnick/webservice/DP-dock/ .
|
Many essential biological activities require interactions between DNA and proteins . These proteins usually use certain amino acids , called DNA-binding sites , to recognize their specific DNA targets . To facilitate the search of its specific DNA targets , a DNA-binding protein often associates with nonspecific DNA and then diffuses along the DNA . Due to the weak interactions between nonspecific DNA and the protein , structural characterization of nonspecific DNA–protein complexes is experimentally challenging . This paper describes a computational modeling study on nonspecific DNA–protein complexes and comparative analysis with respect to specific DNA–protein complexes . The study found that the specific DNA-binding sites on a protein are typically favorable for nonspecific DNA and that nonspecific and specific DNA–protein interaction modes are quite similar . This similarity may reflect an important sampling step in the search for the specific DNA target sequence by a DNA-binding protein . On the basis of these observations , a novel method was proposed for predicting DNA-binding sites and binding modes of a DNA-binding protein without knowing its specific DNA target sequence . Ultimately , the combination of this method and protein structure prediction may lead the way to high throughput modeling of DNA–protein interactions .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/macromolecular",
"structure",
"analysis",
"biophysics/macromolecular",
"assemblies",
"and",
"machines",
"biophysics/structural",
"genomics",
"computational",
"biology/protein",
"structure",
"prediction"
] |
2009
|
From Nonspecific DNA–Protein Encounter Complexes to the
Prediction of DNA–Protein Interactions
|
Recent years have seen a rapid increase in the number of rabies cases in China and an expansion in the geographic distribution of the virus . In spite of the seriousness of the outbreak and increasing number of fatalities , little is known about the phylogeography of the disease in China . In this study , we report an analysis of a set of Nucleocapsid sequences consisting of samples collected through the trial Chinese National Surveillance System as well as publicly available sequences . This sequence set represents the most comprehensive dataset from China to date , comprising 210 sequences ( including 57 new samples ) from 15 provinces and covering all epidemic regions . Using this dataset we investigated genetic diversity , patterns of distribution , and evolutionary history . Our analysis indicates that the rabies virus in China is primarily defined by two clades that exhibit distinct population subdivision and translocation patterns and that contributed to the epidemic in different ways . The younger clade originated around 1992 and has properties that closely match the observed spread of the recent epidemic . The older clade originated around 1960 and has a dispersion pattern that suggests it represents a strain associated with a previous outbreak that remained at low levels throughout the country and reemerged in the current epidemic . Our findings provide new insight into factors associated with the recent epidemic and are relevant to determining an effective policy for controlling the virus .
Rabies is an enzootic disease that causes severe dysfunction to the central nervous system [1] . While cases are relatively rare in developed countries , the virus has significant impact on a global scale , with more than 55 , 000 deaths reported annually [2] , and represents a major public health issue in many countries . More than half of these cases occur in Asia and China has the second highest incidence of rabies after India [3] , [4]; in the last 60 years several rabies epidemic waves have been reported in China and improving the understanding of how these epidemics emerge can help to determine how to best reduce the likelihood of future outbreaks . Human rabies cases in China decreased during the first half of the 1990s with a low of 159 cases reported in 1996 [5] , [6] , [7] but subsequently the number of human rabies cases increased dramatically , with 3 , 302 cases reported in 2007 [5] . At the same time the geographic distribution and the range of infected hosts has also expanded [3] . There are already many published reports on the phylogenetic relationship amongst strains isolated in China that have primarily focused on sample classification and estimation of features such as date of the most recent common ancestor ( TMRCA ) [3] , [4] , [6] , [7] , [8] , [9] , [10] . In this work we have expanded on the work of previous studies by using a more complete sequence set consisting of sequences spanning a 720 nt region of the nucleocapsid gene that encompasses samples from the entire epidemic region . Additionally , we perform a more extensive analysis of the sequence set and examine the epidemic from a phylogeographic perspective . In 2005 , in order to improve rabies control and prevention , the Chinese government implemented a trial surveillance program to monitor rabies at the national level in an attempt to obtain a more comprehensive epidemiological dataset . In addition to recording statistics on human cases , the Institute for Viral Disease Control and Prevention of China CDC cooperated with the provincial CDC laboratories and began collecting samples from dog populations in regions where human rabies cases had been reported; these samples were then screened for presence of the rabies virus by both DFA and RT-PCR detection . The positive samples were then submitted for DNA sequencing and combined with a second subset of selected sequences from publicly available sequences . Although dogs remain the major infection source , contributing 85%–95% of human cases in China [9] , the number of reported incidences caused by wildlife has also increased [11] , [12] . Thus , as part of this study , we also included available samples from wildlife to examine their contribution to the current epidemic . The final sequence set represents the most comprehensive dataset from China to date , representing 210 sequences ( including 57 new samples ) from 15 provinces and covering all epidemic regions . In this study we use this dataset to investigate the dissemination of the virus across China as the epidemic took hold and we analyze the sequence set in terms of genetic diversity , patterns of distribution and evolutionary history .
Data on human rabies cases in China between 1996 and 2008 were collected from the annual reports of Chinese Center for Disease Control and Prevention ( China CDC ) . Human rabies cases in China are defined according to clinical symptoms and subject's case history ( such as a record of close contact with infected hosts via bites or scratches ) and are confirmed by laboratory test where possible . Although not all patients are confirmed by laboratory test , the typical symptoms of human rabies cases are very distinct and misdiagnosis is unusual . Cases are recorded as part of the national infectious disease reporting system set up by the Chinese government for monitoring several diseases including rabies . If a subject at a local health service center , hospital or other health institute is diagnosed with rabies virus as outlined above this institute has the responsibility of immediately reporting the case to the local CCDC who liaise with the national headquarters . Samples were collected as part of a national surveillance program . In this program , reported human cases of rabies were followed up by visits by provincial CDC laboratories to the region . Since 85%–95% of laboratory-confirmed human rabies cases in China could be associated with a dog bite [9] , specimen collection focused on dog brain samples from meat markets . The markets were selected as they are the principal location where farmers in the neighborhood sell dogs they have raised , or stray and trapped dogs , and are therefore representative of the canine population in the surrounding region . The choice of provinces and municipalities where the samples were collected were determined by incidence rate of reported human rabies cases . In every province there were 9 to15 counties selected for sample collection , which were representative regions of low , middle and high incidence rates of rabies for that province . For each county , several meat markets were chosen as sample collection locations . In this way 3275 samples were collected from the brains of dogs in 7 provinces or municipalities ( Guangxi , Guizhou , Anhui , Zhejiang , Jiangsu , Shandong , and Shanghai ) in China between 2003 and 2008 and screened for the presence of the rabies virus [11] . Fifty-eight brain tissue samples that tested positive for the Rabies Virus by both direct immunofluorescence assay ( DFA ) and RT-PCR [10] , [12] , [13] , [14] were selected for nucleotide sequencing of a 720 nt region of the nucleocapsid gene ( 636 nt to 1353 nt ) as described previously [3] . This region was selected as it represented the most variable section of the gene . Sequence coverage was 4× . All sequences were submitted to Genbank and accession numbers are listed in Table S1 . Additional rabies sequences were downloaded from GenBank and a subset was selected based on the following criteria: ( 1 ) that the sequence spanned the 720 nt region of the N gene from 636 nt to 1353 nt; ( 2 ) the full background information ( isolation time/host/location ) was available . Finally , T-COFFEE was used to identify samples with the greatest nucleotide diversity . This reduced the original set of 176 published sequences to a subset of 153 sequences that provided the greatest coverage of geographical regions and host species . These sequences were isolated from dogs , cats , deer , raccoon dogs , striped field mice ( apodemus agrarius ) and ferret badgers ( Melogale moschata ) from 15 provinces that represented the majority of regions with the most serious rabies problem ( 85 . 5% of reported cases between 1996 and 2008 ) . When combined with the newly sequenced samples they formed a final set of 210 sequences ( Table S1 ) . Phylogenetic trees were constructed based on the 720-nt N-gene sequence ( nt704–1423 ) using the Maximum Likelihood ( ML ) method implemented in the PHYML [15] and PHYLIP [16] software packages . For the PHYML and PHYML trees , the site frequencies were estimated and gamma site variation was selected with 4 categories . For PHYML , the gamma shape value was estimated from the data . PHYML does not use an outgroup so trees were estimated with and without Australian bat rabies virus sequence ABL1996 , the topologies of all the PHYLIP and PHYML trees were compared for consistency and no significant differences were observed . The HKY model was selected using MODELTEST [17] and parameter values for the HKY substitution matrix , base composition and gamma distribution of among-site rate variation were estimated . Bootstrap values were determined for 1000 replicates . To examine the distribution of branch lengths within clade I and clade II , the two major clades identified in the tree , a Java program was used to calculate all tip to tip distances for every leaf node within each clade ( downloadable from http://srlab . whiov . ac . cn/wiv_bioinformatics/treeStr . html ) ) . The non parametric Wilcoxon test implemented in R was then used to compare the two distributions . To investigate the patterns of distribution and geographical structure of the rabies virus in China , isolates in the constructed ML tree were assigned a state according to the province in which they were collected and the tree was examined for discordant sample locations . In previous phylogeographic studies these events are referred to as migration events but , for consistency with common rabies terminology , we refer to them in this paper as translocation events . Translocation events were inferred using a parsimony method with DELTRAN optimization [18] on the constructed ML tree implemented in using the PAUP 4 . 0 [19] and MigraPhyla [20] software packages . Statistical significance was determined using an upper-tail Monte Carlo test of 10 , 000 trials for randomized datasets . A sparse false discovery rate ( sFDR ) correction was used to account for multiple comparisons between pairs . Only clade I and II had sufficient number of sequences for the analysis . To examine the relationship between locations the UniFrac software package was used to generate a distance matrix between all pairs of communities ( i . e . provinces ) based on an estimation of the fraction of the branch lengths of the tree which is unique to each community . Principal Component Analysis ( PCA ) was then used to examine the geographical structure of the data by transforming the matrix such that the greatest variation occurred in the initial principal components [21] , [22] . Evolutionary history , including evolutionary rates of populations ( nucleotide substitutions per site per year ) , TMRCA and population growth models was inferred by using the Bayesian - Markov chain Monte Carlo ( MCMC ) method implemented in the BEAST software package [23] , [24] , [25] , [26] . Demographic histories were inferred by Bayesian skyline reconstruction and statistical uncertainty was expressed by 95% confidence intervals of the Highest Posterior Density ( HPD ) . The constant population size , exponential population growth and logistic population growth models were considered in turn and compared using Bayes Factors [27] . Chain lengths of 5×107 and 4×107 were used for clade I and clade II respectively to ensure estimated samples sizes >100 and topology of the final trees was compared with the PHYML and PHYLIP trees generated in the previous section; no significant differences were observed . To investigate whether the observed phylogeographic structure was simply a consequence of sample size or sampling bias we defined a series of distance matrices according to location for difference in ( i ) sample size , ( ii ) geographical distance between locations , ( iii ) number of translocation events ( iv ) UniFrac distance ( v ) Net Relatedness Index ( NRI ) and ( vi ) Nearest Taxon Index ( NTI ) . NRI provides a measure of the dispersion of a locality throughout a tree , whereas NTI is a measure of the clustering at the leaf nodes [28] . Both these quantities were calculated for both clades using our Java phylogenetic analysis package with 1000 random trees with 100 shuffles . To compare the geographic composition of clade I and clade II a two way contingency table was created according to collection date ( 2003–2005 , 2006–2008 ) and location ( SW , E ) and a Pearson's Chi-squared test with a Yates' continuity correction was used to compare pairs .
ML reconstruction of the 211 RABV partial N gene sequences collected in China divided the isolates into four major clades; clade I , clade II , clade III , and clade IV ( Figure 1 ) . Consistent with previous reports , the majority of these samples were located in clade I and II , corresponding to the Asian branch; within both of these clades further branching of the tree identified several distinct subgroups . All the sequences in clade I are from dogs . Although the clade contains sequences covering all the sampled regions , the older sequences are almost exclusively from the southwest whereas the younger sequences are from the east . This is consistent with the recorded spread of the virus ( i . e . , number of human cases ) from the southwest to the east ( Figure 2 ) . Clade II contains all wildlife samples and the Wilcoxon test indicates this clade has shorter branch lengths overall compared to clade I ( Wilcoxon test W = 5002985 , p-value <2 . 2e-16 ) and shows no clear geographical division , consisting of two subgroups each composed of sequences from both the southwest and the east . Another interesting feature is that sample F01 , isolated from ferret badgers in Zhejiang , is placed at the earliest branch of this clade , and subgroup IIC ( consisting primarily of ferret badger sequences ) is placed at the top of a second large group of samples isolated from dogs . Clade III ( n = 24 ) corresponds to the cosmopolitan branch and represents a more general group of strains that includes isolates from dogs , rats , deer and raccoon dog and also shows no clear geographical segregation . Clade IV is confined to samples from northeastern China and forms the arctic-related branch . Spatial dynamic analysis was used to identify structure in the geographic diffusion of the rabies virus in China at the provincial level . Only clade I and clade II had sufficient sequences for the analysis . All possible pair-wise comparisons were examined ( 110 for clade I and 90 for clade II ) and a number of translocation events were identified with a high level of support ( Figure 3 and Table S2 ) . Specifically , in clade I , Jiangsu province appears to be a main source of translocation events with additional events identified as originating in Guizhou and Henan . Statistically significant events were predicted for inter-province translocation from Jiangsu province in the east to several other eastern provinces ( Shandong , Fujian ( p = 0 ) , Beiing ( p = 0 . 004 ) , Anhui ( p = 0 . 006 ) and Shanghai and Zhejiang ( p = 0 . 05 ) ) . In the southwest of China the only events with strong support were Henan to Hunan ( p = 0 . 01 ) and Yunnan ( p = 0 . 03 ) , and Guizhou to Yunnan ( P = 0 . 03 ) and geographic subdivision is by far the strongest signal in these data . Similar to clade I , migratory centers in southwestern and eastern China were also present in the clade II samples with high support . Most notably a translocation event associated with ferret badgers clade II was predicted between Jiangxi and Zhejiang province in the east with strong support ( p<0 . 004 ) ( Figure 3b ) . UniFrac is a method that was originally developed to calculate a distance measure between bacterial communities based on the dispersion of the two communities within an estimated phylogenetic tree . The program finds taxa in the tree that contain samples from the two communities and counts the number of branches that are shared by both , or that are unique to one or the other community . To determine which provinces share similar evolutionary patterns , UniFrac was used to analyze the geographical structure of the tree by generating a distance matrix between all location pairs . PCA was then used to transform the matrix such that the greatest variation occurs in the first component , the next greatest variation in the second component and so on . The first two principal components explained 45% and 63% of the total variation for clade I and clade II respectively . The first two principal components for clade I and clade II are shown in figure 4a and b . Notably , for clade I , PC1 separates east China from southwestern China and all the eastern provinces are located to the right , with the exception of Jiangsu which is placed closer to the southwest provinces . This is consistent with the multiple translocation events that were predicted to originate from Jiangsu to provinces in the southwest . In clade II , there is no apparent geographical subdivision between southwest and east . By using a Bayesian relaxed clock method , exponential population growth and constant population size was determined to be the most appropriate population model for clade I and clade II respectively . The evolutionary rates of each clade based on the selected population model were 1 . 274×10−3 ( HPD95%: 8 . 3705−4-1 . 2515E−3 ) substitutions per site per year for clade I and 9 . 629×10−4 ( HPD95%: 3 . 519−4-1 . 628E−3 ) substitutions per site per year for clade II . The corresponding TMRCA estimates for clade I , clade II and clade III were 15 . 5 years ( about 1992; 95%HPD ( 10 . 5–20 . 1 years ) ) , 48 . 0 years ( about 1960; 95%HPD ( 16 . 1–112 years ) ) and 117 years ( about 1891; 95%HPD ( 75–211 years ) ) respectively; because there were only two sequences for Clade IV the TMRCA was not estimated . For clade I , the virus spread from SW to E China , constantly encountering new hosts , whereas it seems that clade II was already distributed throughout the country , suggesting it was present at low levels and reemerged more gradually . Thus , the pattern of spread was very different for the two clades , and this may explain the differences in the selected population models . The Skyline plots ( Figure 5 ) indicate clade I and clade II possess different demographic transition patterns . For clade I the genetic diversity increased rapidly from 1994 to 1996 , remained relatively stable until 2001 and then underwent a second phase of rapid increase that continued until 2003 . From 2004 to 2005 , the genetic diversity decreased rapidly and continued a general downward trend until 2008 . For clade II , the genetic diversity remained stable until 2000 when it experienced a rapid increase similar to clade I , although not as pronounced . The diversity subsequently decreased , reaching a minimum in 2005 and has remained almost constant since this time . This variation in genetic diversity in both clades is consistent with estimated times of geographic subdivision and identified translocation events . In clade I the rapid increase in genetic diversity between 1994–1996 and 2001–2003 corresponds to a significant increase in the number of branch nodes , with each sub-clade comprised of sequences from one region and little mixing between regions; this pattern was also observed in clade II between 2000 to 2003 ( Figure 5a & b top ) . From 2004 to 2008 , as the genetic diversity dropped , the number of translocation events in clade I and clade II increased; during this period the number of rabies cases continued to increase ( Figure 5a bottom ) so this could not be responsible for the observed drop . After this time fewer translocation events were predicted but the number of recorded rabies cases began to fall and this probably also contributed to the reduction in genetic diversity . To investigate whether the observed results were due to sampling bias we generated six distance matrices based on differences between the locations and performed a pairwise Mantel test to test for correspondence ( Table S3 information for results ) . NRI and NTI are measures of the dispersion of a location throughout the tree and for both clades there was no correlation between the number of samples and these quantities or for the number of samples and the observed translocation events . As a further test of whether clade I and clade II possess distinct geographical structures we formed a two-way contingency table for the sample data based on sample location ( SW or E ) and sample date ( 2003–2005 & 2006–2008 ) and performed a chi-squared test on the sample data within and between clades ( Table S4 ) . The results indicate that for 2003 to 2005 the geographical subdivision for clade I was distinct from clade II ( p = 0 . 027 ) but for 2006 to 2008 there was no difference .
We have performed the most comprehensive study to date of the spatiotemporal dynamics of the rabies virus in China . While previous studies of the current rabies epidemic in China have focused on the phylogenetic relationship amongst canine rabies isolates , we also attempted to investigate the possible role of wildlife . Our identification of two major clades , clade I and II , is consistent with results from previous studies [3] , [8] , however this report is the first detailed investigation of the properties of these clades and the first demonstration of their distinct characteristics . Clade I is the younger of the two clades and shows geographical structure in terms of translocation events and sample dispersal . Conversely , clade II showed no clear geographical structure , consistent with its older estimation of TMRCA . One commonly voiced concern is that the observed increase in the number of cases in China might simply be attributed to an improved surveillance program and misdiagnosis of rabies cases [29] , [30] . This is unlikely for a number of reasons . Firstly , rabies surveillance data in China also includes background information for each incident and 85%–95% of reported cases can be associated with a dog bite [9] . The distinct late-stage symptoms of rabies together with the history of a dog bite means that , although misdiagnosis of other viral encephalitides remains a possibility , it is unlikely [29] . Additionally , the epidemiology of current rabies cases is not consistent with observed encephalitis outbreaks . Furthermore , although the majority of reported human cases are not verified experimentally , in the cases where it has been possible to perform laboratory diagnosis , there is a very strong correlation between clinical diagnosis and experimental verification [31] , [32] , [33] ( and unpublished data ) . Secondly , there is a misconception that the surveillance program that was introduced in China in recent years represents the first effort to implement a comprehensive surveillance program in the country . What the China National Statutory Notifiable Communicable Disease Reporting System actually represents is the first attempt to coordinate efficient and automated collection of a more extensive surveillance dataset at the national level . Prior to this , data on human rabies cases was still collected at the local level , but was collated manually at the national level . The continuity between the reporting systems is supported by data from earlier periods which shows that even during periods of social unrest , details of rabies cases were recorded and the data shows clear evidence of previous epidemics [9] . What is still uncertain is the degree to which rabies is present in canines and wildlife . Currently , there is no national or local surveillance system for monitoring dog and wildlife rabies and previous estimates are based on case reports which are inconsistent and clearly underestimate the incidence in these populations . The sampling across 15 provinces in this study , although limited , is informative . Of the 3275 samples collected , 58 tested positive for the virus , corresponding to 2 . 8% of the dataset . As the goal of the surveillance program was to collect dog brain samples from areas where human rabies cases had been reported , this does not necessarily reflect the situation at the national level . Nevertheless , this percentage is consistent with earlier studies in China [10] , [28] as well as studies in other countries experiencing rabies epidemics where larger numbers of samples were also collected from dog populations in regions where human cases had been reported and which were subsequently tested in the laboratory . For example , Vietnam reported 2 positives out of 72 ( 2 . 8% ) canine samples in the former province of Hà Tây , and 5 positives out of 53 ( 9 . 4% ) samples in Hô Chí Minh city [29]; Guatemala reported 25%–30% positive in dog populations sampled in the town of Todos Santos [30]; and in Bali in 2010 , 144 out of 3 , 300 ( 4 . 8% ) tested samples were positive which decreased to 67 out of 2311 ( 2 . 8% ) in 2011 after the effects of widespread vaccination [31] . As more dog samples are collected as part of the National Surveillance Program it will be possible to combine these data to obtain more accurate estimates of the prevalence of rabies in the dog population in China . Our results indicate the growth of clade I coincided with the spread of the epidemic , whereas clade II was already present throughout the sampled regions at the earliest stages . This suggests that clade II is from an earlier outbreak and existed at low levels throughout the country . This is also consistent with the earlier TMRCA for this clade and the difference in the distributions of branch lengths for the two clades . Our results reveal the existence of both geographic dispersal and translocation events , and statistical tests indicate that it is improbable that the events are a consequence of sampling bias . Given the relatively small number of identified translocation events , it appears that geographic dispersal plays the major role in the spread of the virus . This is also supported by the observation that the branch order in the tree coincides with epidemiology data that shows that the neighboring provinces of Hunan , Guangxi and Guizhou experienced rabies outbreaks sequentially . In southwest China , Hunan seems to serve as a major source of geographic dispersal as these sequences are widely distributed among the southwestern sub-clades . The identification of translocation hotspots for clade I suggests that this mechanism also aids dissemination of the virus , although the reason why Jiangsu should act as a major translocation source is unclear . Also , because there were already cases reported in all the translocation regions , it is difficult to be certain how much translocation contributed to the epidemic . As more samples become available through the national surveillance program , it will be possible to further investigate these factors . We also investigated the relevance of wildlife in the spread of the virus and there were a number of curious results from our study . Firstly , our phylogenetic analyses placed ferret badger sequences at the top of two distinct sub-clades of samples isolated from dogs . If the rabies in wildlife was a consequence of spillover from dogs , then we would expect to find the wildlife isolates mixed in with the dog samples . This hasn't been reported in previous studies which have either focused on dogs and only contained one or two wildlife samples [3] , [4] , [8] , or contained more wildlife sequences but focused primarily on the epidemiology rather than the phylogenetic relationships between the strains [34] . Secondly , our study found the first evidence of a translocation event in wildlife in China . Although this is an isolated finding , the translocation event has high bootstrap support . Previous studies have investigated rabies in ferret badgers in southeast China . While the number of isolated samples is small [34] , [35] , surveillance data indicates the habitat of the species extends across the entire region in which rabies events have been observed [36] . Also , a recent study reported that apparently healthy ferret badgers in Zhejiang province region had high levels of seroconversion , although this is not consistent with reports of rabies in ferret badgers in this region [34] . There are insufficient samples to draw any definitive conclusions as to whether wildlife plays a significant role in the spread of rabies in China , but our results are nevertheless interesting and further studies would be worthwhile . However , given the size of rural China , obtaining sufficient positive samples remains a formidable challenge . It is worth noting that the current epidemic and associated increase in human cases was coincident with many social changes in the country that facilitated the spread of the disease . Firstly , vaccination represents the most effective approach to controlling rabies [37] , [38] and vaccine production was previously restricted to government run laboratories . However , with the introduction of economic reforms , several private companies also began production of vaccines , some of which failed to meet national standards . This had a major negative impact on vaccination efforts in the country [39] , [40] . Secondly , relaxation of control on dog ownership led to a rapid rise in dog population that was largely uncontrolled in the countryside [41] . This increase in dog population was further exacerbated by the economic reform that lead to creation of businesses in rural areas for selling dog meat , resulting in concentrated populations of unvaccinated animals [42] , [43] . In order to develop an effective vaccination program , it would be worthwhile to investigate how these factors might impact the cost and efficiency of such a program . Thirdly , prior to economic reform , public healthcare ensured post-exposure prophylaxis was generally available , but with the advent of private healthcare , costs became prohibitively high for many people in rural areas [42] . While this isn't connected with the spread of rabies , it did have a major impact on the number of human cases . This analysis on population dynamics and patterns of distribution and differentiation of the virus may help the development of a program for the prevention and control of rabies in China . Specifically , the identification of translocation hotspots suggests that these regions should be given priority in order to reduce the likelihood of reintroducing the virus into vaccinated areas . Additionally , as our results indicate that clade II is evidence of a previous epidemic , this means that the virus had maintained low levels throughout the country for an extended period and was able to rapidly reemerge when suitable conditions prevailed . The presence of these two distinct components in the epidemic needs to be taken into consideration when attempting to implement WHO recommendations [2] in regard to vaccination control programs .
|
Rabies is a major problem in developing countries and responsible for more than 55 , 000 deaths annually . More than half of the cases occur in Asia and China has the second highest incidence of rabies after India . Human rabies cases in China decreased during the early 1990s but the virus began to re-emerge in the latter half of the decade and spread rapidly across the country with a corresponding increase in cases . To try and learn more about the epidemic , in 2006 the government implemented a trial surveillance program to sample and screen canine populations in locations where human cases were reported . In this work we selected a subset of samples ( representative of the entire epidemic region ) for sequencing and investigated the history and origin of the virus in China and examined the variation from a geographical perspective . Our results indicate that the epidemic is primarily composed of a younger strain with a geographical dispersion that was consistent with the recorded spread of the virus and a second older strain that corresponds to a previous epidemic . This second group exhibits a different geographical pattern , and it appears that this strain remained at low levels throughout the country and was able to re-emerge as the epidemic took hold .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"public",
"health",
"and",
"epidemiology",
"spatial",
"epidemiology",
"phylogenetics",
"rabies",
"neglected",
"tropical",
"diseases",
"population",
"biology",
"veterinary",
"science",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonotic",
"diseases",
"epidemiology",
"biology",
"evolutionary",
"systematics",
"viral",
"diseases",
"evolutionary",
"biology"
] |
2012
|
The Spatial and Temporal Dynamics of Rabies in China
|
The emerging arthritogenic , mosquito-borne chikungunya virus ( CHIKV ) causes severe disease in humans and represents a serious public health threat in countries where Aedes spp mosquitoes are present . This study describes for the first time the successful production of CHIKV virus-like particles ( VLPs ) in insect cells using recombinant baculoviruses . This well-established expression system is rapidly scalable to volumes required for epidemic responses and proved well suited for processing of CHIKV glycoproteins and production of enveloped VLPs . Herein we show that a single immunization with 1 µg of non-adjuvanted CHIKV VLPs induced high titer neutralizing antibody responses and provided complete protection against viraemia and joint inflammation upon challenge with the Réunion Island CHIKV strain in an adult wild-type mouse model of CHIKV disease . CHIKV VLPs produced in insect cells using recombinant baculoviruses thus represents as a new , safe , non-replicating and effective vaccine candidate against CHIKV infections .
Chikungunya virus ( CHIKV ) is a mosquito-borne , single-stranded , positive-sense RNA virus ( genus alphavirus ) that has caused sporadic outbreaks every 2–50 years of predominantly rheumatic disease , primarily in Africa and Asia . CHIKV recently ( 2004–2011 ) produced the largest epidemic recorded for an alphavirus with an estimated 1 . 4 to 6 million patients , and imported cases reported in nearly 40 countries including Europe , Japan and the USA . The first autochthonous CHIKV infections in Europe ( Italy in 2007 and France in 2010 ) were also seen during this epidemic . Although Aedes aegypti is the traditional vector for CHIKV , the recent outbreak was associated with the emergence of a new clade of CHIKV viruses , which were efficiently transmitted by Aedes albopictus mosquitoes , a vector that has seen a dramatic global expansion in its geographic distribution [1] , [2] . CHIKV is a biosafety level 3 ( BSL3 ) pathogen and has been declared a Category C Priority Pathogen by the National Institute of Allergy and Infectious Disease ( NIAID ) in the United States . The US Army has long recognized that CHIKV could be used as a biological weapon [3] . The word “chikungunya” is derived from the Makonde language ( Tanzania ) and means “that which bends up” referring to the severe joint pain-induced posture of afflicted individuals . CHIKV disease is characterized by acute and chronic polyarthritis/polyarthralgia , which is usually symmetrical and often incapacitating , with other symptoms such a fever , rash , myalgia and/or fatigue often also present during the acute phase . Arthropathy usually progressively resolves over weeks to months , usually without long-term sequelae; however , CHIKV infections can sometimes cause severe disease manifestations and mortality [2] , [4] . CHIKV is an enveloped virus of ∼70 nm and has an RNA genome of ∼11 , 800 bp [5] . Alphaviral RNA encodes two polyproteins; the non-structural polyprotein and the structural polyprotein . The structural polyprotein is translated from a 26S subgenomic mRNA and is processed into the 5 structural proteins; capsid ( C ) , E3 , E2 , 6K and E1 [6] . The viral RNA is encapsidated in a ∼40 nm nucleocapsid , which is tightly enclosed by a host-derived lipid bilayer envelope displaying the viral envelope glycoproteins E1 and E2 . The glycoproteins are arranged in 80 trimeric spikes composed of three assembled E1–E2 heterodimers . The trimeric spikes are essential for budding of new virus particles , host receptor recognition and attachment ( via E2 ) , and cell entry via pH-dependent endocytosis ( via E1 ) . Upon translation of the structural polyprotein , the capsid protein C is autocatalytically cleaved from the structural polyprotein and encapsidates cytoplasmic viral genomic RNA . The remaining envelope polyprotein ( E3E26KE1 ) is further processed in the endoplasmic reticulum ( ER ) . The resulting membrane bound E3E2 ( also known as precursor E2 or PE2 ) and E1 form heterodimers , with three of these heterodimers assembling to form the trimeric spikes . Prior to surface exposure of the trimeric spikes , PE2 undergoes furin-dependent cleavage to release E3 from the trimeric spike [7] , [8] , [9] . At present , no licensed vaccine or particularly effective drug is available for human use for any alphavirus . A number of pre-clinical CHIKV vaccines have been described , including inactivated virus formulations [10] , [11] , [12] , live-attenuated virus vaccines [13] , [14] , [15] , chimeric virus vaccines [16] , DNA vaccines [17] , [18] , a recombinant adenovirus vaccine [19] , subunit protein vaccines [20] , [21] , [22] and a virus-like particle ( VLP ) formulation [23] . A formalin-inactivated alphavirus vaccine has been shown to be immunogenic in humans [24] . However , growth of large quantities of CHIKV for vaccine manufacture is complicated by the requirement for appropriate BSL3 containment . A live-attenuated CHIKV vaccine ( TSI-GSD-218 ) , although immunogenic , in a human phase II study caused side effects including arthralgia [25] . DNA vaccines have so far not been particularly effective at generating antibody responses in humans [26] , which is a concern as antibodies are believed to be required for protection against CHIKV infections [12] , [27] . VLPs mimic the native virus surface architecture and protein conformation , which often makes them potent inducers of protective antibody responses in the absence of adjuvants [21] . A CHIKV VLP-based vaccine was recently produced by DNA transfection of mammalian cells , and provided protection in both mice and non-human primates [23] , [27] . Although this VLP approach is promising , recombinant baculovirus expression systems out-perform systems utilising DNA plasmid transfection in mammalian cells in a number of areas , in particular cost and scalability [28] . Herein we describe the generation and in vivo testing of a CHIKV VLP vaccine generated using a recombinant baculovirus-insect cell expression system [29] . Baculovirus expression in insect cells has proven to be a safe and efficient method for producing heterologous proteins for research , diagnostics and vaccine development . Protein expression in insect cells has the benefit of accurate protein folding and post-translational processing of , for instance , complex glycoproteins [30] . Veterinary baculovirus-produced subunits or VLP vaccines have been on the market for many years [30] . The first human baculo-based vaccine , the cervical cancer VLP vaccine ( Cervarix , GlaxoSmithKline ) received FDA approval in 2007 [31] . A recombinant influenza virus vaccine ( FluBlok , Protein Sciences ) is currently under final review by the Food and Drug Administration of the USA [32] . These products have paved the way for future licensing of new baculovirus-based pharmaceutical products and/or vaccines . Recombinant baculoviruses have also been used successfully to expressed alphavirus proteins [33] , [34] , alphavirus VLPs [35] and functionally active CHIKV subunits [22] . Insect cells are readily adaptable to suspension cultures , making scalability a key benefit of the baculovirus system; insect-cell bioreactors of 2 , 000 l culture volume are in routine use in industry [36] , [37] . Therefore , we believe the CHIKV VLPs described here will make a safe and effective vaccine amenable to large scale and rapid production .
Adherent Spodoptera frugiperda ( Sf21 ) -cells ( Invitrogen ) were maintained as a monolayer tissue culture in Grace's insect cell medium ( Invitrogen ) , supplemented with 10% foetal bovine serum ( FBS , Gibco ) . Sf9-easy titration ( ET ) cells [38] were maintained as a monolayer cell culture using Sf900II ( Invitrogen ) serum-free medium ( SFM ) , supplemented with 5% FBS and 200 µg/ml Geneticin ( Gibco ) . Recombinant baculoviruses were generated according to the Bac-to-Bac baculovirus expression system , using an adapted Autographa californica nucleopolyhedrovirus ( AcMNPVΔp10Δcc ) backbone [22] , [39] . The cloning fragment of the complete CHIKV-S27 structural polyprotein ( Genbank accession # AF369024 ) was synthetically generated ( GeneArt ) and equipped with AttB recombination sites to enable Gateway cloning ( Invitrogen ) . The 3842 bp CHIKV fragment was cloned into pDONR207 ( Invitrogen ) donor plasmid and subsequently transferred to the pFastBacI analogue pDEST8 ( Invitrogen ) . The CHIKV-S27 structural cassette was then recombined into the AcMNPVΔp10Δcc , resulting in Ac-S27 . Recombinant baculovirus titers were determined by end point dilution assays using Sf9-ET cells and expressed in tissue culture infectious dose 50 ( TCID50 ) /ml . To produce CHIKV VLPs , 8×106 Sf21-cells were seeded in a 75 cm2 culture flask and infected with Ac-S27 at a multiplicity of infection ( MOI ) of 10 TCID50 units per cell . Infections were performed under serum-free conditions on a shaking platform at 27°C and cells were incubated at 27°C for 72 h . Next , cells were separated from the medium fraction by low speed centrifugation . The cell fraction was washed in phosphate buffered saline ( PBS ) and finally stored in 200 µl PBS at −20°C . Secreted protein fractions were precipitated from the medium with 7% ( w/v ) polyethylene glycol ( PEG ) -6000 and 0 . 5 M NaCl for 2 h at room temperature ( RT ) . Pellets were resuspended in 1 ml GTNE buffer ( 200 nM Glycine , 50 mM Tris/HCl , 100 mM NaCl , 1 mM EDTA , pH 7 . 3 ) and loaded on a discontinuous 70% ( w/v ) , 40% ( w/v ) sucrose in GTNE gradient . Sucrose gradients were centrifuged at 27 , 000 rpm ( SW55 rotor , Beckman ) for 2 h at 4°C . The 70%-40% interphase band was isolated and resuspended in 5 ml GTNE buffer . The VLPs were pelleted by centrifugation with 30000 rpm , for 30 min at 4°C . The pellet was resuspended in 50 µl GTNE , checked for integrity by transmission electron microscopy and stored at −80°C . VLPs were quantified based on specific E2 protein content , which was determined by Western analysis and via Bradford protein assay ( Biorad ) and calculated using purified E2 subunit [22] as a reference . Protein expression and processing of infected Sf21-cell fractions and purified CHIKV VLP fractions were analysed by sodium dodecyl sulphate polyacrylamide gel electrophoresis ( SDS-PAGE ) and Coomassie Brilliant Blue ( CBB ) staining . The purified VLPs and cell fractions were denatured in a gel loading buffer containing SDS and β-mercaptoethanol , incubated for 10 min at 95°C and clarified by centrifugation for 1 min at 14 , 000 rpm . After electrophoresis , denatured proteins were transferred to an Immobilon membrane ( Millipore ) for analysis by Western blot ( WB ) . Membranes were blocked in 3% skimmed milk in PBS-0 . 1% Tween-60 ( PBST ) for 1 h at RT or overnight ( ON ) at 4°C . Blocked membranes were washed 3×5 min with PBST and subsequently incubated for 1 h at RT with rabbit polyclonal anti-E1 and anti-E2 [22] , 1∶15 , 000 and 1∶20 , 000 diluted in PBST , respectively . Membranes were washed and treated with alkaline phosphatase ( AP ) conjugated , goat anti-rabbit IgG monoclonal antibodies ( Sigma ) , 1∶3000 times diluted in PBST , for 45 min at RT . Membranes were washed 2×5 min with PBST and 1×10 min with AP-buffer ( 100 mM NaCl , 5 mM MgCl2 , 100 mM Tris-HCl , 0 . 1% Tween 20 , pH 9 . 5 ) . Proteins were detected by NBT/BCIP staining ( Roche ) . Infected cell- and medium-fractions were treated with PNGase F ( New England Biolabs ) to determine the glycosylation status of the CHIKV glycoproteins E1 and E2 . Protein samples were treated with 1 µl denaturing buffer in 9 µl MilliQ for 10 min at 95°C . The denatured proteins were subsequently incubated with 2 µl G7 reaction buffer , 2 µl 10% NP40 buffer , 0 . 5 µl PNGase F in 4 . 5 µl MilliQ for 1 h at 37°C . Treated and non-treated protein samples were analyzed by SDS-PAGE and WB . Sf9-ET cells [38] were infected with Ac-S27 and Ac-GFP with a MOI of 10 TCID50/ml in Sf900-II SFM medium . A recombinant AcMNPV expressing CHIKV 6KE1 ( Ac-6KE1 ) [22] was used as a positive control . The medium ( pH = 6 . 4 ) was supplemented with 0 . 2 mg/ml cholesterol ( Sigma ) as described previously [22] . Cell fusion was induced 72 hpi , by treating the cells for 2 min with acidified medium with pH = 5 . 8 , pH = 5 . 5 and pH = 5 . 0 , respectively . Syncytia formation was analyzed 4 h post treatment by fluorescence light microscopy . To determine surface expression of CHIKV-E1 and -E2 , Sf21-cells were infected with Ac-S27 , with a MOI of 10 TCID50/ml in Grace's insect SFM ( Invitrogen ) . Cells were harvested 72 hpi and washed with PBS . Next , cells were incubated with PBS containing 1∶5000 diluted rabbit α-E1 and rabbit α-E2 polyclonal antibodies , for 1 h at RT . Cells were washed 3×5 min with PBS and treated with 1∶1000 diluted goat-anti-rabbit Alexa fluor 488 ( Invitrogen ) for 1 h at RT . Finally , cells were washed and treated with 1∶100 diluted Hoechst stain for 5 min at RT . Cells were analyzed using fluorescence light microscopy . To analyze CHIKV-VLP production in time , Sf21 cells were infected with Ac-S27 and samples were taken at intermediate time points from 4 h to 69 hpi . The medium fraction was analyzed for the amount of VLPs in triplo by enzyme linked immunosorbent assay ( ELISA ) . ELISA plates ( Greiner Bio-One ) were coated with 2 . 5 µg/ml rabbit α-E2 polyclonal antibodies [22] in coating buffer for 2 h at RT . Plates were washed three times in PBST , and medium samples were loaded for 2 h at 37°C . The plates were washed three times and treated with 1∶500 diluted α-E2 monoclonal antibodies ( 52B2 , provided by Lucas Goh ) for 2 h at 37°C . Plates were washed and incubated for 2 h at 37°C with 1∶500 diluted , AP-conjugated , goat-anti-mouse IgG monoclonal antibodies ( Sigma ) . Finally , plates were washed three times and treated with 1 mg/ml phosphatase substrate ( Sigma ) in substrate buffer for 45 min at 37°C . Absorbance was measured at 405 nm using a FLUOstar Optima ( BMG Labtech ) . Copper 400 square mesh grids ( Veco ) were treated by Argon gas discharge and loaded with 10 µl sample for 2 min at RT . Excess liquid was removed and the grids were washed five times with MilliQ . Finally , grids were treated with 2% uranyl acetate for 15 s , excess uranyl acetate was carefully removed using filter paper . The grids were air dried and analyzed with a JEOL JEM 1011 transmission electron microscope . The purified CHIKV VLPs were used as a vaccine . Female C57/BL6 mice ( 6–12 weeks old ) were vaccinated once subcutaneously on the back above the base of the tail with 0 . 1 µg VLPs or 1 µg VLPs in 50 µl RPMI 1640 medium ( Gibco ) . As negative control , a purified fraction of Ac-GFP infected culture media was used . Binary ethylenimine ( BEI ) -inactivated purified CHIKV was used as positive control and produced as described [12] . Where indicated , the VLP suspension was formulated with 10 µg/mouse Quil A ( Iscotec ) prior to injection . The neutralizing ability of the mouse serum was assayed as described [19] . Serum from each mouse was heat-inactivated at 56°C for 30 min and serially diluted in a 96-well plate . Diluted serum was incubated with 200 TCID50/ml CHIKV of the Réunion Isalnd CHIKV isolate for 2 h at 37°C . Vero cells ( 104/well ) were added to the plate and incubated for 5 days at 37°C . The serum dilution yielding >95% protection against CPE was determined by staining the cells with crystal violet . Determination of the CHIKV-specific IgG1 and IgG2c antibody titers was performed by ELISA as described previously [12] . Female C57BL/6 mice ( 6 to 12 weeks old ) were inoculated with CHIKV ( LR2006-OPY1 ) , and viraemia and foot swelling were determined as described previously [12] , [40] . Foot swelling was monitored by measuring the height and width of the metatarsal area of the hind feet using digital callipers and is presented as a group average of the percentage increase in foot height times width for each foot compared with the same foot on day 0 . All animal experiments were approved by the Queensland Institute of Medical Research ( QIMR ) animal ethics committee and adhered to the Australian code of practice for the care and use of animals for scientific purposes ( NHRMC , Australia; 7th edition 2004 ) . Statistical analysis on antibody responses was performed using SAS , specifically one way ANOVA with Tukey post-hoc test . Statistical analysis on viraemia and foot swelling was performed using IBM SPSS Statistics 19 . For comparison of two samples , the t-test was used when the difference in the variances was less than 4 and skewness was greater than minus 2 and kurtosis was less than 2; otherwise , a non parametric test was used , specifically , Mann-Whitney U test if variance was less than 4 or Kolmogorov Smirnov test if greater than 4 .
A recombinant baculovirus ( Ac-S27 ) was generated to produce CHIKV VLPs by expressing the complete CHIKV-S27 structural polyprotein ( C , E3 , E2 , 6K , E1 ) ( Fig . 1A ) . The coding sequence of the structural polyprotein was cloned downstream the polyhedrin promoter in an AcMNPV backbone , after which Sf21 cells were infected with a MOI = 10 TCID50/ml . Glycoprotein expression in the cell fraction as well as in the medium fraction was analyzed by WB using α-E1 ( Fig . 1B ) and α-E2 ( Fig . 1C ) polyclonal antibodies . A recombinant baculovirus expressing GFP ( Ac-GFP ) was used a negative control . Western analysis of the cell fraction yielded protein bands of ∼50 kDa for α-E1 ( Fig . 1B , lane 2 ) and two bands of ∼50 kDa and ∼57 kDa for α-E2 ( Fig . 1C , lane 2 ) . These sizes correspond to predicted molecular masses of E1 ( 47 . 5 kDa ) , E2 ( 47 . 3 kDa ) and its precursor E3E2 ( 54 . 6 kDa ) , respectively . CHIKV glycoproteins were also detected in the medium fractions after PEG-precipitation ( Fig . 1B and 1C , lane 4 ) . The molecular mass of observed protein bands corresponds to mature E1 and E2 . The PEG-precipitated medium fraction containing the VLPs was subjected to discontinuous sucrose gradient purification . The 70% - 40% intermediate phase was isolated and analyzed on WB ( Fig . 1B , lane 5 ) . This resulted in a further concentration of CHIKV- E1 and E2 . In addition , a ∼30 kDa protein band , corresponding to the predicted molecular weight of CHIKV-C , was readily observed on the coomassie-stained SDS-PAGE gel ( not shown ) , suggesting that the VLPs contain a nucleocapsid . To analyze the glycosylation status of the glycoproteins E1 and E2 , the infected cell fraction and the purified VLPs were treated with PNGase F , which enzymatically removes glycan residues from N-glycosylated proteins fractions . CHIKV-E1 is predicted to be N-glycosylated at N141 , whereas E2 is predicted to be N-glycosylated at N263 and N273 ( Fig . 1A ) [41] . PNGase F treatment resulted in an expected reduction in molecular mass of CHIKV-E1 in both the cell fraction and purified VLPs ( Fig . 1C , lane 2–5 ) , indicating that E1 was efficiently N-glycosylated by the insect cells . The size difference between non-treated and treated samples was significantly larger for E2 than that of E1 , which may suggest that E2 is N-glycosylated at the two predicted loci ( Fig . 1C , lane 2–4 ) . In addition , both protein bands ( presumably E2 and E3E2 , based upon earlier observations [22] ) in the double-band pattern that were found using α-E2 detection ( Fig . 1B and 1C ) , appeared to be fully glycosylated . During natural infections , CHIKV-E1 and -E2 are assembled into trimeric spikes , which are expressed at the surface of the host cell , to enable budding of new virus particles . Expression analysis has made clear that E1 and E2 are both glycosylated and that a fraction of PE2 is processed by furin . To analyze whether the glycoproteins were subsequently translocated to the cell plasma membrane , non-permeable Sf21 cells infected with Ac-S27 were treated with α-E1 and α-E2 to enable immunofluorescence analysis . Treated cells displayed ring-like structures ( Fig . 2 , middle and bottom ) , indicating that the glycoproteins are indeed exposed at the surface of the infected cells . The non-infected mock cells did not reveal these ring-like structures ( Fig . 2 , top ) . So far , experiments have shown that CHIKV E1 and E2 are expressed , processed correctly and exposed at the surface of the infected host cell . This suggests that maturation of the recombinant CHIKV structural proteins appears to correspond to what happens during natural virus infection . To test whether CHIKV E1 retains its functionality as a fusion protein , a pH-dependent syncytia formation assay was performed . Sf9-ET cells were infected with Ac-S27 , Ac-6KE1 and Ac-GFP ( negative control ) . Ac-6KE1 is a recombinant baculovirus that expresses individual fusogenic E1 [22] and was used as a positive control . Infected cells were treated for 2 min with acidified medium ( pH = 5 . 8 , pH = 5 . 5 and pH = 5 . 0 ) and screened for syncytia formation 4 h post treatment . Syncytia formation at pH = 6 . 4 , pH = 5 . 8 , pH = 5 . 5 was readily observed in cells expressing CHIKV structural proteins ( Ac-S27 ) ( Fig . 3 , right ) and in the positive control ( Ac-6KE1 ) ( Fig . 3 , middle ) but not in the negative control ( Ac-GFP ) ( Fig . 3 , left ) . Syncytia observed in Ac-S27 infected cells were slightly more abundant and also larger in size , suggesting that E1 in its native conformation , i . e . in trimeric spikes closely associated with E2 , displays increased fusogenic activity as compared to individual E1 expression in Ac-6KE1 infected cells . In contrast to Ac-S27 and Ac-6KE1 , Ac-GFP infected cells were only able to fuse when treated with acidified medium of pH = 5 . 0 ( Fig . 3 , left ) . This was due to the pH-dependent activity of the baculovirus fusion protein GP64 , which becomes active only at values below pH = 5 [42] . The formation of syncytia correlates with the presence of E1 on the surface , individually expressed or expressed as a part of the CHIKV structural polyprotein . Therefore , it can be concluded that E1 retains its fusogenic properties when expressed in Sf-cells . Baculovirus expression of the complete CHIKV structural cassette in insect cells leads to correct processing of the CHIKV glycoproteins E1 and E2 , which both are exposed on the surface of the host cell . Western analysis of the medium fraction indicated that both E1 and E2 are present in the medium and can be concentrated by discontinuous sucrose gradient purification . These findings strongly suggest that VLPs were formed and were secreted in the medium . To analyze VLP production in time , a time-course expression assay was performed and CHIKV-VLPs were detected by a sandwich-ELISA , using α-E2 antibodies ( Fig . 4A ) . Sf21-cells were infected in duplo with Ac-S27 and medium samples were taken within 7 to 9 h intervals . VLP production in Sf21-cells initiated at ∼28 hpi , typical for polyhedrin expression . Production peaked at ∼61 hpi and longer production periods , yielded lower amounts of VLPs probably due to cell death . For the final verification of CHIKV-VLP production , insect cells were infected with Ac-S27 and the medium fraction was subjected to discontinuous sucrose gradient purification . The isolated VLP fraction was analyzed by transmission electron microscopy ( TEM ) ( Fig . 4B ) . Spherical , enveloped particles of ∼65–70 nm , were detected in large numbers and were absent in control infections with Ac-GFP . The isolated VLPs varied in size , within a range of ∼55–80 nm . The diameters of 200 VLPs were measured to determine a relative size distribution ( Fig . 4C ) . The CHIKV VLPs fitted a size of 68±14 nm , which is consistent with the reported size ( 65–70 nm ) of alphavirus virions . The specific E2 protein content of the isolated VLPs was determined at 40 mg/l . To assess the immunogenicity of the CHIKV-VLPs , C57/BL6 mice were vaccinated once with 0 . 1 µg or 1 µg of the VLPs , 1 µg of the VLPs formulated with Quil A adjuvant or 10 µg inactivated CHIKV ( positive control ) . The negative controls were PBS and a GFP control . For the GFP control , Sf21-cells were infected with Ac-GFP and the infection medium was treated under exactly the same conditions as the CHIKV VLPs . All groups receiving CHIKV antigens generated neutralizing antibody titers that are significantly different ( P<0 . 01 ) from both control groups ( PBS and Ac-GFP ) but not significantly different from one another ( Fig . 5A ) . The 1 µg VLP dose induced neutralizing antibody titers comparable with those observed after vaccination with 10 µg of inactivated CHIKV ( Fig . 5A ) . CHIKV-specific IgG1 and IgG2c titers were determined by ELISA , and broadly similar titers were seen for IgG1 and IgG2c ( Fig . 5B , C ) , which contrasts with natural CHIKV infection where IgG2c dominates [12] . At the 1 µg dose , VLPs produced ∼10–20 fold , significantly lower antibody titers ( IgG1 P<0 . 05 , IgG2c P<0 . 01 ) than 10 µg inactivated CHIKV . The 0 . 1 µg VLP dose showed a further ∼10–20 fold reduction in antibody titers compared to 10 µg inactivated CHIKV ( IgG1 P<0 . 01 , IgG2c P<0 . 01 ) ( Fig . 5B , C ) . The addition of Quil A significantly ( P<0 . 05 ) increased IgG2c but not significantly IgG1 titers by ∼5–10 fold ( Fig . 5B–C ) . In summary , a single 1 µg dose of insect-cell derived CHIKV VLPs induced potent CHIKV-specific neutralizing antibody and IgG titers . Vaccinated mice were challenged 6 weeks post-vaccination with a Réunion Island CHIKV isolate using a recently developed adult wild-type mouse model of CHIKV viraemia and arthritis [12] . PBS-vaccinated animals and animals vaccinated with the GFP control group showed similar viraemia; >108 TCID50/ml at 2 d post challenge ( Fig . 6A ) . All animals vaccinated with 1 µg VLPs were completely protected against viraemia ( Fig . 6A ) . Mice vaccinated with 0 . 1 µg VLPs , showed a ∼7 log reduction in viraemia on day 2 , with virus undetectable on the other days ( Fig . 6A ) . Vaccination with 10 µg of inactivated virus also provided complete protection against viraemia as described previously [12] . Arthritis in this model is readily determined by measuring foot swelling and , as expected [12] , the inactivated virus completely protected against foot swelling , whereas animals given PBS showed a mean 60–70% increase in foot swelling ( Fig . 6B ) . The 1 µg dose of VLPs ( with or without Quil A ) provided complete protection against foot swelling , while the 0 . 1 µg dose reduced the peak foot swelling from 60–70% to 20–30% ( Fig . 6B ) . These results illustrate that non-adjuvanted VLPs can provide complete protection against CHIKV-induced arthritis .
In this study , recombinant baculoviruses were used to produce CHIKV VLPs in Sf21 insect cells . The VLP proteins were correctly processed and the VLPs provided complete protection against CHIKV viraemia and arthritic disease in a mouse model after a single dose of CHIKV VLPs . The complete CHIKV structural cassette was cloned downstream the strong polyhedrin promoter of the AcMNPV baculovirus ( Ac-S27 ) . Western analysis on cell and medium fractions of Sf21-cells that were infected with Ac-S27 indicated that glycoprotein processing ( glycosylation , furin cleavage and surface localization ) is efficient and that the functionality of E1 as fusion protein is retained . Western analysis on the precipitated medium fraction and purified VLPs shows that only fully matured and glycosylated E2 was incorporated into the VLPs , which suggest that the recombinant VLPs are homogenous . This contrasts with previous findings that uncleaved E3E2 is present in progeny alphavirus particles [43] , [44] . However , both immature ( presumably E3E2 ) and mature , furin-cleaved E2 fractions were found intracellularly , most likely a result of the very high expression levels of the CHIKV proteins . Alphaviral processing intermediates are commonly found in many different expression systems , including recombinant baculoviruses [33] , [34] , [45] , [46] . The triple-banded E2 pattern previously observed upon individual expression of E3E2 [22] was not found after expression using Ac-S27 , indicating that in this case all glycoproteins were efficiently glycosylated . The postulated number of N-glycosylation sites of E1 ( n = 1 ) and E2 ( n = 2 ) correspond to the protein size shifts after PNGase F treatment . These results demonstrate that the processing efficiency of E3E2 in insect cells increases when CHIKV glycoproteins E1 and E3E2 are co-expressed as part of a polyprotein , when compared with expression of E3E2 by itself [22] . To obtain more insight in the exact number and type of glycan-moieties on both E1 and E2 , in-depth studies are required ( e . g . mass spectrometry ) , which can precisely indicate to what extent the glycoprotein intermediates are processed . In addition to correct processing , CHIKV-E1 and E2 were found exposed on the surface of the infected insect cells . The green fluorescent rings found after immunostaining Ac-S27 infected insect cells ( Figure 2 ) , mark the final stage of processing and translocation of the glycoproteins within the host cell , just prior to the budding of the VLPs . In addition , the retained fusogenic function of E1 was shown by treatment of Ac-6KE1 , Ac-GFP and Ac-S27-infected Sf9-ET cells with acidified culture medium ( pH = 5 . 8 , 5 . 5 and 5 . 0 ) resulting in increased syncytia formation . This was not a consequence of the baculovirus GP64 fusion protein , as the control baculovirus Ac-GFP only formed syncytia at pH = 5 . 0 [42] . The VLPs were efficiently isolated using discontinuous sucrose gradient purification . Transmission electron microscopy ( TEM ) analysis revealed that the VLPs were morphologically similar to CHIKV and other alphavirus VLPs and that they had a similar diameter of 68±14 nm [23] , [35] . The overall baculovirus CHIKV VLP yield ( 40 mg/L ) appeared to be higher than in another study , in which VLPs were produced by DNA-transfection of 293F cells ( 10–20 mg/L ) [23] . This underscores one of the major advantages of the baculovirus-insect cell system for production of recombinant proteins . A further gain in VLP yield is expected in an optimized large-scale insect-cell bioreactor configuration [37] . We show herein that a single vaccination with 1 µg of unadjuvanted CHIKV VLP vaccine was able to completely protect mice from CHIKV-induced viraemia and arthritis in an adult wild-type model of CHIKV arthritis that recapitulates many aspects of the rheumatic human disease , i . e . self-limiting arthritis ( joint inflammation ) , tenosynovitus and myositis [12] . The value of adjuvants in this system remains to be fully explored , with different adjuvants , adjuvant doses and VLP∶adjuvant ratios and their effect on protection needing to be analyzed . The importance of protection studies is highlighted by the ability of Quil A to increase CHIKV-specific IgG titers , but reduce neutralizing titers . A recent human trial also illustrated that aluminum hydroxide adjuvant provided increased immunogenicity for an inactivated Ross River vaccine [47] . The VLPs induced a balanced IgG1/IgG2c response , in contrast with CHIKV infection where IgG2c responses dominate [12] . Whether this would have an important effect on protection is unclear . However , we have recently found that there was no difference in viraemia or arthritis following CHIKV infection of mice deficient in the Fc receptor common gamma chain ( unpublished data ) , suggesting the distinct Fcγ receptor-binding properties of the different antibody isotypes [48] does not play a major role in protection ( at least in mice ) . Interestingly , the 1 µg non-adjuvanted VLP vaccine elicited a similar neutralizing antibody response as compared to 10 µg of inactivated virus ( Fig . 5A ) , yet induced lower IgG2c titers ( Fig . 5C ) . This might suggest that chemical inactivation results in generation of antibodies with reduced neutralizing activity , although further experimentation would be required to confirm this . Previous vaccination studies using CHIKV VLPs have shown that VLP provided protection in mice and non-human primates against CHIKV infection [23] . In those experiments , two immunizations of 19 µg VLPs with adjuvant were needed to induce protection in mice , while our data show that a single vaccination of 1 µg of non-adjuvanted VLPs induces complete protection against viraemia and foot swelling . Although a different CHIKV strain ( West-African isotype , also known as strain 37997 ) was used to produce VLPs in 293F cells [23] , the baculovirus-insect cell expression system may also provide VLPs with better immunogenicity . Glycosylation patterns in insect cells differ from those in mammalian cells , in that lepidopteran insect cells do not process N-glycans to terminally sialylated complex-structures . Differences in glycan processing have been shown to influence glycoprotein immunogenicity [49] , [50] , [51] . Four different CHIKV strains have been described so far , including the East- , Central- and South African ( ECSA ) strain , the West-African strain , the Asian strain and the recent Réunion Island strain . The produced VLPs are of the ECSA strain , while the immunized mice were challenged with the Réunion Island strain . Even though these strains belong to the same ECSA phylogroup , it has been shown that ECSA strain based vaccines are able to cross-neutralize against other CHIKV strains and even other alphaviruses from the same serogroup [12] , [14] , [19] . Therefore , our recombinant CHIKV VLPs would be expected to provide protection against most , if not all , CHIKV strains . The favorable properties of the recombinant baculovirus-insect cell expression system renders it extremely powerful in the production of subunit or VLP based vaccines . Even though baculoviral replication is lytic to insect cells and heterologous protein production is therefore not continuous , the sheer expression levels reached are high , if not the highest , of all eukaryotic expression systems [28] . Baculovirus expression vectors can be quickly generated and therefore this system is ideally suited for generating emergency vaccines ( “pandemic preparedness” ) [52] . In addition , insect cells can easily be scaled up in serum-free suspension culture to large culture volumes [53] . Conditions for optimal stability of VLPs produced under serum-free conditions should be determined when this vaccine candidate is further developed by the industry , but so far we have no indications that storage at −80°C is detrimental to VLP integrity . In conclusion , we have shown that complex structures such as CHIKV VLPs are produced at high levels and were efficiently processed and glycosylated in insect cells using recombinant baculoviruses . More importantly , this is the first study that shows that a single low-dose immunization with 1 µg of non-adjuvanted CHIKV VLPs provided complete protection against viraemia and foot swelling caused by CHIKV infection . We propose CHIKV VLPs produced by insect cells using recombinant baculoviruses to be further developed as a safe and effective vaccine candidate to protect humans against CHIKV outbreaks .
|
Viruses that are transmitted by mosquitoes represent major threats for human health all over the world . One of these viruses is the Chikungunya virus ( CHIKV ) . CHIKV is transmitted by the Asian Tiger mosquito , which is making ground to more temperate regions such as Europe , and thereby increasing the risk of CHIKV infections . The virus causes severe fevers and long lasting joint pains . Unfortunately , there is no vaccine to combat CHIKV infections . This study describes the development of a virus-like particle ( VLP ) vaccine against CHIKV infections , which is produced in insect cells . VLPs are structurally identical to the wild type virus , but these particles cannot replicate due to the absence of the viral genome . The CHIKV VLPs that were produced using the baculovirus-insect cell expression system , were correctly produced and mimic live CHIKV in structural organisation and protein function . Interestingly , a single administration of a low dose ( 1 µg/mouse ) of non-adjuvanted VLPs induced robust neutralizing antibody titers and provided complete protection upon CHIKV challenge against viraemia and disease symptoms . This new effective , safe and scalable vaccine candidate represents a step forward in the prevention of CHIKV infections .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"viral",
"vaccines",
"adaptive",
"immunity",
"viruslike",
"particles",
"emerging",
"infectious",
"diseases",
"immunity",
"virology",
"neglected",
"tropical",
"diseases",
"biology",
"microbiology",
"arboviral",
"infections",
"infectious",
"disease",
"control"
] |
2013
|
Effective Chikungunya Virus-like Particle Vaccine Produced in Insect Cells
|
In female mammals , activation of Xist ( X-inactive specific transcript ) is essential for establishment of X chromosome inactivation . During early embryonic development in mice , paternal Xist is preferentially expressed whereas maternal Xist ( Xm-Xist ) is silenced . Unlike autosomal imprinted genes , Xist imprinting for Xm-Xist silencing was erased in cloned or parthenogenetic but not fertilized embryos . However , the molecular mechanism underlying the variable nature of Xm-Xist imprinting is poorly understood . Here , we revealed that Xm-Xist silencing depends on chromatin condensation states at the Xist/Tsix genomic region and on Rnf12 expression levels . In early preimplantation , chromatin decondensation via H3K9me3 loss and histone acetylation gain caused Xm-Xist derepression irrespective of embryo type . Although the presence of the paternal genome during pronuclear formation impeded Xm-Xist derepression , Xm-Xist was robustly derepressed when the maternal genome was decondensed before fertilization . Once Xm-Xist was derepressed by chromatin alterations , the derepression was stably maintained and rescued XmXpΔ lethality , indicating that loss of Xm-Xist imprinting was irreversible . In late preimplantation , Oct4 served as a chromatin opener to create transcriptional permissive states at Xm-Xist/Tsix genomic loci . In parthenogenetic embryos , Rnf12 overdose caused Xm-Xist derepression via Xm-Tsix repression; physiological Rnf12 levels were essential for Xm-Xist silencing maintenance in fertilized embryos . Thus , chromatin condensation and fine-tuning of Rnf12 dosage were crucial for Xist imprint maintenance by silencing Xm-Xist .
A previous study using XmXm embryos ( parthenogenotes ) showed that loss of H3K9me3 via Kdm4b , which is a H3K9me3 demethylase [12] , or gain of histone acetylation by trichostatin A ( TSA ) treatment , induced Xm-Xist derepression [4] ( S1 Fig ) . More recently , chromatin decondensation was shown to be associated with Xist expression [13] . Thus , we first investigated whether the chromatin condensation states of Xm-Xist/Tsix regions at the 2- and 4-cell stages could be altered by Kdm4b overexpression and TSA treatment using XmXm embryos ( Kdm4b/TSA-XmXm ) ( Fig 1a ) . We also examined the chromatin states of androgenetic embryos with Xist RNA positive alleles [14] . DNA fluorescence in situ hybridization ( DNA-FISH ) analysis around Xist/Tsix genomic regions revealed that Kdm4b/TSA-XmXm embryos showed significantly relaxed chromatin states in both stages compared with Egfp/DMSO ( control ) -XmXm embryos , although the chromatin was the most relaxed in Xp at both stages ( Fig 1b ) . We next examined whether Kdm4b/TSA treatment could induce Xm-Xist derepression in XmXp and XmY embryos ( fertilized embryos ) , respectively , at the 4-cell stage ( Fig 1c ) . RNA combined with DNA-FISH ( RNA/DNA-FISH ) analysis showed that Xm-Xist derepression was observed in 7 . 6% of Kdm4b/TSA-XmY cells and that 13 . 5% of Kdm4b/TSA-XmXp cells showed biallelic expression ( Fig 1d ) . Thus , these results indicated that the loss of H3K9me3 and gain of histone acetylation induced chromatin decondensation at Xm-Xist/Tsix genomic regions , resulting in Xm-Xist derepression . Although these results indicated that the chromatin alterations induced Xm-Xist derepression in XmY and XmXp embryos , the induction efficiency was low compared with XmXm embryos ( Fig 1d and S1b Fig ) . In comparison , a previous study showed that the sole induction of Kdm4b mRNA sufficiently induced Xm-Xist derepression in XmXm 4-cell embryos [4] . Notably , it has been shown that the transcriptional capacity of maternal pronuclei in haploid parthenogenetic embryos ( hPE ) was higher than that of paternal and maternal pronuclei in zygotes [15] . Furthermore , although histone H4 acetylation was predominantly imposed on the paternal pronuclei in zygotes , maternal pronuclei in parthenogenetic embryos exhibited an H4 acetylated state [16] . These results suggested that the absence of the paternal genome during pronuclear formation might provide a transcriptionally permissive state within the maternal genome . Consistent with this notion , very few XmXm embryos showed Xm-Xist derepression at the 4-cell stage , although Xm-Xist was never expressed in XmY and XmXp embryos [4] ( S1b Fig ) . In light of these findings , we speculated that the presence of paternal genome during pronuclear formation would impede Xm-Xist derepression . In order to inspect the possibility , we constructed Kdm4b overexpressing bi-parental embryos wherein the parental pronuclei were of different derivation: the maternal pronucleus was formed by SrCl2 activation whereas the paternal pronucleus was formed by in vitro fertilization . Then , to produce bi-parental embryos , paternal pronuclei were transferred into haploid maternal embryos derived from SrCl2 activation ( Fig 1e ) . At the 4-cell stage , Xist RNA-FISH analysis revealed that the constructed bi-parental embryos with Kdm4b overexpression showed marked increase of the cells with Xm-Xist derepression in XmY embryos ( Fig 1f: 37 . 5%; 4 . 9 fold compared with Kdm4b/TSA-XmY in Fig 1d ) and with bialleleic Xist expression in XmXp embryos ( Fig 1f: 39 . 5%; 2 . 9 fold compared with Kdm4b/TSA-XmXp in Fig 1d ) , whereas control embryos showed no Xm-Xist derepression in XmY embryos and only a cell was biallelic expression in XmXp embryos ( 2% ) ( Fig 1f ) . The combination of TSA with Kdm4b mRNA injection was also able to induce Xm-Xist derepression ( Fig 1f ) . Thus , these results indicated that the presence of the paternal genome during pronuclear formation impeded Xm-Xist derepression by chromatin alterations . However , the question remained whether Xm-Xist could be derepressed irrespective of the presence of the paternal genome during pronuclear formation when the maternal chromatin was sufficiently decondensed before fertilization . To test this , we constructed oocytes with decondensed maternal chromatin derived from hPE at the morula stage ( Fig 1g ) , because a previous study had indicated that Xm-Xist/Tsix genomic regions in XmXm embryos became gradually relaxed during preimplantation phases [13] . The constructed oocytes were subjected to fertilization and resulted in semi-cloned embryos ( Fig 1g ) . At the 4-cell stage , Xist RNA-FISH analysis revealed that 78 . 7% of the cells in XmY semi-cloned embryos exhibited Xm-Xist derepression ( Fig 1h: 10 . 3-fold increase compared with Kdm4b/TSA-XmY in Fig 1c ) and 36 . 8% of cells in XmXp semi-cloned embryos showed biallelic expression ( Fig 1h: 2 . 7-fold increae compared with Kdm4b/TSA-XmXp in Fig 1c ) . In contrast , in control embryos ( spindle-exchanged oocytes ) , we did not observed Xm-Xist derepression in XmY embryos or biallelic expression in XmXp embryos ( Fig 1h ) . Taken together , these results demonstrated that Xm-Xist could be derepressed if the chromatin was decondensed even when the paternal genome was present during pronuclear formation , indicating that chromatin condensation of the maternal genome represents the primary factor for imprinting maintenance to silence Xm-Xist . The condensation could be relaxed by loss of H3K9me3 and gain of histone acetylation . Next , we asked whether the derepression of Xm-Xist during the early preimplantation stages could be stably maintained . To facilitate the analysis of Xm-Xist derepression state in female embryos , we used female embryos devoid of Xp-Xist expression because of a paternal deletion in the repeat-A region [10] ( Fig 2a ) . We first checked Xist expression states by RNA/DNA-FISH analysis at the 4-cell stage , demonstrating that 32 . 1% of Kdm4b/TSA-XmXpΔ cells but only one Egfp/DMSO-XmXpΔ cell ( 3 . 7% ) exhibited an Xist signal ( Xist+ ) ( Fig 2b ) . Furthermore , 10 . 7% of Kdm4b/TSA-XmXpΔ cells showed biallelic expression ( Fig 2b ) . Given that Egfp/DMSO-XmXpΔ embryos showed no Xist cloud and biallelic expression , the results indicated that histone modification alteration induced Xist expression on not only Xm but also XpΔ alleles . We further examined whether Xm-Xist derepression could be stably maintained through late preimplantation stages . In blastocysts , RNA/DNA-FISH showed that 13 . 3% of Kdm4b/TSA-XmY and 24% of Kdm4b+TSA-XmXpΔ embryos exhibited robust Xist expression states ( >50% of cells ) , whereas no Xist expression was found in Egfp/DMSO treated embryos ( Fig 2c ) . We also examined the effect of Kdm4b induction alone on Xist expression . Although Xist expression was induced in XmXpΔ embryos , no embryos of either gender exhibited >50% of Xist positive cells ( S2 Fig ) , indicating that both H3K9me3 loss and the absence of histone deacetylases were required for strong Xist induction . The antisense RNA for Xist , commences around the blastocysts stage [17 , 18] . Therefore , to investigate allele-specific Xist expression , we performed strand-specific reverse transcription polymerase chain reaction ( RT-PCR ) analysis . This demonstrated that Xm-Xist expression was clearly induced in Kdm4b/TSA-XmXpΔ embryos ( Fig 2d ) . Thus , Xm-Xist derepression at early preimplantation phases could last until the late phases of preimplantation . To further examine Tsix expression states , we performed RNA/DNA-FISH analysis using Tsix-specific detection probes ( S3a Fig ) . Kdm4b/TSA treatment resulted in an increase of cells with Xist but not Tsix in XmY and XmXp embryos ( S3b and S3c Fig ) . Quantitative PCR ( qPCR ) analysis also confirmed the lack of Tsix upregulation although some X-linked genes , i . e . , Pgk1 and Plac1 , were downregulated in Kdm4b/TSA-XmY and -XmXpΔ embryos compared with Egfp/DMSO treated embryos ( S3d and S3e Fig ) . We also confirmed H3K27me3 enrichment in some cells in Kdm4b/TSA-XmY or -XmXpΔ blastocysts [19] , indicating that the normal XCI process occur in Kdm4b/TSA treated embryos ( S3f Fig ) . To gain further insights into transcriptome states , we performed RNA deep sequencing ( RNA-Seq ) analysis using an individual XmXpΔ embryo with Kdm4b/TSA , Egfp/DMSO , and wild-type ( WT ) . Notably , hierarchical clustering analysis indicated that the transcriptome states of two Kdm4b/TSA-XmXpΔ ( Kdm4b/TSA-1/2 ) embryos resembled those of WT ( Fig 2e ) , indicative of normal X-linked genes expression in the Kdm4b/TSA-1/2 embryo . Consistent with this , hierarchical clustering based on X-linked genes showed that Kdm4b/TSA-1/2 clustered with WT embryos ( Fig 2f ) . Out of 331 X-linked genes expressed , 23 . 6% and 13 . 9% were upregulated in Egfp/DMSO and Kdm4b/TSA-3/4/5/6 embryos , respectively ( Fig 2g and S1 Table ) . However , only , 6 . 0% were upregulated in Kdm4b/TSA-1/2 embryos ( Fig 2g and S1 Table ) . Although group specific differentially-expressed genes were also identified ( S4 Fig ) , these results indicated that Kdm4b/TSA treatment induced Xm-Xist derepression and reduced the number of upregulated X-linked genes in XmXpΔ embryos . To test whether transient histone alterations in preimplantation phases could lead to stable Xm-Xist derepression in embryonic- and extraembryonic tissues and rescue the lethal phenotype of XmXpΔ embryos without gene manipulation , we conducted embryo transfer experiments and assessed the developmental ability of XmXpΔ embryos . We transferred 362 Egfp/DMSO- and 235 Kdm4b/TSA-blastocysts that were recovered at embryonic day 19 . 5 , identifying 60% and 71% of implantation sites in Egfp/DMSO- and Kdm4b/TSA-groups , respectively ( Fig 3a ) . We obtained 26 XmY pups ( 96 . 3% of total pups ) from the Egfp/DMSO treatment . Unexpectedly , one XmXpΔ pup was also born in the group ( Fig 3b ) . However , we have not yet replicated this result . In contrast , 8 XmXpΔ ( 27 . 6% ) and 21 XmY ( 72 . 4% ) pups were born in the Kdm4b/TSA group ( Fig 3b ) . RT-PCR analysis of embryonic- and extra-embryonic tissue from rescued XmXpΔ embryos exhibited predominant Xm-Xist expression ( Fig 3c ) . The rescued females displayed normal reproduction and gave birth to viable offspring ( Fig 3d ) . Taken together , these results clearly demonstrated that the Xm-Xist compensated for imprinted XCI and exhibited the functional equivalency of Xm-Xist to Xp-Xist in both embryonic- and extraembryonic tissues . One of the remaining questions is the molecular mechanisms involved in loss of the Xm-Xist imprint ( Xm-Xist silencing ) in XmXm morula embryos despite the imprint maintenance in XmY and XmXp embryos [4 , 11] . As the chromatin at Xist/Tsix genomic loci was gradually relaxed during preimplantation development [13] and Xm-Tsix began to be expressed around the morula stage [17] , we investigated the Tsix expression state . RNA-FISH analysis for Xist and Tsix revealed that Tsix was also not detected until the 16-cell stage in XmXp , XmY , and XmXm embryos . In XmXp embryos , the cells showing a Tsix signal or an Xist cloud averaged 23% and 94% , respectively ( S5 Fig ) . XmY embryos contained 34% of cells with Tsix expression and no Xist expression ( S5 Fig ) . In XmXm embryos at the 16-cell stage , an Xist cloud was observed in 34% of nuclei . Notably , the ratio of Tsix expressing cells in XmXm embryos was less than that in XmXp and XmY embryos ( S5 Fig , XmXm: 18% ) . Considering that Xm-Tsix was present in two copies in XmXm embryos , these results implied that Xm-Xist derepression in XmXm embryos might be associated with the repression of Tsix . Previous studies indicated that the dose-dependent Xist activator , RNF12 [6 , 7] , was highly expressed in early preimplantation phases [4] . RNF12 activates Xist via REX1 protein degradation and REX1 plays a role in Tsix elongation [20 , 21] . Thus , we performed single cell qPCR assays against Rnf12 mRNA using XmY , XmXp , and XmXm preimplantation embryos . To identify the sex of the cells in fertilized embryos at the 2-cell stage , DNA-FISH analysis was conducted in the remaining cell in each embryo not used for qPCR analysis . From the 4-cell stage onward , Xist expressing cells were defined as female . The analysis exhibited that Rnf12 expression levels were markedly higher in oocytes and at the 1-cell stage ( Fig 4a ) . Although Rnf12 expression levels gradually decreased in most of the embryos as the embryos developed , the levels of XmXm ( Xist− ) cells tended to be high compared with those of XmXp from the 2-cell stage onward ( Fig 4b ) . At 8-cell and morula stages , the Rnf12 levels of XmXm ( Xist+ ) cells were downregulated compared with XmXm ( Xist− ) and XmXp ( ≥ 2 fold on average ) ( Fig 4b ) . These results suggested that excess Rnf12 might facilitate Xm-Xist derepression via Xm-Tsix repression in XmXm embryos . To test this possibility , we constructed Rnf12-depleted XmXm embryos by siRNA injection ( Rnf12KD-XmXm ) ( S6a and S6b Fig ) . The RNA-FISH analysis for Xm-Xist/Tsix revealed that Rnf12 repression caused a remarkable increase of Tsix+ ( scramble-XmXm: 14% vs . Rnf12KD-XmXm: 64% , Fig 4c ) , and the proportion of Xist+ cells significantly declined ( scramble-XmXm: 59% vs . Rnf12KD-XmXm: 28% , Fig 4c ) . Next , we tested the previous notion in differentiating ES cells , which indicated that RNF12-mediated Xist upregulation was involved in the Rex1 expression state [20] , as determined by Rnf12/Rex1 double knockdown ( KD ) experiments in XmXm embryos ( S6c Fig ) . As expected , the Xm-Xist repression with Tsix upregulation seen in Rnf12KD-XmXm embryos was rescued in XmXm embryos with Rnf12/Rex1 double depletion , although there was no marked effect on Xm-Xist/Tsix expression in Rex1 single KD embryos ( Fig 4d ) . Thus , we concluded that the Rnf12 overdose in XmXm embryos caused Xm-Tsix repression and resulted in Xm-Xist activation . Given the above results from XmXm embryos , we inferred that the decrease of Rnf12 in XmY and XmXp embryos around morula stages might be essential for Xm-Tsix activation to repress Xm-Xist since the chromatin was decondensed . To test this possibility , we constructed Rnf12-overexpressing fertilized embryos . As RNF12 turnover was implied to be quick [4] , we injected Rnf12 mRNA into 2-cell blastomeres ( Fig 4d ) . At the blastocyst stage , we carried out immunofluorescence against RNF12 combined with RNA/DNA-FISH ( IF-RNA/DNA-FISH ) . Of the RNF12 overexpressing cells in XmY , 11 . 1% exhibited Xist cloud states ( Fig 4e ) , whereas embryos with Egfp mRNA never showed Xm-Xist derepression ( Fig 4e ) . In XmXp embryos , biallelic expression was significantly induced in Rnf12 overexpressing cells ( Egfp: 1 . 9% vs . Rnf12 overexpression: 11 . 2% , Fig 4f ) . These results indicated that proper Rnf12 expression levels were required for Xm-Xist silencing in fertilized embryos . Around the morula stage , many pluripotency-factors that were shown to regulate Xist begin to be expressed [21–23] . Therefore , using XmXm morulae , we investigated the involvement of pluripotency-factors in Xm-Xist/Tsix regulation . YY1 and Oct4 were selected for candidate pluripotency-factors based on previous reports [23–25] and we conducted siRNA-mediated KD experiments . qPCR and Xist RNA-FISH analysis revealed that Oct4 depletion induced significant reduction of Xm-Xist ( S7 Fig ) , implying that Oct4 is an important factor for regulating Xm-Xist imprint erasure . Further , Xist/Tsix RNA-FISH analysis in Oct4KD-XmXm embryos revealed that the proportions of cells with Xist and Tsix signals were extremely reduced ( Oct4KD-XmXm: Xist 18% and Tsix 8% , Fig 5a and scramble: Xist 59% and Tsix 14% , shown in Fig 4c ) , indicating that opposed to RNF12 , Oct4 controls not only Xm-Xist derepression but also Xm-Tsix activation . To gain further insights into the effect of Oct4 and Rnf12 depletion in XmXm morulae , we performed RNA-Seq analysis . Out of 7898 genes with > 10 trimmed mean of M values ( TMM ) [26] in at least one group , 280 and 613 genes were differentially expressed ( more than 2-fold ) in Rnf12KD-XmXm and Oct4KD-XmXm embryos , respectively ( Fig 5b and 5c and S2 Table ) , indicating the high impact of Oct4 depletion on the transcriptome . Notably , the differentially expressed genes following Oct4 or Rnf12 depletion were randomly distributed across the chromosomes ( S8 Fig ) . Consistent with the FISH results , Xist and Tsix were down- and upregulated in Rnf12KD-XmXm embryos , respectively ( Fig 5b and S2 Table ) , whereas both were repressed in Oct4KD-XmXm embryos ( Fig 5c and S2 Table ) . Oct4 and Rnf12 expression levels were comparable to those of scramble-XmXm embryos in Rnf12KD-XmXm and Oct4KD-XmXm embryos , respectively ( Fig 5b and 5c and S2 Table ) . The expression states of major histone modifiers including members of the H3K9me3 demethylase Kdm4-family were not dramatically affected ( Fig 5b and 5c and S2 Table ) . However , we found that Tet2 , which is associated with DNA methylcytosine dioxygenase [27] , was markedly downregulated ( 14% of scramble ) only in Oct4KD-XmXm morulae ( Fig 5c and S2 Table ) . To examine the impact of Tet2 depletion on Xm-Xist derepression , we constructed Tet2KD-XmXm embryos and evaluated their Xm-Xist expression states . RNA-FISH analysis indicated that the extent to which Tet2 mediated Xm-Xist repression was modest compared with Oct4 depletion ( S9 and S7c Figs ) . However , as we could not exclude the possibility that the Tet2 KD efficiency might be insufficient , given that DNA methylation was not responsible for Xm-Xist expression [28] , these results suggested that dysregulation of epigenomic factors were not likely to be the primary cause for Xm-Xist repression following Oct4 depletion . The known Xist activators on the X chromosome ( Jpx and Ftx ) [29 , 30] were not detectable in either group by qPCR analysis . Taken together , these results indicated that the mechanism by which Oct4 mediated Xist regulation was different from that underlying Rnf12-mediated regulation . Previous studies demonstrated that H3K9me3 was involved in Xm-Xist silencing and that Tsix transcripts altered H3K27me3 states at Xist promoter regions [31 , 32] . As such , we investigated the two histone modifications at Xm-Xist promoter regions in Oct4KD-XmXm and Rnf12KD-XmXm embryos using embryo-chromatin immunoprecipitation combined qPCR [4 , 13] . Notably , we found nosignificant changes of H3K9me3 modifications in Oct4KD- and Rnf12KD-XmXm embryos compared with scramble-XmXm embryos in the regions examined ( Fig 6a ) whereas , as expected , in Rnf12KD-XmXm embryos , significant hypermethylation of H3K27me3 at the promoter regions compared with scramble-XmXm embryos was observed ( 4 . 7-fold increase , Fig 6b ) . Oct4KD-XmXm embryos also showed this effect , albeit more modest , ( 2 . 4-fold increase compared to scramble-XmXm , Fig 6b ) . The repeat-A regions were also markedly hypermethylated in Rnf12 or Oct4 depleted XmXm embryos ( Rnf12KD: 5 . 5 fold and Oct4: 18 . 4 fold increase compared to scramble-XmXm , respectively , Fig 6b ) . Thus , these results indicate that Oct4 and Rnf12 were involved in the alteration of histone modifications leading to transcriptional active states around Xist regulatory regions . Since Oct4 repression caused not only Xm-Xist but also Xm-Tsix silencing in XmXm embryos , we inferred that Oct4 might also regulate chromatin condensation states at Xm-Xist/Tsix genomic regions . To test this possibility , we conducted DNA-FISH analysis in XmXm morulae . Chromatin condensation at the loci was significantly induced by Oct4 depletion but was not observed upon Rnf12 repression ( Fig 6c ) . These results indicated that Oct4 mediated chromatin relaxation facilitated transcription around Xist/Tsix regions and resulted in Xm-Xist/Tsix activation in XmXm embryos . Next , we investigated whether Oct4 served as chromatin opener in XmXp and XmY embryos . To distinguish Xp and Xm alleles , we conducted RNA/DNA-FISH at the morula stage . Notably , Oct4 depletion significantly induced chromatin contraction in both XmY and XmXp embryos at Xm-Xist/Tsix ( Fig 6d and 6e ) . Given this , we sought to investigate whether Oct4 might affect Xm-Tsix expression states , by Xist/Tsix RNA-FISH analysis in Oct4KD-XmY and -XmXp embryos . Notably , Xist expression states in XmXp embryos were comparable between scramble and Oct4KD embryos ( Fig 6f ) , indicating that Oct4 did not affect Xp-Xist expression . However , as expected , Oct4KD embryos exhibited a significant reduction of cells with Tsix+ in both XmY ( 9 . 0% ) and XmXp ( 6 . 1% ) compared to scramble-XmY ( 21 . 7% ) and -XmXp ( 18 . 2% ) cells counterparts ( Fig 6f ) . Taken together , these results revealed the novel role of Oct4 as a chromatin opener to induce the activation of Xm-Xist/Tsix in XmXm and of Xm-Tsix in XmY and XmXp embryos .
Species-specific imprinted XCI has been observed and one study indicated that human embryos showed no imprinted XCI [33] . Previously , Sado and Sakaguchi proposed that chromatin condensation states in parental genomes differed in each specie and that this might define imprinted XCI [34] . In the current study , we showed that the asymmetric chromatin condensation states of parental Xist/Tsix genomic regions are crucial for the initiation of Xist expression in mice ( Fig 7b ) . In mice , the Xm-Xist imprint is established during oogenesis [13 , 35] . During the phases , a maternal genome state is imposed on many transcriptionally repressive marks such as 5mC of DNA , H3K27me3 , and H3K9me2/3 [4 , 36–38] . Furthermore , HDAC2 , which mediates induced histone deacetylation , is not highly expressed until the full grown oocyte stage [39] . Thus , maternal chromatin becomes condensed during oocyte growth [13] . In mice , the maternal pronucleus is smaller than its paternal counterpart after fertilization , reflective of the maternal genome condensation . As a reflection of the chromatin condensation states in the maternal genome immediately after fertilization , the maternal pronuclear size is smaller than paternal size in mice [4] , whereas in humans , the parental pronuclei size was equal and non-imprinted XIST expression was observed [40] . Our findings disclosed a novel role of Oct4 in Xist regulation in vivo . Maternal Oct4 has been shown to be dispensable for embryonic development [41] . Recently , we found that the Oct4 protein was not localized to the nucleus until the 8–16-cell stage in mice [42] . Moreover , Oct4 overexpression altered chromatin conformation during early preimplantation phases[42] . More recently , Oct4 has been shown to relax chromatin in 8-cell embryos [43] . These findings supported the conclusion in the present study that Oct4 is a functional chromatin remodeler around Xist/Tsix genomic loci . However , the mechanism by which Oct4 induces chromatin decondensation remains unknown . RNA-Seq analysis showed that the expression levels of major chromatin remodelling factor genes such as Caf1 , Brg1 , Ring1b , and Ezh2 [44–46] were not dramatically altered by Oct4 depletion ( S2 Table ) . One of the other possibilities for controlling chromatin remodeling is direct binding of Oct4 around Xist/Tsix regions . In ES cells , Oct4 could bind to XqD regions including Xist/Tsix loci [47] ( S10 Fig ) . Moreover , recent study revealed that Nanog was necessary for an open heterochromatin organization in ES cells by direct binding to major satellite regions [48] . Thus , the direct bindings of Oct4 around Xist/Tsix loci might recruit transcriptional activators or evict transcriptional repressors . RNF12 is an essential factor for imprinted XCI [8] . The role of RNF12 as a dose-dependent Xist activator [7 , 49] is supported by the present study ( Fig 7b ) . At late preimplantation embryos , as shown in previous studies using differentiating ES cells [20 , 21] , Rnf12 controls Xm-Xist expression by silencing Tsix , which was induced by Rex1 in XmXm embryos ( Fig 7b ) . Thus , the primary role of Rnf12 at late preimplantation phases is the silencing of Rex1 leading to Tsix repression . However , the Rnf12 expression levels of XmY and XmXp embryos markedly declined compared with those of XmXm embryos ( Fig 4b ) . Therefore , under the physiological conditions of XmY and XmXp embryos , Rnf12 double dosage never occurs and Tsix can be expressed from the Xm allele to induce chromatin alteration at Xist promoter regions . In contrast , the role of RNF12 in Xp-Xist activation at early preimplantation phases remains a large question . Makhlouf et al . demonstrated that YY1 binds to Xist exon1 loci and can activate Xist in somatic cells [24] . These YY1 binding sites are CpG regions and DNA methylation inhibited this YY1 binding [24] . In support of the importance of YY1 binding sites for Xist activation , a DNA methylome study revealed that a part of the exon1 regions in the sperm genome were hypomethylated [50] , implying YY1 binding in Xp-Xist . Furthermore , Gontan et al . showed the interaction of RNF12 with YY1 [20] . Therefore , the examination of the role of YY1 for Xp-Xist activation will aid in determining the mechanism of RNF12-mediated Xp-Xist activation .
Female B6D2F1 and male C57BL/J mice were purchased from CLEA and Sankyo Labo service ( Japan ) and oocytes and sperm were collected according to standard methods [4] . Repeat-A deletion mice were obtained from RIKEN BRC ( B6 . Cg-Xist<tm5Sado> ) . All animals were maintained and used in accordance with the Guidelines for the Care and Use of Laboratory Animals of the Japanese Association for Laboratory Animal Science and the National Research Institute for Child Health and Development ( NRICHD ) of Japan . All animal experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee of the NRICHD ( Permit Number: 05–006 ) . The production of parthenogenetic and androgenetic embryos was previously described [4] . In brief , oocytes were incubated in Ca-free M16 medium containing 8 mM SrCl2 and 5 μg/mL cytochalasin B ( Sigma-Aldrich ) for 5–6 hours . For production of haploid parthenogenetic embryos ( hPE ) , the cytochalasin B was removed in the activation medium . All embryos were cultured in KSOM medium ( EMD Millipore ) in an atmosphere containing 5% CO2 at 37°C . In the TSA experiment , the embryos were cultured for 25 h in activation and culture media containing 50 nM TSA ( Sigma-Aldrich ) . siRNAs were purchased from Life Technologies; siRNA sequences are described in S3 Table . siRNA injection into ovulated oocytes was conducted using a Piezo drive ( Sutter Instrument Company ) . For expression or FISH experiments , the embryos were collected at 24–26 ( 2-cell ) , 48–50 ( 4-cell ) , 57–59 ( 8-cell ) , and 72–74 ( morula ) h after activation or insemination , respectively . For nuclear transfer , HVJ-E ( Ishihara Sangyo , Japan ) -mediated fusion methods were used for all nuclear transfer experiments . Prior to nuclear transfer , zona pellucida was silted by a grass knife and the 1st polar body was removed to prevent fusion with oocytes by HVJ-E . In male pronuclear transfer experiments , large pronucleus was selected and transferred into hPE . For the preparation of metaphase nuclei of hPE , hPE at the morula stage were incubated with M2 containing 1 μg/mL Nocodazole ( Sigma-Aldrich ) for 4–5 hours and used as donor cells . The reconstructed oocytes were subjected to intracytoplasmic sperm injection . For embryo transfer , pseudopregnant ICR mice ( Clea Japan ) were used as embryo recipients . At E19 . 5 , the embryos were recovered from the uterus . The preparation of in vitro synthesized Kdm4b and Egfp mRNA was described previously [4] . For Rnf12 mRNA synthesis , the full length coding sequence ( CDS ) was amplified by PCR using KOD-Plus-Neo DNA polymerase ( Toyobo , Osaka , Japan ) from 1-cell embryos . The amplified DNA was used as a template for the generation of PCR products with Poly-A tail and a T7 promoter and the products were subjected to in vitro transcription . The primers used for Rnf12 CDS amplification are shown in S3 Table . The qPCR analysis was conducted using TaqMan probes ( Life Technologies ) as described previously [4] . Total RNA from morula embryos ( 72 h after activation ) was extracted using an RNeasy micro kit ( Qiagen ) according manufacturer instructions . Gapdh ( Mm99999915_g1 ) was used as an internal control for normalization of target genes ( Oct4: Mm00658129_gH , Yy1: Mm00456392_m1 , and Rex1: Mm01194090_g1 ) The zona pellucida was removed by treatment with acid Tyrode’s solution ( Sigma ) and single cells from each preimplantation stage were collected using a micromanipulator . Total RNA isolation and cDNA synthesis were performed using a Single Cell-to-CT™ qRT-PCR Kit ( Thermo Fisher ) with slight modifications . In brief , half volumes of all reagents were used in this study . The qPCR analysis using TaqMan probes ( Rnf12: Mm00488044_m1 Xist: Mm01232884_m1 ) was conducted without a cDNA preamplification step . A total of 4 or more embryos were randomly selected from which to collect single cells used in the assay . The remaining cells at the 2-cell stage in fertilized embryos were subjected to DNA-FISH analysis as described below . Embryos were fixed and permeabilised as previously described [13] . In brief , zona pellucida embryos were fixed with 2% PFA in PBS containing 0 . 1% PVA ( Sigma ) for 15 min at room temperature and then permeabilised with 0 . 25% Triton-X in PBS-PVA for 10 min at room temperature . After blocking with 1% BSA , the samples were incubated with the primary antibody RNF12 ( 1:500 diluted by blocking buffer , Abnova , H00051132-M01 ) , H3K27me3 ( 1:200 , Millipore , 07–449 ) , or Oct4 ( 1:200 , Santa Cruz Biotechnology , C-10 ) , respectively . For H3K9me3 ( 1:500 , Abcam , ab8898 ) and H3K9Ac ( 1:500 , Abcam , ab12179 ) detection , fixation and permeabilisation treatments were simultaneously conducted and the primary antibodies were simultaneously incubated . The images were observed using a LSM510 laser scanning confocal microscope ( Carl Zeiss ) . For quantification of the signal intensity , the same laser intensity was applied to each sample and the signals were calculated using U . S . National Institutes of Health ( NIH ) ImageJ software ( http://rsb . info . nih . gov/ij/ ) . Embryo-ChIP ( eChIP ) analysis for preimplantation embryos was based on previous reports . At least 15 XmXm morulae were used per assay . The primer/probe sequences used were described previously [13] . In addition , antibodies for H3K9me3 ( Abcam , ab8898 ) and H3K27me3 ( Millipore , 07–449 ) were used . The samples for RNA-FISH were prepared as previously described [4] . In brief , for Xist detection , the pXist12 . 9 plasmid containing the majority of the Xist cDNA was used ( kindly gifted by T . Sado ) . For Tsix detection , the region ( around 7 kb ) of the Tsix locus from chr X: 103 , 448 , 873 to 103 , 455 , 853 was amplified by PCR and the products were subjected to nick translation ( Abbott Laboratories ) . The region from chr X: 103 , 459 , 241 to 103 , 460 , 958 was amplified by PCR and cloned in the PUC118 vector ( Takara ) , resulting in PCU118-Tsix1 . 7 . The plasmid was also subjected to nick translation along with the PCR products of the 7 kb region . The FISH images were observed using a LSM510 laser scanning confocal microscope using C- . Apochromat 40x/1 . 2 W ( Carl Zeiss ) . The DNA-FISH procedures were based on a previous study [13] . The fixed and permeabilised embryos were treated with RNaseA and then incubated with 0 . 2N HCl containing 0 . 05% tween-20 solution on ice for 10 min . The samples were incubated at 85°C for 10 min and then for overnight at 37°C . BAC DNA probes ( RP23-311P7 and RP23-36C20 ) were prepared by nick translation . For evaluation of chromosome pairing , the probe derived from RP23-311P7 was used . Both probes were used for the chromatin condensation assay . For embryo sexing , the probes of X-chromosome ( XqF4 regions ) and Y-chromosome were purchased from Chromosome Science Labo ( Sapporo , Japan ) . Distance measurements were based on previous reports [13] . Briefly , the signal centroid was calculated by NIH ImageJ software . Each nuclear radius used for distance normalization was calculated using the DAPI-stained area measurement . For image capture of all DNA FISH analyses , LSM510 laser scanning confocal microscopy using a Plan-Apochromat 100×/1 . 46 Oil DIC objective ( Carl Zeiss ) was used . Morula stage embryos were used for RNA/DNA-FISH analysis . The RNA-FISH procedure and image capture were carried out as in the above method and after image capture , the samples were washed with PBS and incubated with RNaseA for 1 . 5 h . After washing , the samples were treated with a solution including 0 . 01N HCl , 0 . 1% Tween20 , and 100 μg/ml Pepsin ( Sigma ) for 7 min at 37°C . After washing , the samples were hybridized with probes at 85°C for 10 min and then overnight at 37°C . The image capture and distance calculations were performed as described above . The IF-RNA/DNA-FISH procedures were based on a previous report [51] . In brief , the embryos were fixed with 2% PFA-PVA for 15 min at RT and then permeabilised with 0 . 25% Triton X-100 in PBS-PVA for 10 min . After washing with PBS-PVA , the samples were blocked in 1% BSA-PBS-PVA containing 1 . 3 U ml−1 RNaseOUT ( Life Technologies ) for 40 min . After washing , the embryos were incubated with primary antibodies ( anti-RNF12 , Abnova , diluted 1:200 in blocking buffer containing RNaseOUT ) for 1 h . After incubation with the secondary antibody , the samples were subjected to RNA/DNA-FISH as described above except that the pepsin treatment in blastocysts was for 4 min . The Hiseq system ( Illumina , Inc . ) was used for RNA-sequencing . In brief , total RNA from each sample ( 30 pooled embryos ) or single blastocysts were extracted using a Qiagen RNeasy Micro Kit ( Qiagen ) , and the remaining DNA was degraded by DNase treatment . In blastocyst samples , a fraction of the total RNA was used for qPCR analysis to screen female samples . For Kdm4b/TSA-XmXpΔ samples , we selected samples with high Xist expression . For construction of sequencing libraries , we used an Ovation Single Cell RNA-Seq System ( NuGEN ) according the manufacturer’s instruction . BAM format data yielded by Tohat 2 . 0 . 11 were subjected to successive analyses using AvadisNGS 1 . 6 ( Agilent Technologies ) . The counts of raw reads allocated for each gene/transcript , which link to UCSC transcripts , were normalized using the TMM method ( AvadisNGS 1 . 6 ) . Normalized values were described as log2 values . For clustering analysis , the R function “hclust” ( https://www . r-project . org/ ) was used to produce unsupervised clustering . The raw data was deposited in SRA ( http://www . ncbi . nlm . nih . gov/sra ) under accession I . D . : PRJNA312739 and PRJNA305455 . The published data of Oct4 ChIP-seq [52] ( GSM566277 ) in ES cells was visualized via the UCSC genome browser ( https://genome . ucsc . edu/ ) using custom tracks .
|
X-inactive specific transcript ( Xist ) is essential a large non-coding RNA for establishment of X chromosome inactivation in female mammals . The aberrant X chromosome inactivation critically affects cellular viability . Therefore , spatiotemporal regulation of Xist expression is required for proper development . In mice , Xist expression is imprinted in early embryonic development and maternal Xist is never expressed during preimplantation phases irrespective of the presence of Xist activator , maternal Rnf12 . Generally , parental origin-specific expression pattern of autosomal imprinted genes is maintained in various types of embryos . However , Xist imprinting for transcriptional silencing of maternal Xist was erased in cloned or parthenogenetic but not fertilized embryos . Here , we dissect the molecular mechanism underlying the variable nature of Xist imprinting . We show that in fertilized embryos , chromatin condensation states are essential maternal Xist repression in early preimplantation phases , whereas at late preimplantation stages , pluripotency factor Oct4 serves as a chromatin opener and the maintenance of Xist silencing depends on Rnf12 expression dosage . Although the Oct4 mediated chromatin decondensation also occurs in parthenogetic embryos , Rnf12 overdose causes maternal Xist derepression at late preimplantation phases . Thus these findings reveal that the chromatin regulation by pluripotency factor and Xist activator dose define Xist imprinting state .
|
[
"Abstract",
"Introduction",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"condensation",
"gene",
"regulation",
"rna",
"analysis",
"condensed",
"matter",
"physics",
"germ",
"cells",
"developmental",
"biology",
"oocytes",
"epigenetics",
"molecular",
"biology",
"techniques",
"embryos",
"mammalian",
"genomics",
"chromatin",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"embryology",
"chromosome",
"biology",
"animal",
"cells",
"genomic",
"imprinting",
"gene",
"expression",
"phase",
"transitions",
"molecular",
"biology",
"animal",
"genomics",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"physics",
"biochemistry",
"rna",
"cell",
"biology",
"ova",
"nucleic",
"acids",
"nucleic",
"acid",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"genomics",
"non-coding",
"rna"
] |
2016
|
Maintenance of Xist Imprinting Depends on Chromatin Condensation State and Rnf12 Dosage in Mice
|
Human African Trypanosomiasis ( HAT ) , a disease caused by protozoan parasites transmitted by tsetse flies , is an important neglected tropical disease endemic in remote regions of sub-Saharan Africa . Although the determination of the burden of HAT has been based on incidence , mortality and morbidity rates , the true burden of HAT goes beyond these metrics . This study sought to establish the socio-economic burden that households with HAT faced and the coping strategies they employed to deal with the increased burden . A mixed methods approach was used and data were obtained through: review of hospital records; structured interviews ( 152 ) ; key informant interviews ( 11 ) ; case narratives ( 12 ) and focus group discussions ( 15 ) with participants drawn from sleeping sickness patients in the south western HAT foci in Kenya . Quantitative data were analysed using descriptive statistics while qualitative data was analysed based on emerging themes . Socio-economic impacts included , disruption of daily activities , food insecurity , neglect of homestead , poor academic performance/school drop-outs and death . Delayed diagnosis of HAT caused 93% of the affected households to experience an increase in financial expenditure ( ranging from US$ 60–170 ) in seeking treatment . Out of these , 81 . 5% experienced difficulties in raising money for treatment resorting to various ways of raising it . The coping strategies employed to deal with the increased financial expenditure included: sale of agricultural produce ( 64% ) ; seeking assistance from family and friends ( 54% ) ; sale/lease of family assets ( 22% ) ; seeking credit ( 22% ) and use of personal savings ( 17% ) . Coping strategies outlined in this study impacted negatively on the affected households leading to further food insecurity and impoverishment . Calculation of the true burden of disease needs to go beyond incidence , mortality and morbidity rates to capture socio-economic variables entailed in seeking treatment and coping strategies of HAT affected households .
Human African trypanosomiasis ( HAT ) also known as sleeping sickness is one of the 17 neglected tropical diseases identified by the World Health Organization ( WHO ) and has also been marked for elimination by the year 2020 [1 , 2] . Two forms of the disease exist depending on the parasite involved; trypanosome brucei rhodesiense that is dominantly found in the Southern and Eastern Africa regions , Kenya included and Trypanosoma brucei gambiense mostly found in Central and Western Africa [3] . It affects the world’s poorest and tends to occur in areas where there are no doctors , no drugs , hunger is greatest and food security least , incomes are lowest , health information is scanty and human need is greatest [4] . In 2002 , calculations showed that HAT was the cause of loss of about 1 . 5 million DALYs , making it to be ranked number 7 in relation to other diseases [5] . It is ranked second among the top five priority zoonoses in Kenya [6] . The African region accounts for close to 90% of all the patients reported [7] . In spite of its significant presence in East and Central African region , there is a dearth of knowledge on the social and economic burden of African typanosomiasis . FAO estimates that Africa loses up to US$1 . 5 billion annually as a result of the disease [3] . African trypanosomiasis reduces labour resources , prevents growth of the livestock industry given that high yielding animals are less likely to survive the disease , affects availability of meat and milk and deprives farmers of draught power [8] . The disease generally occurs in remote rural areas where health systems are weak or non-existent and tends to affect economically active people [9] . The resulting burden on the extended family is heavy , not only because infected individuals become unproductive but also because close relatives have to spend time taking them for treatment and looking after them . Time and money spent on seeking treatment may be a serious drain on the family’s resources [10 , 11] . Left untreated , the final outcome of the disease for the patient is death , but equally devastating is its effect on households , communities and quality of life resulting from its insidious and debilitating nature [12] . Economically , the effects of the disease are costly for young and developing economies like Kenya , which are predominately dependent on agriculture [13] . Studies in Uganda and Democratic Republic of Congo have demonstrated that HAT can have an adverse impact on the functioning of households [12 , 14] . Such adverse consequences include: increased poverty; decline in agricultural activities often leading to famine or lack of basic food security; disruption of children’s education and; generally reversal of role obligations , which more often than not enhance women’s and children’s burdens [10 , 15] . The disability adjusted life years ( DALYs ) has over the years been used to quantify the impact of disease . However , DALY calculations have been criticized for failing to capture the true burden of NTDS [16–19] . Studies conducted to estimate the disability adjusted life years ( DALYs ) of NTDs have focused more on morbidity and mortality rates [5 , 20] . Given that the socio-economic effects of HAT and its coping strategies on households , have not been adequately researched in Kenya , yet it is on the strength of these impacts that policies and control programmes are formulated , this paper documents peoples' experiences to highlight rhodesiense HAT’s direct and indirect effects on affected households and how the coping strategies put in place by households further impoverishes them .
Ethical clearance for this study was obtained from the WHO Ethical committee ( M8/181/4/B . 317 ) and from the Kenya Medical Research Ethical review board ( KEMRI/RES/7/3/1 ) . Ethical considerations were upheld throughout the study including obtaining written/thumb print consent from the respondents before the interviews . The study was carried out in south western Kenya , an area that has been a foci for HAT experiencing epidemics since the late 1980’s . Most HAT cases prior to 1990 were from Lambwe valley in Nyanza province . However , during 1990’s-2002 the majority of cases came from new focus in Teso and Bungoma districts in south western Kenya [21] . The key villages in the study area that recorded high HAT numbers were: Alkudiet , Amongura , Amaase , Amoni , Amukura , Apatit , Bukhwamba , Ikapolok , Katelenyang , Kodedema , Kokoki , Obekai and Obuchun [21] . These were areas experiencing an outbreak of the disease for the first time hence the focus of this study . This south western Kenya HAT foci is part of the Busoga HAT focus that combines both the western Kenya and eastern Uganda foci , covering an area of 3889Km2 and traversing Busia , Bungoma and Teso counties in Kenya . The 1990 epidemic outbreak affected both countries in the Busoga focus [22] . The ecosystem comprises hills , valleys and rivers draining into Lake Victoria harbours and is a habitat for Glossina pallidipes and Glossina fuscipes which are important vectors for animal and human trypanosomiasis , respectively [9] . The vegetation in the uncultivated land in the area is mainly composed of savannah grassland interspersed with Lantana camara bush and Digithonia spp , which form good habitats for tsetse flies . Thick forests and swamps are found along the rivers and streams , which form suitable habitats for Glossina fuscipes , a riverine tsetse species that is the main one infesting the areas [23] . The area has a population of about 6 million people , and is served by 200 health facilities [24] . The area also hosts the Alupe Health facility , the only referral hospital for sleeping sickness in Kenya established in 1970 . Thirty eight of the households are female-headed while the mean household size is 4 . 6 . The population age-sex structure is wide based with more persons in the younger age groups than in the older groups for both sexes [24] . A majority ( 60% ) of the population fall below the official poverty line [13] A retrospective cross-sectional study that applied mixed methods approach was used . To limit recall bias due to arise in any retrospective study , triangulation of data collection methods was used . Former HAT patients from the year 1990–2002 in the 13 villages that recorded high HAT numbers were the main respondents for this study . Data were obtained through both quantitative and qualitative methods . Quantitative methods included review of hospital records from 1990–2002 to identify HAT patients and their villages , the stage of the disease at the time of diagnosis , and duration the patients spent in hospital . To identify past HAT patients , the study adopted the purposeful sampling strategy where the existing KETRI-Alupe HAT database formed the sampling frame . The databases were compiled according to districts , locations , sub-locations and villages , and dated as far back as the 1950s . However , for the purpose of this study , only patients from 1990 to 2002 who resided in the 13 villages that experienced the disease for the first time and recorded high HAT numbers were considered . During this period of study , there were 208 HAT patients in the database . Due to difficulties of tracing some of them as a result of migration and natural attrition , only 152 HAT patients or their guardians were traced with the help of the village chiefs and interviewed using a structured questionnaire . Five research assistants who were taken through a one-day training on the use of the tool helped administer the questionnaires . The quantitative data collected through this method included , socio-demographic and economic data , HAT treatment seeking behaviour and effects of HAT on the household including coping strategies . Qualitative methods included key informant interviews , case narratives and focus group discussions ( FGDs ) . Eleven key informant interviews were held with opinion leaders from key affected villages and health personnel who had attended to the HAT patients to understand the disease in the context of the study area . To be recruited as a key informant , one had to have lived in the study area during the 1990 HAT epidemic and been in a key position in leadership or a health facility . Participants for the case narratives were identified with the help of the health staff and village chiefs . Twelve case narratives were conducted with former HAT patients , who had various experiences that best illustrated the socio-economic impacts of HAT , to enable a richly detailed exploration of individual’s own perceptions and accounts of their experiences with HAT . These were identified and recruited during the administration of the structured questionnaire . In addition , 15 FGDs ( 5 male only , 5 female only and 5 mixed groups ) were held in the study area . These consisted of between 6–12 participants who were identified and recruited on the basis of coming from the same village , having suffered from HAT or were guardians/care-takers of former HAT patients and their immediate neigbours . The FGDs were held to re-visit emerging issues especially with regard to the socio-economic effects of HAT and the coping strategies employed by HAT affected households . In setting up these groups , attention was paid to homogeneity of the participants to give ample room for free discussions . However , in some areas where there were few HAT patients , mixed FGDs were conducted . The discussions focused on what these people knew about HAT and its effects on the HAT patients and their households . A checklist was used to guide these discussions . The first author played the role of a moderator/interviewer and a trained research assistant took detailed notes during the FDGs , key informant interviews and the case narratives . These were also tape-recorded with the consent of the participants . Field data were collected for 3 months , beginning January to March , 2004 . We first approached the participants in person , explained the objectives of the study to them , and fixed an appointment with them for an interview/ group discussion based on their availability and convenience . The interviews were conducted in the homes of the HAT patients or a neutral quiet venue approved by the participants . All interviews lasted between 30 to 120 minutes . To protect the informants from harm in terms of the nature of interviews , we took informed consent , assured them about anonymity , and told them that they were free to withdraw at any time or refuse to answer any specific question . The interviews were conducted in either the national language , Kiswahili , or the local languages . All the recordings of the interviews were transcribed and translated into English . The hand written notes were used to fill in any gaps in the recording to complete the transcripts . Thematic analysis was used to analyse the qualitative data . Thorough familiarity with the responses was gained by reading and re-reading all of the transcripts . This helped to code and categorise the data according to emerging themes . Quantitative data were coded and analysed descriptively using the statistical package for social science .
One hundred and fifty-two former patients who were recorded as HAT cases between1990-2002 , ( Fig 1 ) were interviewed to document the socio-economic impacts of HAT and the coping strategies of their households . Table 1 presents a summary of the socio-demographics of the respondents . At the time of interviews , 88 . 8% ( 135/152 ) of former HAT patients or their guardians had only primary school or no formal education . A majority ( 77% ) were married while 10 . 5% were widowed . Sixty three percent of the respondents were from monogamous households while 30% were from polygamous households . Up to 82% of the respondents were primarily farmers , and 76 . 9% of the respondents earned less than Kenya shillings 4000 ( 52 US dollars ) per month ( US 1 was an average of KES 77 during the period of data collection . Through a questionnaire survey , we determined the duration of illness before HAT diagnosis was made and treatment started . Only 15% of the patients reported a diagnosis of HAT following onset of clinical signs , with the majority ( 57 . 6% ) starting HAT treatment after two months ( Table 2 ) . Additionally , 73 . 7% of the HAT patients were diagnosed at a late stage . Almost an equal number of male ( 12 . 5% ) and female ( 13 . 8% ) HAT patients were diagnosed with the early stage while slightly more males ( 38 . 2% ) than females ( 35 . 5% ) were diagnosed with the late stage of HAT , having sought treatment using various options ( health facility , over-the-counter , traditional medicine , diviners and seeking prayers ) . There is no statistically significant association between sex and stage of diagnosis ( Chi square = 0 . 2165 , p = < . 05 ) . The main symptoms of the disease as mentioned by the respondents include sleep ( 69% ) , fatigue ( 60% ) , feeling cold ( 40% ) , loss of appetite ( 28% ) , body swellings ( 6% ) and miscarriage in pregnant women ( 1% ) . However , in the focus group discussions , mental disturbance , weight loss , itching , rashes , joint pains , headache , stiff neck , nausea , partial blindness , stomach-ache and paleness of skin colour were mentioned . There was consensus among FGD participants on the similarities of HAT and AIDS . They were thus in agreement with the following statement from one of them whose son suffered from HAT: More than a quarter of the respondents ( 30% ) reported that the community members thought that their illness was caused by witchcraft , 21 . 2% of them related it to the tsetse fly , and 34 . 5% cited HIV/AIDS while 14 . 3% mentioned other varied causes such as malaria and malnutrition among others . The treatment options and pathways sought by the respondents and the delays in diagnosis are already a subject of detailed discussion in an earlier paper [25] emanating from this study . In seeking varied alternative modes of treatment for HAT based on the manifested signs and symptoms and perceived causation of the disease households spent various amounts with more than half ( 69 . 8% ) utilizing up to KES 4000 ( Table 1 ) . Others utilized much more as demonstrated by what is expressed by one of the affected: However , one key informant , a nurse , in the HAT hospital clarified that most respondents incurred these medical expenses before proper diagnosis was made . She also clarified that once proper HAT diagnosis had been made , treatment was carried out free of charge , but patients and their households still incurred other indirect expenses such as transport costs to visit the patient and loss of working time and income . The health seeking behavior of the respondents and their families in trying to get treatment for their illness from various options increased financial demands on 93% ( 106 ) of the affected households . Delays in getting a correct diagnosis emanated from both the nature of care seeking by the patients and their families , and from the health system as illustrated in the following excerpts: Cross tabulation of duration taken before correct diagnosis and average amount of money used showed that those who took more than one month before diagnosis utilized over KES 3000 meaning that delays of more than one month led to increased expenses . The average amounts spent were incremental with the time one spent before correct diagnosis . However , after staying for over four months before correct diagnosis , the amount spent reduced ( Table 3 ) . Almost ( 80% ) of the HAT patients once correctly diagnosed , took an average of five weeks in hospital undergoing treatment . This combined with the process of seeking treatment led to various socio-economic impacts and various coping strategies that were utilized to deal with it as illustrated in Fig 2 .
This was a retrospective study hence there would have been some recall bias however , the study made use of triangulation of methods to try and deal with this challenge . Similarly , sleeping sickness is not a frequent disease in the study area and most households were experiencing it for the first time ever , hence they were able to recall their experiences quite vividly . This study has leaned more on the qualitative aspects given that most studies on burden of HAT are quantitative in nature . Much as the findings of the study cannot be generalized to all HAT endemic areas , given that the focus of the study was on the rhodesiense form of HAT , the findings are based on real experiences of HAT patients in seeking treatment thus highlight key issues necessary to be considered when calculating burden of the disease . The data for this study was collected in 2004 for a PhD study . Over the years , the dynamics of HAT in Kenya have changed with only four new cases being reported from the period 2004 to 2016 [40] . However , given the neglect of NTDS at International and national levels , partly due to the limited information on their real burden , and specifically , limited information on the effects of coping strategies on the socio-economic burden of rhodesiense HAT , the findings from this study will be useful in highlighting the real burden of HAT and by implication , other NTDS . This is important in the face of the WHO’s 2030 road map for accelerating work to overcome the global impact of NTDs including HAT elimination [2] . The paper also provides an advance in knowledge about HAT as one of the NTDs and provides a basis on which future research can build upon . The findings of this study show that HAT resulted in exorbitant in-direct health care costs ( ranging from US$60–170 ) for people already living on less than US$1 per day , this leads to households engaging in coping strategies which further impact negatively on the effects of the disease on households . Morbidity from HAT temporarily removes adult labour from the household hence affecting labour availability , causing shifting of household roles with some children forced to be absent from school , or to drop out of school temporarily or permanently to provide household labour . The socio-economic effects of HAT is exacerbated by coping strategies with negative consequences on households , that erode their asset bases and makes them more vulnerable to any other future shocks . It also implies that the burden of HAT goes beyond morbidity and mortality to include other indirect costs that if left out would lead to misinterpretation of the actual burden of the disease and also condemning the families to further poverty . HAT is a disease that has the potential of negating the progress made in the achieving of the SDGs and elimination of HAT and therefore concerted effort should be made to enhance the capacity of health care facilities including provision of appropriate diagnostics to ensure prompt treatment at onset of disease . The community also needs to be empowered to help build and strengthen their livelihood base and assets in order to be resilient in the face of shocks such as those posed by HAT . Continuous sensitization and awareness creation about HAT at the community , national and international level is critical in making progress towards the goal of HAT elimination by 2020 .
|
Sleeping sickness affects people often living in remote rural areas and those who mainly depend on subsistence agriculture . We carried out a study among former sleeping sickness patients in Kenya to find out the socio-economic challenges they faced in seeking treatment and the coping strategies they used to deal with them . This is important because the socio-economic effects of sleeping sickness and its coping strategies have not been adequately researched on yet it is on the strength of these impacts that policies and control programmes are formulated . If the real burden of sleeping sickness is not known , then it will continue to be neglected in terms of the attention it receives world-wide . Sleeping sickness patients and their households spent a lot of money seeking treatment besides facing challenges of disruption of daily activities , food insecurity , neglect of homesteads , poor academic performance/school drop-outs and death . Majority of them faced difficulties in raising the money required for seeking treatment hence resorted to various coping strategies . These negatively impacted on them and their households , already living on less than a dollar per day . There is need to pay attention to these effects of sleeping sickness in establishing the real burden of the disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"children",
"medicine",
"and",
"health",
"sciences",
"education",
"african",
"trypanosomiasis",
"sociology",
"tropical",
"diseases",
"geographical",
"locations",
"social",
"sciences",
"parasitic",
"diseases",
"health",
"care",
"age",
"groups",
"human",
"families",
"neglected",
"tropical",
"diseases",
"africa",
"veterinary",
"science",
"families",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonoses",
"health",
"economics",
"schools",
"protozoan",
"infections",
"trypanosomiasis",
"economics",
"people",
"and",
"places",
"finance",
"kenya",
"biology",
"and",
"life",
"sciences",
"population",
"groupings"
] |
2017
|
The socio-economic burden of human African trypanosomiasis and the coping strategies of households in the South Western Kenya foci
|
Malaria drug resistance contributes to up to a million annual deaths . Judicious deployment of new antimalarials and vaccines could benefit from an understanding of early molecular events that promote the evolution of parasites . Continuous in vitro challenge of Plasmodium falciparum parasites with a novel dihydroorotate dehydrogenase ( DHODH ) inhibitor reproducibly selected for resistant parasites . Genome-wide analysis of independently-derived resistant clones revealed a two-step strategy to evolutionary success . Some haploid blood-stage parasites first survive antimalarial pressure through fortuitous DNA duplications that always included the DHODH gene . Independently-selected parasites had different sized amplification units but they were always flanked by distant A/T tracks . Higher level amplification and resistance was attained using a second , more efficient and more accurate , mechanism for head-to-tail expansion of the founder unit . This second homology-based process could faithfully tune DNA copy numbers in either direction , always retaining the unique DNA amplification sequence from the original A/T-mediated duplication for that parasite line . Pseudo-polyploidy at relevant genomic loci sets the stage for gaining additional mutations at the locus of interest . Overall , we reveal a population-based genomic strategy for mutagenesis that operates in human stages of P . falciparum to efficiently yield resistance-causing genetic changes at the correct locus in a successful parasite . Importantly , these founding events arise with precision; no other new amplifications are seen in the resistant haploid blood stage parasite . This minimizes the need for meiotic genetic cleansing that can only occur in sexual stage development of the parasite in mosquitoes .
The emergence of chloroquine and Fansidar resistance contributed to resurgence of malaria in the 1970s and 1980s [1] , [2] . Today , from an estimated 2 billion global clinical cases , ∼0 . 5 to 1 million individuals die of malaria every year [3] , [4] , [5] . There is a growing concern that decreased effectiveness of artemisinin combination therapies in Southeast Asia will once again lead to even higher morbidity and mortality [6] , [7] , [8] , [9] , [10] . While point mutations and DNA copy number variations have been associated with resistance to previously effective antimalarials [11] , [12] , [13] , [14] , [15] , a detailed understanding of how haploid blood stages of malaria parasites acquire resistance to truly new antimalarials is critical for the effective management of this global disease . Similar to what has been observed in clinical settings , Plasmodium falciparum malaria parasites are able to acquire resistance under controlled laboratory conditions [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] . Although parasites exposed to potent antimalarials do not show protective , real-time transcriptional responses [25] , the targets of novel antimalarials are often definitively revealed in in vitro selected resistant parasites through novel mutations or copy number variations in the parasite genome [20] , [21] , [22] , [24] , [26] , [27] , [28] . Such selections are now routinely used to identify target pathways of new antimalarials , but early molecular steps leading to beneficial mutations remain unknown . Here , we use in vitro selections to understand how haploid malaria parasite populations , under continual antimalarial pressure , correctly acquire protective changes in their genome . These controlled laboratory selections with asexual blood-stage P . falciparum allow step-wise mechanistic dissection of independently evolving parasite cell lines in ways that are not possible in field isolates or other model organisms .
Resistance was achieved by challenging P . falciparum parasites with DSM1 , a new potent and selective inhibitor of dihydroorotate dehydrogenase ( DHODH ) [29] ( see structure in inset of Fig . 1 ) . In the initial DSM1 challenge , populations of 107 parasites developed resistance to 0 . 3 µM DSM1 ( Fig . 1 , Table S1 ) . Four independently-derived clones , exhibiting ∼5-fold resistance , were selected for further investigation ( round 1 clones were designated C , D , E , and F; Table S2 ) . Pair-wise comparative genomic hybridizations of DNA from parent versus DSM1-resistant clones revealed a single ∼2- to 3-fold amplification event on chromosome 6 in all four round 1 clones ( Fig . 2A , Fig . S1 ) . The amplicon units ranged in size from 34 to 95 kb , covering 9 to 23 genes ( Fig . 2B , C ) . As discussed below , the variation in the size of the amplicon unit between independently-selected clones provided a molecular fingerprint of each evolving parasite line . All amplicons in each round 1 clone included the DHODH gene ( Fig . 2C; gene 19 , PlasmoDB gene ID PFF0160c [30] , Fig . 2D ) . DHODH mRNA and protein levels were correspondingly increased ( Fig . S2 ) , and mutations were not detected in the gene itself ( Fig . S3 ) . Whole genome sequencing of the parent Dd2 clone and clone C ( see genome coverage rates in Table S3 ) confirmed the de novo Whole acquisition of the DHODH amplicon and the absence of causal point mutations hidden within individual amplicon units ( Fig . 3A; Table S4 , Fig . S4 ) . In addition , resistance-associated point mutations were not detected anywhere else in the genome ( Table S5 , Table 1 ) . To learn how DSM1-resistant parasite populations efficiently arrived at these unique beneficial amplicons , we mapped the junction regions of each independently-derived DSM1 resistant clone . Based on the boundaries initially identified by mid-density microarray analysis ( Table S6 ) , we sequenced the DNA between adjoining amplicon units assuming a head-to-tail orientation , and identified long homopolymeric stretches of adenine or thymine ( A/T tracks ) between the 3′ end of one unit and the 5′ end of the second ( Fig . S5 ) . These A/T tracks fall mostly in intergenic regions at the edges of the P . falciparum amplicons , with clones C and F sharing exactly the same unit end point ( Fig . 2C ) . The 3′ junction of the remaining two clones D and E exist in two separate introns of PFF0185c ( gene 24 , Fig . 2C and Table S7 ) . Of the 8 independent events studied here ( 2 junctions for each of the 4 independently derived clones analyzed ) , all displayed A/T tracks at the junctions ( 0 events occurred at a non-A/T tracks ) . Since homopolymeric tracks of >10 bp make up 5% of the genome [31] , the probability that all of the 8 independent events would randomly end with an A/T track is 1 in 25 billion . Investigations into the orientation of amplicon units ( i . e . head-to-head or tail-to-tail orientation ) as well as whether they were situated outside of chromosome 6 ( the original DHODH locus ) were expected to provide mechanistic insight into what pathways may be acting at these locations . In a quantitative approach that was not achievable in earlier studies ( either by our group ( Fig . S5 ) or others [32] , [33] ) , we acquired paired-end reads from whole genome sequencing that aligned to the junction regions of clone C and D . Histograms of read coverage displayed junctions that were consistent with both microarray and targeted sequencing results discussed above ( Fig . 3A ) . Computationally-isolated reads from the above analysis failed to reveal recombination of the DHODH loci with A/T stretches elsewhere in the genome since reads from all possible junctions aligned to only two genomic locations: ( 1 ) the region that represents the reference genome match on chromosome 6 ( Fig . 3B , red and yellow arrows ) or ( 2 ) the opposite end of the amplicon unit ( Fig . 3B and C , blue or green arrows; Fig . S6 ) . These data formally prove that the tandem head-to-tail arrangement is the predominant outcome of the initial duplication in DSM1 resistant clones ( Fig . 3D ) . Based on outcomes from round 1 clones , we hypothesize that the initial resistance-conferring duplication around the DHODH locus arises from an imprecise , even chaotic , process involving mitotic rearrangement between random A/T tracks that are sprinkled at a high frequency across the genome . Importantly , there appears to be a second non-A/T based step for expanding P . falciparum amplicon numbers . When a DSM1 resistant parasite carried more than two units in a freshly-generated amplicon , each unit had the same length , genetic content , and junction regions . Conservation of these units in each independently-selected parasite clone suggested that , after an initial fortuitous duplication between random A/T tracks surrounding the DHODH locus , subsequent expansion of the founder amplicon involves precise homologous recombination that overrides chaotic , possibly unproductive A/T track-based mechanisms . This hypothesis was further tested by exposing round 1 clones to higher DSM1 concentrations ( 3 µM or 10 µM DSM1 in round 2 compared to 0 . 3 µM in round 1; Table S8 ) . The resulting independent round 2 clones derived from clones C and D were ∼15- to ∼150-fold more resistant to DSM1 compared to the parent Dd2 ( Fig . 4A , Table S9 ) . Comparative genomic hybridizations also showed an increase of the founder DHODH amplicon in these round 2 clones ( Fig . 4B and C , Table S6 ) . Whole genome sequencing studies of the amplicon unit junctions of round 2 clones ( see genome coverage rates in Table S3 ) again displayed solely the tandem head-to-tail orientation ( Figs . 3B and C and S6 ) . The precise maintenance of the respective founder amplicons in clones C and D is particularly remarkable given that resistance can be conferred by much smaller units as was observed in round 1 clones E and F ( Fig . 2C ) . To test whether the machinery that allows for faithful expansion of the DHODH amplicons would work with the same precision during deamplification , DSM1 resistant parasites were grown without antimalarial pressure over a long period of time . Overall , resistance of both round 1 and 2 clones initially decreased before stabilizing at ∼2-to 3-fold ( Fig . 4D–E , Tables S9 and S10 ) . This observation suggested that there was a measureable fitness cost of maintaining higher levels of the DHODH amplicon , an idea that is consistent with other observations such as the normal growth rate of round 1 clones , the growth defect displayed by round 2 clones ( Fig . S7A ) , and in some cases the increased growth rate following the removal of DSM1 pressure ( Fig . S7B ) . Similar to what has previously been observed with amplified loci on P . falciparum chromosomes 4 , 5 , and 11 [20] , [34] , [35] , the step-wise decrease of DHODH copy numbers in the absence of DSM1 could be captured over time . Although the starting level of resistance differed , a gradual “dialing down” of the amplicon in the population to stable round 1 levels was observed for two independent C-derived round 2 clones ( Fig . 4D–G ) . Furthermore , comparative genomic hybridization of clones isolated from these cultures grown in the absence of DSM1 for 3 months ( DSM1 removal ( DR ) clones ) showed that despite de-amplification , amplicon unit boundaries of C-derived clones were faithfully maintained ( Fig . 4H and I , Table S6 ) . Intriguingly , this implied that the pathway that relies on large stretches of homology to “dial up” the amplicon also controls the reverse action and does not allow the A/T track-based mechanism to disrupt amplicon units that were initially evolutionarily successful .
The present findings underscore the extraordinary capability of the parasite to evolve during a human infection as a haploid asexual population . In nature , during a single human infection , a few hundred parasites entering the liver expand successfully to become many billions in the face of both drug and immune pressure . Once established in the blood , the parasites can increase and decrease in waves even without a reinfection . In order to evolve during these expansions , haploid parasites must do so with minimal damage to their genome . Similarly to what was first proposed for bacteria [36] , the initial random sampling of duplications in the malaria genome under selective pressure serves as an effective first step to locate and identify genetic targets for resistance and generates enough of a foothold for the haploid parasites to proliferate under lethal pressure . The randomness of the initial duplication step in this organism is evident in our detailed molecular characterization of independently-selected resistant parasites from round 1 selections . In addition , these early events also capture the large size of amplicons that are initially sampled ( Fig . 2 ) . Assuming one duplicated region of approximately 50–100 kb per parasite , in principle , it is possible to cover the entire 23 Mb P . falciparum genome with a few hundred parasites . However , this is clearly not the whole story: the large parasite populations of roughly a million cells required for a successful DSM1 resistance event ( Table S1 ) points to possibly extensive number of “trial duplications” that are non-productive or even lethal to the parasite . The success rates of about 1∶10 , 000 , 000 from round 1 selections against this completely novel evolutionary challenge ( DSM1 ) are similar to a previous semi-quantitative estimation of the initial amplification rate in this organism that were inferred from challenges with a traditional aminoquinoline class of antimalarials in clinical use [37] . The few parasites that can identify a productive locus by chance in the first step then rely on a second more efficient step to achieve evolutionarily more robust levels of resistance ( Fig . 5 , Step 2 ) . Based on survival numbers from round 2 selections ( Table S8 ) , this second step appears at least 100-fold more efficient once pseudo-polyploids have been established around a high priority locus . This second process also allows continual fine-tuning of amplicon unit numbers based on the level of antimalarial pressure ( Fig . 4 ) . For a haploid blood-stage parasite , when necessary , pseudo-polyploidy could even allow for the safe introduction of point mutations within the amplified region before amplicon units decrease to single copies ( Fig . 5 , Long Term ) . Indeed , during laboratory selections , amplifications of the target gene often are observed alongside point mutations in the same gene ( [22] , [28] , [35] and our unpublished observations ) . Both during in vitro selection and in natural human infection , these productive genomic alterations must take place independent of meiosis . Meiosis is the stage of the life cycle where “textbook” chromosomal crossover mixes different genomes from coinfections to bring together beneficial new traits and to remove damaged DNA in the progeny . However , the sexual stages at which meiosis occurs are not available to the parasite until the transmission of gamete stages to the mosquito . Prior to this stage of the life cycle , how does the evolving haploid parasite avoid large collateral damage as it is under pressure to change in the human ? Our detailed characterization of clones from carefully-controlled independent experiments reveals a powerful evolutionary strategy to make precise changes in its genome while expanding in the human , away from the mosquito . At its core , the strategy involves the creation of a single significant new genetic amplification in an individual parasite , even as the entire genome is being sampled by a large starting population . Through controlled laboratory experiments , we directly observed that the amplicon responsible for resistance was the only new amplicon in every individual successful DSM1 resistant parasite . By avoiding adventitious new amplicons elsewhere in the genome , collateral damage is minimized during a time when meiotic cleansings are not available to the parasite . This precise genetic modification is not without cost: every event is accompanied by millions of parasites that do not amplify a useful portion of the genome and do not survive . Whether the initial rearrangements are occurring continuously during the life of the parasites or only in response to stress is a question that remains to be answered . The first step in the generation of the DHODH amplicon was clearly mediated by stretches of polyA sequences or polyT sequences ( Figs . 5 ( Step 1 ) and S5 ) . Previously , similar homopolymeric A/T tracks have also been identified at the borders of naturally-occurring P . falciparum amplicons on chromosome 5 [32] , [33] , [38] , solidifying the relevance of the current laboratory based observations to naturally occurring genomic amplifications in this organism . The A/T-based strategy revealed by these data is uniquely matched with the high AT content of the P . falciparum genome , which averages 81% AT but can reach upwards of 90% in introns and intergenic regions [39] . Exactly such approaches are probably not utilized by other Plasmodium species that cause human malaria or by other protozoan parasites . Of note , the genomes of the haploid blood stage of P . vivax , the second most prevalent human malaria species worldwide [40] , averages ∼60% A/T content [41] and Leishmania species that are prone to drug resistance and gene amplifications average ∼40% A/T content [42] , [43] . This A/T-dependent approach likely applies to many successful evolutionary selections of different P . falciparum parasites; genomic amplifications have been observed during the characterization of both lab-adapted and field-derived parasites from various regions of the world [15] , [33] , [34] , [35] , [38] . In the previous studies , however , the exact mechanistic origin of the genomic rearrangements was often ambiguous . First , amplicons were generated in response to antimalarials in clinical use , and independent founder events could not be distinguished from later rearrangements . Second , in some cases , parasites were isolated from clinical infections and thus information on both the clinical drug pressures and the life history of the parasite leading to observed mutational patterns ( including passage through a mosquito and recombination with other genotypes ) were lost . In the present study , since the DHODH amplicons were selected entirely in the asexual blood stage of P . falciparum , we can definitively conclude that the A/T track-mediated step is important in the initial acquisition of a new amplification and not in changing or rearranging amplicons later in evolution . In addition to showing a general strategy of how a population of parasites narrows in on a resistance-conferring DNA locus , data from the present study points to the importance of two distinct biochemical processes that must operate in each parasite for overall evolutionary success . During replication , A/T tracks are known to cause polymerase pausing due to the rigid bend of the DNA structure [44] , [45] . Events that follow could include the creation of a double strand break and recognition by a DNA repair pathway . Alternatively , the rigidness of A/T tracks may prevent adequate histone interactions , leaving DNA open to proteins that may trigger recombination pathways [46] . Recombination pathways generally require large regions of homology to mediate strand invasion but shorter stretches of repetitive bases have also been implicated in the initiation of various mitotic DNA rearrangements [47] , [48] , [49] . Recent studies of E . coli under stress also implicate very short G-rich sequences in template switching between stalled replication forks that leads to the duplication of large genomic regions [50] . In addition to a microhomology-mediated recombination pathway that repairs DNA breaks , a similar replication-based mechanism has been implicated in the generation of complex genomic rearrangements in yeast and humans [49] , [51] , [52] . Both of these processes appear to get by with extremely small stretches of homology ( <10 bp ) , which are significantly shorter than the A/T tracks observed at the borders of P . falciparum amplicons ( ∼30 bp , Fig . S5 ) [32] , [33] , [38] . A/T tracks as large as 60 bp are estimated to make up ∼5% of the parasite genome [31] , [39] , [53] and although these sequences may take part in template switching or microhomology-mediated recombination as in other organisms , their significant length could also be enough to trigger more canonical recombination pathways requiring as little as 50 bp of homology [54] . Our selection studies with DSM1 show a clear preference for pathways that generate DHODH amplifications even though point mutations have been shown to prevent DSM1 binding to a recombinant catalytically active version of DHODH [55] . Given that round 2 clones display a broad range of DSM1 resistance ( Table S9 ) , we had to be sure that hidden point mutations ( either in the DHODH gene itself or at other locations in the genome ) were not contributing to survival in the presence of high levels of DSM1 . Based on the very deep coverage of our whole genome sequencing studies ( Table S3 , Fig . S4 ) , we are confident that even low frequency mutations within large amplicons would be detected . A few additional observations suggest that resistance is truly due to DHODH amplification and there are no other undetected causal mutations in the DSM1 resistant genome: 1 ) parasites maintain sensitivity to a number of additional antimalarials ( Table S11 ) indicating that they are not employing a pleotropic resistance mechanism such as drug efflux , and 2 ) EC50 against DSM1 and DHODH copy number decrease in a parallel fashion ( Fig . 4 ) , which emphasizes the contribution of the chromosome 6 amplicons to the resistance phenotype ( as opposed to changes in other regions of the genome ) . Despite our confidence in the sequencing data , we cannot rule out that variations in amplicon sizes , and related physiological effects , contribute to the relationship between amplicon copy number and drug resistance . Why are genome amplifications favored over the acquisition of point mutations in the DSM1 model ? The ease with which one can find the correct locus that confers drug resistance and the lack of severe penalties for expanding copy numbers in the neighborhood of the DHODH gene may allow the gene amplification path to dominate . Additionally , the pharmacodynamics of drug exposure during selections could also play a role in favoring amplifications over mutations . Continual , unrelenting drug pressure demands an immediate sustained solution from the parasite population with little tolerance for wrong guesses . Although there is a measurable fitness cost of maintaining many amplicons in the absence of drug pressure ( Fig . S7 ) , parasites thrive following the increase in copy numbers of dozens of genes by an order of magnitude . Intuitively , intermittent cycling of increasing antimalarial levels , as is applied in many in vitro selection systems , may provide parasites with a chance to acquire mutations that confer high level resistance , and possibly even lose a relevant amplicon that had served its purpose in the early stages of resistance evolution . Beyond this , the nature of the drug , the nature of the target , and its location in the genome could all contribute to the optimum pathway to resistance since continual , uninterrupted application of some antimalarials during laboratory selections ( as utilized in our selection scheme ) has successfully generated point mutations in various target genes ( [19] , [23] and additional unpublished work ) . The present laboratory-controlled studies show that , in the absence of drug pressure , malaria parasites lose extra copies of amplicons . However , as often seen with field-derived amplicons ( Table S13 and [56] ) , malaria parasites do not always revert back to single copies of the target gene but instead retain a low number of amplicons in the absence of drug pressure ( Fig . 4 ) . This act has important implications for future survival: when a parasite population encounters drug pressure that it has successfully overcome before , the population is poised to rapidly re-amplify relevant amplicons quickly and efficiently without heavy collateral damage associated with A/T-based reshuffling between genes near the target . While the evolution of malaria parasites is studied here in the context of drug resistance as a selection force , the versatile parasite-specific mechanisms that are used to achieve evolutionary success must help the parasites deal with a diverse set of challenges . In the forward direction , acquisition of appropriate beneficial amplifications could help parasites survive antimalarial drugs but also other potential challenges such as host immunity [57] . It may not be a coincidence that the liver stage expansion of an incoming parasite first allows a few hundred parasites to expand to about 100 , 000 to a million parasites before the population faces unexpected immune-reactions or unusual erythrocyte genotypes of the human patient . In addition to a gain of genetic material through asymmetric recombination , the reverse direction could also have public health relevance . For example , deletions of specific genes in changing parasite populations could render rapid diagnosis tests ineffective [58] thereby misguiding diagnosis-based chemotherapy campaigns . The initial two-step evolutionary strategy of P . falciparum identified here , likely driven by two different molecular pathways with different biochemical preferences , assists the parasite in finding productive solutions to new and unexpected evolutionary challenges . The strategy is well suited for a parasite population to evolve with minimum collateral damage in surviving cells , it can act to anticipate and mount a rapid response to repeat threats , and it may offer universal advantages to parasite populations that need to withstand multiple threats beyond drug pressure .
For each experiment , erythrocytic stages of P . falciparum ( previously cloned HB3 or Dd2 ) were freshly thawed from frozen stocks and maintained as previously described [59] , [60] , [61] . Briefly , parasites were grown in vitro at 37°C in solutions of 2 to 2 . 5% hematocrit ( serotype A positive human erythrocytes ) in RPMI 1640 ( Invitrogen ) medium containing 28 mM NaHCO3 and 25 mM HEPES , and supplemented with 20% human type A positive plasma in sterile , sealed flasks , flushed with 5% O2 , 5% CO2 , and 90% N2 . Cultures were maintained with media changes 3 times each week and sub-cultured as necessary to maintain parasitemia below 5% . The highest concentration of DSM1 to which clonal Dd2 and HB3 parasites could develop resistance was determined empirically as previously described [17] . To ensure genetically pure populations , aliquots of 10 infected erythrocytes of each clonal parasite line were allowed to proliferate to about 108 infected erythrocytes . From these populations , 102–107 infected erythrocytes were challenged in flasks with 0 . 1–10 µM DSM1 ( results from 107 are displayed in Table S1 ) . Additionally , 10 infected erythrocytes were challenged with these same concentrations , to ensure that DSM1 was effective and lethal . To confirm that the parasites could proliferate normally under these experimental conditions , one flask of 10 infected erythrocytes did not receive DSM1 . This experiment was performed in triplicate , using three independent biological samples of both Dd2 and Hb3 clones . Media was changed 3 times each week ( receiving fresh DSM1 each time ) and cultures were split 1∶2 once a week to guarantee a continuous supply for fresh erythrocytes during the experiment . Parasite proliferation was monitored by Giemsa-stained thin smear blood samples taken at each media change . Selection flasks were cultured until parasites were observed proliferating or until 90 days , whichever occurred first . Using limiting dilution , 102 to 107 ( Dd2 ) or 107 ( HB3 ) populations of genetically pure parasites ( see above ) were plated across 24 wells of a 96-well plate ( each clone was selected in quadruplicate on a single plate ) . Additionally , a control plate , containing 10 infected erythrocytes per 24 wells , was set up for each clone . To ensure that DSM1 was effective and lethal , the upper half of the control plate was treated with 0 . 3 µM DSM1 . To show that the parasites could proliferate normally under the test conditions , the lower half of the control plate received no DSM1 . Plates were cultured ( as described above ) until parasites were observed proliferating or up to 120 days , whichever occurred first . As soon as parasites were observed ( Round 1 results are displayed in Table S2 ) , the well contents were transferred to a new 10 ml culture flask for expansion , sample storage and sub-cloning . During this expansion , DSM1 resistant parasites were kept under continuous 0 . 3 µM DSM1 pressure and freeze-thawing and culturing for >1month at a time was avoided as much as possible . Four DSM1 resistant clones isolated in round 1 were submitted to another round of selections ( round 2 ) . Parasite populations of 10 and 107 were selected with 1 , 3 . 3 and 10 µM DSM1 as described for round 1 . In addition , 105 parasites were also challenged with 3 . 3 µM ( Round 2 results are displayed in Table S8 ) . Resistant parasites were isolated as described above before sub-cloning for further analysis . To isolate genetically pure populations of DSM1 resistant parasites for further analysis , aliquots of 10–20 infected erythrocytes were plated across an entire 96-well plate . These plates were maintained ( as described above ) and as soon as parasites were observed proliferating , the well contents were extracted from the plate and transferred to a new 10 ml culture flask for further expansion , sample storage and analysis . A parasite solution at 0 . 5–1% parasitemia ( 0 . 5% hematocrit ) from the clone of interest was plated into a 96-well culture plate . An appropriate range of concentrations of DSM1 ( from 0 . 02–200 µM ) , depending on the level of resistance of the parasites being tested , were then added to the parasites ( because of solubility issues , 100× DSM1 concentrations ( in 100% DMSO ) were first diluted 1∶10 into RPMI ( final 10% DMSO ) before being diluted again into the parasite-containing wells ( final 1% DMSO ) ) . Each concentration of interest was performed in triplicate and included solvent-only controls . After incubating for ∼48 hours , wells were pulsed with 0 . 35 µCi each of 3H-hypoxanthine . Following an additional 24–40 hours , well contents were extracted and radioactivity was measured . Parasite proliferation in each test well was expressed as a percentage of the solvent control well . EC50 values were fit using the GraphPad PRISM software , according to the equation: Y = Bottom+ ( Top-Bottom ) / ( 1+10 ( ( LogEC50−X ) * HillSlope ) ) . For microarrays and quantitative PCR ( qPCR ) protocols , clonal asynchronous P . falciparum-infected erythrocytes were lysed with 0 . 15% saponin ( Akros ) for 5 min and genomic DNA ( gDNA ) was extracted using the DNeasy kit ( Qiagen ) according to the manufacturer's instructions . For whole genome sequencing , clonal P . falciparum cultures ( 30 mls in T75 flasks , 3% hematocrit ) were synchronized with 5% sorbitol for two consecutive cycles ( ∼45 hrs apart ) and then once more ( 3–4 hr later ) before harvesting for gDNA purification . These highly synchronous cultures ( ∼3% parasitemia at >90% rings ) were washed with PBS and frozen at −80°C prior to red blood cell lysis with saponin as above . Isolated parasites were washed 3× with PBS before resuspension in 150 mM NaCl , 10 mM EDTA , and 50 mM Tris-HCl pH 7 . 5 . Parasites were lysed with 0 . 1% L-loril sarkosil ( Teknova ) in the presence of 200 µg/ml proteinase K ( Fermentas ) overnight at 37°C . Nucleic acids were then extracted with phenol/chloroform/isoamyl alcohol ( 25∶24∶1 ) pH 7 . 8–8 . 1 ( Acros ) using phase lock tubes ( 5 Prime ) . Following RNA digestion ( with 100 µg/ml RNAse A ( Fermentas ) for 1 hr at 37°C ) , gDNA was extracted twice more as above , once with chloroform , and then ethanol precipitated by standard methods . Spotted DNA microarrays ( used for both CGH and expression analysis ) consisted of 10 , 416 −70mer oligonucleotides designed from the P . falciparum 3D7 sequence with increased coverage for long ORFs [62] . Additional custom oligonucleotides were included in the microarray to increase coverage of genes involved in folate and nucleic acid metabolism . DNA was spotted on poly-lysine coated slides and post-processed using methods described previously [25] , [63] . For hybridizations on spotted DNA microarrays , 5 µg of gDNA from each clone was sheared , labeled with 5- ( 3-aminoallyl ) -2′-deoxyuridine-5′-triphosphate , and coupled to Cy-dyes as was done previously [64] . Uncoupled Cy-dyes were removed using the DNA Clean and Concentrate-5 kit ( Zymo Research ) before hybridization to the microarray at 62°C for 16–18 h . After washing , slides were dried , scanned at 10 µM resolution using the GenePix 4000B scanner and fluorescent images were quantified with GenePix Pro 3 . 0 ( Axon Instruments ) . Further analysis , including normalization and statistical methods were performed as described previously [25] . Spotted microarray data are presented in MIAME-compliant format on the NCBI-based Gene Expression Omnibus ( GEO ) database ( Accession # GSE35732 ) . Commercially manufactured mid-density CGH microarrays containing 385 , 585 oligonucleotide probes ranging in size from 15- to 45-mer were purchased from NimbleGen Systems , Inc . These microarrays are sufficient to detect copy number variations but not single nucleotide polymorphisms ( SNPs ) [65] , [66] . For hybridizations to mid-density microarrays , gDNA was labeled with Cy3 and Cy5-labeled random nanomers ( Trilink Biotechnologies ) and hybridized to the current CGH design Plasmodium_3D7_WG_CGH as described previously [65] except hybridization was performed overnight ( ∼16–18 h ) in a 42°C water bath and microarrays were dried and scanned as above ( at 5 µM resolution ) . Normalization and analysis was performed using NimbleScan version 2 . 6 ( SegMNT CGH ) and plotted using GraphPad PRISM . Mid-density microarray data are presented in MIAME-compliant format on the GEO database ( Accession # GSE37306 ) . For DHODH qPCR , two separate sets of primers were used to amplify a 206 bp amplicon beginning at nucleotide +656 of the DHODH coding sequence ( DHODH front ) , and the second set amplified a 158 bp amplicon beginning at nucleotide +1423 of the DHODH coding sequence ( DHODH rear ) ( see Table S11 for all primer sequences ) . The qPCR protocol was 95°C for 10 min , followed by 39 rounds of 95°C for 15 sec and 60°C for 1 min . For all experiments , we performed melt curves ( 55°C to 85°C in 0 . 5°C steps with 1 s hold at each step ) to ensure a single amplicon was produced , and standard curves ( 10× dilution ladders of Dd2 gDNA ) to determine the amplification efficiency . Relative copy number was determined for 1 ng of gDNA , using the Pfaffl method [67] according to the equation ( Etarget ) ΔCt , target ( control−test ) / ( Eref ) ΔCt , reference ( control−test ) , where Seryl t-RNA Synthetase ( PF07_0073 ) and 18 s Ribosomal RNA ( MAL13P1 . 435 ) served as reference genes . DSM1 resistant clones served as the test , and the Dd2 parent served as the control . Significance was determined from multiple experiments with one-way ANOVA analysis and values from individual clones were compared using the Tukey's Multiple Comparison Test in GraphPad PRISM . Plasmodium falciparum Dihydroorotate dehydrogenase ( DHODH ) , Genbank Accession Number: AB070244 .
|
Malaria parasites kill up to a million people around the world every year . Emergence of resistance to drugs remains a key obstacle against elimination of malaria . In the laboratory , parasites can efficiently acquire resistance to experimental antimalarials by changing DNA at the target locus . This happens efficiently even for an antimalarial that the parasite has never encountered in a clinical setting . In this study , we formally demonstrate how parasites achieve this feat: first , individual parasites in a population of millions randomly amplify large regions of DNA between short sequence repeats of adenines ( A ) or thymines ( T ) that are peppered throughout the malaria parasite genome . The rare lucky parasite that amplifies DNA coding for the target of the antimalarial , along with dozens of its neighboring genes , gains an evolutionary advantage and survives . In a second step , to withstand increasing drug pressure and to achieve higher levels of resistance , each parasite line makes additional copies of this region . This second expansion does not rely on the random A/T-based DNA rearrangements but , instead , a more precise amplification mechanism that retains the unique signature of co-amplified genes created earlier in each parasite . Generation of multiple copies of the target genes in the parasite genome may be the beginning of other beneficial changes for the parasite , including the future acquisition of mutations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"genome",
"evolution",
"parasite",
"evolution",
"parasitology",
"biology",
"genomics",
"microbiology",
"drug",
"discovery"
] |
2013
|
Asexual Populations of the Human Malaria Parasite, Plasmodium falciparum, Use a Two-Step Genomic Strategy to Acquire Accurate, Beneficial DNA Amplifications
|
The speed of stem cell differentiation has to be properly coupled with self-renewal , both under basal conditions for tissue maintenance and during regeneration for tissue repair . Using the Drosophila midgut model , we analyze at the cellular and molecular levels the differentiation program required for robust regeneration . We observe that the intestinal stem cell ( ISC ) and its differentiating daughter , the enteroblast ( EB ) , form extended cell-cell contacts in regenerating intestines . The contact between progenitors is stabilized by cell adhesion molecules , and can be dynamically remodeled to elicit optimal juxtacrine Notch signaling to determine the speed of progenitor differentiation . Notably , increasing the adhesion property of progenitors by expressing Connectin is sufficient to induce rapid progenitor differentiation . We further demonstrate that JAK/STAT signaling , Sox21a and GATAe form a functional relay to orchestrate EB differentiation . Thus , our study provides new insights into the complex and sequential events that are required for rapid differentiation following stem cell division during tissue replenishment .
In metazoans , the digestive tract supports organismal growth and maintenance . Genetic disorders or microbial dysbiosis that prevent the digestion and absorption of nutrients are major causes of morbidity and mortality in humans . In mammals , mature intestinal cells are short-lived and constantly replaced by newborn differentiated cells . This is ensured by the existence of fast-cycling intestinal stem cells ( ISCs ) [1] . Although ISC division is important , failure in or improper differentiation into mature intestinal cells can equally cause a wide range of disorders that compromise organ function , such as intestinal cancer [2] and microvillus inclusion disease [3] . There is a great extent of similarity in intestinal functions and maintenance between flies and mammals [4] . Over the past decade , research has revealed the extreme plasticity of the Drosophila ISCs . For instance , stem cell activity and epithelial renewal can be adjusted in response to i ) changes in nutrient availability [5–7] , ii ) physiological requirements for reproduction [8–10] , iii ) aging [11–13] , iv ) intestinal damage or infection [14–16] , and v ) body injury [17 , 18] . Thus , both local and remote signals coordinate ISC activity to ensure intestinal homeostasis . In the adult Drosophila midgut , ISCs differentiate into either polyploid absorptive enterocytes ( ECs ) or diploid secretory enteroendocrine cells ( EEs ) ( Fig 1A ) . Recent studies indicated that EC and EE are generated through distinct mechanisms [19–21] . A post-mitotic and intermediate differentiating cell called enteroblast ( EB ) is differentiated into EC in a Notch-dependent manner [22 , 23] , while the production of EE through a so-far not molecularly characterized enteroendocrine mother cell ( EMC ) requires only low levels of Notch signaling [24] . ISCs and EBs ( referred to as progenitor cells ) reside basally next to the visceral muscles , while ECs cover the apical brush border ( Fig 1B ) . In both flies and mammals , Notch signaling plays the same central roles in the choice of an absorptive or secretory fate in the intestinal lineages [25] [26] . Drosophila ISCs express the Notch ligand , Delta ( Dl ) , which turns on Notch activity in its sibling cells for EB fate commitment [23 , 25] ( Fig 1C ) . Moreover , JAK/STAT signaling [14 , 27] , the transcription factors Escargot ( Esg ) [28–30] , Sox21a [31 , 32] , GATAe [33] , and Dpp signaling [34–36] have recently been shown to regulate progenitor differentiation . While stem cell proliferation has been the focus of most studies , the cellular mechanisms that mediate proper conversion of the expanded stem cell pool into mature intestinal cells especially during regeneration , are currently missing . Moreover , an integrated view of intestinal regeneration has not been established . Here , we investigate the cellular and genetic basis underlying efficient differentiation of progenitor cells during intestinal regeneration . Our data uncover that enhanced cell-cell contact between an ISC and its differentiating daughter , consolidated by cell adhesion molecules , is required for efficient Notch signaling and rapid progenitor differentiation into EC during regeneration . We further identify a regulatory cascade involving , sequentially , JAK/STAT signaling , Sox21a and GATAe , that functions in EBs and is required for rapid differentiation . Our integrated study of intestinal regeneration provides new insights into stem cell differentiation that likely apply to other systems .
To understand the molecular and cellular mechanisms underlying intestinal regeneration , we analyzed the behaviors of progenitors in the gut of flies orally infected with the gram-negative bacterium Erwinia carotovora carotovora 15 ( Ecc15 ) . Oral infection with Ecc15 causes damage to the intestinal epithelium , which is quickly repaired through activating ISC proliferation and progenitor differentiation to maintain tissue integrity [15] . Unless otherwise noted , we focused our study on the anterior midgut , a region where the relatively low overall cell density allows better identification of individual cells . Interestingly , progenitor pairs with extended contact were observed in intestines following infection ( Fig 1D and 1E ) , visualized by β-catenin staining ( Armadillo ( Arm ) in Drosophila ) . Most differentiating EBs were in extended contact with at least one cell with strong esg>GFP signal ( Fig 1F ) . Furthermore , ingestion of dextran sulfate sodium ( DSS ) , which damages the intestinal epithelium and activates regeneration [16] , also led to formation of progenitor pairs with extended cell-cell contact ( S1A Fig ) . Since extended progenitor contact was not observed in basal homeostatic conditions , these results indicate that increased progenitor contact area is a general feature of the regenerating intestine . Both Ecc15 infection and DSS treatment activate stem cell proliferation . To investigate if division of stem cells is required for the formation of extended progenitor contact , we used colcemid , a microtubule-depolymerizing drug , which blocks dividing cells in metaphase . The presence of colcemid suppressed the formation of extended contact that is normally induced by Ecc15 infection ( S1B and S1C Fig ) . This suggests that during regeneration stem cells first proliferate before generating progenitors with increased cell-cell contact . However , the formation of extended cell-cell contact was not affected in Sox21a mutant gut where EB to EC differentiation is blocked ( S1D and S1E Fig ) . Moreover , clusters of ISCs induced upon expression of a Notch RNAi using the progenitor specific driver esgTS also showed extended contact ( S1F Fig ) . Thus , the formation of increased cell contact during regeneration appears to be a stem cell intrinsic behavior , which occurs independently of developmental signals regulating terminal differentiation . Dl/Notch signaling plays a central role in determining the ISC and EB cell fate and is further involved in EB differentiation into EC . Since this signal transduction requires cell-cell contact between the signaling sending and receiving cells , the change in contact area likely affects Dl/Notch signaling dynamics [37] . We hypothesized that the extended progenitor contact observed in epithelial damage-induced regenerating intestines could promote efficient differentiation by enhancing Dl/Notch signaling . This led us to further analyze Notch signaling state in progenitor pairs of regenerating intestines by applying cell-type specific markers . To unambiguously identify ISCs , we used an endogenous Dl-GFP fusion line with an ISC restricted expression [38] ( Fig 1G ) . EBs were visualized by Su ( H ) -lacZ , a reporter gene of Notch activity [39 , 40] . We first confirmed the increase in progenitor contact upon bacterial infection ( Fig 1G and 1H ) . In line with previous results [5 , 25 , 41] , ISCs in unchallenged conditions largely undergo asymmetric division , which generates another self-renewing ISC ( Dl-GFP+ ) and a committed EB ( Su ( H ) -lacZ+ ) ( Fig 1G and 1I ) . When we quantified all progenitor combinations , including single Dl-GFP+ cells , Dl-GFP+—Dl-GFP+ pairs , Dl-GFP+—Notch+ pairs , Notch+—Notch+ pairs and single Notch+ cells , in both unchallenged and bacteria-infected ( Ecc15 , 12 hours post infection ) intestines , we uncovered a significant increase of the Dl-GFP+—Dl-GFP+ pairs in infected guts ( Fig 1I and 1J ) . Notably , the ratio of Dl-GFP+—Dl-GFP+ pairs increased from 5% in unchallenged intestines to around 40% in infected guts . This change was accompanied by a reduction in the proportion of single Dl-GFP+ cells in regenerating intestines , suggesting that most of them had recently divided . We also observed a drop in the ratio of Dl-GFP+—Notch+ pairs , from 62% to 42% . Collectively , these data are consistent with the notion that increased contact directly arises from newborn progenitor pairs that have just completed mitosis . Interestingly , 57% of the Dl-GFP+—Dl-GFP+ pairs had one cell showing weak but specific Notch activity in Ecc15-infected intestines as revealed by the expression of the Su ( H ) -lacZ reporter ( Fig 1K–1M; S2A and S2B Fig ) . Use of an antibody against Dl confirmed that both cells , including the one with weak Su ( H ) -lacZ expression , were indeed stem cells as defined by the expression of the Dl marker ( S2C Fig ) . We further excluded the possibility of EE differentiation , since the EE marker Prospero ( Pros ) was never observed in such Dl-GFP+—Dl-GFP+ pairs with Notch activity ( n>50 ) ( S2D Fig ) . Importantly , ISCs undergoing mitosis were never found to express Su ( H ) -lacZ reporter ( n>30 ) ( S2E Fig ) , indicating that Notch activity was established in one cell of a newly formed Dl-GFP+—Dl-GFP+ pair after mitosis ( S2F and S2G Fig ) . Although Dl-GFP+—Dl-GFP+ pairs formed during infection can arise either from single Dl-GFP+ cells or from Dl-GFP+—Notch+ progenitor pairs , these results support a symmetric expansion of the Dl+ cell pool followed by diverting a subset of them to be committed into EB and quickly differentiated , likely due to the presence of the extended cell-cell contact ( Fig 2L ) . We and others have previously shown that the transcription factor Sox21a is both necessary and sufficient for the differentiation of EB to EC [31 , 32] . This was supported by the observations that ISC progenies are blocked at the EB stage in the absence of Sox21a , while overexpressing Sox21a in progenitors induced their precocious differentiation into mature EC . However , it is not yet established through which mechanisms Sox21a regulates progenitor differentiation , especially during regeneration following intestinal damage . To further analyze the role of Sox21a in EB differentiation , we first monitored its expression levels in intestinal cells at different stages of their differentiation using a sGFP ( superfolder green fluorescent protein ) -tagged Sox21a transgene that is controlled by its own regulatory sequences [42] . Since it has been shown that Sox21a is expressed in both ISC and EB , cells that express Sox21a-GFP but not Su ( H ) -lacZ reporter are expected to be ISCs . As expected for a transcription factor , Sox21a-GFP signal was localized to the nuclei ( Fig 2A and 2B ) . In unchallenged 5–7 day-old adults , Sox21a-GFP was expressed in both ISC and EB , with higher levels in ISC than in EB within an ISC-EB pair ( Fig 2A , 2A’ and 2C ) . Oral infection with Ecc15 induced a marked increase of Sox21a-GFP expression in the progenitors ( Fig 2B and 2B’ ) . However , quantification of the relative intensity of Sox21a-GFP levels between cells within each ISC-EB pair indicated that regenerating intestines now expressed stronger Sox21a-GFP in the EB than in its sibling ISC ( Fig 2C ) . Careful examination further revealed that Sox21a-GFP levels started declining in middle-sized EBs at a stage when Notch activity remained high . Moreover , Sox21a-GFP expression was totally shut down before the Notch activity reporter ( Fig 2B ) . Intestinal regeneration can be triggered by activating JNK or Ras/MAPK signaling in progenitors [12 , 43 , 44] . Interestingly , activating either pathway in progenitor cells with esg-Gal4 tub-Gal80ts for 36 hours also elevated Sox21a-GFP expression ( S3A–S3E Fig ) , consistent with a previous observation [45] . However , akin to infection , stronger Sox21a-GFP expression was again seen in EBs , supporting the specific role of Sox21a in EB for EC differentiation . Collectively , this analysis shows that Sox21a displays a dynamic expression pattern , which coincides with the process of intestinal progenitor differentiation . As shown previously [31] , expressing Sox21a in progenitor cells with the esgTS driver for only 36 hours led to their differentiation into EC ( Fig 2D–2F ) . Thus , over-expressing Sox21a provides a useful framework to unravel the sequence of events underlying stem cell differentiation into mature ECs . Taking advantage of this approach , we monitored the cellular changes resulting from overexpressing Sox21a in progenitors . Cells that were precociously differentiating towards ECs ( termed “differentiating EBs” ) showed an increase in cell size , became polyploid and expressed Pdm1 , a marker for ECs [31] . Differentiating EBs accounted for around 40% of the esg>GFP+ progenitors expressing Sox21a , while they were barely seen in midgut progenitors of wild type flies ( Fig 2G ) . They retained a weak esg>GFP signal and were located at a similar basal position as genuine ISC/EB ( S3F Fig ) . In addition , each esg>GFP+ nest contained a slightly increased number of cells ( Fig 2F and 2H ) , consistent with the notion that Sox21a also promotes a low level of ISC proliferation . In homeostatic guts , most progenitor cells have limited membrane contact ( Fig 2D ) . In Sox21a overexpressing guts , nearly all the differentiating EBs displayed an extended plasma membrane contact , with at least one cell with a strong esg>GFP signal ( Fig 2E and 2E’; S3F Fig ) . Increased membrane contact was also reflected by the increased GFP signal at the membrane in progenitor cells expressing the membrane-tethered mCD8::GFP ( Fig 2E’ ) . To quantify changes in membrane contact , we measured contact area relative to the size of the smaller cell in more than 200 progenitor pairs . Maximal cell-cell contact between progenitors was reached as early as 36 hours after Sox21a expression was induced ( Fig 2I and 2J ) . Similarly to the situation observed with infection , expressing Sox21a using the progenitor driver esg-Gal4 tub-Gal80ts also induced Dl-GFP+—Dl-GFP+ pairs with one cell exhibiting Notch activity with a high frequency ( Fig 2K ) . The observation that over-expressing Sox21a induces progenitors to differentiate towards ECs rather than EEs [31] , suggests that Sox21a could enhance Notch activity . To test this idea , we analyzed how Sox21a impacts Notch signaling . Interestingly , Dl-GFP signal was enriched at the extended ISC-EB contact area upon Sox21a expression , in contrast to their even distribution in wild-type ISCs ( Fig 3A and 3B ) . Furthermore , transcriptomic analysis using FACS-sorted progenitors revealed a four to seven fold increase of Dl mRNA in intestinal progenitors expressing Sox21a for only 12 or 24 hours ( Fig 3C; S1 Table ) . The increase of Dl mRNA levels was further validated by quantitative PCR ( qPCR ) from dissected guts ( S4 Fig ) . Thus , expressing Sox21a in progenitors increased Dl transcription and the amount of Dl protein at the ISC-EB interface , which are both expected to reinforce efficient Notch signal transduction . Finally , knocking down Dl with two independent RNAi lines both suppressed Sox21a-induced differentiation ( Figs 3D and 3E and 4A; S5A and S5B Fig ) . Collectively , our data indicate that Sox21a mediates rapid differentiation at least in part by up-regulating Dl and by increasing progenitor contact , two features likely to enhance Notch signaling . Extended progenitor contact in regenerating guts suggested that cell adhesion molecules might be involved in rapid progenitor differentiation . E-cadherin ( E-cad ) forms a complex with β-catenin at adherens junctions and mediates homophilic cell-cell adhesion [46] . Using an endogenous E-cad-GFP fusion , we first confirmed that E-cad was enriched at the ISC-EB interface in Sox21a expressing intestine . Similarly , the localization of Arm was almost identical to E-Cad ( Fig 4B and 4C ) . Simultaneous depletion of either E-cad or arm for 36 hours completely suppressed the precocious differentiation phenotype seen in intestines expressing Sox21a ( esgTS>Sox21a ) ( Fig 4A; S5C and S5D Fig ) . In these intestines , progenitor pairs displayed reduced contact confirming an essential role of E-cad in the formation of extended cell-cell contact during rapid differentiation . Since E-cad can impact Wg/Wnt signaling [47] , the phenotype observed was possibly due to a requirement of Wg signaling for Sox21a-induced differentiation . However , neither activating ( with constitutively active β-catenin ) nor blocking Wg signaling ( with dominant-negative Pangolin ) affected Sox21a-induced differentiation ( S5H and S5I Fig ) . We conclude that E-cad-mediated cell-cell adhesion is required for Sox21a-induced differentiation , independently of Wg signaling , in line with a previous report [48] . Furthermore , another cell adhesion molecule , Connectin ( Con ) , which mediates homophilic cell-cell adhesion both in vitro and in vivo [49 , 50] , was up-regulated in progenitor cells expressing Sox21a in our progenitor-specific transcriptomic analysis ( Fig 4D; S1 Table ) . Using an antibody against Connectin , we confirmed its enrichment at the extended contact between progenitors in esgTS>Sox21a gut ( Fig 4E; S6A Fig ) . Like E-cad , Connectin was also crucial for Sox21a-induced differentiation , as its depletion abolished the occurrence of differentiating EBs ( Fig 4A and 4F; S5E Fig ) . To determine to which extent cell-cell adhesion can contribute to differentiation , we also overexpressed Connectin in progenitors . As expected , overexpressing Connectin using esgTS altered the morphology of progenitor cells with the formation of interconnected progenitors , sometimes forming big clusters ( Fig 4G; S6A–S6C Fig ) . These changes were also associated with an increase in ISC proliferation but not a blockage of EB differentiation ( Fig 4G’; S6H Fig ) . Consequently , progenitor tumors were not observed in esgTS>Connectin intestines despite the presence of large esg>GFP+ clusters . Surprisingly , many progenitors in esgTS>Connectin intestines displayed the same characteristics of differentiating EBs of esgTS>Sox21a intestines , including a weak esg>GFP signal , increased cell size , and extended contact with neighboring esg>GFPstrong progenitors ( Fig 4G’ ) . Moreover , regions with clusters of esg>GFP+ cells were devoid of EEs and progenitors in such regions were differentiating towards ECs as judged from their large cell size and polyploid nuclei ( Fig 4G and 4G’; S6B and S6C Fig ) . Further experiments indicated that Connectin overexpression in the progenitors increased the expression levels of Notch activity reporter Su ( H ) -lacZ ( Fig 4I ) and promoted EB-EC differentiation rather than EE differentiation ( Fig 4H–4L; S6I and S6J Fig ) . Therefore , an increase in cell adhesion by over-expressing Connectin in progenitors can promote their differentiation toward ECs , by enhancing Notch activity . Nevertheless , results obtained with the esgF/O system [14] did not support an essential role of E-cad or Connectin in basal intestinal turnover ( S6K–S6M Fig ) . Unexpectedly , knocking-down Connectin in the progenitors induced both mild stem cell proliferation and progenitor differentiation in the absence of a challenge ( S6D–S6H Fig ) , suggesting Connectin may have other functions in the maintenance of ISC under normal conditions . We conclude that the formation of extended cell contact between progenitors through adhesion molecules is required for Sox21a-induced rapid differentiation . Importantly , we show that this process is specifically required for the rapid differentiation of progenitors but not for basal low-level intestinal turnover during homeostatic conditions . Thus , our study reveals specific mechanisms that have evolved to accelerate the differentiation program . Having analyzed the cellular changes that enhance Notch signaling activity in the ISC-EB transition during intestinal regeneration , we went on to investigate how Sox21a contributes to the processes of differentiation from EB to EC . In Drosophila , Notch deficient stem cells over-proliferate , leading to the formation of tumors composed mostly of ISCs and intermingled with EEs [22 , 23 , 25 , 51] , while the over-activation of Notch signaling drives progenitors to differentiate into ECs [22 , 25] . The similarities between the function of Notch signaling and that of Sox21a in terminal differentiation led us to investigate their relationship . While Sox21a is expressed in both ISC and EB , Notch activity is only found in EB [22 , 23] . Thus , Sox21a expression in ISC is independent of Notch signaling . In addition , Notch activity was not blocked in a Sox21a mutant [31] . Several observations indicate that Sox21a and Notch signaling function interdependently for terminal differentiation . We first observed that Sox21a was required for the differentiation of progenitors into ECs upon Notch over-activation ( Fig 5A and 5B ) , and for the formation of tumors composed of ISCs and EEs in Notch deficient clones ( Fig 5C–5E ) . Conversely , the forced differentiation of progenitors into ECs upon expression of Sox21a was blocked in the absence of functional Notch signaling ( S7A–S7D Fig ) . Thus , over-expression of Sox21a cannot overcome the requirement of Notch signaling for the differentiation of EB . Collectively , this led us to conclude that Sox21a and Notch signaling encompass two parallel systems that need to cooperate to ensure terminal differentiation . Both Sox21a and JAK/STAT have been shown to be mandatory for the EB-EC differentiation , and knockdown of either of these two factors results in the formation of EB containing tumors . Mechanistically , the formation of these tumors is caused by a feed-back amplification loop whereby the differentiation-defective EBs secrete growth factors stimulating ISC proliferation [14 , 27 , 31 , 32] . We have previously shown that over-expression of Sox21a can partially rescue the differentiation defect caused by the loss of JAK/STAT in MARCM clones , suggesting that Sox21a functions downstream of JAK/STAT signaling in EB differentiation [31] . We further analyzed the relationship between JAK/STAT signaling and Sox21a by using this time an EB specific driver , Su ( H ) GBETS ( Fig 6A–6D ) . Knocking down Sox21a specifically in EB caused the accumulation of EBs and an increase in ISC mitosis resulting in strong tumor formation , recapitulating the Sox21a mutant phenotype ( Fig 6D–6F ) . In contrast , EB-specific depletion of Stat92E , the gene encoding the JAK/STAT transcription factor , had only mild consequences with the formation of small-sized EB tumors ( Fig 6C , 6E and 6F ) . Unexpectedly , expressing a dominant-negative form of Domeless ( DomeDN ) , the receptor of the JAK/STAT signaling cascade , using the same EB driver did not affect the differentiation process and did not lead to any tumor formation ( Fig 6B , 6E and 6F ) . However , silencing the JAK/STAT pathway in both ISC and EB with esgTS , using the same UAS-DomeDN or the UAS-Stat-RNAi constructs led to the formation of massive progenitor tumors ( Fig 6G and 6H ) . The observations that EB tumor formation requires the inactivation of JAK/STAT in both ISC and EB , and that knock-down of Sox21a only in EBs is sufficient to cause tumors , are consistent with the notion that JAK/STAT signaling is required earlier than Sox21a in the course of EB-EC differentiation . Supporting this hypothesis , we observed that expressing Sox21a could suppress tumor formation caused by loss of JAK/STAT signaling in progenitors ( Fig 6I–6L ) , indicating that Sox21a can promote progenitor differentiation in the absence of JAK/STAT signaling . Thus , the observations that JAK/STAT signaling functions earlier than Sox21a , and that Sox21a can override the differentiation blockage caused by the loss of JAK/STAT signaling confirm and extend our previous observation based on mosaic analysis [31] that Sox21a acts downstream of JAK/STAT pathway in EB-EC differentiation . The role of the Dpp signaling pathway in EB differentiation has been controversial , with studies supporting that it is essential to this process while others suggest it is dispensable [34–36] . The observation that several genes encoding components of Dpp signaling , including the receptor thickveins ( tkv ) , the transcription factors schnurri and Mothers against dpp ( Mad ) , are down-regulated in Sox21a mutant EBs , supported a role of Dpp signaling in EB to EC transition ( Fig 7A ) . We therefore explored its role in the differentiation process and its relationship with Sox21a . Interestingly , Dpp signaling was specifically induced in differentiating EBs in esgTS>Sox21a intestines as revealed by a reporter gene of Dpp signaling , Dad-GFPnls ( Fig 7B and 7C ) . The expression of the Dad-GFPnls reporter was much stronger in differentiating EBs than in mature ECs that were already differentiated prior to the activation of Sox21a ( Fig 7C ) , highlighting a role of Dpp signaling in the EB to EC transition . Importantly , Dpp signaling was mandatory for Sox21a-induced rapid differentiation , as depleting the key component Mad abolished progenitor differentiation of esgTS>Sox21a intestines ( Fig 4A; S5F Fig ) . These results support a role of the Dpp pathway in EB differentiation downstream of Sox21a . Nevertheless , overexpressing the Dpp transcription factor Schnurri , which is known to promote progenitor differentiation into EC in the midgut [36] , did not rescue the differentiation defect of Sox21a mutant clones ( Fig 7D and 7E ) . We conclude that Dpp signaling is required downstream of Sox21a for rapid differentiation but is not sufficient to rescue Sox21a deficiency . GATAe encodes a transcription factor , which is expressed in all the cell types of the fly midgut . It has recently been shown to be important for ISC proliferation and , to a lesser extent , for EB differentiation [33] . It also has a role in ECs to maintain the regionalization of the intestine [33 , 52] . Our RNA-seq experiments with FACS-sorted EBs revealed that the expression of GATAe was decreased in the absence of Sox21a ( Fig 7A ) , suggesting that this transcription factor could contribute to the differentiation program downstream of Sox21a . We therefore investigated in further detail the role of GATAe in the differentiation process and its relationship with Sox21a . We first observed that over-expressing GATAe under the control of esgTS driver led to the precocious differentiation of progenitors into Pdm1-positive ECs ( Fig 8A–8C ) , consistent with the notion that GATAe can promote progenitor differentiation [33] . The fast differentiation of progenitors was supported by the presence of many mature EC that still kept residual esg>GFP signal ( Fig 8B ) , reminiscent of progenitor cells in esgTS>Sox21a intestines . Consistent with [33] , loss of GATAe did not block terminal differentiation in basal conditions , since both Pdm1-positive ECs and Pros-positive EEs were found in stem cell clones deficient for GATAe ( S8A and S8B Fig ) . Similarly , EB-specific depletion of GATAe using the EB-specific driver Su ( H ) GBETS did not lead to EB tumors , but rather to a slight accumulation of late-stage EBs as judged from their appearance . These results suggest that GATAe may contribute to the rate of EB differentiation . To test this idea , we analyzed the contribution of GATAe to rapid epithelial turnover induced by ingestion of bacteria . Both wild-type flies and flies with EB-specific depletion of GATAe were orally infected with Ecc15 for 2 days , a treatment that increases the pool of progenitors undergoing differentiation , and were further let to recover for 3 days . At this timepoint , the midgut of Su ( H ) GBETS>w control flies subjected to bacterial challenge had already returned to a homeostatic condition where only nascent EBs with small nuclei were found ( Fig 8D ) . In sharp contrast , accumulation of EBs was observed along the midgut of Su ( H ) GBETS>GATAe-IR flies ( Fig 8E ) . These EBs had a larger cell size than EBs found in wild-type flies suggesting that they were stuck in the process of differentiation . Consistent with a role of GATAe for accelerated EB differentiation , simultaneously depleting GATAe abolished progenitor differentiation in esgTS>Sox21a intestines ( Fig 4A; S5G Fig ) . Moreover , expressing GATAe with the esgTS driver suppressed the differentiation defect and tumor formation induced by the loss of either Sox21a or Stat92E ( Fig 8F–8L ) , revealing that GATAe acts downstream of Sox21a and JAK/STAT . Collectively , our study not only confirms that GATAe contributes to the differentiation process [33] , but further reveals its critical role during regeneration as opposed to basal conditions . Our data also show that JAK/STAT-Sox21a-GATAe forms a sequential relay orchestrating the EB-EC differentiation process .
Key questions in stem cell biology are how the pool of stem cells can be robustly expanded yet also timely contracted through differentiation to generate mature cells according to the need of a tissue , and what are the underlying mechanisms that couple stem cell proliferation and differentiation . Over the last years , the mechanisms underlying intestinal stem cell activation have been extensively studied in both flies and mammals [1 , 4] , while the genetic control of progenitor differentiation , especially during regeneration , has only recently begun to be understood [26 , 28 , 31] . The transcription factor Sox21a has recently been the focus of studies in fly intestines [31 , 32 , 45] . Using a Sox21a-sGFP transgene , we uncovered its dynamic expression pattern in intestinal progenitors . Higher levels of Sox21a were found in ISC during homeostatic conditions but in EB during regeneration , supporting the roles of Sox21a in both ISC maintenance and EB differentiation at different conditions . The highly dynamic expression pattern of Sox21a revealed by this sGFP-tagged transgene per se argues against accumulation and perdurance of GFP fusion protein . Indeed , immunostaining using an antibody against Sox21a also indicated stronger Sox21a expression in ISC in homeostatic condition and global activation of Sox21a in progenitors under DSS-induced regeneration [45] . However , Chen et al . , ( 2016 ) suggested that Sox21a levels are always higher in EB than in ISC by applying another antibody against Sox21a . The inconsistency between these studies may have arisen from the differences in EB stages examined or the sensitivity of respective detection approaches . In this study , we have analyzed the cellular processes required for efficient progenitor differentiation during regeneration and uncovered three main findings revealing: i ) the importance of extended contact between a stem cell and its differentiating daughter , ii ) the existence of specific mechanisms allowing fast differentiation during regeneration , and iii ) the characterization of a genetic program instructing the transition from EB to EC . These results together led us to propose a molecular framework underlying intestinal regeneration ( Fig 9 ) that is discussed below step by step . By studying the mechanisms of Sox21a-induced differentiation , we found that ISC establishes extended contact with its differentiating daughter within a progenitor pair . Increased interface contact was not only observed upon Sox21a expression but also during regeneration after bacterial infection and DSS-feeding . Since the presence of extended contact is rare in intestinal progenitors under homeostatic conditions , we hypothesize that extended contact between progenitors is related to increased epithelial renewal as a mechanism to elicit optimal juxtacrine Notch signaling to accelerate the speed of progenitor differentiation . The observations that down-regulation of the cell adhesion molecules E-Cadherin or Connectin suppresses rapid progenitor differentiation upon regeneration , and that overexpression of Connectin is sufficient to promote differentiation , underline the importance of increased cell-cell contact in rapid differentiation . Our study shows that one early role of Sox21a is to promote the formation of this contact zone , possibly through transcriptional regulation of Connectin . Further studies should identify the signals and pathways leading to the change of contact between progenitors to adjust the rate of differentiation . Intestinal progenitors with extended contact in non-homeostatic midguts have been observed in some studies [14 , 41] , but their role and significance have not been analyzed . Previous studies have also shown that progenitor nests are outlined by E-Cadherin/β-Catenin complexes [23 , 48] , yet it was not known whether different degrees of progenitor contact are associated with their ISC versus EB fate . Consistent with our results , recent modeling analyses suggested a positive correlation between the contact area of progenitor pairs and the activation of Notch signaling [53 , 54] . Thus , it seems that an increase in the contact area between intestinal progenitors is a hallmark of progenitors that are undergoing accelerated differentiation towards ECs . Another study by Choi et al . ( 2011 ) has suggested an inhibitory role of prolonged ISC-EB contact to restrict ISC proliferation . Collectively , these studies and our findings suggest that the strong contact between ISC and EB promotes on one hand the efficient differentiation of EBs into mature intestinal cells while on the other hand preventing stem cells from over-dividing . Thus , we hypothesize that alteration in the contact zone provides a mechanism for ensuring both the appropriate speed of differentiation and the timely resolution of stem cell proliferative capacity . A second finding of our study consists in revealing the existence of specific mechanisms accelerating differentiation for tissue replenishment . In addition to the extended contact discussed above , we observe a difference in the pattern of ISC division between homeostatic and highly regenerative intestines . The modes of ISC division in Drosophila have been the topic of intense discussion , and the general consensus is that it is associated with an asymmetric cell fate outcome , in which one cell remains an ISC and the other engages in differentiation [5 , 41 , 55 , 56] . In line with these previous studies , our results support the notion that asymmetric cell division is the most prevalent mode of ISC division under homeostatic conditions , where the rate of epithelial renewal is low . However , our use of ISC- and EB-specific markers shows that upon rapid regeneration an ISC divides into two cells both expressing the ISC marker Dl-GFP but with one cell showing weak Notch activity . Similarly to other Notch-mediated cell-fate decision systems [57] , our study suggests that the two resulting Dl-GFP+ cells from a symmetric division stay in close contact and compete for the stem cell fate . While our study is not the first to postulate the existence of symmetric ISC division [5 , 55] , the use of reliable ISC- and EB-specific markers allows us to better visualize this process . Applying a dual-color lineage tracing system to unravel the final fate of respective cells in a Dl+—Dl+ pair could reinforce the existence of symmetric stem cell division . This is nevertheless technically challenging to apply here since all the current available lineage-tracing settings require a heat shock to initiate the labeling , which affects intestinal homeostasis . Importantly , we show that the genetic program required for fast intestinal regeneration differs from the one involved in basal intestinal maintenance . Our study indicates that GATAe , Dpp signaling , and the cell adhesion molecules E-cadherin and Connectin are not critical for progenitor differentiation when the rate of epithelial renewal is low , whereas their roles become crucial upon active regeneration . We speculate that many discrepancies in the literature can be reconciled by taking into consideration that some factors are required only for rapid differentiation but not in basal conditions . For instance , the implication of Dpp signaling in differentiation has been disputed , since Zhou et al . ( 2015 ) focused on bacterial infection-induced regeneration while the other two studies dealt with basal conditions [34–36] . Our study here points to a clear role of Dpp signaling in the differentiation process upon regeneration . Therefore , better defining the genetic program that allows adjusting the speed of differentiation would be of great interest . Cell fate determination and differentiation involve extensive changes in gene expression and possibly also gradual change of cell morphology . The EB to EC differentiation in the adult Drosophila intestine provides a model of choice to study this process . This transition includes changes in cell shape , an increase in cell size , DNA endoreplication leading to polyploidy and the activation of the set of genes required for EC function ( S9A–S9C Fig ) . In this study , we have integrated a number of pathways ( Notch , JAK/STAT and Dpp/BMP ) and transcription factors ( Sox21a and GATAe ) into a sequential framework . We further show that Sox21a contributes to the EB-EC transition downstream of JAK/STAT but upstream of Dpp signaling and GATAe . The recurrent use of several factors , namely JAK/STAT , Sox21a and GATAe at different processes including ISC self-renewal and EB-EC differentiation is likely to be a general feature during cell fate determination , and somehow also complicates the study of differentiation . Future work should analyze how each of the factors interacts with the other in a direct or indirect manner . It would be interesting as well to further study how these factors shape intestinal regionalization as the gut exhibits conspicuous morphological changes along the length of the digestive tract [52 , 58] . Several of the findings described are likely to apply to the differentiation program that takes place in mammals . Since Notch signaling plays major roles in stem cell proliferation and cell fate specification from flies to mammals [57 , 59] , it would be interesting to decipher whether in mammals changes in progenitor contact also impact differentiation speed and whether a specific machinery can accelerate progenitor differentiation when tissue replenishment is required .
Fly strains were kept on a standard medium ( maize flour , dead yeast , agar and fruit juice ) . esg-Gal4 , tub-Gal80TS , UAS-GFP ( referred to as esgTS ) ; Su ( H ) GBE-Gal4 , tub-Gal80TS , UAS-GFP ( referred to as Su ( H ) GBETS ) ; esg-Gal4 , tub-Gal80TS , UAS-GFP; UAS-Flp , Act>>Gal4 ( referred to as esgF/O ) ; MARCM tester FRT2A: y , w , hsFlp; tub-Gal4 , UAS-CD8::GFP; FRT2A , tub-Gal80; MARCM tester FRT82B: y , w , hsFlp , tub-Gal4 , UAS-nlsGFP;;FRT82B , tub-Gal80; Su ( H ) -lacZ , Sox21a6 , UAS-Sox21a , UAS-Sox21a-RNAi ( BL53991 ) , UAS-Stat92E-RNAi ( BL31318 and 35600 ) and UAS-N-RNAi ( VDRC100002 ) have been described before [31] . Dl-GFP , UAS-hep , UAS-RafACT , UAS-lacZ , UAS-NICD , UAS-armS10 , UAS-PanDN , UAS-E-cad-RNAi ( BL27689 ) , UAS-Con-RNAi ( BL28967 ) , UAS-Dl-RNAi ( BL28032 and 34322 ) , UAS-arm-RNAi ( BL31304 and 31305 ) , UAS-Mad-RNAi ( BL31315 ) , UAS-GATAe-RNAi ( BL 34907 ) , UAS-Notch-RNAi ( VDRC , KK ) and UAS-mCherry-RNAi were obtained from Bloomington Drosophila Stock Center ( BDSC ) . Sox21a-GFP was from VDRC stock center . UAS-shn , UAS-Stat92E and UAS-domeDN ( gift from Michael Boutros ) , FRT82B , NeurIF65 ( gift from Allison Bardin ) , Dad-GFPnls ( gift from Fisun Hamaratoglu ) , UAS-Con ( gift from Rob White ) and FRT82B , GATAe1 ( gift from Takashi Adachi-Yamada ) were also used . UAS-GATAe was generated in this study . Dl-GFP ( BL59819 ) encodes an endogenously GFP-tagged Dl protein resulting from recombination mediated cassette exchange of a Mi{MIC} insertion in the Dl coding intron [38] . This line is homozygous lethal . Sox21a-GFP transgenic line derives from a GFP-tagged fosmid clone containing a large genomic region including Sox21a [42] . In most cases , the driver lines ( esgTS or Su ( H ) GBETS ) were crossed to the w1118 strain , UAS-mCherry-RNAi , or UAS-lacZ , and the progenies were used as control for overexpression experiments . w1118 flies carrying one copy of esgTS ( esgTS>w ) were used as wild type to visualize the contact between progenitors in different conditions . Erwinia carotovora carotovora15 ( Ecc15 ) was grown in LB medium at 29°C with shaking overnight , and harvested by centrifugation at 3000g at 4°C for 30 minutes . The pellet was then suspended in the residual LB , and bacterial concentration was adjusted to OD600 = 200 . Flies older than 3 days were first dry-starved in an empty tube for 2 hours , and then transferred into a classical fly food vial containing a filter paper that totally covers the food and was soaked with a solution consisting of 5% sucrose and Ecc15 at OD200 ( 1:1 ) , or 6% DSS ( average MW 40 kDa , sigma ) treated flies were kept at 29°C until dissection . Colcemid treatment was done as reported previously [34] . 200ug/ml colcemid ( Sigma ) was added to 5% sucrose to pre-treat the flies for 12 hours , and then an Ecc15 infection was performed in the presence of 200ug/ml colcemid . Flies were transferred overnight into a classical fly food vial containing a filter paper soaked with a solution consisting of 5% sucrose to clean the digestive tract . Then , 10–15 intestines of mated adult females were dissected in phosphate-buffered saline ( PBS ) , and fixed for at least one hour at room temperature in 4% paraformaldehyde ( PFA ) in PBS . They were subsequently rinsed in PBS+0 . 1% Triton X-100 ( PBT ) , permeabilized and blocked in 2% BSA PBT for one hour , and incubated with primary antibodies in 2% BSA PBT for overnight at 4°C . After one hour of washing , secondary antibodies and DAPI were applied at room temperature for two hours . Primary antibodies used are: mouse anti-Pros ( DSHB , 1:100 ) , mouse anti-Arm ( DSHB , 1:100 ) , mouse anti-Dl ( DSHB , 1:100 ) , mouse anti-βPS ( DSHB , 1:100 ) , mouse anti-Con ( DSHB , 1:4 ) , rabbit anti-pH3 ( Millipore , 1:1000 ) , Chicken anti-GFP ( Abcam , 1:1000 ) , rabbit anti-βGal ( Cappel , 1:1000 ) , mouse anti-βGal ( Sigma , 1:1000 ) , and Rat anti-mCherry ( Life Technologies , 1:500 ) . Alexa488- , Alexa555- or Alexa647-conjugated secondary antibodies ( Life Technologies ) were used . Nuclei were counterstained by DAPI ( Sigma , 1:10’000 ) . All the images were taken on a Zeiss LSM 700 confocal microscope at BIOP in EPFL . Images were processed using Image J and Adobe Photoshop software . Shown in figures are maximal intensity projections of all the confocal z stacks . To generate the UAS-GATAe construct , the following primers ( caccATGGTCTGCAAAACTATCTC and TTAGTTATTCGATGATCGCTC ) were used to amplify the 2 . 2kb GATAe-PA coding regions from cDNA clone LD08432 purchased from DGRC . The PCR product was first cloned into pENTR-D-TOPO ( Life Technologies ) vector , and then swapped into pTW destination vector to make UAS-GATAe . Transgenic flies were established by standard P element-mediated germ-line transformation ( BestGene Inc . ) . At least three independent transgenic lines were tested for expression level . The TARGET system was used in combination with the indicated Gal4 drivers to conditionally express UAS-linked transgenes [60] . Flies were grown at 18°C to limit Gal4 activity . After 3–5 days at 18°C , adult flies with the appropriate genotypes were shifted to 29°C , a temperature inactivating the temperature-sensitive Gal80’s ability to suppress Gal4 , and dissected after indicated time of transgene activation . Mosaic analysis with a repressible cell marker ( MARCM ) technique was used for clonal analysis [61] . For clone induction , 3-5-day-old flies with the appropriate genotypes were heat-shocked for 30 min at 37 . 5°C in a water bath . The flies were immediately transferred into a new tube and kept at 25°C or indicated temperature until dissection . Overexpression experiments were performed by combining the UAS-linked transgenes with the FRT2A , the FRT2A , Sox21a6 , or the FRT82B , NeurIF65 chromosome . Note that UAS-linked transgenes were only expressed in the clones indicated by the presence of GFP . esgTS virgin females were crossed to either w1118 as control or UAS-Sox21a for overexresspion at 18°C . Eclosed flies were maintained at 18°C for 5–7 days . Around 50 flies for each biological replicate were dissected in ice-cold 1xPBS made with DEPC-treated water under a dry-ice chilled dissecting microscope , within a one-hour time frame . Proventriculus , hindgut and midgut/hindgut junction were removed to collect only midgut esg>GFP positive cells . Two biological replicates were performed , and the activation of Sox21a expression was done by shifting esgTS>Sox21a flies to 29°C for 12 hours and 24 hours , respectively . Cell dissociation , FACS sorting , total RNA isolation and mRNA amplification were performed as described [31] . RNA-seq was performed on a Hi-Seq2000 ( Illumina ) with 100 nt single-end sequencing , and sequencing data was analyzed as described before [31] . Sequencing data will be deposited in public database prior to publication . Total RNA was extracted from dissected whole guts ( 12–15 guts per sample ) using Trizol and cDNA was synthesized using the PrimeScript RT reagent Kit ( TaKaRa ) . 0 . 5μg total RNA was used for reverse transcription with oligo dT , and the 1st strand cDNA was diluted 10–20 times with water to be further used in real time PCR . Real time PCR was performed in triplicate for each sample using SYBR Green ( Roche ) on a LightCycler 480 System ( Roche ) . Expression values were calculated using the ΔΔCt method and relative expression was normalized to Act5C . Results are shown as mean ± SEM of at least 3 independent biological samples . Statistical analysis was performed in Prism Software using the unpaired t test . Primers used for qPCR are as follows . All analyses were done with GraphPad Prism . Unpaired t test were used unless otherwise noted . p values are indicated by *p < 0 . 05 , **p < 0 . 01 , ***p < 0 . 001 , ns: p > 0 . 05 . Shown are means and SEM . Data are representative of at least three experiments . Each dot represents one gut except Fig 5E . Sample size is also indicated in the figures .
|
Adult tissue/organ function is maintained by stem cells . Key question in stem cell biology is how the pool of stem cells can be robustly expanded yet timely contracted through differentiation according to the need of a tissue . Over the last years , the mechanisms underlying stem cell activation have been extensively studied , while the genetic control of progenitor differentiation , especially during regeneration , is still poorly understood . Using the fruit fly Drosophila midgut as model , we investigate the cellular changes and the genetic program required for efficient progenitor differentiation during intestinal regeneration . We first detect the presence of extended cell-cell contact between a stem cell and its differentiating daughter in regenerating intestine , compared to homeostatic conditions . The extended cell-cell contact is consolidated by cell adhesion molecules and enhances Notch signaling in the differentiating progenitors leading to their fast differentiation into enterocytes . We further uncover a genetic program , involving the JAK/STAT and Dpp signaling , the Sox21a and GATAe transcription factors , which acts in the differentiating progenitors to instruct their terminal differentiation . Thus , our study presents an integrated view of stem cell differentiation during tissue regeneration and the findings here are likely to apply to mammals .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"cancers",
"and",
"neoplasms",
"cloning",
"animals",
"notch",
"signaling",
"cell",
"differentiation",
"oncology",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"stem",
"cells",
"experimental",
"organism",
"systems",
"differentiated",
"tumors",
"molecular",
"biology",
"techniques",
"drosophila",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"dpp",
"signaling",
"cascade",
"molecular",
"biology",
"insects",
"arthropoda",
"gastrointestinal",
"tract",
"signal",
"transduction",
"anatomy",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"organisms",
"signaling",
"cascades"
] |
2017
|
A genetic framework controlling the differentiation of intestinal stem cells during regeneration in Drosophila
|
The retinoblastoma tumor suppressor ( Rb ) is a potent and ubiquitously expressed cell cycle regulator , but patients with a germline Rb mutation develop a very specific tumor spectrum . This surprising observation raises the possibility that mechanisms that compensate for loss of Rb function are present or activated in many cell types . In particular , p107 , a protein related to Rb , has been shown to functionally overlap for loss of Rb in several cellular contexts . To investigate the mechanisms underlying this functional redundancy between Rb and p107 in vivo , we used gene targeting in embryonic stem cells to engineer point mutations in two consensus E2F binding sites in the endogenous p107 promoter . Analysis of normal and mutant cells by gene expression and chromatin immunoprecipitation assays showed that members of the Rb and E2F families directly bound these two sites . Furthermore , we found that these two E2F sites controlled both the repression of p107 in quiescent cells and also its activation in cycling cells , as well as in Rb mutant cells . Cell cycle assays further indicated that activation of p107 transcription during S phase through the two E2F binding sites was critical for controlled cell cycle progression , uncovering a specific role for p107 to slow proliferation in mammalian cells . Direct transcriptional repression of p107 by Rb and E2F family members provides a molecular mechanism for a critical negative feedback loop during cell cycle progression and tumorigenesis . These experiments also suggest novel therapeutic strategies to increase the p107 levels in tumor cells .
The retinoblastoma gene Rb was initially identified as a prototypic tumor suppressor through its association with hereditary retinoblastoma; mutations in Rb or in genes that play a role in the regulation of Rb function are found in virtually all types of human cancers . The best-described function of Rb is to act as a transcriptional co-factor: Rb regulates the activities of numerous transcription factors and recruits chromatin remodeling complexes to control the expression of genes involved in the control of cell cycle progression , differentiation , and senescence . It is generally thought that the E2F family of transcription factors , consisting of both activating members ( E2F1 , E2F2 , E2F3a ) and some of the repressing members ( E2F3b , E2F4 , E2F5 ) , are the most critical downstream mediators of Rb function in the control of cell cycle progression ( reviewed in [1]–[3] ) . Although Rb is expressed in nearly all cell types [4] , patients and mice carrying heterozygous mutations for the Rb gene are not strongly predisposed to a broad range of tumors [5]–[9] . Perhaps most strikingly , conditional deletion of Rb in the mouse retina is insufficient to induce retinoblastoma [10]–[13] , in sharp contrast to what is observed in human patients . After it was found that Rb is a member of a three-gene family , along with p107 and p130 , it was quickly hypothesized that one or both of these other Rb family members may be able to compensate for the absence of Rb in specific cell types . Indeed , Rb/p107 and Rb/p130 double knock-out mice develop retinoblastoma [10]–[14] . The ability of p107 to compensate for loss of Rb has since been observed in numerous cell types , beyond the mouse retina [13]–[19] . The observation that the presence of p107 or p130 is able to suppress some phenotypes in the absence of Rb has raised the question of what molecular mechanisms underlie this compensatory activity . Of the three Rb family members , p107 is thought to be mostly regulated at the transcriptional level [20]–[22]; p107 mRNA and protein levels are generally low in non-cycling cells , and expression increases as cells enter late G1 and S-phase [22] , at a time when the protein is being functionally inactivated through phosphorylation . Because loss of Rb often results in increased levels of p107 mRNA in mammalian cells [18] , [23]–[26] , an appealing model is that the absence of Rb directly affects p107 transcription , resulting in genetic compensation rather than general functional redundancy . The 5′ regulatory region of the human p107 gene contains two consensus E2F consensus binding sites ( TTTSSCGC where S is G or C ) [27] that are almost completely conserved among mammals ( Figure 1A ) . These tandem E2F sites contribute to the appropriate cell-cycle induction of the human p107 promoter in reporter assays [22] . In addition , E2F transcription factors directly bind to the p107 promoter in a cell cycle-dependent manner suggesting a model in which activating E2Fs activate the p107 promoter in late G1 and S while repressing E2Fs are associated with the p107 promoter in G0 and early G1 [28] , [29] . However , many of these reporter assays were performed using a minimal p107 promoter transiently expressed in tumor cells lines , and may not fully recapitulate the regulation of the endogenous allele . In addition , chromatin immunoprecipitation ( ChIP ) experiments did not identify the exact sequences bound by E2F in the p107 promoter , did not rule out that E2F could bind to other sequences , and did not determine if the two consensus sites were bound differently by different E2F family members . Moreover , from these experiments , it is still unclear how the cell cycle-dependent regulation of p107 contributes to the cellular functions of p107 . Finally , as Rb controls the activity of multiple transcription factors , whether the E2F binding sites or other transcription factor binding sites [30]–[32] ( Figure 1B ) mediate the repressive effects of Rb on the p107 promoter is still unknown . Traditional knockout studies , which delete an entire gene and modify multiple functional interactions in cells , may often obscure the importance of individual regulatory loops . To investigate the mechanisms underlying transcriptional regulation of p107 in vivo , we used gene targeting in mouse embryonic stem cells ( mESCs ) to engineer point mutations in the two E2F binding sites in the endogenous p107 promoter . Disruption of this specific cis-acting node in the Rb/E2F network has allowed us to show that the tandem E2F binding sites in the p107 promoter dynamically regulate p107 levels in wild-type and Rb-deficient cells and control p107 function during S phase .
To understand the functions of the tandem E2F consensus binding sites in the mouse p107 promoter , we first generated a series of four luciferase constructs in which inactivating point mutations [33] were introduced into the two consensus E2F sites , either individually or together , into a construct containing 900 bp of the mouse p107 promoter ( Figure 1B ) . These constructs were transfected into wild-type mouse embryonic stem cells ( mESCs ) , which contain high levels of activating E2F transcription factors as well as limited Rb family function , due to hyperphosphorylation of the Rb family proteins through high activity of Cyclin/Cdk complexes [34]–[36] . Because mESCs are rapidly cycling , we reasoned they would provide a good system to investigate the role of these consensus E2F sites in the activation of the p107 promoter . In mESCs , the p107-1* construct , which contains a mutation in the more distal consensus E2F binding site , expressed less than half of the luciferase activity of the wild-type construct , whereas the p107-2* construct , which retains the more distal site but contains a mutant proximal site , expressed nearly 70% of the wild-type activity ( Figure 1C ) . Simultaneous mutation of the two sites in the p107-1*2* construct resulted in even lower expression , suggesting that both sites contributed to some extent to the activation of the p107 promoter , although the distal site appeared to mediate the majority of the activation . We also found that exogenous E2F3 efficiently enhanced the activity of the wild-type and the p107-2* constructs and to a much lesser extent that of the p107-1* and p107-1*2* mutant promoters ( Figure 1C ) . These observations further suggested that the distal E2F consensus binding site mediated the majority of the activation of p107 by E2F but did not exclude that some activation of p107 could be mediated through the proximal site in this context . We next investigated the potential role of E2F-mediated transcriptional repression in the control of p107 transcription in quiescent cells , where p107 levels are normally low . Because mESCs do not stably arrest in G0 , the four reporter constructs were transfected into wild-type MEFs that were then serum-starved for 24 hours in order to induce cell cycle exit in G0 . In this system , the p107-1* construct showed a significant trend towards increased reporter expression , suggesting that the distal site mediated a significant amount of repression of the p107 promoter that could not be compensated by the presence of the proximal site ( Figure 1D ) . However , the significant increase in luciferase activity found with the p107-1*2* construct over the p107-1* ( Figure 1D ) construct also suggested that both sites contributed to repression of p107 to some extent in this context . Together , these results are indicative of a model for the mouse p107 promoter in which the distal site is most significant for both the activation and repression of the p107 promoter , while the proximal site contributes to a lesser extent to both functions ( Figure 1E , right ) . In contrast , previous experiments using the human promoter had generated a model in which the distal consensus E2F binding site had a more significant role in repression of p107 while the proximal site was more important for activation [22] , [27] ( Figure 1E , left ) . This difference between the mouse and the human promoters could potentially result from a single polymorphism between the mouse and human sequences in the proximal consensus E2F binding site ( Figure 1A ) ( see Discussion ) . Nevertheless , these results with plasmid reporters also underscore the fact that the two tandem E2F consensus sites may perform different functions in different contexts , raising the question of their respective roles in the control of the endogenous p107 gene . To examine the role of the consensus E2F binding sites in the regulation of the endogenous p107 promoter , we generated a series of targeting vectors designed to knock-in the same series of mutations as those described for the luciferase vectors into the endogenous p107 locus ( Figure 2A ) . These vectors were electroporated into mESCs in order to generate heterozygous cells . Targeting was confirmed by both 5′ and 3′ Southern analysis , as well as by sequencing of amplified genomic DNA ( Figure 2B and data not shown ) . Wild-type control mESCs were generated through targeting events in which the Neomycin resistance cassette was correctly targeted but the E2F binding sites remained untargeted . Targeted cells were then infected with an adenovirus expressing the Cre recombinase in order to remove the resistance cassette , followed by a second round of targeting ( Figure 2C ) . This procedure generated homozygous cells of three knock-in genotypes: p107E2F-1*/1* cells with mutations in the distal site of both p107 alleles , p107E2F-2*/2* cells with mutations in the proximal site , and p107E2F-1*2*/1*2* cells with mutations in both E2F sites on both alleles of p107 . Targeting was verified by Southern and sequencing analysis ( Figure 2D and data not shown ) . To determine the role of the consensus E2F binding sites in controlling endogenous p107 expression , we first examined p107 mRNA levels in control and homozygous mutant mESCs . In a pattern similar to the luciferase assays , quantitative RT-PCR ( RT-qPCR ) analysis showed that p107E2F-1*/1* cells expressed 40-50% of the p107 mRNA expressed by wild-type cells while p107E2F-2*/2* cells expressed 70%; p107E2F-1*2*/1*2* cells expressed the lowest amounts of p107 in mESCs ( Figure 2E ) . The relative levels of p107 mRNA in these mutant cells reflected actual p107 protein expression ( Figure 2F ) , providing additional evidence that p107 levels in these cells are regulated largely at the transcriptional level and suggesting that E2F activity is involved in this transcriptional control . Importantly , the basal p107 promoter remained active despite the knock-in mutations , validating this knock-in approach to investigate the functional importance of discrete elements in the p107 promoter . We next examined if the point mutations introduced into the p107 promoter affected the binding of E2F transcription factors and p107 , the member of the Rb family with the highest level of expression in cycling cells [37] , to the p107 regulatory regions , an experiment that was not possible without the knock-in mutant cells . Using quantitative ChIP analysis , we found that both E2F3 and E2F4 bound to the wild-type p107 promoter in mESCs ( Figure 2G ) . The p107E2F-1*/1* and p107E2F-1*2*/1*2* cells demonstrated no binding for these two E2F family members to the p107 promoter ( Figure 2G ) , consistent with the decreased levels of p107 expression observed in these cells ( Figure 2E–2F ) . In all cases , there was significant binding of both E2Fs to the promoter of a control gene , B-Myb ( Figure 2G ) . Rb family members are largely inactivated by hyperphosphorylation in undifferentiated mESCs [36] . As expected , we did not detect any significant binding of p107 on the p107 or B-Myb promoters in wild-type or knock-in mutant cells under these conditions ( Figure 2G ) while p107 binding could be observed in asynchronously cycling mouse fibroblasts at the same promoter regions ( data not shown ) . These experiments show that E2F transcription factors require the consensus E2F binding sites for binding to the p107 promoter region . They also suggest that the distal E2F consensus binding site is the major mediator of E2F activity controlling p107 transcription in cycling mESCs . To understand the function of the E2F binding sites in the control of p107 during a more normal cell cycle and in G1 arrest , we grew MEFs from chimeric embryos generated from homozygous p107E2F-1*/1* , p107E2F-1*2*/1*2* , and control mESCs as indicated in Figure 3A . Due to lower numbers of chimeric embryos derived from the p107E2F-1*/1* cells , experiments were only performed in this genotype after immortalization . Primary control and p107E2F-1*2*/1*2* MEFs were first rendered quiescent in low serum . We found that p107E2F-1*2*/1*2* MEFs expressed two times more p107 mRNA than did the wild-type cells in the same conditions ( Figure 3B ) . This fold increase , while somewhat variable depending on the MEF line ( see below ) , was always observed and is similar to the increased levels of reporter activity observed with the p107-1*2* luciferase construct in quiescent MEFs ( Figure 1D ) . A similar increase was observed with p107 protein levels in these cells ( Figure 3C ) . These data showed that p107 expression was repressed in G0 through the E2F binding sites present in its proximal promoter region . Accordingly , we found a significant decrease in E2F4 binding to the p107 promoter in quiescent immortalized p107E2F-1*/1* and p107E2F-1*2*/1*2* MEFs compared to controls ( Figure 3D ) , underscoring the role of the distal E2F binding site in the control of p107 repression in G0-arrested cells . p107 levels are low in quiescent cells and we did not observe any significant binding of p107 to the B-Myb promoter in wild-type or p107E2F-1*/1* and p107E2F-1*2*/1*2* quiescent immortalized MEFs ( Figure 3D ) . We found some binding of p107 to its own promoter in wild-type cells just above the non-specific signal found with the control antibody , and there was a trend towards decreased binding in the knock-in mutant cells ( Figure 3D ) , even though p107 levels are higher in the mutant cells ( Figure 3C ) . p130 binding to the p107 promoter was decreased in p107E2F-1*/1* and p107E2F-1*2*/1*2* MEFs compared to controls; interestingly , p130 binding to the B-Myb promoter was also decreased in the mutant cells ( Figure 3D ) . It is possible that increased p107 levels in the mutant quiescent cells may alter the composition of the protein complexes between members of the Rb and E2F families . One obvious candidate whose binding to the p107 and B-Myb promoters could also be affected and could influence p130 binding in the mutant cells is Rb itself . Detection of murine Rb at the promoters of E2F target genes has proven challenging , especially in cells with low levels of Rb such as MEFs [29] , [38]–[40] . While we have not measured Rb binding to the p107 promoter in quiescent cells , we were able to detect Rb binding to the p107 and Mcm3 promoters in cycling immortalized MEFs . We found that Rb binding to the p107 promoter was decreased to close to background levels in knock-in mutant cells but not changed at the promoter of the Mcm3 gene ( Figure 3E ) . Because of the low intensity and the variability of the ChIP signal for Rb , we cannot exclude that Rb may be binding to other sites in the p107 promoter . However , altogether , these observations support a model in which Rb/E2F complexes bind to the p107 promoter through the E2F consensus binding sites . To determine whether the de-repression observed in the p107E2F-1*2*/1*2* MEFs was due to the loss of a repression complex involving Rb family members , control and p107E2F-1*2*/1*2* MEFs were infected with retroviruses stably expressing shRNA molecules directed against Rb or p130 ( Figure 4A , top ) . As previously shown [18] , [20] , Rb knock-down in quiescent MEFs resulted in an increase in p107 protein levels ( Figure 4A , top ) . In contrast , we did not observe an increase in p107 expression in cells with a p130 knock-down ( Figure 4A , top , and data not shown ) . We could not functionally test if low levels of p107 expression altered its own transcription , although knockdown of p107 has no effect on the expression of an eGFP transgenic reporter for p107 in either cycling or quiescent MEFs ( [20] and unpublished observations ) . Based on these observations , we sought to determine the consequences of knocking-down Rb in wild-type and p107E2F-1*2*/1*2* cells for p107 levels . Expression of shRNA molecules in wild-type and mutant MEFs resulted in a significant knock-down of Rb mRNA levels ( Figure 4B , left ) . As expected , decreased Rb levels led to increased p107 mRNA levels in wild-type quiescent MEFs . In contrast , low levels of Rb did not result in a further de-repression of p107 mRNA expression in p107E2F-1*2*/1*2* MEFs ( Figure 4B , right ) . Moreover , the degree of de-repression that occurred in wild-type cells upon Rb knockdown was similar to that seen through point mutations in the E2F binding sites ( Figure 4B , right ) . These experiments strongly suggested that Rb represses p107 through the two E2F binding sites in the p107 promoter in quiescent MEFs . In cycling mESCs ( Figure 2E–2F ) and in quiescent MEFs ( Figure 3B and 3C ) , p107 transcript levels correlate with p107 protein levels . In contrast , although quiescent wild-type MEFs with Rb knock-down and p107E2F-1*2*/1*2* MEFs either with or without Rb knock-down all have similar p107 mRNA levels ( Figure 4B ) , our initial immunoblot analysis suggested that p107 protein levels were higher in p107E2F-1*2*/1*2* MEFs with Rb knock-down compared to control wild-type cells with knock-down or knock-in mutant cells with wild-type Rb levels ( Figure 4A ) . Additional experiments confirmed and quantified these observations ( Figure 4C ) . In order to explore the potential post-transcriptional regulation of p107 levels in the absence of Rb , we treated wild-type MEFs in the presence and absence of Rb knockdown with cycloheximide , an inhibitor of translation . We found that p107 levels in the Rb knockdown cells remained more constant in the presence of cycloheximide than in wild-type cells treated with cycloheximide ( Figure 4D ) . These data suggest that loss of Rb function may control p107 levels post-transcriptionally , at least in certain contexts . Nevertheless , these observations also support a model in which the transcriptional control of p107 expression by Rb is largely through the two E2F binding sites in the p107 promoter . The increased levels of p107 protein found in quiescent p107E2F-1*2*/1*2* MEFs that are further increased in the presence of Rb knockdown led us to ask what functional effect these increased levels of p107 may have on the transcription of other E2F target genes . We performed RT-qPCR analysis on several E2F target genes , and found that , as expected , the expression of some of the genes examined–B-Myb , Cyclin A , and Cyclin E–was increased in wild-type MEFs in which Rb has been knocked down ( Figure 4E , left , shows the data for B-Myb , similar data for Cyclin A and Cyclin E are not shown ) ; the expression of these same genes was not increased in p107E2F-1*2*/1*2* MEFs in which Rb has been knocked down ( Figure 4E , left ) . Interestingly , Cdc6 expression was elevated in both wild-type and p107E2F-1*2*/1*2* MEFs upon knock-down of Rb , although a much larger increase in Cdc6 mRNA expression is observed in p107−/− MEFs with additional knockdown of Rb ( Figure 4E , right ) . Lastly , E2F1 expression was unchanged in wild-type and p107E2F-1*2*/1*2* MEFs with or without Rb expression , but was de-repressed in p107−/− MEFs with Rb knockdown ( Figure 4E , center ) . These results indicate that , in the absence of Rb , increased levels of p107 are able to repress the expression of some , but not all E2F target genes in quiescent MEFs . We next investigated the role of the E2F binding sites in the cell-cycle dependent activation of p107 transcription . We first found that asynchronously cycling p107E2F-1*2*/1*2* MEFs expressed ∼10% less p107 mRNA than did the wild-type cells ( Figure 5A ) , a decrease which was barely observable at the protein level ( Figure 5B ) . We speculated that because p107E2F-1*2*/1*2* MEFs display a 1 . 5-2-fold de-repression of p107 when they are in G0/G1 , this could mask a decrease in p107 expression at other phases of the cell cycle . To investigate this possibility , we expanded and stained control and mutant immortalized MEFs with Hoechst33342 , a DNA intercalating agent that enabled the cells to be FACS-sorted by DNA content ( Figure 5C ) . We found that wild-type G1 cells expressed 2–3 times more p107 than did cells in G0 , and this level increased up to 10-fold in S phase . On the other hand , while both p107E2F-1*/1* and the p107E2F-1*2*/1*2* cells showed some cell-cycle dependent induction of p107 , this induction was lower than the induction of p107 in wild-type cells during S-phase ( Figure 5D ) . To examine the regulation of p107 transcription via the two E2F binding sites during cell cycle re-entry from G0 , we synchronized primary wild-type and p107E2F-1*2*/1*2* MEFs through serum starvation , and then stimulated cell-cycle re-entry through the addition of serum in the medium . We found that p107E2F-1*2*/1*2* MEFs expressed higher levels of p107 than wild-type cells initially and that the mutant cells failed to increase p107 expression as much as the wild-type cells during cell cycle re-entry ( Figure 5E ) , supporting the fact that the two E2F binding sites in the p107 promoter are critical for p107 up-regulation during S phase progression . Despite these differences in p107 levels , MEFs of both genotypes re-entered the cell cycle with similar kinetics , as determined by measuring the mRNA levels of the highly cell cycle regulated gene Cdc6 ( Figure 5F ) and by the two-dimensional analysis of DNA content by PI staining and BrdU incorporation ( Figure 5G ) . One potential reason for the similarity of the cell cycle profiles in control and mutant cells is that p107 protein levels reached wild-type levels in p107E2F-1*2*/1*2* mutant MEFs progressing through S phase ( Figure 5H ) in these re-entry experiments . These observations corroborated our findings in cells with Rb knock-down that other mechanisms exist to increase p107 levels in some contexts , beyond the control of p107 transcriptional control by Rb/E2F complexes . Nevertheless , these experiments also confirmed that the transcriptional control of p107 expression in cells re-entering the cell cycle is under the control of the two E2F binding sites in its proximal promoter region . The lower levels of p107 RNA found in p107E2F-1*2*/1*2* cells at the G1/S transition and during DNA replication provided a system to test the role of p107 specifically during S phase progression . As discussed above , we found that the kinetics of Cdc6 induction were largely similar between primary wild-type and p107E2F-1*2*/1*2* mutant MEFs , with only a slight increase in Cdc6 maximal levels and a slight acceleration of Cdc6 induction in the mutant cells ( Figure 5F ) . Interestingly , however , when we repeated the same experiment with MEFs immortalized through knockdown of p19ARF , we found that p107E2F-1*2*/1*2* mutant MEFs induced Cdc6 mRNA levels more rapidly ( Figure 6A ) , suggesting that these mutant MEFs re-entered the cell cycle more quickly . BrdU/PI analysis further suggested that immortalized p107E2F-1*2*/1*2* mutant MEFs generally entered S phase earlier than control MEFs in this context ( Figure 6B ) . Together , these results suggest that p107 plays a critical role in controlling the kinetics of entry into S-phase in immortalized cells . These defects in cell cycle re-entry led us to investigate if altered p107 levels may change the length of the cell cycle in asynchronously proliferating primary cells . To investigate the importance of the regulation of p107 transcription for cell cycle control , we performed cell proliferation assays comparing wild-type , p107E2F-1*2*/1*2* , and p107−/− MEFs . BrdU/PI analysis of these asynchronously growing MEFs did not reveal any significant differences in the cell cycle profiles between the wild-type and p107E2F-1*2*/1*2* MEFs , although slightly more knock-in mutant cells were in G0/G1 and slightly fewer in G2/M than wild-type cells within this analysis; significantly more p107−/− MEFs were in S-phase than in either of the other genotypes ( Figure 6C ) . Despite this absence of difference in BrdU incorporation , we found that p107E2F-1*2*/1*2* MEFs proliferated similarly to p107−/− MEFs and more rapidly than wild-type cells ( Figure 6D ) . Thus , decreased levels of p107 mRNA specifically during S phase and in asynchronously cycling cells , even in primary MEFs , are sufficient to recapitulate the phenotype of p107−/− MEFs . Together , these observations indicate that p107 plays an important role at specific points during S phase in mammalian cells .
The classical view of the E2F family of transcription factors is that they are necessary to drive cell-cycle progression by binding to the promoters of and activating genes necessary for S phase , including those needed for nucleotide synthesis and DNA replication ( reviewed in [3] , [41] ) . However , it has also long been observed that the list of E2F target genes also includes negative regulators of the cell cycle , including p16 , Rb , and p107 . Traditional knock-out studies obscure the careful balance of these positive and negative feedback loops and the relative importance of each individual target . The knock-in of potential binding sites for specific transcription factors has not been extensively used [42] , [43] but proves here to be an extremely informative approach to dissect the functional role of specific nodes in complex regulatory networks . Our experiments demonstrate that E2F family members and Rb control p107 transcription largely through two tandem E2F binding sites in the proximal promoter of the p107 gene . Our data also identify functional differences between the two sites and the E2F activity bound to these sites in different contexts ( see model in Figure 6E ) . E2F transcription factors make up a diverse family whose members can all recognize the same consensus sequence . While some E2F transcription factors activate target gene expression , others repress transcription , either dependent upon or independent from their association with Rb family members . Many E2F target genes , like p107 , reveal an even more complex promoter structure that includes two , or sometimes more , E2F consensus sites , and each site could have distinct functions in the control of the target gene [44]–[46] . Why certain promoters have several E2F binding sites and what dictates if a site serves to repress or activate transcription is not understood . In the specific example of p107 , our data show that the distal E2F binding site is favored by both E2F3 and E2F4 in mESCs , indicating that activating and repressor E2Fs may act through the same binding site in vivo . Our data also indicate that binding to the consensus site is context-dependent: while E2F4 binding in mESCs requires the presence of at least one of the two binding sites , an E2F4 binding activity is still retained in MEFs with mutations in both E2F binding sites . This observation suggests the existence of secondary E2F binding sites and/or the presence of co-factors , including Rb family members , which may help tether E2Fs to a promoter region ( Figure 6E ) . This residual binding activity may explain why p107 levels are still increasing during cell cycle re-entry in p107E2F-1*2*/1*2* mutant MEFs . In the future , the availability of cells with knock-in mutations in individual binding sites in the p107 promoter will provide novel tools to dissect how transcription factors and chromatin-remodeling enzymes interact with Rb/E2F complexes to regulate the expression of E2F target genes in different cellular contexts . Similar to the residual binding of the E2F transcription factors to the mutant p107 promoter , we observed above background levels of Rb , p130 , and p107 bound to the p107 promoter in p107E2F-1*2*/1*2* mutant MEFs . Rb family members have been well described to bind to many other protein binding partners , including ATF-1 [31] and Sp1 [30] . The mouse p107 promoter contains consensus sequences for both ATF and Sp1 ( Figure 1B ) , and we cannot exclude that these sites , or others , may mediate E2F-independent binding of all three Rb-family members to the p107 promoter . Interestingly , we also found that the p107E2F-1*2*/1*2* mutant MEFs demonstrated reduced binding of p130 to the B-Myb promoter during quiescence . Expression analysis of B-Myb did not reveal substantial de-repression in these mutant MEFs , and increased levels of p107 were able to repress B-Myb expression in the absence of Rb . Together these results are consistent with the model that all three Rb family members are able to regulate B-Myb , and that these complexes may shift in response to altered Rb family levels . Furthermore , the trend towards increased Rb binding to the Mcm3 promoter in the p107E2F-1*2*/1*2* mutant MEFs ( Figure 3E ) was observed at other promoters ( B-Myb and Cyclin A , data not shown ) , which may further suggest that p130 is displaced by Rb in these cells . The tandem E2F consensus sites in the human p107 promoter had previously been demonstrated to have differential functions over control of p107 [27] . In contrast , our experiments show that the distal site is most important for both activation and repression of the p107 promoter in mouse cells . This discrepancy could be explained by the presence of a single point mutation in the mouse promoter , which makes the proximal site a less perfect E2F consensus site . We found that correcting the proximal site in the mouse p107 promoter ( TTTGTCGC → TTTGGCGC ) did increase the activity of this promoter in luciferase assays approximately 3 fold relatively to the parental construct ( unpublished data ) . However , in the absence of an intact distal site , the activity of even the “corrected” construct was substantially lower than that of a wild-type construct , further emphasizing the relative importance of this site in the mouse p107 promoter . These results also suggest that the human promoter , with two perfect E2F consensus sites , should generally be more responsive to E2F transcription factors than the mouse promoter . Therefore , this discrepancy alone is unlikely to explain why p107 is up-regulated and able to compensate for loss of Rb function in the mouse retina but not the human retina [47] or why mice and patients with an Rb mutation develop a distinct tumor spectrum . While the E2F sites are highly conserved across mammalian species , the rest of the p107 promoter is not , and these other evolutionary changes may further impact promoter regulation by E2F and other factors . Interestingly , in humans , single nucleotide polymorphisms ( SNP ) of unknown frequency have been identified in the p107 promoter , including one that deletes one of the 4 Ts in the distal E2F binding site; while this may not affect E2F binding ( only 3 Ts are required in the consensus sequence ) , this polymorphism may potentially alter the physical orientation of the two sites relative to each other and influence p107 transcription . Future experiments will continue to dissect the mechanisms regulating p107 transcription , including the interactions between E2F family members , Rb family members , and other transcription factors that bind to the p107 promoter . These interactions may also help to explain the tissue-specificity of expression of p107 in vivo [20] , [48] . While several binding partners for p107 and its expression profile in various cell types are well known , the unique cellular functions of p107 are still poorly understood [2] , [49] . Overexpression of p107 can arrest some cell types in G1 [50] , but loss of p107 function often results in no visible phenotypes , probably because of functional compensation by Rb and p130 [13] , [17] . Interestingly , loss of p107 in neural progenitors results in the activation of Notch signaling and increased proliferation [51] , [52] . p107−/− mice on a BALB/cJ background also display a myeloid hyperplasia and mutant MEFs derived from these mice demonstrate accelerated proliferation [53] . Furthermore , an insertional mutagenesis screen for tumor suppressor genes identified p107 as a tumor suppressor in B-cell lymphoma [54] . However , the mechanisms underlying these loss-of-function phenotypes are still unclear . Some studies have suggested that p107 may have a particular function during the progression from late G1 to S phase [55] , [56] , when p107 levels are highest . We show here that expression of abnormally low levels of p107 during S phase results in increased cell numbers in a proliferation assay . Importantly , while the p107 protein is still expressed and p107 levels found during S phase in p107E2F-1*2*/1*2* cells are similar to those found in wild-type cells in G0/G1 , and while BrdU incorporation failed to detect a significant difference between knock-in mutant cells and wild-type cells , the consequences for the cell cycle are as strong as in p107 null cells . This observation indicates that p107 levels are probably critical during very specific stages of the cell cycle , including during the DNA replication phase , and this role for p107 during S phase should be the focus of future studies to elucidate the cellular functions of this cell cycle regulator . Unexpectedly , in immortalized MEFs in which p19ARF has been knocked-down , the importance of p107 expression during S-phase is even more evident , as the kinetics of cell cycle re-entry are altered in the p107E2F-1*2*/1*2* cells compared to wild-type cells , whereas these kinetics are unchanged in primary MEFs . This difference between the function of p107 during S phase in immortal and primary MEFs is potentially related to an additional function of p19ARF during S phase [57] . This function could stem from a network of interactions between p107 , p19ARF , E2F1 and c-Myc during S phase [58]–[63] and these findings could reveal cooperation between appropriate expression of p107 and p19ARF in the control of S phase . In the absence of Rb , de-repression and/or activation of p107 transcription is thought to result in higher levels of p107 that may then suppress the functions of activating E2F family members [64] , [65] . It is interesting to note that a similar mechanism has evolved independently in plants [66] and that this type of negative feedback loop also exists in the control of the cell cycle in yeast [67] , strongly suggesting that this genetic circuitry is a universal component of the regulatory networks ensuring proper cell cycle progression . Here , we investigated the mechanisms by which this feedback loop is activated in mammalian cells using a mouse genetics approach . While it has long been hypothesized that direct regulation of the p107 promoter by Rb is the mechanism by which compensatory upregulation of p107 occurs in the absence of Rb , our mutant cells enabled us to distinguish effects of Rb loss on transcriptional and post-transcriptional compensatory expression of p107 . Recently it has been shown that the p107 protein may be more stable in non-cycling human hepatocellular carcinoma cell lines when Rb is absent than when Rb is present [68] . Similarly , we found that in quiescent MEFs , the p107 protein becomes more stable when Rb is knocked-down , suggesting that transcriptional regulation alone may not fully explain compensatory levels of the p107 protein . Clearly , this increase in p107 protein levels in cycling cells or Rb mutant cells independent of p107 transcription could serve as a feedback mechanism to limit the effects of decreased Rb levels . The relative contribution of transcriptional and post-transcriptional mechanisms of p107 up-regulation in different contexts will be the focus of future studies . We hypothesized that disrupting the transcriptional feedback loop between Rb/E2F and p107 would prevent compensation in the absence of Rb , such that the low p107 levels observed in p107E2F-1*2*/1*2* MEFs would cooperate with decreased Rb expression , potentially recapitulating some of the phenotypes observed in Rb;p107 double knock-out MEFs . Additional cell cycle assays comparing primary and immortalized knock-in mutant MEFs , both with and without Rb knock-down , failed to reveal conditions in which the knock-in mutation would cooperate with loss of Rb in allowing cells to grow in conditions that were permissive to growth of Rb;p107 double knock-out MEFs ( data not shown ) . These results suggest that , in most contexts , the lower levels of p107 observed specifically during S-phase are insufficient to overcome the higher levels of p107 observed at other stages of the cell cycle , at least in terms of reducing the ability of p107 to compensate in the absence of Rb . Instead , we found that the increased levels of p107 observed in quiescent MEFs were able to repress expression of some genes in the absence of Rb even better than the increased levels of p107 found in wild-type MEFs in the absence of Rb . Interestingly , the ability of increased levels of p107 to compensate for absence of Rb in repressing E2F target genes during quiescence depends on the particular gene . Increased levels of p107 protein produced by transcriptional de-repression and increased protein stability resulted in repressed levels of B-Myb , Cyclin A , and Cyclin E but had no further effect on Cdc6 ( Figure 4E and data not shown ) . These results are consistent with several studies suggesting that the Rb family members regulate both distinct and overlapping target genes [23] , [69] , [70] . These observations also suggest that further increasing p107 levels in Rb mutant cells through a variety of mechanisms may enhance the compensatory abilities of this Rb family member . A remaining question is why humans develop retinoblastoma upon loss of Rb while mice do not . It has recently been demonstrated that mouse retinal progenitors deficient for Rb display increased levels of p107 mRNA whereas human retinal progenitors do not increase the amount of p107 expressed when Rb is knocked down in culture [47] . This observation , and the fact that Rb/p107 double mutant mice develop retinoblastoma [71] supports the idea that p107 levels are important for its tumor suppressor activity in Rb-deficient retinal cells . However , no system has been devised to test whether the increased expression of p107 observed in Rb mutant cells directly contributes to the ability of p107 to compensate for the loss of Rb , or if this increased transcription is merely coincidental to a constitutive overlapping function shared by Rb and p107 . In other words , it is possible that basal levels of p107 in Rb mutant cells would be sufficient to suppress retinoblastoma development . A similar question can be asked in other cell types in which p107 loss of function by knock-out enhances tumor development in Rb mutant cells and during development [13] , [15] , [72] . Our results indicate the converse , i . e . that even higher levels of p107 may more completely compensate in the absence of Rb , in terms of target gene expression . The generation of mESCs and MEFs in which the E2F binding sites in the p107 promoter were singly mutated were intended to discretely separate the ability of p107 to be activated in the absence of Rb from other sources of transcriptional regulation . However , these genetic studies clearly demonstrated that in MEFs , a single E2F site is critical for both the repression of p107 in non-cycling cells and the activation of p107 in cycling cells , making the separation of these two activities impossible . In future experiments , mice carrying point mutations in the E2F binding sites in the p107 promoter or mice carrying fragments of the human p107 promoter may help explore the necessity of p107 up-regulation in the prevention of retinoblastoma in mice by distinguishing between overlapping and compensatory expression patterns of p107 .
Mice were maintained according to practices prescribed by the NIH at Stanford's Research Animal Facility accredited by the AAALAC . Recombineering in bacteria [73] was used to generate the p107E2F-1* , p107E2F-2* , and p107E2F-1*2* BACs . Briefly , BAC clone RP23-163J20 ordered from BACPAC ( http://bacpac . chori . org/ ) was transformed into the EL250 strain and the heat-inducible recombinase present in this strain enabled the insertion of a Neomycin resistance cassette into intron 1 of p107 by homologous recombination . Approximately 500 bp of the wild-type promoter sequence was cloned into pBluescript . Mutation of the E2F sites was performed using blunt end primers carrying the wild-type sequence or the point mutations . The 500 bp fragments were then targeted to the p107 BAC with the conditional Neomycin resistance cassette . To generate the targeting vectors , we excised approximately 4 kb upstream of the E2F binding sites and 4 kb downstream of the Neomycin resistance cassette from the BAC DNA into a targeting vector backbone carrying a DTA cassette . mESCs and MEFs were cultured as described previously [74] . For expression analysis in asynchronous MEFs , 3×105 cells were plated per 6 cm culture dish . For BrdU/PI analysis in asynchronous MEFs , 2 . 5×105 cells were plated per 6 well . Extracts were collected 48 hours later . For quiescent cell analysis were plated at higher density: 8×105 per 6 cm dish , 3×105 per 6 well , or 1 . 5×105 per 12 well . The following day cells were washed twice with PBS and then cultured in DMEM supplemented with 0 . 1% serum for at least 72 hours . For extracts from synchronized MEFs , cells were rendered quiescent and cultures were re-stimulated with DMEM supplemented with 20% serum as above after at least 72 hours in low serum conditions . Extracts were collected at various time points after stimulation . For proliferation assays , MEFs were plated at either 2 . 5×104 or 5×104 cells per 12 well on day 0 . Cells were counted in duplicate and were given fresh media every other day . For cycloheximide experiments , wild-type MEFs were rendered quiescent for 3 days as above . At time 0 , cycloheximide ( Calbiochem , 25 mg/ml in methanol ) or methanol was added to cells at a concentration of 30 µg of cycloheximide per ml of DMEM with 0 . 1% serum . To construct the plasmid reporters , primers were designed to amplify the p107 promoter from the targeted BAC clones ( wild-type and mutants ) to enable cloning directly into the multiple cloning site of PGL3-basic ( Promega ) . The reverse primer was positioned to include all sequence upstream of the translation start site . 1 . 4×104 mESCs ( V6 . 5 ) and 1 . 25×104 MEFs were plated in wells of 48-well plates and transfected one day later . For mESCs , luciferase activity was read two days after transfection following the manufacturer's instructions ( Promega ) . For quiescent MEFs , luciferase activity was read 24 hours after the withdrawal of serum . In all luciferase assays , 500 ng of each p107 construct was co-transfected with 125 ng of a Renilla luciferase vector . For exogenous E2F3 experiments , 100 ng of DP1 ( a gift from the Dyson lab ) was co-transfected with either 100 ng of pCDNA empty vector or 100 ng of CMV-E2F3 . Transfections were carried out using the Fugene6 Reagent ( Roche ) . Rb knock-down was achieved using the pSicoR lentivirus [75] , as described [18] . The sequence in the Rb cDNA that is targeted by the shRNA molecules is 5′-TGAGAGCAAGGATGTCTCA-3′ . p19ARF and p130 knock-down was achieved using the pSiripp retrovirus [18] , as above . The sequence in the mouse p19 cDNA that is targeted by the shRNA molecules is: 5′-CACCGGAATCCTGGACCAG-3′ . The sequences in the mouse p130 cDNA that is targeted by the shRNA molecules are: 5′-TCACTCTGCTCTGTTACGT-3′ and 5′- GATGTGGCGAATGACCGAG-3′ . DTA- p107E2F-1* and DTA- p107E2F-1*2* targeting vectors were electroporated into V26 . 2 mESCs ( C57BL/6 ) and DTA- p107E2F-2* was electroporated into J1 mESCs ( 129Sv/J ) . Genomic DNA from targeted clones was screened by 5′ and 3′ Southern analysis , details of which are available upon request . The E2F binding sites of clones targeted with the neomycin resistance cassette were sequenced using the primers described for p107 promoter ChIP below . Clones that were appropriately targeted by the neomycin resistance cassette but that retained wild-type binding sites were used as wild-type controls . Heterozygous mESCs for each construct were infected with Ad-Cre and plated for single colonies . Colonies were picked and screened for neomycin sensitivity . E2F binding sites were again sequenced to rule out loss a larger loss of the chromosome . Clones retaining heterozygous sequence for the binding site ( s ) , as well as the loxP site in intron 1 were re-targeted and rescreened by Southern analysis and sequencing . p107E2F-1*/1* and p107E2F-1*2*/1*2* homozygous mutant mESCs and controls ( described above ) were injected into wild-type blastocysts by the Stanford Transgenic Research Facility . MEFs were derived 11 days post-implantation . Pure populations were obtained through selection with 600 µg/ml of Geneticin ( Invitrogen ) . MEFs were generated from one p107E2F-1*/1* clone , two independently targeted p107E2F-1*2*/1*2* clones , as well as two independent control clones . Where indicated , immortalized MEFs were generated through retroviral infection with a vector that expresses shRNA molecules against p19ARF [18] . RNA was extracted from frozen cell pellets with TRIzol ( Invitrogen ) . TaqMan or SYBR green quantitative PCR was performed as described previously [20] , [76] . Rb , p107 , TBP , and CDC6 primers were described previously [76] , [77] . Other primer sequences are as follows: for B-Myb , forward primer , 5′-TTA AAT GGA CCC ACG AGG AG-3′ and reverse primer , 5′-TTC CAG TCT TGC TGT CCA AA-3′; for E2F1 , forward primer , 5′-TGC CAA GAA GTC CAA GAA TCA-3′ and reverse primer , 5′-CTT CAA GCC GCT TAC CAA TC-3′ . All relative expression analyses were calculated relative to TBP ( TATA binding protein ) . Immunoblots were detected as described previously [74] . Quantitative immunoblot analysis was performed using an Odyssey Infrared Imager from LI-COR Biosciences . Antibodies used were as follows: rabbit anti-p107 ( sc-318 , Santa Cruz Biotechnology ) , mouse anti-Rb [78] , mouse anti-p130 ( BD 610621 ) , goat anti-MCM6 ( sc-9843 ) , and mouse anti-PCNA ( sc-56 ) . Loading was verified with antibodies against mouse anti-alpha-Tubulin ( Sigma T9026 ) , by anti-β-Actin ( Sigma A5441 ) , or by staining of total protein with Ponceau . Quantitative chromatin immunoprecipitation ( ChIP ) for E2F3 , E2F4 , p107 , and p130 was performed as described previously [76] . Rb ChIP was performed as described in the Farnham laboratory protocol ( http://genomics . ucdavis . edu/farnham/pdf/FarnhamLabChIP%20Protocol . pdf ) with a few modifications . Chromatin was sonicated using a Virsonic probe sonicator at setting 2 at 20% output power for 8 cycles of 15 seconds . The chromatin was pre-cleared before being diluted and bound by 4 µg of the primary antibody overnight at 4°C . Each ChIP was then incubated with 8 µg rabbit anti-mouse IgG ( MP Biomedicals #55436 ) as a secondary for 1 hour . The nucleoprotein complexes were pulled down by Pansorbin cells ( Calbiochem , Cat# 507862 ) . The DNA was digested with Proteinase K and RNaseA and then purified by a Qiagen QIAquick PCR Purification Kit . Additional details are available upon request . Antibodies used for immunoprecipitations were as follows: E2F3 ( sc-878X ) , E2F4 ( sc-1082X ) , p107 ( sc-318X ) , p130 ( sc-317X ) , p16 ( sc-467 ) , Rb [78] , and normal mouse IgG ( sc-2025 ) . p107 , B-Myb , and Mcm3 promoter binding were assessed through quantitative PCR using SybrGreenER Mastermix ( Invitrogen ) . p107 forward , 5′-GGT CCA TCT TCT TAT CCC ATT CCG-3′; p107 reverse , 5′-CTT CGG GGT TTT CTT TTC CCT C-3′; B-Myb forward , 5′-CTC GTG TCT TGT ACG CTT CGC C-3′; B-Myb reverse , 5′-CAC GTT CCC AGG AAC TGC AGC T-3′; Mcm3 forward , 5′- AGC CAA TCA TAA CGC GTC TC-3′; Mcm3 reverse , 5′-CAG CTC CAC ATC ATC CAG CA-3′; actin forward: 5′-GCT TCT TTG CAG CTC CTT CGT TG-3′; actin reverse , 5′-TTT GCA CAT GCC GGA GCC GTT GT-3′ . For primary and immortal MEF synchronized cell cycle analysis , cells were pulsed with BrdU for 2 hours prior to trypsinization . For primary MEF asynchronous cell cycle analysis , cells were pulsed with BrdU for 4 hours . BrdU and propidium iodide staining and analysis was performed as described previously [79] , and analyzed on a BD FACSCalibur instrument . Data was analyzed using FlowJo software ( Tree Star ) . Subconfluent , immortalized MEFs were trypsinized and resuspended in DMEM +10% serum at 10×106 cells/ml . Hoechst33342 was added at 30 mg/ml and incubated at 37°C for 1 hour in the dark . Cells were spun and resuspended in 1 ml of DMEM with fresh Hoechst , strained through a 40 mm cell strainer and then sorted at the Stanford Shared FACS Facility . Statistical significance was assayed by unpaired Student's t-test , except where otherwise indicated . *: p-value<0 . 05; **: p-value<0 . 005; ***: p-value<0 . 0005; ns: not significant . Mean and standard error are shown .
|
The retinoblastoma tumor suppressor Rb belongs to a family of cell cycle inhibitors along with the related proteins p107 and p130 . Strong evidence indicates that the three family members have both specific and overlapping functions and expression patterns in mammalian cells , including in cancer cells . However , the molecular mechanisms underlying the functional differences and similarities among Rb , p107 , and p130 are still poorly understood . One proposed mechanism of compensation is a negative feedback loop involving increased p107 transcription in Rb-deficient cells . To dissect the mechanisms controlling p107 expression in both wild-type and Rb-deficient cells , we have engineered inactivating point mutations into the E2F binding sites in the endogenous p107 promoter using gene targeting in mouse embryonic stem cells . Gene expression and DNA binding assays revealed that these two sites are essential for the control of p107 transcription in wild-type and Rb mutant cells , and cell cycle assays showed their importance for normal functions of p107 . These experiments identify a key node in cell cycle regulatory networks .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology/gene",
"expression"
] |
2010
|
Tandem E2F Binding Sites in the Promoter of the p107 Cell Cycle Regulator Control p107 Expression and Its Cellular Functions
|
The incidence of human brucellosis in Kyrgyzstan has been increasing in the last years and was identified as a priority disease needing most urgent control measures in the livestock population . The latest species identification of Brucella isolates in Kyrgyzstan was carried out in the 1960s and investigated the circulation of Brucella abortus , B . melitensis , B . ovis , and B . suis . However , supporting data and documentation of that experience are lacking . Therefore , typing of Brucella spp . and identification of the most important host species are necessary for the understanding of the main transmission routes and to adopt an effective brucellosis control policy in Kyrgyzstan . Overall , 17 B . melitensis strains from aborted fetuses of sheep and cattle isolated in the province of Naryn were studied . All strains were susceptible to trimethoprim-sulfamethoxazole , gentamicin , rifampin , ofloxacin , streptomycin , doxycycline , and ciprofloxacin . Multilocus variable number tandem repeat analysis showed low genetic diversity . Kyrgyz strains seem to be genetically associated with the Eastern Mediterranean group of the Brucella global phylogeny . We identified and confirmed transmission of B . melitensis to cattle and a close genetic relationship between B . melitensis strains isolated from sheep sharing the same pasture .
Agriculture is a key component of Kyrgyzstan's economy and livestock play a major role in the daily lives of the population . Sixty four percents of the population live in rural areas and rely on agriculture for their livelihoods . Up to 76% of the rural population of the country is classified as poor . [1] . Since independence in 1991 , veterinary support ceased then largely and the incidence of diseases transmitted from animals to humans ( zoonoses ) has increased dramatically in many regions in Kyrgyzstan . Brucellosis , anthrax , rabies and echinococcosis are public health concerns and constitute a serious risk to the human and the livestock health . The incidence of brucellosis has increased steadily and Kyrgyzstan has now one of the highest human brucellosis incidences worldwide ( annual incidence: 77 . 5 new cases per 100 , 000 people in 2007 ) [2] . Currently , Kyrgyz communities are concerned about the effective reduction of the brucellosis burden in people and livestock . The latest species identification of Brucella spp . cultures in Kyrgyzstan was done in the 1960ies . Both B . abortus and B . melitensis were isolated from cattle . B . melitensis infections in cattle were thought to be a spill-over from sheep . Smirnov and Tretyakova noted that abortions in cows after immunization with S19 were most often seen in herds that were infected with Brucella spp . B . melitensis was isolated from vaccinated and non-vaccinated sheep [3] , [4] . the authors concluded that B . melitentsis steadily adapted to sheep [5] . At present , the circulating genotypes of Brucella spp . are not known . This is true for virtually all Central Asian regions . Bacteriological confirmation of Brucella spp . -induced abortions is almost absent , because owners do not report suspected abortions to the veterinary services . Here we report recently isolated Brucella spp . strains from sheep and cattle , which were collected in addition to a representative national study on brucellosis sero-prevalence in humans and livestock [6] . and to cost of disease studies in Kyrgyzstan ( data not shown ) . The results contribute to the understanding of the main transmission routes and effectively inform brucellosis control policy in Kyrgyzstan .
The study was performed in the province of Naryn oblast , which has the highest human brucellosis incidence in Kyrgyzstan and most of its population has an income through selling of animals and animal products . First primary isolations of Brucella strains from aborted fetuses were done at the veterinary laboratory of the Naryn province in November 2008 . All public and private veterinarians were informed about the ongoing project on brucellosis . Farmers were informed beforehand and asked to report abortions through local village veterinarians; leaflets with information were distributed through veterinarians and announcement was published in the province newspaper . Abortions from sheep and cattle were collected during the lambing seasons of 2009 and 2010 . In general , the lambing season starts in January and continues until March and April and thus first abortions can occur in late November/December . Veterinarians brought the collected specimens – aborted sheep and cattle fetuses - dissected on site - to the Naryn laboratory . Stomach content was collected in tubes and liver , spleen , kidney , lung , heart and other tissues were collected in plastic bags . Veterinarians collected accompanying basic information on the animals and farms such as geographic position and keeping of other than affected animals . Two weeks after the abortion , a visit to the affected farm allowed blood sampling of farm animals for serology ( data not shown ) and an interview with the livestock holders to obtain epidemiological data with a questionnaire . Total number of fetuses collected by the veterinarians was 125 from whole district and positive isolates by the Urease and Oxidase were selected for further study . Primary cultures were done at the Naryn zonal Center for Veterinary Diagnostic and specimens were frozen . When culture was negative , frozen specimens were re-cultured at the Republican Center for Veterinary Diagnostic in Bishkek . Stomach content and organs of the aborted fetuses were cultured onto Brucella selective agar ( bioMérieux , Switzerland ) and onto own produced Brucella selective agar ( with agar , horse serum and antibiotics from Oxoid , Switzerland ) . Strains were cultured on Brucella agar at 37°C with 10% CO2 for 2 days [7] . For the investigation of the sensitivity of the cultures to , phenotypic antibiotic resistance to 7 different drugs was assessed by the standard E-tests ( bioMérieux , Switzerland ) on Mueller-Hinton blood agar ( MHS2 , bioMérieux SA , France ) and their minimum inhibitory concentrations ( MIC ) were determined additionally . The following antibiotics were tested: trimethoprim-sulfamethoxazole ( SXT ) ( 1 . 25+23 . 75 µg ) , gentamicin ( GM ) ( 10 µg ) , rifampicin ( RA ) ( 30 µg ) , ofloxacin ( OFX ) ( 1 µg ) , streptomycin ( S ) , ( 15 µg ) , doxycyclin ( D ) , ( 30 µg ) , and ciprofloxacin ( CIP ) . ( 5 µg ) , Inducible clindamycin resistance test ( “D-zone” test ) was also carried out for all isolates . Results were interpreted according to the Clinical and Laboratory Standards Institute ( CLSI ) guidelines; for the purpose of this study , intermediate results were classified as resistant . DNA was extracted from one loopful of bacterial cells grown for 48 h on chocolate agar , and single colonies were isolated by using the tissue protocol of the QIAamp DNA minikit ( Qiagen , Germany ) . DNA concentrations were measured by UV spectrophotometry ( Shimadzu , Japan ) . Multiple Loci Variable Number of Tandem Repeat Analysis ( 16 locus MLVA ) typing was performed with the 17 isolates according to the protocol initially proposed by Le Flèche et al . [8] . and modified by Al Dahouk [9] . to include 1 additional locus , bruce19 . The protocols are available online on the MLVA-NET for Brucella ( http://mlva . u-psud . fr ) . In brief , the assay comprised the typing of eight mini-satellites of the so-called panel 1 ( bruce06 , bruce08 , bruce11 , bruce12 , bruce42 , bruce43 , bruce45 , and bruce55 ) , three micro-satellites of the panel 2A ( bruce18 , bruce19 , and bruce21 ) , and five micro-satellites of the panel 2B ( bruce04 , bruce07 , bruce09 , bruce16 , and bruce30 ) . The 16 published VNTR loci were PCR-amplified in parallel and the numbers of tandem repeats determined after electrophoresis on agarose gel . DNA extracts of B . melitensis 16MT and vaccine strain Rev1 were used as positive controls . The obtained MLVA patterns of each sample were then matched with an online database ( http://minisatellites . u-psud . fr/MLVAnet/querypub1 . php ) for identification . A small amount of a colony of each pure culture was transferred to a FlexiMass target well using a disposable loop and overlaid with 1 . 0 µl alpha-cyano matrix solution ( CHCA; 40 mg alpha-cyano in 33% acetonitrile , 33% ethanol , 33% ddH2O and 1% trifluoroacetic acid ) . The spotted solution was air-dehydrated during 1–2 min at room temperature and analysed with MALDI-TOF MS Axima Confidence spectrometer ( Shimadzu-Biotech Corp . , Kyoto , Japan ) . The reference strain Escherichia coli K12 ( GM48 genotype ) was used as a standard for calibration and as reference measurement for quality control . Mass spectrometry ( MS ) analyses were performed in positive linear mode in the range of 2 , 000–20 , 000 mass-to-charge ratio ( m/z ) with delayed , positive ion extraction ( delay time: 104 ns with a scale factor of 800 ) and an acceleration voltage of 20 kV . For each sample , 2×50 averaged profile spectra were stored and used for analysis . All spectra were processed by the MALDI MS Launchpad 2 . 8 software ( Shimadzu Biotech ) with baseline correction , peak filtering and smoothing . A minimum of 20 laser shots per sample were used to generate each ion spectrum . For each bacterial sample , 50 protein mass fingerprints were averaged and processed . Spectra were analyzed using SARAMIS ( Spectral Archive and Microbial Identification System , AnagnosTec GmbH ) at default settings . Cladistic analysis were based on the peak patterns of all analyzed strains submitted to single-link clustering analysis using SARAMIS with an error of 0 . 08% and a m/z range of 2 , 000 to 20 , 000 Daltons . Allelic diversity was calculated using the formula below , where xi is therelative frequency of the ith allele at the locus , n the number of isolates in the sample and ( n/ ( n-1 ) is a correction for bias in small samples [10] . VNTR data was the basis for the phylogenetic analysis using SAS ( Statistical Analysis Systems Inc . Cary , USA ) proc cluster using the unweighted pair-group method with arithmetic averages , ( UPGMA ) . For the assessment of the phylogenetic place of the Kyrgyz isolates , strains were selected from the online database by Maquart [11] , [12] . ( 1471-2180-9-145-S1 . xls; http://www . biomedcentral . com/1471-2180/9/145 ) . Isolates were selected to reflect the diversity of geographical origin and the different biovars . Phylogenetic trees were drawn using SAS proc tree . The Ethics Committee of the University and the state of Basel has approved this study without restrictions in the meeting of January 11 , 2007 ( Reference number 02/07 ) . The project conforms with the ethics requirements on animal testing ( Published in Schweiz . Ärztezeitung , 2006 , Band 87 , S . 832–837 ) by the Swiss Academy of Medical Sciences and the Swiss Academy of Natural Sciences . Animal owners were asked for consent to test aborted fetuses of their livestock for brucellosis .
Livestock systems and management of herds from which B . melitensis were isolated varied between owners . Livestock owners kept cattle and small ruminants together and practiced seasonal transhumance to high-altitude pastures . They sometimes also kept entrusted animals from several owners and traded actively animals . During the lambing seasons 2009 and 2010 in Naryn , 125 aborted fetuses ( 112 from sheep and 13 from cattle ) were collected in the 4 villages and the city of Naryn ( Figure 1 ) . The rate of isolation for sheep was 8 . 9% and for cattle is 15% but the difference is not statistically significant . Urease and oxidase positive cultures were selected and 17 out of 23 isolates were confirmed B . melitensis by MALDI-ToF MS and MLVA-16 ( Figure 2 ) . The dendrogram is based on the MLVA-16 genotyping assay showing the relationship of the 15 sheep and two cattle isolates of Brucella melitensis . For each locus showing variability , the number of tandem repeats is presented . Additional information is provided on the type of sample , the local strain designation , the serial number of the animal owner and the name of the village in Naryn oblast Numbers in brackets indicate repeated isolates from the same animal . Isolates not indicated as primary were frozen prior to cultivation . Of the 17 isolates , 15 were isolated from sheep and two from cattle . All strains were susceptible to the tested antibiotics . The allelic diversity of VNTR ( h ) was low , with only three loci showing variation in the numbers of repeats . For locus 4 it was 0 . 6 , for locus 16 0 . 16 and 0 . 49 for locus 30 ( Table 1 ) . All other loci did not show any variation . Eight out of 17 strains grouped into 6 different clusters . However , it has to be noted that more than one isolate was obtained from four animals . Isolates of cluster 2 were found in herds of two different owners in sheep and cattle . With regard to the geographical location , the Kyrgyz isolates are closest to strains from Kazakhstan , Israel and Iraq which are all biovar3 ( Figure 3 ) [11] .
B . melitensis isolates from Kyrgyzstan appear to be close to the so-called Eastern Mediterranean group ( Figure 3 ) [11] , but a more detailed analysis and more isolates are required to conclusively determine the position of Kyrgyz Brucella in the global phylogeny . All B . melitensis isolates from Naryn Oblast were closely related according to VNTR patterns . Isolates belonging to second cluster from the top ( Strain No . 3–6 ) ( Figure 2 ) were found in the herd of two owners of sheep and cattle , indicating that strains circulated between farms and were transmitted between small ruminants and likely to cattle during communal grazing . These two owners live 45–50 km apart . The owner of the cattle lives in the city of Naryn and his cattle graze on a summer pasture with several other animals suggesting rural/urban spill over through sharing of common pasture . The 8 isolates ( sixth cluster from the Top in Figure 2 ) from sheep stem from Jer-Kochku and Lakhol , two villages 10 km apart . The animals from which they originated use the same pasture for grazing , except for the two strains from Kulanak which is located at more than 80 km from Jer-Kochku and Lakhol . This may indicate a contact relationship between Kulanak , Jer-kochku and Lakhol ( Figure 1 ) . Owner 1 had sheep in which three B . melitensis genotypes are present . A better understanding of the contact network of each animal owner could possibly further explain genetic diversity . Multiple strains were isolated from liver , spleen and heart in three animals ( Figure 2 ) . Isolates from different organs of the same animal had always the same VNTR pattern , hinting to a likely monoinfection . The isolation of B . melitensis in sheep and cattle is the first recent confirmation by culture since the 1960ies in Kyrgyzstan . It was expected because brucellosis in cattle was not a problem a decade ago and increasing sero-prevalences and brucellosis abortions in cattle were observed during the past years . It was therefore speculated that cattle may be a spill-over host of B . melitensis from small ruminants . More isolates are needed to further consolidate this finding . If confirmed , this may have policy implications for ongoing pilot mass livestock vaccination campaigns , considering cattle vaccination . We found no antibiotic resistance and therefore the standard regimen used in Kyrgyzstan ( i . e . , Gentamicin plus Doxicycline ) is likely to be adequate for humans . However , human isolates should be tested as well . The use of antibiotics in livestock is clearly not recommended . This study confirms ongoing transmission of B . melitensis in sheep and likely to cattle in the province of Naryn in Kyrgyzstan . The high genetic homogeneity indicates rather clonal expansion and ongoing transmission , confirming serological observations [6] . The role of cattle in the transmission of B . melitensis transmission should be examined more closely . Further studies on human brucellosis strain characteristics are needed to confirm sheep as the suspected principal source of livestock to human transmission [6] . For this purpose more discriminatory methods than VNTR may be needed . Further collection of isolates from aborted fetuses including information on contact networks are needed to monitor the success of the ongoing mass vaccination campaign and to allow calibrating VNTR dynamics in space and time .
We conclude that B . melitensis is endemic in Naryn oblast and sheep are apparently the main host . B . melitensis is also transmitted to cattle . In the study period we observed no abortions in goats and hence consider them less important for brucellosis transmission in Naryn oblast . Our findings confirm an earlier serological study , which related human brucellosis sero-prevalence to sheep but not to goat and cattle [6] .
|
Brucellosis is a bacterial disease causing abortion in cattle , sheep , and goats . It is transmissible to humans by direct transmission and the consumption of untreated milk . Brucellosis has become more and more frequent in Kyrgyzstan in the last decades , and its control has been made a priority . Knowing the bacterial strain circulating is important for the understanding of the transmission and the selection of interventions . The latest identification of Brucella in Kyrgyzstan dates from the 1960s . We report the molecular characterization 17 strains identified as Brucella melitensis from Naryn oblast . Strains were mainly isolated from sheep but also from cattle . All strains were susceptible to a series of antibiotics . We hence identified and confirmed transmission of B . melitensis among sheep which is likely the most important host species . We found close genetic relationship between B . melitensis strains isolated from cattle sharing the same pasture with sheep . Our results support the strategy of pursuing a mass vaccination of livestock in Kyrgyzstan . Further research is needed to identify the most important circulating strains in humans .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Conclusion"
] |
[
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"veterinary",
"epidemiology",
"veterinary",
"microbiology",
"medical",
"microbiology",
"epidemiology",
"biology",
"microbiology",
"public",
"health",
"veterinary",
"science"
] |
2013
|
Molecular Epidemiology and Antibiotic Susceptibility of Livestock Brucella melitensis Isolates from Naryn Oblast, Kyrgyzstan
|
Many emerging infections are RNA virus spillovers from animal reservoirs . Reservoir identification is necessary for predicting the geographic extent of infection risk , but rarely are taxonomic levels below the animal species considered as reservoir , and only key circumstances in nature and methodology allow intrinsic virus-host associations to be distinguished from simple geographic ( co- ) isolation . We sampled and genetically characterized in detail a contact zone of two subtaxa of the rodent Mastomys natalensis in Tanzania . We find two distinct arenaviruses , Gairo and Morogoro virus , each spatially confined to a single M . natalensis subtaxon , only co-occurring at the contact zone’s centre . Inter-subtaxon hybridization at this centre and a continuum of quality habitat for M . natalensis show that both viruses have the ecological opportunity to spread into the other substaxon’s range , but do not , strongly suggesting host-intrinsic barriers . Such barriers could explain why human cases of another M . natalensis-borne arenavirus , Lassa virus , are limited to West Africa .
Most emerging RNA virus infections originate from wild animals [1] . Fortunately , outbreaks of many such infections are geographically restricted , e . g . MERS coronavirus in the Middle East , Marburg filovirus in central and southern Africa , and Nipah henipavirus in south-east Asia . This restriction is most likely due to dependence on particular host reservoir species ( single or multiple ) for persistence in nature–those hosts themselves having restricted distributions . Identification of the reservoir host and its geographic range are therefore essential for informed public health responses [2] , dramatically illustrated by the 2014 outbreak of Zaire-Ebola virus that unexpectedly emerged in West Africa [3] . Repeated detection of a particular virus in a particular animal species , and not in other species in sympatry , usually implicates that species as the main reservoir . However , species with a wide geographic range are often ( cryptically ) genetically subdivided into subtaxa , yet it is rarely assessed whether a local intraspecific taxon may represent the reservoir instead of the entire species . Such associations between intraspecific animal taxa and particular viral taxa could explain why the distribution of some viruses appears smaller than the range of the reservoir species , for example in the case of distinct hantaviruses of the widespread rodent species Peromyscus leucopus [4 , 5] , P . maniculatus [4 , 6 , 7] and Oligoryzomys flavescens [8] , and the Simian Immunodeficiency viruses ( SIV ) and Simian Foamy viruses ( SFV ) of chimpanzees ( Pan troglodytes ) [9–12] . However , it is difficult to corroborate these associations . Experimental infections require housing individuals of the particular subtaxa in biosafety laboratories , and will in any case reflect capacity of the virus to infect the host , rather than whether the host has a reservoir status in nature: whether an infection may be persistently transmitted in a particular host population depends on more factors than the ability to propagate in a host body after manual inoculation . For example , the route and timing of viral shedding in concordance with the host’s population dynamics may be important determinants of the infection’s invasion and persistence probabilities in a host population [13 , 14] . Inferring subtaxon-virus associations from observations in nature may be more appropriate , but when spatial gaps or coinciding geographic barriers occur between sampling points of the distinct subtaxa , host-extrinsic factors such as isolation-by-distance or host movement barriers are indistinguishable as explanations of the spatial separation of viruses . Therefore , the association between virus and host taxa must be evaluated in areas where distinct host ( sub ) taxa carrying distinct viral taxa are in direct physical contact . This situation can be found in secondary contact zones . These are formed when vicariant subtaxa that had allopatrically diverged in the past , re-expand into secondary contact during favourable environmental conditions [15] . Commonly , this contact results in the production of fertile hybrids across a delineated and stable hybrid zone [16] . These limited zones are often maintained by a balance between dispersal and ( endogenous ) selection against hybrids , so that distinctive genepools co-exist in the face of gene flow [17] . For example , using fine scale sampling across the European house mouse hybrid zone , it was demonstrated that strains of Murine cytomegalovirus and of the protozoan Cryptosporidium tyzzeri are each associated with a distinct Mus musculus subspecies [18 , 19] . For RNA viruses , it might be expected such secondary contact zones provide the optimal ecological conditions for an evolutionary host shift to the closely related taxon across the zone , as even for host-specific RNA viruses their high mutation rates might ensure rapid adaptation to the exposed novel host . However , this has not yet been evaluated in nature . Host shift potential of RNA viruses has previously mainly been studied by comparing genealogical histories of virus phylogenetic groups and of their corresponding hosts [20–26] . This has allowed identification of e . g . phylogenetic distance between host taxa as an important constraint for a host shift; yet it remains unclear whether such a constraint may last when closely related subtaxa carrying different RNA viruses physically meet , for example at a secondary contact zone . Here , we characterize a secondary contact zone of subtaxa of the African rodent Mastomys natalensis to better understand host-imposed constraints to the distribution patterns of the rodent’s arenaviruses . Arenaviruses are bi-segmented RNA viruses and those of the genus Mammarenavirus are typically hosted by rodents . Only a few can successfully infect humans , and while human-to-human transmission is possible , it has so far never resulted in a sustained epidemic [27 , 28] . Lassa mammarenavirus ( LASV ) may cause a severe haemorrhagic fever in humans , and with about 200 , 000 cases and 3 , 000 deaths annually [29] it has a major public health impact [30] . LASV’s main natural reservoir host species is the Natal multimammate mouse Mastomys natalensis [31–33] , although recently LASV and LASV-related strains have also been detected in other rodent species [34] . While this common rodent occurs throughout most of sub-Saharan Africa , LASV and Lassa fever in humans is restricted to West Africa and has never been detected east of Nigeria ( Fig 1 ) . Instead , five other arenaviruses have so far been detected from M . natalensis in various other regions ( Fig 1 ) : Mopeia virus ( MOPV ) in Mozambique [35] , Morogoro virus ( MORV–a strain of MOPV ) in Tanzania [36] , Luna virus ( LUV ) in Zambia [37] , Gairo virus ( GAIV ) in Tanzania [38] and recently an unnamed Mobala-like virus in east-Nigeria [39] . These have never been detected in humans . Mammarenaviruses are in general considered rodent-host specific . The majority has only been detected in a single rodent species , and on several occasions distinct arenaviruses have been found in distinct rodent species captured at the same sites [45–49] . On the other hand , five arenaviruses have been detected in multiple , yet closely related , rodent species [34 , 50–53] , suggesting that the level of host specificity may vary between arenaviruses . M . natalensis is one of the most widespread and common mammals in sub-Saharan Africa; it occurs in all terrestrial habitats apart from dense forests and ( semi- ) deserts [54 , 55] . Its mitochondrial DNA can be divided into six matrilineages that differ up to 3 . 8% at the cytochrome b gene and that are each geographically confined to distinct regions [43] . When superimposing the distributions of these matrilineages and M . natalensis-borne arenaviruses , each arenavirus seems to be restricted to the range of a single matrilineage , the Lassa fever endemic area roughly matching the range of M . natalensis matrilineage A-I ( Fig 1 ) . This pattern suggests that host intraspecific structure , as approximated by the matrilineal pattern , may constrain the geographic ranges of arenaviruses including LASV . Recently , Olayemi et al . ( 2016 ) concluded this is not the case , as they detected LASV in three M . natalensis individuals carrying the A-II mitochondrial lineage in eastern Nigeria . However , no nuclear markers were typed , and the A-II mitochondrial lineage was observed in less than 25% of the animals in the two localities where these LASV positive animals were detected , located at the edge of A-I matrilineage distribution . Moreover , a different and new arenavirus was found in an eastern locality across the river Niger where the A-II mitochondrial lineage was found in all individuals . Therefore , it remains unclear whether the observed pattern was due to LASV dispersal into the range of the host taxon associated with the A-II matrilineage , or due to A-II mitochondrial introgression into the range of the host taxon carrying LASV . Indeed , introgression of mitochondrial lineages into other taxa is a common phenomenon , therefore the spatial distribution of mitochondrial lineages often does not closely match that of the ( multi-locus inferred ) taxa themselves [56] . In central Tanzania , M . natalensis–borne GAIV and MORV are known to occur in close proximity [36 , 38] , and the subtaxa potentially represented by M . natalensis matrilineages B-IV and B-V are estimated to be in secondary contact in that region [43] . We previously found within the B-V matrilineage range that nuclear markers are further substructured in relation to an urban-rural contrast , and that the region varies in landscape features that likely translate into spatially varying M . natalensis densities [57] . Such areas of low host densities and/or inter-host contacts might in themselves present a sufficient barrier for virus transmission [58] , and should therefore be taken into account when distinguishing host-intrinsic and host-extrinsic factors of viral distribution patterns . In this study , we sampled M . natalensis at a fine scale across the spatial transition zone between GAIV and MORV , multilocus genotyped hosts and their arenaviruses , and estimated landscape connectivity between localities . In the context of this natural laboratory , we are able to discern the contributions of host genetic structure and landscape features on the spatial distribution of arenaviruses .
We sampled small mammals at twelve localities along the road between Dar es Salaam and Dodoma , spaced approximately 20 km apart and spanning the distribution range boundary of the B-IV and B-V M . natalensis matrilineages [43] and Gairo and Morogoro arenaviruses ( Fig 1 ) . Part of the sampling overlaps with data presented in [38] and [57]; see Table 1 . At each locality small mammals were captured in Sherman live traps baited with a mixture of peanut butter and maize flour . Traps were set in a 1 ha square grid of 10x10 traps in fallow lands . At each locality minimum two grids were constructed minimum 500 m , maximum 2 . 5 km apart . If trapping success was low after two nights , we set additional grids . For each sampled grid maximum 20 M . natalensis individuals were euthanized by Isoflurane inhalation . Blood was drawn either from the retro-orbital sinus or the punctured heart with a capillary tube and preserved on pre-punched filter papers ( Serobuvard , LDA 22 , 106 Zoopole , France ) , and organ samples were preserved in RNAlater and ethanol . RNAlater samples were kept at 4°C for maximum six weeks prior to storage at -80°C . When more than 20 M . natalensis were captured in a grid , supernumerary animals were sedated through Isoflurane inhalation , blood and toe-clips sampled on filter paper and in ethanol , respectively , and each was released at point of capture . Molecular screening of arenaviruses was augmented with six dried-blood sample collections from previously published rodent-trapping work [57 , 59–61] ( see details in Table 1 ) . M . natalensis genotyping for microsatellite and cytochrome b markers ( see below ) was augmented using three of these additional collections ( see Table 1 ) . All animal work was approved by the University of Antwerp Ethical Committee for Animal Experimentation ( 2011–52 ) , and followed regulations of the Research Policy of Sokoine University of Agriculture as stipulated in the “Code of Conduct for Research Ethics” ( Revised version of 2012 ) . Euthanasia of small mammals was performed using an overdose of Isoflurane or via cervical dislocation . DNA was extracted from toe or liver samples using the DNeasy Blood & Tissue Kit ( Qiagen ) . Fifteen microsatellite loci [62] were genotyped as described in [57] . However , only those samples for which more than 10 loci were successfully amplified were considered for downstream analyses . Parts of cytochrome b ( on the maternally inherited mitochondrion ) and smcy ( on the paternally inherited Y chromosome ) were amplified in PCRs and Sanger sequenced in one direction . See further PCR details in in S1 Text . We analysed the population genetic structure of M . natalensis microsatellite genotypes using the Bayesian clustering algorithm implemented in the program STRUCTURE v2 . 3 . 2 . [63 , 64] , using the same settings as described in [57] . In brief , genetic clusters are sought in which deviation from genetic disequilibria are minimised , with proportions of each microsatellite genotype assigned to each of K clusters . The analysis was replicated 25 times for each K value , allowing for admixture and using a prior on shared sampling location ( at the locality-level ) . Modes in STRUCTURE outputs were distinguished using CLUMPAK; similar level-K replicates are placed in the same mode , within-mode cluster labels are standardized , and assignments across modes and K levels calculated [65] . Similarity between clusters at different K-levels and modes was assessed by eye and for visual clarity given the same colour . Cytochrome b sequences were aligned and compared to published M . natalensis sequences by constructing a Maximum Likelihood phylogenetic tree in RAxML ( GTR substitution model , gamma rate variation , 1000 bootstraps ) [66] . Each sequence was then assigned to lineage B-IV or lineage B-V as described in [43] . Smcy sequences were aligned in Geneious 6 . 1 using the Geneious alignment algorithm with a 5 . 0/-9 . 203 match/mismatch cost model . Arenavirus RNA was screened in RNA extracted from dried blood samples ( pooled by two ) using two independent one-step reverse transcription-PCRs ( RT-PCRs ) targeting the same 340 nucleotide ( nt ) portion of the RNA-dependent RNA polymerase gene ( L segment ) , but with different primers with different target affinities ( see details in S1 Text ) . For pools positive for this viral gene ( and a subset of 347 negative samples ) , additional RNA was extracted from individual kidney biopsies preserved in RNAlater using the Nucleospin RNA II kit ( Macherey-Nagel ) when available . From these RNA extract parts of the GPC gene ( 979 nt or 234 nt ) and NP gene ( 558 nt or 450 nt ) were amplified . All amplicons were Sanger sequenced in both directions . See S1 Text for further details on these assays . We aligned our L , NP and GPC sequences with arenavirus sequences available in GenBank ( all from rodents , except Lujo virus ) in Geneious 6 . 1 based on the translated amino acid sequences ( Blosum62 cost matrix ) . We removed the short non-coding parts , and constructed the phylogenetic trees of each partial gene sequence in MrBayes , ( GTR substitution model , gamma rate variation: 6 categories , rate parameters estimated separately for each codon ) . Since we were only interested in topology and not in dating nodes , we minimised parameters by using a uniformly distributed strict clock prior on branch lengths . We let 4 MCMC chains run for 1 million iterations after the standard deviation of the split frequency reached 0 . 01 . The replicate analyses without assuming a clock model ( unconstrained branch lengths using an exponential prior probability distribution ) did not significantly differ in likelihood or topology ( S3 Fig ) . As a quantitative measure of the potential for M . natalensis to move between localities , i . e . a measure of environmental barriers , we estimated landscape resistance pairwise between localities of the transect ( Fig 1 , localities A to L , and excluding all other sampled localities ) , using the methodology of [57] but applied over a larger geographic extent . In brief , ten field experts in M . natalensis ecology translated landscape elements of Tanzania’s land cover layer of 1997 [67] to exponentially increasing categories of M . natalensis habitat quality . After integrating linear landscape elements into this habitat quality layer ( rivers and roads of three different width categories ) , the resulting expert opinion layer was modelled using Circuitscape [68] as a conductive surface from which resistance values between pairs of polygons ( minimum polygons drawn around sampling sites in each locality ) are calculated , in analogy with circuit theory . The centroids of each locality and the earth ( great-circle ) distance between them were calculated in the R package ‘fields’ [69] . We based the mean genetic distance between arenavirus samples of each locality on a concatenation of the three arenavirus gene sequences ( to a total of 1848 nucleotides ) and then calculated this mean distance between localities in MEGA 5 . 2 ( Tamura 3-parameter model ) [70] . The correlation between the genetic distance matrices and the landscape resistance distance matrix or earth distance matrix was calculated using simple Mantel tests in the R package ecodist [71] ( 1 , 000 , 000 permutations for bootstrapping ) . Partial Mantel tests of the same package were used to correlate the genetic distance matrices to the landscape resistance distance matrix while ‘partialling out’ the influence of earth distance between localities . Sampling coordinates were transformed to a flat surface using gnomonic projection . The centroids of the twelve transect localities ( Fig 1 , A to L ) were then orthogonally projected onto their regression line to form a one-dimensional transect . Narrow clines are robust to such mapping details [72] . For each locality , frequencies of the M . natalensis matri- ( cytochrome b ) and patri- ( smcy flank ) lineages and the average assignment to either of the microsatellite clusters in the integrated Q matrix of the K = 2 STRUCTURE scenario were tabulated . Numbers of MORV and GAIV in the total arenavirus infected ( RT-PCR positive ) animals were tabulated per locality . We fitted clines to these observations using the software Analyse [73] . To evaluate whether by chance we sampled more related animals in some localities than others , we calculated Li’s relationship coefficient r [74] between pairs of host genotypes within each locality in SPAGeDi [75] . See details in supplementary Methods in S1 Text .
Genetic sequences generated in this study are deposited in GenBank: Original data deposited in the Dryad repository: http://dx . doi . org/10 . 5061/dryad . 5n00k [76]:
Of 1 , 289 M . natalensis individuals sampled in 12 localities , a random subset was genotyped , as well as 39 additional M . natalensis individuals from one southern and two northern localities in Tanzania ( “wide-scale localities”; Fig 1 , Table 1 ) . We identified two M . natalensis cytochrome b matrilineages ( B-IV and B-V sensu Colangelo et al . ( 2013 ) [43]; S1 Fig ) , as well as a bi-allelic SNP in the Y chromosome smcy-gene intron , suggesting two distinct patrilineages ( later referred as A and T lineages ) ( Table 1 ) . When allowing for two microsatellite clusters ( K = 2 ) in STRUCTURE , all replicate runs ( 25/25 ) converged to the same spatial pattern of genetic structure ( Fig 2 ) , strongly supporting a division of our sample into two replicable genetic disequilibrium-minimising groups . Outwith the transect , in northern localities Shinyanga and Itigi all M . natalensis carried the mitochondrial B-IV lineage and belonged to microsatellite cluster 1 ( yellow ) , while in the southern locality Lihale all mice carried B-V mitochondrial lineage and belonged to microsatellite cluster 2 ( blue ) ( Table 1 , Fig 2 ) . Along the sampled transect the proportions of the two matrilineages , patrilineages and microsatellite cluster memberships changes sharply between localities B and E ( Fig 3 , Table 1 ) . In locality C , Berega , both maternal and paternal lineages are present and autosomal hybrid genotypes dominate ( Table 1 , Fig 2 ) , indicating ongoing hybridization between two M . natalensis subtaxa . The clines for all three M . natalensis genomic compartments ( mitochondrion , Y chromosome and autosomes ) have narrow confidence intervals and very similar estimated cline centre positions ( near locality C ) and cline widths ( Fig 3 ) . The consistency across genomic compartments and the relatively narrow estimated cline widths ( 20 . 0 , 21 . 0 and 21 . 6 km , respectively–Fig 3 ) indicate a multilocus barrier to gene flow between the two M . natalensis subtaxa . Together with the observations from the wide-scale localities , we can conclude M . natalensis matrilineages B-IV and B-V correspond with genome-wide genetic structure in this region . We will therefore subsequently refer to these genome-wide clusters as M . natalensis subtaxa B-IV and B-V . The B-IV mitochondrial lineage , but not the smcy A allele , is observed at low frequencies throughout the sampled transect localities in B-V’s range ( localities E-L , up to 140 km from the estimated clines centres ) , indicating wide scale but low level mitochondrial introgression ( Table 1 ) . Low level introgression of both B-V mitochondrial lineage and smcy T allele was also observed in the sampled transect localities in the B-IV subtaxon range ( localities A and B ) , but this range was only sampled up to 40 km from the estimated clines centers . For STRUCTURE K>2 , the two subtaxa’s microsatellite clusters are further hierarchically substructured with consistent spatial pattern ( Fig 2 ) . At K = 3 ( major modes ) or K = 4 ( minor modes ) , animals from locality B ( Chakwale ) form a sub-cluster embedded within subtaxon B-IV . This substructure may be due to significantly higher levels of relatedness within this locality than in others ( S1 Text and S2 Fig ) . At K = 3 ( minor modes ) or K = 4 ( major modes ) , animals from locality H ( Morogoro ) also form a sub-cluster embedded within subtaxon B-V , consistent with previous inference over a subset of this dataset [57] . These animals do not show high relative relatedness patterns ( S1 Text and S2 Fig ) . At K = 5 ( minor modes ) and K = 6 ( major modes ) , a subset of the animals at the southern locality Lihale are consistently distinguished . These animals are significantly more related to each other than animals in other localities ( S1 Text and S2 Fig ) . A total of 53 arenavirus positive samples were found in 1 , 167 dried blood samples ( DBS ) from the transect-localities and in 392 blood samples from additional collections ( Table 1 ) . A further 6 out of 347 kidney samples tested ( from individuals with negative DBS ) were arenavirus RT-PCR positive . Phylogenetic reconstruction of parts of the L , GPC and NP genes showed positive samples contained the arenaviruses MORV and GAIV ( Fig 4 ) . Bayesian phylogenetic trees based on the partial L ( Fig 4A ) , GPC ( Fig 4B ) and NP gene ( Fig 4C ) showed four clades for MORV that were each specific to a single or set of adjacent localities ( Fig 1 ) . One clade ( MORV-III ) was however not well supported by the phylogenetic tree based on NP sequences ( Fig 4C ) . MORV-III monophyly was also not supported by phylogenetic reconstruction without branch length constraints ( S3 Fig ) . The remaining topology was similar under clock and non-clock models ( S3 Fig ) . Along the transect GAIV was only detected in majority B-IV host subtaxon localities and MORV only where the B-V host subtaxon was in the majority ( Table 1 , Figs 1 , 3 and 4 ) . The only exception was the hybrid-host-rich locality C , Berega , where both GAIV and MORV were detected at low prevalence ( 1 . 5% and 0 . 2% , respectively; Table 1 ) . GAIV was also found in Mbulu and Shinyanga in north Tanzania , consistent with it being present across the range of the northern B-IV subtaxon ( Fig 1 ) . The spatial frequency clines of the two arenaviruses thus coincide with their host’s genotypic clines , with estimated virus cline centre and width falling within the 95% confidence intervals of those of the host clines ( Fig 3 ) , although virus cline confidence intervals are much broader due the lower number of virus observations . The association between viral type and host taxa on either side of the cline centre is highly significant ( χ2 = 42 , df = 1 , p = 9 . 1 x 10−11 ) . Both virus and host transition zones centre around locality C ( Fig 3 ) . Pairwise landscape resistance varies across the localities within the zone centre’s confidence intervals ( B-C-D-E-F , 13 . 6 , 12 . 2 , 30 . 6 , 7 . 9 , Fig 3 , S1 Table ) with maximum between D-E . Across the transect as a whole , J-K-L comparisons have higher resistance estimates than D-E . Neither of these regions of high host-movement-resistance are likely to match the true centre of the GAIV-MORV transition zone nor appear to genetically structure the arenavirus associated with the B-V host subtaxon: MORV occurs on both sides of D-E and the clade MORV-II occurs in localities J as well as K . A partial mantel test correlating mean arenavirus ( GAIV+MORV ) genetic distance between pairwise localities to the landscape resistance , while partialing out the earth distance matrix , was not significant ( R2 = -0 . 13 , p = 0 . 95 ) . Mantel tests ( both simple and partial ) lead to a high number of false-positive correlations [78] , making this lack of correlation more striking . To summarise: away from the host contact zone landscape resistance is stronger than that within the zone , and even at its strongest , does not correlate with intraspecific arenavirus genetic substructure . In contrast to the large-scale geographic association between arenavirus species and M . natalensis subtaxa , there was no geographic match between population genetic structure of M . natalensis within the B-V taxon ( Fig 2 ) and the sublineages of MORV ( Fig 1 and Fig 4 ) . M . natalensis-B-V from locality H ( Morogoro ) show genetic separation from surrounding localities , but their MORV strains are of the same sublineage as those of adjacent localities G and I ( Fig 1 , Fig 2 , Fig 4 ) . M . natalensis from all other B-V localities belong to the same population genetic cluster , while carrying four distinct MORV sublineages ( Fig 1 , Fig 2 and Fig 4 ) . A partial mantel test correlating mean MORV genetic distance between pairwise localities ( excluding GAIV specific localities A and B ) to the landscape resistance , while partialling out earth distance was significant ( R2 = 0 . 02 , p = 0 . 001 ) but weaker than a simple mantel test correlating mean genetic distance to earth distance ( R2 = 0 . 38 , p<0 . 001 ) , indicating isolation-by-distance . GAIV’s range was not sampled broadly enough for equivalent analyses in relation to structure within M . natalensis-B-IV subtaxon .
We show evidence suggesting host-intrinsic factors determine the spatial distributions of two arenaviruses . This paves the way for experimental studies investigating what such factors might be . Candidate factors should include a wide range of possibilities from direct host immune defense differences to indirect effects , for example differences in infection-mediated behavior that enhance viral transmission . Interactions with other infections/symbionts should not be ignored: the host taxa may have , for example , diverged not only in their genomes but also in their gut microbiomes . It should also be borne in mind that host-intrinsic effects may be associated with hybrids , which can show vigorous immune response due to heterosis [88] . The combination of mechanisms involved is unlikely to be simple . For example , the configuration of α-dystroglycan , the ( known ) main cell entry receptor of Old-World mammarenaviruses is invariant across several species of the Mastomys genus [89] , ruling this out as a simple explanation . Replicating relevant aspects of the population-level process of viral transmission in the laboratory will therefore be challenging . As mentioned in the introduction , the human Lassa fever endemic area roughly matches the distribution range of M . natalensis A-I matrilineage , but recently , Olayemi et al . ( 2016 ) found LASV in three M . natalensis individuals carrying the A-II mitochondrion and concluded that LASV may spread to the rest of the A-II matrilineage range ( which extends up to eastern Democratic Republic of Congo; Fig 1; [43] ) [39] . However , the observations in Olayemi et al . ( 2016 ) actually appear consistent with the current study: 1 ) frequencies of matrilineages A-I and A-II ( only mitochondria were typed ) gradually changed along a west-east axis; 2 ) LASV and no other arenavirus was found in two localities where A-I predominated over A-II ( 1/9 and 2/9 genotyped animals carrying A-II , the remainder A-I ) ; 3 ) a different , Mobala-like arenavirus , but not LASV , was found in localities where A-II mitochondrion predominates ( A-II found in 3/3 genotyped animals , and lies east of a locality where A-II was found in 17/19 animals ) . As the authors note , M . natalensis subtaxa associated with A-I and A-II matrilineages thus likely form a hybrid zone in eastern Nigeria , potentially coinciding with the river Niger . Comparing these findings with our multilocus fine-scale data from Tanzania , we would predict that the locality where LASV was detected in three individuals with A-II mitochondria lies west of the hybrid zone’s centre , across which these A-II mitochondrial copies have introgressed , as is common in contact zones [56] and as we also observed in this study . We would thus predict multilocus ( not mitochondrial ) genotyping would cluster those three individuals in the “A-I subtaxon” . The convergent spatial patterns in Olayemi et al . ( 2016 ) and this study are thus consistent with a general association of the arenaviruses of M . natalensis to particular subtaxa , implying that LASV is restricted to the West African range of the subtaxon corresponding with A-I matrilineage . However , firstly the potential dispersal barrier effect of the river Niger on M . natalensis nuclear gene flow and LASV transmission should be evaluated as alternative explanation . Secondly , the role of other rodent hosts in the spatial spread of LASV needs to be clarified . It has recently become clear that several strains of LASV may be harboured by rodent species other than M . natalensis , namely the closely related M . erythroleucus and Hylomyscus pamfi [34] , and divergent LASV-related strains have been found in distantly related Mus baoulei and Mus cf . setulosus [90] . Two important questions thus remain: ( 1 ) is LASV a generalist whereas other African arenaviruses , especially GAIV and MORV , are specialists ? This would explain why LASV is found in a wide array of species , including humans , but fails to explain why the distribution of LASV appears largely bordered by the distribution of M . natalensis A-I matrilineage . Therefore: ( 2 ) Do hosts other than M . natalensis A-I subtaxon contribute to the long-term persistence of LASV in nature ? M . erythroleucus has been found carrying LASV in an area just outside of M . natalensis’ range ( coastal Guinea ) and where human Lassa fever cases have also been reported–implying that either humans or M . erythroleucus managed to import LASV and establish a transmission chain without involvement of M . natalensis [34] . On the other hand , it is clear that these importations have not ( yet ) occurred very far outside of the M . natalensis A-I matrilineage range , despite the continuous distribution of M . erythroleucus , H . pamfi and of course humans through to other parts of Africa . It therefore seems that either: 1 ) M . erythroleucus and H . pamfi are not able to sustain a long-term persistent LASV transmission , similar to the situation in humans [27 , 28] . Such less efficient transmission could e . g . be due to differences in intra-host infection dynamics and population ecology of these species in comparison to M . natalensis . 2 ) LASV is sustainably transmitted in M . erythroleucus and H . pamfi populations , and LASV’s distribution in reality , and unnoticed , expands throughout the ranges of M . erythroleucus and H . pamfi ( though note that at least M . erythroleucus is known to be subdivided into subtaxa similar to M . natalensis [91] ) . Perhaps they carry LASV strains less pathogenic to humans elsewhere; such a situation has previously been postulated to explain the absence of reported Lassa fever patients in regions in Mali where a particular LASV strain has only recently been found to be common in M . natalensis [32] . 3 ) LASV strains in M . erythroleucus and H . pamfi are the result of adaptive host switching events that occurred only recently , and have yet to expand through the rest of the rodents’ ranges . It is clear more surveillance of arenaviruses in rodents , especially in the area bordering the Lassa fever endemic area , is needed to fully answer the pressing question: should we expect LASV and its associated diseases to emerge in the rest of Africa ?
|
Reservoirs of zoonotic viruses are usually equated with a particular wildlife species . It is rarely assessed whether genetic groups below the species level may instead represent the actual reservoir , though this would have major implications on estimations of the zoonosis’ spatial distribution . Here we investigate whether geographically and genetically distinct subtaxa of the widespread African rodent Mastomys natalensis carry distinct arenaviruses , by sampling in detail across a contact zone of two of these subtaxa . Ongoing hybridization shows that individuals of the subtaxa are in direct physical contact , in principle allowing viral exchange , yet neither of the two arenaviruses -Gairo and Morogoro virus- were found to have crossed the zone . Such intraspecific genetic barriers to arenavirus spatial spread have important implications for our understanding of the related Lassa arenavirus , a pathogen potentially lethal to humans of which Mastomys natalensis is also the main reservoir . Although Lassa virus appears to infect several secondary hosts , its distribution is restricted to West Africa and matches that of another M . natalensis subtaxon . Our data thus indicates that it is because of M . natalensis intraspecific distinctions that the human Lassa fever endemic area has not expanded to the rest of sub-Saharan Africa .
|
[
"Abstract",
"Introduction",
"Methods",
"Data",
"Archiving",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"rna",
"extraction",
"microbiology",
"vertebrates",
"geographical",
"locations",
"animals",
"mammals",
"viruses",
"rna",
"viruses",
"phylogenetic",
"analysis",
"mitochondria",
"molecular",
"biology",
"techniques",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"extraction",
"techniques",
"africa",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"sequence",
"alignment",
"bioinformatics",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"people",
"and",
"places",
"biochemistry",
"rodents",
"cell",
"biology",
"arenaviruses",
"viral",
"pathogens",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"amniotes",
"organisms"
] |
2017
|
When Viruses Don’t Go Viral: The Importance of Host Phylogeographic Structure in the Spatial Spread of Arenaviruses
|
Resistance emergence against antileishmanial drugs , particularly Sodium Antimony Gluconate ( SAG ) has severely hampered the therapeutic strategy against visceral leishmaniasis , the mechanism of resistance being indistinguishable . Cysteine leucine rich protein ( CLrP ) , was recognized as one of the overexpressed proteins in resistant isolates , as observed in differential proteomics between sensitive and resistant isolates of L . donovani . The present study deals with the characterization of CLrP and for its possible connection with SAG resistance . In pursuance of deciphering the role of CLrP in SAG resistance , gene was cloned , over-expressed in E . coli system and thereafter antibody was raised . The expression profile of CLrP and was found to be over-expressed in SAG resistant clinical isolates of L . donovani as compared to SAG sensitive ones when investigated by real-time PCR and western blotting . CLrP has been characterized through bioinformatics , immunoblotting and immunolocalization analysis , which reveals its post-translational modification along with its dual existence in the nucleus as well as in the membrane of the parasite . Further investigation using a ChIP assay confirmed its DNA binding potential . Over-expression of CLrP in sensitive isolate of L . donovani significantly decreased its responsiveness to SAG ( SbV and SbIII ) and a shift towards the resistant mode was observed . Further , a significant increase in its infectivity in murine macrophages has been observed . The study reports the differential expression of CLrP in SAG sensitive and resistant isolates of L . donovani . Functional intricacy of CLrP increases with dual localization , glycosylation and DNA binding potential of the protein . Further over-expressing CLrP in sensitive isolate of L . donovani shows significantly decreased sensitivity towards SAG and increased infectivity as well , thus assisting the parasite in securing a safe niche . Results indicates the possible contribution of CLrP to antimonial resistance in L . donovani by assisting the parasite growth in the macrophages .
Infection of Leishmania has several clinical manifestations as visceral , cutaneous and mucocutaneous forms of leishmaniasis and visceral leishmaniasis ( VL ) , among these , VL is the deadly form in the absence of proper treatment . In India , particularly the states of Bihar , adjoining areas of West Bengal and Jharkhand themselves carry about half the burden of the world’s account of VL [1] . Sodium antimony Gluconate ( SAG ) , having a chemotherapeutic background of 60yrs against VL , is now obsolete in the endemic areas of Bihar due to widespread resistance to antimonials [2] . The emergence of SAG resistance along with the limited availability of safe and cost-effective antileishmanial agents has worsened the situation and raised the chemotherapeutic challenges . Although , there has been a significant advancement in the treatment of VL , the question of resistance still remains unanswered . The resistance phenomenon has been studied mostly in laboratory mutants that differ a lot from field isolates [3 , 4 , 5 , 6] . Earlier some studies based on closely related metal arsenic were carried out to understand the resistance mechanism but was less worthy as it differs from antimony’s working mechanism in several aspects such as in increasing intracellular calcium and not affecting the glutathione level etc . [4] . Other resistance based studies were mostly carried out on the laboratory prepared mutants . While some studies lay emphasis on clinical isolates , but these were based on several biochemical , biophysical and immunological investigations [5 , 6] . Actual mechanism could be interpreted by exploring clinical isolates on a molecular level and characterizing the differentially regulated proteins in the resistant field isolates [7 , 8] . Phenotypes , genomic and proteomic level approaches have been applied to investigate the resistance at cellular and molecular level [9 , 10 , 11 , 12] . In order to understand the mechanism at protein level differential proteomics of sodium antimony gluconate ( SAG ) sensitive and SAG resistant clinical isolates was done wherein several cytosolic as well as membrane proteins were found to be differentially expressed in a SAG resistant strain of L . donovani [13] . CLrP was spotted as one of the up-regulated protein in the membrane fraction [13] . CLrP is a member of the superfamily of Leucine-rich repeat ( LRR ) proteins . With the purpose of gaining an in-depth knowledge about CLrP and to understand its association with SAG resistance , CLrP has been characterized and over-expressed in sensitive isolate of L . donovani ( Ld ) to analyze its potential to modulate the parasite’s behavior towards SAG .
Clinical isolates were procured from VL patients from Kala–Azar Medical Research centre , Muzaffarpur , Bihar , India . The isolates from patients , who had responded to chemotherapy by SAG were termed as SAG-sensitive ( SAG-S ) , whereas those who remain unresponded to SAG were termed as SAG-resistant ( SAG-R ) . SAG-S and SAG-R isolates used in the present study were 2001 ( S1 ) and 2039 ( R1 ) , 1216 ( R2 ) , 761 ( R3 ) , whereas Dd8 ( S2 ) strain ( MHOM/IN/80/DD8 ) served as reference strain . All the isolates were maintained in vitro in RPMI-1640 medium ( 10%FCS ) ( Sigma , USA ) at 25°C and their virulence have been retained through regular passage in hamsters , so as to maintain their chemosensitivity profiles as described elsewhere[14] . Genomic DNA of L . donovani was isolated from 108 cultured promastigotes and subjected to RNase ( 100μg/ml ) treatment [15] . CLrP gene was amplified using Taq DNA polymerase ( Sigma Aldrich ) lacking 3’-5’ exonuclease activity in a thermocycler ( Bio-Rad ) under conditions at one cycle of 95°C for 5min , 30 cycles of 95°C for 1min , 60°C for 45s , 72°C for 1 . 5min , and finally one cycle of 72°C for 10min ( Table A in S1 Text ) . Amplified PCR product was gel eluted from the gel ( electrophoresed in agarose gel ) by Gen Elute columns ( Qiagen ) . Eluted CLrP was cloned in pTZ57R/T ( T/A ) cloning vector ( Fermentas ) and transformed into competent DH5α cells . LdCLrP was further sub cloned at the BamHI and EcoRI site of vector pET-28a ( + ) ( Novagen ) and transformed in Escherichia coli BL21 Strain . The positively transformed cells were inoculated into 5ml test tube culture medium ( Luria Bertani ) and allowed to grow at 37°C in a shaker at 220 rpm . Cultures in logarithmic phase ( at OD600 of ~0 . 5–0 . 6 ) were induced for 5hrs with 1mM isopropyl-ß-D-thiogalactopyranoside ( IPTG ) at 18°C . The lysis of induced as well as uninduced cells was done in SDS- sample buffer ( 5X stock ( 0 . 313M Tris-Hcl ( pH6 . 8 ) , 50% glycerol , 10%SDS ) [16] . These SDS-PAGE separated proteins were transferred onto a nitrocellulose membrane as described elsewhere [17] . Blocking of membrane was done using 3% skimmed milk for 1h followed by a 2h incubation with 1:2500 dilution of mouse anti-His antibody ( Novagen ) . This step was followed by incubation with 1:10 , 000 dilution of goat anti-mouse HRP conjugate antibody ( Bangalore Genei ) for 1h . All the incubations were performed at room temperature . The blot was developed using an ECL kit ( GE Biosciences ) . For further purification of rLdCLrP , 200ml of LB medium containing 35μg/mL kanamycin was inoculated with E . coli BL21 strain positively transformed with LdCLrP+pET28a+ and grown till O . D . 600 reaches to 0 . 6 at 37°C . The culture was induced by addition of 1mM ( IPTG , Sigma ) and then incubated for 8 hrs at 18°C . The rLdCLrP was purified by 6-His Tag fusion peptide derived from the pET28a+ vector by affinity chromatography using Ni2+ chelating resin . The cells were resuspended in 5mL of lysis buffer [10mM Tris-HCl ( pH 8 . 0 ) , 200mM NaCl] containing 1:200 dilution of the protease cocktail inhibitor ( Sigma ) with a 30 mins incubation on ice followed by sonication for 10×20s ( with 30s interval between each pulse ) . After sonication the cells were centrifuged at 15 , 000g for 30min , and the collected supernatant was incubated at 4°C for 1h with the 2ml of Ni-NTA Superflow resin ( Qiagen , Hilden , Germany ) equilibrated prior with lysis buffer . After washing with buffer ( 10mM Tris-HCl , 200mM NaCl ) containing 10 , 20 , 30 and 50mM concentrations of imidazole , the purified rLdCLrP was eluted with elution buffer ( 10mM Tris-HCl , 200mM NaCl , and 300mM imidazole , pH 7 . 5 ) . Estimation of the protein content in eluted fractions was carried out by the Bradford method and were analysed in 12% SDS-PAGE . For protein homology analysis and structure prediction , Hhpred ( toolkit . tuebingen . mpg . de/hhpred ) was used . HHpred constructs a Hidden Markov Model based profile using query sequences against known databases to search for closest possible homolog with HMM profile-profile comparison . PDB database was chosen here for profile analysis and homolog searching . Top scoring PDB was selected as template for further model building . Structure of query sequence was constructed using Modeller9 . 11 [18] interfaced with Hhpred server based on template 1z7x . Molecular modeling revealed that each conserved fragment consists of a short beta strand and helical coil . Motif analysis was done with ClustalW [19] . Prediction with TMpred [20] programs indicated two transmembrane regions ( residues 47 to 70 and 160 to 183 ) might be present in the protein . However HMMTOP [21 , 22] and TMHMM ( www . cbs . dtu . dk/services/TMHMM/ ) servers predicted no transmembrane domains in the amino acid sequence of L . donovani LLR . PSI-BLAST was implied to search for potential sequence homologues searches against the nr database for 10 iterations . The E-value cutoff was set to 10–5 . This allows detection of far-off members of the sequence family that simple pair wise comparison would fail to disclose . A sequence alignment was generated using alignment program MUSCLE [23] . Sequences with less than 90% identity with each other were selected only and mapped to the uniprot id ( HH filter http://toolkit . tuebingen . mpg . de/hhfilter ) . Phylogenetic tree was calculated using phylogeny . fr ( phylogeny . lirmm . fr ) with default values of parameters ( Fig A in S1 Text ) . The preparation of soluble L . donovani promastigote antigen ( SLD ) has been described elsewhere [24] . Briefly , mid-phase promastigote ( 109 ) ( 3 to 4 days old culture ) were washed 4 times in cold 1×PBS , further resuspended in 1×PBS with protease inhibitor mixture ( Sigma-Aldrich ) , and further ultrasonicated and centrifuged at 40 , 000g for 30 min . The quantification of protein content of the supernatant was done and the sample was prepared in SDS PAGE sample buffer [25] . Polyclonal antibodies of rLdCLrP have been raised in rabbit ( New Zealand white rabbit ) with the purified recombinant CLrP as described elsewhere [15] . For Western blot analysis , purified rLdCLrP and whole cell lysate ( WCL ) ( Leishmania as well as E . coli induced and uninduced ) was resolved in 12% SDS-PAGE and transferred onto nitrocellulose membrane [17] . The membrane was incubated with antiserum to rLdCLrP ( raised in rabbit ) at a dilution of 1:5000 for 2 hrs at room temperature ( RT ) after 2 hrs blocking in 5% BSA . After washing with PBS containing 0 . 5% Tween 20 ( PBS-T ) ( 1× ) the membrane was incubated with Rat anti-rabbit IgG HRP conjugate ( Invitrogen , Carlsbad , USA ) for 1h at RT at a dilution of 1:10 , 000 . Blot has been developed by using diaminobenzidine+imidazole+H2O2 ( Sigma ) . Cellular localizations of CLrP have been verified by immunofluorescence using anti-rLdCLrP antibody raised in rabbit . L . donovani parasites ( S1 ) were plated on 18mm cover-slips prior coated with poly-L lysine . These samples were fixed using 4% paraformaldehyde . 0 . 5% Triton X-100 has been used to permeabilized two of them followed by 1×PBS washing . Both the permeabilized as well as one non-permeabilized cover-slips were incubated with anti-CLrP . These were further treated with secondary anti-rabbit FITC-conjugate ( Bangalore Genei ) after washing with 1×PBS . Rhodamine-Concanavalin A ( Vector Labs ) has been used to treat non-permeabilized samples followed by anti-rabbit FITC-conjugate for 1h at RT ( 1×PBS washing after each incubation ) . Cover-slips , mounted upside down on glass slides with Fluorescent Mounting Media ( CALBIOCHEM ) , were visualized under a fluorescence microscope ( Eclipse 80i Nikon ) using 100X oil objective ( 1 . 4 NA ) . Cells transfected with rLdCLrP+pXG-‘GFP+ were also observed directly under the same fluorescent microscope [26] . To analyze propinquity of DNA with CLrP , a ChIP assay was done as described by with slight modifications [27 , 28] . L . donovani ( S1 ) was grown to log phase ( 3×107cells/ml ) ( ~40ml ) and were fixed for 5min with 1% formaldehyde at 25°C . 2 . 5ml of 2 . 5M glycine was added to stop the fixation process , and was further incubated for 5min at 25°C and then washed with 1×PBS . Cells were further resuspended in 2ml ChIP lysis buffer {50mM Tris-HCl ( pH8 . 0 ) , 150mM NaCl , 1% Triton X-100 , 0 . 1% Sodium deoxycholate and protease inhibitor cocktail ( Sigma ) } . Cells were subjected to sonication ( QSONICA Misonix ) with 10s pulse at 10% amplitude followed by a 1min pause after each pulse . Sheared cells were visualized under microscope and then pelleted at 12000×g for 10min at 4°C . Protein quantification of the supernatant ( chromatin ) has been done and 1 . 0mg protein was used for each ChIP reaction assay . To analyze the DNA proximity with CLrP , cross linked chromatin prepared was subjected to ChIP using anti-rLdCLrP and anti-actin ( positive control ) [29] . Experiments were performed using pre-immune serum and an irrelevant antibody GRP78 ( a kind gift from Dr . E . Handman ) as a negative control . Reaction mixtures were incubated for 2h at 4°C with shaking . 5mg Protein-A Sepharose beads was added to this after their pre-blocking with 10μg/ml salmon sperm DNA and 1% ( w/v ) in ChIP lysis buffer for 2h at 25°C followed by another incubation for 2h at 4°C ( with shaking ) . Washing of beads was done twice with ChIP lysis buffer , high salt lysis buffer ( same as lysis buffer , but also containing 500mM NaCl ) , and then with Tris-EDTA ( 10mM Tris-HCl , pH8 . 0 , 1mM EDTA ) . The elution of ChIP complexes were done by adding 200μl ChIP elution buffer ( 50mM Tris-HCl , pH 8 . 0 , 1% SDS , 10mM EDTA ) for two times . The combined supernatants were incubated for 5h at 65°C with 16μ of NaCl ( 5M ) to reverse-crosslink DNA and protein components . The resultant DNA has been precipitated overnight at -20°C with three times volume of absolute ethanol . Pellet obtained after centrifugation at 12000×g at 4°C for 30 min , was further resuspended in 100μl Tris-EDTA , 11 μl of 10X proteinase K buffer and 1μl proteinase K from 20mg/ml stock ( MBI Fermentas ) , followed by incubation at 55°C for 1h . DNA was recovered by silica-KI method , and analyzed by PCR for two different genes . For each PCR reaction , 1 μl purified ChIP DNA was used as the template , whereas for control PCR , DNA was isolated from 10% reverse cross linked input chromatin of actin ( positive control ) and GRP-78 , pre-immune serum ( negative control ) . To check post translational modification in LdCLrP , deglycosylation of whole cell lysate of L . donovani has been carried out with deglycosylation mix ( New England Biolabs ) as per the manufacturer’s denaturing protocol , followed by western analysis through anti-CLrP antibody . RNA of log phase promastigotes ( S1 , S2 , R1 , R2 , R3 ) ( 107 parasites ) was isolated using Tri reagent ( Sigma , Aldrich ) followed by DNase treatment and quantified . cDNA was synthesized using First-strand cDNA synthesis kit ( Fermentas ) . qRT-PCR was carried out with 12 . 5μl of SYBR green PCR master mix ( TAKARA ) , 1μg of cDNA , and 200nM primer ( Table A in S1 Text ) in a final volume of 25μl . qRT was performed with the following conditions: initial denaturation at 95°C for 10min x 40 cycles , each consisting of denaturation at 95°C for 1 min , annealing at 52°C for 1 min and extension at 70°C for 1min followed by 80°C for 10sec . A melt curve of 87 cycles was set at 52°C for 10sec . Quantifications were normalized to the Ld-actin gene . A no-template control cDNA was included to eliminate contaminations or non-specific reactions . The comparative CT method was used to calculate differences in gene expression [30] . Results are expressed as the degrees of difference between ΔCT values of test and comparator sample ( S1 ) to get ΔΔCT [31] . The normalized expression ratio was calculated as 2- ΔΔCT [32] . The expression profile of protein was analyzed in the whole cell lysate of L . donovani and fold expression was calculated considering the densitometric values of bands at ~150kDa , ~68kDa , ~41kDa and ~27kDa . The graph was plotted with the fold expression of densitometric values normalized with Ld-actin ( loading control ) . LdCLrP gene was PCR amplified from LdCLrP+pET28a ( + ) construct ( Table A in S1 Text ) and the product was cloned into Leishmania expression vector pXG-‘GFP+ at BamHI and EcoRV site [33] . S1 promastigotes of late log phase were washed with transfection buffer ( 21mM HEPES , pH7 . 5 , 137mM NaCl , 5mM KCl , 0 . 7mM Na2HPO4 , 6mM glucose ) . Transfection of the parasites with 20μg of LdCLrP+ pXG-‘GFP+ and pXG-‘GFP+ alone , was performed in a Gene Pulsar ( Bio-Rad ) . The transfectants were allowed for 24h recovery and then were selected with G418 at 5 , 10 , 20 , 50μg/mL [33] . The expression outline of the protein in S1 ( CLrP+pXG-‘GFP+ ) ( T ) and vector control i . e . S1 ( pXG-‘GFP+ ) ( VC ) was observed by the fold expression when maintained at 0μg/mL ( T0 & VC0 ) , 20μg/mL ( T20 & VC20 ) and 50μg/mL ( T50 & VC50 ) of G418 . Fold expression in different clinical isolates as compared to S1 ( WT ) was done by the densitometric study through chemidoc software ( BIORAD ) . The densitometric values were normalized with Ld-Actin ( loading control ) [29] . 2×106 promastigotes of both the transfectants were seeded in culture flask and each were counted for 8 days in Neubauer’s chamber . Growth curve was obtained as number of parasites versus days . To check the role of membrane localization of CLrP in macrophage infectivity , log phase L . donovani promastigotes of wild type ( WT ) as well as transfectants ( VC and T ) ( 5×105 cells each ) were pelleted down and incubated with anti-CLrP and pre immune serum ( PIS ) for 30 mins and then added to well plates containing 104 J774 macrophages [34] . After 4hrs free parasites were removed and allowed to grow for 24 hrs , the cells were washed two times in 1×PBS , fixed in 100% methanol followed by Geimsa staining . At least 100 macrophages were counted per well for calculating % infected macrophages . The results were plotted as the number of parasites per 100 macrophages . IC50s values for Sb ( III ) and Sb ( V ) of VC and T maintained at 0μg ( VC0 , T0 ) , 20μg ( VC20 & T20 ) and 50μg ( VC50 & T50 ) of G418 were determined in the same manner as described elsewhere [35] . Log [inhibitor] vs . response- variable slope of log dose/response data on the drug has been used to get non linear regression for 50% inhibitory concentrations ( IC50s ) of Sb ( V ) and Sb ( III ) [36] . Further one way ANOVA test and a post Tukey test were applied for statistically analyzing the data and are presented as means and standard deviations ( SDs ) of three determinations from three separate experiments . P value of less than 0 . 05 was considered significant . The study was approved by the Ethics Committee of the Kala-azar Medical Research Centre , Muzaffarpur , India ( Protocol # EC-KAMRC/Vaccine/VL/2007-01 ) with the prior consent of the human subjects and Institutional Animal Ethics Committee ( IAEC ) of CDRI for conducting the experiments on the animals ( 25/08/Para/ IAEC dated 03 . 08 . 2011 ) . The protocol and the guidelines of IAEC were bound to the National Guideline of CPCSEA ( Committee for the purpose of Control and Supervision on Experiments on Animals ) under the Ministry of Environment and Forest , Government of India .
1866bp CLrP gene has been cloned into T/A vector . The cloned sequence [http://www . ncbi . nlm . nih . gov/nuccore/JQ653307 . 1 ( Accession no . JQ653307 . 1 ) ] has a close identity of 99 . 19% to L . infantum , 95 . 18% to L . major and 92 . 04% to L . mexicana ( Table B in S1 Text ) . The motif LX3LX2L/CX2LX2LXLX2CX2L is found to be well conserved in the L . donovani homologue ( Fig B in S1 Text ) . Similar LRR repeat motif was found in amino acid sequence of protein for the homologue of this protein in L . infantum encoded by LinJ34 . 0570 gene along with other LRR repeats found in another organism including human , Arabidopsis and yeast ( Fig A and B in S1 Text ) [37 , 38] . For expression and purification of rCLrP further sub-cloning in bacterial expression vector pET28a+ was done , protein was purified and eluted at 300 mM imidazole concentration . The eluted rCLrP was ~71kDa in size as affirmed by the western blot with antibody raised in rabbit ( Fig C in S1 Text ) . Western blot analysis of SLD as well as membrane fraction of L . donovani promastigote with the polyclonal anti-rLdCLrP antibody detected band of ~68kDa , ~41kDa and ~27kDa protein in the SLD wherein the membrane fraction an additional band of ~150kDa was detected ( Fig 1A ) . To assess the purity of the SLD and membrane fraction western of both fractions was further analyzed with anti-GRP78 antibody ( Fig 1A1 ) . Dual localization of the protein in the nucleus ( Fig 1B1–1B4 ) as well as in the membrane of the parasite ( Fig 1B5–1B8 ) has been detected in immunolocalization study of CLrP . The nuclear localized CLrP was checked for its association with chromatin material with ChIP assay . The Eluted DNA of ChIP assay was probed with PCR of known genes of L . donovani i . e . Ld60sRL23a ( 60s Ribosomal L23a , 438bp ) and LdTPI ( Triose Phosphate Isomerase , 763bp ) ( Fig 1C and 1D ) . Deglycosylation of whole cell lysate of S1 resulted in a band shift of 150kDa band confirming the glycosylation of CLrP at the membrane entity ( Fig 1E ) . The expression pattern of CLrP was investigated in SAG resistant and sensitive strains of L . donovani through qRT-PCR ( Fig 2A ) . For comparison Ld-actin was employed as internal control which showed a comparable expression level among the clinical isolates ( Fig D in S1 Text ) . The study revealed two fold increased expression of CLrP in all the three resistant isolates as compared to sensitive isolates . The expression level of CLrP was further investigated at protein level ( Fig 2B and 2C ) . In all the resistant strains there was ~2 . 2 fold increase in the expression of membrane protein ( 150kDa ) , wherein the fold expression of 68kDa band increased by ~2 . 6 fold . 41kDa band was observed to be 1 . 7 fold higher in resistant isolates . 27kDa band showed no significant difference in its expression among the isolates . Ld-actin was used as loading control . To assess the role of CLrP in alteration of sensitivity profile of WT , if any , subcloning of CLrP gene was done in pXG-‘GFP+ and then transfected into WT . The immunoblot analysis with WCL of T with anti-GFP antibody exhibited prominent bands at 95kDa and 27kDa , in addition faint bands of ~68kDa and ~54kDa were also evidenced in the same blot which indicates the detachment or parting off of GFP protein from the CLrP+GFP entity after expression . ( Fig 3A ) . VC exhibited a band at mol wt of 27kDa ( Fig 3A ) . The western blot analysis of the soluble as well as the membrane of WT , T and VC parasite ( each grown at 0 μg/mL , 20μg/mL and 50 μg/mL of G418 ) with anti-CLrP revealed the protein expressions with several band patterns ( Fig 3B ) . In the western analysis of membrane fraction with anti-CLrP the 150kDa band was found to be 1 . 5 fold and 1 . 8 fold increased with 20 μg/mL ( T20 ) and 50 μg/mL ( T50 ) of G418 respectively ( Fig 3Bm ) . The intensity of the band ( 150kDa ) remained the same in WT , VC and T growing at 0 μg/mL G418 . A similar pattern was observed for 27kDa expression wherein the fold expression increased up to 1 . 2 to 1 . 9 fold under 20 and 50 μg/mL G418 pressure respectively . The expression pattern of 68kDa and 41kDa band showed no significant difference under all concentration of G418 . The expression pattern of CLrP was again studied in the soluble fraction ( Fig 3Bs ) . There is no significant difference in the 68kDa band expression at 0μg/mL and 20μg/mL of G418 , but a significant difference of 1 . 5 fold existed ( parasites grown at 50μg/mL ) when compared to the respective band of WT . The fused protein expression appeared at 20μg/mL of G418 in T , with significant difference of 2 . 2 fold expression when compared to its expression at 50μg/mL G418 ( Fig 3Bs ) . In all transfectants GFP-tagged CLrP was absent without G418 pressure . The growth curve of transfectants [T and VC] were comparable to the WT ( Fig 3C ) . Transfectants maintained at 0 , 20 , 50μg of G418 were assessed for their in vitro SAG sensitivity with Sb ( V ) and Sb ( III ) in macrophage-amastigote system and promastigotes respectively . Fig 4A and 4B exhibits the sensitivity pattern of the transfectants to Sb ( III ) and Sb ( V ) . T50 has IC50 i . e . 221 . 66±13 . 14648 for Sb ( V ) whereas that of VC has 92 . 50±5 . 79 as IC50 . T20 ( maintained at 20 μg/mL of G418 ) also revealed higher IC50 ( 168 . 60±6 . 17 ) as compared to VC20 ( 78 . 55±6 . 53 ) whereas T0 ( growing without G418 ) exhibited IC50 comparable to VC and WT . IC50 value for Sb ( V ) of T50 was 2 . 4 fold amplified than the corresponding VC , whereas for T20 it was 1 . 8 fold increased . There was a similar sensitivity pattern of transfectants to Sb ( III ) and Sb ( V ) . The sensitivity of T50 [IC50 = 96 . 44±7 . 19] to Sb ( III ) was found greater than VC50 [IC50 = 17 . 67±1 . 71] , whereas at 20μg/mL of G418 it was 67 . 76±4 . 82 and VC had similar IC50 at all concentrations of G418 . T50 showed 5 . 13 fold elevated IC50 for Sb ( III ) as compared to VC50 . Without G418 the IC50s of transfectants were analogous . The membrane localization of the protein and glycosylation of CLrP further prompted us to check its infectivity . The infectivity of the macrophage was checked with WT , VC and T parasites blocked with the anti-CLrP antibody , pre immune serum ( PIS ) was used as a control . Infectivity of S1 and VC was lowered by 2 . 2 and 2 . 5 respectively , when compared to control ( PIS treated ) . Parasitic burden ( PB ) of T was significantly higher than of WT and VC ( Fig 5 ) . Anti-CLrP antibody blocking of CLrP reduced the PB of Wt , VC and T . PIS has no significant effect on the PB . AntiGRP78 an irrelevant antibody showed no significant difference in the infectivity of T .
CLrP belongs to the protein superfamily of LRR , which shows a strong selection pressure for the conservation of this motif among several organisms [37 , 39] . In eukaryotes , the LRR motif is found in proteins encoded by the disease resistance ( R ) genes of the plant immune system [40] and by the toll and toll-like genes of Drosophila and mammals , which are implicated in innate immune responses [41 , 42] . Also internalins of L . monocytogenes have distinctive leucine-rich repeats ( LRR ) which play crucial role in interaction with human intestinal mucin MUC2 [43] . The upregulated expression of CLrP has been recently demonstrated in a SAG resistant isolate of L . donovani ( 13 ) . Persistent sensitivity profile of clinical isolates employed in this study has been checked earlier through in vitro and in vivo ( golden hamsters ) experiments , confirming the patient’s response to SAG [44] . In order to interpret the role of CLrP in resistance , gene was cloned , expressed and the recombinant protein ( ~71 kDa ) was purified to homogeneity , which exhibited very close homology to several Leishmania spp . LdCLrP has 13 . 56% homology with humans that indicates its potential to be explored as drug target . Further , bioinformatic analysis revealed that out of the whole CLrP , N-terminus part of the protein containing 248 amino acid ( ~27kDa ) shows a novel entity , whereas the remaining part of 374 amino acid ( ~41kDa ) exhibited a molecular pattern of LRR motif ( Fig E in S1 Text ) . Western analysis of Ld promastigotes WCL with polyclonal anti-rLdCLrP antibody has depicted 68kDa , 41kDa and 27kDa bands in the soluble fraction of LdS1 which apparently undergoes proteolytic cleavage probably separating the LRR motif of 41kDa from the novel 27kDa motif . A similar profile was observed in the membrane fraction of LdS1 , however , an additional prominent band of ~150kDa was also detected therein . The band patterns were not followed in the purified recombinant protein ( ~71kDa ) , revealing the protein is not autocatalytic . Earlier in the proteomics study , the protein was detected only at higher molecular weight ( ~150kDa ) in the membrane fraction [13] , and no other cleaved part of the protein was distinguished , not even of its actual size ~68kDa , in any of the membrane as well as soluble fraction indicating the membrane entity of the protein somehow playing a crucial role in SAG resistance . Deglycosylation of whole cell lysate of L . donovani depicts the post translational modification only in the 150kDa entity as a mobility shift of 150kDa band occurred on SDS-PAGE gel wherein no change is observed in the position of any other entity of the protein ( 68kDa , 41kDa and 27kDa ) . Localization study of the protein shows the dual existence of the protein in nuclear as well as in the membrane of parasite . The presence of protein in the membrane as well as the presence of ~150kDa band in western analysis reaffirms the identification of protein in the membrane fraction in proteomics studies [13] . Nuclear existence of CLrP was positively probed by Ld60sRL23a and LdTPI confirming its interaction with DNA . LxLL motif of CLrP ( 299th position ) , is well known for nuclear receptor cofactor interactions involved in transcriptional regulations [45] . The above evidences pointed out towards the proteolytic cleavage and post translational modification of the protein throughout the cell . CLrP transcript contour revealed ~2 fold expression of CLrP transcripts in resistant isolates , verifying the differential proteomics finding [13] . As in Leishmania the gene expression is rarely regulated at the transcriptional level and L . donovani strains have been shown to generate extensive aneuploidy [46 , 47] , the increase in the CLrP transcripts may therefore be attributable to the ploidy phenomenon . Besides the differential expression of proteins one other phenomenon of drug resistance includes gene amplification / duplication , an example of which is well known in cultured mammalian cells for the acquisition of resistance to MTX3 ( Methotrexate ) [48 , 49] . Leishmania also adapted themselves for resistance to several cytotoxic compounds often via amplifying or deleting a number of specific loci coding for either drug targets or drug transporters [50 , 51 , 52] . CLrP’s expression profile further depicts the significantly different expression of 150kDa , 68kDa and 41kDa bands among the clinical isolates wherein 2 . 1 fold increased expression of 150kDa band was observed in resistant isolates as compared to sensitive ones justifying the proteomics results [13] . On the other hand , the 68kDa and 41kDa band were over-expressed by 2 . 5 and 1 . 7 fold respectively in the resistant clinical isolates , which were not noticeable in the proteomics results [13] . The 27kDa band also does not reveal any significant difference among the isolates . As the 150 kDa band is the only glycosylated entity of protein , which is differently overexpressed , it may be argued to have some association with the development of resistance . Nevertheless , it will be too early to directly correlate glycosylation pattern with resistance mechanism as there may be several other players in the pathway ( s ) associated with the development of SAG resistance in Leishmania [44 , 53 , 54] . To check the potential of CLrP to alter the SAG sensitivity profile of parasite , CLrP was over-expressed in WT [33] . Similar to WT , a pattern of several bands was observed in GFP-CLrP overexpressed cells by western blotting using anti-GFP antibodies , evidencing the proteolytic cleavage of CLrP . This ensures the CLrP expression in its original form . Transfectants were further analyzed in vitro in macrophage-amastigote model [Sb ( V ) ] as well as in the promastigotes form [Sb ( III ) ] for their SAG sensitivity profile . Our results indicate that overexpression of CLrP as GFP conjugate in WT Leishmania promastigotes as well as amastigotes has decreased the SAG sensitivity and their sensitivity pattern towards Sb ( III ) and Sb ( V ) were almost parallel . Thus , increasing concentrations of CLrP and parasite modulation towards SAG revealed its potential in SAG resistance . In Leishmania , CLrP being a member of the LRR super—family is ideally placed on the membrane of the parasite to interact with the macrophages , and there is mounting evidence that the parasite uses a variety of ligands to interact with the host such as membrane bound PPG and LPG [55] . Both of them share LRR motif which plays a crucial role in macrophage invasion [56] . Apparently , the parasite has learned to exploit its LRR motif in host interactions . Hence , the protein macrophage interaction consequences were further investigated for its infectivity . Blocking CLrP in S1 and VC lowered its infectivity to more than 2 folds when compared with control ( PIS treatment ) , whereas treatment of T50 with anti-CLrP also decreased its infectivity . VC showed a similar pattern of infectivity that of WT , but infectivity of both decreased 1 . 6 times as compared to T50 . These results indirectly indicate that overexpression of CLrP is related to the development of resistance . SSG ( Sodium Stibogluconate ) resistant parasites have been reported to increase the parasitic burden in vivo but its direct link to any molecular marker still remains to be investigated [57] . Also , LRR has been reported to provide a scaffold implicated in assisting several pathogens-associated molecular patterns as well as surface receptors , therefore over-expression of the protein may facilitate the interaction of parasites to macrophages and thus its invasion by the parasites . The results indicate that the CLrP upregulation is associated with SAG resistance , but it may be linked indirectly . There could be several possibilities for induction of resistance due to CLrP overexpression , such as its association or competition with the cascade ( s ) involved . The results further seek a longitudinal study to trace the routes of the protein journey and to decipher the consequences of post-translational modification at the functional level and its actual association with SAG resistance . This study further demonstrates the ability of CLrP in increasing the infectivity of parasites which can smoothen the parasite invasion in macrophages and further modulating parasite behavior for SAG . Validation of the present finding depends on the down-regulation of CLrP , but multi-copy of the CLrP dispersed in the genome of Leishmania assures the infeasibility of knocking out its gene . Hence , this study exclusively depends on the consequences of upregulation of CLrP . As the Indian subcontinent largely depends on combination chemotherapy ( SAG with other drugs ) , the parasites seem to have learned to modulated itself in resisting these combinations under several laboratory conditions [58] . Understanding of resistance mechanism is therefore essential to strengthen our arsenal of chemotherapeutic strategies .
|
Leishmania causes complex of pathologies called Leishmaniasis and among the several forms visceral leishmaniasis is the precarious one as being fatal , if left untreated . Emergence of resistance against several antileishmanials particularly Sodium Antimony Gluconate ( SAG ) has severely battered the therapeutic strategy against VL and the resistance mechanism is still vague . Thus , to apprehend the underlying mechanism , previously , a differential proteomics of SAG unresponsive versus SAG sensitive isolates of L . donovani was done wherein overexpression of Cysteine Leucine Rich protein ( CLrP ) , a member of Leucine rich repeat superfamily , was observed . To scrutinize its involvement in the SAG resistance mechanism , which is till date not investigated , the characterization of CLrP was carried out which revealed its post-translational modification along with its dual existence in the nucleus and in the membrane of the parasite . Further investigation using a ChIP assay confirmed its DNA binding potential . Over-expression of CLrP in sensitive isolate of L . donovani significantly decreased its SAG sensitivity . CLrP overexpressed parasites have increased infectivity . These results point out towards the CLrP’s contribution to antimonial resistance in L . donovani by facilitating parasites growth through macrophages . Further studies are required to depict CLrP as a potential drug target to strengthen the present arsenal against L donovani .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Over-Expression of Cysteine Leucine Rich Protein Is Related to SAG Resistance in Clinical Isolates of Leishmania donovani
|
Brain-wide interactions generating complex neural dynamics are considered crucial for emergent cognitive functions . However , the irreducible nature of nonlinear and high-dimensional dynamical interactions challenges conventional reductionist approaches . We introduce a model-free method , based on embedding theorems in nonlinear state-space reconstruction , that permits a simultaneous characterization of complexity in local dynamics , directed interactions between brain areas , and how the complexity is produced by the interactions . We demonstrate this method in large-scale electrophysiological recordings from awake and anesthetized monkeys . The cross-embedding method captures structured interaction underlying cortex-wide dynamics that may be missed by conventional correlation-based analysis , demonstrating a critical role of time-series analysis in characterizing brain state . The method reveals a consciousness-related hierarchy of cortical areas , where dynamical complexity increases along with cross-area information flow . These findings demonstrate the advantages of the cross-embedding method in deciphering large-scale and heterogeneous neuronal systems , suggesting a crucial contribution by sensory-frontoparietal interactions to the emergence of complex brain dynamics during consciousness .
The brain is a complex system with restless activity dynamics [1] . These dynamics are generated and maintained by interactions among variety of neural populations inside the brain . How brain-wide interaction produces specific dynamics underlying cognitive functions is a central question in neuroscience [2 , 3] . While recent theories propose that complex dynamics generated by integrative and recurrent interactions among brain areas form a universal foundation of conscious brain state [4] , it is unknown how complex dynamics support consciousness and what specific circuits in the brain generate the complex dynamics . Recent experimental results of high-dimensional recordings highlighted the importance of various different aspects of dynamics in individual cortical areas underlying sensory , motor , or cognitive functions [5–9] . A challenging point in characterizing brain-wide dynamics is that the complexity and diversity , which many reductionist model-based approaches neglect , may be an irreducible feature supporting cognitive processes [4] . A dilemma is that overly realistic and detailed simulations often require a number of unknown parameters and can obscure physiological principles [3 , 10] . It is , therefore , not straightforward to instantiate an appropriate reductive model that capture dynamical complexity and diversity across multiple brain areas [3] . Another stream of research directly evaluates statistical features in data . Although some approaches evaluate complexity of neural activity ( e . g . the correlation dimension [11] or the algorithmic complexity [12] ) and others evaluate neural interaction ( e . g . the functional connectivity [13] or the Granger causality [14] ) , there are few methods that relate these two features in a consistent interpretable manner . Here we introduce an alternative approach that can address how brain-wide nonlinear neural interaction generates complex dynamics , with minimal assumptions . We focus on a generic mathematical property termed delay-embedding [15 , 16] in nonlinear dynamical systems , which enables topological reconstruction of global attractor dynamics based on local temporal sequences . We developed an extended delay-embedding method to provide an accurate estimate of dynamical complexity , i . e . , the dimensionality of a reconstructed attracter , under the presence of confounding factors in real data , based on a random projection technique . Relying only on the topological aspect of reconstructed attractor dynamics , the method allows us to shortcut several specific model assumptions and a-priori dimensionality reduction of data , and yet provides a clear explanation for how nonlinear interactions generate complexity of dynamics . We first illustrate the principle of our approach with simple models ( where we know the true underlying dynamical system ) , showing that it reliably estimates interactions among observed nodes and the dynamical complexities that reflect the interactions , despite variations in signal timescale and noise . Next , we apply the method to wide-field electrophysiological recordings from awake and anesthetized monkey cortex , reporting a novel hierarchical organization of cortical areas as a universal correlate of conscious brain state . In this hierarchy , the dynamical complexity increases along the directed cross-area interaction from the visual to frontoparietal areas . This can reconcile two contrasting views on conscious process—concerning whether conscious processes are localized to specific “workspace” areas [17 , 18] or distributed across the entire system [4 , 19 , 20]—from the viewpoint of information coding by dynamical systems . The high dynamical complexity in the workspace area is accompanied by structured interactions across cortical areas under conscious states . Remarkably , we find that brain-state distinction is reflected in the high-dimensional temporal sequences , but not in snapshots of neural activity , emphasizing the functional importance of dynamics . Finally , we discuss possible physiological mechanisms that account for the current results , proposing a likely ( but to date not thoroughly examined ) contribution of bottom-up interaction ( that originates from sensory cortex ) to generate the state-dependent complexity of frontoparietal dynamics .
In dynamical systems , causally interacting variables ( say , electrode signals x , y and z ) share a trajectory in the state space , ( x , y , z ) [Fig 1A , ( 1 ) ]—that is , each time point corresponds to a location on the common set of realizing states ( “attractor” ) in this space . It means that once we specify the time evolution of a variable , the other variables’ behaviors are also constrained via this set ( e . g . , the dynamics of multiple electrodes can be estimated based on a single electrode’s dynamics ) . Indeed , mathematical theorems guarantee that the temporal sequence of a single variable [“local observation” , Fig 1A , ( 2 ) ] has sufficient information about the underlying high-dimensional system dynamics . This is sufficient for complete reconstruction of the original global attractor topology [Fig 1A , ( 3 ) ]; the global state has one-to-one mapping to a local temporal sequence , or equivalently , a location within the delay-coordinate state-space [Fig 1A , from ( 3 ) to ( 1 ) ] . This property is called “embedding , ” a general principle in deterministic dynamical systems . Based on this property , a protocol for inferring causation in real systems was recently proposed in ecology [21] , using nearest-neighbor forecasting techniques [22 , 23] . It allows us to detect causal interactions and their directionality because a “downstream-node” generally has sufficient information to estimate the dynamics of the “upstream-nodes” but not vice versa ( more details are described in the next section ) . However , this protocol alone does not tell us how the causal interactions shape emergent dynamics , which is the central issue in the present study ( and in many other neuroscience studies ) . In addition , we show that the estimate of dynamical complexity is severely affected by signal timescale , when we reconstruct the state-space based on the standard delay-coordinate ( as in the previous work [21] , which focused on detecting causality but not on accurately quantifying the attractor complexity ) . Therefore it is problematic to naïvely apply the original protocol to a highly heterogeneous system such as the brain [e . g . , different cortical areas have different timescales of dynamics [24]] . We extended the previous approach to link the causal interaction and the dynamical complexity via an embedding-based relationship between time-series variables ( hereafter called cross-embedding ) . The key idea is to utilize the fact that the embedding is invariant to the coordinate transformation—instead of the standard delay coordinates , here we use a random projection of them to reconstruct the attractor dynamics [Fig 1A , ( 4 ) ] . It is also intuitively understandable that virtually any random projection does not break the topology of the attractor ( except for special cases ) , which is a consequence of random projection theory [25] . A critical requirement for embedding is that the dimensions , d , of the state-space for reconstruction have to be greater than the attractor’s dimension , dA ( Fig 1B; see the figure legend for additional comments on the relationship between d and dA ) . We make use of this necessity condition for quantifying the effective attractor dimensionality in local dynamics , which is relevant to interaction from other areas . Before moving on to real data analysis , the following two sections provide simplified examples to describe the general cross-embedding principle and demonstrate how the proposed method dissociates the effective dimensionality from other factors such as the timescale of dynamics . The major innovation of our cross-embedding framework is in the simultaneous reconstruction of directionality in interaction ( simply denoted by directionality hereafter ) and a new dynamical complexity measure ( denoted by complexity ) . This is achieved by extending the causality estimation protocol by Sugihara et al . [21] to the dynamical dimensionality domain . It leads to a core theoretical prediction that we refer to as downstream complexity: “dynamical complexity increases along directed interaction . ” Note that we mention the “downstream” in terms of functional network interaction , not directly associated with the biological or anatomical definition of “downstream areas . ” To have an intuition about this , let us consider a simple model example ( Fig 1C ) , where two recurrent systems ( e . g . , brain areas x and y ) interact in an asymmetric manner ( y affects x but not vice versa ) . The aim is to identify their interaction based on two observed time series [x ( t ) and y ( t ) ] from each system ( Fig 1D ) ( which corresponds to inferring interaction among cortical areas based on the electrode signals recorded from them ) . Since the interaction is only from y to x , the history of x has information about y , but not vice versa . We call it “x embeds y” in this paper . [More formally , we have an one-to-one mapping from attractor manifold Mx to My , which are respectively reconstructed from the time series x ( t ) and y ( t ) as random projections of the delay-coordinate reconstructions . ] This generally happens if and only if y causally influences x; it means , conversely , we have no one-to-one mapping from My to Mx ( Fig 1E ) . It is a direct consequence of a mathematical theorem proven by Stark [26] . Furthermore , armed with this mathematical insight , it is predicted that the downstream attractor , Mx , is generically more “complex” than the upstream one , My . This is because , as x is perturbed by y , x’s dynamics live in a state space not of x alone , but of ( x , y ) , which generally has higher dimensionality than y alone . Despite the fact that we only assume the validity of delay embedding here , we have a strong theoretical prediction: asymmetry in causal interaction should be accompanied by a hierarchy of dynamical complexities , which we directly and quantitatively test using electrophysiological recordings from monkeys , as described later . For quantitative assessments of the cross-embedding relationships , we introduced two measures summarizing the cross-embedding relationships: directionality and complexity ( the full details of the analyses are described in Materials and Methods ) . First , we defined the directionality of interaction from y to x by the difference in embeddedness of y by x minus that of x by y ( Fig 1F ) . Here the embeddedness between the variables is quantified , essentially by seeing whether the time indices of nearby data points on attractor Mx are also clustered on My ( Fig 1E; see also Materials and Methods ) . This embeddedness takes a value in between 0 and 1 . Given a sufficient data length , a positive value of directionality from y to x indicates an asymmetric causal interaction directed to x [21] . Next , we introduced a new measure , complexity , of interaction from y to x by the required dimensions of Mx to identify the corresponding locus on My . The embeddedness is given as a function of reconstructing dimensions and therefore the complexity was defined as the dimensions at which the embeddedness is saturated ( Fig 1F; Materials and Methods ) . Coordinate randomization is crucial to avoiding systematic errors in complexity that might arise due to a finite data size ( see next section ) . All the analyses are based on pairwise node-to-node relationships , and thus computations are scalable to high-dimensional networks . As the previous study [21] did , we allow the data to include some noise , although the embedding theorem in the mathematically rigorous sense assumes deterministic dynamical systems . In the practical data analysis below , we do not assume that the data are noise-free , but expect that the data contain some noticeable components that reflect underlying deterministic dynamics . If we could reconstruct the attractor dynamics by the randomized delay-coordinate state space , even under some approximations , we are able to estimate the directional information flow and dynamical complexity . Note that incomplete recovery of upstream dynamics does not necessarily indicate a breakdown of our deterministic assumption; the present cross-embedding method is generally data-expensive because required data length exponentially increases with the dimension of a downstream attractor . Nevertheless , it turns out that our method can still extract meaningful features of an underlying dynamical system even with finite data size and in the presence of small noise , as shown in the next section ( see also S4 Fig ) . In realistic setups that include observation noise and limited data points with various length of signal autocorrelation ( or “timescale” ) , quantifying the attractor complexity is not necessarily straightforward , as already pointed out in dynamical systems studies [27 , 28] . Therefore , we examined the robustness of our complexity measure against variation in signal timescales and the level of observation noise . If the estimate is systematically affected by those signal properties , it complicates the interpretation of the complexity measure because different brain areas are known to exhibit different timescales [24] and experimental observations inevitably involve noise . We compared the present complexity measure to the five other related measures ( Materials and Methods ) : ( i ) correlation dimension [29] , ( ii ) dimensionality derived from principle component analysis ( PCA ) , ( iii ) permutation entropy [30 , 31] , ( iv ) algorithmic ( Kolmogorov-Chaitin ) complexity [32 , 33] , and ( v ) the optimal embedding dimension using the standard ( non-randomized ) delay-coordinate [using the algorithm by [21] , which was not specifically designed to measure the complexity] . The measures ( i ) and ( ii ) were applied to the attractor manifold that was reconstructed within the random delay-coordinate for each electrode signal . Using the simulated dynamics with an artificial system ( Materials and Methods ) , we confirmed two points . First , the present complexity measure is almost orthogonal to the timescale ( Fig 2A ) and observation noise ( Fig 2B ) . We also confirmed that the randomization of the coordinates worked for avoiding the overestimation of complexity for relatively slow dynamics ( Fig 2C ) . Second , the conventional complexity measures can be sensitive to the signal properties such as timescale or observation noise . In particular , while the measures ( i ) - ( iv ) have been widely used in neuroscience [11 , 12 , 34–39] , we need to be careful with their interpretation in the presence of such confounding factors . In contrast , our new measure is advantageous for fair comparison of dynamical complexity across heterogeneous brain dynamics . Together , our new measure based on cross-embedding has advantages in providing a reliable measure of dynamical complexity under potential heterogeneity in timescale or observation noise . Note that the primary aim of introducing coordinate randomization is the accurate and robust estimation of complexity , but not of directionality . Directionality of interactions can be estimated accurately without using coordinate randomization if we select the embedding dimensions appropriately . We also confirmed the robust estimation of complexity and directionality under the variation in system noise ( S4 Fig ) . Based on the cross-embedding framework , we explored the structure of physiological large-scale neural dynamics that concern distinct brain states . We recorded the large-scale brain dynamics from monkey electrocorticography ( ECoG ) recordings from most of the hemispheric cortical surface ( Fig 3A and 3B ) . We performed the aforementioned cross-embedding analysis on the resting-state ECoG signals ( Fig 3C–3F ) , as well as comparing conscious and unconscious brain states . The complexity and the directionality were first computed for each electrode pair ( 128×128 = 16384 pairs in total ) . Although the embeddedness hardly reaches the theoretical maximum value of one ( 1 ) with real data under finite recording time , the asymmetry of embedding relation between an electrodes pair was often clear and statistically significant ( Fig 3E and 3F ) . To characterize inter-areal interactions , we then averaged all the complexity and directionality observed within each set of areas . Notably , the complexity varied depending on the targets , but not on the source areas of directed interactions ( Fig 3G ) ; this is consistent with the feature that the current complexity measure converges to the attractor dimensions of the target with sufficient data ( see Materials and Methods ) . Overall , the awake state showed higher complexity than the anesthetized state ( Fig 3H; p<10−5 , sign test , paired samples ) . On the other hand , the directionality depended both on source and target areas ( Fig 3I ) . Although the lower and higher visual cortices tended to show similar properties under the awake condition ( Fig 3I , left ) , no clear structure was observed in the anesthetized state ( Fig 3I , right ) . The structure of directionality changed drastically between the awake and the anesthetized conditions , which was further confirmed by the small correlation between these values between the two conditions ( Fig 3J; ρ = 0 . 01 , p<10−4 , Spearman rank correlation test ) . We also found that the structure of directionality was less clear when we used an insufficient number of reconstruction dimensions ( e . g . , d = 1 , Fig 3K ) , indicating the importance of dynamical history , not just momentary activity levels . Remarkably , the present cross-embedding analysis extracts a correlate of the brain-state change much more clearly than the conventional characterization based on correlation . For example , in terms of correlation among momentary electrode signals , the awake and anesthetized states had similar network structures ( Fig 3L ) , but with weaker correlations in the awake state ( p<10−5 , sign test , paired samples with absolute correlation coefficients ) . This was confirmed by pair-wise correlation coefficients correlated between the awake and anesthetized states ( Fig 3M; ρ = 0 . 18 , p<10−5 , Spearman rank correlation test ) ; the correlation between the two conditions was significant even if the analysis was restricted to electrode pairs straddling different cortical areas to remove trivial distance dependency ( ρ = 0 . 52 , p<10−5 , Spearman rank correlation test; S5A Fig ) . These results are in contrast to the drastic structural change in the cross-embedding-based directionality values between the two conditions ( Fig 3I and 3J ) . Difference in the dynamics was also not obvious when the dimensionality of the multi-electrode signal was reduced by PCA ( S2 Fig ) . Furthermore , we found that electrode pairs showing only weak correlation ( correlation coefficient ≈ 0 ) can have strong asymmetric interaction in terms of directionality values ( Fig 3N ) ; in both awake and anesthetized conditions , the absolute values of correlation coefficients , and directionality for individual electrode pairs , were overall not strongly correlated ( -0 . 0394<ρ<-0 . 0307 , 95% confidence interval of Spearman rank correlation ) although they showed weak negative correlation within the three behavioral conditions ( Awake-eyes-closed , ρ = -0 . 02 , p<10−29; Anesthesia ( ketamine-medetomidine ) , ρ = -0 . 04 , P<10−46; Anesthesia ( propofol ) , ρ = -0 . 02 , P<10−4; Spearman rank correlation test ) . These results demonstrate that cross-embedding and correlation extract different aspects of dynamics . Specifically , brain-state dependent cross-areal interaction detected by the directionality was not reflected in conventional correlation-based statistics . This fact may explain why some studies [37 , 40–42] but not others [13 , 43] have reported changes in the resting-state functional connectivity depending on arousal level . To relate the cross-embedding properties to the electrode loci on the cortex , we averaged the complexity and directionality values over all source areas ( Fig 4 ) . It demonstrated three major findings: First , dynamical complexity was much higher in frontoparietal areas than occipital areas in the awake state ( Fig 4A ) . Second , we found the information directionality in the awake cortex from occipital to frontoparietal areas , indicating that latter areas were downstream of the functional network ( Fig 4B ) . This covariation of the directionality and complexity measures is consistent with the prediction of the embedding theorem that receiving additional information via directed interaction increases dynamical complexity . Third , these characteristic structures of directionality and complexity seen in the awake cortex disappeared under anesthesia ( Fig 4C and 4D ) . In particular , the complexity in the frontoparietal cortex was drastically reduced , resulting in a more uniform distribution of complexity across cortical regions ( Fig 4C ) . The electrodes within the same area generally shared these trends ( Fig 4E ) . To summarize , the distinct brain states seemed to be selectively characterized by the downstream complexity ( the higher dynamical complexity downstream of directed interaction ) of a whole-brain functional network . Next , we questioned how universally these dynamical properties were observed across experimental setups and conditions . For example , the complexity of dynamics could reflect only the conscious-unconscious difference ( i . e . , the level of consciousness ) in the brain state , or could be affected by other cognitive experiences , such as visual stimulus or body movement ( i . e . , the contents of consciousness ) [44] . To dissociate these possibilities , we compared the current results with two additional experimental conditions: “Awake-eyes-open” ( where monkeys freely viewed their environment , but their body movements were restricted ) and “Food-reaching conditions” ( where monkeys were seeing and grabbing the food by moving their arms , Materials and Methods ) . We found that the difference in complexity between inside and outside visual cortex distinguished the awake from the anesthetized brain state across individual animals , regardless of substantial differences in behavioral or sensory experiences ( Fig 5A ) . These results suggest that high complexity in those areas reflects the conscious brain state rather than specific cognitive processes induced during visual or motor experiences ( cf . table in Fig 6A ) . Notably , high complexity was consistently accompanied by downstream areas with high directionality across the conditions ( Fig 5B ) . The increase in complexity in conscious conditions was significant not within the visual cortex ( i . e . , the upstream; p>0 . 2 , Wilcoxon rank sum test ) , but only outside of the visual cortex ( i . e . , the downstream; p<10−5 , Wilcoxon rank sum test ) . Moreover , further analysis revealed that the complexity and directionality were significantly correlated across the electrodes in all conditions ( Fig 5C; 0 . 49<ρ<0 . 62 , p<10−86 , Spearman rank correlation test ) . This suggests that the co-occurrence of positive directionality ( network downstream ) and high complexity is a fundamental property shared between conscious and unconscious states , while the range of complexity and spatial structure of interactions differ between these brain states . As we mentioned earlier , this relationship between the complexity and network directionality , although not reported in the previous neuroscientific studies ( to our best knowledge ) , agrees with the theory of directionally-coupled deterministic dynamical systems ( as shown in Fig 1C ) , where the downstream system must have more complex dynamics than the upstream system . Additional analysis supported this view by revealing that the reduction in the top-down interaction ( from outside to inside visual cortex ) dominates the increase in bottom-up interaction ( from inside to outside visual cortex ) during the change from the anesthetized to awake-eye-closed condition ( Fig 5D; p<10−5 , sign test , paired samples ) . To summarize , we found ( i ) that an increased complexity downstream is generally observed in neural dynamics ( regardless of brain state ) , and ( ii ) that this property emerges across the global cortical network under conscious brain states , which is likely to be a result of enhanced bottom-up interaction ( we further examine this point in a later section ) . It was also confirmed that details and operations of the analyses ( spatial differentiation , denoising , and orthogonality of the coordinate ) did not change the results ( S1A–S1D Fig ) . Finally , the high complexity observed in the network downstream is not an artifact of unreliable estimation of complexity for weakly interacting nodes because the resulting complexity structure was qualitatively and quantitatively unaffected when we restricted our analysis to the strongest interacting nodes ( S1E Fig ) . The results were cross-validated , thus random variability in data does not explain our results . Together , these results revealed a robust structure of the large-scale cortical hierarchy under consciousness , where the visual cortex was located upstream exhibiting simple dynamics , whereas the frontoparietal cortex was located downstream exhibiting complex dynamics ( Fig 6B and 6C ) . Therefore , we conclude that the downstream complexity revealed by the present cross-embedding analysis reflects a universal property of neuronal dynamics that can dissociate the conscious and unconscious states . In previous studies , the distinct brain states have often been described in terms of their typical timescales of neural dynamics: e . g . , slow oscillations in anesthesia/sleep vs . high-frequency fluctuation in wakefulness [45–47] . The present findings on the hierarchy and brain-state dependency of complexity are in contrast to the characterization based on timescale because our complexity measure was designed to be invariant to the timescale . Although its timescale-invariance was confirmed with the simulation ( Fig 2A ) , one might still consider the possibility that the change in complexity reflects the difference in the timescale of neural dynamics . To directly exclude this possibility , we replicated the analyses using a different size of the unit delay in the state-space reconstruction . If the complexity depended on timescale , this manipulation should yield different results . However , we found that the main results , including the spatial distributions of complexity and directionality , were robust to the substantial change in the unit delay size ( Fig 7A ) . It should be noted that the use of the different unit-delay can affect the embedding performance ( Fig 7B ) and the absolute value of directionality ( changed by 0 . 36 times ) , possibly due to the limitation in data length [an interpretation of why 20-Hz unit-delay yielded better embedding performance is that signals with moderately high frequency ( e . g . , >5 Hz ) had large contribution to reconstructing the attractor topology under the noise and limitation of data length] . Nevertheless , the average directionality from visual to other areas was still larger in the awake than anesthetized conditions ( p<10−7; Wilcoxon rank sum test ) , suggesting the qualitative robustness of the present results . In addition , the complexity values themselves were quantitatively maintained across the different unit-delay sizes ( Fig 7C ) , which is reasonable since the complexity reflects the attractor dimensionality . Although one might speculate that the complexity values were robust to the unit delay if they reflected a temporally scale-free property such as ratio of time-constants between source and target , it is unlikely since the data suggest that the complexity primarily depended on the target areas , not on the combination of source and target ( Fig 3F ) . Therefore , our results strongly support the view that this emergent hierarchy of complexity accurately reflects the hierarchy of attractor dimensions , which is invariant to the timescale of dynamics . If timescale is not critical , what determines this hierarchy ? Our data suggest that it is the dimensionality of an attractor underlying observed temporal sequences . For example , the correlation structure ( which was derived from the momentary signal values ) was not sensitive to the difference in brain states , as we have seen before in this paper ( Fig 3L ) . This insensitivity is not caused by the difference in analytical procedure between correlation and cross-embedding because the brain state-dependency was not clear even with the cross-embedding analysis when we limited the embedding dimension to one ( Fig 3K ) . Rather , it is critical to take into account the high-dimensional pattern of temporal sequence , not momentary “snapshots . ” Altogether , the current results suggest that the brain-state change from the conscious to unconscious condition is reflected in relatively complex dynamics of neural activities , which is not reduced to conventional statistics such as timescale or correlation . What causes the increased complexity in the awake frontoparietal cortex ? As directionality from visual to frontoparietal cortex also increased under the awake conditions , a likely mechanism would be that the increased bottom-up interaction caused the complex frontoparietal dynamics . Alternatively , it is possible that increased complexity in the frontoparietal cortex caused the change in the estimate of directionality in our cross-embedding analysis . To examine this , we explored what mechanism can ( or cannot ) account for the present results , using a simplified network model ( Fig 8 ) . Here , we consider three large clusters of the nervous system: two ( visual and frontoparietal cortices ) correspond to those observed in the present ECoG experiment while the last one is assumed an unobserved system ( e . g . , subcortical structure ) . The contributions of cortical and subcortical mechanisms to the anesthesia-induced loss of consciousness is currently a central subject of debate [48–52] . The unconscious ( anesthetized ) state was modeled by a network having only weak long-range coupling among the three clusters , but is moderately coupled within individual clusters ( Fig 8A ) , which follows the experimental findings [40] . We simulated individual nodes with a linear-nonlinear model including self-feedback ( Materials and Methods ) . Note that our focus here is not on modeling details of nervous systems but on illustrating the basic mechanisms of how complexity and directionality change depending on the network properties . The increase in complexity could be caused by three different mechanisms ( Fig 8B–8D ) : ( i ) increases in the bottom-up interaction from the visual to the frontoparietal cluster , ( ii ) increased extra-cortical input from the subcortical to the frontoparietal cluster , and ( iii ) change in intrinsic connectivity within the frontoparietal cluster . Simulation confirmed that all the three mechanisms lead to an increase in complexity within the frontoparietal cluster ( Fig 8B–8D , fourth column ) . However , they differ in directionality related to the bottom-up interaction from the visual to the frontoparietal cluster ( Fig 8B–8D , fifth column ) . Only the first model ( with increased bottom-up interaction ) predicted the increase in the visual-to-frontoparietal directionality , replicating the present results . In contrast , the other two models showed no increase in this directionality measure , which is reasonable because those models do not assume any mechanistic change in terms of the visual-to-frontoparietal interaction . Adding the baseline correlation among nodes did not affect these qualitative results . It should be noted that the results here do not reject partial contributions by ( ii ) or ( iii ) , or the possibility that the changing complexity reflects the variation in indirect interactions between the visual to the frontoparietal areas via subcortical system . Whether direct or indirect , these simulation results suggest that the sensory cortex contributes to shaping the conscious cortical dynamics , posing important constraints on the models of brain-state dependent nervous activities .
We have introduced an embedding-based computational method that bridges cross-area interactions and emergent dynamics with a clear theoretical underpinning , but without making specific or reductive assumptions about brain systems . The method is based only on generic dynamical properties and is readily applicable to systems that are difficult to model . Indeed , it has successfully captured features that can be easily overlooked by conventional reductionist approaches , such as spatiotemporal variations in complexity itself ( e . g . , Fig 4A and 4C ) . In addition , unlike PCA or power spectrum analyses in temporal frequency domains , this method can fully exploit nonlinear and directed interactions between brain areas . One caveat of the present method is that it may not always detect veritable underlying causal interactions if the data length is restricted . Despite this limitation , the embedding-based characterization of interactions is advantageous over existing prediction-based methods , such as Granger causality or transfer entropy , because it is not confounded by self-predictability [21] ( S3B Fig ) . That is , a node’s own historical trace is generally sufficient to predict its future regardless of the presence or absence of causal interactions from other nodes in a deterministic system . It is important to discriminate the embedding in a rigorous mathematical sense from that for practical data analysis . Embedding in a rigorous mathematical sense is not defined in the presence of noise . However , in practical applications , embedding-based analysis can extract underlying dynamical features with a resolution limited by the noise . Indeed , the method successfully estimated the dimensionality of an attractor and the structure of interactions in the system ( as we showed in S4 Fig , and as demonstrated by Sugihara et al . [21] ) unless noise is too large . We believe that the present cross-embedding based analysis yields new insights from the practical viewpoint , in purpose of analyzing deterministic aspects of in neural dynamics . This view is supported by the fact that the present results are fully consistent with a mathematical property of a deterministic system , in which the cortical areas estimated as network downstream have higher complexity than the upstream areas . This structure was highly robust to different versions of analytic procedures and parameters ( S1 Fig ) . Based on this method , we simultaneously characterized the large-scale cortical interaction and the dynamical complexities embedded in individual area activities . It revealed that the awake brain has a hierarchical structure of the dynamical complexity , where the frontoparietal areas had more complex dynamics than visual areas . Intriguingly , this hierarchy was linked to the directed cross-area interaction from visual to frontoparietal areas . To our best knowledge , this is the first study reporting clear cortical hierarchy in terms of dynamical complexity , as well as its relationship to the global cortical interaction . Moreover , we found that this hierarchy was universal across different behavioral/sensory conditions and disappeared after the loss-of-consciousness induced by either of two different anesthetization methods . These results indicate that this hierarchical structure is correlated with the level of consciousness rather than its specific contents reflecting perception or action . The present results demonstrate that the correlation structure was similar ( correlated ) between awake and anesthetized conditions at a whole-brain scale ( Fig 3L and 3M ) , albeit with slightly larger absolute correlation strength in the anesthetized condition . On the other hand , some previous studies show that the brain states can be differentiated based on correlation-based analysis [53 , 54] . Although the exact reason behind this discrepancy is beyond the focus of this work , one potential reason is the difference in natures of signals to be analyzed . For example , the correlation based on slow fMRI signal might correspond to a nonlinear sum of the ECoG-based correlation across a wide range of time-delays . We do not exclude the possibility that the correlation-based measure can differentiate the two conditions . Instead , we emphasize that the correlation and directionality are conceptually distinct measures—we can intuitively expect that two electrodes can be highly correlated despite week directionality in their interaction if the two sites are bidirectionally interacting with the equal strength . Moreover , we observed that the area pairs showing only weak correlation can have large value in the directionality measure , and vice versa ( Fig 3O ) . This result was also robust to the exclusion of intra-areal electrode pairs ( S5B Fig ) . From the theory and modeling , we proposed that the increasing complexity at the frontoparietal cortex under awake conditions is likely to be a consequence of enhanced ( direct or indirect ) bottom-up interaction from the visual cortex . Another way of accounting for the current data is to assume that the frontoparietal cortex is always downstream of the visual cortex , and that the former simply reflects input from the latter in the unconscious state . In this case , we do not observe any hierarchy among them under unconscious state since the visual and frontoparietal cortices have exactly the same dynamical complexity . When the transition to the conscious state causes some additional input or self-recurrence in the frontoparietal cortex , it no longer simply reflects the visual cortex and can have more complex dynamics . Although theoretically interesting , this hypothesis requires several physiologically questionable assumptions . First , there is currently no evidence supporting the idea that the visual cortex drives the entire cortical dynamic in the unconscious state . Second , it assumes no local recurrent connection within the frontoparietal area under unconsciousness , which is not consistent with the experimental findings ( e . g . , [40] ) . Nevertheless , we do not reject such the possibility in this paper . To directly test which mechanism is the most likely , we can stimulate the visual cortex ( using transcranial magnetic stimulation or other methods ) and compare the efficacy of bottom-up signal transmission from the visual to frontoparietal cortex , among different brain states . If the frontoparietal cortex simply reflects the visual cortex under unconsciousness , the former should precisely reflect the stimulation pattern in the latter . On the other hand , if the bottom-up interaction varies among the distinct brain states ( as we hypothesized in the above ) , the stimulation in the visual cortex would have fewer effects on the frontoparietal cortex in the unconscious than the conscious state . Whichever hypothesis is true , our present finding emphasizes the link between frontoparietal dynamical complexity and bottom-up interaction . It inspires a novel debate on the organization of the state-dependent cortical dynamics by suggesting that the interaction originates from the sensory cortex , which seems to be of much greater importance than previously expected . In a current theory of consciousness , the high complexity and integrative interaction have been proposed as two important aspects of the conscious experience [4] , but their relationship remains elusive . A previous study that quantified dynamical complexity [11 , 38] did not focus on spatial distribution of the complexity measures across brain areas , or on , its relationship to information transfer . Conversely , while previous studies have reported that functional connectivity can differentiate conscious and unconscious brain states [40 , 55–57] , this was not linked to local dynamical characteristics . Hence , the present result establishes for the first time tight relations between these two concepts by quantifying them in a unified framework , and by revealing structural correlates of cognitive brain states . Our findings on an area-specific consciousness analogue in the frontoparietal cortex is concordant with its proposed role as a workspace for information sharing [17] . However , the present embedding-based approach conceptually advances this model by suggesting that its highly complex dynamics are the substrate of embedding the information across wide-spread brain areas , specifically in conscious states ( Fig 4B ) . Our data and model suggest that , although the frontoparietal area is a key region , its characteristic dynamical complexity is likely to be a consequence of integrating brain-wide information including the sensory cortex . In this sense , the present results are also consistent with the view that finds the basis of consciousness in brain-wide interaction , rather than in a specific area [4 , 19 , 20] . A particularly unique aspect of the cross-embedding concept is that the information integration does not necessarily require observation of spatially distributed dynamics , but can be achieved by localized complex dynamics that embeds the global state . More generally , the cross-embedding property highlights the universal importance of neuronal dynamics in information coding . Embedding describes the observability of information in one brain area by another area , fulfilling a prerequisite for the actor-critic models of brain systems [58] . While a similar concept has been proposed in terms of anatomical connectivity as an internal monitoring property [59] , it has not been linked to neuronal dynamics . Complex dynamic in one area that embeds a high dimensional attractor is generally capable of differentiating a large number of states that characterize its entire “upstream-area” . For example , the embedding dimensions outside of the visual cortex were typically larger than one ( Fig 6B ) , indicating that substantial information is embedded not only in single momentary states , but also in complex temporal sequences of neuronal activity . How the brain utilizes such temporally encoded information remains an open problem , nevertheless , temporal sequences in non-stationary attractor dynamics have been suggested to subserve behaviorally relevant information [60 , 61] . Based on the current delay-embedding framework , one can extract more in-depth information about an attractor’s topology [15 , 16] and its characterization would facilitate an understanding of dynamic information coding and messaging across brain areas . Together , these results suggest that computational approaches based on dynamical cross-embedding will be generally applicable to study complex neuronal interactions in large-scale physiological dynamics of the brain .
All the experimental procedures were approved by the RIKEN Ethics Committee . We used ECoG data that were recorded from four monkeys . 128 channel ECoG signals were recorded from a wide field of a hemisphere for each monkey brain , covering the lower and higher visual cortices , temporal cortex , parietal cortex , motor and somatosensory cortex , premotor cortex , and lateral and medial prefrontal cortices; the medial prefrontal cortex was recorded from two of the four male monkeys ( three Macaca fuscata and one Macaca mulatta ) . The details on the recording apparatus have previously been reported [62] , and part of the dataset is available online ( http://neurotycho . org ) . For the food-reaching experiment , the data were extracted from periods where the monkeys were moving their arms . Chewing noises were eliminated from the main analysis by excluding the post-food-grabbing periods ( typically 2–15 s ) where the channel-averaged signal was higher than the 75th percentile of the whole data . We did not apply pre-filtering to the ECoG signals before the subsequent analyses , except for eliminating the 50±2 . 5 Hz component , which included line noise due to the recording apparatus . We separated the recorded signals for the anesthetization experiment into blocks corresponding to awake-eyes-open , awake-eyes-closed , and deep-anesthesia periods , and applied embedding analysis within each block . The total data length was 400 s . Data were analyzed using random coordinate cross-embedding ( see next section for details ) , which we developed to reliably measure attractor dimensions . The inter-subject average was obtained by averaging the results across electrodes within individual areas of each subject and experiment , before they were averaged across subjects and experiments . The data from the medial prefrontal cortex were analyzed for the two subjects for which they had been recorded . In this section , we explain the technical details of the cross-embedding algorithm while the problem formulation and the theoretical principle themselves are presented in Results . The algorithm was based on the convergence cross-mapping ( CCM ) method , which was proposed in ecosystem analysis [21] as an extension of nonlinear state-space reconstruction [22 , 23 , 63] . Although the original method was shown to characterize the causal interaction in ecological system with 2~5 variables , its applicability to analysis of dynamical complexity in large-scale and heterogeneous systems ( such as the brain ) has been not clear . The CCM alone does not tell us how the causal interactions are related to the emergent neural dynamics and functions . Moreover , we find that the state-space reconstruction based on the original delay-coordinate is severely affected by the heterogeneity in signal timescale across the system . Therefore , we developed a method to simultaneously quantify the causal interaction and dynamical complexity , by utilizing random projection in the reconstructed state space . We quantified the pairwise dynamical relationships based on a delay-embedding theorem in nonlinear dynamical systems including external forces [26 , 64–66] . We first reconstructed the attractor manifold for each node x in the delay-coordinates , x_tdmax= ( xt , xt−τ , … , xt− ( dmax−1 ) τ ) , where dmax represents the maximum number of dimensions ( number of delay-coordinates ) to be considered , t is the time point , and τ is the unit delay length . Although the original protocol uses this delay-vector for subsequent nonlinear forecasting analysis [22 , 23 , 63] , we found that it makes the results vulnerable to the time-scale heterogeneity in the system . We avoided this problem by projecting delay vector x_tdmax to a randomized coordinate space by multiplying a square random matrix , R , from the left to obtain a transformed vector: xtdmax= R x_tdmax , which is equivalent to considering the time series convolved with preset random filters [In our dataset , the distribution of random numbers did not have much effect on the results , at least for widely-used distributions ( e . g . , Gaussian or uniform distribution ) ; in the current analysis , we used Gaussian distribution centered at zero with standard deviation of one] . A d-dimensional delay vector xtd was constructed by selecting the first d ( ≤ dmax ) components of xtdmax . The topological embeddedness for signal y by another signal x was quantified based on the correlation coefficient , ρyt , y^ ( xtd ) , between the true ( yt ) and forecast ( y^t ( xtd ) ) signals , where y^t=∑t′ s . t . xt′d∈B ( xtd ) w ( | xt′d−xtd | ) yt′ with the k-nearest neighbor set B ( xtd ) of xtd in the delay-coordinate space . We set k = 4 , weight w ( | xt′d−xtd | ) ) =e−| xt′d−xtd |/Σxt'd∈B ( xtd ) e−| xt′d−xtd | , and |xt'd−xtd| as the square distance between xt'd and xtd in the data analysis . Note that ρyt , y^ ( xtd ) =1 means perfect embedding , where observing xtd completely eliminates the uncertainty of estimating state y and the conditional probability density distribution P ( yt|xtd ) becomes a delta function . The relative embeddedness was derived by subtracting the embeddedness with d = 1 from that with d = d* to extract the coupling of complex temporal structures , which is not reflected in correlation . All the analyses were conducted after normalizing the signal variance to one , in order to avoid heterogeneity in signal amplitude affecting the estimation accuracy . All estimation accuracy was measured with two-fold cross-validation , in which 1000 uniformly subsampled time points ( t ) from the latter-half of the data were predicted based on the embedding relation to the former-half , to avoid the confounding effects of over-fitting [27] . This procedure of random projection was introduced to avoid systematic biases in estimating the embedding dimension based on finite data sets . Adjacent components ( e . g . , xt , and xt−τ ) typically have some non-zero correlation in a standard construction of delay-coordinate vector x_tdmax , depending on the signal timescale . A greater number of dimensions may be required to untangle the attractor manifold due to non-negligible correlation among the delay coordinates , leading to an overestimation of complexity . In addition , since the magnitude of overestimation depends on the signal timescale , the estimated complexities will be contaminated by variations in signal autocorrelation [27 , 67] . This problem does not arise after the effect of signal autocorrelation [67 , 68] is eliminated by applying the above-mentioned coordinate-randomization procedure . Indeed , the simulation demonstrated that estimated complexity typically became more accurate with the randomization procedure ( Fig 2C ) . The intuition behind random matrix multiplication is as follows . When the studied signal has relatively slow dynamics , the signal values at two adjacent time points , x ( t ) and x ( t-τ ) , can be similar if unit delay τ is small compared to the time constant of the signal . In this case , observing x ( t-τ ) in addition to x ( t ) does not practically add much information about the system state . As a consequence , small τ can lead to overestimation of dimensionality in standard delay coordinates . Hence , estimated dimension sensitively depends on the choice of τ in pre-randomized delay coordinates . Using the randomized delay coordinates alleviates this problem . In this case , each observation is linear sum of observations at across widely different time delays , where a weighted sum of observations at intermediate delays always well characterizes the system dynamics . As we explained above , summed observations at small delays do not add much information but are not harmful . Summed observations at large delays can add variability if the system is very noisy ( because noise can accumulate in time ) but generally does not bias characterization of system’s state . Hence , the choice of the maximal delay requires much less tuning than the choice of τ in the standard delay coordinates . The complexity of dynamics concerning pairwise interaction was quantified using minimum embedding dimension d* that yielded ≥ 95% of optimal estimation accuracy maxd ρy , y^ ( xtd ) , which is an extension of the false-neighbor method [27 , 28 , 68] to the present cross-embedding . ( Our ECoG results were robust when complexity was instead defined as the minimum embedding dimension that yielded 90% or 100% of the optimal estimation accuracy . ) Note that in real data with a limited sample size , the accuracies sometimes did not have plateaus but have a peak followed by a gradual decrease [23] . In this case , complexity corresponds approximately to a dimension that maximizes estimation accuracy . Estimating from the upstream to downstream nodes is generically incomplete ( correlation coefficient < 1 ) because the latter does not embed the former; even in such cases , estimation accuracy depends on the embedding dimension , and we can quantify complexity with the same procedure . While the method cannot estimate the attractor dimension of the target electrode if the cross-embedding method detects no interactions , this only occasionally happened in our analyses of the ECoG data . As expected from this characteristic , our results were qualitatively and quantitatively the same when we limited our analyses only to most strongly interacting pairs of electrodes . The directionality in interaction from x to y was quantified by the difference between optimal estimation accuracies: ρy , y^ ( xtd ) −ρx , x^ ( ytd ) . This takes a positive value when there is directed interaction from x to y . Note that stochastic components within an upstream system is not embedded by downstream dynamics since the embedding requires deterministic nature of dynamics . However , the directionality of causality between as pair of nodes should be detected by a node asking the other node whether there is any asymmetry of observability ( conditional entropy ) of a node by the other node [21] , even under some stochasticity due to system noise ( S4 Fig ) . It is worth noting that detecting a clear causal relationship by Transfer entropy [69 , 70] , Granger causality [71] or spectral Granger causality [72 , 73] is not straightforward unless the system’s dynamical properties are well known , due to the confounding effects of phase delay [74] or self-predictability in deterministic dynamics [21] . In the present analysis , we used unit delay τ = 20 ms and embedding dimension dmax = 30 in the main analysis . The optimal unit-delay depends on noise/dynamical property , and again , has to be selected generally depending on each dataset to be studied . Note that our point in Fig 7 is not to show that 20-ms delay is “the optimal” but to demonstrate that the complexity and relative relationships in terms of network upstream-downstream is robust to selection of unit delay . The embedding dimension , dmax , does not affect the results in theory , as long as our methodological assumptions hold and if we set dmax sufficiently larger than true dimensions of attractor . In practice in the presence of noise in the system , however , the embedding performance slowly falls if we choose too large dmax . However , this dmax dependency is not very sensitive because intermediate delays always provide good signal about system’s state and , while large delays gradually add some variability , they do not bias the estimate . Since we do not know a-priori the maximum dimensionality of attractors in general , dmax has to be selected ad hoc manner depending on data to be studied . In this study , we set dmax = 30 empirically by observing that the accuracy of nearest-neighbor mapping ( measure of “embeddedness” ) saturates , at most , around dimensions 10 . In the current data , we used total data points in each segment at order of 10000 to obtain robust results . When we decrease the number of data points ( e . g . , 6000 or 2000 ) , the accuracy of nearest-neighbor mapping were degraded , although the relative patterns ( e . g . , asymmetry in mapping accuracy ) tended to be preserved . Regarding how to estimate the sufficient data length , one rough estimate is based on the dimension of an attractor , d . For example , if n data points per dimension are needed for accurate prediction , the total points required would be ~nd . While the number of required data points also depends on spatial roughness of the attractor , this provides a reasonable estimate . A more sophisticated estimate may be possible by fitting the attracter using a manifold learning algorithm . Although we normalized the data before the present analyses to make the results comparable to the correlation analysis and to rule out the possibility that the heterogeneity in signal amplitude variability affected the results , the embedding analysis is , in principle , scale invariant . The mathematical concept of embedding relies only on the topological property of attractor dynamics , which has by definition nothing to do with the scale of the signal amplitude ( e . g . , variance ) . In this sense , the principle of method is scale invariant . In practical implementation , however , the result of the analysis can be affected by the scale of signal under a limited data length . For a concrete example , we used a nearest neighbor model to construct a mapping from a reconstructed attractor to the other one . In this case , the results depend on how we define the “neighbor” of a data point . Unless we have infinitely long data , the data points are distributed on the attractor with some sparcity . Keeping the number of total data points fixed , the larger signal scale leads to the smaller absolute density of data points ( number of data points in a unit cube/sphere in the state space ) . If the neighbor of a data point were defined with the unit cube/sphere around it , the number of other data points inside the neighbor would be affected by the absolute density of data points , and thus the result would depend on the scale of signal . On the other hand , in the present method , we defined the neighbor not by unit cube/sphere but by the simplex spanned by k-nearest points of each data point with keeping k constant . In this method , the number of the data points to construct nearest-neighbor mapping does not depend on the scale . We also confirmed that the normalization procedure was not critical in the present finding ( S1E Fig ) . The correlation dimension was computed with the algorithm proposed by [29]; the PCA dimension was quantified by the number of dimension explaining 90% of the all variance; the permutation entropy was derived using the algorithm proposed by [31]; the algorithmic complexity was derived with the method proposed by [75]; the complexity with standard delay-coordinate was given as the optimal embedding dimension in the algorithm proposed by [21] . The example dynamics in Fig 1C–1F were generated based on the following coupled Rössler oscillators: Tξ˙1=− ( η1+ζ1 ) ( ξ2+11 ) /2 , Tξ˙2=− ( η2+ζ2 ) , Tη˙1=ξ1+0 . 2η1 , Tη˙2=ξ2+0 . 22η2 , Tζ˙1=ζ1 ( ξ1−5 . 7 ) +0 . 2 , Tζ˙2=ζ1 ( ξ1−5 . 6 ) +0 . 2 , where the dots indicate temporal differentiation; x→ = ( ξ1 , η1 , ζ1 ) , y→ = ( ξ2 , η2 , ζ2 ) , and ( xobs , yobs ) = ( ξ1 , ξ2 ) in Fig 1C . Parameter T controls the timescale of the dynamics . The signals were generated using a fourth-order Runge-Kutta method with a time step of 0 . 01 , and subsampled with a time step of 0 . 1 in the embedding analysis , where we used τ = 4 and dmax = 20 . In Fig 8 , the brain dynamics were simulated with the following difference equations: xi ( t ) =f ( ∑jJijxj ( t−1 ) ) +bi ( t ) , Where xi ( t ) is activity of ith node at time step t , Jij is the connectivity from jth to ith node , and f ( x ) = x exp[3 ( 1 –x ) ] . The activation function , f , is interpreted as a qualitative model of balanced local circuit , although our aim here is not to provide a detailed model of nervous systems but to demonstrate the basic mechanisms with a simplified example . We set Jii = 1 for all i . Inter-node connectivity ( Jii ) was randomly sampled from a normal distribution [mean = 0 , and standard deviation = 0 . 1/N ( within cluster ) or 0 . 5/N ( between cluster ) , where N is the total number of nodes in the whole network] , or set as zero , depending on the models; in the model 3 , the connectivity within the frontoparietal cluster was doubled , except for the self-feedback ( Jii ) . bi ( t ) is the bias input; to add the baseline correlation , we assumed the bias term irrelevant to brain state , and modeled it with a sinusoidal function with random amplitudes and phase . We ran the simulation for 1000 time steps , with 5 nodes for each clusters for illustrative purpose ( increasing the number of node had not much effect on the simulation results; note also that the number of nodes does not directly reflect the number of ECoG electrodes ) . After the generating simulated dynamics , we performed the same cross-embedding analysis as the one used in the main ECoG results ( except for that here we used the unit time-delay of 1 time step ) .
|
Advances in recording technologies have enabled the acquisition of neuronal dynamics data at unprecedented scale and resolution , but the increase in data complexity challenges reductionist model-based approaches . Motivated by generic theorems of dynamical systems , we characterize model-free , nonlinear embedding relationships for wide-field electrophysiological data from behaving monkeys . This approach reveals a universality of inter-areal interactions and complexity in conscious brain dynamics , demonstrating its wide application to deciphering complex neuronal systems .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Untangling Brain-Wide Dynamics in Consciousness by Cross-Embedding
|
Dogs , with their breed-determined limited genetic background , are great models of human disease including cancer . Canine B-cell lymphoma and hemangiosarcoma are both malignancies of the hematologic system that are clinically and histologically similar to human B-cell non-Hodgkin lymphoma and angiosarcoma , respectively . Golden retrievers in the US show significantly elevated lifetime risk for both B-cell lymphoma ( 6% ) and hemangiosarcoma ( 20% ) . We conducted genome-wide association studies for hemangiosarcoma and B-cell lymphoma , identifying two shared predisposing loci . The two associated loci are located on chromosome 5 , and together contribute ~20% of the risk of developing these cancers . Genome-wide p-values for the top SNP of each locus are 4 . 6×10-7 and 2 . 7×10-6 , respectively . Whole genome resequencing of nine cases and controls followed by genotyping and detailed analysis identified three shared and one B-cell lymphoma specific risk haplotypes within the two loci , but no coding changes were associated with the risk haplotypes . Gene expression analysis of B-cell lymphoma tumors revealed that carrying the risk haplotypes at the first locus is associated with down-regulation of several nearby genes including the proximal gene TRPC6 , a transient receptor Ca2+-channel involved in T-cell activation , among other functions . The shared risk haplotype in the second locus overlaps the vesicle transport and release gene STX8 . Carrying the shared risk haplotype is associated with gene expression changes of 100 genes enriched for pathways involved in immune cell activation . Thus , the predisposing germ-line mutations in B-cell lymphoma and hemangiosarcoma appear to be regulatory , and affect pathways involved in T-cell mediated immune response in the tumor . This suggests that the interaction between the immune system and malignant cells plays a common role in the tumorigenesis of these relatively different cancers .
Lymphoma and angiosarcoma are both malignancies of the hematological system , originating from lymphocytes and hematopoietic stem cells , respectively . Lymphomas are a heterogeneous group of diseases , estimated to be the eighth leading cause of human cancer deaths in the US in 2014 [1] . The majority is classified as non-Hodgkin lymphoma ( NHL ) and , among these , diffuse large B-cell lymphoma ( DLBCL ) and follicular lymphoma are the most common [2] . Angiosarcoma is a highly aggressive cancer accounting for 1–5% of adult spontaneous sarcomas [3 , 4] but its rarity limits genetic studies . Equivalents of both lymphoma and angiosarcoma occur spontaneously in pet dogs . Sixty-eight percent of golden retrievers , one of the most popular dog breeds in the US , die from cancer [5] . Approximately 13% of golden retrievers develop lymphoma [5] , and approximately 50% of these cases are of B-cell origin , within which the most common subtype is the canine equivalent of DLBCL [6–9] . Twenty percent of golden retrievers develop hemangiosarcoma [5] , which is clinically and histologically similar to human visceral angiosarcoma [10 , 11] . Large-scale population-based epidemiological studies and several genome-wide association studies ( GWAS ) of human lymphoma cases have shown increased familial risks and germ-line risk factors in the human population [12–15] . These studies provide clear evidence for heritable predisposing mutations for B-cell NHL subtypes in certain human populations , but also point to the heterogeneous nature of B-cell NHL . In this study , we have used the relatively limited genetic diversity in golden retrievers to facilitate the identification of susceptibility loci . Dogs have been used successfully to map complex diseases including systemic lupus erythematosus , obsessive-compulsive disorder and osteosarcoma [16–19] . Dogs spontaneously develop diseases that are also common in humans , and , as dogs receive modern health care , have recorded family structures and share the living environment with humans , they make an excellent model to study these diseases [8] . In addition , due to recent breed creation , purebred dogs have megabase-sized haplotypes and linkage-disequilibrium ( LD ) blocks , allowing GWAS in dogs to be performed with 10-fold fewer SNPs than in humans [20 , 21] . Power calculations and proof of principle studies have shown that 100–300 cases and 100–300 controls can suffice to map risk factors contributing a 2–5 fold increased risk in dogs [16 , 20] . Strong bottlenecks in the evolutionary history of the dog have led to genetic homogeneity within breeds , allowing for relatively efficient identification of germ-line mutations , and allowing for effective clinical trials to study the effect of those germ-line mutations on outcome or response to therapy [22] . Here we present the combined results of GWAS of B-cell lymphoma and hemangiosarcoma in 356 golden retrievers . While originally performed as two separate studies , the major associated regions colocalized , which prompted us to combine the datasets . Our analysis revealed two major loci on canine chromosome 5 , associated with both diseases and together accounting for ~20% of the disease risk in this cohort . Neither associated region is explained by coding mutations , but RNA-Seq analysis of differential gene expression in B-cell lymphomas suggests that the risk alleles at the two loci significantly alter expression of genes involved in the T-cell mediated immune response . These results highlight the importance of regulatory mutations , as well as the interaction between the immune system and malignant cells in cancer development , and may explain why these two different diseases unexpectedly share the same predisposing germ-line risk factor .
To search for inherited risk factors predisposing to hemangiosarcoma in golden retrievers , we performed GWAS by genotyping 148 hemangiosarcoma cases and 172 cancer-free golden retrievers >10 years old using the canineHD Illumina 170k SNP array [23] . Since dog breeds contain high levels of cryptic relatedness and complex family structures , it was necessary to apply a method to control for the population stratification [24] ( Methods ) , and a final dataset of 142 hemangiosarcoma cases and 172 controls , and 108 , 973 SNPs was used for the association analysis . The quantile-quantile plot ( QQ-plot ) showed an inflation factor λ of 0 . 959 , indicating that the population stratification had been well controlled ( Fig . 1A ) . SNPs with p-values below 1 . 45×10−4 significantly deviate from the expected distribution , and as the Manhattan plot of p-values estimated by GCTA [25] shows , the main association signal comes from chromosome 5 , with other less significantly associated peaks on chromosomes 11 and 13 ( Fig . 1A ) . For the chromosome 5 peak , the top SNP ( regression odds ratio ( ORregres ) = 1 . 23 , p-value = 1 . 09×10−6 ) was located at 29 , 892 , 306 bp , 85 kb upstream of TRPC6 and in strong LD ( r2 > 0 . 8 ) with 10 other significantly associated SNPs ( Table 1 ) . The four most associated SNPs are all in high LD with each other . The next three significantly associated SNPs are all located within the STX8 gene , around 33 . 8–34 . 1 Mb; two more significantly associated SNPs are in LD with SNPs at 33 Mb ( Table 1 ) . A separate GWAS for B-cell lymphoma in golden retrievers was performed using 41 cases and the same 172 controls as for the hemangiosarcoma study . Since the case sample size was relatively small , stricter cutoffs were used to control for population stratification , but due to careful selection of controls based on pedigrees , all of the 41 cases and 172 controls , and 109 , 579 SNPs remained in the dataset for the association analysis . The QQ-plot revealed that although no SNPs reach genome-wide significance for this small dataset of cases , there are three SNPs with p-values below 1×10−4 that deviate from the null distribution . These three SNPs are located on chromosome 5 at 33 . 4–33 . 9 Mb , and have ORregres of 1 . 36–1 . 39 . The hemangiosarcoma dataset showed a strong association on chromosome 5 . The B-cell lymphoma signal was considerably weaker and no SNP reached genome-wide significance , but the association signals overlapped with the hemangiosarcoma signal on chromosome 5 . Therefore , we combined the datasets to assess if the two diseases had common predisposing risk factors . After quality and relatedness control , 183 cases ( 142 hemangiosarcoma cases and 41 B-cell lymphoma cases ) , 172 controls , and 109 , 407 SNPs were analyzed for the association . The QQ plot deviated from the null distribution at 2 . 2×10−4 , identifying 35 significantly associated SNPs ( best p-value = 4 . 63×10−7 , Fig . 1C , Table 1 ) , of which 20 were located on chromosome 5 between 29 . 6 Mb and 34 . 1 Mb . Sixteen SNPs out of these 20 SNPs were identical to the significantly associated SNPs from the hemangiosarcoma analysis , all of them with more significant p-values in the combined study , confirming their importance in B-cell lymphoma . The associated SNPs in this region clustered in two peaks located 4 Mb apart . The top SNPs in the two regions were located at 29 , 892 , 306 bp and 33 , 854 , 327 bp , with p-value of 4 . 63×10−7 and 2 . 66×10−6 , respectively . Importantly , the two loci located 4 Mb apart were tagging different risk haplotypes . For the combined dataset , the top SNP in each region shows high LD ( r2 > 0 . 8 ) with SNPs within the same peak , but low LD ( r2 < 0 . 2 ) to the associated SNPs in the other peak ( Table 1 , Fig . 2 A , D , S1–S3 Fig . ) . To further confirm that these loci are not in linkage , we conducted conditional association analyses , which included the genotype of the top SNP of one peak as a covariate ( Methods ) , and the results also indicate that the two peaks are independent signals ( S1–S3 Fig . ) . Detailed analyses of the associated risk haplotypes in the separate and combined datasets shows that the 29 Mb risk alleles are mostly predicting hemangiosarcoma predisposition , although the association is stronger in the combined dataset compared to hemangiosarcoma alone . The 33 Mb region is associated with disease in both datasets , and interestingly , the top SNPs differ in the hemangiosarcoma and combined , vs the B-cell lymphoma dataset ( Table 1 , Fig . 2D , E ) . The respective top SNP from each analysis , located 8 . 7 kb apart , are in high LD ( r2>0 . 8 ) with several SNPs around them , but not with each other ( r2 = 0 . 45 , combined dataset ) . They are tagging two different haplotypes in the 33 Mb region . SNPs in the B-cell lymphoma risk haplotype are not significantly associated with hemangiosarcoma ( Table 1 ) and p-values drop in the combined analysis compared to B-cell lymphoma alone , suggesting that this is an independent haplotype only predisposing to B-cell lymphoma . The SNPs of these two haplotypes are interspersed along the genome ( S1 Table ) . To define the exact risk haplotypes and their boundaries , r2-based clumping analysis was performed by PLINK [26 , 27] , and r2-based block definition and association analysis was performed by Haploview [28] ( Methods ) . These analyses identified risk and non-risk haplotypes in both loci . In the 29 Mb region two associated haplotype blocks were seen: a 9-SNP block ( “29 . 7Mb-shared” ) spanning 182 Kb , and a 4-SNP block ( “29 . 9Mb-shared” ) spanning 26 kb ( Table 2 , Fig . 2 ) . The risk haplotypes largely appear in the same dogs , suggesting the possibility of selection in this region ( S2A Table ) . In the 33 Mb region , a 5-SNP haplotype block ( “33Mb-shared” ) spanning 266 kb was identified in the combined dataset ( Table 2 , Fig . 2 , S1 Table ) . An additional , B-cell-lymphoma-specific haplotype was identified at 33 Mb ( “33Mb-BLSA” ) , which consists of 4 SNPs spanning over 887 kb . An r2-based haplotype analysis of the chromosome 5 region including both peaks using the combined dataset showed no long-range haplotype spanning two peaks , thus further confirming the independence of these two peaks . Notably , the BLSA-33Mb risk haplotype is in LD ( r2 = 0 . 75 ) with 4 SNPs in the 29 Mb region ( Fig . 2E ) . Those SNPs are interspersed with the top SNPs at 29 Mb identified in the combined analysis . The risk haplotypes at the 29 Mb locus have a high frequency ( Fig . 2C , S3 Table ) ; almost half of all cases are homozygous for the risk haplotype as compared to 25% in the control dogs for the 29 . 7Mb-shared risk haplotype . The frequencies are similar for the 29 . 9Mb-shared haplotype . For both haplotypes , the percentage of dogs homozygous for the risk allele is considerably larger among the cases compared to controls ( S3 Table ) . In contrast , the risk haplotypes at the 33 Mb locus have a much lower frequency; only 7% in dogs with B-cell lymphoma and 4% in dogs with hemangiosarcoma are homozygous risk , while about a third are heterozygous for the 33Mb-shared risk haplotype . In comparison , not a single control dog is homozygous risk , and one in five are heterozygous for this risk haplotype ( Fig . 2 , S3 Table ) . The disparate frequency of the risk alleles at the two loci also supports a hypothesis of two distinct risk factors . The separate B-cell lymphoma risk haplotype ( 33Mb-BLSA ) is also rare; 2% of B-cell lymphoma and 1% of hemangiosarcoma cases are homozygous for this haplotype and 34% and 11% , respectively , are heterozygous . In contrast , no control dog is homozygous for the risk haplotype and 8% are heterozygous for the risk haplotype . The 33Mb-BLSA risk haplotype appears to be tagging a newer variant that occurred on the existing , shared risk haplotype . Every 33Mb-BLSA risk allele is carried with a 33Mb-shared risk allele , such that dogs homozygous for the 33Mb-BLSA risk haplotype are also homozygous for the 33Mb-shared risk haplotype , and all dogs heterozygous for the 33Mb-BLSA risk haplotype have at least one copy of the 33Mb-shared risk haplotype . This is a significant deviation from what would be expected if the two haplotypes were unlinked ( pChiSq = 7 . 3×10−50 ) ( S2A Table ) . To determine the proportion of disease risk explained by the genotypes of these two loci , we performed a restricted maximum likelihood ( REML ) analysis using GCTA software [25] ( Methods ) . All the autosomes together explain 43 . 2% ± 17 . 1% of the phenotype ( p-value = 5 . 6 × 10−4 ) , and the SNPs within 25–40 Mb on chromosome 5 explain 22 . 4% ± 10 . 7% ( p-value = 2 . 7 × 10−5 ) of the phenotype in the combined analysis ( S4 Table ) . These results suggest that the two risk loci on chromosome 5 account for ~20% of the phenotypic variance of these cancers in the golden retriever breed . Two approaches were taken to evaluate potential candidate genes within the regions of association . In summary , no protein-coding changes associated with either risk or non-risk haplotypes were found , but the risk haplotypes at both loci had a strong effect on the expression level of genes that play important roles in the immune response , especially T-cell mediated responses . Specifically , we first examined the coding exons of genes within the most strongly associated regions for risk-haplotype-concordant non-synonymous germ-line mutations using ~40x coverage of Illumina sequence from nine individuals ( Methods ) . At the 29 Mb locus , KIAA1377 harbored two SNPs that would lead to amino acid substitutions if they were translated but they are likely intronic , ANGPTL5 has one coding mutation , and TRPC6 has two mutations in the 5’ UTR ( S5 Table ) . For NTN1 , STX8 , and WDR16 , genes near the 33 Mb locus , one non-synonymous mutation was found in WDR16 and two in NTN1 ( S5 Table ) . However , none of those mutations was associated with the risk haplotype while deviating from the mammalian consensus . Secondly , since no coding changes were identified , we investigated whether the risk haplotypes were associated with transcriptional changes in tumors . We generated RNA-Seq data from 22 hemangiosarcoma and 22 B-cell lymphoma samples . The gene expression in the hemangiosarcoma samples reflected their high levels of contamination by stroma cells , which is typical for hemangiosarcoma tumors , and no conclusions could be drawn . The B-cell lymphoma samples were more homogeneous , and were grouped into “higher-risk” and “lower-risk” categories depending by how many copies of the risk allele they possessed . Briefly , for the 29 Mb locus , 12 dogs homozygous for the risk haplotype were designated as the higher-risk group and compared to the lower-risk group consisting of mostly heterozygous dogs ( eight heterozygous dogs and two dogs with no copy of the risk haplotype ) . The same individuals were higher-risk or lower-risk for both 29 . 7Mb-shared and 29 . 9Mb-shared haplotypes . The results show that the risk haplotype at 29 Mb had a clear cis-regulatory effect ( Fig . 3A , Table 3 , S6 Table ) , and most significantly altered the expression of TRPC6 , the closest gene to 29 . 9Mb-shared ( logFCrisk = −7 . 46 , p-value = 7 . 45 × 10−17 , FDR = 1 . 37 × 10−12 , Table 3 , Fig . 3A ) . The expression of the TRPC6 transcript was virtually undetectable in the tumors of dogs in the higher-risk group ( all dogs are homozygous for the risk haplotype ) . TRPC6 encodes a transient receptor potential channel , which mediates calcium ion ( Ca2+ ) influx [29] and plays a significant role in T-cell activation through at least two pathways; 1 ) the PLCγ pathway regulated by the T-cell receptor , and 2 ) the PI3K pathway that is mediated by co-stimulation through CD28 [30 , 31] . For the 33 Mb locus , a higher-risk group of mostly heterozygous dogs ( one homozygous and five heterozygous for the 33Mb-shared risk haplotype ) were compared to the lower-risk group of 16 dogs carrying no copy of the 33Mb-shared risk haplotype ( Methods ) . Five of the six higher-risk dogs carried the 33Mb-BLSA risk haplotype , which is consistent with the genotyping data where all dogs carrying the 33Mb-BLSA risk haplotype also carry the 33Mb-shared risk haplotype ( S2B Table ) . Having at least one copy of the 33Mb-shared risk haplotype at 33 Mb significantly changed the expression levels of 100 genes located elsewhere in the genome ( Table 3 , S6 Table ) . None of the 100 genes were within 1 Mb of any of the significantly associated loci in either the hemangiosarcoma , B-cell lymphoma , or combined GWAS . Unsupervised clustering ( S4 Fig . ) did not group the samples relating to their haplotypes , suggesting that the differential gene expression associated with the risk haplotypes is not the key differentiator of tumors . A knowledge based Ingenuity Pathway Analysis ( IPA ) [32] of the 100 genes based on the 33Mb-shared haplotype identified a large number of common biological functions including differentiation , activation and cell-to-cell signaling in the immune system ( S7 Table ) . The 33Mb-shared risk allele was shown to mediate overall decreases in immune cell activation ( Fig . 3B , S7 Table ) . Eighteen significant canonical pathways were identified ( S8 Table ) , and of the top four pathways ( p-value < 0 . 005 ) three directly implicate T-cell responses . Several upstream regulators , including IL-2 ( z-score = −2 . 97 , p-value = 5 . 62×10−14 ) , CD3 ( z-score = 2 . 02 , p-value = 3 . 34×10−13 ) , TCR ( z-score = −2 . 83 , p-value = 6 . 31×10−13 ) , ZBTB7B ( z-score = 2 . 21 , p-value = 1 . 13×10−9 ) and IL-15 ( z-score = −2 . 63 , p-value = 2 . 96×10−9 ) were identified , all of which play an important role in the activation , acquisition of effector functions and lineage differentiation of T-cells [33–35] ( S9 Table ) .
GWAS of human DLBCL using thousands of human patients have detected a few candidate loci , which together only account for a small fraction of the genetic risk [12 , 14 , 15] . For human angiosarcoma , no GWAS has been performed due to the rarity of the disease . Here we performed GWAS for canine B-cell lymphoma and hemangiosarcoma using fewer than 400 dogs for both diseases combined , and identified two loci of strong effect accounting for about 20% of the disease risk . This study illustrates the advantages of mapping a complex trait within a canine breed , in which a small number of risk factors with a strong effect are present as a result of the strong bottlenecks at breed creation , and the relative genetic homogeneity within the breed . The fact that one of the two risk factors on chromosome 5 ( 29 Mb ) is very common in the U . S . golden retriever population may relate to the use of popular sires . It also could be an example of a strong genetic risk factor accumulating either through drift or selective breeding for a nearby locus . It was unexpected and remarkable to discover that two rather different cancers , B-cell lymphoma and hemangiosarcoma are linked to the same inherited risk factors , as shown by the increased strength of association when combining the two datasets . While surprising , this could be explained by previous observations that hemangioblasts have the ability to generate both hematopoietic stem cells and endothelial cells [36] , and that canine hemangiosarcoma is likely to originate from hemangioblasts [37] . Another remarkable finding is that only two loci appear to explain 20% of the total disease risk . This may be partly due to the homogenous genetic background present within this dog breed , but may also result from the effect size of the individual risk factors . While the risk loci on chromosome 5 explain as much as 20% of the risk , no coding mutations were identified . Instead , we found that the risk haplotypes of both loci are significantly associated with gene expression changes , implying that the mutations in regulatory regions play an important role in cancer , which is often the case in other common diseases [38] . Several candidate loci fall just above or below the significance threshold in our current analyses . Since all autosomes together can explain an additional ~21% of the risk , incorporation of additional cases and controls in the future will likely identify more risk loci with genome-wide significance . In this context we note that the 41 B-cell lymphoma cases alone produced a relatively weaker signal for the chromosome 5 locus at 29 Mb , suggesting that for this high-frequency risk allele at ORallelic ~2 . 0 , a higher sample number would be needed to reach genome-wide significance , as our original power calculations predicted that at least 100 cases and 100 controls are required for mapping such alleles at less than 4% false positive rate with 80% power [20] . We find the existence of at least four disease-associated haplotypes in the two nearby chromosome 5 regions intriguing , and speculate that there may be genes in the region affecting traits for which dogs are bred in this population . In small , inbred populations like dog breeds , one popular individual can have many offspring , allowing certain haplotypes to become relatively common . We note that no coding changes agree with the risk haplotypes , suggesting the presence of regulatory mutations . To identify the actual causative mutations additional bioinformatics analysis , validation genotyping in a larger sample set and functional analysis of key candidate variants will likely be necessary . It will also be useful to survey the frequency of the risk haplotypes in different golden retriever populations , for example those from the US and Europe where disease frequencies are reported to vary . RNA-Seq data from B-cell lymphomas demonstrated an almost complete reduction of TRPC6 transcript suggesting cis-regulation by the 29 Mb risk haplotype , which also reduced the expression of three other genes in the region BIRC3 , ANGPTL5 , and KIAA1377 . BIRC3 encodes an anti-apoptotic protein associated with B-cell malignancies and other cancers [39] , ANGPTL5 is a member of the angiopoietin growth factor family [40] , while KIAA1377 is a novel centrosomal protein required for cytokinesis [41] . TRPC6 encodes a transient receptor potential channel , which mediates calcium ion ( Ca2+ ) influx [29] . Interestingly , TRPC6 is not normally expressed in B-cells [42] , but has been reported to play an important role in T-cell activation [30 , 43] . The expression levels of TRPC6 have been shown to significantly alter levels of intracellular Ca2+ elevation and T-cell activation , which are mediated by at least two pathways; the PLCγ pathway regulated by the T-cell receptor , and the PI3K pathway that is mediated by co-stimulation through CD28 [30 , 31] . Notably , the 33 Mb risk allele also suppressed the expression levels of many genes that are involved in the activation of immune responses , particularly T-cell activation . The regulation from the 33 Mb region appears to be trans-regulatory , but the exact mechanism to elicit this effect is unknown at present . One possibility is that a cis-regulatory effect of the risk haplotype on an undiscovered lincRNA in this region could be mediating the trans-regulatory effect . The different effects of the combined risk haplotype and the B-cell lymphoma specific haplotype at this locus cannot be distinguished without further work . Notably , several of the suggested top upstream regulators of the 100 genes affected by the 33Mb haplotype are possible targets of NF-κB [44] , which could suggest that the effect of the risk haplotype could be mediated by pathways affected by NF-κB . Because of the altered gene expression , we hypothesize that the germ-line mutations tagged by the risk haplotypes in the associated loci lead to T-cell dysfunction that plays an important role in B-cell lymphoma and hemangiosarcoma development . The expression levels of T-cell markers , such as CD28 and CD3 epsilon , were not affected by the risk haplotypes , so the expression reduction in TRPC6 and other genes involved in T-cell activation was not due to the absence of T-cells within the tumor . We also did not observe any expression differences in markers for NK cells and dendritic cells , such as CD3 zeta , CD11b , CD11c , CD56 , and CD68 . This is important to note , as the expression levels of certain chemotaxins and receptors , including CCL5 , CCL19 , CCL22 , and CCR6 , which attract lymphocytes , macrophages and/or dendritic cells [45–47] were decreased in dogs carrying the 33Mb-shared risk haplotype . In previous studies , different quantities of these cells in B-cell lymphoma have been linked to diagnostic and prognostic significance in humans as well as dogs [48–55] . In conclusion , we have identified two loci explaining ~20% of the risk for both hemangiosarcoma and B-cell lymphoma in US golden retrievers . While the discovery of the mutation ( s ) and the related mechanisms that lead to tumorigenesis is dependent on future studies , this study demonstrates the power of dogs for mapping germ-line risk factors with strong relevance for human cancer , as well as the importance of non-coding inherited risk factors in cancer predisposition . The strong correlation between the germ-line risk haplotypes and the expression changes that are indicative of immune dysfunction generates a novel hypothesis of how germ-line risk factors contribute to tumorigenesis . This novel hypothesis warrants further investigations both in canine and human lymphoma and angiosarcoma .
All of the golden retrievers in the study were recruited from the privately owned pet population in the US . The owner voluntarily agreed to participate in the study , and a signed consent form was obtained for each participant . All the work described is in accordance to ethical guidelines and is included in the ethical approval protocols on “canine research” , MIT CAC 0910–074–13 ( Lindblad-Toh ) . Diagnosis of B-cell lymphoma was confirmed by histological examination of the tumor as well as by PARR assay [56] . Diagnosis of hemangiosarcoma was obtained by one or more of the following methods: histological examination of formalin fixed tumor tissue , examination of cell surface markers by flow-cytometry , and by the pathology reports that were submitted by the dog owner or their veterinarian , which confirmed hemangiosarcoma diagnosis . Some of the hemangiosarcoma cases that had acute and extensive abdominal hemorrhage with an ultrasound report of multiple cavitated and blood-filled tumors in more than one organ , and those having the characteristic right atrial tumor were included in the study without histological confirmation . Controls were confirmed to be cancer-free by owner questionnaire at the point of sample submission , and by periodic health updates . The age when a dog was last confirmed as healthy was used to determine inclusion . All control dogs’ pedigrees were carefully checked before picking dogs for genotyping to avoid introducing stratification . Cases’ pedigrees were also checked to avoid including closely related individuals when possible . Genomic DNA was isolated from whole blood and was genotyped for 170 , 000 SNPs using the Illumina 170K canine HD array [23] at the Broad Institute of MIT and Harvard , or at GeneSeek Inc ( Lincoln , NE ) . To successfully control for the population stratification present in the dataset , we took an analysis approach based on a method described by Price et al . [24] First , the genome-wide SNP dataset was analyzed by PLINK [27 , 57] ( PLINK1 . 9 was used whenever possible , otherwise PLINK1 . 07 ) to apply standard quality filters including genotyping rate per SNP ( >95% ) and per individual ( >95% ) , and minor allele frequency ( MAF , >5% ) . Chromosome X was excluded because of the risk of it not being handled correctly in mixed model genetic relatedness calculations . Secondly GCTA [25] was used to estimate a genetic relationships matrix ( grm ) to remove excessively related individuals , and to calculate the principal components of the whole-genome SNP genotype data per individual by the EIGENSTRAT method [58] , which was used as a covariate in the final step . Finally , GCTA [25] was used to test for the disease-genotype association with adjustment for the IBS matrix and for the first principal component , both calculated by GCTA . The threshold for genome-wide significance for each association analysis was defined based on the 95% confidence intervals ( CIs ) calculated from the beta distribution of observed p values , a method adopted from the study by the Wellcome Trust Case Control consortium [59] . Sex was used as a covariate . For the conditional analysis to address the independence of the two peaks on chromosome 5 , the genotype of a top SNP of one peak/haplotype was used as the first covariate and sex was used as the second covariate . For the GWAS of hemangiosarcoma , we genotyped 148 hemangiosarcoma cases ( 107 histologically confirmed cases , and 41 presumed cases including 16 with tumor in the right atrium of the heart ) , and 172 healthy controls > 10 years of age . After quality control and removal of excessively related individuals ( grm value > 0 . 75 ) , the final dataset analyzed for the hemangiosarcoma association included 142 cases , 172 controls and 108 , 973 SNPs . For the GWAS of B-cell lymphoma , we genotyped 41 histologically confirmed B-cell lymphoma cases and they were compared to the 172 healthy controls used for the analysis of hemangiosarcoma . To control for population stratification in this small dataset , grm value of 0 . 25 was used as the cut-off to remove dogs related at greater than the half-sibling level within the cases , and in the controls . After the filtering , the final dataset analyzed for the B-cell lymphoma association included 41 cases , 172 controls and 109 , 579 SNPs . For the combined analysis , after quality control and removal of excessively related individuals ( grm value > 0 . 75 ) , the final dataset analyzed for the association included 183 cases ( 142 hemangiosarcoma cases and 41 B-cell lymphoma cases ) , 172 controls , and 109 , 407 SNPs . We further independently validated the genotypes of the 24 top SNPs in a subset of 250 dogs by Sequenom ( miscalling rate 0 . 0038 ) . The haplotype blocks in the associated loci were defined with boundaries that were commonly identified by the clumping analysis using PLINK [26 , 27] and r2 based LD analysis by Haploview [28] . PLINK clumping analysis was performed by setting parameters as follow: association p-value for the index SNP < 1 × 10−4 , r2 > 0 . 8 or 0 . 9 , and a physical distance limit of 1 Mb . The Haploview analysis was performed by calculating pair-wise r2 values for the SNPs between 28 Mb and 36 Mb on chromosome 5 with a 2 Mb distance limit , and haplotype blocks were defined by r2 > 0 . 8 or 0 . 9 . The haplotype blocks commonly identified by both analyses were used for further analysis . Haplotypes of each block , their allelic frequencies , chi-square test , allelic odds ratio and p-values ( Praw ) were obtained using PLINK . Each haplotype was then tested for association significance by running a permuted chi-square test for 107 iterations using PLINK . Estimation of the phenotypic variance explained by genetic variance was performed by REML analysis using GCTA [60] , following online instructions on the GCTA website . In our analyses , the variance of the genetic factor was determined by the genotypes of SNPs on all autosomes , on each autosome separately , and within the associated region ( 25–40 Mb ) on chromosome 5 . Sex was used as a covariate . The estimate of variance explained on the observed scale is transformed to that on the underlying scale by the estimated disease prevalence of the general population . A p-value for each analysis is calculated based by performing a log-likelihood ratio test . We estimated prevalence as 0 . 20 for hemangiosarcoma , 0 . 0625 for B-cell lymphoma [5] , and 0 . 2625 for being affected by either cancer , as it is extremely rare for one dog to have both cancers . Whole-genome paired-end sequencing was performed for germ-line DNA from nine golden retrievers , of which six were from the GWAS cohort . For each sample , approximately 1 billion 101 base-pair paired-end reads at 40x coverage were generated using Illumina HiSeq 2000 . Picard pipeline [61] was used for data quality filtering and alignment of the reads to the canFam3 . 1 reference genome . The Genome Analysis Toolkit’s ( GATK’s ) UnifiedGenotyper [62] was then used to make genotype calls from the cleaned alignments . The resulting variants were then annotated based on the conservation across species using SEQscoring [63 , 64] , annotated and analyzed for predicted effect by using snpEff [65] , and were visually examined by IGV [66] to look for variants likely to cause biological changes , and that are concordant with the disease-associated haplotypes . One variant was evaluated with SIFT [67] . Twenty-two canine nodal B-cell lymphoma and twenty-two hemangiosarcoma samples ( one tumor sample per dog ) were analyzed by high-density RNA sequencing ( 20 million paired end reads ) . Total RNA was isolated from a whole frozen naïve ( untreated ) tumor tissue or cryopreserved single cell suspension of naïve tumor cells . Indexed Illumina sequencing libraries were constructed , size selected to 320 bp +/- 5% , and 50 base-pair paired-end reads were generated by Illumina HiSeq 2000 . To estimate the abundance of different genes expressed in our samples , we first aligned the read data to canFam3 . 1 using TopHat [68] v1 . 4 . 1 . The mate inner distance was set to 100 bp , and the maximum intron length was set to 500 , 000 bp . We then used HTSeq [69] v0 . 5 . 3p9 set for non-strand-specific data to perform read counting on genes . For a gene annotation , we used the canFam3 . 1 annotation supplemented with RNAseq data [70] . The expression levels were compared using edgeR [71] v3 . 0 . 8 to examine the relative gene expression changes associated with the presence or absence of approximately one copy of the risk haplotypes at 29 Mb or 33 Mb locus in the tumors . Given the high frequency of the risk allele , the 29 Mb “higher-risk” and “lower-risk” groups were defined as follows: a higher-risk group containing 12 dogs homozygous for risk haplotype; and a 29 Mb lower-risk group containing eight heterozygous dogs and two dogs with no copy of the risk haplotype ( all dogs haplotypes were identical for the 29 . 7Mb-shared and 29 . 9-shared Mb ) . Because very few dogs were homozygous for the risk haplotype at the 33 Mb , the 33 Mb higher-risk and lower-risk groups were defined as follows: a higher-risk group of six dogs ( five heterozygous and one homozygous for the 33Mb-shared risk haplotype ) ; and a lower-risk group of 16 dogs with no copy of the risk haplotype . The groups were largely the same if defined from the 33Mb-BLSA risk haplotype , but the shared haplotype was used for group definition to be consistent with hemangiosarcoma analysis . B-cell lymphoma RNA was isolated from either tumor cells in suspension , or from a tumor biopsy that contained more stromal tissue ( lymphocyte content > 90% , of those 85–100% were malignant cells ) . This known variable was applied as a blocking factor in edgeR analysis to reduce its influence in detecting the differences in gene expression . Expression differences between the groups with p-value and false discovery rate ( FDR ) of less than 0 . 05 were considered significant findings . Unsupervised clustering was performed using normalized FPKM values for the annotated genes , calculated for each sample using CuffNorm from Cufflinks 2 . 2 . 1 . These values were then used as a feature vector and the dendrogram was created using the R v2 . 15 functions “dist” and “hclust” . A knowledge-based functional analyses of the significant expression changes by the 29 Mb risk allele in 27 genes , and by the 33 Mb risk allele in 100 genes were performed by Ingenuity Pathway Analysis ( IPA ) [32] . Of the 27 and 100 genes examined , IPA mapped 25 and 89 genes respectively . The parameters for the core analysis were set to consider direct and indirect relationships of genes and endogenous chemicals at predicted and experimentally observed confidence levels . The p-values for the downstream functions and canonical pathway analyses were corrected for multiple testing by the Benjamini-Hochberg procedure , and resulting p-values less than 0 . 05 were considered significant . When the analysis of downstream functions or upstream regulators identified a gene set with “bias” in the direction of expression changes , significance was determined by the combination of a p-value of less than 0 . 05 and an activation z-score of less than-2 . 00 or greater than 2 . 00 , following Ingenuity Systems’ recommendation . False discovery rate ( FDR ) cutoff was set to 0 . 05 and fold change ( FC ) cutoffs were 1 and-1 ( in log2 ) . All the p-values reported in this study were obtained by using the programs mentioned in each analysis method . Briefly , the p-values in GWAS analysis were obtained by using GCTA , with a mixed model approach to account for population stratification , and a 0–1 quantitative response variable to represent the case-control status . The significance of the slope coefficient of a SNP , which represents the effect size of the SNP is calculated by the standard t test based on the variance of the slope coefficients of the study cohort [72] . For case-control data , Haploview utilizes a simple chi-square test to calculate the phenotype-haplotype association p-values ( Praw ) [28] , and the association significance p-value ( Pperm ) was obtained as the empirical probability of observing chi-square values in permutation tests that exceeded the best observed chi-square value using PLINK1 . 07 . The p-values obtained by edgeR to identify differentially expressed genes were calculated by fitting gene-wise generalized linear models , and then conducting likelihood ratio tests for the risk haplotype [71] . The p-values by IPA for the canonical pathways and downstream biological functions were calculated using Fisher’s Exact Test , comparing the proportion of genes from the provided list mapping to a function or pathway to the proportion genes in the IPA database in that function or pathway [32] . The p-values were then corrected for multiple testing by the Benjamini-Hochberg procedure [32] . The upstream regulator analysis calculates the “overlap p-values” using Fisher’s Exact Test , which measures whether there is a statistically significant overlap between the observed gene set and the genes that are regulated by a particular transcriptional regulator [32] . GWAS data are available on the Broad Institute’s website ( www . broadinstitute . org/ftp/pub/vgb/dog/HSA_BLSA_PlosGenetics2014_paper/ ) . WGS and RNA-Seq data are available via the NCBI BioProject site ( WGS: PRJNA247491 , RNA-Seq: PRJNA267721-267742 ) .
|
To shed light on the genetic predisposition to cancers of the hematologic system , we performed genome-wide association analysis of affected and non-affected pet dogs . Dogs naturally develop the same diseases as humans , including cancer , and the relatively limited genetic diversity within different breeds makes genetic studies easier compared to in humans . By doing genome-wide association , we identified loci predisposing to hemangiosarcoma and B-cell lymphoma . To our surprise , we found two shared loci predisposing to both diseases . Within these two regions we identified several partially overlapping haplotypes , predisposing somewhat differently to the two cancers . We found no coding mutations that followed the risk or non-risk haplotypes suggesting that regulatory mutations exert the effect on disease . We also looked at gene expression in B-cell lymphomas , comparing samples from individuals with risk or non-risk haplotypes . This analysis showed differential expression associated with the haplotypes at both loci , suggesting the risk haplotypes are associated with an effect on T-cell response .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Genome-wide Association Study Identifies Shared Risk Loci Common to Two Malignancies in Golden Retrievers
|
Mutation is fundamental to evolution , because it generates the genetic variation on which selection can act . In nature , genetic changes often increase the mutation rate in systems that range from viruses and bacteria to human tumors . Such an increase promotes the accumulation of frequent deleterious or neutral alleles , but it can also increase the chances that a population acquires rare beneficial alleles . Here , we study how up to 100-fold increases in Escherichia coli’s genomic mutation rate affect adaptive evolution . To do so , we evolved multiple replicate populations of asexual E . coli strains engineered to have four different mutation rates for 3000 generations in the laboratory . We measured the ability of evolved populations to grow in their original environment and in more than 90 novel chemical environments . In addition , we subjected the populations to whole genome population sequencing . Although populations with higher mutation rates accumulated greater genetic diversity , this diversity conveyed benefits only for modestly increased mutation rates , where populations adapted faster and also thrived better than their ancestors in some novel environments . In contrast , some populations at the highest mutation rates showed reduced adaptation during evolution , and failed to thrive in all of the 90 alternative environments . In addition , they experienced a dramatic decrease in mutation rate . Our work demonstrates that the mutation rate changes the global balance between deleterious and beneficial mutational effects on fitness . In contrast to most theoretical models , our experiments suggest that this tipping point already occurs at the modest mutation rates that are found in the wild .
Mutation is fundamental to evolution . Without it , evolution cannot occur , because mutation provides the genetic variation necessary for selection and genetic drift . Each new mutation in an individual can increase its fitness , decrease its fitness , or have no effect on its fitness . Unfortunately , most mutations with fitness effects are deleterious , and fitness-increasing beneficial mutations constitute only a small fraction of all possible mutations [1] . The mutation rate can itself evolve , because it is subject to genetic change in the "mutation rate genome" , the part of a genome encoding DNA replication and repair systems [2 , 3] . Here , we characterize the long-term effects of a range of mutation rates on adaptation , as well as the evolution of the mutation rate itself , by evolving multiple replicate populations of asexual Escherichia coli in a minimal medium in the laboratory . Evolutionary adaptation under increased mutation pressure in large non-recombining populations like ours has been explored in past work ( all mutations that occur in our E . coli laboratory strain's genome are linked ) . The joint effects of mutation and linkage on selection ( and the related topics of diversity and the evolution of sex ) have been much studied since Fisher [4] and Muller [5] ( [6–10] , recently reviewed in [11–14] ) . Under increased mutation pressure , multiple clones within a population may acquire new mutations , and then compete with each other for fixation . While relevant studies show that the speed of adaptation can increase with the genomic mutation rate [10 , 15–18] , they leave open the possibility that extremely high mutation rates could hinder adaptation . This possibility is raised by a variety of models that predict declining fitness in populations with extreme mutation rates . An early , influential , but simple model predicted that a population's fitness will decrease when the rate of mutation increases beyond a critical “error threshold” [19] whose value depends on model details . Other models of populations evolving at high mutation rates are more realistic and take into account phenomena like beneficial mutations and demography . However , they also predict that adaptation can be slowed and eventually reversed at sufficiently high mutation rates by the effects of deleterious mutations [20–24] . Many studies have documented the evolution of increased mutation rates [25–31] , which can evolve in certain conditions . For example , after a recent environmental change that creates opportunities for novel adaptations and new beneficial mutations [32 , 33] , a cell with a mutator allele is more likely to produce large-effect beneficial mutations than a cell with a wild-type mutation rate . Because of their improved fitness , cell lineages with newly acquired beneficial alleles ( and their linked mutator alleles ) can increase in frequency in the population . Thus , hypermutation can readily evolve when mutator alleles hitchhike to fixation with beneficial mutations [34–37] . In the long term , however , hypermutation can be detrimental , because most non-neutral mutations have deleterious consequences [1] . Thus , an individual with a higher mutation rate may accumulate more deleterious mutations overall , which can result in lower fitness . For this reason , selection has been predicted to reduce mutation rates [38] . However , there are several potential reasons why mutation rates may not decline all the way to zero . One of them is that the physiological mechanisms required to improve replication fidelity and DNA repair carry a fitness cost [39–42] . Another is that the power of selection to reduce the mutation rate is limited by population size via the so-called drift-barrier [43 , 44] . Experimental observations of evolved reductions in the mutation rate have been reported , but are relatively infrequent [27 , 31 , 45–50] ( reviewed in [51] ) . While some previous experiments explored the adaptive responses and mutation rate changes that can take place under increased mutational pressure [46–48 , 50] , they focused on one or two mutation rates , and did not include genomic analyses ( except [50] ) . Here , we sought to provide a uniquely comprehensive empirical data set across a range of mutation rates , including whole genome population sequencing data , mutation rate data , and fitness measurements in a number of environments . To do so , we engineered four isogenic E . coli K12 MG1655 derivative strains with increased mutation rates and evolved eight replicate populations of each strain for 3000 generations in a serial-transfer experiment . Genomic mutation rates differed more than a hundred-fold among these strains and ranged from U = 0 . 00034 to U = 0 . 036 point mutations per genome per generation by one method of estimation . During evolution , we periodically characterized the growth rate and stationary population density of each population . We also assayed the fitness of evolved populations in a variety of stressful environments . High-throughput population sequencing allowed us to characterize how far our populations spread through sequence space , and to study the mutations occurring in each population .
We used a population’s maximum growth rate during exponential growth as a proxy for fitness , and we refer to relative fitness as the difference between an evolved and a reference population , usually the ancestral population . We first measured the fitness of the ancestral strains relative to the E . coli K12 MG1655 strain and found that the ancestral strains had similar fitness values , with the exception of the MRL strain , which had much lower fitness than the other strains ( S3A Fig , linear mixed effects analysis , Χ2 ( 4 ) = 248; MRS: 0 . 34±0 . 02; MRM: 0 . 32±0 . 02; MRL: 0 . 11±0 . 02; MRXL: 0 . 25±0 . 02 , relative fitness ± s . e . m . , p<2×10−16 ) . We next measured the fitness of each evolving replicate population relative to its ancestor at several time points during the experiment ( Fig 2 , S3B Fig , S4A Fig ) . At generation 3000 , replicate populations with higher mutation rates showed a greater increase in fitness , except for the MRXL strain , which had the smallest fitness increase of all strains ( p = 0 . 002 , linear mixed effects analysis , Χ2 ( 4 ) = 16; simultaneous tests for general linear hypotheses that relative fitness is unchanged at generation 3000: MRS:0 . 7±0 . 2 , p = 0 . 004; MRM:1 . 0±0 . 2 , p<4×10−5; MRL:1 . 1±0 . 2 , p<1×10−6; MRXL:0 . 6±0 . 2 , p = 0 . 02 , evolved fitness difference from ancestor ± s . e . m . , significance ) . In replicate populations with lower mutation rates ( MRS and MRM ) fitness only began to rise substantially after 1000 generations , while in replicate populations with higher mutation rates ( MRL , MRXL ) this fitness increase began earlier ( Fig 2A; linear mixed effects analysis , Χ2 ( 4 ) = 17 , p = 0 . 002; simultaneous tests for general linear hypotheses that relative fitness is unchanged at generation 1000: MRS: -0 . 01±0 . 07 , p = 1 . 0; MRM:0 . 07±0 . 07 , p = 0 . 76; MRL:0 . 40±0 . 07 , p<0 . 0001; MRXL:0 . 57±0 . 07 , p<0 . 0001 , evolved fitness difference from ancestor ± s . e . m . , significance ) . A second growth curve metric that integrates information about the lag phase , growth rate , and carrying capacity yielded similar results ( S1 Text ) . The delays in fitness gains in populations with lower mutation rates are also reflected in reduced fitness variation among replicate populations at any one point in time ( S4B Fig ) . MRL and MRXL populations seem to form two clusters with either high or low ( ancestral-like ) fitness at the end of the experiment ( Fig 2B ) . In principle , the fitness of populations with lower fitness might either not have increased at all , or it might have increased at first and subsequently decreased again . Examination of individual fitness trajectories ( S4A Fig ) shows that most MRXL replicate populations with low fitness at generation 3000 gained and then lost fitness again . We sequenced a sample of each heterogeneous evolving population rather than a clone isolated from each population , so that we could estimate the genetic diversity within each sequenced population . We sequenced 100 populations in total: the four ancestor populations with different mutation rates , and eight replicates evolved from each ancestor at generations 1000 , 2000 , and 3000 . Specifically , we sequenced the four ancestor populations after one day of growth just before having split them into their replicate populations . The mean sequence coverage for populations was 364-fold ( standard deviation 98 ) , with 99% ( 83% ) of the samples having at least 100-fold ( 200-fold ) coverage across at least 95% of the genome . In virtually all sequenced samples less than 1% of the genome had no sequence coverage ( S5 Fig ) . We identified the frequency of SNPs in each sequenced population and their annotations using breseq , which has been widely used in microbial studies and has been optimized for bacterial data [56] ( S6 Fig ) . We discovered several SNPs at non-zero frequency in these sequenced ancestral populations ( 1 , 3 , 2 , and 64 loci for MRS , MRM , MRL , and MRXL , respectively; S7 Fig ) , some of which may have been transferred to the evolving populations . Previous studies had suggested that there may be biases in the mutational spectra caused by the reduced efficacy of the mutL and dnaQ gene products [57–59] , but our sequence data shows that any such bias is weak or absent in our strains ( S8 Fig ) . One can view an evolving population as a cloud of mutant individuals in sequence space . We suspected that this mutant cloud would be spread out further in genotype space–indicating greater standing diversity–for populations with a higher mutation rate . To test whether this was indeed the case , we first defined the center of an evolving population as its consensus sequence and then computed the average distance of each population to this consensus . A strain’s mutation rate affected the size of its mutant cloud ( linear mixed effects analysis , cube root of diversity taken to ensure homoscedasticy , Χ2 ( 4 ) = 79; p = 3×10−16; see Methods ) , such that higher mutation rates led to a larger cloud ( Fig 3A; MRS:0 . 005±0 . 001; MRM:0 . 007±0 . 001 MRL:0 . 011±0 . 001; MRXL:1 . 018±0 . 001 , cube root of diversity ± s . e . m . ) . Similarly , higher mutation rates led to higher levels of mean nucleotide site diversity ( Fig 3C ) . We also expected evolving replicate populations with higher mutation rates to accumulate more high frequency derived alleles than those with lower mutation rates because populations with more variation are expected to adapt faster . Here , we defined a high frequency derived allele at a given site as a derived allele found in more than 50% of the population . Again , strain identity affected the number of high frequency derived alleles ( linear mixed effects analysis , Χ2 ( 4 ) = 69 , p = 4×10−14; see Methods ) , such that strains with higher mutation rates accumulated more high frequency derived alleles ( Fig 3B; MRS:0 . 8±0 . 3; MRM:1 . 8±0 . 8; MRL:12±2; MRXL:68±13 , number of high frequency derived alleles ± s . e . m . at generation 3000 ) . Most new mutations are thought to be effectively neutral or deleterious , and only a small fraction are beneficial in a given environment [1] . To identify putatively beneficial mutations in our replicate populations , we developed a statistical test that identifies genes in which more replicate populations contain high frequency derived alleles of any one gene than one would expect by chance alone ( Methods , S9 Fig ) . In addition to identifying beneficial mutations , this approach can also identify artifacts such as mutational hotspots or the violation of independence across samples . Our test identified 20 genes with putatively beneficial mutations . Statistical analysis places the highest confidence for the existence of beneficial mutations in eight of these genes ( clpB: 6/32 populations , p<6×10−6 , cspC: 3/32 populations , p = 3×10−5 , mreB: 5/32 populations , p<6×10−6 , pykF: 16/32 populations , p<6×10−6 , rnb: 7/32 populations , p<6×10−6 , rpoC: 6/32 populations , p<6×10−6 , topA: 9/32 populations , p<6×10−6 , and ygeN: 4/32 populations , p = 4×10−5 ( S9A Fig , S10 Fig ) . Most of the mutations are nonsynonymous or nonsense mutations , and are thus likely to affect gene function ( S9A Fig ) . Furthermore , between 70 and 100% of the observed mutations in any one gene occurred at different sites in different replicate populations ( clpB: 100% , cspC: 100% , mreB: 91% , pykF: 95% , rnb: 78% , rpoC: 100% , topA: 70% , and ygeN: 100% ) , indicating that they occurred de novo and independently from each other . The two most commonly mutated genes were pykF and topA , which encode pyruvate kinase and topoisomerase A , respectively . Pyruvate kinase is a key enzyme in glycolysis , and topoisomerase A can affect the superhelicity of DNA . Both genes have repeatedly acquired beneficial mutations in previous experiments with E . coli B in glucose minimal medium [60–65] . Similarly , mutations in cspC , a stress protein , confer a fitness advantage for E . coli populations evolving at 37°C and higher [62 , 66] . Finally , mutations in the RNA polymerase gene rpoC and the cytoskeletal gene mreB have also been commonly found in laboratory evolution [62 , 67 , 68] . Surprisingly , multiple MRXL replicates showed the same nucleotide change in 12 of the 20 putatively beneficial genes . As previously discussed , some mutations arose in the ancestor MRXL population before we split it into its replicate populations ( S7 Fig ) . Specifically , mutations in 10 of the 12 genes with the same nucleotide change across the MRXL replicates were also found in the ancestral population ( at a frequency between 3% and 22% ) , before we had split this population into our replicates ( S9B Fig ) . Only two genes ( yfeZ and rrlH ) showed no evidence for such identical , pre-existing mutations , although such mutations may have existed below the detection limit of our sequencing coverage . We cannot conclude that these 12 putatively beneficial genes have a beneficial effect in the MRXL populations , because our statistical test relies on the assumption that the mutations occurred and were subject to selection independently . To know with certainty the phenotypic effects of any of these mutations would require additional empirical data from allelic replacement experiments . Thus far , the only phenotype we studied was population growth in one environment–the glucose minimal medium in which we conducted the entire experiment . To expand our analysis to other environments , we used Biolog Phenotype MicroArrays , which help measure the growth and respiration activity of a bacterial strain in multiple environments ( [69] , but see [70] for caveats ) . These microarrays determine the ability of our strains to grow in the presence of 96 stressful compounds that include antibiotics and heavy metals . We exposed our evolving replicate populations to these stressors only after completion of laboratory evolution , i . e . , the populations could not have adapted to them during the evolution experiment . We selected two populations at random from the MRS , MRM , MRL , and MRXL replicate populations at the end of evolution . Remarkably , the two selected MRXL replicates failed to grow in every single one of the 96 environments , as did the MRXL ancestor . One possible explanation is that the MRXL strain is inherently more sensitive to novel environments , including the medium used in the assay . The remaining populations grew in 42–60 ( 43 . 8%-62 . 5% ) of the environments , depending on the population . In order to identify any link between mutation rate and growth in these 96 environments for the MRS , MRM , and MRL replicates , we identified the molecules in which an evolved replicate population grew better ( and worse ) than its ancestor ( S11 Fig ) . All replicate populations were better able to tolerate stressful conditions than their ancestors in some of the tested conditions ( between 8% to 30% ) , which suggests that some ( fortuitously ) beneficial mutations have occurred . The two MRL replicate populations we tested tolerated stressful conditions better than the MRS and MRM replicate populations . However , the MRS replicate populations were able to tolerate more stressful conditions than the MRM populations . In sum , these analyses establish no simple association between ancestral mutation rate and stress tolerance after evolution . To study the evolutionary dynamics of growth in stressful conditions over the course of the experiment , we periodically tested the growth of the ancestor and all evolving replicate populations in two stressful conditions: the antibiotic nitrofurantoin ( a specific , “narrow” stressor ) and acidic media ( a broader stressor ) . Nitrofurantoin is a nitrofuran antibiotic with multiple mechanisms of action . Resistance to nitrofurantoin is conferred by mutations in two genes ( nfsA and nfsB ) , and has a fitness cost in the absence of the antibiotic [71] . Thus , resistance mutations are unlikely to exist at appreciable levels as part of the standing variation in populations not exposed to nitrofurantoin . E . coli is known to tolerate acidic conditions due to several acid resistance systems , one of which depends on the alternative sigma factor σS , encoded by the rpoS gene [72 , 73] . For both nitrofurantoin-containing and acidic media , we computed the fold change in growth ( the cell density after 24 hours ) of the evolved populations relative to their ancestors , controlling for changes in carrying capacity ( see Methods , Fig 4 , S12 Fig ) . We found that replicate populations with the highest mutation rates grew more slowly in nitrofurantoin than their ancestor . In contrast , populations at low and intermediate mutation rates grew faster than the ancestor ( see Methods , p = 5×10−8 , linear mixed effects analysis , Χ2 ( 4 ) = 40; MRS:0 . 85±0 . 15; MRM:0 . 49±0 . 15; MRL:0 . 94±0 . 15; MRXL:-0 . 29±0 . 15 , log fold change ± s . e . m . , positive log fold change indicates evolved strains grew better , and negative log fold indicates ancestor strains grew better ) . The ancestral mutation rate also affected growth in acidic media . In contrast to nitrofurantoin , all strains showed increased growth , but qualitatively similar to nitrofurantoin , the MRXL replicates showed the smallest growth rate increase at low pH ( p = 2×10−8 , linear mixed effects , Χ2 ( 4 ) = 42 ) ; MRS: 1 . 23±0 . 09; MRM:1 . 39±0 . 09; MRL:1 . 56±0 . 09; MRXL:1 . 15±0 . 09 , fold change ± s . e . m . , fold change > 1 indicates evolved strains grew better , and fold change < 1 indicates ancestor strains grew better ) . In sum , growth in two stressful conditions , nitrofurantoin-containing and acidic media , improved with increasing mutation rates ( and thus increasing diversity ) , except for the MRXL replicates which showed a relative reduction in growth . We measured the mutation rate of one randomly selected clone from each evolved replicate population at generation 3000 , and of the population’s ancestral strain . To this end , we used fluctuation assays for mutations that cause rifampicin resistance , and estimated the genomic mutation rate U using Drake's approach [74] . The mutation rates of the evolved MRXL replicates decreased on average by 556% , reaching 18% of the ancestor’s mutation rate ( 2%-42% , depending on the replicate ) ; Fig 5 , S3 Table ) . At the end of the evolution experiment , the mutation rates in the MRXL populations were no longer statistically distinguishable from those of the MRL replicates ( Wilcoxon rank sum test , p = 0 . 38 ) . The evolved populations’ mutation rates for the MRM and MRL strains also tended to decrease ( MRM: 49% of the ancestor's mutation rate on average , ranging between 11%-90% of the ancestor; MRL: 87% , range 32%-170% ) . In contrast , the replicates from the MRS strain increased their mutation rate somewhat , to 206% of the ancestor's mutation rate ( range 16%-1000% ) . Having estimated the mutation rates for the ancestor and evolved populations , we also wanted to examine whether prominent theoretical models that predict declines in mean population fitness at high mutation rates apply to our populations ( S2 Text ) . While some of the models we studied ( e . g . , that of mutational load ) predict a small reduction in fitness at the highest mutation rates we employed , none of them could account for the magnitude of the loss of adaptation we found in several of the MRXL replicates ( S2 Text; Fig B , Fig C in S2 Text ) . Because mutation rates changed between the beginning and the end of the experiment , we wondered whether the final mutation rates were correlated with our measured phenotypes . We found significant correlations between a replicate's mutation rate and its effective population size , standing genetic diversity , and number of high frequency derived alleles , but no correlations between a replicate's mutation rate and its final relative fitness , or normalized cell density after 24 hours of growth in acidic medium or medium containing nitrofurantoin ( Spearman's rank correlation , S13 Fig ) . Interpretation of these results requires caution for two reasons . First , for any one population , we do not know exactly when ( during the 3000 generations of evolution ) the mutation rate changed from its ancestral value . Second , we compared the mutation rate of a single randomly-selected clone from populations which can have considerable genetic diversity , and thus potentially also show diversity in mutation rates . Despite these caveats , we found that the correlations between a representative clone's mutation rate and our other metrics are consistent with our previous analyses and figures ( Fig 4 , S2 Fig , S12 Fig ) , which simply considered the effects of ancestral mutation rate ( strain identity ) . We call the set of genes potentially involved in modulating the mutation rate the "mutation rate genome" . We wondered whether this part of the whole genome was a preferential target of mutation or selection in our experiments . To find out , we first identified a set of 96 genes potentially involved in modulating the mutation rate ( S2 Table ) from the literature and EcoCyc [49 , 75–77] . If mutations or selection did not preferentially affect the mutation rate genome , the amount of genetic change we observe in it would be proportional to its length relative to the rest of the genome . This is indeed the case: We counted the number of synonymous mutations occurring at any frequency in any replicate population at generation 3000 , and observed no statistically significant increase in the incidence of such genetic change in the mutation rate genome for any of our evolving strains ( S14A Fig ) . We also found no difference in mean diversity between synonymous sites in the mutation rate genome relative to the rest of the genome ( S14B Fig ) . Although the mutation rate genome is not a preferential target of genetic change , its genes still accumulated many non-synonymous and nonsense changes , which are the kinds of changes that are especially likely to affect protein function ( S15 Fig ) . To identify mutant alleles putatively associated with the decrease in mutation rates we had observed in MRXL replicate population after 3000 generations ( Fig 5 ) , we identified nonsynonymous or nonsense mutations in the mutation rate genome with an allele frequency of at least 50% in any MRXL evolved replicate population . Mutations in ten genes met these criteria ( rpoS , umuC , dinB , dinG , dps , glyS , glyW , mutL , phr , and vsr ) , and two were found in multiple replicate populations ( rpoS: 7 of 8; umuC: 2 of 8 ) . The rpoS gene encodes the alternative sigma factor , σS , which activates the stress response in E . coli ( reviewed in [78] ) . Populations with rpoS mutations can hold a fitness advantage in nutrient-limiting environments [79] , but at a cost to fitness in a variety of stressful environments [28 , 80] . Because σS is a bacterial transcription factor , it can only affect the mutation rate indirectly , by changing the expression of proteins directly involved in DNA copying , repair , and proofreading . For example , σS modulates the expression of the error prone DNA Polymerase IV encoded by dinB [81] . We found the same rpoS N124D mutation in 2 . 5% of the individuals in the ancestral MRXL population and in all eight evolved replicates . ( This mutation reached 40 . 1% in MRXL1 and 100% in the rest of the replicates . ) Thus , the mutation was likely distributed to the eight replicate populations from the ancestor , and either increased in frequency due to its direct fitness effects , or because it was hitchhiking with a beneficial mutation . The MRXL4 and MRXL6 replicate populations each acquired different non-synonymous mutations in umuC , which encodes DNA polymerase V . Each of the remaining genes with high frequency mutant alleles in a single replicate population were involved in DNA repair and replication ( dinB , dinG , glyS , glyW , mutL , phr , vsr ) or protection of DNA in stationary phase ( dps ) in a single replicate population and could have also affected the evolved mutation rate .
Here , we studied the effects of mutational pressure on evolutionary adaptation and the evolution of the mutation rate itself . To this end , we engineered four isogenic E . coli K12 MG1655 derivative strains with increasing mutation rates ( MRS , MRM , MRL , and MRXL ) , and evolved them for 3000 generations . Our smallest ( wild-type ) ancestral mutation rate ( MRS: U = 0 . 00034 per genome per generation ) was somewhat smaller than rates estimated in E . coli strains using a similar experimental approach ( U = 0 . 0025 ) [39 , 74] , but similar to those estimated in a wildtype E . coli B strain using a sequencing approach ( U = 0 . 00041 ) [82] . At the opposite extreme was our strain with the highest ancestral mutation rate ( MRXL ) . We originally expected this strain to have a mutation rate approximately 4500-fold higher than wildtype [35] , consistent with the large effects that mutations in the dnaQ and mutL genes have on the mutation rate [75] . However , our measurements of this rate demonstrated that it was lower than expected ( MRXL: 139-fold higher than our wildtype; U = 0 . 036 per genome per generation ) . The discrepancy could in principle be due to the acquisition of an anti-mutator allele during the transfer of the strain between laboratory locations . Alternatively , our mutation rate could be an underestimate for technical reasons discussed in the Methods . The mutation rate for our MRXL strain was also somewhat lower than that of a hypermutable clone which spontaneously evolved from an E . coli B strain [50] ( mutT: U = 0 . 061 ) . The mutation rate of our hypermutable MRXL strain is low enough that we expected its populations to be viable [21] . In sum , we conducted our experiments with strains having a range of viable mutation rates , from wildtype ( MRS ) , to a 16- , 22- , and 139-fold higher mutation rate ( MRM , MRL , and MRXL ) . We first characterized the general patterns of adaptation in our four strains , and found that their fitness increased significantly by generation 3000 for all replicate populations . Previous experimental evolution studies in constant environments have observed fitness gains that are initially large but decrease over time [17 , 18 , 83 , 84] , which is consistent with diminishing returns epistasis , in which the size of the fitness gain in an evolving population depends on its current fitness , such that populations with lower fitness can improve their fitness to a greater extent [85 , 86] . However , our fitness trajectories differ from those predicted by diminishing returns epistasis in two ways . First , they do not show a decreasing fitness gain over time [18] . Second , the mean fitness of replicate populations with small or modestly high mutation rates ( MRS , MRM ) did not immediately improve , but unexpectedly remained largely unchanged for the first 1000 generations ( compared to [87] ) . While delayed adaptive response is consistent with a lower overall beneficial mutation supply rate , it may not be sufficient to explain our observations . We expected to wait just 44 generations for a new beneficial mutation to establish in our slowest-evolving replicate population ( S3 Text ) . It also raises the possibility that even at moderately high mutation rates , contingent evolution [88] , in which the timing and the order of mutational events affects a population’s adaptive evolution , may be important in our populations . An instance of such contingent evolution has been documented in E . coli [89 , 90] , but the higher mutation rate of some of the strains used in our evolution experiment makes contingent evolution a less likely explanation for delayed adaptation . We next characterized the effect of mutational pressure on adaptation . We found that strains with higher ancestral mutation rates increased in fitness more than those with lower mutation rates , except for MRXL populations , which we will discuss below . These observations are in agreement with theory [15 , 91] and previous experimental studies which found that large asexual populations of E . coli [17 , 50] and yeast [15] with high mutation rates outperformed those with low mutation rates . If we just consider relative fitness after 1000 generations , our data from our four strains are consistent with expectations: MRS and MRM populations have lower mean relative fitness than MRL and MRXL . It is only thereafter that the fitness of MRXL populations stops increasing , such that they have lower mean fitness at generation 3000 than the MRS , MRM , and MRL strains . We do not actually observe the loss of fitness on average across the MRXL replicate populations , but rather a prolonged period in which fitness remains unchanged as a whole . Interestingly , however , the fitness of several MRXL replicate populations decreases from its maximum and arrives at a value that is approximately equal to that of the ancestral population . This is reminiscent of models of extreme mutational pressure developed over the past forty years that predict reduced adaptation and eventual extinction [19 , 20 , 22 , 92–94] . However , these models predict a loss of fitness only at higher mutation rates than we observed , and require unrealistic assumptions ( S2 Text ) , together emphasizing the importance of additional theoretical work . Another possibility is Hill-Robertson interference [7] , which can reduce the rate of adaptive evolution by background selection—negative selection against deleterious alleles that removes the most deleterious lineages from a population—and can reduce genetic diversity [8 , 12] . Empirical evidence supports the action of this mechanism in natural populations of several eukaryotic species ( reviewed in [13 , 14] ) . However , because background selection removes deleterious mutations from a population , it cannot alone reduce the fitness of a population and it can therefore not explain the loss of fitness we observed in the three MRXL replicates . Overall , our observations support the notion that reduced adaptation can manifest itself at smaller mutation rates than previously thought ( U = 0 . 036 in the MRXL strain ) , even though more than 1000 generations may be needed to manifest its effects . This observation is all the more striking , because the mutation rate itself had decreased dramatically for all MRXL populations after 3000 generations ( and much less so in the MRS , MRM , MRL populations ) . While a lowering of the mutation rate has been previously observed [46–48 , 50] and predicted to be favored in some conditions [38 , 40 , 42 , 43 , 95] , its extent and consistency across multiple of our evolving populations is remarkable . The mutation rate decrease probably did not occur very early during evolution , because the MRXL populations show greater genetic diversity than all other populations throughout the experiment ( Fig 3 ) . The decreasing mutation rate , together with the observation that the MRXL populations failed to adapt after more than 1000 generations , suggests that the maladaptive effects of hypermutation begin at even lower mutation rates than those in our initial MRXL strain . While we cannot predict whether our hypermutable populations would eventually go extinct , the observation that their mutation rate can decrease adaptively makes this less likely . Indeed , recent mutation accumulation experiments with small bacterial populations suggested that populations with higher mutation rates tend to go extinct more often and have reduced fitness than populations with lower mutation rates [47] . Of the several "mutation rate genome" genes mutated in MRXL strains , only rpoS was found in all eight evolved MRXL replicate populations . rpoS encodes for the stress response modulator σS that can indirectly affect the mutation rate through transcriptional changes . However , we cannot definitively identify the proximal mechanisms driving the drop in mutation rates using bioinformatics alone . Future experimental studies to evaluate the effect of each "mutate rate genome" mutant allele on the mutation rate and fitness would be necessary . We emphasize that all our experiments use asexual populations , and that the evolutionary dynamics of mutation rates and adaptation may be different in sexual , recombining populations . For example , in our non-recombining populations , any mutator allele remains completely linked to the ( mostly deleterious ) mutations it helps bring forth , resulting in indirect negative selection on the mutator allele . However , such an allele and its associated mutations can become unlinked in recombining populations , which reduces the strength of indirect selection on the mutator allele ( see [33 , 39] for reviews ) . Additionally , beneficial and deleterious alleles can become unlinked in recombining populations , which can lead to increased levels of adaptation and diversity ( see [13 , 14] for reviews ) . We also characterized the effect of mutational pressure on the ability of an evolving population to grow better ( or worse ) than its ancestor in a variety of chemically novel environments , which contain chemical agents that include heavy metal stressors , antibiotics , or acids . Importantly , our populations were never exposed to any of these conditions during the evolution experiment . A priori , we reasoned that two outcomes were possible . First , populations with high mutation rates may grow better in novel environments , because they can accumulate more beneficial mutations while evolving in their original environment , and these mutations may also be beneficial in novel environments through pleiotropy . High mutation rate populations can also generate more genotypic diversity , which in turn increases the chances that a population harbors a clone with a latent beneficial mutation that allows it to grow better in a novel environment . Such latent beneficial mutations can indeed occur , as demonstrated by the classic fluctuation test , which relies on such mutations to estimate mutation rates towards resistance to lethal selection [96 , 97] . Second , populations with high mutation rates may grow worse in novel environments , because they may accumulate more mutations that are either beneficial or neutral in the current environment , but deleterious in a novel environment . Such latent deleterious mutations do indeed exist [36 , 70 , 98] . In sum , strains with high mutational pressure may harbor more latent beneficial alleles , but also more latent deleterious alleles , and it is not clear a priori which dominates in their effect on fitness . We conducted two tests on how mutational pressure can affect growth in novel conditions . In the first , we measured the growth of eight evolved replicate populations ( two each from MRS , MRM , MRL , and MRXL ) in 96 chemically novel environments . This test did not yield a clear association between mutation rate and growth for our MRS , MRM , and MRL populations . However , it yielded a very clear pattern for our MRXL populations: They were not able to grow in any one of these environments , which illustrates that at the highest mutation rates we consider , latent deleterious mutations outweigh beneficial ones in both the ancestor and evolved populations . One possible explanation is that the MRXL strain is inherently more sensitive to novel environments , including the assay environment . Because the MRXL ancestor population could not grow at all , we were unable to further quantify the effect of the highest mutation rate in these 96 novel environments . In the second test , we periodically measured growth of all 32 replicate populations ( relative to their ancestors ) in two stressful conditions: the antibiotic nitrofurantoin ( a specific narrow stressor ) and an acidic medium ( a broader stressor ) . For both , we found that strains with higher ancestral mutation rates could grow better than those with lower mutation rates , except for MRXL replicate populations , which grew worst of all populations . This experiment shows that latent beneficial alleles may predominate at low and intermediate mutational pressure , but no longer at high mutational pressure . Our observations are consistent with a previous study showing that multidrug resistance in E . coli is favored by intermediate mutation rates [99] . In sum , a modest increase in mutation rates can provide an evolutionary advantage in both the constant environment of our long-term laboratory evolution experiment and in novel environments . This advantage disappears at the highest mutation rates ( U = 0 . 036 ) we considered , where populations show signs of decaying adaptation and poor performance in novel environments . These mutation rates are below those commonly considered to limit adaptation , and highlight the need for additional theoretical work . Our observations show that biological systems may be more sensitive to mutational pressure than simple theoretical models suggest , at least when the effects of mutations are allowed to accumulate over many generations . This observation may improve the prospects of using elevated mutagenesis to drive pathogen or tumor populations to extinction [20 , 100–104] , if high mutation rates can be sustained for a sufficiently long amount of time .
We utilized four isogenic E . coli strains derived from K12 MG1655 that have increasing mutation rates . We refer to these strains as the MRS , MRM , MRL , and MRXL strains , corresponding to small ( S ) , medium ( M ) , large ( L ) , and extra-large ( XL ) mutation rates . Strain genotypes are summarized in Table 1 . We obtained MRM and MRXL strains from our previous experiments ( therein called the single- and double-mutator , respectively ) [35] . MRM has a non-synonymous ( A120T ) mutation in the mutL gene relative to the E . coli wild type . This gene is involved in the methyl-directed mismatch repair system . We previously constructed the MRXL strain by P1 transduction of the dnaQ gene from the E . coli CSH116 strain , which has a non-synonymous mutation , T15I , in the dnaQ gene , into the MRM strain [35] . The dnaQ gene encodes the epsilon subunit of DNA Polymerase III; mutations in this gene can disrupt proofreading . We constructed the mutator strain MRL by replacing the mutL region in MRM with the mutL region from ES4 with a kanamycin resistance gene inserted upstream of the region , using the method of Datsenko and Wanner [105] . We then excised the kanamycin resistance gene using pCP20 [106] , which left a small scar immediately upstream of the mutL gene . We constructed the low mutation rate strain MRS by using P1 transduction to replace the error-prone mutL region in MRM with the wildtype allele from CAG12073 [107] . We confirmed the mutation rates of these ancestral strains using fluctuation tests [108] ( see "Mutation rate measurements and calculations" for details ) , and found that the MRM , MRL , and MRXL strains had 16- , 22- , and 139-fold higher mutation rates to rifampicin resistance than MRS . We also confirmed our manipulations of the MRS , MRM , MRL , and MRXL strains by sequencing their genomes ( S1 Table ) . In additional to these strains , we used the strain E . coli K12 MG1655 , which we obtained from Yale's Coli Genetic Stock Center . See Fig 1 for an overview . We evolved eight independent replicates from populations starting from a single clone for each of the MRS , MRM , MRL , and MRXL strains for 175 days ( 2907 generations ) in 48-well plates ( Fluka 15758-500G-F ) in 2 mL of Davis Minimal broth [111] supplemented with 1000 mg/L glucose ( ‘DM1000’ ) at 37°C with shaking at 400 rpm in a microtiter plate shaker ( Stuart Microtiter 51505 ) . Each plate held 24 populations arranged in a checkerboard pattern , such that each well was surrounded only by wells with blank medium , and the populations were assigned to the 24 wells at random by a custom R script . We diluted each culture 100 , 000-fold every 24 hours into fresh DM1000 medium , which allows almost 17 generations of growth per day . Every 7 days , we archived each evolving population by adding 400 μL of 50% glycerol to 800 μL of stationary phase culture and freezing at -76°C , estimated the cell density by plating subsamples onto LB ( Difco 244620 ) plates with 1 . 5% agar ( Sigma A1296-1KG ) , and froze cell pellets from 800 μL of stationary phase culture for eventual genome sequencing . We delayed the start of the MRS replicates by 63 days for technical reasons . We controlled for contamination in several ways . First , if we observed growth in an empty well , we repeated the transfer from the previous day's 48 well plate stored at 4°C . Second , we periodically checked each evolving culture for contamination by confirming its resistance profile and approximate mutation rate using spot tests . In short , we spotted 25 μL from each evolving culture onto tetracycline ( 10 μg / mL ) and rifampicin ( 100 μg / mL ) plates , and incubated overnight at 37°C . MRS and MRXL replicates can grow on tetracycline , and the replicates with higher mutation rates display more colonies on rifampicin . Cross-contamination occurred once , which prompted us to restart the experiment from the most recent set of uncontaminated glycerol stocks ( day 98 for MRM , MRL , and MRXL replicates; day 35 for MRS replicates ) . Third , we examined the genome sequence data for cross-contamination , but detected no evidence for cross-contamination in it . For populations that do not have a constant number of cells , the effective population size is given by the harmonic mean of population sizes over the course of the dilution and growth cycles of the experiment . Previous studies have estimated the effective population size only from the size of the bottleneck measured during one dilution [112 , 113] . In contrast , because we recorded the census size of the population at carrying capacity ( Nmax ) every 7 days , we were able to estimate the effective population size as the harmonic mean of the population sizes both at the beginning and at the end of a cycle of growth and dilution . To obtain Nmax , d at any one day d , we counted the number of cells in each evolving replicate population in stationary phase just before transferring the population into fresh media . We did so by plating serial dilutions in duplicate on LB agar plates and incubating overnight at 37°C . We discarded plates with fewer than 20 or more than 700 colonies for the purpose of this analysis . Because at the end of each growth cycle we diluted our cultures 100 , 000-fold into fresh medium , a total of G = log2z ( 105 ) = 16 . 61 cell generations ( floor ( log2 ( 105 ) ) ≈16 complete cell generations ) elapsed during each growth cycle , and the minimum population size was Nmin , d = Nmax , d × 10−5 . Thus , during each generation g of each growth cycle , a population assumed population sizes Nd , g={2gNmin , d , if0≤g≤16Nmax , d , ifg=17 . ( Because the precise number of generations in each dilution cycle is log2 ( 105 ) = 16 . 61 , we included the final number of cells Nd , 17 = Nmax , d in this calculation ) . We then determined the nominal effective population size ( Ne ) of a replicate population during its entire lab evolution as Ne=25×18/∑d=125∑g=0181Ng , d which is the harmonic mean of all the population sizes . We calculated it for all 25 days on which we collected population size data . The number 18 corresponds to the total number of generations g for which we computed population sizes during any one of these 25 days . We also estimated the effect of linkage on reducing the effective population size due to background selection or interference selection [53–55 , 114] . To this end , we used the R functions ( GordoNe and GoodNe , respectively , available from [53] ) , where we take the size of the deleterious selective effect as s = 0 . 03 [115] and use U as an upper bound on Ud to obtain rough estimates of linkage's effect on effective population size . We periodically obtained a proxy for the fitness of the evolving strains by measuring growth curves of the archived populations . For each time point , we restarted all evolving populations as well as three replicates from each ancestral population and three replicates of wild type E . coli K12 MG1655 from glycerol stocks in 2 mL of DM1000 and incubated them overnight at 37°C with shaking . We then diluted the overnight cultures 50-fold into 200 μL final volume of DM1000 in 96-well plates ( TPP 92096 ) , and incubated them in a plate reader ( Tecan Infinite Pro F200 ) for 18 hours at 37°C with shaking . During this time , we read the absorbance at 600 nm every 10 minutes . We fit the classic logistic equation describing population growth to the data [116] , using the Growthcurver R package [117] , and defined the relative fitness of each population as revo—ranc . Here , revo is the growth rate of the evolved population and ranc is the mean growth rate of the three replicates of the ancestor grown in the same plate reported in units of cell divisions per hour . We measured each growth curve three times . We used the R package lme4 v1 . 1–12 [118] to perform a linear mixed effects analysis of the relationship between the evolved fitness relative to the ancestor and the mutation rate class . In this analysis , we chose the mutation rate classes as fixed effects , and the replicate population as well as the 96-well plate as random effects . For all linear mixed effects analyses conducted in this paper , we observed no deviations from homoscedasticity according to Levene's test for homogeneity of variance [119] implemented in the R package car v2 . 1–2 [120] . Also , all residuals were normally distributed unless otherwise specified . We obtained significance values using a likelihood ratio test of the full model against a null model that did not contain the fixed effects . Using the data from the above growth curve experiments , we also compared the fitness of the ancestor populations against each other by obtaining the relative fitness of the ancestors as ranc—rK12 , where ranc is the growth rate of the ancestor population and rK12 is the mean growth rate of the three replicates of E . coli MG1655 K12 grown in the same plate . We performed a linear mixed effects analysis of the relationship between the ancestral fitness relative to E . coli K12 and the mutation rate class using the lme4 package , as just described . In this analysis , we chose the mutation rate classes as fixed effects , and the identity of the original glycerol stock and of the 96-well plate as random effects . We used the R package multcomp v1 . 4–6 [121] to test whether fitness values had changed from the ancestral or reference state of 0 . We sequenced samples from the four ancestral populations ( day 0 , generation 7 ) and from each of the 32 evolving replicate populations at days 63 , 119 , and 175 ( generations 1046 , 1977 , and 2907 ) . In total , we thus sequenced 100 populations ( 4 + 3×32 ) . For simplicity , we hereafter designate these time points as generations 0 , 1000 , 2000 , and 3000 . For each , we isolated the DNA directly from cell pellets obtained from the evolving populations using Qiagen's DNeasy Blood and Tissue kit ( cat . No 69582 ) , with modifications as previously described [122] . We used the TruSeq DNA PCR-Free kit ( Illumina FC-121-3002 ) to prepare and barcode the libraries for paired-end sequencing , as previously described [122] . Importantly , we used no PCR steps in preparing the libraries . We employed qPCR with Roche’s FastStart Essential DNA Green Master kit ( Cat no . 06402712001 ) to quantify the libraries , which were then mixed in equimolar amounts for sequencing . We sequenced the populations ( paired-end , 125 bp ) on a single lane of Illumina’s HiSeq 2500 v2 . We used breseq v0 . 26 . 1 [56] to align the reads , and call and annotate the variants relative to the E . coli K12 MG1655 reference genome NC_000913 . 3 [110] , downloaded from the NCBI ( https://www . ncbi . nlm . nih . gov/nuccore/556503834 ) on January 6 , 2015 . We developed scripts in R to identify the alterations that occurred in the evolved populations , but were not fixed in their ancestors . We determined mutational spectra by identifying all mutations that occurred at any detectable frequency in each population at every sequenced time point , and classified them into the following categories: A→C , A→G , A→T , C→A , C→G , or C→T . ( Because DNA is double-stranded , the remaining possible point mutations are covered by their reverse complements , e . g . , T→G corresponds to A→C . ) We computed the relative frequencies of each mutational class for each replicate population , and used these to perform a principal component analysis ( PCA ) in R with prcomp , which uses singular value decomposition for the PCA . All data are available from the Dryad Digital Repository: https://doi . org/10 . 5061/dryad . mh206 . To quantify the movement and spread of a population as a “cloud” of sequences in sequence space , we first defined the center of this mutational cloud at any given site n in the genome as the majority allele en∈{A , C , G , T} , i . e . , the allele whose frequency p exceeded 0 . 5 ( all sites we analyzed had one such allele ) . We defined the center of the mutational cloud of genomes as the location in genotype space defined by the majority allele at each site . It can also be viewed as the location of the population’s consensus sequence . If we denote the fraction of a population with the majority allele at site n as p , then the distance of the population to a given majority allele at site n can be thought of as the fraction of the population not having the majority allele , which is given as Cn = ( 1—p ) . We define the population spread metric C as the average of Cn over all sites in the genome . A related quantity is the approximate sequence distance D that an evolving population has moved from its ancestral genotype , i . e . , D=∑n{0 , ifan=en1 , otherwise where an∈{A , C , G , T} is the ancestral allele of the population at generation 0 . In other words , D corresponds to the total number of sites at which the majority allele is different from the ancestral allele . We also computed each population's average genome-scale nucleotide site diversity [123 , 124] using the pairwise alignment position nucleotide counting approach [125 , 126] . We estimated the proportion of pairwise nucleotide differences at each site n as pn=mp ( 1−mp ) m ( m−1 ) /2 , where mp is the number of reads corresponding to the majority allele and m is the total number of reads at site n . We estimated the average nucleotide diversity for the L positions in our genome having non-zero coverage as π=∑n=1LPnL . We used the R package lme4 v1 . 1–12 [118] to perform a linear mixed effects analysis of the relationship between the cube root of C or D ( taken to ensure homoscedasticy ) and the mutation rate class . In this analysis , we chose the mutation rate classes as fixed effects , and the time points and each of the 32 evolving replicates as random effects . We obtained significance values using a likelihood ratio test of the full model against a null model that did not contain the fixed effects . We identified putatively beneficial mutations as mutations that occurred in a genomic region more often than one would expect by chance alone . To identify such mutations , we used a numerical approach that focuses on a given gene g among a larger set of genes or genomic regions G ( e . g . , a gene among the set of all genes ) , and asked whether more replicate population experienced a high-frequency genetic change than expected by chance . To this end , we first counted the number ng of replicate populations with a mutation in gene g that had reached a frequency greater than 50% at generation 3000 . If all sites in the genomes of all samples were equally likely to experience a mutation , and if different genes were likely to experience mutations only in proportion to their length , then the probability pg that any one gene g receives such a mutation in any given replicate would depend only on the length of the gene lg , pg=lg∑γ∈Glγ . The ng total mutations found in gene g could be distributed in ( 32ng ) ways across the 32 replicate populations R = {MRS1 , …MRS8 , MRM1 , …MRM8 , MRL1 , …MRL8 , MRXL1 , …MRXL8} . For example , for ng = 2 , the mutations could be distributed across the 32 replicate populations in ( 322 ) = 496 ways , i . e . , one could occur in MRS1 and the other in MRS2 , one could occur in MRS1 and the other in MRS3 , etc . We computed the probability of observing the ng mutations in any given set of replicates as the probability that gene g was mutated in each member of the set of replicates times the probability that it was not mutated in any of the other replicates . For ng = 2 and replicate populations ri , rj ∊R , this quantity is given by the binomial distribution adjusted to account for the number of observed mutant genes in the replicate populations , nri and nrj pri , rj=nrinrjpgng ( 1−pg ) ( ntot−nri−nrj ) , where ntot=∑r∈Rnr . The probability of observing exactly ng = 2 mutations in gene g in any pair of replicate populations is the sum of the probabilities that ng mutations occurred in each of the 496 pairs , and is given by Pg=∑ri , rj∈R , ri≠rjpri , rj . This quantity Pg is our null expectation that two replicates acquire mutations in gene g , if each replicate population's mutations were randomly distributed across its genome . We were interested in genes containing mutations in improbably many replicate populations , which we identified as those genes having less than a 0 . 005 percent chance of finding ng replicate populations with a high frequency derived allele in the gene . We performed analogous analyses for ng > 2 . For example , for genes in which we observed three replicates with mutations ( ng = 3 ) , we computed the probability that three replicates ri , rj , and rk each contained a mutation in gene g as pri , rj , rk=nrinrjnrkpgng ( 1−pg ) ( ntot−nri−nrj−nrk ) . Similarly , the probability of observing a mutation in exactly 3 replicate populations is given by Pg=∑ri , rj , rk∈R , ri≠rj≠rkpri , rj , rk . We estimated the mutation rate of a single clone isolated from each ancestor and from each evolved replicate population through fluctuation assays that screened for mutants resistant to rifampicin [127] , which can be caused by mutations in the rpoB gene . Specifically , we performed the following procedure for each replicate population . We isolated a single random clone and incubated it overnight in 2 mL DM1000 in 48 well plates at 37°C with shaking . We diluted the resulting overnight culture 100 , 000-fold to yield a culture with approximately 1000 cells in 100 μL . We then transferred 100 μL of the diluted culture into 5–7 sterile 50 mL tubes ( Sarstedt 62 . 547 . 254 ) containing 30 mL of DM1000 , and incubated for 48 hours at 37°C with shaking . We estimated the number of cells in each tube by plating dilutions on LB agar plates , and estimated the number of resistant cells in each tube by plating dilutions on LB agar plates supplemented with 50 mg/mL rifampicin ( Sigma R3501-5G ) . We calculated the mutation rate to rifampicin μrif using the method and program provided by Philip Gerrish [108] . We obtained the genomic mutation rate U using Drake's approach [74] by first determining the "correction factor" C , which counts the number of single nucleotide mutations in rpoB that show rifampicin resistance . By counting all possible nucleotide changes underlying the amino acid changes in rpoB previously shown to confer rifampicin resistance [57] , we determined that C = 71 . Finally , we estimated the genomic mutation rate as U = Lμrif/C , and the mutation rate per base pair as μ = μrif/C , where L = 4641652 is the number of nucleotides in the E . coli K12 genome . This mutation rate may be an underestimate because we neglected other types of mutations ( e . g . , indels ) and mutations in other genes that may lead to rifampicin resistance . Our phenotype screening revolved around the density of cells after growth in various chemicals . Specifically , we determined the cell density after 24 hours of growth using Biolog Phenotype MicroArrays PM11C , PM12B , PM13B , and PM14A MicroPlates ( Biolog , Inc . , Hayward CA , USA ) , which assay the sensitivity of bacteria to diverse chemicals that range from antibiotics to heavy metals . We screened the ancestors at generation 0 , and two randomly selected replicates of the evolved populations ( MRS1 , MRS6 , MRM3 , MRM4 , MRL2 , MRL7 , MRXL3 , MRXL4 ) at the final time point of laboratory evolution . To do so , we streaked population samples from glycerol stocks onto LB agar plates , incubated them at 37°C for 24 hours , restreaked the resulting colonies onto fresh LB agar plates ( 37°C , 24 hours ) , and repeated this streaking and incubation procedure once more . We resuspended the colonies from the final ( third round ) plates in IF-0 solution ( Biolog , Inc . , Hayward CA , USA ) to a final absorbance reading at 600 nm of approximately 0 . 18 ( 200 μL suspension in a 96-well TPP plate , ref 92096 ) . We diluted this suspension 6-fold using IF-0+dye ( Biolog , Inc . , Hayward CA , USA ) , and diluted the resulting suspension 201-fold using IF-10+dye ( Biolog , Inc . , Hayward CA , USA ) . We added 100 μL of the final solution to each well of a Phenotype MicroArray and incubated the array in the dark at 37°C for 24 hours , taking absorbance readings at 600 nm after 10 minutes and 24 hours . We then computed the Biolog phenotype BS , C = AS , C , 24h−AS , C , 10m , where AS , C , 24h and AS , C , 10m are the absorbance readings for each sample S and compound C at 600 nm after 24 hours and 10 minutes . To determine the minimum threshold for detection of growth in a given compound C , we computed the absolute difference between the readings in a given well across all pairs of samples ( i , j ) after 10 minutes ( before cells had started to grow and divide ) , i . e . , Ai-j , C , 10m = |Ai , C , 10m−Aj , C , 10m| . The values of Ai-j , C , 10m quantify the expected experimental noise of wells with no growth . We found that Ai-j , C , 10m<0 . 097 for more than 99% of sample pairs . Based on this observation , we considered differences between readings smaller than the threshold value Athresh = 0 . 097 as due to experimental noise . Each compound in the Biolog Phenotype MicroArrays we used occurs in four wells at increasing concentrations . For further analysis , we used data only from the concentration ( the well ) that showed the highest variation in the difference between matched evolved and ancestor strains across all samples . We considered a sample to have evolved tolerance to a compound C if it improved its phenotype after 3000 generations of evolution more than expected based on experimental noise , i . e . , BSEvo , C − BSAnc , C > Athresh . Likewise , we considered that a sample had lost tolerance if its phenotype had degenerated after 3000 generations of evolution , i . e . , if BSAnc , C − BSEvo , C > Athresh . We note that both cellular growth and respiration contribute to the Biolog phenotype BS , C , because respiration can occur independently of cellular growth [70 , 128] . We were also interested in observing the evolutionary dynamics of phenotypes over time . The phenotypes we selected for this analysis are the cell density after 24 hours of growth of the evolved populations relative to their ancestors in two conditions: a narrow antibiotic ( nitrofurantoin ) stress , and a broader environmental ( low pH ) stress . Specifically , we chose DM1000 medium with 1 . 5–2 . 4 μg/mL nitrofurantoin for the narrow antibiotic stress , and acidic DM1000 ( pH 4–5 . 25 ) for the broad stress . Nitrofurantoin is one of the phenotypes where evolved populations gained tolerance in the Biolog analyses , and acid stress has been well-studied in E . coli [72] . To control for changes in cell density at stationary phase , we also performed a control measurement in the standard medium , DM1000 . Specifically , we measured the growth of evolved replicate populations at days 28 , 63 , 91 , 119 , 147 , and 175 ( generations 465 , 1046 , 1511 , 1977 , 2442 , and 2907 , hereafter designated as generations 500 , 1000 , 1500 , 2000 , 2500 , and 3000 ) . To do so , we inoculated each population in triplicate from glycerol stock in DM1000 and grew it at 37°C for at least 18 hours . We then diluted the resulting culture 50-fold into DM1000 medium immediately before adding 10 μL from the diluted culture to 190 μL media with nitrofurantoin , low pH , or just DM1000 in a 96-well plate . We incubated the resulting 96-well plates for 24 hours , and then measured the absorbance at 600 nm . We computed the normalized fold change in cell density in each condition at six time points by obtaining the average value of G = ( AX , E / AX , A ) / ( ADM1000 , E / ADM1000 , A ) for the three replicate cultures , where AX , Y is the absorbance reading in condition X ( e . g . , 2 . 2 μg/mL nitrofurantoin or acidic DM1000 medium ) of a given replicate before evolution ( Y = A for ancestral ) or after evolution ( Y = E for evolved ) . The denominator , ( ADM1000 , E / ADM1000 , A ) , removes the effect of changes in the evolved carrying capacity , which otherwise could confound cell density changes observed in stressful media with evolved cell density changes in the medium without stressor . We considered that evolution had increased cell density relative to the ancestor when the numerator of G was greater than its denominator , ( AX , E / AX , A ) > ( ADM1000 , E / ADM1000 , A ) . In order to quantify the relationship between the normalized fold-change in cell density G and ancestral mutation rate , we performed a linear mixed effects analysis using the R package lme4 v1 . 1–12 [118] to obtain the relationship between G and the mutation rate class . In this analysis , we chose the mutation rate classes ( MRS , MRM , MRL , and MRXL ) as fixed effects , and the measurement time points , the experimental condition , and each of the 32 evolving replicates as random effects . For the nitrofurantoin and pH stressors , we used data from the experimental condition with the most variability between replicates ( 2 . 3 μg/mL nitrofurantoin and pH 5 . 25 ) in this analysis . We tested for homoscedasticity using the R package car v 2 . 1–2 [120] , and found that the untransformed pH data and the log-transformed nitrofurantoin data were homoscedastic ( Levene's test , pH 5 . 25: F3 , 188 = 1 . 71 , p = 0 . 17; nitrofurantoin 2 . 3 μg/mL: F3 , 188 = 1 . 25 , p = 0 . 29 ) .
|
Mutation is of central importance in biology . It creates genetic variation , the raw material of evolution by natural selection , It can improve traits and organisms , but can also lead to phenomena like cancerous cells and antibiotic resistant pathogens . Increasing the mutation rate can accelerate evolutionary adaptation , even over many thousands of generations in a constant environment . Our study describes the laboratory evolution of asexual Escherichia coli strains with a range of mutation rates at levels found in the wild ( from wild type to strong mutator ) . Unexpectedly , evolutionary adaptation was most limited in the populations with the highest mutation rate . Our work suggests that deleterious mutations can begin to limit adaptation at lower mutation rates than previously thought .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"genome",
"evolution",
"population",
"genetics",
"oncology",
"mutation",
"nonsense",
"mutation",
"evolutionary",
"adaptation",
"population",
"biology",
"population",
"density",
"molecular",
"evolution",
"evolutionary",
"genetics",
"carcinogenesis",
"population",
"metrics",
"evolutionary",
"processes",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"computational",
"biology"
] |
2018
|
High mutation rates limit evolutionary adaptation in Escherichia coli
|
This study evaluates the diagnostic accuracy and cost-effectiveness of the Kato-Katz and Mini-FLOTAC methods for detection of soil-transmitted helminths ( STH ) in a post-treatment setting in western Kenya . A cost analysis also explores the cost implications of collecting samples during school surveys when compared to household surveys . Stool samples were collected from children ( n = 652 ) attending 18 schools in Bungoma County and diagnosed by the Kato-Katz and Mini-FLOTAC coprological methods . Sensitivity and additional diagnostic performance measures were analyzed using Bayesian latent class modeling . Financial and economic costs were calculated for all survey and diagnostic activities , and cost per child tested , cost per case detected and cost per STH infection correctly classified were estimated . A sensitivity analysis was conducted to assess the impact of various survey parameters on cost estimates . Both diagnostic methods exhibited comparable sensitivity for detection of any STH species over single and consecutive day sampling: 52 . 0% for single day Kato-Katz; 49 . 1% for single-day Mini-FLOTAC; 76 . 9% for consecutive day Kato-Katz; and 74 . 1% for consecutive day Mini-FLOTAC . Diagnostic performance did not differ significantly between methods for the different STH species . Use of Kato-Katz with school-based sampling was the lowest cost scenario for cost per child tested ( $10 . 14 ) and cost per case correctly classified ( $12 . 84 ) . Cost per case detected was lowest for Kato-Katz used in community-based sampling ( $128 . 24 ) . Sensitivity analysis revealed the cost of case detection for any STH decreased non-linearly as prevalence rates increased and was influenced by the number of samples collected . The Kato-Katz method was comparable in diagnostic sensitivity to the Mini-FLOTAC method , but afforded greater cost-effectiveness . Future work is required to evaluate the cost-effectiveness of STH surveillance in different settings .
The reliable mapping , surveillance and evaluation of infectious diseases relies upon two key factors: ( i ) accurate methods of diagnosis and ( ii ) optimal strategies to sample the population . For the soil-transmitted helminths ( STH: Ascaris lumbricoides , Trichuris trichiura and hookworm ) , the commonly used diagnostic technique is the Kato-Katz method [1] . This technique allows for the quantification of intensity of infection on the basis of fecal egg counts . Whilst this method is used widely due to its simplicity and need for minimal equipment , it has low sensitivity arising mainly from the non-random distribution of eggs in stool and day-to-day variation in egg output [2]–[7] . The sensitivity of the method is improved by duplicate readings of samples and collecting samples over consecutive days [8] , but this increases effort and cost [9] . An alternative to the Kato-Katz method is a new flotation and translation-based technique , the FLOTAC method [10] , which exhibits greater sensitivity for detecting STH species compared to the Kato-Katz method [11]–[14] . However , FLOTAC requires use of a centrifuge which may be unavailable in field laboratories and also consists of more procedural steps . The recently developed Mini-FLOTAC overcomes this constraint and includes a closed chamber for flotation and mixing , and a separate reading disc [15] . A study in Tanzania and India demonstrated that Mini-FLOTAC was more sensitive for STH diagnosis than either a direct smear or the formol-ether concentration technique , while other work has shown Mini-FLOTAC and Kato-Katz to be comparable for hookworm diagnosis in a very high prevalence setting in Tanzania [16] . The choice of diagnostic method should not only take into account ease of use and test performance but should also consider costs [17] . Previous studies have examined the costs of alternative methods for the diagnosis of clinical malaria [18]–[20] , but few studies have been conducted for helminth diagnosis . A cost analysis of FLOTAC and Kato-Katz in Zanzibar showed that the additional time requirements for FLOTAC preparation and the specialist equipment required resulted in higher costs compared to the Kato-Katz method [9] . Costs will also be influenced by the sampling platform for the collection of stool samples . Surveys of STH are typically conducted in schools since school-aged children are the natural targets for control and because of the practical advantages of conducting school surveys [1] , [21] . However , alternative platforms to schools have recently been proposed for STH surveys , including household surveys conducted as part of the transmission assessment surveys ( TAS ) used to assess whether lymphatic filariasis is below a pre-defined prevalence threshold [22] . In order to inform the choice of sampling strategy , there is a need to evaluate the relative cost of school-based surveys compared to community surveys . In the present study , we evaluate the diagnostic accuracy and cost and cost-effectiveness of the duplicate Kato-Katz and Mini-FLOTAC methods in western Kenya where mass treatment had recently been provided as part of the national school deworming programme . Such a low transmission setting will become increasingly important as control programmes effectively reduce infection levels . We use Bayesian latent class models to estimate test sensitivity and specificity in the absence of a gold standard [23]–[25] . This approach has the advantage that it can adjust for the conditional dependence between tests that are based upon a common biological phenomenon ( direct observation of eggs ) [26] . Our economic analysis not only evaluates the costs and cost-effectiveness of diagnosis , but also explores the cost implications of collecting samples during school surveys compared to household surveys .
The collection of stool samples and cost data was embedded in a larger study investigating the impact of deworming on malaria-specific immune responses and risk of clinical malaria ( ClinicalTrials . gov: NCT01658774 ) in 20 schools . Written informed consent from child participants was provided by a parent or guardian on the child's behalf . Ethical approval was obtained from the Kenya Medical Research Institute and National Ethics Review Committee ( SSC No . 2242 ) , the London School of Hygiene and Tropical Medicine ( LSHTM ) Ethics Committee ( 6210 ) , the Makerere School of Public Health , Institutional Review Board ( IRB00005876 ) . The study took place in Bumula District ( 0 . 52747 , 34 . 4395 ) , Bungoma County , Western Province , Kenya ( Figure 1 ) during July 2013 . The district is located at 1320 m elevation . Rainfall ( annual average of 2428 mm ) is seasonally bimodal , with the long rains occurring from March–May and the short rains from October–December . Average annual minimum and maximum temperatures are 11 and 24°C , respectively . The population of the area consists of indigenous Bukusu people and numerous Luhya who settled in recent years . The economy is primarily rural subsistence agriculture , with some families growing sugar cane as a cash crop . Cattle and sheep are commonly kept . The population is serviced by Bumula District Hospital , which has a catchment area of about 180 , 000 people and approximately 250 km2 . Historically , STH infections were highly prevalent in western and coastal Kenya [27] , [28] but a national school deworming programme launched in 2009 has helped to reduce infection levels . As part of this programme , school children were treated with 400 mg of albendazole in June 2013 , and thus the study was conducted 20–36 days following delivery of mass treatment delivered through schools . Drug efficacy was not formally investigated as the treatment fell outside of the WHO recommended 14–21 days for assessing anthelmintic drug efficacy [29] . Eighteen schools with the highest prevalence of STH infection , assessed during pre-treatment screening surveys conducted in January 2013 , were included in the present study . Data collection was originally planned to take place in schools , however a national teachers strike , June 25–July 17 , meant this was not initially possible . So not to delay work , it was decided to collect stool samples from the homes of children enrolled in school for 14 of the 18 schools , with children found to be infected with STH species in the previous screening surveys purposively sampled ( household sampled , n = 504 ) . Once the schools reopened , samples from children purposively selected in the remaining four schools were collected at the schools ( school sampled , n = 148 ) . These differences in sample collection provided the unforeseen opportunity to estimate the cost of sampling in both households and schools . In a subset of children ( n = 233 ) , stool samples were collected over two consecutive days to evaluate if multiple sampling improved test performance . Each stool sample was examined using the Kato-Katz method and the Mini-FLOTAC method . The Kato-Katz method was performed using a 41 . 7 mg template , according to the WHO recommendation , and examined in duplicate with different technicians reading each sample . Mini-FLOTAC was performed using 2 g of stool and flotation solution ( FS ) 2 ( saturated sodium chloride ) to detect STH [30] . For details on how to conduct each method , see [31] . The intensity of infection was expressed by eggs per gram ( EPG ) of feces . For the Kato-Katz method , a multiplication factor of 24 was used . For Mini-FLOTAC , a multiplication factor of 10 was used , based on calculation dilution ratio/volume , where the dilution ratio is 2 grams of faeces to 38 ml of flotation solution , or 2∶40 ( 1∶20 ) , and the volume read in the reading disk is 2 ml . 1∶20/2 ml . Quality control was performed on 10% of all samples , where they would be re-examined by a second microscopist to check for discrepancies . Any discrepancies necessitated examination by a third microscopist . Results were recorded by hand on data collection sheets and entered into Microsoft Excel version 12 ( 2007 , Microsoft Corporation; Redmond , WA , USA ) . Statistical analysis was performed on STATA version 10 ( College Station , TX , USA ) and WinBUGs 1 . 4 . 1 software ( Imperial College and MRC , UK ) . Analysis was based on Bayesian latent class modeling , which is increasingly used to evaluate diagnostic sensitivity for a number of parasite infections , especially in the absence of a ‘gold standard’ reference test [32]–[34] . They are particularly well suited to such problems as they can incorporate prior scientific information about the sensitivities and specificities of the tests and the prevalence of the sampled population , thus overcoming problems of non-identifiability [26] , [35]–[37] and can be expanded to account for conditional dependence between tests [35] . In our analysis , each school/community is considered as a separate population k with its own ( true but unobserved or latent ) infection prevalence ( πk ) . Each population is subjected to two diagnostic tests , j ( j = 1 , 2 ) ; + and denote positive and negative test results from test j , and and denote true numbers of infected and non-infected . We define Sj and Cj to be the sensitivity and specificity of test j where and ; common sensitivities and specificities of each diagnostic test are assumed across all populations , and in the first instance are assumed to be conditionally independent . Results from each diagnostic test were cross-classified , and the joint distribution assumed to be multinomial with four categories corresponding to all possible combinations of the results in two tests . The multinomial probabilities were expressed as functions of the true prevalence of infection and of the sensitivities and specificities of the two tests . Sensitivity and specificity over two days was considered a direct function of one day sensitivity/specificity . As both tests are based upon a common biological phenomenon ( direct observation of eggs ) they can be considered conditionally dependent , which must be accounted for in order to obtain unbiased estimates of test accuracy . The models were thus extended to include covariance between tests for infected individuals and for non-infected individuals , following the method of Dendukuri and Joseph [35] . A detailed description of the model is given in Supplementary Information S1 . The diagnostic performance of the methods was further assessed in terms of positive predictive value ( PPV , proportion of true positive results detected ) , negative predictive value ( NPV , proportion of true negative results detected ) and accuracy ( proportion of readings that have given a valid result ) based upon modeled prevalence , sensitivity and specificity . Finally , the comparison of methods in estimating EPG was made using the Wilcoxon Rank Sum test . Financial and economic costs were estimated for diagnosis using both the Kato-Katz and Mini-FLOTAC methods . Since both household and school sampling were undertaken , costs were also estimated for each sampling method . Costs were estimated using an itemized , ingredients-based approach where individual costs and quantities were recorded [38] . Quantities used for each of the categories were obtained through observation in the field and from accounting records provided by KEMRI . Evaluation of costs was undertaken from the perspective of the provider , here defined as the government . The time frame was for one round of surveillance: nine days for household sampling and four days for school sampling . A wastage value of 10% was applied to all consumables . The costs for all activities were categorized into four categories: personnel , materials , transport and facility . Financial and economic costs were classified separately for each of the cost categories . Financial costs are those that represent the accounting cost of a good or service , representing the actual amount paid . Economic costs can represent opportunity cost , meaning the benefits forgone of a resource not being used in its next best alternative use [39] . Financial and economic costs were combined to provide an overall cost for personnel , materials , transport and facility , for each sampling and diagnostic method . Costs were collected in Kenyan Shillings and converted to US dollars using an average of the last year of exchange rates , which ranged from $82 . 23 to $86 . 80 ( www . oanda . com ) . No annualization or discounting was made due to the time frame of one round of surveillance . All costs obtained were for the year 2013; accordingly no inflation or deflation factor was used in the analysis . Current guidelines for STH control focus on the prevalence levels of any STH species , rather than individual STH species . Therefore , cost-effectiveness was calculated for any STH species rather than for each species . Three outcomes were estimated: ( i ) cost per sample tested; ( ii ) cost per case of STH infection detected by each test; and ( iii ) cost per STH infection correctly classified . Cost per case tested was defined as the total cost of sampling and diagnostic activities per individual tested in the given scenario . Cost per case of STH infection detected was calculated by dividing the total costs for each diagnostic and sampling scenario by the number of positive cases identified in each scenario . Cost per STH infection correctly classified was estimated by dividing the cost per child tested by the related diagnostic test accuracy as estimated in the latent class model . Probabilistic sensitivity analysis ( PSA ) allows simulation of a model where uncertain input parameters are sampled within their specified distributions , assessing the combined effect of parameter uncertainty on outcome measures . PSA was conducted to determine how variance in key input parameters affected the cost outcomes of the four survey scenarios . PSA was applied to the cost per case tested , cost per case detected and cost per infection correctly classified . A 10% variance was applied to salaries and per diems of laboratory technicians and per diems of field workers . Cost per case detected will be influenced by the underlying prevalence of infection , therefore prevalence was varied within a distribution of 0–80% . The number of samples collected and diagnosed was varied within the observed maximum and minimum values . Diagnostic test sensitivity and accuracy were also varied within a distribution of observed values in this study and previous studies [16] , [30] , [40] . Microsoft Excel and Palisade @Risk ( www . Palisade . com ) were used in the analysis , and @Risk simulations of 1000 iterations were conducted for the PSA of each surveillance scenario .
Overall , 93 of the 657 samples were positive for at least one STH species on both tests ( 14 . 2% ) , 485 samples tested negative for any STH species on both tests ( 73 . 8% ) , 45 samples tested positive with Kato Katz and negative with Mini-FLOTAC ( 6 . 8% ) and 34 tested negative with Kato Katz and positive with Mini-FLOTAC ( 5 . 2% ) . When considering combined results from consecutive days , children were classified as infected if they were positive on either day , and uninfected if testing negative on both days: 23 . 5% of the 132 children included were classified as positive for any STH species by both tests , 60 . 6% as negative by both tests , 9 . 1% as positive by Kato Katz only and 6 . 8% by Mini-FLOTAC only . Estimates of sensitivity and specificity for both tests ( both singly and over two consecutive days ) are provided in Table 2 and further data on PPV , NPV and test accuracy are provided in Supplementary Information . For all three individual STH species , sensitivity and specificity for each test are comparable , lying between 47 . 3% and 53 . 3% when conducted just once , and increasing to 72 . 2–78 . 2% when conducted on two consecutive days . For diagnosis of infection with any STH species over two consecutive days , the sensitivity estimates for Kato Katz and for Mini-FLOTAC were 76 . 9% ( 95% Bayesian credible interval ( BCI ) : 62 . 2–88 . 3% ) and 74 . 1% ( 95% BCI: 59 . 8–86 . 6% ) , respectively . Estimates of specificity were generally above 93% , and even 99% in some instances . Significant correlations between the test outcomes for all three STH species in both infected ( ρ = 0 . 37 [95% BCI: 0 . 06–0 . 59] for any STH species ) and uninfected children ( ρ = 0 . 52 [95% BCI: 0 . 03–0 . 84] ) suggest that the two tests were conditionally dependent , highlighting the inappropriateness of using a combined reference standard for evaluation . Model results also suggest that PPV , NPV and accuracy do not differ significantly between tests for all three STH species . For A . lumbricoides and T . trichiura , for which positivity rates were very low ( 26 and 6 of 657 samples by Kato Katz , 24 and 9 of 657 samples by Mini-FLOTAC , respectively ) , there was no notable increase in accuracy when the tests were repeated over two consecutive days . However , for hookworm ( and thus any STH species ) , increased sensitivity resulted in some improvement in accuracy when tests were repeated over two consecutive days: for any STH , accuracy increased from 79 . 0% ( 95% BCI: 67 . 1–87 . 3% ) to 85 . 7% ( 95% BCI: 76 . 3–93 . 3% ) for Kato Katz and from 78 . 8% ( 95% BCI: 67 . 5–87 . 5% ) to 85 . 9% ( 95% BCI: 76 . 3–93 . 3% ) for Mini-FLOTAC , although notably the 95% BCI do overlap . To demonstrate the potentially large influence of prior distributions and assumptions on model parameters and model fit , a sensitivity analysis focusing on the prior distributions of tests' sensitivity and specificity was conducted . Less restrictive prior distributions on specificity resulted in lower values of sensitivity , but did not improve model fit . Choice of prior had less influence on sensitivity and did not qualitatively influence parameter values . Adjusting for conditional dependence for the combined result of two consecutive tests did not improve the model fit , and neither did fitting unique sensitivity/specificity values . Finally , allowing sensitivity and specificity to vary as a function of prevalence did not improve model fit , and resulting parameter estimates remained comparable . Analysis of EPG estimates demonstrated that there was no statistical difference in the intensity of hookworm infection estimated using the two methods on single sample ( z = 0 . 506 , p = 0 . 612 ) . Mean EPG increased for each STH species when sampling occurred over two days , but this difference was non-significant ( z = −1 . 78 , p = 0 . 082 ) . Table 3 presents the costs per child tested , cost per case detected ( by the given method ) and cost per case correctly classified ( based on Bayesian latent class modelling ) for each diagnostic and sampling scenario , for both single and consecutive day sampling . The Kato-Katz method was cheaper than the Mini-FLOTAC , regardless of the sampling approach ( US$ 10 . 14 vs . US$ 13 . 11 for school-based sampling and US$11 . 99 vs . US$14 . 96 for community-based sampling ) . Two day sampling doubled the costs per child tested . Table 3 also presents the number of cases which tested positive for at least one STH , based on each survey diagnostic and sampling scenario and the associated cost per positive case detected . Again , the Kato-Katz method was cheaper than the Mini-FLOTAC method for each sampling method . Interestingly , community-based sampling was associated with lower costs than school-based sampling , due mainly to a greater number of positive cases identified under the community-based sampling . Latent class analysis of each diagnostic test produced accuracy estimates ranging from 78 . 8 to 85 . 9% ( Supplementary Information S3 ) . Accuracy for each test was then incorporated in the costing analysis to generate a cost per STH infection correctly classified ( either as negative or positive ) ( Table 3 ) . The cost per case correctly classified was higher than the cost per child tested , but the relative cost-effectiveness of the different diagnostic and sampling scenarios remained comparable , with the Kato-Katz method being the most cost-effective diagnostic method . The tornado diagram in Figure 2 presents the percentage change in the cost per child tested and cost per case correctly classified for the input parameters of interest in each survey scenario . Cost per child tested in community sampling was most strongly influenced by the number of household samples collected , while in school-based sampling the cost per child was most strongly influenced by the number of samples diagnosed with either Kato-Katz or Mini-FLOTAC . A similar pattern of parameter influence on cost-effectiveness was observed for the cost per case correctly classified ( data not shown ) . The additional influence of the diagnostic accuracy parameter was noted , although less influential than the aforementioned parameters . Uncertainty in the prevalence estimate for all four scenarios affected most the cost per case of any STH species detected . Sensitivity analysis also demonstrated that the cost per case detected decreased for each scenario as prevalence rates of any STH increased , with costs rapidly declining and eventually reaching a cost threshold ( Figure 3 ) .
Suitable surveillance methods are needed to accurately estimate prevalence and intensity of infection , and thus guide disease control programming and track progress towards programme goals . Current recommendations for STH surveys include school-based sampling and parasitological diagnosis using the Kato-Katz method [1] , although recently Mini-FLOTAC has been proposed as an alternative technique . Our use of Bayesian latent class modeling showed that in a post-treatment setting in western Kenya , the Kato-Katz and Mini-FLOTAC methods exhibited comparable diagnostic accuracy for detection of any STH species over single and consecutive day sampling . Furthermore , our economic analysis showed that use of the Mini-FLOTAC method was more costly and less cost-effective , whether the samples were collected through school surveys or through household sampling . An advantage of the Mini-FLOTAC method is that it reduces the exposure of the technician to the sample fecal matter , but has the disadvantage of requiring different flotation solutions for STH and for Schistosoma mansoni , increasing costs further . In contrast , S . mansoni can be read alongside STH using the Kato-Katz method . The observed sensitivity of each method is lower than that presented in a previous study comparing Kato-Katz and Mini-FLOTAC , which reported a sensitivity of up to 91% and 97% for detecting hookworm using Kato-Katz and Mini-FLOTAC methods respectively [16] . There are two likely reasons for the lower accuracy reported here . First , Barda et al conducted their study in a treatment-naïve high intensity setting , whilst infection intensities were considerably lower in the current study . It is widely acknowledged that sensitivity of coprological techniques can be poor in low infection intensity settings . Second , previous analyses have relied on using a combined reference standard . Without a reliable gold standard , the true infection status of a population is unknown , and accordingly , sensitivity and specificity cannot be estimated directly , thus introducing bias in comparing the accuracies of new diagnostic tests . This is especially true when the tests are based on the same biological phenomenon and thus likely to be highly correlated [23] , [26] . Bayesian latent class models have the advantage of overcoming this bias by allowing for the estimation of accuracy when true infection prevalence is unknown , under the assumption that sensitivity and specificity are the same in all tested populations . However , intensity of STH and therefore diagnostic sensitivity may differ between populations . It should be noted that simulation studies investigating this approach have shown that when there is a true difference in test sensitivity between populations , results will be biased towards the sensitivity of the test in the population with the highest infection prevalence , thus potentially over-estimating sensitivity in low prevalence settings [41] . Additionally , sensitivity will likely vary across STH species . However , this analysis focused on sensitivity and costs for detection of any STH as that is of primary relevance to STH control and treatment programmes . The finding that the cost per child tested is lower for Kato-Kato method compared to the Mini-FLOTAC is consistent with a previous costing study in Zanzibar [9] . Our sensitivity analysis illustrated the non-linear trend of decreasing costs with increasing prevalence of infection , with significantly high costs estimated at prevalences below 10% . This result highlights that identifying positive cases will become more expensive as control programmes are successful in reducing infection levels , and programmes and funders should be aware that surveillance costs may increase over the life of a control programme . Our study additionally provides , for the first time , insight into the cost-effectiveness of diagnosis as previous studies have only estimated costs , and shows that the Kato-Katz method has greater cost-effectiveness in correctly classifying infection status . Previous research on the cost-effectiveness of helminth surveillance has focused on the geographical targeting at the start of control programmes , either in diagnosing individuals infected with S . haematobium [42] or sampling strategies for identifying schools requiring mass treatment for S . mansoni [43] and STH [44] . The work by Sturrock and colleagues [43] , [44] employed Monte Carlo simulation to derive a pseudo gold standard data set , using parameters from empirical data in order capture the spatial and demographic heterogeneities in infection patterns . A similar simulation approach was employed by Smith and colleagues [45] who evaluated the performance of different sampling methods for trachoma surveys . We suggest that the inclusion of diagnostic accuracy , using Bayesian latent class modelling , and the collection of assocated cost data would be an important advance in evaluating the cost-effectiveness of surveillance strategies for STH and other negelcted tropical diseases ( NTDs ) . For example , a combination of simulation , cost-effectiveness and field studies could provide useful insight into the value of transmission assessment surveys for lymphatic filariasis [22] in assessing STH in different epidemiological and programmatic settings . Notwithstanding the value of our adopted approach , a number of study limitations are worth highlighting . Diagnosis by any microscopy technique is labour intensive and inevitably incurs human error . Although study technicians were rotated to retain alertness , the number of slides processed may still lead to reading errors . This is particularly likely when hookworm is present , as the eggs desiccate after 20–40 minutes in Kato-Katz thick smears [46] . Further limitations are that costs may have been underestimated because personnel performed multiple duties , with technicians undertaking both sample collection and diagnostic preparation . The cost estimates are also limited in their generalizability to an extended time frame of surveillance , as the short survey rounds ( four and nine days ) cannot provide an estimate of long-term costs . The different sampling methods used ( school vs . community-based ) suggest differences in cost but not in the parasitological profile of sampled children . The number of positive cases detected in school or community-sampling is unrelated to factors such as school enrolment , as the children sampled by either method were from the same study population , either sampled in their homes or at the schools themselves . Additionally , the distances travelled from school or community sampling locations to the diagnostic facility did not vary enough to alter transport costs . Daily fuel expenditure was equal across both sampling methods , so fuel costs were not significantly influenced by variation in sampling method . Finally , the recent national deworming may have affected the number of positive cases identified by either diagnostic method , influencing costs for case detection . In regard to the generalisability of findings , the relationship between costs and prevalence as shown in Figure 3 suggests that costs of detection would decrease with increasing prevalence of infection , and thus the costs of surveys conducted in high transmission are likely to be lower than the results presented here . In conclusion , our evaluation shows that the Kato-Katz and Mini-FLOTAC methods were comparable to one another in diagnostic sensitivity , yet Kato-Katz afforded greater cost-effectiveness . We encourage the wider use of simulation , cost-effectiveness and field studies to evaluate the cost-effectiveness of diagnostic and sampling strategies for STH surveillance in a variety of settings and for the wider surveillance of different NTDs . To this end , we provide the code for the Bayesian latent class modeling ( Supplementary Information S1 ) and a costing template for use in future studies ( Supplementary Information S2 ) .
|
Accurate methods of diagnosis and optimal strategies to sample the population are essential for the reliable mapping and surveillance of infectious diseases . The current standard for detection of soil-transmitted helminths ( STH ) entails use of the Kato-Katz diagnostic method . Alternative diagnostic methods , such as flotation techniques like the Mini-FLOTAC , have been developed with hopes of achieving greater sensitivity and ease of use . Here , we evaluate the diagnostic accuracy of the Kato-Katz method and the Mini-FLOTAC method for detecting STH infection . We use Bayesian latent class modeling to calculate the diagnostic accuracy in the absence of a gold-standard method for STH detection . Stool samples were collected from school-age children using school-based and community-based sampling . We present cost estimates for use of the Kato-Katz and Mini-FLOTAC diagnostic methods in combination with both sampling methods , providing cost data for the various survey scenarios . Sensitivity was comparable between the Kato-Katz and Mini-FLOTAC methods for detection of any STH species over a single day ( Kato Katz: 52 . 0% , Mini-FLOTAC: 49 . 1% ) and consecutive days ( Kato-Katz: 76 . 9% , Mini-FLOTAC: 74 . 1% ) . Costs were lowest in scenarios using the Kato-Katz method; cost per child tested and cost per case correctly classified for school-based sampling with the Kato-Katz diagnostic were $10 . 14 and $12 . 84 respectively . The lowest cost per case detected was $128 . 24 with community-based sampling and use of Kato-Katz . Further work is required on the cost-effectiveness of diagnostic and sampling methods for STH surveys and surveillance of other neglected tropical diseases ( NTDs ) in various settings . To this end , we provide the model code used in the diagnostic analysis and a costing template for STH surveys .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"cost-effectiveness",
"analysis",
"infectious",
"disease",
"epidemiology",
"economic",
"analysis",
"tropical",
"diseases",
"social",
"sciences",
"parasitic",
"diseases",
"health",
"care",
"research",
"design",
"global",
"health",
"neglected",
"tropical",
"diseases",
"population",
"biology",
"infectious",
"disease",
"control",
"ascariasis",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"health",
"economics",
"epidemiology",
"hookworm",
"diseases",
"economics",
"helminth",
"infections",
"survey",
"methods",
"survey",
"research",
"trichuriasis",
"biology",
"and",
"life",
"sciences",
"soil-transmitted",
"helminthiases"
] |
2014
|
Diagnostic Accuracy and Cost-Effectiveness of Alternative Methods for Detection of Soil-Transmitted Helminths in a Post-Treatment Setting in Western Kenya
|
The Kaposi's sarcoma-associated herpesvirus ( KSHV ) protein kinase , encoded by ORF36 , functions to phosphorylate cellular and viral targets important in the KSHV lifecycle and to activate the anti-viral prodrug ganciclovir . Unlike the vast majority of mapped KSHV genes , no viral transcript has been identified with ORF36 positioned as the 5′-proximal gene . Here we report that ORF36 is robustly translated as a downstream cistron from the ORF35–37 polycistronic transcript in a cap-dependent manner . We identified two short , upstream open reading frames ( uORFs ) within the 5′ UTR of the polycistronic mRNA . While both uORFs function as negative regulators of ORF35 , unexpectedly , the second allows for the translation of the downstream ORF36 gene by a termination-reinitiation mechanism . Positional conservation of uORFs within a number of related viruses suggests that this may be a common γ-herpesviral adaptation of a host translational regulatory mechanism .
Translation initiation of eukaryotic mRNAs is dependent on the 5′ mRNA cap and proceeds by ribosomal scanning until recognition of an AUG codon in a favorable context [1] , [2] . As a consequence of the translation machinery not engaging start codons at internal positions within the mRNA , eukaryotic transcripts generally encode only one functional protein . For the majority of mRNAs the most 5′-proximal AUG is selected , however strategies exist to bypass upstream start codons to enable downstream initiation . For example , leaky scanning can occur if the nucleotides flanking the 5′-proximal AUG are not in the Kozak consensus sequence ( ccRccAUGG ) , allowing the 40S ribosomal subunit to instead engage a downstream methionine codon [2] , [3] . Alternatively , when an upstream AUG is followed shortly thereafter by an in-frame termination codon , ribosomes can reinitiate translation , albeit with reduced efficiency , at a downstream AUG . These upstream open reading frames ( uORFs ) presumably permit translation of a downstream gene because factors necessary for initiation have not yet dissociated during the short elongation period . Notably , uORFs are common regulatory elements in eukaryotic transcripts , and generally function to reduce translation of the major ORF [3] , [4] . Additional , although rare , examples of internal ORF translation also exist , for example after ribosome shunting over a highly structured upstream sequence [5]–[8] , or upon direct 40S recruitment via internal ribosome entry sites ( IRESs ) [9]–[13] . Viruses do not encode translation machinery and thus operate under the constraints of host protein synthesis . However , the compact nature of viral genomes has resulted in the evolution of specialized strategies to maximize their coding capacity . Examples of such mechanisms include translation of a large polyprotein that is cleaved into multiple proteins , ribosomal frameshifting and non-canonical translation mechanisms such as those described above [14] . Accordingly , many viral mRNAs do not conform to the one protein per mRNA cellular paradigm and require specialized mechanisms to subvert the translational constraints of the host . Kaposi's sarcoma-associated herpesvirus ( KSHV ) is the etiologic agent of several human cancers including multicentric Castleman's disease , primary effusion lymphoma and Kaposi's sarcoma ( KS ) , one of the early AIDS-defining illnesses [15]–[17] . KSHV is a double-stranded DNA virus of the γ-herpesvirus subfamily , possessing a ∼165-kb genome and encoding an estimated 80 viral proteins [17] , [18] . The viral genes closely resemble those of their cellular counterparts in that they have canonical transcriptional promoters , consensus pre-mRNA splice sites and 3′-end formation signals . However , one notable departure from the cellular paradigm is the scarcity of poly ( A ) sites distributed throughout the genome , with a single signal often allocated to several consecutive ORFs . These gene clusters give rise to viral transcripts with polycistronic coding potential , although in general only the 5′-proximal gene is translated on each mRNA [19]–[21] . Most genes are positioned as a 5′ cistron by the use of multiple transcriptional start sites upstream of common poly ( A ) signals and/or alternative splicing [21] , [22] . To date , the only described KSHV mechanism to enable translation of a 3′-proximal ORF is an IRES identified in the coding region of vCyclin ( ORF72 ) , which allows for expression of vFLIP ( ORF71 ) [23]–[25] . A previously described tricistronic KSHV mRNA encompasses three partially overlapping open reading frames that are expressed with lytic kinetics ( ORF35 , 36 , and 37 ) . However , the mechanism of translation initiation of the 5′-distal ORF36 and ORF37 proteins has remained unresolved [26] , [27] . The function of the protein product of the 5′-proximal ORF35 is ill defined , although it shares limited sequence similarity with the α-herpesvirus UL14 gene product , which has described heat shock protein-like properties and functions to inhibit apoptosis during host cell infection [28] , [29] . The second gene , ORF36 , encodes a serine/threonine kinase that activates the cellular c-Jun N-terminal kinase ( JNK ) signaling pathway and phosphorylates the viral transcriptional transactivator K-bZIP , two processes involved in the progression from early to late viral gene expression [27] , [30] , [31] . Moreover , ORF36 sensitizes KSHV-infected cells to ganciclovir , an anti-viral drug shown to reduce KSHV replication in cultured cells and in patients [32]–[35] . The 3′-proximal ORF37 expresses SOX ( shutoff and exonuclease ) , a viral protein responsible for promoting widespread degradation of host mRNAs and also thought to assist in viral DNA replication and packaging [36]–[38] . Here , we demonstrate that the ORF35–37 transcript is functionally bicistronic , supporting translation of both ORF35 and ORF36 , whereas ORF37 is expressed from a previously uncharacterized monocistronic transcript . The polycistronic locus lacks IRES activity , and both proteins are expressed in a cap-dependent manner . Interestingly , translation of ORF36 occurs via a reinitiation mechanism after engagement of one of two overlapping short uORFs located in the 5′-untranslated region ( UTR ) , which also regulate the relative expression levels of these proteins . Thus , KSHV uses a host strategy normally reserved to repress translation of the major ORF to instead permit expression of a 3′-proximal cistron on a viral polycistronic mRNA . Analysis of homologous genetic loci from additional γ-herpesviruses similarly revealed the presence of dual short upstream ORFs ( uORFs ) , suggesting this may be a conserved mechanism of translation initiation among these viruses .
Two potential functionally polycistronic mRNAs are transcribed from the KSHV ORF34–37 genetic locus during lytic replication: a minor transcript encompassing ORFs 34 , 35 , 36 , and 37 ( ORF34–37 ) and a major transcript encompassing ORFs 35 , 36 and 37 ( ORF35–37 ) ( Figure 1A ) [26] , [27] . Although both ORF36 and ORF37 proteins play important roles in the viral lifecycle , no transcripts were reported in which these ORFs were present as the 5′-proximal cistron [26] , [27] . To confirm this observation , we searched for transcripts produced from this locus in a B cell line ( TREx BCBL1-RTA ) that harbors KSHV in a latent state but can be stimulated to engage in lytic replication . RNA isolated from cells infected latently or lytically for 8–36 h was Northern blotted with riboprobes specific for ORF36 or ORF37 . In infected cells , the ORF36 probe recognized transcripts co-migrating with or larger than the polycistronic ORF35–37 mRNA but did not reveal any smaller , potentially monocistronic species ( Figure 1B ) . Results from ORF36 5′ rapid amplification of cDNA ends ( RACE ) experiments were in agreement with its transcript initiating upstream of ORF35 at nucleotide position 55567 as previously reported by Haque et al . ( Figure 1A , data not shown ) [18] . In contrast , the ORF37 probe reacted with transcripts ≥3 . 4 kb and an additional ∼1 . 7 kb transcript that co-migrated with the control ORF37 monocistronic mRNA ( Figure 1C ) . Analysis of transcription start sites by 5′ RACE ( data not shown ) , as well as similar observations in a related γ-herpesvirus further supported the presence of an ORF37 monocistronic transcript [39] . Thus , ORF37 is most likely translated by the canonical cap-dependent scanning mechanism and is present as a silent cistron on the ORF35–37 polycistronic mRNA . We next sought to evaluate directly whether the ORF35–37 transcript could support translation of ORF36 as a downstream gene . 293T cells were first transfected with a plasmid expressing the coding sequence of ORF35–37 downstream of the native viral 72-nt 5′ UTR , and lysates were Western blotted using polyclonal antisera specific for ORF36 or , as a control , ORF37 . The ORF36 protein was readily translated from this polycistronic construct , whereas the ORF37 protein was detected only in cells transfected with the monocistronic ORF37 plasmid ( Figure 1D , 1E ) . In these and all subsequent experiments , Northern blotting of the mRNAs produced from each transfection confirmed that the transcripts were of the expected size and of equivalent abundance across experiments ( Figure 1D , 1E ) . ORF35 is conserved between the α , β , and γ-herpesvirus subfamilies but its function remains unknown and antibodies are not available to detect it in KSHV-infected cells [40] . ORF35 is predicted to encode a 151-amino acid protein , and its start site resides in a favorable Kozak context . Nonetheless , we considered the possibility that ORF35 is not translated , instead serving as a portion of the 5′ UTR for ORF36 . In order to directly compare the levels of ORF35 and ORF36 protein produced from the bicistronic construct , we engineered in-frame HA tags at the 5′ or 3′ end of each respective gene , maintaining the native viral 5′ UTR ( 5′ UTR HA-ORF35-ORF36-HA ) . Monocistronic versions of each HA-tagged gene were also generated as controls ( 5′ UTR HA-ORF35 , ORF36-HA ) . Importantly , Western blotting with HA antibodies revealed that the ORF35 protein is produced from both the monocistronic and bicistronic constructs ( Figure 2A ) . Although our data indicated that the ORF35–37 transcript is functionally bicistronic , it was still formally possible that ORF36 translation occurred from a low-abundance monocistronic transcript generated by a cryptic internal promoter or splice site ( s ) in the DNA plasmid . To address this possibility , we transfected cells directly with in vitro transcribed monocistronic or bicistronic mRNAs , and performed anti-HA Western blots to detect each protein ( Figure 2B ) . Again , both ORF35 and ORF36 protein were produced from the bicistronic 5′ UTR HA-ORF35-ORF36-HA mRNA , as well as from the appropriate control monocistronic mRNA , confirming that this locus is functionally polycistronic . The only other known example in KSHV of translation of a downstream ORF from a polycistronic mRNA occurs via an IRES [23]–[25] . We therefore used an established dual luciferase assay to determine whether an IRES similarly resides upstream of ORF36 . The dual luciferase construct consists of a 5′-proximal Renilla luciferase gene that can be constitutively translated via a cap dependent mechanism , followed by a 3′-distal firefly luciferase gene , which is not normally translated . The two genes are separated by a defective encephalomyocarditis virus ( ΔEMCV ) to prevent translational read-through [11] , [41] . Sequences of interest are then inserted between the ΔEMCV and the firefly luciferase gene , and IRES activity leads to the translation of firefly luciferase . Sequences encompassing ORF35 , ORF35–36 or ORF34–36 as well as two known IRES elements ( EMCV and KSHV ORF72 ) were cloned into the dual luciferase construct . The capped and polyadenylated in vitro transcribed mRNA was electroporated into lytically infected TREx BCBL1-RTA cells ( Figure 3A ) . The integrity of the mRNAs was verified by Northern blotting ( data not shown ) . After 4 h , the ratio of firefly/Renilla luciferase activity was measured to determine whether IRES activity was detectable in the context of lytic infection . Although both the EMCV and ORF72 control IRES elements supported translation of firefly luciferase , none of the sequences upstream of ORF36 possessed detectable IRES activity ( Figure 3B ) . We next sought to determine whether ORF36 translation was instead initiated via a cap-dependent mechanism by inserting a strong 40 nucleotide hairpin ( Hp7; ΔG = −61 kcal/mol ) after nucleotide 32 within the 72 nucleotide native 5′ UTR of the 5′ UTR HA-ORF35-ORF36-HA construct ( Figure 3C ) [42] . Stable hairpin structures ( ΔG<−30 kcal/mol ) present near the 5′ cap dramatically reduce translation initiation by stalling the pre-initiation complex [42] . Translation of both ORF35 and ORF36 was markedly reduced in the presence of Hp7 following either DNA or RNA transfection ( Figure 3D , S1A ) . Thus , recognition of the 5′ cap and subsequent 40S scanning are critical for translation of both ORF35 and ORF36 . It is notable that ORF36 protein production is robust given that its translation requires the pre-initiation complex to bypass the relatively strong Kozak context surrounding the ORF35 start codon ( AgaAUGG ) and to scan through 424 nucleotides of upstream sequence . To determine whether the context of the ORF35 start codon influences the expression of ORF36 , we mutated the preferred nucleotide ( A ) at position −3 to the least preferred nucleotide ( U ) ( 35 KCS wkn; Figure 3E ) . As expected , ORF35 expression was reduced; however , surprisingly , this mutation this did not significantly alter ORF36 expression , arguing against a pure leaky scanning mechanism to explain ribosomal access to the ORF36 start site ( Figure 3F ) . Direct transfection with in vitro transcribed mRNAs confirmed that this result was not due to induction of an alternative promoter ( Figure S1B ) . Thus , the relative strength of the ORF35 start site does not dramatically influence ORF36 translation , suggesting that there is an alternative mechanism in place that disfavors initiation at the 5′ gene . We searched for features of the ORF35–37 sequence that might contribute to translational start site selection . Within the 5′ UTR we noticed two short upstream ORFs ( uORFs ) . The first nine codon uORF , dubbed uORF1 , spans KSHV nucleotides 55603 to 55629 and has an AUG residing in a relatively weak Kozak context ( CguAUGA ) [18] . The second 11 codon uORF ( uORF2 ) spans KSHV nucleotides 55626–55658 and overlaps with both the 3′ end of uORF1 and the ORF35 start codon ( Figure 4A ) . To determine the contribution of uORF1 towards ORF35 and ORF36 translation , we mutated the uORF1 start site ( Δ1 ) ( Figure 4A ) . ORF35 expression was elevated in the Δ1 mutant ( Figure 4B ) . We confirmed that the HA tag at the 5′ end of ORF35 did not alter this translational regulation by showing similar results upon repositioning of the HA tag internally within ORF35 ( Figure S2 ) . Thus , ORF35 expression undergoes modest negative regulation by ribosomal engagement at the uORF1 start codon , although this does not appear to influence ORF36 expression . The uORF2 start codon is in a more favorable Kozak context than that of uORF1 , and disruption of the uORF2 AUG ( AUG→UUG; Δ2 ) or weakening the Kozak context of its start codon ( KCS2 wkn ) increased ORF35 translation and severely decreased translation of ORF36 in both DNA and RNA transfection experiments ( Figure 4C–D , S3 ) . Notably , the Δ2 mutant was designed to ensure the uORF1 stop codon remains intact , permitting the independent analysis of uORF1 and uORF2 . Unlike uORF1 , uORF2 therefore plays a key role in regulating expression of both genes in this polycistronic mRNA , likely due to the strong context flanking the uORF2 AUG as compared to the uORF1 start codon . Although a few rare uORFs have been found to function in a sequence-dependent manner [43]–[47] , for most characterized uORFs it is the act of translation rather than the peptide sequence that mediates their function . The fact that 45% of the uORF2 amino acid sequence is altered in the construct bearing the HA tag at the 5′ end of ORF35 is in agreement with the amino acid sequence of uORF2 not being the primary determinant of its activity . Indeed , rebuilding the uORF2 mutants into a construct in which the HA tag was moved to an internal position in ORF35 yielded indistinguishable results ( Figure S4 ) . The above findings suggested that engagement of the translation machinery at either uORF1 or uORF2 rather than the sequence of the uORF-encoded peptide mediates their regulatory function . We therefore sought to confirm that these uORFs were indeed recognized by the translation machinery . Due to their small size , uORF-generated peptides tend to be highly unstable and are very difficult to detect . To circumvent this problem , we made a single nucleotide change in each uORF to place them in frame with ORF35 lacking its AUG ( Δ35 ) , thereby generating uORF-ORF35 fusions ( Figure 4E ) . Thus , restoration of ORF35 expression is a direct readout translation initiation from the uORF start codon . In both cases , the uORF fusions restored ORF35 expression to levels corresponding to the relative strength of the Kozak consensus sequence of each uORF ( Figure 4F ) . As expected , only the uORF2 fusion abrogated expression of ORF36 ( Figure 4F ) . Finally , to determine whether additional cis-acting elements within ORF36 are required for its translation after uORF2 engagement , we replaced the ORF36 gene with a GFP reporter ( Figure 4G ) . GFP protein was expressed robustly as a downstream gene from this construct , arguing against a requirement for an element within ORF36 for its translation ( Figure 4H ) . Similar to our results with ORF36 , disruption of uORF2 compromised expression of GFP ( Figure 4H ) , supporting a uORF2-dependent mechanism as the primary pathway enabling translation of a downstream gene from this locus . Translation of a major ORF following engagement at a uORF generally occurs via a termination-reinitiation event . The length of a uORF is important for reinitiation , as it is thought that some of the translation initiation accessory factors have not yet dissociated prior to termination at the uORF stop codon [4] . In this regard , translation of the downstream ORF decreases dramatically if the time required to complete translation of the uORF is increased , for example by increasing the ORF length or inserting secondary structure to stall the ribosome [48] , [49] . Therefore , we reasoned that if ORF36 translation initiates using the same 40S ribosomal subunit involved in translation of uORF2 , then artificially elongating uORF2 should inhibit ORF36 expression . This experiment was performed on the construct backbone with the ORF35 HA tag located internally to mimic the wild type length of uORF2 . Indeed , extension of uORF2 from 11 to 64 codons ( uORF2-long ) resulted in a dramatic drop in ORF36 expression ( Figure 5A–B ) . The rate-limiting step of reinitiation is postulated to be the re-acquisition of the pre-initiation complex ( eIF2-GTP-Met-tRNAi ) during ribosomal scanning , and thus a sequence of sufficient length must be present downstream of the uORF for this to occur [3] , [4] . We therefore evaluated how the distance between the uORF2 stop codon and the subsequent start codon influences reinitiation within the viral mRNA . Start codons in a favorable Kozak context were inserted at two positions between the uORF2 stop codon and the ORF36 start site . We hypothesized that start codons located close to uORF2 would not be as efficiently recognized , and therefore they would not inhibit ORF36 expression . However , more distally located start codons should better engage the initiation machinery , thereby preventing translation from occurring at the authentic ORF36 start site . In agreement with this prediction , a start codon positioned 16 nucleotides downstream of uORF2 did not strongly inhibit ORF36 expression , whereas a methionine positioned 246 nucleotides after termination of uORF2 severely compromised ORF36 expression ( Figure 5C–D ) . These data support the conclusion that engagement of the ORF36 start codon is dependent on the reacquisition of the pre-initiation complex after termination of uORF2 translation . Translation reinitiation at the internal ORF36 start codon could occur either after linear scanning of the 40S complex through the 332-nucleotide intercistronic region between uORF2 and ORF36 or through shunting of the complex past this sequence and its subsequent positioning proximal to ORF36 . To distinguish between these possibilities , two strong hairpins ( Hp7 ) that impede scanning were inserted within the 5′-proximal or 3′-proximal coding region of ORF35 ( Figure 5E ) . If the 40S ribosomal subunit were shunted past these internal sequences , one or both of the hairpins ( depending on the location of the shunting sites ) should not compromise ORF36 translation [5] , [50] . However , we observed a significant reduction in ORF36 expression in the presence of either hairpin , arguing that the 40S complex scans in a linear fashion through ORF35 ( Figure 5F ) . One potential caveat is that the insertion of the hairpins might dramatically alter the RNA folding landscape , disrupting a secondary structure required for shunting . To exclude this possibility , the single natural methionine codon present within the coding region of ORF35 , was mutated to an arginine ( MidMut; Figure 5G ) . If this internal sequence were bypassed via shunting after uORF2 termination , the natural start codon should not be able to compete with the ORF36 AUG for the pre-initiation complex . However , we found that ORF36 expression was increased from the MidMut construct , arguing against a shunting mechanism and further suggesting that this methionine normally engages a fraction of the scanning ribosomes before they can reach the ORF36 start codon ( Figure 5H ) . Translation of the peptide generated cannot be directly monitored due to the fact that it is only eight amino acids . Collectively , these data support a model in which the preferential recognition of uORF2 diverts ribosomes past the ORF35 start codon , whereupon they scan in a linear fashion and reacquire the pre-initiation complex before reinitiating translation at a downstream start codon . To confirm that uORF2 regulates ORF36 expression during lytic KSHV infection , we engineered a uORF2 point mutant ( BAC16-Δ2; ATG→TTG ) and a revertant mutant rescue ( BAC16-Δ2-MR; TTG→ATG ) within the recently described KSHV BAC16 ( Figure S5 ) [51] . BAC16-WT , BAC16-Δ2 and BAC16-Δ2-MR were transfected into iSLK-PURO cells bearing a doxycycline-inducible RTA expression system to enable lytic reactivation [52] . Immunoblot analysis using polyclonal anti-sera specific for ORF36 revealed that while ORF36 was readily detectable at 48 h post-lytic reactivation in cells infected with WT or the mutant rescue virus , deletion of the uORF2 start codon severely compromised ORF36 expression ( Figure 6 ) . In contrast , the uORF2 mutation had no effect on the levels of the KSHV latent protein LANA or the lytic protein ORF57 , confirming its specificity for ORF36 ( Figure 6 ) . Thus , uORF2 plays a critical role in enabling expression of the ORF36-encoded viral protein kinase during lytic KSHV infection . We examined whether the loci analogous to KSHV ORF35–37 in several additional γ-herpesviruses also possessed uORFs within their 5′ UTRs ( Table S1 ) . Indeed , we identified two 6–12 codon uORFs within the predicted 5′ UTR of the locus in Epstein Barr virus ( EBV ) , herpesvirus saimiri ( HaSV-2 ) and ateline herpesvirus 3 ( AtHV-3 ) and one 11 codon uORF in good context within the 5′ UTR of the rhesus rhadinovirus ( RRV ) locus ( Figure 7A , 7B ) . The fact that the uORF positioning but not the coding sequence is conserved supports the hypothesis that their regulatory contribution relies on their ability to engage translation complexes , rather than the actual peptide produced . Furthermore , eight of the nine ORF35 homologs examined contain ≤2 internal methionine codons , as would be predicted if a termination-reinitiation mechanism was used to translate the downstream gene ( Table S1 ) . Interestingly , in all cases where two uORFs are present , the first uORF is within a weaker Kozak context than the second uORF , which overlaps the start codon of each ORF35 homolog ( EBV BGLF3 . 5 , SaHV-2 ORF35 , AtHV-3 ORF35 and RRV ORF35 ) . Thus , the conservation of uORFs at this genetic locus suggests that using uORFs to enable expression of a 3′-proximal gene may be a conserved strategy for translational control among these viruses . However , whether these loci indeed encode a functional polycistronic mRNA and are regulated by a similar uORF-based mechanism remains to be experimentally verified .
In this study , we describe a novel functionally bicistronic viral mRNA that is translated via a unique adaption of ribosomal reinitiation . In other characterized examples of viral translation via a reinitiation mechanism , expression of the downstream gene is significantly tempered as a consequence of ribosomal engagement at an upstream start codon [43] , [53]–[56] . Aside from being bicistronic , translation from the KSHV ORF35–37 transcript is unusual in that the protein product of ORF36 is at least as robustly expressed as the 5′ ORF35 despite the fact that the ORF35 start codon is in a favorable sequence context . We reveal that a key mechanism underlying this phenotype involves the position of a short uORF overlapping the start codon of ORF35 , which enables translation of ORF36 ( Figure 8 ) . These findings provide the first example of cap-dependent non-canonical translation in KSHV and illustrate a novel strategy to translate polycistronic mRNA . Several lines of evidence support the notion that ORF36 is expressed in a cap-dependent manner as a 3′-proximal cistron . No transcript of an appropriate size with ORF36 as the 5′-proximal cistron was detected in KSHV-infected cells , in agreement with the results of 5′ RACE that indicated its transcription starts upstream of ORF35 [26] . In addition , ORF36 protein expression was detected after transfection of an in vitro transcribed bicistronic RNA transcript . Finally , interfering with scanning from the 5′ mRNA cap via insertion of a hairpin blocked ORF36 translation , consistent with our failure to detect IRES activity within the locus . This is in contrast with the sole functionally bicistronic KSHV mRNA described to date , where an IRES is present within the coding region of ORF72 allows for ORF71 expression in a cap-independent manner [23]–[25] . Our results indicate that the ORF36 start codon is accessed via a termination-reinitiation event after translation of uORF2 . The most 5′ uORF ( uORF1 ) resides in a weaker context than uORF2 , which overlaps the ORF35 start codon . Importantly , because the stop codon of uORF1 overlaps with the start site of uORF2 , engagement of these uORFs is mutually exclusive . Therefore , preferential initiation at uORF2 likely drives the enhanced translation of ORF36 by causing ribosomes to bypass the favorable ORF35 start codon . After translating uORF2 , ribosomes continue to scan through the following 332 nucleotides to reinitiate at ORF36 . In support of this model , lengthening uORF2 to decrease the efficiency of reinitiation abrogated ORF36 expression . Furthermore , weakening the context surrounding the uORF2 start codon enhanced ORF35 expression , suggesting that the ORF35 start site is primarily reached by ribosomes that have bypassed the AUG of uORF2 , likely by leaky scanning . This provides a rare example of a uORF enhancing translation of a downstream major ORF . To date , the only described short uORF that enables access to the start codon of a downstream gene in a polycistronic transcript was identified in hepatitis B virus ( HBV ) . The HBV uORF , dubbed C0 , weakly inhibits the 5′-proximal C ORF while stimulating translation of the 3′-proximal J and P proteins [6] , [57] . However , the termination-reinitiation event described for HBV may be facilitated by a shunting mechanism , as non-linear scanning was found to occur in the homologous region in the related duck hepatitis B virus [58] . This appears not to be the case for ORF36 because insertion of strong hairpins within the coding region upstream strongly compromises ORF36 expression , suggesting that the ribosomes are scanning continuously from the 5′ mRNA cap to the ORF36 start codon . uORFs are common features found in the 5′ UTRs of many mammalian mRNAs [59] . They are widely recognized as cis-regulatory elements and their presence generally correlates with reduced translation of the major ORF by causing initiation to instead occur by leaky scanning or a low-efficiency reinitiation event , which is agreement with the function of uORF1 as a negative regulator of ORF35 [4] , [59] , [60] . A few cases have been described in which the ability of the uORF to repress downstream translation is dependent on the amino acid sequence of the encoded peptide [43]–[47] . For example , a uORF present in the 5′ UTR of the human cytomegalovirus gp48 gene attenuates downstream translation in a sequence-dependent fashion , likely by delaying normal termination and preventing leaky scanning by the 40S ribosomal subunit to reach the downstream AUG [43] . However , in general , engagement of the translation apparatus rather than the translated product itself represses translation of the major ORF . Indeed , regulation of the ORF35–37 transcript appears independent of the uORF peptide sequence because the 5′ HA-tagged construct had two amino acids mutated within uORF2 yet still functioned to permit translation of ORF36 . Moreover , uORFs in homologous regions of the genome in related γ-herpesviruses lacked amino acid conservation . However , individual amino acid substitutions in all of the uORF1 and uORF2 codons would be required to formally rule out a role for the encoded peptides in the translational control of this mRNA . Factors that influence the ability of a terminating ribosome to resume scanning remain poorly understood . It has been shown using chimeric preproinsulin mRNAs that efficient reinitiation progressively improves upon lengthening the intercistronic sequence up to 79 nucleotides [61] . Sufficient intercistronic sequence length is thought to be necessary to allow time for the scanning 40S ribosomal subunit to reacquire eIF2-GTP-Met-tRNAi prior to encountering the downstream start codon , although at what point the sequence length becomes inhibitory is not known [4] , [49] . In the context of the viral ORF35–37 transcript , the ribosome is able to reinitiate translation with a high frequency despite scanning 332 nucleotides after terminating translation of uORF2 , indicating that intergenic regions significantly longer than 79 nucleotides still enable reinitiation . Interestingly , a prior report identified a translational enhancer element within the tricistronic S1 mRNA of avian reovirus that functions to increase expression of a downstream cistron . This occurs as a consequence of sequence complementarity to 18S rRNA , which is reminiscent of the prokaryotic Shine-Dalgarno sequence [62] , [63] . A similar strategy of having 18S rRNA complementarity within a bicistronic mRNA was also found to enhance the ability of the minor calicivirus capsid protein VP2 to be translated by reinitiation [56] , [64] . Whether enhancer elements exist in the KSHV uORF-ORF36 intercistronic region to facilitate translation at the downstream cistron remains to be determined . However , no critical reinitiation element exists downstream of the ORF36 start codon , as replacement of these sequences with GFP does not block its translation . This is distinct from the termination-reinitiation mechanism described for certain retrotransposons , which require complex downstream secondary structures [65] . The question arises as to what benefit is conferred by this finely tuned strategy of translational control for both ORF35 and ORF36 . One possibility is that ORF35 and ORF36 are required at different points during lytic infection and that during the course of viral replication , conditions arise that favor translation of one protein versus the other . This type of regulation occurs in the well-characterized Saccharomyces cerevisiae GCN4 locus , where four short uORFs modulate reinitiation at the major ORF depending on the level of eIF2α phosphorylation [66]–[68] . Indeed , certain types of cell stress have also been shown to influence non-canonical translation of the cytomegalovirus UL138 gene [69] . Alternatively , the uORFs may confer a tight level of regulation to ensure that ORF36 is not synthesized at deleterious levels during infection . For example , an EBV mutant that over-produces BGLF4 ( the ORF36 homolog ) exhibited defects in viral replication [39] . Determining if and how this non-canonical mechanism of translational control influences the KSHV lifecycle will be an important future endeavor .
pcDNA3 . 1 ( + ) -ORF35–37 was generated by PCR-amplifying the ORF35–37 genetic locus from the KSHV-BAC36 ( kindly provided by G . Pari [70] ) and cloning it into the EcoRI/NotI sites of pcDNA3 . 1 ( + ) ( Invitrogen ) . pcDNA3 . 1 ( + ) -5′ UTR-HA-ORF35 was assembled in a two-step process starting with the addition of the N-terminal HA tag after the native start ATG ( nucleotide sequence: GCTTACCCATACGATGTAC CTGACTATGCG ) to the coding sequence amplified from the KSHV genome as above , followed by an overlap extension PCR to insert the 72 nucleotide ( nt ) native 5′ UTR . The final product was then inserted into the pcDNA3 . 1 ( + ) EcoRI/NotI restriction sites . pcDNA3 . 1 ( + ) -ORF36 was constructed by PCR-amplification of the ORF36 coding sequence or to add the in frame C-terminal HA tag ( GCTTACCCATACGATGTACCTG ACTATGCGTGA ) followed by insertion into EcoR1/Not1 restriction sites . pCDEF3-ORF37 is described elsewhere [37] . HA-ORF35-ORF36-HA was amplified from the KSHV-BAC36 using primers with additional HA tag sequences and inserted into the EcoR1/Not1 sites of pcDNA3 . 1 ( + ) . This was followed by scarless insertion of the native 5′ UTR via two-step sequential overlap extension PCR [70] . To construct 5′ UTR-ORF35iHA-ORF36-HA , a backbone construct consisting of 5′UTR ORF35-ORF36-HA was first generated by PCR-amplification from the KSHV-BAC36 with HA tag sequences solely for ORF36 and inserted into the EcoR1/Not1 sites of pcDNA3 . 1 ( + ) . This construct was then linearized by inverse PCR at nucleotide position 55795 followed by ligation-independent cloning using InFusion ( Clonetech ) with primers consisting of an HA tag flanked by 15 base pair regions of vector overlap . A stable hairpin structure ( Hp7 sequence: GGGGCGCGTGGTGGCGGCTGCAGCCGCCACCACGCGCCCC , [42] ) was inserted into the 5′ UTR at nucleotide position 55599 , or within the ORF35 coding region at nucleotide position 55662 and at position 55862 [18] . For the 5′ UTR HA-ORF35Δ96-HA-GFP construct , HA-GFP was inserted between the NotI/XbaI restriction sites in pcDNA3 . 1 ( + ) , and the 5′ UTR-HA-ORF35 Δ96 fragment was then inserted between the EcoRI/NotI restriction sites upstream of HA-GFP . Two bicistronic , dual luciferase constructs , a negative control ( ΔEMCV; mutated IRES sequence ) and a positive control ( ΔEMCV element+functional EMCV ) were kindly provided by P . Sarnow ( Stanford University ) [11] , [41] . ORF72 , ORF34–36 , ORF35–36 and ORF35 PCR amplicons were inserted into the EcoRI restriction site downstream of the ΔEMCV element and upstream of firefly luciferase . The primers used to generate these constructs are listed in Table S2 . Where specified , parental plasmids were subjected to site-directed mutagenesis using the QuikChange kit ( Stratagene ) as per the manufacturer's protocol . The context of the ORF35 start codon was weakened by mutating the wild type AgaAUGG to UgaAUGG ( 35 KCS wkn ) . uORF1 and uORF2 mutants ( designated Δ1 and Δ2 ) were generated by substituting the AUG start codon with AGA or UGA , respectively . The uORF2 Kozak context was weakened by mutating the wild-type AccAUGA to UuuAUGA ( KCS2 wkn ) . The ORF35 start codon was disrupted by mutating the wild type AUG to AGA ( Δ35 ) . The uORF1 fusion to Δ35 was generated by mutating the uORF1 stop codon UGA to UGG ( uORF1-Δ35 ) . The uORF2 fusion to Δ35 was generated by deleting one nucleotide ( A ) located immediately prior to the ORF35 start codon ( uORF2-Δ35 ) . Two codons within in the ORF35 coding region were converted to AUGs in a strong context: ( 1 ) AccAACU to AccAUGG and ( 2 ) AauUUUG to AauAUGG . The native AUG residing at location 55778-80 within the ORF35 coding region was mutated to an AGA ( MidMut ) [18] . uORF2 was lengthened from 11 to 64 codons by mutating the first UAA stop codon to AGA , the second UAA stop codon to CAA , the third UGA stop codon to CGA , and the fourth and fifth UAG stop codon to CAG , resulting in the use of the next downstream stop codon ( uORF2-long ) . The KSHV BAC16 was modified as described previously [51] use a two-step scarless Red recombination system [71] . Briefly , BAC16 was introduced in GS1783 E . coli strain by electroporation ( 0 . 1 cm cuvette , 1 . 8 kV , 200 Ω 25 µF ) . A linear DNA fragment encompassing a kanamycin resistance expression cassette , an I-SceI restriction site and flanking sequence derived from KSHV genomic DNA was generated by PCR and subsequently electroporated into GS1783 E . coli harboring BAC16 and transiently expressing gam , bet and exo . Integration of the KanR/I-SceI cassette was verified by PCR and restriction enzyme digestion of the purified BAC16 DNA . The second recombination event between the duplicated sequences resulted in the loss KanR/I-SceI cassette and the seamless recirculation of the BAC16 DNA , yielding kanamycin-sensitive colonies that were screened by replica plating . BAC16 DNA was purified from chloramphenicol-resistant colonies using the NucleoBond 100 ( Machery-Nagel ) as per the manufactures instructions . Human embryonic kidney 293T cells were maintained in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) ( Gibco ) . The iSLK-PURO KSHV-negative endothelial cell lines [51] , [52] were maintained in DMEM supplemented with 10% FBS , penicillin ( 100 U/ml , Gibco ) and streptomycin ( 100 µg/ml , Gibco ) . To induce lytic reactivation of KSHV , iSLK-PURO cells were treated with doxycycline ( 1 µg/ml , BD Biosciences ) and sodium butyrate ( 1 mM , Sigma ) . TREx BCBL1-RTA [72] cells were maintained in RPMI supplemented with 10% FBS , L-glutamine ( 200 µM , Invitrogen ) , penicillin ( 100 U/ml ) , streptomycin ( 100 µg/ml ) and hygromycin B ( 50 µg/ml , Omega Scientific ) . To induce lytic reactivation of KSHV , TREx BCBL1-RTA cells were split to 1×106 cells/ml and induced 24 h later with 2-O-tetradecanoylphorbol-13-acetate ( TPA; 20 ng/ml , Sigma ) , doxycycline ( 1 µg/ml ) and ionomycin ( 500 ng/ml , Fisher Scientific ) [73] . For DNA transfections , constructs ( 1 µg/ml ) were transfected into subconfluent 293T cells grown in 12-well plates , either alone or in combination with 0 . 1 µg/ml GFP as a co-transfection control using Effectene reagent ( Qiagen ) or Lipofectamine 2000 ( Invitrogen ) following the manufacturers protocols . For RNA transfections , 3 µg/ml of mRNA in vitro transcribed using the mMessage mMachine kit ( Ambion ) and polyadenylated with yeast poly ( A ) polymerase ( Epicentre Technologies ) was transfected into ∼90% confluent 293T cells grown in 12-well plates using Lipofectamine 2000 . TREx BCBL1-RTA cells were transfected with 20 µg of DNA per 107 cells via electroporation ( 250 V , 960 µF ) with a Gene Pulser II ( Bio-Rad , Hercules , CA ) . For BAC transfections and reconstitution , ∼70% confluent iSLK-PURO cells were grown in a 24-well plate followed by transfection with 500 ng of BAC DNA via FuGENE 6 ( Promega ) , after 6 h , a further 500 ng BAC DNA was transfected with Effectene , following the manufacturers protocols and subsequently selected with 800 µg/ml hygromycin B to establish a pure population . iSLK-PURO-BAC16 cells were then induced with doxycycline ( 1 µg/mL ) and sodium butyrate ( 1 mM ) to enter the lytic cycle of KSHV replication . Luciferase activities were determined using the dual-luciferase assay system ( Promega ) and a bench-top luminometer according to manufacturer's protocol . IRES activity was calculated by obtaining the firefly/Renilla activity ratios for each of constructs containing the putative IRES sequences or the positive controls and dividing them by the ratio obtained from the ΔEMCV negative control . The value of fold activation represents at least three independent experiments with triplicate samples in each electroporation . Error bars represent the standard deviation between replicates . Protein lysates were prepared in RIPA buffer [50 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl , 1% ( v/v ) Nonidet P-40 , 0 . 5% ( w/v ) sodium deoxycholate , 0 . 1% ( w/v ) sodium dodecyl sulfate ( SDS ) ] containing protease inhibitors ( Roche ) , and quantified by Bradford assay . Equivalent quantities of each sample were resolved by SDS-PAGE , transferred to a polyvinylidene difluoride membrane and incubated with the following primary antibodies: mouse monoclonal GFP ( 1∶2000 , BD Biosciences ) , mouse monoclonal HA ( 1∶2000 , Invitrogen ) , rabbit polyclonal ORF36 ( 1∶5000 , kindly provided by Y . Izumiya [27] ) , goat polyclonal horseradish peroxidase ( HRP ) -conjugated actin ( 1∶500 , Santa Cruz Biotechnology ) , rabbit polyclonal SOX J5803 ( 1∶5000 , [38] ) , rabbit polyclonal ORF57 ( 1∶5000 , kindly provided by Z . Zheng [74] , rabbit polyclonal LANA #6 ( 1∶1000 ) or mouse monoclonal S6RP ( 1∶1000 , Cell Signaling ) followed by incubation with HRP-conjugated goat anti-mouse or goat anti-rabbit secondary antibodies ( 1∶5000 dilution ) ( Southern Biotechnology Associates ) . Total cellular RNA was isolated for Northern blotting using RNA-Bee ( Tel-Test ) . The RNA was then resolved on 1 . 2–1 . 5% agarose-formaldehyde gels , transferred to Nytran nylon membranes ( Whatman ) and probed with 32P-labeled DNA probes made using either the RediPrime II random prime labeling kit ( GE Healthcare ) or the Decaprime II kit ( Ambion ) . Strand-specific riboprobes specific for ORF36 and ORF37 were synthesized using the Maxiscript T7 kit ( Ambion ) with 32P-labelled UTP . The probes used for Northern blot analysis spanned the following regions according to the nucleotide positions described by Russo et al . [18]: ORF35 probe: 55639–56091 , ORF36 full-length probe: 55976–57310: ORF36-specific probe: 56093–56805: and ORF37 probe: 57273–58733 . Results in each figure are representative of at least three independent replicates of each experiment . The uORF1 and uORF2 alignments were generated from data obtained from the NIAID Virus Pathogen Database and Analysis Resource ( ViPR ) online through the web site at http://www . viprbrc . org .
|
Kaposi's sarcoma-associated herpesvirus ( KSHV ) is the etiologic agent of multicentric Castleman's disease , primary effusion lymphoma and Kaposi's sarcoma . KSHV expresses a number of transcripts with the potential to generate multiple proteins , yet relies on the cellular translation machinery that is primed to synthesize only one protein per mRNA . Here we report that the viral transcript encompassing ORF35–37 is able to direct synthesis of two proteins and that the translational switch is regulated by two short upstream open reading frames ( uORFs ) in the native 5′ untranslated region . uORFs are elements commonly found upstream of mammalian genes that function to interfere with unrestrained ribosomal scanning and thus repress translation of the major ORF . The sequence of the viral uORF appears unimportant , and instead functions to position the translation machinery in a location that favors translation of the downstream major ORF , via a reinitiation mechanism . Thus , KSHV uses a host strategy generally reserved to repress translation to instead allow for the expression of an internal gene .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology",
"viruses",
"and",
"cancer",
"viral",
"enzymes",
"biology",
"microbiology",
"viral",
"replication"
] |
2013
|
Dual Short Upstream Open Reading Frames Control Translation of a Herpesviral Polycistronic mRNA
|
Given the disposability of somatic tissue , selection can favor a higher mutation rate in the early segregating soma than in germline , as seen in some animals . Although in plants intra-organismic mutation rate heterogeneity is poorly resolved , the same selectionist logic can predict a lower rate in shoot than in root and in longer-lived terminal tissues ( e . g . , leaves ) than in ontogenetically similar short-lived ones ( e . g . , petals ) , and that mutation rate heterogeneity should be deterministic with no significant differences between biological replicates . To address these expectations , we sequenced 754 genomes from various tissues of eight plant species . Consistent with a selectionist model , the rate of mutation accumulation per unit time in shoot apical meristem is lower than that in root apical tissues in perennials , in which a high proportion of mutations in shoots are themselves transmissible , but not in annuals , in which somatic mutations tend not to be transmissible . Similarly , the number of mutations accumulated in leaves is commonly lower than that within a petal of the same plant , and there is no more heterogeneity in accumulation rates between replicate branches than expected by chance . High mutation accumulation in runners of strawberry is , we argue , the exception that proves the rule , as mutation transmission patterns indicate that runner has a restricted germline . However , we also find that in vitro callus tissue has a higher mutation rate ( per unit time ) than the wild-grown comparator , suggesting nonadaptive mutational “fragility” . As mutational fragility does not obviously explain why the shoot—root difference varies with plant longevity , we conclude that some mutation rate variation between tissues is consistent with selectionist theory but that a mechanistic null of mutational fragility should be considered .
In some animals , the germline is segregated early in development , thereby preventing many ( i . e . , somatic ) mutations from being transmitted to progeny [1] . Given this , classical theory of senescence posits that organisms have no vested interest in keeping the somatic mutation rate under control after the age of reproduction [2] . The same logic predicts that , given the reduced temporal longevity of any mutation , the shorter lived the organism , the higher the somatic mutation rate could be [3] . A higher somatic rate in mouse than human [4] , for example , is consistent with such expectations . As the future potential longevity of new germline mutations is longer than that of somatic ones , special protection for the germline from mutation can sometimes be expected . In humans , for example , there is an unusually low per-cell-division mutation rate in the male germline [5] . More generally , somatic rates are typically reported to be higher than germline rates [4 , 6–9] . The extent to which such a theoretical framework , based on the potential longevity of a mutation , enables understanding of the variation in mutation rates between tissues across phyla is poorly understood , not least because of a dearth of data in many major groups ( somatic mutation rates , either per cell division or per unit time , are for example hard to measure [10] ) . Whether plants are potentially informative in this debate is at first sight doubtful , not least because whether they have any clearly distinct soma and germline is contentious [11 , 12] . Leaving this uncertainty to one side , plants provide some relatively clear predictions and exceptional opportunities . We might , for example , expect a difference between root and stem , as stem alone has the prospect of being a germline progenitor . The transmissibility of mutations is not the only issue , however . Whereas germline mutations have a high prospective temporal longevity ( i . e . , they can be passed to the next generation and thence onwards ) , prospective longevity also varies between nongermline tissues . For example , petals and leaves , although ontogenetically related , have different longevities , the petal being highly transient and thus potentially under reduced selection to minimize mutation rates . Similarly , perennials and annuals will differ in the longevity of true somatic mutations and might differ in the proportion of premeiotic mutations that are transmissible . If so , annuals and perennials may differ also in the extent to which the shoot might have a reduced mutation rate , much as short- and long-lived mammals differ in their somatic rates [4] . Any model of selectively optimized mutation rate differences also predicts that variation between samples is not simply owing to stochastic variation . Plants provide excellent opportunities to test this given their branched structure and hence numerous biological replicates whose age can be ascertained . In this context , although we expect different branches to harbor different mutations [13–16] , we do not necessarily expect some branches to be significantly more or less mutagenic than others; i . e . , we expect heterogeneity to be between different tissues , not between biological replicates of the same tissues . The development of multiple branches in any growing season from the same plant permits an unusually well-controlled resource to test for homogeneity of mutation rates . Between-branch differences are also important because bud breeding has long been a classical way to establish a good variety in perennial crops , especially in the important fruits and ornamental plants [17–19] . Indeed , the plant has long been viewed as a metapopulation [20] , in which each branch evolves independently such that interbranch variation could prevent pest populations from adapting to all branches on individual host trees [21] . In this context as well , understanding the nature of between-branch mutation accumulation heterogeneity is of importance . Above , we presume one model for between-tissue differences in mutation rate , this being a model in which the variation is understood as the product of selection on the rate of mutation accumulation . An alternative possibility is that the mutation rate is “fragile”—i . e . , easily perturbed by , for example , intra-organismic local environment or growth conditions [22] , but not necessarily in a selectively advantageous manner . If so , we might expect that the mutation rate of artificial callus tissue , raised in the lab , might be mutationally different from a field-grown comparator . Here then , we attempt to define an extensive platform for consideration of the architecture of intra-organismic mutation accumulation in plants . To this end , across numerous species we ask whether roots and shoots have the same mutation rates , whether leaves and petals have different rates , whether we can detect between-branch heterogeneity , and whether tissue culturing modifies mutation rates . Consistent with an adaptive model , we find that roots commonly have higher mutation rates ( per unit time ) than shoots in perennials but not in an annual , petals have higher rates than leaves , and variation is between tissues rather than between biological replicates . However , we also find that callus has very high mutation rates . In addition , we assume that the apparent differences in mutation rate are just that and are not owing to differential degrees of purifying selection . We test for this possibility but find no support for it . In sum , the in vivo evidence is largely consistent with an adaptive mutation rate model , but in vitro data support the viability of a mechanistic null of context-dependent mutational fragility . Note that we make no attempt to directly measure the per-cell-division rate , as the issue at stake is the net mutation accumulation . That is , a system that reduces the number of cell divisions in , for example , germline but not the per-cell-division mutation rate can be of equivalent selective consequence as one that reduces the per-division rate but not the number of such divisions . In this context , we measure net mutation accumulation with the “rates” being comparable between comparators but not necessarily defined in absolute terms ( i . e . , not per cell division ) .
Prima facie , a selectionist model of mutation rate adaptation might predict a lower rate of mutation in stem compared with root , as mutations ( which are mostly deleterious ) are more likely to be transmissible to a subsequent generation if they occur in shoot . One could indeed argue that the root—shoot difference is the closest many plants get to anything resembling an unambiguous soma—germline distinction . Moreover , despite a distinct phenotype , the shoot and root share much in common about their organization of stem cell niches [12 , 23] , which also resembles that of animal stem cell niches [23] . New leaf and new root at the terminal branch in a perennial plant are the best organs for comparison because they have an equal separation age ( i . e . , time to common cell ancestor in the embryo ) . If all else is equal , the average mutation number per leaf sample ( or part leaf with a total of >80 mg for genome sequencing ) should be comparable with the number per root sample with a similar weight from the same plant . Note that this assumes that their cell sizes are similar , and hence , a similar number of cells are sampled . We note that 200–500 ng of DNA per sample is used in library preparation and is sequenced . This ensures that even in tissues where DNA is hard to extract ( e . g . , root ) or where the ratio of DNA mass to tissue mass is low , the total amount of sequenced DNA is approximately invariant . Four pairs of leaf and root samples were collected from three perennials ( Prunus persica , P . mume , Salix suchowensis ) and one annual ( Brachypodium distachyon ) species ( Table 1 ) . A total of 96 leaf and 74 root samples were sequenced . Mutations were called following stringent quality control and by reference to the ancestral state derived by the allelic state elsewhere in the same plant ( S1 Fig ) . To minimize miscalling , we employ two different calling methods and require mutations to either be called by both or , if called by just one , to be confirmed by other means ( see Materials and methods ) . We find that there are more mutations per sample in roots than in leaves in the perennials . The mean mutations per root and per leaf sample are 29 . 8 versus 3 . 74 in peach ( Brunner-Munzel [BM] test , P < 2 . 2 × 10−16 ) and 25 . 4 versus 12 . 9 in plum , respectively ( BM test , P = 3 . 8 × 10−5 ) . This contrast was also evident in perennial shrub willow ( 2 . 24 in root versus 1 . 05 in leaf , BM test , P = 0 . 01 ) , which has exactly grouped roots and leaves from the same cuttings ( S2 Fig and S1 Table ) . However , although an absolutely higher number was also observed in the annual B . distachyon ( 4 . 75 in root versus 3 . 17 in leaf ) , the ratio is more modest , and indeed the difference is not significant ( BM test , P = 0 . 48 ) . It is notable that the ratio is most extreme in one of the two long-lived species ( peach ) , near parity for the annual , and intermediate for the other perennials ( relatively short-lived shrub willow and plum ) . A visual way to observe the root—shoot difference is by observation of the topology of an ontogenetic tree ( like a phylogenetic tree but reflecting mutations through development ) constructed from all the leaf and root mutations of peach tree PXL . This displays the very evident differences of mutation patterns between the shoot apical meristems ( SAMs ) and root apical meristems ( RAMs ) ( Fig 1 ) , with root having very long “branch” lengths , consistent with very different mutation rates per unit time from the two tissues . In the prior Results section , we found a general trend for relatively low stem mutation rates ( per unit time ) in perennial species but not in an annual . Might there be a reason that annuals and perennials are different in this regard ? One possibility is that , as in animals , a short-lived species has less interest in restraining the mutation rate of all somatic tissues [3 , 4] . However , unlike soma in animals , stem mutations are potentially transmissible; thus , we might in addition predict that annuals should have a higher stem rate ( and a stem—root relative rate near parity ) if they also transmit relatively few premeiotic mutations . Consistent with this possibility , theoretical models predict that the contribution of somatic ( pregametic ) mutation could outweigh that of gametic mutation , especially in modular plants with small populations [14] , such as long-lived trees . Prior data on the rate of evolution of perennials and annuals are undecisive on this issue , as they do not assess the relationship between intra-organism mutation accumulation and the relative transmissibility of mutations . Whereas the per-generation mutation rate in long-lived perennials could be as high as 25 times as that in short-lived annuals [24] , on the per-year scale , the long-lived perennials apparently evolved slower than short-lived annuals , as suggested by the generation-time hypothesis [25 , 26] . These data do not address the ratio of mutation accumulation to mutation transmission . When considering transmissibility of mutations we can consider two metrics . In both , we estimate the number of mutations that are premeiotic in the parent but transmitted to progeny ( Nt ) . We can then consider this in proportion to the number of mutations observed in either the offspring ( No ) or the parent ( Np ) . We start by considering the first ratio ( Nt/No ) in a perennial . To this end , 14 fruits from the tree GL2 ( Fig 2A–2C ) were harvested and germinated . The leaf DNA was extracted from these seedlings and sequenced . Based on the sequenced genotypes , 10 seedlings were self-pollinated products , and one was an outcome of putative inner cross between branches B2 and B5 . Both mitotic ( 113 ) and “not premeiotic” mutations ( 47 ) can be unambiguously identified from the 11 self-pollinated products ( Table 2 ) , indicating that the majority of mutations observed in the offspring ( 71% and 66% in younger tree GZ; Table 2 and S2 Table ) are derived from premeiotic mutations in peach trees . Note that the “not premeiotic” mutations are defined as all mutations from the meiotic progeny that are specific to meiotic progeny . Some of them may be generated by mitosis just before meiosis , and some may be from early development of the progeny . The other 3 seedlings were outcrossed products between GL2 and different peach trees , in which only the transmitted mitotic mutations can be easily determined . Further tests revealed an average of 3 . 86 premeiotic mutations per seed in the 21 fruits of the plum tree MHG1 ( S3 Table ) , indicating that premeiotic mutations are a major source of genetic variation in perennial species . The Nt/No ratio is much lower in annuals . Sixteen whole-genome-sequenced progeny of Brachypodium sample WD2 indicated that only 24% of mutations in these seedlings are derived from premeiotic somatic mutations , with 0 . 69 premeiotic somatic mutations on average in any given progeny ( S4 Table ) . This proportion is much lower than that seen in trees ( for GL2 it is 71% [Table 2] , for GZ 66% [S2 Table] , higher than 24% in the WD2 [S4 Table] , χ2 = 32 . 835 , d . f . = 1 , P = 1 . 003 × 10−8 ) . Note that for all cases , both the annuals and perennials , the progeny DNA was sampled after approximately 1 month of growth . The alternative ratio , the proportion of premeiotic mutations that get transmitted/total premeiotic mutations , Nt/Np , is also lower in annuals . In total , we did 317 PCRs for 49 mutations in a total of 115 seedling samples in Arabidopsis , rice , and Brachypodium ( S3 and S4 Figs , Fig 3A ) . We found 1 . 72% ( 1/58 per seed ) , 3 . 0% ( 3/100 per seed ) , and 6 . 29% ( 10/159 per seed ) of premeiotic mutations to be transmissible , respectively ( Nt/Np ) . This proportion of transmissible mutations is significantly lower ( Pearson’s χ2 test with Yates continuity correction , χ2 = 187 . 53 , d . f . = 1 , P < 2 . 2 × 10−16 ) than that observed in trees ( approximately 51 . 6% overall , with 154/305 = 50 . 5% per progeny in GL2 , 61/124 = 49 . 2% per progeny in GZ , and 17/21 per progeny = 81 . 0% in Maoping ) . Note the plum MHG1 ( 84/319 = 26 . 3% mutations per progeny transmitted ) was not included in this comparison , as its progeny are from outcrossing whereas all others are selfing , which would make the comparison unfair because a mutation in selfing progeny has a greater chance of being transmitted . It is then all the more striking that the absolute rate in this outbred individual is higher than in the selfing annuals . The high number of somatic mutations in annuals is similarly reflected in the large number ( 3 . 32 per leaf ) in rice ( S4 and S5 Figs ) with low transmissibility . The low transmission in annuals is similarly reflected in the spatial location of the few mutations that are transmitted , these typically arising in the vicinity of the sites of gametogenesis . For example , the Brachypodium sample WD2 has eight branches derived from three major branches , from which 29 leaf samples and seven glume/lemma samples ( each from a spikelet and each branch usually growing three spikelets in general ) were sampled ( Fig 3A ) . In addition , 42 seeds were collected from different spikelets within the same branch as the glume samples . In the 36 leaf and glume samples , 77 novel mutations were detected . Among the 77 mutations , 22 were selected for further PCR and Sanger sequencing to see if they were transmissible . Just under one-quarter ( 5/22 = 23% ) of them were confirmed to be present in any of the five seeds . Further whole-genome sequencing of 16 progeny suggested this rate was actually lower ( 8% ) , with only six mutations ( five were those confirmed by PCR ) present in any of six seeds among 40 mutations within the parental branches bearing those seeds ( Fig 3A ) . Five out of the six mutations are present in the seven glumes that have been sampled and sequenced . This indicates that mutations physically closer to the site of gametogenesis ( i . e . , in glumes ) have a greater chance of being transmissible ( χ2 test with Yates correction , χ2 = 16 . 9 , d . f . = 1 , P = 3 . 86 × 10−5 ) . As an independent assessment of quality control , we note that the transmitted mutations identified in the sequenced genomes exactly match the PCR-verified results . We conclude that annuals transmit proportionally fewer of their premeiotic mutations than do perennials and that a lower proportion of mutations reported in the progeny are premeiotic in origin . This in turn could explain , from an adaptive model , why the root—shoot mutation rate ratio is high in perennials but not in annuals: if premeiotic mutations have little chance of transmission and there is no possible accumulation year on year ( i . e . , in annuals ) , then relatively little is to be gained by reducing the shoot mutation rate . All the data presented here support an adaptive framework well . In perennials , the shoot mutation rate is relatively constrained because they transmit shoot-accumulated mutations , whereas the shoot mutation rate is relatively unconstrained in annuals because they transmit relatively few mutations and the plant will shortly be dead . In this context , one observation appears , prima facie , to be an exception and counter to the selectionist model . In woodland strawberry , an initial plant sends out runners that can occasionally produce lateral buds that initiate new plants with shoots , leaves , and fruit ( Fig 3B ) . Thus , every gamete in a runner-propagated mature plant has an ontogenetic cell lineage history that runs back through the runner to the parent plant and thence to the initial seed . We might therefore expect that runner , as the progenitor of all plants and hence of all seeds , to have an especially low mutation rate . However , we find that there are 4 . 75 mutations accumulated per runner versus 1 . 93 per leaf ( BM test , P = 8 . 2 × 10−7; Table 1 ) . Counting each mutation once , we find 0 . 67 mutations per daughter plant , and 2 . 33 mutations per node for runners . Why is the runner rate higher ? The example of nontransmissibility of somatic mutations in annuals suggests a related explanation . What if mutations that occur in runners are for the most part not passed on to the lateral buds , as the cell lineage permitted to develop into lateral buds is spatially restricted ? Were this the case , most cells in the runner would be more like root in having no ontogenetic future in gametes and hence would be under relatively relaxed selection . By contrast , cells of the shoot of the plant would still have a potential future in gametes . Does then the runner contain an effective germline ? Analysis of mutation accumulation patterns ( Fig 3B ) provides strong support for the possibility of two separate cell lineages , one that is ontogenetically restricted to the runner and one that is not . Because we know which was the first plant , we know both spatially and temporally where all the new mutations initially occurred and when they are subsequently found . We find a pattern in which mutations found in the runners are normally restricted to the runner , but with one exception . Consider the first three mutations ( numbered 1 to 3 Fig 3B ) . These appeared in runner site B1-S1 and can be detected in the subsequent runner sites of 1a-S1 , 1c-S1 , and 1d-S2 but could not be detected in the subsequent shoots and leaves produced from the lateral buds ( lateral buds 1a to 1e , resultant shoot/leaves 1a-1 , 1a-2 , etc . ) . Similar ontogenetic restriction was also found for all subsequent runner mutations ( 4 and 11 ) , bar one . The one exception is mutation 5 , which occurs prior to stem/runner 1a but is henceforth seen everywhere: in all progeny runners and in the products of the five lateral buds ( 1a to 1e ) . It is notable that at positions 1c-S1 and 1d-S1 in the runner , we find all of the runner mutations ( 1–5 and 11 ) , but in the lateral bud progeny ( 1c , 1d , 1e ) mutation 5 alone is seen ( excepting mutations that arose in the shoots/leaves , e . g . , mutation 12 ) . Providing a statistical test for the exceptionalism of mutation 5 requires a few assumptions . But let us suppose that each lateral bud has , as observed , only one mutation seen in the runner from which it is derived ( this could be owing to a small initial cell population founding a lateral bud ) . At 1a , there are five runner mutations , and we can then suppose that one ( and only one ) of these was transmitted to descendent shoots and leaves . We attach no probability to this first selected mutation being number 5 , as this is only relevant post hoc . Instead , we ask what the probability is that for the subsequent plants/lateral buds the same one mutation ( whichever it is ) is selected at random , this being the null . At 1b , the runner has mutations 1 to 5 , so the probability of any prespecified mutation being the mutation in the lateral bud is 1/5 . At 1c to 1e , these five mutations are joined by mutation 11 . Thus for each of these , again , assuming one successful mutation , the probability that the prespecified mutation is in the lateral bud is 1/6 . Thus , the probability of the initially successful mutation ( in our case , mutation 5 ) alone being selected at each lateral bud is 1/5 × ( 1/6 ) 3 = 0 . 00093 . This provides strong reason to reject the null of random cell selection in the production of lateral buds and , conversely , supports the possibility that in runner , there is a segregated germline . Although highly significant , the above calculation comes with numerous caveats . We assume only one mutation can be transmitted to lateral buds . However , a correct null of random cell selection ( rather than random mutation selection ) would make it even less likely that all subsequent lateral buds would have the same prespecified mutation , as nontransmission must be an alternative part of parameter space under such a null , there being no reason to suppose that every cell has a mutation . Inclusion of a nontransmission possibility thus renders the likelihood of the same mutation being transmitted to the lateral bud every time even less likely . However , we do not know the number of progenitor cells in the lateral bud or the proportion of cells with at least one mutation and so cannot specify this null correctly . Furthermore , we assume all mutations to have occurred in different cells and so are themselves independent in any model of random cell selection . This need not be true . Despite the above caveats , it is most parsimonious to suppose that one mutation ( 5 ) uniquely occurred in a germline lineage within the runner and that only mutations in this cell lineage make up the shoot derived from lateral buds . Other lineages may make up further runners ( 1–4 , 11 ) but are restricted from lateral buds and hence are not in shoots , leaves , and gametes . Because of such a restriction , a relatively unconstrained mutation rate can be expected . Thus , the one prima facie exception may be the exception that proves the rule . The stem—root difference is consistent with a selectionist view of mutation rate variation within a plant . The same model could also predict that longer-lived terminal tissues might have lower mutation rates than shorter-lived tissues , just as a soma in short-lived species has a higher rate than a soma in longer-lived species . The leaf—petal difference is here a potentially informative test . Petals comes from the second whorl formed by the floral meristems [27] and have a similar cell division profile to leaves [28] . As the floral meristem shares a similar organization with SAM , it has been suggested that flowers and shoots are homologous structures , with floral organs being viewed as modified leaves [27] . But differences do exist; for example , unlike the stem cell fate in SAM , which is indeterminate ( i . e . , not determined by its cell lineage but by its position ) and grows indefinitely , the stem cells in floral meristem are determinate and will cease growth upon the formation of four whorls [27] . We consider two sampling strategies to examine the leaf—petal difference . First , using whole tissues , we observe that peach petals have a higher rate than leaf samples ( 11 . 31 versus 6 . 19; BM test , P = 0 . 007 ) . Second , we consider sampling of tissue by using microholes . To this end , two peach leaves and four petals were sampled from different branches of the tree HY1 . In two leaves , 16 tiny holes , each containing about 1 , 000 cells , were punched ( S6 Fig and S5 Table ) . After amplifying and sequencing those samples , 59 mutations were identified , with an average of 3 . 69 mutations per hole sample in leaves , much lower than 1 , 567 mutations per hole in petals . Although this latter result agrees qualitatively with the prior result using whole peach leaf and petal samples , the ratio is clearly much higher when using the microhole methodology . The mutation number seen in petals is so high we must suspect a technical artifact . Our method involves each petal sample being compared to all other petal and leaf samples ( i . e . , between different tissues ) from the same tree , and the mutations being called are those unique to a single hole sample and not in any other leaves or petals . This is potentially prone to false positives , as it requires few consistency checks and could be liable to sequencing artifacts introduced during amplification . We can be confident that the numbers are not sequencing errors , as 39 mutations that we can retest via Sanger sequencing 36 ( 92% ) are verifiable . However , we may be doing little more than confirming amplification artifacts . To be confident of a qualitative difference between leaf and petal , we therefore also ask about mutations that are shared between different microholes but are specific to a single flower/leaf . Such mutations are unlikely to be amplification or sequencing artifacts . We observed 73 and 16 mutations in two flowers ( the four petals belong to two flowers from two branches ) that were present in at least two microhole samples ( both supported by at least five reads for the mutation allele ) . This contrasts with 4 and 2 shared mutations in two leaf samples from two branches ( one-sided comparison of Poisson rates , P = 0 . 000974 ) . Although this does not resolve the cause of the remarkably high mutation number called singly in microholes of petals , it reinforces the conclusion that petal has more mutation accumulation than leaf and , as such , is consistent with highly relaxed selection in very short-lived petals . We have provided evidence that different plant tissues have different rates of mutation accumulation , the variation being consistent with an adaptive optimal allocation model . A further prediction is that the variation should be deterministic and hence that between biological replicates there should be no more heterogeneity than expected under a null of equal rates . We address this issue by asking whether different branches also differ in their mutation rates . A key problem in any such analysis is controlling for heterogeneity in the number of new mutations held on a branch that results from something as trivial as different ages of branches . To circumvent this , we consider 75 terminal branches on a young peach tree ( DHQ1 ) in which we can be confident that all the branches sampled are of approximately the same age . We then consider the number of mutations that are unique to any given branch . If the null model is correct , the distribution of these numbers should be a Poisson function , and hence the dispersion ( = Variance/Mean ) should be no different from unity . We find the dispersion is D = 1 . 031 . Significance we tested via simulation ( 10 , 000 replications ) , deriving a mean D in simulants of 1 . 0 ± 0 . 162 ( SD ) . The observed dispersion is thus no higher than expected by chance ( from simulation , P = 0 . 454 ) . Although then branch-specific mutations can be found ( each branch has on average 0 . 45 branch-specific mutations ) , we see no evidence for between-branch heterogeneity in rates . The in vivo evidence is broadly consistent with selectionist models in which we expect a lower mutation rate in cells in which any mutation has a larger potential future impact ( longer-lived terminal tissues or potential germline tissue ) and the variation observed is deterministic . But might there also be variation that is nonadaptive and better explained by mutational fragility ? We address this by comparing plants grown under very different conditions but over the same time span . The artificial condition is tissue culture , which we compare with the same plant grown in the wild . We considered a 1 . 1-mg callus derived from a single rice seed . This was cultured to 657 . 3 mg ( about 10 cell divisions ) and then divided into five groups with 10 seedlings regenerated from each of them ( S7 Fig ) . When the leaves were sequenced , the mutations specifically generated during culturing can be identified . This results in an average of 357 mutations for each seedling ( each regenerated plant was grown for approximately 2–3 months before sampling ) , which is approximately 47-fold higher than the number of mutations accumulated among different tillers in the same plant ( S6 Table and S7 Fig ) , even though the wild-grown plants are possibly older ( grown for about 3–4 months before sampling ) , indicating a high rate of mutation per unit time in the callus . We have presented evidence for differences in rates of mutation accumulation between different tissues in the same plant that we have postulated to be owing to differences in the mutation rate . However , an alternative possibility is that different tissues have the same mutation rate but differential degrees of purifying selection . Although this is unlikely to explain more-extreme differences , this has been a potential issue in the debate as to whether plants have a mutationally protected germline: prior data suggesting this [11 , 12 , 29] have been argued to alternatively be explained by purifying selection removing mutations , not by mutation not generating them [16 , 30] . Here then , we ask whether intra-organismic purifying selection is an important problem ( nota bene: we do not attempt to directly ask if any putative germline has a low mutation rate ) . We adopt two approaches . First , we ask whether the rate of mutational accumulation decreases as a function of age , taking advantage of our ability to determine , given the branching structure of a plant , when any given mutation arose . Second , we compare the transmissibility of harmful ( nonsynonymous ) and less harmful mutations . In neither case do we detect a signal of purifying selection . We thus presume that our measures of mutation accumulation are not profoundly confounded by intra-organismic selection . In addition , this supports the evidence that plants might have an effective germline [11 , 12 , 29 , 37]—i . e . , an early segregating and slowly dividing germline that accumulates few premeiotic mutations—as the low number of germline mutations cannot be easily explained by purifying selection .
We postulate that if selection is acting on the rate of mutation accumulation in plants , then stem should commonly have a lower rate than shoot; that highly ephemeral structures such as petals should have higher rates than ontogenetically related but longer-lived structures ( i . e . , leaves ) ; and that mutation accumulation rates should be deterministic ( i . e . , no more between biological replicate variation than expected under a null ) . We find all to be upheld and , in addition , that the one exception ( runner mutation rates in strawberry ) may well be the exception that proves the rule , as it appears to have a segregated germline . The high mutation rate in tissue culture , however , suggests that mutation rate is quite easily altered by changes in local environment . Evidence that stress can cause increases in the mutation rate may well partially explain the callus result [34] and in turn suggests that we need to be cautious in our interpretation of between-tissue differences . The variation that we observe suggests that in plants , mutation accumulation rates are deterministically variable between different parts of the same plant . This does not argue in favor of either hypothesis ( selection versus fragility ) , assuming that given tissues have consistent mutational microenvironments . Although then some of the heterogeneity that we observe is predictable from an adaptive model ( e . g . , stems have lower rates than roots ) and some predictable on anatomical grounds ( different mutations in different branches ) , we cannot definitely rule out nonadaptive mutational fragility . We propose that in testing adaptive theories of intra-organismic mutation rate variation , the alternative hypothesis should be that heterogeneity reflects mutational fragility that is conditional on local context . The petal—leaf differences we see could , in principle , be consistent with either hypothesis . By contrast , evidence for variation in the shoot—root difference as a function of the proportion of shoot mutations that are transmissible argues against microenvironment and in favor of the adaptive hypothesis . This does not rule out the possibility that the root microenvironment is more mutagenic than the shoot environment , but any such effect cannot obviously explain why the root/shoot ratio varies with the proportion of shoot mutations that are transmissible . We conclude that prima facie , at least some of the mutation rate variation observed best fits a selectively driven model , whereas some is just consistent with such a model . Regardless of whether the variation that we have observed is predictable in an adaptive framework , we can in addition ask whether the difference in mutation rates ( leaf—petal , root—shoot ) that we find reflects differences per cell division or differences in the number of cell divisions [35] . As far as adaptive theories are concerned , reducing either parameter would be an effective means to reduce the net rate of mutation accumulation . The root—shoot differences we suggest probably reflect differences in rates per cell division . Our strategy required a mutation to be called when the majority of a leaf ( or a root ) shares a single given mutation . The requirement means that this mutation is most likely to be derived from a single or a few cell divisions , and this division must occur at a very early stage of leaf ( or root ) development . When defining a mutation with ≥5 reads in a total of 40× genome coverage ( the maximal mutant reads is 20× in a diploid , assuming no bias ) , on average the mutation must be shared by ≥25% of cells in a sample . In other words , our sampling strategy will detect mutations that occur in one of a few early cell divisions for a leaf or root . Therefore , the mutations observed from such samples most probably reflect the mutation rates of SAMs or RAMs per cell division . Similarly , the high rate in the callus is probably best explained by an increased mutation rate per cell division . Thus , although the concentration in focus has been on strategies to minimize the number of cell divisions to protect the germline , the possibility of modification of the per-cell-division mutation rate , as seen in humans [5] , should not be discounted . The suggestion that the differences between shoot and root might reflect per-cell-division differences also accords with prior evidence suggesting that root and shoot have similar growth profiles . Estimation of the mitotic index , for example , indicates that the duration of mitotic cycle is roughly the same in shoots and roots [36] . This does not , however , take into account the finding that both SAM and RAM contain some cells that divide much slower than others [11 , 12] . This is likely to influence mutation accumulation because somatic mutation occurrence is correlated with the number of divisions [35] , regardless of the per-cell-division rate . Prior work has suggested that plants might have an effective germline [11 , 12 , 29 , 37] , which may be early segregating and slow dividing , thus accumulating few premeiotic mutations . We did not seek to test this hypothesis directly , but our lack of evidence for intra-organismic selection argues against the hypothesis that the reduced mutation rate observed by others may be more apparent than real . Although we did not set out to test the germline hypothesis , our analysis of woodland strawberry strongly supports the possibility that runners in this species have two discrete cell lineages: one that can be propagated to future runners but not to lateral buds and one that can be propagated to all ( i . e . , a germline ) . The relatively high mutation rate in runners makes sense , if this is the case , as most of the runner is “root-like” in having no gametogenic future . Our data , however , have little to say as to whether this germline is mutationally protected . There is no further mutation in runner that cosegregates ( ontogenetically ) with the germline one ( mutation 5 ) that would be consistent with a low rate . However , after mutation 5 in runner , we see only one further mutation ( mutation 11 ) , all others occurring after lateral bud development . The chance that this new mutation would not be germline must be very high , even if there is no mutation rate difference . We suggest that this system would be valuable for further interrogation of the hypothesis of a mutationally protected germline and for mechanisms of cell lineage sequestration , not least because runner also makes a helpful control for the possibility that root might have a high mutation rate owing to its subterranean environment .
We collected a total of 22 plant individuals , including seven peach ( P . persica ) and four wild peach ( P . mira ) trees , two plum ( P . mume ) trees , one woodland strawberry ( F . vesca ) , one shrub willow ( S . suchowensis ) , four rice ( O . sativa ) , one B . distachyon , and two A . thaliana individuals ( S7 Table ) . The sampled individuals cover a life span range from several months to hundreds of years and three distinct genera . One of the peach trees was sampled in Maoping , Guizhou Province , China , and the others were from Nanjing , Jiangsu Province . The wild peach trees were sampled in Nyingchi , Tibet . The A . thaliana individuals were derived from two seeds of a single Col-0 plant . Three rice individuals , including one O . sativa L . cv . Pei-Ai 64s ( PA1 ) , two O . sativa ssp . indica cv . Kasalath ( KA1 ) , and cv . Dee-geo-woo-gen ( DG1 ) , were obtained from the International Rice Research Institute ( IRRI ) . The plum trees were sampled in Nanjing . The shrub willow YAF1 was kindly provided by Jiangsu Forestry Science Academe in China . The woodland strawberry was obtained from Nanjing Agricultural University , which was the same accession ( Hawaii 4 ) as the reference genome . The seeds of B . distachyon diploid inbred line Bd21 ( WD2 ) were obtained from South China Agricultural University . In total , 480 leaves were sampled from the terminal branches of 21 plants . For rice DG1 , B . distachyon WD2 , willow YAF1 , plum MHG1 , and peach PXL , 25 , 8 , 22 , 32 , and 13 root samples were collected , respectively . One bark sample was also prepared for PXL . Seven lemma samples were collected for WD2 before maturing . Four stem samples were obtained from strawberry FH1 . For a wild peach tree GL2 , 14 fruits were also sampled at the same date as its leaves . Those fruits were treated with gibberellin to accelerate germination . For a plum tree MHG1 , 21 fruits were sampled 7 months after the leaf sampling . The age of peach trees was estimated using a growth cone . The ages of wild peach trees were estimated based on work by Wang and colleagues [33] . DNA was extracted using the CTAB method [38] . About three-quarters of leaf DNA samples were extracted using a single leaf or part of a single leaf , weighing approximately 0 . 08–0 . 7 g . DNA samples for fruits of MHG1 were extracted from the seeds after carefully removing the seed coats . For progeny of GL2 , the DNA samples were extracted from the leaves of seedlings after growing for approximately 1–2 months . The root and bark DNA samples were extracted after careful cleaning . To obtain microscale plant samples , we used a Harris micropunch ( 0 . 5-mm diameter ) to harvest a defined area of leaf . Genomic DNA of microscale samples was amplified with a Qiagen REPL-g single-cell kit following the kit instructions . All plant DNA was fragmented into an insert size of about 300–350 bp and sequenced on the Illumina Hiseq4000 platform with 150-bp paired-end reads at BGI . Detailed statistics of sequencing results are provided in S1 Data . Whole-genome sequences and annotations for peach [39] and woodland strawberry [40] were downloaded from Genome Database for Rosaceae ( GDR , https://www . rosaceae . org , version 2 . 0 . a1 ) . Both peach and woodland strawberry have a compact genome ( about 240–260 M ) and a qualified reference genome that is both of high accuracy and completeness . The peach genome was initially sequenced using Sanger reads and assembled into eight chromosomes [39] , this being subsequently improved with additional linkage maps and NGS reads [41] . The woodland strawberry was initially sequenced with NGS reads , assembled into seven chromosomes [42] , and improved by dense targeted capture linkage maps [43] . The reference genome and annotations for the plum tree [44] were downloaded from http://prunusmumegenome . bjfu . edu . cn , mirrored at https://github . com/lileiting/prunusmumegenome . The rice reference genome [45] was downloaded from the Rice Annotation Project Database ( RAP-DB , http://rapdb . dna . affrc . go . jp/ , version IRGSP-1 . 0 ) , and the Arabidopsis reference genome [46] was obtained through The Arabidopsis Information Resource ( TAIR , http://www . arabidopsis . org/ , version 10 ) . Each sample was sequenced to a cleaned depth over 40× with qualified bases ( base quality ≥ 20 ) over 90% after removing adaptors and low-quality reads ( i . e . , reads containing more than 50% low-quality bases ) . Cleaned reads were mapped to each reference genome using BWA-mem 0 . 7 . 10-r789 [47] with default settings . The resulting BAM files were then sorted and processed with MarkDuplicates in Picard package ( version 1 . 114 ) to remove noninformative PCR duplicates . A local realignment step was also implemented using RealignerTargetCreator and IndelRealigner in GATK package version 3 . 5 . 0 [48] to reduce false variant calls due to alignment errors around insertions/deletions ( indels ) . The rice root samples were susceptible to bacterial contamination , which resulted in lower effective coverage . We excluded those samples with extremely low coverage ( <45% ) from further analysis and only used those samples to exclude false positives . The MHG1 individual was found to be from grafting; thus , the branches and the root were analyzed separately as independent systems . Single-nucleotide variants ( SNVs ) and small indels were called using two distinct algorithms implemented in GATK: UnifiedGenotyper ( UG ) and HaplotypeCaller ( HC ) . Only reads with a mapping quality over 20 ( i . e . , less than 1% error rate ) were considered . The initial candidate mutations were called by comparing the samples within the same branch against all other branches based on the branching topology ( S1 Fig ) . A variant would be called as a candidate mutation if the allele is different from that in the comparator branches , which we presume reflects the ancestral state . This parallel comparison approach has been demonstrated to be robust against sequencing or mapping artifacts and has a relative low false-negative rate [26] . We compiled a series of criteria for filtering and evaluating the initial candidates ( S1 Fig ) . Those criteria deal with all respects of sequencing , mapping , or calling errors . First , we filtered candidates with low variant quality ( quality score < 50 given in VCF file ) , low depth ( no sample carries ≥5 putative mutated reads ) , or many missing calls ( no variant calls in more than 5 samples ) . For mutations only found in a single sample , we required the focal sample ( the sample assumed to carry a mutation ) to contain at least five reads . For mutations shared by >1 sample , at least one sample should fit this criteria , and other samples should have no fewer than three reads carrying the same mutations . Variants that failed any of these criteria were assumed to be sequencing errors . We also removed candidates that were biased in read strands ( only have forward or reverse strands ) , a signature of erroneous mapping artifacts from duplications . Second , we masked the remaining candidate sites that ( 1 ) have missing calls but no more than four samples , ( 2 ) have two or three reads with the same “mutated” alleles of base quality over 20 ( termed as “mimic reads” hereafter ) among all compared ( control ) samples , or ( 3 ) could only be captured by UG caller , this having the higher false-positive rate . Candidates passing all those criteria were considered the “confidence set , ” otherwise they were treated as the “evaluation set” ( S1 Fig ) . We further manually investigated all candidates in the confidence set and part of the evaluation set , from most evidence to least . For each candidate mutation site , the Integrative Genomics Viewer ( IGV ) was applied to review the mapping states across all related samples . Loci found to have resulted from spurious mapping artifacts or contamination ( detected by BLAST search in NCBI Nucleotide collection database using the aligned reads ) were discarded . An additional round of inspection was performed for indels and SNVs around indels . We first extracted reads mapping to each candidate region and then realigned them to the reference sequence with ClustalW2 . The regenerated alignments were saved in FASTA format and further revised in MEGA6 to get the best possible alignments . From this , we confirmed whether the candidate is a true variant or just a misalignment artifact . A candidate “mutation” is considered false if it ( 1 ) is an alignment artifact , mostly found in regions containing indels , regions divergent between the reference genome and analyzed genome , regions harboring large genomic rearrangements , duplications ( which easily cause the wrong placement of reads ) , etc . ; ( 2 ) is a preexisting variant ( i . e . , one actually present in all other samples ) but happened to be called only in some samples ( this situation is most likely due to some subtle differences in reads—e . g . , slightly more sequencing errors in some—covering the candidate site between different samples , which cause some samples to pass the threshold and be called by the caller while others happened to fail ) ; ( 3 ) is a contamination artifact , either from impurities on the sample’s surface or “sample bleeding ( index hopping ) ” of multiplexed samples; or ( 4 ) resulted from sequencing errors , mainly found in regions with homopolymers or tandem repeats , which have dubious lengths among different samples and thus are less likely to be bona fide mutations . It was found that the false discovery rate increased rapidly with increasing numbers of missed calls because there was more sequencing bias when relatively few samples were properly amplified and sequenced . The same situation was found for more “mimic reads . ” Assuming all “mimic reads” were only from sequencing errors , for two mimic reads to be present in compared samples would require a probability less than ( 1% base error rate × 1/4 the same allele by chance ) 2 = 6 . 24 × 10−6 . In practice , we found that the presence of more than one mimic read was mostly a signature of false mutation calls due to sequencing or mapping artifacts . Another false-positive source was from misalignments around indels , which could be witnessed as a high error rate in candidates called by UG alone . The UG algorithm directly calls variants from the alignments and thus is capable of capturing most SNVs but could have a high false-positive call rate due to misalignment , especially around indel sites . The HC algorithm has fewer positive SNV calls and performs better in indel detection compared to UG , as HC implements a local reassembly algorithm . However , it was found that the HC caller occasionally lost a few SNVs , possibly because of the non-lossless GVCF mode or the reassembly process . We integrated results from the two callers in later analyses ( S1 Fig ) . Through these mechanisms , we minimized both the false-positive rate and false-negative rate caused by the variant callers [49] . This was confirmed during the manual inspection stage . From the SNV mutations identified in this study , we found around 95 . 5% of SNVs could be called by both HC and UG callers , whereas 2 . 6% were only called by HC and 1 . 9% were only called by UG . For indel mutations , around 83 . 4% of indels were found in both call sets , whereas 15 . 8% could only be called by HC and 0 . 8% were only called by UG . In general , the manual inspection suggested the confidence set could capture over 90% of candidate callable mutations ( calculated as “Manually confirmed mutations in confidence set” / “Manually confirmed mutations in both confidence and evaluation set” ) for base substitutions within accessible regions , whereas ignoring the evaluation set would only cause a false-negative rate of no more than 10% . As the evaluation set was manually investigated from higher confidence to lower , it was mostly likely that all callable mutations were captured in our analysis . During the filtering stage , we also observed a certain number of cross-sample contaminations . Those contaminations were only found within different individuals that were sequenced in the same sequencing lane . A small number of reads that were believed to belong to one individual could be observed in another individual , especially in genomic regions with ultrahigh read coverage ( e . g . , >100× ) . This contamination was unlikely to be owing to early experimental mistakes , as each sample was processed independently during the DNA extraction and library construction stage . We could also rule out the possibility of read-assign errors , as the barcodes used for each sample were very different . Therefore , we conclude that those cross-contaminations are most likely a result of cluster-detection errors during the sequencing stage , known as “sample bleeding” [50] . These contaminants were removed by comparing against unrelated samples within the same lane . The “topology-based” method could miss mutations that occurred originally in soma but were fixed across different branches . We searched for these heterogeneous sites that have a variant allele present in only some ( “M” ) of all “N” samples , hereafter referred as the frequency-based method ( S1 and S8 Figs ) . The frequency-based method then compares every possible combination of M samples ( focal samples supposed to carry the mutation allele ) with the remaining N-M samples ( treated as “control” samples ) using the same criteria used in the “topology-based” method ( like a comparison between two “branches” ) . Variant alleles present in over 0 . 8 * N samples were not considered as mutations because they ( 1 ) could well be preexisting variants for which not all samples were properly genotyped , owing to sequencing/mapping/calling biases , and ( 2 ) could not be distinguished from somatic recombination events ( for further logic and illustration , see S8 Fig ) . This method was more prone to various analytical errors , as the “control” samples were often inaccurate , which could miss true mutations if mutated samples were included in the control group ( see Site4 in S8 Fig ) , while generating false candidates if insufficient samples were included in the control group ( see Site5 in S8 Fig ) . Therefore , we only considered the most confident sites present in several samples defined as before but with no evidence in all other samples ( e . g . , no mimic reads allowed ) . Results from the frequency-based method could also be used to correct any errors in topology records ( S1 Fig ) . For instance , the relationship of five primary branches of GL2 tree are almost indistinguishable ( Fig 2A ) and could only be treated as five independent branches , whereas , based on mutations shared between them , the frequency-based analysis suggested the branches B1 and B2 are actually ontogenetically closer ( Fig 2B , same for B4 and B5 ) . Only substitutions and small indels ( e . g . , <100 bp ) were investigated in this study . Substitutions include SNV and multiple-nucleotide variants ( MNVs ) , whereas indels contain pure insertions ( INSs ) , deletions ( DELs ) , and complex replacements ( RPLs; i . e . , nonequal-size base substitutions ) . A full list of all identified mutations can be found in S2 Data . Although new mutations are expected to be heterozygous , we did not filter with heterozygosity as a requirement; rather , we require a mutation to be different from the ancestral state . This decision was based on the premise that there exist several situations that could cause real mutations to be witnessed in an apparent or real “homozygous” state . These include subsequent somatic recombination leading to the loss of the nonmutated haplotype , sequencing bias in which only the mutated haplotype gets sequenced , mapping issues in which only the mutated haplotype is properly mapped , etc . As it happens , 99 . 0% of mutations we identified in this study were heterozygous . We used PCR and conventional Sanger sequencing to validate 89 mutations in 122 mutated samples , 59 progeny samples , and 274 control samples . The mutated samples were confirmed at a rate of 96 . 7% ( 118/122 ) . The unconfirmed instances could reflect false positives or a failure of the PCR to amplify the mutant allele . No mutant allele was found in control samples . The number of callable sites and the false-negative rate were estimated using a simulation method similar to that described previously [26 , 51] . The read-depth distribution for each group was based on the real mutations identified . For each tree , we generated 1 , 000 synthetic mutation sites in one or several branches according to each topology . The leaf and root samples were simulated separately because they had different read-depth distributions . The same pipelines were then used to detect these synthetic mutations . The fraction of callable sites in the genome for each tree was then estimated as the fraction of callable simulated mutated sites ( S8 and S9 Tables ) . Supposing a mutation is inherited by all the cells in a branch , and then each cell has a genotype of Aa ( “A” is the wild allele , and “a” is the mutated allele ) . Fifty percent of all the gametes produced by this branch thus are expected to carry the “a” allele , and 50% carry “A . ” The probability of the absence of “a” in all of the fertilized eggs would be ( 1/2 ) n , where n denotes the number of seeds . Statistics and correlation test were performed in R [52] . BM test was implemented in the R package “lawstat . ”
|
Whereas there has been considerable attention paid to understanding differences in the mutation rate between different species , much less is known about variation in the mutation rate within individuals of multicellular species . In animals , evidence suggests that the segregated germline has a lower mutation rate than somatic cells , which accords with an adaptive model of intra-organism mutation rate variation . Here , we consider related questions in plants , which , although not having an early segregated germline , present numerous opportunities to test such an adaptive model . In particular , such a model would predict a lower rate in shoots than in roots and in relatively long-lived leaves compared with ontogenetically related but more ephemeral petals . In addition , we expect that mutation rate variation should be deterministic , such that there is no more heterogeneity in mutation rates between similarly aged branches than expected by chance . By sequencing several hundred genomes for numerous different species , we find all predictions to be supported , with the proviso that the root—shoot difference is witnessed in perennials but not in an annual . This last difference we show to be explicable , as annuals transmit relatively few premeiotic mutations . Although the adaptive model is then parsimonious , we caution that the mutation rate in plants can be easily affected by local conditions , as evidenced by large differences between lab-grown callus and field-grown plants .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"biotechnology",
"plant",
"anatomy",
"engineering",
"and",
"technology",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"trees",
"mutation",
"substitution",
"mutation",
"plant",
"science",
"peach",
"trees",
"plant",
"genomics",
"plants",
"extraction",
"techniques",
"flower",
"anatomy",
"research",
"and",
"analysis",
"methods",
"bioengineering",
"plant",
"genetics",
"leaves",
"somatic",
"mutation",
"dna",
"extraction",
"eukaryota",
"cell",
"biology",
"petals",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"plant",
"biotechnology",
"organisms"
] |
2019
|
The architecture of intra-organism mutation rate variation in plants
|
Increased availability of Next Generation Sequencing ( NGS ) techniques allows , for the first time , to distinguish relapses from reinfections in patients with multiple Buruli ulcer ( BU ) episodes . We compared the number and location of single nucleotide polymorphisms ( SNPs ) identified by genomic screening between four pairs of Mycobacterium ulcerans isolates collected at the time of first diagnosis and at recurrence , derived from a collection of almost 5000 well characterized clinical samples from one BU treatment center in Benin . The findings suggest that after surgical treatment—without antibiotics—the second episodes were due to relapse rather than reinfection . Since specific antibiotics were introduced for the treatment of BU , the one patient with a culture available from both disease episodes had M . ulcerans isolates with a genomic distance of 20 SNPs , suggesting the patient was most likely reinfected rather than having a relapse . To our knowledge , this study is the first to study recurrences in M . ulcerans using NGS , and to identify exogenous reinfection as causing a recurrence of BU . The occurrence of reinfection highlights the contribution of ongoing exposure to M . ulcerans to disease recurrence , and has implications for vaccine development .
Buruli ulcer ( BU ) is a neglected necrotizing skin and bone disease caused by the enigmatic pathogen Mycobacterium ulcerans that occurs in riverine regions of West and Central Africa . The clonal nature of M . ulcerans has complicated molecular analyses of the epidemiology of the pathogen , as genotyping methods with sufficient resolution have been lacking [1] . Using insertion sequence element single nucleotide polymorphism ( ISE-SNP ) typing , a technique in which two relatively small regions ( 1 , 431 and 1 , 871 bp ) of the M . ulcerans genome are screened for polymorphisms , strains could be distinguished to the level of the river basin in West Africa [2] . However , such molecular genotyping techniques lack sufficient resolution to distinguish relapse from reinfection with an unrelated exogenous strain of M . ulcerans among BU recurrences . The World Health Organization ( WHO ) defines a relapse as a recurrence of BU within one year after termination of antibiotic treatment [3] . A recurrence that appears after that period is consequently considered a reinfection . The contribution of relapses versus reinfections to BU recurrences and their biological basis are to date unknown . Until the routine use of antibiotics ( rifampicin and streptomycin ) was advocated by the WHO in 2004 , surgery was the mainstay of BU therapy . After surgical treatment only , a recurrence rate of 6% was reported in Benin [4 , 5] . Higher recurrence rates were reported in Ghana ( 16%-35% ) [6 , 7] , Ivory Coast ( 17% ) [8] , Uganda ( 20% ) [9] , and Australia ( 32% ) [10] . When specific antibiotics are used very few , if any , recurrences are observed [11–13] . The introduction of Next Generation Sequencing ( NGS ) now allows for the first time to distinguish relapses from reinfections in patients with multiple BU episodes . Similar studies have been conducted for Mycobacterium tuberculosis [14 , 15] and other monomorphic bacterial infections , such as Clostridium difficile [16] . In the present study we compared the number and location of SNPs identified by genomic screening between four pairs of M . ulcerans isolates collected at the time of first diagnosis and at recurrence , derived from a collection of almost 5000 well characterized clinical samples from one BU treatment center in Zagnanado , Benin , between 1989 and 2010 .
We defined an episode as a clinical suspicion of BU and a recurrence as the presence of two episodes separated by at least six months . Four such patients were identified for this study . We compared the number of SNP differences separating the paired isolates and a random selection of 36 isolates from 36 patients living in the same geographical area and diagnosed with BU in the same time frame ( 1998–2008 ) as the patients with multiple episodes . We also compared genomic relatedness between six patients with two M . ulcerans cultures isolated from the same disease episode . This genetic background helped to avoid misclassifying any second episodes with similar M . ulcerans strains prevalent in the patient’s environment as relapses rather than reinfections . As such a total of 58 M . ulcerans isolates obtained from 46 patients was included in this study ( Fig 1 ) . DNA was obtained by harvesting the growth of three Löwenstein-Jensen slants followed by heat inactivation , mechanical disruption , enzymatic digestion and DNA purification on a Maxwell 16 automated platform , a technique modified from Käser et al . [17] . Whole genome sequencing of the isolates was performed using an Illumina HiSeq 2000 DNA sequencer and an Illumina MiSeq DNA sequencer with Nextera XT or TruSeq ( Illumina Inc . , San Diego , CA , USA ) library preparation and 2x36bp 2x100bp 2x150bp 2x250bp sequencing paired-end chemistry . Sequencing statistics are provided in S1 Table . The sequencing data analysis was done using the Nesoni software [18] . Firstly , reads were filtered to remove those containing ambiguous base calls , any reads <50 nucleotides in length , and reads containing only homopolymers . All reads were furthermore trimmed removing residual ligated Nextera of TruSeq adaptors and low quality bases ( <Q10 ) at the 3' end . Average read lengths after clipping are summarized for all isolates in S1 Table . Bowtie2 v2 . 1 . 0 [19] was used with default parameters to map clipped sequence reads sets to the Ghanaian M . ulcerans Agy99 reference genome ( Genbank accession number: CP000325 ) . Due to the unreliability of read mapping in repetitive regions , all ISE elements ( IS2404 and IS2606 ) were hard masked in this reference genome . Average read depths after mapping to M . ulcerans Agy99 are summarized for all isolates in S1 Table . We compared the number and location of SNPs between isolates collected at baseline and at recurrence . At each of the loci called as a variant in any read set , Nesoni was used to generate a multi-way summary of consensus allele calls at the corresponding locus in all other read sets of the investigated panel . By concatenating all these loci a multiple SNP sequence alignment was generated containing all 282 variant loci across the Agy99 reference chromosome sequence . A maximum likelihood ( ML ) phylogenetic tree was constructed from this alignment using RAxML v8 . 0 . 19 [20] under a GTR model of evolution ( no rate heterogeneity ) and with an ascertainment bias likelihood correction for SNP data . The resulting tree was visualized in Figtree v1 . 4 . 0 [21] with nodes of interest highlighted . A haplotype network was derived using the median joining algorithm [22] as implemented within SplitsTree v . 4 . 13 . 1 [23] with default settings . This network was subsequently visualized with Hapstar v . 0 . 7 [24] . The open source geographic information system Quantum GIS ( QGIS v1 . 8 . 0 ) [25] was used to generate the illustration of the geographical distribution of all included M . ulcerans genomes in Fig 2 . The geographical locations of the residences of BU patients at the time of their first consultation are shown . The river layer ( Ouémé river and its tributaries ) was digitized from declassified Soviet military topographic maps b31-03 , b31-09 , and b31-15 ( scale 1:200k ) . The administrative borders of African countries were rendered from the Global Administrative Unit Layers data set of FAO [26] . This retrospective study on stored isolates was approved by the Institutional Review Board of the Institute of Tropical Medicine . Read data for the study isolates have been deposited in the NCBI Sequence Read Archive ( SRA ) under accession n° PRJNA296792 .
Among the 4951 clinically BU suspected patients who consulted the BU treatment centre of Gbemotin in Zagnanado in southern Benin between 1989 and 2010 , we identified 100 who presented with multiple BU episodes ( Fig 3 ) . A majority of 93 patients had two disease episodes while 7 had three episodes . Twenty recurrence patients received streptomycin and/or rifampicin during their first BU episode . The distribution of patients that had received ( partially ) effective antibiotics is shown in the S1 Fig . Only for seven of the 100 recurrence patients were we able to successfully culture isolates from each of two or three disease episodes , owing to the limited sensitivity of culture for isolation of M . ulcerans from skin biopsies [27] . These mycobacterial isolates were stored at ≤70°C in Dubos broth enriched with growth supplement and glycerol . However , paired cultures were found to be viable for only four of these seven patients . The first two patients of these four patients each had three paired isolates while the other two patients each had two ( Table 1 ) . We assessed the paired M . ulcerans isolates of four patients ( letter-coded from A to D ) with two BU episodes each ( Table 1 ) . Each of these 4 patients presented for the first time at the BU treatment centre of Gbemotin between 1999 and 2004 . Patients A & B had lesions at the same location during both episodes , while patients C & D had lesions at another body site . The time between the diagnoses of both episodes ranged from seven months to almost two years . All patients underwent surgery , while patient D also received specific antibiotics , which were introduced for the routine management of BU in 2004 [3] . Comparative genomic analysis identified two patients ( B & C ) with no detectable genetic differences ( 0 SNPs ) between isolates originating from two disease episodes . Patient A had one SNP difference between his first episode isolate and both second episode isolates although at different positions , suggesting that micro-evolution took place . For patient D however the isolates of both disease episodes were differentiated from each other by no less than 20 SNPs , which were found distributed throughout the genome . This is a genetic distance similar to that observed between different isolates from the same geographic region as illustrated in the haplotype network ( Fig 2B ) and the phylogenetic tree ( red nodes , S2 Fig ) . The difference in SNPs between control patients varied from 0 to 53 SNPs . To exclude cross-contamination of the M . ulcerans culture obtained from the second episode of patient D , the genome obtained from a strongly positive biopsy that was treated on the same day as the biopsy of patient D was sequenced as well and found to differ by 21 SNPs . There were 2 clusters of 5 and 4 patients having identical M . ulcerans genomes , who lived in an area of respectively 70 km² ( 4 . 7–27 km between villages ) and 14 km² ( 3–18 km between villages ) . Four patients with multiple M . ulcerans isolates from the same BU episode had identical paired genomes while one such patient differed in 1 SNP and another one in 6 SNPs between the paired genomes with respectively 8 and 7 days between the times of sampling of the biopsies from which the M . ulcerans cultures were isolated ( Fig 2B ) . Among the total of 24 substitutions that were identified in patient A and patient D , four occurred in intergenic regions , while 20 were found in coding sequences resulting in 3 synonymous changes ( i . e . ‘silent’ changes ) and 17 non-synonymous changes ( i . e . resulting in a change in amino acid ) , in genes encoding proteins with various functions ( S2 Table ) .
This study is the first to study recurrences in M . ulcerans using NGS , and to identify exogenous reinfection as causing a recurrence of BU . In relapses , paired isolates are genetically more related to each other than to isolates from other patients living in the same region and infected during the same time-period . In reinfections on the other hand , paired isolates are potentially not more related to each other than to isolates from other patients living in the same region and infected during the same time-period . Our results suggest that the second BU episode of patients A , B and C was most likely due to relapses . The second BU episode of the 4th patient was probably due to a reinfection . This patient is also the only one who had received specific antibiotics during his first BU episode . We can however not be entirely certain that patients A , B and C were not infected a second time by an identical M . ulcerans strain , as we identified other genetically identical clusters among patients living in the same area and time period . As the mode of transmission of M . ulcerans remains enigmatic , detailed investigation of these genetic clusters may provide leads to a common point source of exposure . To our knowledge , this study is the first using NGS to assess recurrences in M . ulcerans , which is important to understand BU epidemiology . In BU , when disease re-occurs within one year after the end of treatment it is assumed to be a relapse which is considered the primary endpoint in several studies [11–13] and in patient management . This study for the first time indicates that exogenous reinfection plays a role in recurrence of BU . However , the restricted phylogeny presented here based on four patients should be interpreted with some circumspection because of the limited sample size , which is due to the overall low recurrence rate at this treatment center , combined with the low sensitivity of in vitro culture of M . ulcerans . In other mycobacterial infections , most notably infection with M . tuberculosis , the number of different SNPs between relapse and reinfection pairs can be large , with a clear distinction between pairs with a small difference ( ≤6 SNPs ) , classified as probably relapses , and those with a large difference ( ≥1306 SNPs ) , deemed to be reinfections [15] . Epidemiologically linked M . tuberculosis populations have been reported with a mean SNP difference of 3 . 4 demonstrating high genomic stability [28] . Walker and colleagues [29] reported that in patients with an epidemiological link the divergence between their M . tuberculosis genomes generally does not exceed 14 SNPs , with most patients having fewer than 5 SNP differences during one disease episode . During recurrences of Clostridium difficile infection , paired isolates ≤2 SNPs apart were considered relapses while paired isolates >10 SNPs apart were considered reinfections [16] . In the present study we detected only one genomic difference between the relapse pairs , reaffirming the high genomic stability previously reported for M . ulcerans [1] . Apparent reinfections could theoretically result from differential sampling of an initially mixed infection . Patient D could possibly have had a simultaneous infection by two M . ulcerans strains although the results from the 6 patients with multiple isolates from one BU episode suggest that different M . ulcerans strains causing a single disease episode is , at best , an infrequent occurrence . The same was observed among four Cameroonian BU patients with multiple isolates from one BU episode [30] . This suggests mixed infection would have been an unlikely explanation for the genetic differences between the paired isolates of patient D . The occurrence of reinfection highlights the contribution of ongoing exposure to M . ulcerans to disease recurrence . Since the delay between relapse ( 7 , 9 . 5 , and 22 . 5 months ) and reinfection ( 9 months ) episodes overlap , the use of NGS on cultured isolates is required to distinguish these two scenarios . BU recurrences within a period of one year after antibiotic treatment that are considered relapses by WHO [3] may therefore also be reinfections . Since BU transmission is probably not human-to-human and the mode of transmission from the environment is not yet clarified , epidemiological links in support of transmission routes are speculative at best . However , we expect to be able to determine a SNP-threshold which can be interpreted as an epidemiological link consistent with a common source of infection in an ongoing study with a greater number of M . ulcerans genomes from the Ouémé river valley . The definition of transmission clusters could help to unravel the enigmatic transmission routes of M . ulcerans . NGS of paired M . ulcerans strains collected from patients with multiple episodes of BU has sufficient resolution to distinguish relapse from reinfection . Our results on a small number of patients suggest that after surgical treatment without antibiotics the second episodes were due to relapse rather than reinfection . Since specific antibiotics were introduced for the treatment of BU , the only patient with a culture available from both disease episodes had M . ulcerans isolates with a greater genetic distance , suggesting this patient was most likely reinfected .
|
We compared the whole genomes of four pairs of Mycobacterium ulcerans isolates collected at the time of first diagnosis and at recurrence , derived from a collection of almost 5000 well characterized clinical samples from one BU treatment center in Benin . Our findings suggest that after surgical treatment—without antibiotics—the second episodes were due to relapse rather than reinfection . Since specific antibiotics were introduced for the treatment of BU , the one patient with a culture available from both disease episodes had M . ulcerans isolates with a larger genomic distance , suggesting that the patient was most likely reinfected rather than having a relapse . To our knowledge , this study is the first to assess recurrences in M . ulcerans using whole genomes , and to identify exogenous reinfection as causing a recurrence of BU . The occurrence of reinfection highlights the contribution of ongoing exposure to M . ulcerans to disease recurrence , and has implications for vaccine development .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
A Genomic Approach to Resolving Relapse versus Reinfection among Four Cases of Buruli Ulcer
|
Because most efforts to understand the molecular mechanisms underpinning fungal pathogenicity have focused on studying the function and role of individual genes , relatively little is known about how transcriptional machineries globally regulate and coordinate the expression of a large group of genes involved in pathogenesis . Using quantitative real-time PCR , we analyzed the expression patterns of 206 transcription factor ( TF ) genes in the rice blast fungus Magnaporthe oryzae under 32 conditions , including multiple infection-related developmental stages and various abiotic stresses . The resulting data , which are publicly available via an online platform , provided new insights into how these TFs are regulated and potentially work together to control cellular responses to a diverse array of stimuli . High degrees of differential TF expression were observed under the conditions tested . More than 50% of the 206 TF genes were up-regulated during conidiation and/or in conidia . Mutations in ten conidiation-specific TF genes caused defects in conidiation . Expression patterns in planta were similar to those under oxidative stress conditions . Mutants of in planta inducible genes not only exhibited sensitive to oxidative stress but also failed to infect rice . These experimental validations clearly demonstrated the value of TF expression patterns in predicting the function of individual TF genes . The regulatory network of TF genes revealed by this study provides a solid foundation for elucidating how M . oryzae regulates its pathogenesis , development , and stress responses .
Fungal pathogenesis requires well-orchestrated regulation of multiple cellular and developmental processes in response to diverse stimuli from the host and the environment . Transcription factors ( TFs ) function as key regulators of such processes . Identification of TF genes , which typically represent 3–6% of the predicted genes in eukaryotic genomes , has been greatly facilitated by genome sequencing [1] . High-throughput methods for gene expression analysis have enabled studies on how TF genes are globally regulated under diverse conditions [2]–[4] . A combination of these approaches has uncovered putative roles and potential interactions of TFs in animals and plants [3] , [5] . Although DNA microarrays have been successfully used to study global gene expression patterns , this approach may not be sensitive enough to accurately analyze low-abundance transcripts , including those from many TF genes [6] . Quantitative RT-PCR ( qRT-PCR ) has been shown to be five times more sensitive than microarrays [4] , serving as an effective means for accurate quantification of TF transcripts . The rice blast fungus Magnaporthe oryzae , one of the most devastating pathogens of rice and related grass species , undergoes sequential developmental changes to successfully infect host plants and complete the disease cycles . These processes include conidiogenesis , conidial germination , appressorium formation , penetration peg formation and infectious growth . Extensive studies have been performed to identify and characterize the genes that participate in these developmental changes and pathogenicity in M . oryzae [7]–[11] . Recent functional analyses of several M . oryzae TF genes demonstrated their critical roles in processes such as conidiation ( COS1 , MoHOX2 , MoHOX4 , and COM1; [12]–[14] ) , appressorium formation ( MoHOX7 , MoLDB1 , and Con7p; [12] , [15] , [16] ) , infectious growth ( Mig1 , Mstu1 , MoHOX8 , and MoMCM1; [12] , [17]–[19] ) , oxidative stress ( Moatf1; [20] ) , and light regulation ( Mgwc-1; [21] ) . However , how M . oryzae TF genes are globally regulated and coordinated at the transcriptional level has not been studied . To address this knowledge gap , we analyzed expression patterns of 206 TF genes under 32 conditions , including infection-related developmental stages and various abiotic stresses , using qRT-PCR . To test the utility of expression profiles for predicting the role of individual TF genes in development and pathogenicity , mutants of selected TF genes were characterized . The resulting data clearly demonstrated their value . All the data from this study are publicly available through the Fungal Transcription Factor Database ( http://ftfd . snu . ac . kr/magnaporthe ) , an online platform designed to systematically identify and catalog TF genes in fungi [22] .
The data extraction pipeline of FTFD identified 495 putative TF genes ( 4 . 5% of the 11 , 054 protein-coding genes in M . oryzae ) using the InterPro terms associated with DNA-binding motifs . The proportion of TF genes in the total proteome of 23 other fungal and Oomycetes species ranged from 2 . 4% ( Laccaria bicolor ) to 6 . 4% ( Rhizopus oryzae ) ( Table S1 ) . Interestingly , 26 genes ( 5 . 3% of the TF genes ) belonging to 9 different TF families appeared to be M . oryzae–specific based on the lack of orthologs in other species , which was determined using basic local alignment search tool ( E<10−50 ) and InParanoid algorithm [23] ( Table S2 ) . According to the InterPro classification [24] , 495 M . oryzae TF genes were grouped into 44 families with the following four families dominating ( Figure 1 ) : fungal-specific Zn2Cys6 ( 141 genes; 28 . 5% ) , C2H2 zinc finger ( 89 genes; 18 . 0% ) , HMG ( 48 genes; 9 . 7% ) , and OB-fold ( 47 genes: 9 . 5% ) . Furthermore , 49 genes possessed more than one DNA-binding domains; among these , 29 of 35 homeodomain-like TF genes belonged to six different families . TFs with multiple DNA-binding domains are not unique to M . oryzae and have been detected in animals and plants [1] , [25] . A few genes , such as tubulins , actins , and elongation factors , have been used as references for normalizing M . oryzae gene expression data generated using RT-PCR or qRT-PCR [12] , [20] , [26]–[32] . To identify the most stable reference gene under all the conditions used in our study , we evaluated seven candidate genes: β-tubulin [12] , [31] , [32] , actin2 [20] , [29] , [30] , glyceraldehydes-3-phosphate dehydrogenase ( GAPDH ) [28] , cyclophilin ( CYP1 ) [26] , [27] , elongation factor1β ( EF1β ) , α-tubulin , and ubiquitin extension protein ( UEP1 ) ( Table S3 ) . One of the widely used methods for identifying stably expressed genes is to calculate the cycle threshold ( Ct ) . These seven genes showed a relatively narrow range of Ct mean values across all conditions ( Figure S1A and B ) . To evaluate the stability of gene expression , we employed the GeNorm software [33] . Under all conditions tested , these candidate genes exhibited a high degree of expression stability with relatively low M values ( less than 0 . 1 ) , which are far below the default limit of M≤0 . 15 [33] ( Figure S2A ) . For all samples , the most stable gene was β-tubulin with M value of 0 . 049 , indicating that β-tubulin can be used as a stable reference gene under multiple conditions ( Figure S2 B ) . To further validate the results obtained using GeNorm , we also employed Normfinder [34] and BestKeeper [35] , which showed almost identical patterns ( data not shown ) . We analyzed the expression patterns of 206 M . oryzae TF genes at multiple developmental stages and under various stress conditions that M . oryzae likely encounters during infection of host plants . These genes were chosen mainly based on their predicted significance and belong to 10 families , including one dominant and well-conserved family in fungi , plants , and animals ( Zinc finger proteins [36] ) , two fungal specific families ( Zn2Cys6 and APSES [37] ) , and those that are known to be involved in development ( Homeobox [12] and bHLH [38] ) , cell differentiation ( Myb [39] ) , and cell cycle ( Forkhead [25] ) ( Table 1 ) . The conditions analyzed included: ( A ) three developmental stages ( conidiation , conidial germination , and appressorium formation ) ; ( B ) two in planta infection stages at 78 hours post inoculation ( hpi ) and 150 hpi; and ( C ) 26 abiotic stress conditions ( Table S4 ) . The quality of RNA samples was evaluated using two pathogenicity genes with well- known expression patterns . The expression patterns of MPG1 [40] , a developmentally regulated gene , and DES1 [26] , which is up-regulated in the early stage of infection and under H2O2 stress , were consistent with published data ( Figure S3A and B ) . We analyzed the abundance of transcripts of 206 TF genes under 32 conditions , and fold changes relative to levels in vegetatively grown mycelia were calculated using the 2−ΔΔCt method [41] . Through a hierarchical clustering based on gene expression patterns , 185 of 206 TF genes were categorized into 4 groups with distinct expression patterns ( Figure 2A ) . Group I contained 47 genes that were up-regulated preferentially at all infection-related developmental stages and under carbon ( C ) -starvation conditions and included the previously characterized TF gene MoHOX7 , which regulates appressorium formation [12] . Genes in Group II ( 39 ) , including Mgwc-1 [21] , MoCRZ1 [9] , and Mstu1 [18] were induced preferentially by abiotic stresses . Group III contained 63 genes that were activated mainly at 78 and 150 hpi and under C-starvation and abiotic stresses caused by methyl viologen , H2O2 , MnCl2 , Congo red , FeSO4 , and uric acid . None of the TF genes in this group have been characterized . Group IV consisted of 36 genes that were up-regulated by abiotic stresses , but not during 3 developmental stages , and included COS1 [14] and MoHOX1 [12] . The number of TF genes with significantly altered expression varied widely depending on the conditions ( Figure 2B ) . Most TF genes were up-regulated ( >2-fold ) in response to treatment with methyl viologen ( 191 , 92 . 7% ) and H2O2 ( 119 , 57 . 8% ) . More than 50% of the TF genes were up-regulated during conidiation and/or in conidia ( 112 , 54 . 4% ) , host infection at 78 hpi ( 139 , 67 . 5% ) and 150 hpi ( 141 , 68 . 4% ) . In contrast , less than 20% of the TF genes were induced in response to changes in nutrient conditions ( i . e . , minimal medium , nitrogen starvation , and thiamine treatment ) and pH ( 4 and 8 ) . Under ionic stress , MnCl2 induced the expression of most genes , whereas LiCl caused the down-regulation of the 47 . 3% of the genes ( Figure 2B ) . Less than 20% of the genes were down-regulated in most of the conditions tested , except conidial germination ( 43 , 20 . 8% ) , appressorium formation ( 54 , 26 . 1% ) , LiCl ( 100 , 48 . 3% ) , and 4 min UV irradiation ( 103 , 49 . 8% ) ( Figure 2B ) . To identify TF genes that potentially control infection-related fungal development , we analyzed TF expression patterns during conidiation and/or in conidia , conidial germination , and appressorium formation . We identified 127 genes ( 61 . 7% ) that were up-regulated during at least one of these developmental stages ( Figure 3A ) . Expression of 70 genes was up-regulated at a single stage only: 57 ( conidiation and/or in conidia ) , 5 ( conidial germination ) , and 8 ( appressorium formation ) . MoHOX2 , a previously reported conidiation-specific TF gene [12] , belonged to the first group . Thirty-one genes were found to be up-regulated at all three stages , and interestingly and included MGG_00021 . 6 , a gene that is present exclusively in M . oryzae ( Table S5 ) . To colonize host plants successfully , pathogens must overcome host-generated , defense-associated compounds such as reactive oxygen species ( ROS ) [42] , [43] . To test the potential correlation between infectious growth in planta and oxidative stress responses , we compared the expression profiles under these conditions ( Figure 3B ) . During infectious growth , 139 ( 67 . 5% ) and 141 ( 68 . 4% ) genes were up-regulated at 78 hpi and 150 hpi , respectively with 117 ( 71 . 8% ) being up-regulated at both time points . Treatment with H2O2 or methyl viologen up-regulated 117 genes ( 71 . 8% ) , in which 61 . 5% of them ( 72 ) were also induced during in planta proliferation ( Figure 3B ) . To further analyze this correlation , PCA was conducted with the data from five infection-related conditions and oxidative stresses caused by H2O2 and methyl viologen . The data from 78 hpi and 150 hpi and these oxidative stress conditions were separated from those collected during conidiation and/or in conidia , conidial germination , and appressorium formation ( Figure 3C ) , further supporting a close relationship between infectious growth and oxidative stress responses . To validate the functional significance of these 72 genes during infectious growth and oxidative stress responses , we retrieved mutants in four genes , ATMT4413 ( MGG_06279 . 6 , Zn2Cys6 family ) , ATMT0047A6 ( MGG_04951 . 6 , Zn2Cys6 family ) , ATMT0662D4 ( MGG_04521 . 6 , GATA family ) , and ATMT0334A5 ( MGG_06434 . 6 , Myb family ) , from a M . oryzae T-DNA insertion mutant library [44] . Compared to wild-type strain , three of the mutants ( ATMT4413 , ATMT0047A6 , and ATMT0662D4 ) with an insertion upstream of the open reading frame ( ORF ) showed increased sensitivity to 2 . 5 mM H2O2 ( Figure 4A ) . These mutants also exhibited impaired infectious growth in rice , resulting in decreased virulence . However , one mutant ( ATMT0334A5 ) , with a T-DNA insertion at the 206 bp downstream from the stop codon of MGG_06434 . 6 , was insensitive to 2 . 5 mM H2O2 and was nearly identical with wild-type strain KJ201 in terms of infectious growth and virulence ( Figure 4A ) . Because all four mutants had a T-DNA insertion outside of ORF , we hypothesized that the phenotypes observed , except that of ATMT0334A5 , were most likely caused by reduced expression of the tagged genes . To test this hypothesis , we examined their expression using qRT-PCR . The level of transcripts from the disrupted gene in the mutants in ATMT4413 , ATMT0047A6 , and ATMT0662D4 was reduced to 60% , 20% and 50% , respectively , of the corresponding wild-type level ( Figure 4B ) . These results supported a strong correlation between expression profiles and function and suggested the involvement of largely overlapping sets of TFs in controlling pathogenicity and ROS stress responses . Two members of the fungal-specific APSES family , MoAPS1 ( MGG_09869 . 6 ) and MoAPS2 ( MGG_08463 . 6 ) ( Figure S4 ) are up-regulated specifically during conidiation and/or in conidia ( Figure 5A ) . Deletion of these genes ( Figure S4C and S4E ) caused a significant reduction in conidiation . In addition , the ΔMoaps1 and ΔMoaps2 mutants showed reduced vegetative growth ( Figure 5C ) and infectious growth in rice sheath cells ( Figure 5D ) , resulting in 50% reduction in virulence . However , conidial germination and appressorium formation were normal ( Figure 5B ) . All of the mutant phenotypes of ΔMoaps1 and ΔMoaps2 were restored by genetic complementation . To further validate the utility of predicting functional roles based on expression profiles , we studied T-DNA insertion mutants of eight additional conidiation-specific TF genes ( see Table S6 ) . All eight mutants were defective in conidiation or conidial morphology with some additional defect in conidial germination , appressorium formation or pathogenicity ( Figure S5 ) . Conidiation of four mutants , ATMT0094A6 ( MGG_06243 . 6 , Zn2Cys6 family ) , ATMT0104A6 ( MGG_02474 . 6 , C2H2 family ) , ATMT0068B3 ( MGG_01426 . 6 , Myb family ) , and ATMT 0349D2 ( MGG_02755 . 6 , GATA family ) , was significantly reduced , and one previously reported mutant , ( ATMT0651A4 ( MoHOX2 ) ) [12] , did not produce any conidia . The remaining three mutants , ATMT0052B2 ( MGG_06355 . 6 , Zn2Cys6 family ) , ATMT0591D1 ( MGG_09263 . 6 , Zn2Cys6 family ) , and ATMT0034B1 ( MGG_06507 . 6 , C2H2 family ) , produced abnormally shaped conidia . Taken together , the phenotypes of both groups of mutants strongly support the value of expression patterns of TF genes in predicting their functions . In M . oryzae , conidiogenesis is generally divided into four stages: ( A ) generation of conidiophores; ( B ) formation of a single-celled young conidium at the tip of conidiophore; ( C ) maturation of a three-celled conidium; and ( D ) multiplication of conidia in a sympodial manner [45] . To investigate expression patterns of these 57 genes at these stages , we collected samples at four different time points after induction of conidiation ( Figure S6A ) . The time point at 0 h corresponded to submerged mycelial cultures in liquid CM which inhibits conidiogenesis [45] , [46] . No conidia were observed at 6 h after induction of conidiation . Whereas , one to three-celled conidia were detected ( 5 . 3±3 . 1×104 conidia/plate ) at 12 h . After 18 h , many of typical three-celled conidia were detected ( 26 . 7±1 . 5×104 conidia/plate ) . Finally , conidia were produced abundantly ( 756 . 7±20 . 5×104 conidia/plate ) at 24 h time point ( Figure S6B ) . These observations were illustrated in Figure S6C . To test whether these samples were suitable for stage-specific gene expression profiling during conidiogenesis , we examined expression patterns of three well known conidiogenesis-related genes , COS1 [14] , CON7 [16] , and ACR1 [47] . Fold change in expression was calculated by dividing the expression level at 6 to 24 h by that at 0 h . Expression of all three genes increased during conidiation and/or in conidia . Increased COS1 transcripts were first detected at 6 h . Levels of Con7 and ACR1 transcripts increased ( ≥2 fold ) after 12 h . In particular , the amount of ACR1 transcripts at 24 h was 17 times higher than that at 0 h ( Figure S6D ) . These results are consistent with data in previous studies [14] , [16] , [47] , supporting that our samples were suitable for detailed gene expression analyses during conidiogenesis . All 57 conidiation-specific TF genes showed increased transcripts ( ≥2 fold ) at more than one stage ( Table S6 ) . Seven genes ( MGG_07319 . 6 , MGG_00139 . 6 , MGG_02447 . 6 , MGG_07681 . 6 , MGG_09263 . 6 , MGG_01833 . 6 , and MGG_06243 . 6 ) showed increased transcript levels at all four time points compared with that at 0 h , while 21 genes increased transcripts at only one of the time points ( one gene at 6 h , one at 12 h , 10 at 18 h , and nine at 24 h ) . The rest of the genes had increased transcripts at two to three time points ( three at 6 h , 12 at 18 h , 14 at 12 h , 18 h , and 24 h , nine at 12 and 18 h , one at 18 h and 24 h , and two at 12 h and 18 h ) . This data clearly showed differential expression of all 57 conidiation-specific TF genes conidiogenesis , suggesting their involvement in this process . To investigate the regulatory network controlling the expression and interactions of these 57 genes during conidiation and/or in conidia , we examined their expression in six TF gene deletion mutants . These mutants showed conidiation-related phenotypes such as no conidial production ( ΔMohox2 [12] ) , smaller conidia ( ΔMohox4 [12] ) , and reduced conidial production ( ΔMoaps1 ( this study ) , ΔMoaps2 ( this study ) , ΔMoleu3 [48] , and ΔMonit4 [48] ) . We compared gene expression profiles of these 57 genes in the six mutants with those in KJ201 to determine if and how their gene expression was affected by each mutation ( Figure 6 ) . Sixteen genes ( Figure 6 ) were not affected by any of the mutations . Among the remaining 41 genes , TF116 ( MGG_02474 . 6 , C2H2 family ) and TF192 ( MGG_03711 . 6 , Zn2Cys6 ) were down-regulated in all mutants , suggesting that their expression requires the mutated genes , whereas three genes , including TF035 ( MGG_07319 . 6 , GATA type ) , TF220 ( MGG_06243 . 6 , Zn2Cys6 ) , and TF269 ( MGG_09829 . 6 , Zn2Cys6 ) , were up-regulated in all mutants . Expression of several genes were up- or down-regulated only in one mutant: TF094 ( MGG_00373 . 6 , C2H2 ) and TF150 ( MGG_06507 . 6 , C2H2 ) in ΔMohox2; TF206 ( MGG_04951 . 6 , Zn2Cys6 ) , TF260 ( MGG_09263 . 6 , Zn2Cys6 ) , TF231 ( MGG_07131 . 6 , Zn2Cys6 ) in Δ Mohox4; TF241 ( MGG_07681 . 6 , Zn2Cys6 ) , TF246 ( MGG_08094 . 6 , Zn2Cys6 ) , and TF268 ( MGG_09825 . 6 , Zn2Cys6 ) in ΔMoaps1 ;TF271 ( MGG_09950 . 6 , Zn2Cys6 ) , MoFOK1 , MoHOX3 in ΔMoAPS2; TF263 ( MGG_09312 . 6 , Zn2Cys6 ) , TF117 ( MGG_02505 . 6 , C2H2 ) , and MoHOX8 in ΔMonit4 . In addition , expression of TF134 ( MGG_02845 . 6 , C2H2 ) , TF008 ( MGG_10837 . 6 , bHLH ) , and TF276 ( MGG_10528 . 6 , Zn2Cys6 ) seems to require both MoHOX2 and MoHOX4 , while MoHOX1 requires only MoAPS2 and is down-regulated in ΔMoaps1 , ΔMoleu3 and ΔMonit4 . Based on the results shown in Figure 6 , we developed a model for the regulatory network controlling the expression of conidiation-specific TF genes ( Figure 7 ) .
Advances in tools for analyzing global gene expression profiles have facilitated the identification of genes potentially associated with specific processes and the characterization of regulatory networks controlling their expression . To test whether expression patterns of TF genes under diverse conditions help predict the functional roles of individual genes and potential regulatory interactions among them , we analyzed expression of 206 M . oryzae TF genes under 32 conditions using qRT-PCR . Expression profiles and functional validation of several genes selected based on their expression patterns clearly demonstrate the value of TF gene expression patterns in predicting their function . This comprehensive expression data of TF genes , publicly available through FTFD , will serve as a new community resource in analyzing the functions of and potential interactions among individual TF genes . Previous studies based on microarrays [49] , [50] , SAGE [51] , or RNA-seq [52] revealed many genes that potentially play important roles under specific conditions in M . oryzae . However , despite the biological significance of TF genes , relatively few have been characterized in M . oryzae and their regulation and genetic interactions have not been systematically investigated . In this study , we adopted qRT-PCR to address this deficiency . This method is labor intensive but has been shown to be robust in accurately quantifying TF transcripts [4] . We have identified differentially expressed TF genes under 32 conditions with most of them being up-regulated under at least one of these conditions ( Figure 2 ) . Conidiation in plant pathogenic fungi , including M . oryzae , plays a central role in their life and disease cycles and epidemics . However , little is known about the molecular changes underpinning conidiation in M . oryzae . The developmental complexity of conidiation was suggested by the fact that 8 . 5% of the protein-coding genes in M . oryzae are differentially expressed during conidiation and/or in conidia based on a whole-genome microarray experiment [46] . Approximately 25% of the predicted genes are differentially expressed during conidiation in Neurospora crassa [53] and that ∼1 , 000 genes in Aspergillus nidulans are involved in conidiation [54] . Thus , it is likely that a relatively large numbers of TF genes are involved in controlling and coordinating the expression of many genes that participate in producing conidia . Our analysis revealed that more TF genes were up-regulated during conidiation and/or in conidia ( 112 genes ) than during conidial germination ( 51 genes ) and appressorium formation ( 52 genes ) . However , most of the genes induced during conidial germination and appressorium formation were also induced during conidiation and/or in conidia , suggesting that the same general transcription regulators probably control multiple developmental changes . In total , 57 genes were considered conidiation-specific . These 57 genes were differentially expressed at one or more stages of conidiation , including conidiophore formation , conidia formation , and multiplication of conidia in a sympodial manner ( Figure S6 ) . The importance of many of these genes ( 41 out of 57 ) in conidiation was implied by their modified expression in one or more mutants that are defective in conidiation . Compared with the patterns observed in the wild-type strain KJ201 , three genes were up-regulated while two genes were down-regulated in all the mutants during conidia production and/or in conidia . We hypothesize that these TFs act as major regulators of transcription throughout conidiation . These genes are interesting candidates for functional studies via mutagenesis . Results from this gene expression analysis in the multiple mutant backgrounds led to a model for a regulatory network controlling the expression of conidiation-specific TF ( Figure 7 ) . This model will serve as a useful roadmap in studying the regulation of conidiation . Interestingly , most of the TF genes induced by oxidative stresses were also induced during in planta growth ( 72 genes , Figure 3B ) ; this finding is consistent with the accumulating evidence suggesting that fungal pathogens must overcome plant-generated ROS for successful invasion [20] , [26] , [42] , [55] . Our results also indicate that in vitro oxidative stress conditions mimic those that the fungus encounters in planta , and that in planta invasion and in vitro oxidative stress responses share common transcriptional regulatory factors . Nitrogen starvation is known to be one of the important environmental cues for appressorium formation and in planta growth of M . oryzae [50] . Donofrio et al [50] reported that one GATA family TF gene , NUT1 ( MGG_06050 . 6 ) , was highly up-regulated in both nitrogen starvation condition and inside infected rice , suggesting NUT1 is a global nitrogen regulator . We also found that 13 TF genes were up-regulated in response to nitrogen starvation as well as during host infection ( data not shown ) . Moreover , one of the M . oryzae specific TF gene ( MGG_00021 . 6 , Zn2Cys6 ) and one Myb family TF gene ( MGG_06898 . 6 ) showed up-regulation at all three developmental stages , two infection stages , and nitrogen starvation , suggesting that these TF genes function as general regulators controlling multiple processes in M . oryzae . One of the most important outcomes of this study is demonstrating the value of expression data in predicting the putative function of individual TF genes . Those TF genes induced during conidiation and/or in conidia were used to test their value . MoHOX2 , which plays a critical role in conidial production [12] , was identified as a conidiation-specific TF gene . Further , T-DNA insertional mutants in seven of these genes were defective in conidiogenesis . Targeted mutagenesis of two fungal-specific TF genes of the APSES family , which are up-regulated during conidiation and/or in conidia , also caused defects in conidiation . In a second test involving four mutants in the TF genes induced both during infection and under oxidative stress also showed that the mutants displayed increased sensitivity to oxidative stress and severely reduced infectious growth in rice ( Figure 4A ) . Results from both tests strongly supported the predictive value of expression patterns in functional studies . Considering that similar TF expression profiles were observed between in planta infectious growth and oxidative stress , a high throughput in vitro assay system that screens for mutants defective in growth under oxidative stress can serve as a surrogate platform for quickly identifying candidate pathogenicity genes . Metal ions , such as MnCl2 and FeSO4 , induced expression of many TF genes . The effect of metal ions in fungal biology and pathogenicity is not clearly understood . However , a recent study suggested that ferrous ion is required for the normal function of the DES1 gene in M . oryzae [26] . In mammals , manganese ion induces apoptosis by causing endoplasmic reticulum stress and mitochondrial dysfunction [56] , [57] . Comprehensive expression profiles of TF genes in the presence of metal ions or other abiotic stresses will help decipher not only how fungal responses to such stresses are controlled at the transcriptional level , but also their roles in fungal biology and pathogenicity . Functional characterization of fungal genes requires a well-standardized platform that assays diverse phenotypes . However , only a few phenotypes , such as mycelial growth , reproduction , and pathogenicity , have been evaluated in gene functional studies with filamentous fungi [44] , [58] , [59] . When mutants of N . crassa in 103 TF genes were evaluated , only less than half of the resulting mutants exhibited clear phenotypes [59] , which can be attributed to overlapped functions among TFs , limited phenotype assays , or a combination of both . In clusion of 26 abiotic stress conditions to profile expression patterns has helped the establishment of a novel phenomics platform for large-scale gene functional studies in M . oryzae and other pathogenic fungi . This platform will help systematically decipher the functional roles of TF genes in fungal development , pathogenicity , and abiotic stress management .
Annotated genomes of 21 fungal and 2 Oomycete species ( Table S1 ) were used to compare of the number and types of TF genes . Putative TF genes in version 6 of the M . oryzae genome ( http://www . broadinstitute . org/annotation/fungi/magnaporthe ) were identified using the annotation pipeline in FTFD which annotates fungal TFs based on the InterPro database using DNA binding motifs [22] . To identify M . oryzae specific TF genes ( orphan genes ) , a combination of BLAST matrix [60] and InParanoid algorism [23] was used . We applied a cutoff e-value of less than 10−50 for protein similarity for BLAST matrix searches and the default parameter for InParanoid . M . oryzae KJ201 ( wild-type strain ) and all mutants used in this study were obtained from the Center for Fungal Genetic Resource ( CFGR ) at Seoul National University , Seoul , Korea . All strains were grown at 25°C for 14 days on oatmeal agar . Conidia and germinated conidia were harvested as described previously [61] , and appressoria were collected 6 h after dropping a conidial suspension ( 5×104 conidia/ml ) on a hydrophobic surface . For infected plant samples , after inoculating rice seedlings ( 3–4 leaf stage ) with 20 ml of a KJ201 conidial suspension ( 1×105 conidia/ml ) , leaves were collected at 78 hpi and 150 hpi . Prior to exposing fungal cultures to various types of stress , cultures of 100 ml liquid CM ( complete medium ) inoculated with 1 ml of a conidial suspension ( 5×104 conidia/ml ) were incubated at 25°C for 4 days in an orbital shaker ( 120 rpm ) . The resulting mycelia were harvested using a 0 . 45-µm filter , washed with sterilized distilled water , transferred to fresh liquid CM and minimal medium ( MM ) [62] as a control , and CM or MM containing each treatment ( Table S4 ) for 4 h culture . All mycelial samples were harvested from three replicates of three biological repeats , immediately frozen using liquid nitrogen , and stored at −80°C until processed . For harvesting samples at different time points during conidiogenesis , a previously described procedure [46] was slightly modified . Actively growing wild-type mycelia were inoculated into liquid CM , and incubated at 25°C on a 120 rpm orbital shaker for 4 days . The resulting mycelia were fragmented using spatula and pressed through two-layers of cheese cloth . The mycelia were collected using two-layers of miracloth ( Calbiochem , California , USA ) and washed three times with one liter of sterilized distilled water . After resuspending the harvested mycelia in 10 ml sterilized distilled water , 400 µl of the suspension was spread on each 0 . 45 µm pore cellulose nitrate membrane filter ( Whatman , Maidstone , England ) placed on V8-Juice agar plate . The plates were incubated at 25°C with constant light . The whole tissue on the membrane filters was collected at 0 h , 6 h , 12 h , 18 h , and 24 h after inoculation by disposable scraper ( iNtRON Biotechnology , Seoul , Korea ) . All samples were harvested from three replicates of three biological repeats , immediately frozen using liquid nitrogen , and stored at −80°C until processing . Total RNA was extracted using an Easy-Spin Total RNA Extraction Kit ( iNtRON Biotechnology , Seoul , Korea ) , and 5 µg of RNA was reverse-transcribed to cDNA using the Prom-II Reverse Transcription System ( Promega , Madison , WI , USA ) according to the manufacturer's instructions . The resulting cDNA preparations were diluted to 12 . 5 ng/µl and kept at −20°C . A total of 206 primer pairs were designed using the 3′-end exon region of the target genes ( GC contents = 45–55% and Tm = 60 ) ( Table S7 ) . qRT-PCR reactions were performed using a MicroAmp Optical 96-Well Reaction Plate ( PE Biosystems , Foster City , CA , USA ) and an Applied Biosystems 7500 Real-Time PCR System . Each well contained 5 µl of Power 2× SYBR Green PCR Master Mix ( Applied Biosystems , Warrington , UK ) , 2 µl of cDNA ( 12 . 5 ng/µl ) , and 15 pmol of each primer . The thermal cycling conditions were 10 min at 94°C followed by 40 cycles of 15 s at 94°C and 1 min at 60°C . All amplification curves were analyzed with a normalized reporter threshold of 0 . 1 to obtain the threshold cycle ( Ct ) values . To identify an appropriate reference gene for normalizing the expression levels of individual TF genes , GeNorm v . 3 . 4 [33] , Normfinder [34] and BestKeeper [35] were used . Expression levels of the chosen reference gene , β-tubulin , were measured in more than two replicates for each PCR run , and their average Ct value was used for relative expression analyses . To compare the relative abundance of target gene transcripts , the average Ct value was normalized to that of ß-tubulin for each of the samples as 2−ΔCt , where −ΔCt = ( Ct of the target gene – Ct of ß-tubulin ) . Fold changes of transcripts in samples representing developmental stages and infectious growth relative to those in mycelial samples in liquid CM were calculated as 2−ΔΔCt , where −ΔΔCt = ( Ct of the target gene –Ct of ß-tubulin ) test condition - ( Ct of the target gene – Ct of ß-tubulin ) CM [41] . qRT-PCR was conducted twice with three replicates , and all data are presented . The fold changes of transcripts from various stress-exposed mycelial samples compared to those in untreated samples ( CM or MM ) were calculated as 2−ΔΔCt , where −ΔΔCt = ( Ct of the target gene –Ct of ß-tubulin ) treated condition - ( Ct of the target gene – Ct of ß-tubulin ) untreated condition . Pearson's correlation coefficient and Spearman's rank were used to measure the similarity between gene expression profiles and the similarity between samples , respectively . A heat map of the clustered genes and samples was generated by complete linkage . A principle component analysis ( PCA ) was conducted to reduce the dimensions and to understand the relationships between the TF genes and the experimental conditions . PCA was performed using SPSS software v . 12 . 0 ( SPSS Inc . , Chicago , IL , USA ) . To build a model for the regulatory network controlling the expression of conidiation-specific TF genes based on their expression patterns in six TF gene deletion mutants , we used NodeXL ( http://nodexl . codeplex . com ) . Assays for measuring the sensitivity to exogenous oxidative stress were performed on CM agar amended with 2 . 5–5 mM H2O2 or methyl viologen . Radial colony growth was measured on day 6 after inoculation . Infection assays with rice sheath and 3-week-old rice seedlings were conducted as described previously [63] . Gene disruption ( Fig . S4B and D ) and fungal transformation were conducted as described previously [61] . Putative mutants were confirmed by Southern blot analysis . Vegetative growth , pigmentation , conidiation , conidial size , conidial germination , appressorium formation , and infection assays on onion epidermis , rice sheath cells , and rice seedlings were conducted as described previously [12] , [63] .
|
Rice blast disease , caused by Magnaporthe oryzae , destroys rice crop enough to feed 60 million people every year and has served as a model pathosystem for understanding host-parasite interactions . However , little is known about how M . oryzae globally regulates and coordinates its gene expression at the whole-genome scale . We analyzed the expression patterns of 206 M . oryzae genes encoding transcription factors ( TFs ) under 32 conditions , including infection-related developmental stages and various abiotic stresses , using quantitative real-time PCR . We focused on identifying the TF genes that are induced during the two most important infection-related morphogenetic changes; conidiation and infectious growth in rice . We identified 57 conidiation-specific TF genes and functionally characterized ten of them . Our data also showed that infectious growth in planta and oxidative stress responses in vitro involve largely overlapping groups of TFs . Comprehensive TF expression data and functional validation provided new insights into the regulatory mechanism underpinning pathogenicity and stress responses in M . oryzae . These data will also serve as a guide in studying the role of individual TF genes and the coordination of their expression in controlling development , pathogenicity , and abiotic stress responses in M . oryzae .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biology",
"agriculture"
] |
2013
|
Global Expression Profiling of Transcription Factor Genes Provides New Insights into Pathogenicity and Stress Responses in the Rice Blast Fungus
|
In most eutherian mammals , sex chromosomes synapse and recombine during male meiosis in a small region called pseudoautosomal region . However in some species sex chromosomes do not synapse , and how these chromosomes manage to ensure their proper segregation is under discussion . Here we present a study of the meiotic structure and behavior of sex chromosomes in one of these species , the Mongolian gerbil ( Meriones unguiculatus ) . We have analyzed the location of synaptonemal complex ( SC ) proteins SYCP1 and SYCP3 , as well as three proteins involved in the process of meiotic recombination ( RAD51 , MLH1 , and γ-H2AX ) . Our results show that although X and Y chromosomes are associated at pachytene and form a sex body , their axial elements ( AEs ) do not contact , and they never assemble a SC central element . Furthermore , MLH1 is not detected on the AEs of the sex chromosomes , indicating the absence of reciprocal recombination . At diplotene the organization of sex chromosomes changes strikingly , their AEs associate end to end , and SYCP3 forms an intricate network that occupies the Y chromosome and the distal region of the X chromosome long arm . Both the association of sex chromosomes and the SYCP3 structure are maintained until metaphase I . In anaphase I sex chromosomes migrate to opposite poles , but SYCP3 filaments connecting both chromosomes are observed . Hence , one can assume that SYCP3 modifications detected from diplotene onwards are correlated with the maintenance of sex chromosome association . These results demonstrate that some components of the SC may participate in the segregation of achiasmate sex chromosomes in eutherian mammals .
The proper distribution of chromosomes into daughter cells during meiosis depends on the essential phenomena of pairing , synapsis , recombination , and segregation . During early prophase I homologous chromosomes associate in pairs and are held by a proteinaceous structure , the synaptonemal complex ( SC ) [1–4] . Moreover , homologous chromosomes undergo a process of reciprocal recombination whose cytological manifestation is chiasmata . Recombination between homologues along with the existence of mechanisms that maintain sister chromatid cohesion are responsible for ensuring the proper segregation of homologous chromosomes during first meiotic division [5 , 6] . It is currently known that these phenomena are intimately related and that they occur in an ordered fashion . Thus , homologous recognition , pairing , and synapsis are promoted by the initiation of recombination events involved in the repair of programmed DNA double strand breaks ( DSBs ) made by SPO11 protein at the very beginning of first meiotic prophase [7–9] . Furthermore , the assembly of the SC is necessary for the correct completion of recombination and the formation of crossovers [3 , 10 , 11] . On the other hand , at least one chiasma per bivalent is necessary to ensure that homologues remain associated from the disorganization of the SC at diplotene until they segregate at anaphase I . Although this plan is followed by a great majority of species , there are some groups of organisms that show variations in the sequence or even the occurrence of the meiotic hallmarks ( for review see [12] ) . Thus , synapsis precedes recombination initiation in flies and nematodes [13 , 14]; SC is not formed in dipteran males and fission yeast [15 , 16]; recombination does not occur in Drosophila males and lepidopteran females [17 , 18]; and in most hemipterans sex chromosome segregation is postponed to the second meiotic division [19 , 20] . Sex chromosomes are especially prone to get out of the rules of meiosis [21] . In most mammals , sex chromosomes only share a small region of homology named pseudoautosomal region ( PAR ) [22 , 23] , to which synapsis and recombination are restricted . The occurrence of recombination in the PAR allows sex chromosomes to remain associated until they segregate at anaphase I . However , there are some mammalian species in which the X and Y chromosomes do not form SC . This situation is especially well characterized in marsupials [24–28] , in which we have recently reported that a particular structure formed by SC proteins , called dense plate , is involved in maintaining the association of the X and Y chromosomes from pachytene until they segregate at anaphase I [29] . The lack of synapsis has also been reported in some species of eutherian mammals , especially among gerbils and voles [30–34] . In these species sex chromosomes do not form SC , but they are associated during first meiotic prophase and segregate properly during first meiotic division . It has been proposed that in the absence of synapsis , the association of sex chromosomes could be maintained by telomeric or distal heterochromatic associations [30 , 33 , 34] . Nevertheless , the nature of the mechanisms that promote sex chromosome pairing and segregation in these species remains unclear . To shed light on these mechanisms , we have investigated the sequence and the nature of X and Y chromosome association during male meiosis in the Mongolian gerbil ( M . unguiculatus ) , an eutherian mammal that presents asynaptic sex chromosomes [31] . For this purpose we have analyzed the location of SYCP1 and SYCP3 proteins of the SC [35–37] , as well as RAD51 and MLH1 proteins , which are involved in meiotic recombination [38 , 39] , and γ-H2AX , a histone variant related to both DSBs′ repair and meiotic sex chromosome inactivation [40 , 41] . Our results show that even though sex chromosomes in M . unguiculatus neither synapse nor recombine , they pair and remain associated until anaphase I . We have observed structural modifications in their axial elements ( AEs ) that involve SYCP3 protein , which could be responsible for maintaining sex chromosome association . Since similar results have been reported in marsupials [29] , one can assume that the SC plays a crucial and ancient role in the segregation of achiasmate chromosomes .
We first studied the location of SYCP3 protein , the main component of the AE and lateral elements ( LEs ) of the SC [35 , 36] , on squashed spermatocytes ( Figure 1 ) . At leptotene , the signal of SYCP3 is detected as short filaments dispersed in the nucleus ( Figure 1A ) . During zygotene these filaments , corresponding to the AEs , begin to associate in pairs to form thicker filaments ( Figure 1B ) . The typical ”bouquet” arrangement of telomeres is only seen at early zygotene ( Video S1 ) , and it usually does not include all the telomeric ends . At pachytene autosomes are associated all along their length ( Figure 1C; Video S2 ) . The trajectories of their LEs are clearly discerned , and several twists along each bivalent are detected ( Figure 1C , inset ) . During diplotene , LEs separate ( Figure 1D; Video S3 ) , and the SYCP3 signal on the desynapsed LEs becomes thinner at the end of this stage ( Figure 1E ) . At diakinesis SYCP3 is still associated to chromosomes as a discontinuous array of speckles that occupy the region between sister chromatids ( Figure 1F ) . SYCP3 also forms aggregates and irregular bars in the cytoplasm from this stage until the end of first meiotic division . Sex chromosomal AEs are not distinguishable from that of the autosomes during leptotene ( Figure 1A ) or zygotene ( Figure 1B ) . The location and morphology of sex chromosomal AEs become evident just at pachytene . At this stage , sex chromosomes are located at the nuclear periphery and occupy a particular domain—the sex body , which presents a higher degree of chromatin condensation compared to the autosomes ( unpublished data ) . The AEs of both X and Y chromosomes are distinguishable one adjacent to the other and inside the sex body . However , they are not in contact , either laterally or distally ( Figure 1C and 1C′; Video S4 ) , and they do not show any kind of modifications like thickenings or excrescences , as it is usually found in other mammals [23] . The position of the centromeres along sex chromosomal AEs reveals that the X chromosome is submetacentric and the Y chromosome is metacentric . During diplotene sex chromosomes remain associated and located at the nuclear periphery . However , as sex chromosomes increase their condensation their AEs fold ( Figure 1D and 1D′ ) . At late diplotene , sex chromosome axes become tangled , and the SYCP3 signal shows an intricate morphology making it difficult to discern each sex chromosome inside the sex body ( Figure 1E and 1E′ ) . At diakinesis , X and Y chromosomes are distinguishable ( Figure 1F and 1F′ ) , and SYCP3 labeling spreads throughout the Y chromosome , while it forms an irregular line running all along the X chromosome . Moreover , sex chromosomes are in contact by an end-to-end association ( Figure 1F′ ) . This conformation is maintained until metaphase I . To test the asynaptic nature of the sex chromosome association in M . unguiculatus , we carried out the double immunolocalization of SYCP3 and SC central element protein SYCP1 [35 , 37] on spermatocyte spreads ( Figure 2 ) . SYCP1 is not detected at leptotene ( Figure 2A ) , while at zygotene short stretches of signals appear between the AEs of homologous chromosomes located either at distal or interstitial regions ( Figure 2B ) . Interestingly , SYCP1 association to the chromosomes starts before all AEs are completely formed . At pachytene , synapsis is completed in the autosomal bivalents , and the signals of SYCP1 and SYCP3 are coincident along the bivalents ( Figure 2C ) . At diplotene , SYCP1 dissociates from the bivalents and the LEs begin to separate ( Figure 2D and 2E ) . Contrary to what was observed in squashes , sex chromosome AEs can be identified on spreads during zygotene ( Figure 2B and 2B′ ) . This is most probably due to the higher degree of chromatin dispersion produced with this technique . At this stage sex chromosomal AEs are completely formed , and their thickness is similar to that of the autosomal AEs . The AEs of X and Y chromosomes can be found either closely located or separated in the nucleus at mid zygotene , and even at late zygotene sex chromosomes still remain separated in 41 . 7% of the cells ( n = 60 ) ( Figure S1 ) . Nevertheless , from early pachytene onwards they are always closely related ( Figure 2C ) . Two features indicate that the assembly of the SC central element is not involved in this association: first , sex chromosomal AEs do not show any physical contact; and second , SYCP1 protein is completely absent from the X and Y chromosomes ( Figure 2C′ ) . However , the ends of sex chromosome AEs are distally connected at the beginning of diplotene ( Figure 2D and 2D′ ) . This contact may involve any of the ends of each sex chromosome with the other sex chromosome or even the two tips of each chromosome , as shown for sex chromosomes of other species of Gerbillidae [30 , 31 , 33] . Nevertheless , at late diplotene the four chromosome ends are always connected ( Figure 2E and 2E′ ) . Noticeably , no SYCP1 signal is found in these regions of distal association . As observed in squashed preparations , the morphology of sex chromosomal AEs is modified from diplotene onwards when studied on spreads . Thus , AEs become irregular and folded at diplotene ( Figure 2D and 2E′ ) , and at diakinesis SYCP3 expands to form a massive and intense signal that seems to cover the whole Y chromosome ( Figure 2F and 2F′ ) . In mammals the initiation of SC assembly is dependent on the occurrence of previous recombination events [9 , 41] . Therefore , we wondered whether the absence of synapsis between sex chromosomes could be due to the absence of such recombination events in them . We analyzed the location of RAD51 , a protein related to early repair of DSBs , and MLH1 protein , which is related to the last steps of recombination leading to the formation of crossovers ( Figure 3 ) [42 , 43] . RAD51 is detected on the autosomal AEs during zygotene and early pachytene as dots on or very close to the AEs/LEs ( Figure 3A ) . The number of RAD51 dots decreases at mid pachytene ( Figure 3B ) , and it is absent from the autosomes at late pachytene ( Figure 3C ) . Sex chromosomes , particularly the X chromosome , also exhibit RAD51 , as several dots on the AEs ( Figure 3A and 3B′ ) . At mid pachytene ( Figure 3B and 3B′ ) , the number of dots on the AEs decreases and is undetectable at late pachytene ( Figure 3C and 3C′ ) . It is interesting to note the persistence of many RAD51 foci on the sex chromosomes at mid pachytene , while most of RAD51 foci have disappeared from autosomes ( Figure 3B ) . Additionally , the detection of a faint RAD51 labeling on the sex chromatin at early-mid pachytene is also intriguing ( Figure 3B′ ) . This labeling persists and becomes even more intense in some late pachytene spermatocytes ( Figure 3C′ ) . MLH1 is only detected at late pachytene on autosomal SCs . Most bivalents present one MLH1 focus , but some of them may present two foci ( Figure 3D ) . However , MLH1 is not detected on the sex chromosomes ( Figure 3D and 3D′ ) . Given the striking modification of SYCP3 location during late stages of prophase I , we analyzed SYCP3 distribution during late stages of first meiotic division to ascertain its potential role in sex chromosome segregation ( Figure 4 ) . At metaphase I , SYCP3 protein remains associated with autosomes at the region of sister chromatid contact ( the interchromatid domain ) ( Figure 4A and 4B ) . This pattern of localization is similar to that described in mouse and other mammals and can be related to a role for SYCP3 in maintaining sister chromatid cohesion [44 , 45] . Nevertheless , this protein does not accumulate in the centromeric regions , as occurs in mouse [44] . The pattern of SYCP3 localization on the X chromosome is visualized as a sinuous and irregular line that runs along its interchromatid domain ( Figure 4A and 4A′′ ) . In contrast , SYCP3 signal on the Y chromosome is not restricted to the interchromatid domain , but occupies almost the entire width of the chromatin , excepting the pericentromeric region , in which the protein is present only as a thin filament ( Figures 4A′ , 4C′ , 4D′ , and S2 ) . As described above , this pattern of SYCP3 distribution appears during diakinesis and is subsequently maintained in later stages . However , once sex chromosomes are pulled to the spindle poles at metaphase I , it becomes evident that the extensive labeling of SYCP3 involves not only the Y chromosome but also the most distal segment of the X chromosome long arm ( Figures 4A′ , 4C′ , 4D′ , and S2 ) . At metaphase I , sex chromosomes are associated and properly bioriented . However , we observed two different configurations . In the first configuration , both arms of the X chromosome are in contact with the Y chromosome ( Figure 4A ) , and there is a clear continuity between SYCP3 signals on the X and Y chromosomes . Interestingly , SYCP3 signal may overpass the length of the short arm of the X chromosome and contact with the massive SYCP3 labeling that covers the distal region of the X long arm and the Y chromosome ( Figure 4A′ and A′′′ ) . In the second configuration , the bridge of SYCP3 signal breaks in the short arm of the X chromosome , but remains intact between the end of the X long arm and the Y chromosome ( Figure 4C and 4C′′; Video S5 ) . Therefore , the association of both chromosomes seems to be maintained by SYCP3 and not just by a direct contact of the chromatin of the X and Y chromosomes . At the beginning of anaphase I , SYCP3 dissociates from the chromosomes but does not disappear abruptly since it is still detectable during early anaphase I at the interchromatid domains , mainly close to the centromeres ( Figure 4D ) . The SYCP3 bridge between X and Y chromosomes also persists in anaphase I . As observed in metaphase I , at the beginning of anaphase I SYCP3 filaments may appear either connecting both X and Y chromosomal ends ( Figure 4D and 4D′′; Video S6 ) or just the long arm of the X chromosome with either one or both of the Y chromosome arms ( unpublished data ) . Later in anaphase I , SYCP3 disappears from the interchromatid domain of autosomes and is only detectable in some pericentromeric regions ( Figure 4E and 4E′′ ) . SYCP3 is also visible as filaments located inside and outside the spindle area that are not in contact with the chromosomes . Nevertheless , one of these filaments is usually detected connecting the chromosomes of opposite anaphase I poles ( Figure 4F and 4F′′ ) . SYCP3 is still detected at telophase I as filaments present near the centromeric regions of chromosomes and as longer filaments dispersed over the protoplasm ( Figure 4G ) . Some of these filaments are visible during interkinesis ( Figure 4H ) , while no SYCP3 labeling is detectable during the second meiotic division ( unpublished data ) . Taking into account the pattern of SYCP3 distribution on the sex chromosomes up to metaphase I , it is likely that some of the SYCP3 filaments found during anaphase I associate the sex chromosomes . With the aim of identifying the X and Y chromosomes inside these chromatin masses and their relation to SYCP3 filaments , we carried out the double immunolabeling of SYCP3 and γ-H2AX ( Figure 5 ) , a phosphorylated form of histone variant H2AX . In mouse , γ-H2AX decorates the entire nucleus during leptotene and early zygotene in response to DSBs and is thereafter restricted to the chromatin of the sex body from late zygotene until diplotene , where it is related to the process of meiotic sex chromosome inactivation [40 , 41] . We found that in M . unguiculatus γ-H2AX occupies the whole nucleus at leptotene ( Figure 5A ) and zygotene ( Figure 5B ) . From pachytene onwards γ-H2AX is almost exclusively located on the sex chromosomes . However , contrary to what occurs in mouse , it does not disappear at diplotene but remains detectable until telophase I ( Figure 5C–5I ) . Thus , the signal of the γ-H2AX allowed us to unequivocally identify the sex chromosomes during the late stages of the first meiotic division . The simultaneous labeling of SYCP3 and γ-H2AX corroborates that SYCP3 occupies almost the entire width of the Y chromosome during diakinesis up to metaphase I ( Figure 5E–5G′; Video S7 ) . At the beginning of anaphase I , the massive SYCP3 signals found on the Y chromosome and the distal region of the X chromosome disorganize and the sex chromosomes migrate to opposite poles ( Figure 5H ) . Moreover , the SYCP3 filament detected during anaphase I actually bridges the sex chromosomes ( Figure 5H and 5H′′′; Video S8 ) . Additionally , we observed that the sex chromosomes are always lagged during anaphase I migration . However , this feature seems not to be due to a chromatin association of the sex chromosomes since no chromatin bridges are detected either with DAPI ( Figure 5H′ and 5I′ ) or γ-H2AX staining ( Figure 5H′′′ and 5I′′′ ) .
Our analysis of the sequence of SC assembly in the Mongolian gerbil revealed that both X and Y chromosome assemble an AE , but they do not synapse . A first explanation for this behavior is that the mechanisms that promote chromosome synapsis in mammals , i . e . , occurrence and repair of DNA DSBs [9 , 41] , do not take place on the sex chromosomes of M . unguiculatus . It has been demonstrated that the disruption of DNA DSB occurrence and/or repair severely impairs synapsis in mouse [9 , 46 , 47] . Moreover , it has been reported that in the grasshopper Stethophyma grossum large portions of the autosomes remain unsynapsed during first meiotic prophase due to the absence of DNA DSBs [48] . However , our results on the location of RAD51 in the sex chromosomes of M . unguiculatus suggest that asynapsis is not due to the absence DSB recombinational repair . A second explanation is that sex chromosomes in the Mongolian gerbil do not share a region of homology . Thus , although sex chromosomes can initiate the processes that ultimately culminate in the synapsis with the homologous chromosome , they are unable to complete this process because they have no homologous partner . In this sense , the absence of synapsis between sex chromosomes appears to be a recurrent feature among the species of the family Gerbillidae . This is the case of Psamommys obessus [30 , 33] , Gerbillus campestris , M . libycus , M . shawi , and M . crassus [32] . On the other hand , some species present sex chromosomes with synapsis and recombination , as in G . chiesmani , G . nigeriae , G . hoogstrali , and Taterillus pygargus [31] . However , the synapsing regions in these species seem to be originated , as in many other eutherian mammals , by recent translocations of autosomal segments to both the X and Y chromosomes [49 , 50] . Previous analyses on the sex chromosomal phylogeny of Gerbillidae have shown that the X chromosome of M . unguiculatus could be one of the most primitive among this family [49] , reinforcing the idea that the asynaptic condition of sex chromosomes would be an ancient feature of this group . Therefore , the absence of PAR is to us the most plausible explanation for the absence of synapsis between the X and the Y chromosomes . However , it has been reported that in some species , the marsupial Macropus eugenii for instance , sex chromosomes do not synapse even though they share a region of homology [51] . In the same way , our current knowledge of the human X and Y chromosomes reveals that they still share many segments with different degrees of homology that lay out of the regions usually involved in the formation of SC [52 , 53] . Therefore , in the absence of direct DNA sequence comparison it is not possible to rule out the possibility that some homology is still shared between sex chromosomes in M . unguiculatus . Nonetheless , these homologous regions could be degenerated or reorganized in such a way that they would not be able to promote synapsis any longer , i . e . , they would lack a sort of ”functional homology” . Our data indicate that pairing of sex chromosomes takes place during zygotene , and they remain associated at pachytene . However , the lack of the PAR between sex chromosomes in M . unguiculatus poses an interesting question about the mechanisms that could be involved in bringing and maintaining them together during the first meiotic prophase . As regards the first topic , one could assume that the polarization of telomeres during the bouquet stage plays an important role in the initial approach of sex chromosomes . Nevertheless , this mechanism would not be sufficient to ensure sex chromosome pairing , since they can appear close together at the very beginning of zygotene , before autosomal AEs are completely formed , and on the contrary , they can remain separated in the nucleus until late zygotene , well after the resolution of the bouquet . Another possibility , although highly speculative , is that sex chromosome pairing is based on a mechanism of homologous sequence recognition . As mentioned above , the absence of a functional PAR does not imply that there is not homology at all between sex chromosomes . Provided that a certain degree of homology could be conserved , it is possible that the mechanisms of DNA repair mediated by RAD51 and other proteins could promote the approaching and recognition of X and Y chromosomes , although , as stressed before , structural or genetic factors would hamper the formation of a SC . In this sense , this residual homology could not be as efficient as a PAR in promoting the recognition of sex chromosomes , explaining their erratic behavior during zygotene . Once sex chromosomes recognize each other they remain intimately paired throughout pachytene , even though SC is not formed . In other Gerbillidae species , sex chromosomes present some kind of distal connections between the ends of their AEs at pachytene [30 , 32 , 33] . These associations may be autologous or heterologous , and it has been claimed that they would be ultimately responsible for ensuring sex chromosome association [30] . However , this may not be the case for sex chromosomes in M . unguiculatus since X and Y chromosomes do not present distal contacts of their AEs at pachytene , a stage that lasts several days in the Mongolian gerbil [54] . Distal connections between the tips of the AEs are only found from early diplotene onwards . Therefore , other mechanisms must be discussed . Nevertheless , it is possible that all the combinations of distal association previously reported [30 , 32 , 33] may respond to a sequential and random clustering of telomeres , which culminates in the association of all AE tips at late pachytene/early diplotene , and that the differences found between species are simply due to different timing in the association of sex chromosomal ends . An alternative explanation is that the particular chromatin condensation of the sex body may contribute to maintain the association of X and Y chromosomes . It is currently known that sex chromosomes are transcriptionally inactive during most of the first meiotic prophase in mammals and a huge number of proteins , including γ-H2AX , are specifically associated to and/or modified in the sex body [41 , 55 , 56] . In this sense , it has been demonstrated that disruption of H2AX in mouse abolishes meiotic sex chromosome inactivation and sex body formation [40] . Furthermore , in H2AX-depleted mice sex chromosome pairing is severely disturbed and X and Y chromosomes are often located separately in the nucleus . Our results on the location of γ-H2AX indicate that sex chromosomes in the Mongolian gerbil are inactivated and form a compacted chromatin mass at the nuclear periphery . Therefore , it is possible that the chromatin conformation acquired during first meiotic prophase could have an important role in maintaining sex chromosome association in absence of SC formation . The correct segregation of chromosomes during first meiotic division depends on their proper alignment and biorientation at the metaphase I plate . In M . unguiculatus , as occurs in other Gerbillidae species , X and Y chromosomes appear distally associated at metaphase I . However , the absence of a chiasma between sex chromosomes , as revealed by the absence of MLH1 , poses an intriguing question about the mechanism that maintains their association . In P . obessus , this association is mediated by distal blocks of heterochromatin [30] . However , our results in the Mongolian gerbil reveal this distal association may not be mediated just by chromatin interactions . We suggest that the physical connection mediated by SYCP3 protein , starting at the latest stages of the first meiotic prophase , may be responsible for maintaining sex chromosomes connected ( Figure 6 ) . This SYCP3 link would prevent X and Y chromosomes to separate each other before they biorientate at the metaphase I plate . Afterward , as soon as each chromosome is pulled to the spindle poles , sex chromosomes would tend to separate displaying an early segregation . This would explain our finding of different configurations in the association of sex chromosomes at metaphase I . Since sex chromosomes always appear as laggards at anaphase I it seems that their movement to the poles is somehow obstructed . Physical links between segregating half bivalents at anaphase I have been detected in a wide range of species [57] but are specially well characterized in crane flies [58] . In these species , the presence of elastic tethers between the ends of segregating chromosomes has been reported , giving rise to the stretch of the chromosome arms [59] . The nature of such bridges remains obscure , but it was proposed that they could be formed by chromatin fibers [58] or even by elements related to the attachment of telomeres to the nuclear envelope [59] . In this way , it is likely that the SYCP3 filaments that connect sex chromosomes in M . unguiculatus , by establishing a physical link between their ends , could act also as a tether that retards sex chromosome separation . However , we do not favor the idea that chromatin could be also involved in this connection because we were unable to detect any link by either DAPI or γ-H2AX labeling . The critical feature in this context is how the AEs components derive in such a structure . Previous studies have shown that the elements of the SC can be transformed into a variety of structures that may remain associated to chromosomes until anaphase I [60 , 61] . Furthermore , in vitro experiments have shown that both SYCP3 and SYCP1 are able to self assemble into filaments and polycomplex-like structures , respectively [62 , 63] . Our observations indicate that in the Mongolian gerbil SYCP3 reorganization initiates at diplotene , concomitant with the initiation of sex chromosome end-to-end connections . An electron-dense irregular network has been detected at late pachytene on the sex chromosomes in other gerbil species , soon after the establishment of sex chromosome distal associations [31 , 32] . Therefore , it is likely that other Gerbillidae species also present a similar SYCP3 reorganization . Although the timing of these changes may differ between species , they could be a part of a conserved program of sex chromosome modification at the late stages of first meiotic prophase . Interestingly , in marsupial mammals a structure derived from the sex chromosomal AEs , the dense plate , is also involved in both pairing [24] and segregation [29] of the achiasmate sex chromosomes . Although differences exist between the organization and behavior of the SYCP3 structures characterized in marsupials and in the Mongolian gerbil , these findings indicate an evolutionarily conserved role of SC components not only in synapsis but also in chromosome segregation when synapsis and/or recombination do not take place . Finally , the finding that SYCP3 may form conspicuous aggregates in the cytoplasm , which is a feature common to other species of mammals , is also remarkable [44] . In the Mongolian gerbil these structures , mainly bars and filaments , are detected from diakinesis up to interkinesis . Our results suggest that the origin of these structures may be multiple: ( i ) aggregates formed at later stages of the first meiotic prophase ( diakinesis ) , and which remain out of the nucleus until interkinesis; ( ii ) filaments in the spindle area that could derive directly from the SYCP3 on the autosomes at the beginning of anaphase I; and ( iii ) filaments that presumably dissociate from the SYCP3 link between the X and Y chromosomes during anaphase I and telophase I . The formation of these filaments could be related to the tendency of SYCP3 to form filaments when over expressed in vitro [62] . Different animal groups challenge the rule that synapsis and recombination are required for proper segregation . It is well known that insects represent a wide range of segregation mechanisms of achiasmate chromosomes [12 , 17 , 20 , 64] . The observations presented here and those of previous authors indicate that , at least in mammals , special chromatin conformations ( heterochromatinization and/or inactivation ) and modification of SC components may be key mechanisms to explain the proper association and segregation of achiasmate chromosomes [29 , 30 , 33 , 34 , 65] . Since the occurrence of such chromosomes is a feature found in almost all groups of organisms , they may represent universal backup mechanisms to ensure the correct outcome of meiosis in the absence of synapsis and recombination .
Slides were incubated overnight at 4 °C with the following primary antibodies diluted in PBS: mouse monoclonal anti-SYCP3 ( Abcam , 12452 ) at a 1:100 dilution; rabbit anti-SYCP3 ( Abcam , 15093 ) at a 1:50 PBS dilution; rabbit anti-SYCP1 ( Abcam , 15087 ) at a 1:100 dilution; mouse monoclonal against histone H2AX phosphorylated at serine 139 ( γ-H2AX ) ( Upstate , 05–636 ) at a 1:3 , 000 dilution; rabbit anti-RAD51 ( Calbiochem , PC130 ) at a 1:50 dilution; mouse monoclonal anti-MLH1 ( Pharmingen , 551091 ) at a 1:10 dilution; and a human anti-centromere serum that recognizes centromeric proteins ( Antibodies Incorporated , 15–235 ) at a 1:100 dilution . Slides were rinsed 3 × 5 min in PBS and subsequently incubated with secondary antibodies in a moist chamber at room temperature for 1 h: fluorescein isothiocyanate ( FITC ) -conjugated goat anti-mouse IgG; Texas Red ( TR ) -conjugated goat anti-mouse IgG; FITC-conjugated goat anti-rabbit IgG; TR-conjugated goat anti-rabbit IgG; and TR-conjugated goat anti-human IgG . All secondary antibodies were from Jackson ( Jackson ImmunoResearch Laboratories ) and used a 1:100 dilution . Slides were subsequently rinsed in PBS 3 × 5 minutes , stained with DAPI , and mounted with Vectashield ( Vector ) . For double detection of two antibodies raised in mouse , we followed the procedure previously described [24] . Observations were made on an Olympus BX61 microscope equipped with a motorized Z axis . Images were captured with an Olympus DP70 digital camera using the analySIS software ( Soft Imaging System , Olympus ) and processed by using public domain ImageJ ( National Institutes of Health , http://rsb . info . nih . gov/ij ) and Adobe Photoshop 7 . 0 software .
|
Meiosis is a special kind of cell division that leads to the formation of gametes . During meiosis the number of chromosomes must be halved in the daughter cells , and to do this properly , most organisms use an amazing strategy: during the first of the two meiotic divisions , homologous chromosomes associate in pairs , undergo a reciprocal genetic interchange , and then each member of the pair segregates into a different daughter cell . Genetic exchange , called meiotic recombination , is a key process to ensure that homologous chromosomes remain tightly associated until they segregate . In general , sex chromosomes are subjected to the same processes as the rest of chromosomes . But , of course , exceptions exist . This is the case in the Mongolian gerbil , a mammal whose sex chromosomes pair and segregate during male meiosis without undergoing meiotic recombination . We have found that they are able to do this because some proteins of a meiosis-specific structure , the synaptonemal complex , are reorganized to maintain sex chromosomes associated until they segregate into daughter cells . This kind of behavior resembles the situation found in marsupials and some insect species , indicating a recurrent role of synaptonemal complex components in chromosome segregation when meiotic recombination is missing .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"mongolian",
"gerbil",
"(meriones",
"unguiculatus)",
"cell",
"biology"
] |
2007
|
Meiotic Pairing and Segregation of Achiasmate Sex Chromosomes in Eutherian Mammals: The Role of SYCP3 Protein
|
Entamoeba histolytica is an enteric pathogen responsible for amoebic dysentery and liver abscess . It alternates between the host-restricted trophozoite form and the infective environmentally-stable cyst stage . Throughout its lifecycle E . histolytica experiences stress , in part , from host immune pressure . Conversion to cysts is presumed to be a stress-response . In other systems , stress induces phosphorylation of a serine residue on eukaryotic translation initiation factor-2α ( eIF2α ) . This inhibits eIF2α activity resulting in a general decline in protein synthesis . Genomic data reveal that E . histolytica possesses eIF2α ( EheIF2α ) with a conserved phosphorylatable serine at position 59 ( Ser59 ) . Thus , this pathogen may have the machinery for stress-induced translational control . To test this , we exposed cells to different stress conditions and measured the level of total and phospho-EheIF2α . Long-term serum starvation , long-term heat shock , and oxidative stress induced an increase in the level of phospho-EheIF2α , while short-term serum starvation , short-term heat shock , or glucose deprivation did not . Long-term serum starvation also caused a decrease in polyribosome abundance , which is in accordance with the observation that this condition induces phosphorylation of EheIF2α . We generated transgenic cells that overexpress wildtype EheIF2α , a non-phosphorylatable variant of eIF2α in which Ser59 was mutated to alanine ( EheIF2α-S59A ) , or a phosphomimetic variant of eIF2α in which Ser59 was mutated to aspartic acid ( EheIF2α-S59D ) . Consistent with the known functions of eIF2α , cells expressing wildtype or EheIF2α-S59D exhibited increased or decreased translation , respectively . Surprisingly , cells expressing EheIF2α-S59A also exhibited reduced translation . Cells expressing EheIF2α-S59D were more resistant to long-term serum starvation underscoring the significance of EheIF2α phosphorylation in managing stress . Finally , phospho-eIF2α accumulated during encystation in E . invadens , a model encystation system . Together , these data demonstrate that the eIF2α-dependent stress response system is operational in Entamoeba species .
Entamoeba histolytica is an intestinal parasite that is the causative agent of amebic dysentery and amoebic liver abscesses . It is transmitted by the cyst form of the pathogen in fecally-contaminated food and water , making it prevalent in the developing world where sanitation practices are substandard . There are 173 million people that live in regions with untreated water sources and one billion people that carry out open defecation practices [1] . Thus , there is considerable risk for transmission of E . histolytica . E . histolytica is passed from human to human without the utilization of an intermediate host . The parasite’s latent stage , a cyst , is able to withstand extreme conditions in the external environment as well as the acidic pH of the host stomach . The cyst exits the stomach and enters the small intestine , where unknown triggers cause excystation . The emerging active trophozoites continue down the digestive system until they reach the large intestine , where they establish infection , feed on bacteria and host cell material , and divide by binary fission . Trophozoites can also invade the colonic epithelial lining and cause extraintestinal complications of infection including liver abscess . During infection the parasite may experience stress , in part , due to immune pressure from the host . This stress can include heat shock , osmotic shock , nutrient deprivation , and/or exposure to reactive oxygen species , nitrogen species , or high oxygen . To survive , the parasite must elicit a cellular response to counter these stresses . E . histolytica does not readily encyst in axenic culture . Thus , E . invadens , a related reptilian intestinal parasite that can be induced to encyst in vitro , has been widely used as a model system [2 , 3 , 4 , 5 , 6] . Conversion to latency in E . invadens is accompanied by increased expression of the heat shock protein , BiP/GRP78 [3] . Thus , encystation is likely to be a stress response in Entamoeba species . In many systems , stress is controlled , in part , by the phosphorylation of the alpha subunit of eukaryotic initiation factor 2 ( eIF2 ) [reviewed in 7] . This factor is a heterotrimeric protein complex consisting of alpha ( α ) , beta ( β ) , and gamma ( γ ) subunits . In normal growth conditions , eIF2 forms a ternary protein complex with GTP and Met-tRNAi . The Met-tRNAi is then delivered to the ribosome to initiate translation . Once Met-tRNAi is delivered , the bound GTP is hydrolyzed to GDP . To become reactivated , eIF2-GDP binds to the guanine exchange factor , eIF2B , and the GDP is released , allowing for the binding of a new GTP . Nucleotide exchange is considered the rate-limiting step of translation initiation [7] . During stress , eIF2 kinases become activated and phosphorylate a key serine residue on the alpha subunit ( eIF2α ) to generate a phosphorylated form of the protein ( phospho-eIF2α ) . This phosphorylation induces a conformational change in eIF2 , causing it to become a competitive inhibitor of eIF2B . This leads to a general decline in protein biosynthesis; however , paradoxically , the expression of a subset of genes is up-regulated . This subset of genes assists the cell in countering stress . In other eukaryotic pathogens , stage conversion to a latent form is accompanied by phosphorylation of eIF2α . For example , stress induces the parasite , Toxoplasma gondii , to convert from a replicating tachyzoite form to a latent bradyzoite form and phosphorylation of eIF2α occurs during this stage transition [8] . Phospho-eIF2α also regulates the formation of latent sporozoites in Plasmodium spp . [9] and the transition of promastigotes to amastigotes in Leishmania [10] . In non-parasitic organisms , such as yeast [11] and Dictyostelium [12] , phosphorylation of eIF2α stimulates the formation of latent spores . Genomic data suggest that E . histolytica and E . invadens possess the components of this stress-response system [13] . However , the role of eIF2α phosphorylation in the Entamoeba stress response has never been characterized . In this study , we show that phosphorylation of E . histolytica eIF2α ( EheIF2α ) occurs in response to certain stress conditions , namely long-term serum starvation , long-term heat shock , and oxidative stress . Phosphorylation of EheIF2α is accompanied by a reduction in global translation . We also show that expression of non-phosphorylatable or phosphomimetic forms of EheIF2α in E . histolytica influences translation and the ability to counter stress . Finally , we demonstrate that phosphorylation of E . invadens eIF2α ( EieIF2α ) accompanies encystation . Together , these data support the hypothesis that Entamoeba species possess an eIF2α-based stress response system that controls protein synthesis , and possibly encystation .
An alignment of the E . histolytica eIF2α ( EheIF2α ) and E . invadens eIF2α ( EieIF2α ) amino acid sequences with that of five different organisms showed that they shared low sequence identity and moderate sequence similarity across the entire protein with other eIF2α proteins , even when compared to the factor from other eukaryotic pathogens ( Fig 1A and 1B ) . EheIF2α and EieIF2α were most similar to each other . Though overall shared homology was low , there was strong sequence identity surrounding the conserved regulatory serine residue , which occurs at amino acid position 59 in the Entamoebae ( Fig 1C ) . Thus , serine-59 is likely to be the residue phosphorylated during stress . Conservation around this residue suggests that the machinery for an eIF2α-based stress-response system may be present in E . histolytica and E . invadens . To determine if the level of phospho-eIF2α changes during stress , cells were exposed to a variety of stress conditions . Only previously-established stressors were chosen: short- and long-term serum starvation [14] , short- and long-term heat shock [15] , glucose deprivation [16] , and oxidative stress [17] . To confirm that the applied stresses were not inducing significant cell death , which would confound our studies , viability was assessed . Only long-term serum starvation and oxidative stress resulted in statistically significant cell death , albeit the mortality was not complete ( Fig 2 ) . To track the level of total and phospho-eIF2α during stress , antibodies that specifically recognize the phosphorylated form or total Entamoeba eIF2α were generated in rabbits and authenticated by Western blotting ( S1 Fig ) . Western blotting also revealed that there was a basal level of phosphorylated EheIF2α in control unstressed trophozoites ( Fig 3 ) . While all of the stress conditions induced an increase in the level of phospho-EheIF2α , only trophozoites that experienced long-term serum starvation , long-term heat shock , or oxidative stress exhibited a statistically significant increase in the level of phospho-EheIF2α ( Fig 3 ) . This was not simply due to cell death , since long-term heat shock , which caused minimal cell mortality , induced one of the most dramatic increases in the level of phosphorylated EheIF2α . These data suggest that an eIF2α-based response system exists in E . histolytica and is activated in a stress-specific manner . In other systems , eIF2α-based control of stress is accompanied by a reduction in global translation [8] . Therefore , we examined global translation in control and stressed E . histolytica cells by characterizing the abundance of high-density polyribosomes using sucrose gradient ultracentrifugation . We chose two conditions of stress; one that induced phosphorylation of EheIF2α ( i . e . , serum starvation ) , and one that did not induce phosphorylation of EheIF2α ( i . e . , glucose deprivation ) . Consistent with the known function of phospho-eIF2α , serum starvation resulted in a significant reduction in dense polyribosomes and an increase in free ribosomes and monosomes when compared to control cells ( Fig 4A and 4B ) . As expected , glucose deprivation did not result in a decrease in dense polyribosome-bound transcripts ( Fig 4C ) . These results support the premise that E . histolytica possesses an eIF2α-based stress response system that reduces translation . To further examine the function of phospho-EheIF2α , we generated E . histolytica cell lines that conditionally overexpress mutant forms of EheIF2α . The cDNA encoding EheIF2α was mutagenized in two ways using PCR . The codon for serine ( S ) at position 59 was changed to that of alanine ( A ) or aspartic acid ( D ) to produce non-phosphorylatable ( EheIF2α-S59A ) or phosphomimetic ( EheIF2α-S59D ) forms of EheIF2α , respectively [18] . Wildtype ( unaltered ) cDNA was designated EheIF2α-S59 . To distinguish exogenous EheIF2α from the endogenous form , the 5’ ends of wildtype and mutated cDNAs were also modified to include sequence encoding an N-terminal FLAG epitope peptide sequence ( DYKDDDDK ) [19] followed by a 5-glycine flexible region . Modified PCR products were inserted into the E . histolytica expression vector pGIR209 , which confers G418 ( neomycin ) resistance and allows for tetracycline-inducible expression of exogenous genes [20] . A standard electroporation protocol [21] was utilized to introduce the expression vector into trophozoites that had been previously transfected with another plasmid , pGIR308 . This partner plasmid encodes the tetracycline repressor protein , which is necessary for tetracycline inducibility . Authentic transfection was confirmed by purification and sequencing of the episomal expression plasmids from stably transfected cell lines [22] . A previously established cell line that conditionally expresses an irrelevant protein , luciferase , was used as a control [20] . Expression of exogenous protein was induced by the addition of 5 μg mL-1 tetracycline to the culture medium for a minimum of 24 h . Western blot analysis using anti-FLAG and anti-Entamoeba eIF2α showed successful induction of exogenous protein with little to no expression prior to the addition of tetracycline ( Fig 5 ) . Interestingly , in the cell line over-expressing the wildtype version of eIF2α , FLAG-tagged EheIF2α-S59 appeared to be itself heavily phosphorylated . It is not known why exogenously expressed eIF2α is so readily phosphorylated; but , the observation is in accordance with another study in which over-expression of wildtype eIF2α in human A375 melanoma cells resulted in an increase in the basal level of phosphorylated eIF2α [23] . The mutated versions of eIF2α , themselves , cannot be phosphorylated and their expression did not seem to induce an increase in the level of endogenous phospho-eIF2α ( Fig 5 ) . To confirm that the exogenously expressed EheIF2α variant proteins were functional , we monitored polyribosome abundance in the transgenic cell lines after 24 or 72 h of tetracycline induction . After 24 h of induction , there was no change in polyribosome abundance in any cell line ( S2 Fig ) . After 72 h of induction , polyribosome abundance remained high in cells expressing luciferase ( control ) ( Fig 6A ) . This was expected given that expression of this irrelevant protein should not interfere with the translation machinery . Compared to control cells expressing luciferase , there was no statistically significant difference in the abundance of high-density polyribosomes in cells expressing wildtype EheIF2α-S59 ( Fig 6B ) . However , the polyribosome peaks ( Fig 6B ) in the EheIF2α-S59-expressing cells were not as well defined as those in the luciferase-expressing control cell line ( Fig 6A ) . Given that nucleotide exchange on eIF2α represents the rate-limiting step in translation initiation [7] , an overabundance of the factor may have relieved this inhibition causing an increase in the translation of mRNAs . Cells expressing the phosphomimetic variant , EheIF2α-S59D , exhibited a statistically significant decrease in high-density polysomes after 72 h of tetracycline-induction ( Fig 6D ) . Given that the phosphorylation of eIF2α down-regulates translation , this result was expected . Surprisingly , cells expressing EheIF2α-S59A also exhibited a decrease in high density polysomes after 72 h of tetracycline-induction , albeit not significantly ( Fig 6C ) . The observation was unforeseen since the presence of unphosphorylated eIF2α is not normally correlated with a decrease in translation . Currently , these unique data cannot be explained , but suggests that the non-phosphorylatable variant of eIF2α is behaving in a dominant negative fashion in E . histolytica cells . Since polyribosome profiling provides a snapshot of translation at the mRNA level , we wanted to confirm the alterations in biosynthesis in the mutants at the protein level . Therefore , we used a second method known as SUrface SEnsing of Translation ( SUnSET ) . This non-isotopic technique uses anti-puromycin antibody for the immunological detection of puromycin-labelled proteins [24] . When added to live cell cultures , puromycin , a tyrosoyl-tRNA analog , becomes incorporated into actively translating proteins . Subsequent Western blot analysis of whole cell lysates with anti-puromycin antibody reveals the extent of active protein production in the cell . SunSET has been previously used to monitor protein biosynthesis in E . histolytica [25 , 26] . To determine if SUnSET could be used to track protein synthesis in E . histolytica in our hands , wildtype cells were incubated with puromycin before or after incubation with cycloheximide , an inhibitor of protein biosynthesis . Western blotting using anti-puromycin antibody revealed that puromycin was readily incorporated into proteins in E . histolytica ( Fig 7A; “Puro” ) . The incorporation was specific since there was minimal background staining of lysates from cells that were not treated with puromycin ( Fig 7A; “Control” ) , cells treated only with cycloheximide ( Fig 7A; “Cyclo” ) or from cells first treated with cycloheximide followed by exposure to puromycin ( Fig 7A; “Cyclo + Puro” ) . Exogenous protein expression was induced in the transgenic cell lines and then the cells were subjected to SUnSET analysis . Cells overexpressing EheIF2α-S59 exhibited the highest incorporation of puromycin ( Fig 7B ) indicating the highest level of protein biosynthesis . In this cell line , increased expression may have been the result of decreased pressure on the rate-limiting step in translation . Consistent with the polyribosome profiles , incorporation of puromycin into proteins in the cell lines expressing EheIF2α-S59A or EheIF2α-S59D was low ( Fig 7B ) . The mutant cell lines were then assessed for their ability to survive the various stress conditions applied previously . After 24 h of tetracycline induction followed by application of stress , there was no statistical difference in survivability among the cell lines ( Fig 8A ) . This was not surprising since there was also no detectable defect in protein biosynthesis at 24 hours ( S1 Fig ) . To further explore the ability of the eIF2α variants to offer protection during stress we chose one condition , namely long-term serum starvation , which was able to induce the highest levels of EheIF2α phosphorylation . Cells were treated with tetracycline for 24 h and then serum-starved in the presence of tetracycline for a total of 72 h . Although , expression of all of the EheIF2α variants seemed to offer some protection during stress , only cells that expressed EheIF2α-S59D exhibited statistically significant ability to survive long-term serum starvation . Given that this variant is the phosphomimetic version of eIF2α , it is possible that this cell line was pre-conditioned to handle stress , perhaps by already expressing an important subset of stress-responsive genes . Since expression of the conserved 70 kDa heat shock protein ( hsp70 ) increases during encystation [3] , stage conversion is presumed to be a stress response in the Entamoeba species [3] . Since our antibodies cross-reacted in the E . invadens system ( S1 Fig ) , the level of total and phospho-EieIF2α was assessed during encystation by Western blotting . Similar to E . histolytica , E . invadens trophozoites possessed a basal level of phospho-eIF2α ( Lanes T , Fig 9A and 9B ) . After 24 h into encystation , there was a significant increase in the level of the phosphorylated form of the factor ( Lane 24 in Fig 9A; Fig 9C ) . The level of phospho-EieIF2α remained high through 72 h into encystation ( Lane 72 in Fig 9B; Fig 9C ) . It was likely that there were some un-encysted trophozoites remaining in the population at 72 h of encystation . Since detergent-resistance is a hallmark of E . invadens cysts , we also probed the 72 h population after treatment with sarkosyl . Sarkosyl is a detergent that lyses trophozoites and immature cysts , resulting in a population that consists of detergent-resistant mature cysts . Phospho-EieIF2α was enriched after removal of detergent-sensitive cells ( Lane 72C in Fig 9B; Fig 9C ) demonstrating that cysts possess the phosphorylated factor at higher levels than trophozoites . In all cases , equal protein load was measured by Coomassie staining ( Fig 9A and 9B ) . Actin was not used as a load control because actin transcripts are developmentally regulated [5] . Overall , our data show that there is correlation between encystation and the appearance of phospho-EieIF2α in E . invadens suggesting that phospho-EieIF2α plays a role in encystation in the Entamoeba species .
This study is the first to demonstrate that stress and encystation can induce phosphorylation of eIF2α in the Entamoebae . Specifically , phospho-eIF2α was significantly increased in E . histolytica as a result of long-term serum starvation , long-term heat shock , or oxidative stress and in E . invadens after induction of encystation . During long-term serum starvation of E . histolytica , phosphorylation of eIF2α was accompanied by reduced translation . This suggests that a phospho-eIF2α-based stress response system exists in E . histolytica . In further support of this , a transgenic cell line expressing a phosphomimetic form of eIF2α exhibited higher viability during long-term serum starvation than cell lines expressing other versions of eIF2α or an irrelevant protein , luciferase . In other words , phosphorylation of eIF2α seems to promote E . histolytica survival during stress . Analysis of E . histolytica genome data suggests that this pathogen possess other components of this stress response system . There are putative homologs for eIF2β ( EHI_153480 ) and eIF2γ ( EHI_132880 ) . Furthermore , E . histolytica possesses two presumptive eIF2α kinases ( eIF2K ) ( EHI_109700 , EHI_035950 ) [13] . Currently , the E . histolytica eIF2α kinases have not been authenticated , nor have the conditions that lead to their activation been discerned . Nonetheless , the occurrence of each of the three subunits of eIF2 , as well as kinases , in genome sequences indicates that this translation factor has a conserved role in this pathogen in delivering Met-tRNAis to the translation machinery . In support of this , cells expressing the phosphomimetic variant of EheIF2α exhibited reduced polyribosme abundance . As expected , long-term serum starvation , a condition that induces phosphorylation of eIF2α , also exhibited reduced polyribosome abundance . The ability of the parasite to respond to oxidative stress is critical for virulence functions [reviewed in 27] including the pathogen’s ability to survive in host liver [28] . The observation that eIF2α becomes phosphorylated during oxidative stress in E . histolytica is in accordance with other studies that demonstrate that phosphorylation of eIF2α protects mammalian cells during oxidative stress [29 , 30] . Our findings are also consistent with the observation that protein biosynthesis is inhibited during oxidative stress in E . histolytica [25] . It has been suggested that oxidation of components of the parasite’s translation machinery ( e . g . , ribosomal proteins , elongation factors ) [25] , and enzymatic down-regulation of almost all tRNA species [31] are responsible for reduced protein biosynthesis during H2O2 exposure . Our data show that phosphorylation of eIF2α may also contribute to changes in gene expression and protein biosynthesis during oxidative stress in the pathogen . It has been reported that serum starvation induces expression of a long non-coding RNA , EhslncRNA [32] and alters ribosome biogenesis by regulating the expression of rRNAs and ribosomal proteins [33] . To date , there have been no genome-wide transcriptomic analyses of serum-starved E . histolytica cells . Therefore , the identification of the entire set of genes that are down-regulated or up-regulated during this type of stress remains to be seen . However , transcriptomic analyses have been performed on E . histolytica cells exposed to oxidative stress . Specifically , incubation with H2O2 induced statistically significant down-regulation of 102 genes and up-regulation of 184 genes [34] . More recently , it has been shown that this differential gene expression may be regulated by a novel transcription factor , which binds to a specific H ( 2 ) O ( 2 ) -regulatory motif ( HRM ) in the promoters of genes [35] . Our data suggest that phosphorylation of eIF2α may halt global translation allowing time for the parasite to reconfigure gene expression during exposure to H2O2 . The eIF2α response to stress appears to be specific to the condition applied since phospho-EheIF2α did not significantly increase in other presumptive states of stress including short-term serum starvation , short-term heat shock , or glucose deprivation . Interestingly , transcriptomic analyses of E . histolytica cells exposed to short-term heat shock results in a significant yet general decline in gene expression accompanied by differential expression of the alleles encoding the heavy subunit of the Gal/GalNAc lectin , a parasite surface protein responsible for interaction with host cells [15] . The current study suggests that such short-term heat shock-induced changes in gene expression do not depend on phosphorylation of eIF2α . Unlike in most glucose-starved mammalian cells [36 , 37] , phospho-EheIF2α did not accumulate in glucose-starved E . histolytica trophozoites . While it is accepted that glucose starvation is a “metabolic stressor” in E . histolytica [16] , it is a form of stress that apparently does not induce the expression of the conserved 70 kDa heat shock protein ( hsp70 ) [16] . Instead , glucose starvation in E . histolytica induces dramatic changes in the expression of other genes and an increase in virulence [16] . Additionally , glucose starvation does not increase the ability of E . histolytica trophozoites to survive a subsequent challenge such as heat shock or oxidative stress [16] . It is intriguing that a form of stress ( i . e . , glucose starvation ) that does not induce hsp70 expression nor protect cells from subsequent stressors [16] also does not induce phosphorylation of eIF2α ( this study ) . Given that standard E . histolytica culture medium is poorly-defined , it is also possible that other carbohydrates serve as a nutrient source in the absence of glucose , which may minimize the stress . This would explain a lack of hsp70 accumulation [16] , a lack of phospho-EheIF2α accumulation ( this study ) , and maintenance of high-density polysomes ( this study ) . E . histolytica is likely well-adapted to handle a low glucose environment as this would be a common condition in the host large intestine [38] . Our data suggest that the parasite’s response to a low glucose environment does not seem to rely on phosphorylation of eIF2α . Cells that express the phosphomimetic version of eIF2α exhibited a statistically significant higher ability to survive at least one condition of stress , long-term serum starvation . In general , global translation is reduced by phosphorylation of eIF2α . However , it is known that certain mRNAs escape this inhibition and are selectively expressed even when phospho-eIF2α accumulates . For example , in Toxoplasma , 500 gene transcripts are preferentially associated with high-density translation-active polysomes during stress [18] . In mammalian cells , activating transcription factor 4 ( ATF4 ) [39] , C/EBP-homologous protein ( CHOP ) [39] , and the α isoform of inhibitor of Bruton's tyrosine kinase ( IBTKα ) [40] are among those proteins preferentially expressed after phospho-eIF2α accumulates . Interestingly , depletion of IBTKα by short hairpin RNA technology reduces viability of cells during ER stress [40] . Thus , proteins that escape translational control by phospho-eIF2α may be among those key players that determine cell fate during stress . It is possible that in E . histolytica , expression of the phosphomimetic version of eIF2α , induced preferential expression of proteins , some of which may be protective during stress . This could explain increased viability of trophozoites during long-term serum starvation . Identification of those mRNAs associated with high-density polyribosomes during stress will provide further insight . In the current study , the expression of wildtype and mutant forms of eIF2α was exogenous . Due to polyploidy , methods for homologous recombination and gene replacement are not yet available for E . histolytica . As such , in all of the transgenic cell lines , endogenous eIF2α was present . Despite this limitation , we observed phenotypes in the transgenic cell lines . As expected , cells over-expressing wildtype or phosphomimetic variants of eIF2α exhibited increased or decreased translation , respectively . Thus , these exogenous proteins were able to apparently compete with endogenous eIF2α in a predicted fashion . Surprisingly , in cells expressing the non-phosphorylatable variant , EheIF2α-S59A , polyribosome abundance and protein biosynthesis were decreased . This phenotype was unexpected because non-phosphorylated eIF2α is generally considered a driver of translation . However , in other systems , the phenotypic outcome of expression of S-to-A variants of eIF2α has not been uniform . For instance , exogenous expression of the non-phosphorylatable variant of eIF2α increases [35S]-methionine incorporation into murine 3T3 L1 cells [19] . Overexpression of eIF2α-S51A can transform NIH 3T3 fibroblasts [41] but not 3T3 L1 cells [19] . In systems where gene replacement at both alleles can be achieved , expression of S-to-A variants causes dramatic outcomes including changes in morphology [42] , ER stress [42] , increases in intracellular reactive oxygen species [29] and cell death [42] . Perhaps EheIF2α has additional unidentified and non-canonical roles in E . histolytica and overexpression of the S59A variant leads to a unique phenotype . It might also be possible that the alanine-bearing mutant is behaving in a dominant-negative fashion by titrating other important components of the translation initiation system . Like several other ancient protozoa [43] a canonical ER-based unfolded protein response ( UPR ) may be incomplete in E . histolytica [44] . Therefore , the unusual phenotype of the EheIF2α-S59A-expressing cell line is likely not the result of overexpression of an exogenous protein that leads to ER-stress . In support of this , overexpression of the control protein , luciferase , did not result in a similar decline in polysome abundance ( Fig 6A ) . A thorough examination of the eIF2α binding partners in mutant and control cells will be necessary to fully understand the phenotype . We demonstrate that the level of phospho-EieIF2α increases during encystation in E . invadens . It is well-known that eIF2-based systems are used in eukaryotes for the conversion to latent or dormant forms . Phosphorylation of eIF2α is responsible , in part , for stage conversion in Toxoplasma gondii [8] , Plasmodium spp . [9] Leishmania spp . [45] , yeast [11] , and Dictyostelium discoideum [12] . The accumulation of phospho-EieIF2α during encystation of E . invadens ( this study ) is consistent with several previous studies that suggests translation declines during encystation [46 , 47] . Specifically , the abundance of high-density polyribosomes [46] and the incorporation of exogenous amino acids [47] are decreased in encysting E . invadens cells . Furthermore , encystation is accompanied by the aggregation of ribosomes into structures known as a chromatoid bodies [47] . Chromatoid bodies are RNA- and ribosomal-containing cytoplasmic granules that arise during stress . They are reminiscent of stress granules that accumulate in an eIF2-dependent manner in other systems [48] . Overall , our data , along with data that show that the levels of stress response proteins , such as hsp70 [3] and hsp90 [49] , fluctuate during stage conversion in the Entamoebae , support the notion that encystation is a stress response . Like E . histolytica , E . invadens possesses two presumptive eIF2α kinases ( EIN_052050; EIN_096010 ) . Expression of the former is developmentally-regulated [5] . Specifically , there is a statistically significant increase in transcript of the former kinase at 24 h of encystation . Interestingly , this surge in transcript level corresponds with the appearance of phospho-eIF2α ( Fig 9A ) . Thus , EIN_052050 is a candidate for regulating the levels of phospho-eIF2α during stage conversion . It remains to be seen if phosphorylation of eIF2α is necessary and/or sufficient to induce stage conversion to the cyst form and further studies with the model organism will provide insight . In conclusion , we have shown that phospho-eIF2α accumulates in E . histolytica during a variety of stress conditions and in E . invadens during encystation . In E . histolytica , the accumulation of the phosphorylated version correlated with a decrease in translation . To the best of our knowledge , this is the first example of translational control in this pathogen .
Amino acid sequences of eIF2α for six model species were aligned individually to Entamoeba histolytica eIF2α to examine sequences surrounding the key serine reside . The sequences were also analyzed using a Standard Protein BLAST v 2 . 3 . 1 ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ? PAGE=Proteins ) and the BIOSUM 62 algorithm for a “positive” or similarity score , as well as an identity score to form a similarity and an identity matrix . Entamoeba histolytica ( strain HM-1:IMSS ) and Entamoeba invadens ( strain IP-1 ) were cultured axenically in TYI-S-33 medium in 15 mL glass screw cap tubes at 37°C [50] or 25°C [6] , respectively . Cells were passaged into fresh media every 72 to 96 h . Log-phase E . histolytica trophozoites were incubated on ice for 10 min to release the cells from the glass surface . Centrifugation was performed at 500 x g for 5 min to pellet cells . The cell pellet was resuspended in the appropriate stress medium as follows and incubated at 37°C unless otherwise noted . To induce short-term serum starvation , cells were cultured in TYI-S-33 medium without the addition of adult bovine serum , penicillin-streptomycin , and Diamond’s Vitamins for 1 h prior to analysis . To induce long-term serum starvation , cells were incubated in the same medium for 24 h prior to analysis [14] . Short-term heat shock was induced by incubating trophozoites in complete growth medium in a 42°C water bath for 4 h [15] . To assess long-term heat shock , cells were incubated at 39°C for 24 h . To induce glucose deprivation , trophozoites were incubated for 12 h in TYI-S-33 medium without glucose [16] . To induce oxidative stress , 500 μM hydrogen peroxide ( Fisher Scientific ) was added to cells in normal TYI-S-33 medium followed by a 45 min incubation [17] . Viability was determined with microscopy using Trypan Blue exclusion ( VWR ) . Stage conversion ( encystation ) was induced by incubating E . invadens trophozoites in 47% LYI-LG , a standard encystation medium with reduced glucose , osmolarity and serum [2 , 3 , 4 , 5 , 6] . To purify mature detergent-resistant cysts from trophozoites , the 72 h population was incubated in 0 . 05% ( v/v ) sarkosyl in PBS at room temperature for 20 min [51] . Antibody development was outsourced ( Pierce Biotechnology , Inc . , Rockford , IL , USA ) and was approved by the Clemson University Institutional Animal Care and Use Committee ( Protocol number: AUP2015-021 ) . Briefly , rabbits were immunized with synthetic phosphorylated polypeptide- ILMSEL ( pS ) KRRFRS and with the unphosphorylated polypeptide EMGTYVALKEYDDIQGMIP targeting phosphorylated or total eIF2α , respectively . These antibodies were purified by ELISA and confirmed against the synthetic polypeptide used in the initial immunization . SDS-PAGE and Western blot analysis were used to test the specificity of the antibodies and were performed as described previously [52] . Briefly , E . histolytica or E . invadens trophozoites ( 4x104 ) were collected by centrifugation , resuspended in NuPage LDS buffer ( Life Technologies , Carlsbad , CA , USA ) and heated for 5 minutes at 95°C . Cell samples were loaded onto a precast 12% Bis-Tris polyacrylamide gel ( Life Technologies , Carlsbad , CA , USA ) . The gel was electrophoresed at 200V and the separated proteins were transferred to a polyvinylidene difluoride membrane ( PVDF; Life Technologies ) for 1 . 5 hours at 12V in Towbin buffer . The membrane was blotted with 5% w/v powdered milk in TBST ( 50 mM Tris , 150 mM NaCl , 0 . 5% ( v/v ) Tween 20 ) for 30 minutes at 37°C . The membranes were incubated overnight ( 4°C ) with rabbit preimmune serum ( Day 0 ) at a dilution of 1:1000 in TBST , serum collected 72 days after inoculation ( Day 72 ) at a dilution of 1:1000 in TBST , or affinity purified antibody at a dilution of 1:500 in TBST . The blots were washed extensively with TBST and incubated for 1 h at 22°C with commercially available horseradish peroxidase-conjugated goat anti-rabbit ( dilution factor 1:5000 in TBST ) ( Fisher Scientific , Fair Lawn , NJ , USA ) . After washing with TBST , the membrane was developed using the Enhanced ChemiLuminescence Western blotting detection system ( Thermo Scientific , Hercules , CA , USA ) according to the manufacturer's instructions . Protein was quantified by scanning densitometry ( ImageJ , version 1 . 47 , National Institute of Health , USA ) . SDS-PAGE and Western blotting of cell lysates generated from stressed E . histolytica cells or encysting E . invadens cells were performed as described above . For E . histolytica-derived samples , a mouse anti-actin commercial antibody was used at a dilution of 1:5000 ( Abcam , Cambridge , MA , USA ) to confirm equal load . For E . invadens-derived samples , staining of gels with Bio-Safe Coomassie G-250 Stain ( Bio-Rad Laboratories , Hercules , CA ) was used to confirm equal load . To halt translation , stressed and unstressed E . histolytica trophozoites were treated with cycloheximide ( 100 μg mL-1 ) for 10 min at 37°C . Cells were collected by centrifugation ( 500 x g for 5 min at 4°C ) . The cell pellet was suspended with cold 1X PBS buffer , washed and resuspended in Breaking/Polysome Buffer ( BPB ) ( 10mM Tris-HCl ( pH 7 . 4 ) , 300 mM KCl , 10 mM MgCl2 , 1% ( v/v ) Triton-X-100 , 2 mM DTT , 1 mg/ml heparin , 50 μg/ml cycloheximide , and 0 . 04 units/μl RNase Out ) in the presence of protease inhibitors . Lysis was achieved by passing cells twice through a 27-gauge syringe needle . Lysates were clarified by centrifugation at 14 , 000 × g for 5 min . Samples were loaded onto a 15–45% sucrose gradient in BPB without RNase Out , heparin , DTT or Triton as described previously [53 , 54] . Ultracentrifugation was performed at 230 , 000 x g for 2 h . Gradients were fractionated and the fractions were analyzed for polyribosome abundance by spectrophotometry ( 254 nm ) and the absorbance was corrected for cell count . The E . histolytica eIFα gene is predicted to be intron-less . Therefore , genomic DNA was purified ( Wizard Genomic DNA Purification Kit , Promega ) and the EheIF2α gene was isolated by PCR . During PCR , nucleotides encoding a BglII restriction site , a FLAG tag [19] , and a 5-glycine flexible region were added to the 5’ end of the gene and nucleotides encoding SalI were added to the 3’ end of the gene . Site-directed mutagenesis of the codon for serine ( S ) ( TCA; amino acid position 59 ) to the codon for alanine ( A ) ( GCA ) or aspartic acid ( D ) ( GAT ) was carried out using a PCR-based protocol using the QuikChange Kit ( Stratagene , Santa Clara , CA ) according to manufacturer’s instructions . Successful mutagenesis was confirmed by sequencing . Wildtype and mutated EheIF2α coding sequences were digested with BglII and SalI and ligated into the E . histolytica expression vector , pGIR209 [20] ( gift of Dr . W . A . Petri , University of Virginia , Charlottesville , VA ) , which had been digested with BglII and SalI . This vector allows for the inducible expression of exogenous proteins via the addition of tetracycline to the medium and is co-transfected with a second vector , pGIR308 [20] ( gift of Dr . W . A . Petri , University of Virginia , Charlottesville , VA ) , which encodes the tetracycline repressor . Exponentially growing trophozoites of E . histolytica , harboring pGIR308 , were transfected with the engineered pGIR209 vector as described [21] . As a control , amoebae were also transfected with pGIR209 containing the gene encoding luciferase [20] . Transfected amoebae were maintained by adding 6 μg mL-1 G418 ( pGIR209 ) and 15 μg mL-1 hygromycin ( pGIR308 ) selection agents to the medium . Expression of exogenous proteins was induced by the addition of 5 μg mL-1 tetracycline to the culture medium for 24 to 72 h prior to all studies and confirmed by Western blotting which was performed as described [52] with rabbit anti-luciferase ( Invitrogen ) , anti-FLAG ( Sigma ) , affinity purified anti-eIF2α or affinity purified anti-phospho-eIF2α at dilutions of 1:2500 , 1:5000 , 1:1000 , or 1:1000 , respectively . Viability of the mutants during stress and measurements of polyribosome abundance were carried out as described above after 24 h or 72 h post induction of exogenous protein biosynthesis . SUnSET analysis has been previously used to assess translational machinery in E . histolytica [25 , 26] . To determine if SUnSET could be used to assess protein synthesis in E . histolytica in our hands , we assessed the incorporation of puromycin into wildtype trophozoites . Cells ( 2x106 ) were incubated with 10 μg mL-1 puromycin ( Sigma-Aldrich ) for 15 min before or after incubation with 100 μg mL-1 cycloheximide for 10 min . All incubations were held at 37°C . Cells were then pelleted and proteins were precipitated using 20% ( v/v ) TCA and incubating on ice for 10 min . Proteins were isolated via centrifugation at 2200 x g for 5 min and washed with 5% ( v/v ) TCA . The protein pellet was resuspended in 2X SDS running buffer and incubated in boiling water for 10 min . The lysate was frozen at -80°C until analyzed via Western blot as described above . Mouse anti-puromycin monoclonal antibodies ( Sigma-Aldrich ) were used at a 1:2500 dilution . As a loading control , samples were stained with Bio-Safe Coomassie G-250 Stain . The protocol was repeated in the transgenic cell lines after a 72 h induction period .
All values are given as means ± standard error of at least 3 trials . To compare means , statistical analyses were performed using GraphPad Prism v . 6 . 05 software with a one-way analysis of variance ( ANOVA ) and a Tukey-Kramer multiple-comparison test . In all cases , P values of less than 0 . 001 were considered highly statistically significant , while P values of less than 0 . 01 or 0 . 05 were considered statistically significant .
|
Entamoeba histolytica is the causative agent of amoebic dysentery and liver abscess and is prevalent in underdeveloped countries that lack proper sanitation . Infection is acquired by ingestion of the cyst form in contaminated food or water . During infection , the parasite experiences stress including demanding growth conditions and host immune pressure . Conversion to the infective cyst may be induced by such stress . In other organisms , stress causes a decrease in protein biosynthesis by inducing phosphorylation of eIF2α , which participates in translation initiation . We exposed E . histolytica to six different stress conditions and observed that some of these conditions ( long-term serum starvation , long-term heat shock , and oxidative stress ) induced an increase in the level of phospho-eIF2α . Long-term serum starvation was also accompanied by a decrease in mRNA translation . A cell line expressing a mutant version of eIF2α that behaves as a phosphomimetic exhibited decreased translation and increased survival during long-term serum starvation . Finally , phospho-eIF2α accumulated in cysts of E . invadens , a reptilian pathogen that readily encysts in vitro . Together , these data demonstrate that the eIF2α-dependent stress response system is operational in Entamoeba and may regulate encystation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Statistical",
"analysis"
] |
[
"phosphorylation",
"antimicrobials",
"trophozoites",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"oxidative",
"stress",
"drugs",
"microbiology",
"carbohydrates",
"parasitic",
"protozoans",
"polyribosomes",
"parasitology",
"organic",
"compounds",
"glucose",
"apicomplexa",
"protein",
"expression",
"protozoans",
"tetracyclines",
"antibiotics",
"pharmacology",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"entamoeba",
"histolytica",
"proteins",
"chemistry",
"molecular",
"biology",
"ribosomes",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"organic",
"chemistry",
"monosaccharides",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] |
2016
|
Phosphorylation of Eukaryotic Initiation Factor-2α during Stress and Encystation in Entamoeba Species
|
Gene regulatory networks ( GRNs ) evolve as a result of the coevolutionary processes acting on transcription factors ( TFs ) and the cis-regulatory modules they bind . The zinc-finger TF zelda ( zld ) is essential for the maternal-to-zygotic transition ( MZT ) in Drosophila melanogaster , where it directly binds over thousand cis-regulatory modules to regulate chromatin accessibility . D . melanogaster displays a long germ type of embryonic development , where all segments are simultaneously generated along the whole egg . However , it remains unclear if zld is also involved in the MZT of short-germ insects ( including those from basal lineages ) or in other biological processes . Here we show that zld is an innovation of the Pancrustacea lineage , being absent in more distant arthropods ( e . g . chelicerates ) and other organisms . To better understand zld´s ancestral function , we thoroughly investigated its roles in a short-germ beetle , Tribolium castaneum , using molecular biology and computational approaches . Our results demonstrate roles for zld not only during the MZT , but also in posterior segmentation and patterning of imaginal disc derived structures . Further , we also demonstrate that zld is critical for posterior segmentation in the hemipteran Rhodnius prolixus , indicating this function predates the origin of holometabolous insects and was subsequently lost in long-germ insects . Our results unveil new roles of zld in different biological contexts and suggest that changes in expression of zld ( and probably other major TFs ) are critical in the evolution of insect GRNs .
Gene regulatory networks ( GRNs ) depend on the coevolution of transcription factors ( TFs ) and their relationship with the cis-regulatory modules ( CRMs ) they bind [1 , 2] . In insects , the detailed role of a number of TFs and CRMs have been well-described , particularly during the embryogenesis of the fruit fly Drosophila melanogaster [3] . In metazoans , the period following fertilization is typically characterized by rapid and near-synchronous mitotic divisions and cleavages that occur under conditions of minimal cellular differentiation . Cleavages typically depend on maternally supplied factors and zygotic genome transcription is constrained during this early period of development [4] . A conserved process of metazoan embryogenesis is the maternal-to-zygotic transition ( MZT ) , which is characterized by two critical steps: 1 ) the elimination of a maternal set of mRNAs and proteins and; 2 ) the beginning of zygotic transcription , which leads to the zygotic genomic control of development [5] . In D . melanogaster , the majority of the first set of zygotic transcripts are regulated by zelda ( zld ) [6] , a zinc finger TF with particular affinity for promoter regions containing TAGteam sites—heptamers constituted by CAGGTAG and its variants [7–9] . zld ( Dm-zld , in D . melanogaster ) binding sites have been identified in D . melanogaster embryos ( cycles 8 to 14 ) by chromatin immunoprecipitation coupled with high-throughput sequencing ( ChIP-Seq ) [8] . Dm-zld regulates a large set of genes involved in important processes such as cytoskeleton organization , cellularization , germ band development , pattern formation , sex determination and miRNA biogenesis [6] . Dm-zld was also suggested to participate in larval wing disc development [10] and its overexpression during wing imaginal disc formation led to wing blisters in adults , an indicative of improper adherence of ventral and dorsal wing epithelia [11] . Nevertheless , while zld´s functions have been thoroughly investigated in D . melanogaster MZT , its roles in other organisms and biological processes remain elusive . D . melanogaster displays a long germ type of embryonic development , during which all segments are simultaneously generated along the whole egg . In contrast , short germ insects generate anterior ( e . g . head ) segments early in development , while the remaining segments are patterned from the posterior region , the growth zone ( GZ ) . Since short germ development is considered to be the ancestral mode of development , short-germ insects have been established as developmental model systems [12 , 13] . The short-germ red flour beetle Tribolium castaneum ( Tc ) was the first beetle species to have its genome completely sequenced [14] . T . castaneum displays a short life-cycle and is amenable to gene silencing via RNAi [15] , gene overexpression [16] , specific tissue expression [17] and fluorescence labeling during early development [18] . Several developmental processes that have been investigated in T . castaneum were lost or extensively modified in the D . melanogaster lineage , such as GZ formation [19] , extensive extra-embryonic morphogenesis [20] and the formation of a morphologically complex head during embryogenesis [21] . Early development of T . castaneum is similar to most other insect groups , in which synchronous rounds of division are followed by nuclear migration to the egg cortex and cellularization , forming the so called uniform blastoderm [18 , 22] . Taken together , all the genetic and morphological information on T . castaneum early development , along with the established techniques mentioned above , make this beetle an ideal model to understand the evolution of zld´s function during insect development . In the present work , we provide the first comprehensive analysis of zld orthologs across a wide range of species . Further , we provide functional analysis of a zld ortholog in a non-Drosophillid insect , the short germ beetle T . castaneum ( thenceforth referred as Tc-zld ) . Among our main results are: 1 ) The identification of some previously overlooked conserved domains in Zld; 2 ) An inference of the evolutionary origin of Zld , based on phyletic analysis in various hexapods and crustaceans; 3 ) The identification of a conserved set of 141 putative Tc-zld targets ( i . e . genes with upstream TAGteam sequences ) , enriched in TFs , whose homologs that have been demonstrated to be zld targets in D . melanogaster; 4 ) Identification of key roles played by Tc-zld during the MZT; 5 ) Identification and experimental validation of two new biological roles of zld in T . castaneum: segment generation from the posterior GZ during embryogenesis and postembryonic imaginal disc development; 6 ) Demonstration that zld is also involved in GZ patterning in the hemipteran Rhodnius prolixus , supporting a conserved role in the GZ of a distant short-germ species . Altogether , our results unveil new roles of zld as a pleiotropic TF acting in various developmental processes across distantly-related insects .
While previous studies reported that zld is involved in the MZT of D . melanogaster [6] , its evolutionary history remains unclear . We investigated the phyletic distribution of zld and found a single ortholog in all the inspected insect genomes , including the beetle T . castaneum ( Fig 1; S1 Table ) , indicating that insects are sensitive to increased copy number of this important regulator , which is interesting considering that different TF families are particularly prone to expansions across insect lineages [23] . Further , zld homologs were also found in some ( but not all ) collembolan and crustacean genomes ( Fig 1B; S1 Table ) . The canonical domain architecture of Zld has been reported as comprising a JAZ zinc finger ( Pfam: zf-C2H2_JAZ ) domain and a C-terminal cluster of four DNA binding C2H2 zinc finger domains ( zf-C2H2 ) [6 , 11 , 24] . However , we performed a sensitive and detailed analysis of Zld proteins from multiple species and found other notable conserved protein domains and structural features . Firstly , we found two additional zf-C2H2 domains , N-terminal to the JAZ domain ( Fig 1A ) . The most N-terminal zf-C2H2 is absent or partially eroded ( i . e . without the conserved cysteines and histidines ) in some species ( Fig 1B ) , including D . melanogaster . We define this N-terminal zf as ZF-Novel , since it has not been reported in previous studies . This observation was confirmed by inspecting Dm-zld alternative splicing isoforms . On the other hand , the second N-terminal zf-C2H2 domain is conserved in virtually all extant insects , but absent or degenerate in the other Pancrustaceans ( e . g . Daphnia magna; partially conserved , with one lost cysteine ) ( Fig 1 ) . Further , between this second zf-C2H2 and the JAZ domain , there is a strikingly conserved acidic patch that is characterized by an absolutely conserved motif of the form [DE]I[LW]DLD ( Fig 1 ) , which is predicted to adopt an extended conformation ( using the JPRED software [25] ) amidst surrounding disordered regions . A related conserved acidic motif was also found in the chordate protein CECR2 , C-terminal to the DDT and WHIM motifs , which constitute a helical domain involved in setting the nucleosome spacing in conjunction with the ISWI ATPase during chromatin remodeling [26] . An analogous conserved acidic patch is also seen in the HUN domain which functions as a histones chaperone [27] . Taken together , these observations raise the possibility that this Zld acidic region interacts with positively charged chromatin proteins such as histones . A largely disordered region , between the JAZ finger and the cluster of 4 widely-conserved zf-C2H2 domains , has been shown to be important for Dm-zld transactivation in vitro [24] . Our analysis of evolutionary constraints on the protein sequence also revealed a motif of the form hP[IVM]SxHHHPxRD , which appears to be under selection for retention despite the strong divergence in this region . Hence , it is possible that it specifically plays a role in transactivation . Further , between the two N-terminal zf-C2H2 domains , there is a highly conserved RYHPY motif , which could be involved in nuclear localization ( Fig 1A ) , as predicted for other TFs [28] . Given the conservation of these additional domains , we hypothesize that they also play important roles in Zld functions that were previously attributed exclusively to the C-terminal domains . Aiming to elucidate the origins of zld , we performed extensive sequence searches and were unable to find homologs with the conserved canonical domain architecture outside of Pancrustacea , indicating that zld is an innovation of this lineage . We detected clear zld homologs across insects , including the termite Zootermopsis nevadensis ( Order Isoptera ) , the scarce chaser Ladona fulva ( Order Odonata ) , the mayfly Ephemera danica ( Order Ephemeroptera ) and Machilis hrabei ( Order Archaeognatha ) . zld homologs were also found in crustaceans belonging to different classes ( Daphnia magna , Hyalella azteca and Eurytemora affinis ) , as well as in the collembola Folsomia candida . Curiously , we found no zld in the genome of Orchesella cincta ( collembola ) and Daphnia pulex ( crustacean ) , suggesting either that it is not absolutely conserved outside of insects or missing due to incompleteness of the deposited genomes . Specifically , we carefully searched the genomes of other non-insect arthropods , including chelicerates ( e . g . the tick Ixodes scapularis and the spider Parasteatoda tepidariorum ) , and found no proteins with the canonical Zld domain organization ( Fig 1B ) . General searches on Genbank against non-Pancrustacea arthropod proteins also returned no Zld orthologs . Although BLAST searches with Zld proteins against the nr and refseq databases recovered a number of significant hits in several distant eukaryotes , the similarity is almost always restricted to the C-terminal cluster of four Zf-C2H2 domains , which are very common across several TF families ( e . g . glass , earmuff/fez , senseless/gfi-1 and jim ) . Taken together , our results support the early emergence of zld in the Pancrustacea lineage , with subsequent losses in particular species . Importantly , all the insect genomes we inspected have exactly one zld gene , indicating that this gene became essential in hexapods . Previous studies in D . melanogaster using microarrays and ChIP-Seq revealed that Dm-zld regulates the transcription of hundreds of genes during early embryogenesis [6 , 9 , 29] . In D . melanogaster , enhancers bound by Dm-Zld are characterized by a consensus sequence CAGGTAG ( i . e . TAGteam sequence ) , which is overrepresented in early zygotic activated genes , including TFs involved in AP and DV patterning [7 , 8] . Since the TAGteam motif identified in D . melanogaster is conserved in A . aegypti [30] , we investigated whether we could predict Tc-zld targets by detecting TAGteam motifs in the upstream regions of T . castaneum genes . Firstly , an ab initio approach using DREME [31] was employed to analyze 2kb upstream regions of all T . castaneum protein-coding genes . This analysis uncovered a motif ( i . e . GTAGGTAY ) that is nearly identical to the TAGteam motif ( Fig 2A ) . We used the D . melanogaster genome and experimental data [6 , 9 , 29] to validate our approach and found a significant overlap between experimental and predicted zld targets in D . melanogaster ( Fig 2B ) . The putative T . castaneum motif was then used to screen the T . castaneum genome , resulting in the identification of 3 , 250 putative zld targets , representing ~19% of the T . castaneum genome ( Fig 2C , S2 Table ) . Comparison of the putative Tc-zld targets with 1 , 087 genes regulated by Zld during D . melanogaster MZT [8 , 29] allowed the identification of 141 D . melanogaster genes for which one-to-one orthologs figured among the putative Tc-zld targets ( hypergeometric distribution , P < 4 . 5x10-4; Fig 2C , S3 Table ) . Functional analysis of this gene set using DAVID [32] uncovered the enrichment of important categories , including a highly significant cluster of 26 homeobox TFs ( S4 Table ) and other significant clusters comprising genes involved in regionalization and segment specification , imaginal disc formation and metamorphosis ( S4 Table ) . Interestingly , this gene set included multiple developmental regulators such as anteroposterior ( AP ) , gap , pair-rule , homeotic and dorsoventral ( DV ) genes ( S2 Table ) . Since several of these genes are involved in early developmental processes , we focused our initial analysis of Tc-zld´s function during embryogenesis . Since zld is maternally expressed in the germ line in D . melanogaster , we compared its transcription in T . castaneum female ovaries and carcass by qRT-PCR . T . castaneum early development starts with synchronous divisions during the first three hours of embryogenesis ( at 30°C ) , followed by nuclear migration to the egg cortex and membrane segregation of nuclei into separate cells , approximately 7–8 hours after egg lay [18 , 22] . We found that Tc-zld is highly expressed in the ovaries ( Fig 3A ) , supporting the transcription of Tc-zld in the germ line . The abundance of Tc-zld transcripts is also higher in the first three hours of development than in the next two 3-hour periods ( i . e . 3–6 and 6–9 hours ) ( Fig 3B ) , suggesting that Tc-zld mRNA is maternally provided and degraded after the first 3 hours of development . An antibody against the transcriptionally active form of RNA polymerase II , previously used in other ecdysozoan species [33 , 34] , showed that zygotic transcription in T . castaneum begins between three and six hours of development , shortly after the nuclei have reached the periphery ( Fig 4 ) . In situ hybridization confirmed maternal ubiquitous expression of Tc-zld in the first three hours of development ( Fig 3C and 3D ) and showed a progressive confinement of Tc-zld mRNA to the posterior region of the egg , where the germ rudiment will be formed ( Fig 3E and 3F ) . Between 6 and 9 hours of development , Tc-zld expression is observed at the embryonic tissue ( Fig 3G ) , with higher levels at the posterior region ( Fig 3H ) , where the GZ will generate new segments [19 , 35] . In addition , Tc-zld is also expressed in the ventral serosa , during the serosal window closure ( Fig 3J ) . Later in development , Tc-zld expression is still detected at the GZ ( Fig 3I ) , at the head lobes and a single gnathal segment ( Fig 3K ) and , subsequently , in the nervous system ( Fig 3L ) . Although zld is maternally provided , and zygotically expressed at the neural progenitors of both D . melanogaster and T . castaneum [6 , 11] , biased posterior expression is a feature so far described only in short-germ insects like the latter . It has been shown that injection of dsRNA in T . castaneum females leads to reduced expression of a given gene in the females and their offspring , in a phenomenon called parental RNAi ( pRNAi ) effect [15] . We injected zld dsRNA in females and analyzed Tc-zld transcriptional levels during embryogenesis by qRT-PCR . After zld pRNAi , Tc-zld mRNA levels were reduced in the first two weeks of egg laying , severely impairing larval hatching ( Fig 5A–5C ) . Importantly , identical knockdown phenotypes during embryogenesis were obtained by using a second , non-overlapping , dsRNA construct ( S1 Fig ) . Further , morphological analyses showed that cellularization was severely disrupted in over 50% of the zld dsRNA embryos ( Fig 5D ) , similarly to what was previously reported in D . melanogaster zld mutants [6 , 11] . The remaining zld pRNAi embryos were not severely affected during cellularization and developed beyond that stage . Finally , we also found that some putative conserved target genes were down-regulated in embryos after zld pRNAi , such as the early zygotic genes involved in AP patterning , the serosal gene Tc-zerknullt [36] ( Fig 5E and 5F ) and the gap gene milli-pattes [37] ( Fig 5G and 5H ) . Changes in the spatial distribution of transcripts from predicted dorsoventral target genes were also observed after Tc-zld RNAi . In wild-type ( WT ) embryos , the TF Dorsal forms a dynamic transient gradient , which activates Tc-cactus ( Tc-cact ) and Tc-short-gastrulation ( Tc-sog ) at the ventral region [38 , 39] . After Tc-zld pRNAi , the expression of Tc-cact and Tc-sog is observed in two lateral domains , in contrast to the expression at the single ventral domain in dsneo RNAi embryos ( Fig 5I , 5J , 5K and 5L ) . These results suggest that Tc-zld is required for proper activity of Dorsal at the ventral-most region of the embryo . As discussed above , Tc-zld is highly expressed at the posterior region of the embryo ( Fig 3 ) and likely associated with segmentation and regionalization ( S4 Table ) . Further , several putative Tc-zld targets are involved in posterior segmentation , such as caudal ( Cdx ) , even-skipped ( Eve ) and several Hox genes ( e . g . Ultrabithorax , Abdominal-A and Abdominal-B ) . Tc-eve , for example , is essential for the establishment of a genetic circuit required for posterior segmentation [40] . Interestingly , Tc-zld pRNAi embryos showed a continuous Tc-eve expression domain instead of the typical stripe patterning required for WT segmentation ( Fig 5O and 5P ) . Elegant studies on T . castaneum GZ patterning showed that cell proliferation is not essential for segment generation , which rather occurs by coordinated cell movement and intercalation [19 , 41] . We then evaluated whether Tc-zld regulates genes involved in cell intercalation , such as Toll2 , Toll6 and Toll8 [42 , 43] . Interestingly , Toll7 ( TC004474 ) and Toll8 ( Tollo:TC004898 ) are among the common zld targets conserved between D . melanogaster and T . castaneum ( S3 Table ) . Since Tc-Toll7 is expressed during early segmentation [43] , we compared its expression in dsneo and Tc-zld RNAi embryos . While anterior expression of Tc-Toll7 is apparently unaffected , the striped expression at the posterior region is lost when Tc-zld expression is reduced ( Fig 5M and 5N ) . Further support for the loss of posterior segmentation after Tc-zld RNAi is also provided by the analysis of expression of the segment-polarity gene , Tc-gooseberry ( Tc-gsb ) ( Fig 5Q and 5R ) . In summary , Tc-zld regulates the expression of several genes that are critical for early AP ( zen , mlpt ) and DV ( sog , cact ) patterning and , in a second phase , genes required for posterior elongation ( e . g . Toll7 , gsb ) . While pRNAi diminishes maternal and zygotic expression of Tc-zld in T . castaneum [15] , embryonic dsRNA injections ( eRNAi ) may affect only the zygotic component , since eggs can be injected after the MZT [38 , 44] . To investigate if Tc-zld is specifically required for embryonic posterior patterning , we injected Tc-zld dsRNA in transgenic embryos expressing nuclear GFP ( nGFP ) , as previously described [18 , 19] . Embryonic injections of Tc-zld dsRNA after the MZT ( see Methods for details ) impaired segment generation from the GZ , while dsneo-injected embryos developed like WT ones ( Fig 5S and 5T , S1 and S2 Movies ) . In addition , expression of the predicted target Tc-eve , a key TF involved in GZ patterning [35] , has been largely down-regulated upon Tc-zld eRNAi , as previously observed for zld pRNAi ( Fig 5O and 5P ) . In summary , our results imply that Tc-zld is involved not only in the MZT , early patterning and nervous system formation , as described for D . melanogaster [6] , but also play roles in segment generation from the GZ , a structure found only in embryos of short-germ insects like T . castaneum . Although we demonstrated the involvement of zld in T . castaneum GZ , the conservation of this regulatory mechanism in other species remained unclear . Thus , we sought to analyze the functions of the zld ortholog in the hemipteran R . prolixus ( Rp; Rp-zld gene ) , which is a hemimetabolous insect and lacks the complete metamorphosis present in holometabolous species such as T . castaneum [45 , 46] . Rp-zld knockdown via pRNAi resulted in two types of embryonic phenotypes: 1 ) severe defects in gastrulation and lack of any appendage development; 2 ) embryos that developed only the anterior-most embryonic regions comprising the head , gnathal and thoracic segments ( Fig 6 ) . These results support a role for zld in posterior elongation that was likely present in the common ancestor of hemimetabolous true bugs and holometabolous insects , if not earlier . To our knowledge this is the first direct description of zld function in insects other than D . melanogaster . A recent report showed zld as maternally transcribed in the hymenoptera Apis mellifera ( also a long-germ insect ) , while zygotic transcripts are concentrated at the central region of the embryo during blastodermal stages [47] . Further , during the preparation of this manuscript , zld has been also proposed to be required for MZT and cellularization in the wasp Nasonia vitripennis [48] . DAVID analysis of D . melanogaster orthologs of the putative Tc-zld targets uncovered a cluster of 29 genes involved in imaginal disc development ( S4 Table; GO:0007444 ) . Among these genes are several homeodomain TFs , such as distalless ( Dll ) , Abdominal A ( Abd-A ) , Abdominal B ( Abd-B ) , zen , Engrailed ( En ) , caudal ( cad ) , defective proventriculus ( dve ) , mirror ( mirr ) , araucan ( ara-iroquois ) and Drop ( dr ) , as well as other TFs such as dachsund ( dac ) , taranis ( tara ) and Lim1 , PoxN , kn , sob , drm , Awh , dp . As the first step towards the characterization of the post-embryonic role of zld , we analyzed Tc-zld expression by qRT-PCR in larvae of third ( L3 ) , fifth ( L5 ) and seventh ( L7 ) stages , and first pupal stage ( P1 ) ( Fig 7A–7D ) . Interestingly , Tc-zld expression increases during successive larval stages and sharply decreases after pupal metamorphosis ( Fig 7E ) . This suggests that Tc-zld might be required for late larval stages , which take place during larvae-pupae metamorphosis , such as growth and patterning of structures derived from imaginal discs in D . melanogaster ( e . g . antennae , legs , fore- and hindwings ) [49 , 50] . Further , we found that three out of five predicted Tc-zld targets , namely Dll , Wg and Lim-1 , also displayed an increase in expression during late larval and pupal development ( Fig 7E ) . To investigate zld´s post-embryonic roles , we injected two non-overlapping Tc-zld dsRNA constructs into early ( L3 ) and late larval ( L6 ) stages , as previously described [50] . qRT-PCR confirmed that Tc-zld was down-regulated after dsRNA injection ( Fig 8A ) . Tc-zld dsRNA injections at early larval stages ( L3 ) led to over 50% lethality during pupal stages . Atypical adult pigmentation in the pupal head and reduction in the wing size were observed after early Tc-zld dsRNA injection , suggesting that proper Tc-zld expression is required for metamorphosis and wing growth ( Fig 8B and 8C ) . Interestingly , Tc-zld dsRNA injections at late larval stage ( L6 ) displayed a different phenotype when compared with early larval dsRNA injections ( L3 ) . L6 larvae injected with dszld and dsneo reached adulthood at comparable rates ( Fig 8D ) . Specifically , Tc-zld dsRNA adults showed a series of morphological alterations in tissues undergoing extensive morphological changes during metamorphosis , such as fore- and hindwings , antennae and legs ( Figs 8E , 8F , 9 and 10 ) . The most visible effect of Tc-zld dsRNA beetles was a failure of the forewings ( elytra ) to enclose the hindwings , leading to the exposure of the dorsal abdomen ( Fig 8E and 8F ) . Elytra , which are highly modified beetle forewings , have been proposed to be an important beetle innovation , being required for protection against mechanical stresses , dehydration , and predation [51] . In line with this hypothesis , Tc-zld dsRNA adults with exposed abdomens started to die a few days after metamorphosis , probably due to dehydration . Next , we performed a detailed morphological analysis to investigate if patterning defects resulting from Tc-zld dsRNA occurred in the sclerotized elytra ( forewing ) or in the hindwing ( Fig 9A–9D ) . This analysis showed that the parallel vein pattern of the elytra ( Fig 9A ) is disrupted after Tc-zld dsRNA in comparison to the control ( Fig 9A and 9B ) . Nevertheless , hindwings of Tc-zld dsRNA beetles showed no signs of abnormal venation ( Fig 9C and 9D ) . To address if fore or hindwing shapes have changed upon Tc-zld dsRNA knockdown , we applied the recently developed Proximo-Distal ( PD ) index , a morphometric analysis consisting of the measurement of the wing length and width at two positions [52] . While the PD index values of forewings ( elytra ) of Tc-zld and controls were similar , a slight but significant increase in the hindwing PD index was observed ( Fig 9E ) . Neither fore- nor hindwings length were altered upon Tc-zld knock down ( Fig 9F ) . In conclusion , Tc-zld is required for proper venation pattern , but not shape , of the elytra . Previous analysis of zld expression in D . melanogaster showed expression in wing imaginal discs , particularly where mitotically active cells are located [11] . Moreover , Dm-zld overexpression during wing imaginal disc formation leads to adult wing blisters or tissue loss [10 , 11] . Nevertheless , to our knowledge , our study provides the first direct evidence of zld´s role in insect wing formation . Interestingly , all Tc-zld dsRNA beetles that showed this ‘opened wing’ phenotypes ( Fig 8F ) also displayed defects in the legs and antennae . Antennae and legs share similar developmental GRNs , involving the so called serial homologs . Distaless ( Dll ) , one of the putative Tc-zld targets , is essential for appendage segmentation [53] . Tc-Dll is also expressed during late embryogenesis on the distal part of the leg and , as its name suggests , disruption of its function leads to the absence of distal leg and antennae segmentation [54] . Interestingly , we found that Tc-zld RNAi resulted in a significant decrease of Tc-Dll mRNA levels ( Fig 10A ) , indicating that this gene is indeed downstream of Tc-zld . Insect antennae possess three primary segments: scape , pedicel and flagellum . In T . castaneum , the adult antennae display eleven segments , out of which nine form the flagellum . The three most distal flagellar segments are enlarged and form the club , while the six intermediate flagella are called the funicle ( Fig 10B ) [55] . In mild phenotypes , dszld caused a joint malformation in distalmost flagellar segments resulting in a fusion of the club ( Fig 10C ) . On the other hand , strong Tc-zld dsRNA phenotypes resulted in fusion of the scape and pedicel , leading to severe loss of flagellar joints and formation of a single truncated segment ( Fig 10D ) . In contrast , no differences in scape and pedicel were observed . T . castaneum legs originate during late embryogenesis and can be recognized as a small outgrowth of the body wall , the limb bud . In the adult stage , there are three pairs of segmented legs with six segments: coxa , femur , trochanter , tibia , tarsus and pretarsus [56] . In T . castaneum the tarsus is subdivided in smaller segments ( i . e . the tarsomeres ) , five in prothoracic and mesothoracic legs and four in metathoracic legs . Tarsal segmentation occurs during beetle metamorphosis and this subdivision of the tarsus evolved in the common ancestor of insects , since the tarsus is not subdivided in non-insect hexapods [57] . After Tc-zld RNAi , tarsal segments were absent or fused , resulting in leg shortening; in dsneo insects , legs were identical to that of WT animals ( Fig 10E and 10F ) . This indicates that some of the Tc-zld targets are involved in segment development or joint formation . Interestingly , a large domain of Dll expression is observed in beetle leg tarsus [58] , and Tc-Dll knockdown in beetles also generated legs with tarsomere deletion [58 , 59] . Besides the conservation of zld roles in the MZT , our results uncovered two new biological roles of zld in T . castaneum: regulation of segment generation from the posterior GZ during embryogenesis and; patterning of imaginal disc derived structures . But what do posterior GZ patterning , MZT and imaginal disc development have in common ? All these processes require an accurate temporal and spatial coordination of complex GRNs to properly pattern cell populations . Recently , Dm-zld has been demonstrated to be a key factor in the establishment of the early chromatin architecture , particularly for the formation of topologically associating domain ( TAD ) boundaries [60] . During D . melanogaster MZT , zld resembles a pioneer TF marking the chromatin of earliest expressed genes [8 , 61] . Dm-zld also increases chromatin accessibility of the most important TFs involved in DV and AP patterning ( Dorsal and Bicoid , respectively ) [62–65] . Further , the addition or removal of Dm-zld binding sites influences the timing of activation of Dorsal early zygotic targets in D . melanogaster [62 , 63] , suggesting that Zld acts as a developmental timer . Our results in T . castaneum showed an extensive and ubiquitous maternal contribution of Tc-zld mRNAs , followed by zygotic Tc-zld expression along the embryonic rudiment , particularly at the posterior region ( Fig 3 ) . It is possible that Tc-zld mediates a progressive anteroposterior opening of the chromatin in T . castaneum , shortly before these posterior GZ cells undergo convergent extension movements required for germ band elongation [19 , 41] . Loss of Tc-zld expression might lead to lack of convergent extension due to loss of Toll7 and eve expression and , ultimately , segmentation failure ( Fig 5 ) . Since zld is also important for the development of the posterior region of the hemimetabolous insect R . prolixus ( Fig 6 ) , zld´s role in posterior region dates back at least to the last common ancestor of Paraneoptera . Tc-zld expression was also observed at the ventral serosa ( Fig 3J ) during embryonic stages . While we cannot rule out that Tc-zld plays a role during normal serosal development , lack of segmentation after Tc-zld eRNAi cannot be explained by loss of Tc-zld expression in this tissue , since embryos without serosa , e . g . Tc-zen1 RNAi , do not display phenotypic defects under normal developmental conditions [36 , 66] . Future studies are warranted to determine if Tc-zld plays a specific role in the beetle serosa , particularly by regulating some of its predicted target genes involved in immune responses ( e . g . Toll , cactus ) . Tc-zld might also play a similar role in the regulation of leg and antenna segmentation during metamorphosis ( Fig 10 ) . The number of tarsomere segments in the leg and intermediate funicle are reduced after Tc-zld dsRNA injection , suggesting that the post-embryonic role of Tc-zld in appendage segmentation might also involve the regulation of complex GRNs , as observed in the posterior embryonic region . The evolutionary success of hexapods is attributed to a combination of features: their segmented body plan and jointed appendages , which were inherited from their arthropod ancestor and; wings and holometaboly , two features that arose later in insect evolution [67] . It is interesting to note that the specific Pancrustacea gene zld is required for most of these processes , such as embryonic segment formation , wing ( elytra ) patterning , and appendage ( antennae and leg ) formation during beetle development . Like several other zinc finger proteins that show tremendous lineage-specific diversity in eukaryotes , zld appears to have specifically arisen within Pancrustacea and risen to play an important role as a “master TF” . Hence , future studies focusing on how this TF was integrated to the conserved backdrop of developmental TFs and existing GRNs of arthropods would be of great interest .
T . castaneum beetles were cultivated in whole-wheat flour . For sterilization , the flour was kept for 24h at -20°C and another 24h at 50°C . The beetles were maintained inside plastic boxes of approximately 15x15cm with humidity between 40–80% . Protein sequence analyses were performed using BLAST [68] and PFAM searches using HMMER3 [69] against proteins available on Genbank [70] and Vectorbase [71] . Dm-Zld ( NP_608356 ) was used as initial query for sequence searches . Genomic data from the following genomes were obtained from the Baylor College of Medicine Human Genome Sequencing Center: Eurytemora affinis , Hyalella azteca , Blattella germanica , Catajapyx aquilonaris , Machilis hrabei , Libellula ( Ladona ) fulva and Ephemera danica . BLAST results and domain architectures were manually inspected . In silico searches in the T . castaneum genome for over-represented motifs were performed using the DREME software [72] , part of the MEME suite software toolkit [73] . The motif with highest identity with Dm-Zld binding site ( CAGGTAY ) was compared to previously described motifs in the FlyFactorSurvey [74] database with TOMTOM . Further , this motif was used to scan 2000 bp upstream of all predicted T . castaneum genes with FIMO [14 , 75 , 76] . Upstream regions and orthology information were retrieved from Ensembl Metazoa Biomart ( http://metazoa . ensembl . org/biomart/ ) [77] . Two non-overlapping PCR fragments containing T7 promoter initiation sites at both ends were used as templates for Ambion T7 Megascript Kit ( Cat . No . AM1334 ) following the manufacturer instructions ( for details see S1 Fig ) . The amount and integrity of the dsRNA samples were measured by spectrophotometry and agarose gel electrophoresis , respectively . For parental RNAi ( pRNAi ) analysis , ~0 . 5μl of dsRNA were injected from a solution containing 1μg/μl of dsRNA into adult female beetles [36] . Eggs were collected for four egg lays ( 2 day each ) and zld´s down regulation was estimated by quantitative Real Time PCR ( see below ) . Egg injections were performed as previously described [38 , 44] . Briefly , for the analysis of Tc-zld zygotic role , embryos containing nuclear-localized green fluorescent protein ( GFP ) were collected for one hour and let to develop for an additional three hours ( 30°C ) [19 , 35] . After this period , twenty embryos were dechorionated with bleach ( 2% solution ) , aligned onto a glass slide and covered with Halocarbon oil 700 ( Sigma ) . Embryos were immediately microinjected at the anterior region with zld or neo dsRNA at 1 μg/μL concentration with the help of a Nanoinject II instrument ( Drummond Scientific Company ) . After injection , a single nGFP embryo was photographed every five minutes during the following 16 hours ( 25°C ) in a Leica DMI4000 inverted microscope using a GFP filter . Single photographs were used to generate a movie using Windows Movie Maker ( S1 and S2 Movies ) . Phenotypes of all injected embryos ( neo or zld dsRNA ) were scored at the end of the experiment . Larvae were injected with zld or neo dsRNA as previously described [50] . Knockdown phenotypes in pupae and adult beetles were generated by injection of zld or neo dsRNA solutions at a concentration of 1 μg/μL in the dorsal abdomen of individuals on third and sixth larval instars ( n = 40 ) . Following injection , larvae were reared in flour at 30°C and collected periodically for RNA extraction and phenotype annotation . Adult beetles were then fixed in ethanol 95% overnight for further morphological analysis . Immunostainings have been performed as previously described [38] . Antennae , legs , elytra and wings were dissected using forceps and placed in a petri dish for observation . Phenotypic analyses and documentation were performed under a Leica stereoscope model M205 . The methods for wing and elytra measurement and PD index were performed according to described by [52] . Leica AF Lite software was used for the wing measurements . Image properties were adjusted in Adobe Photoshop CS4 . For experiments using embryos , total RNA was isolated from 100 mg of eggs collected from specific development stages ( 0–3 , 3–6 and 6–9 hours after egg laying ) , ovary and carcass ( whole beetle without ovary ) using Trizol ( Invitrogen ) , according to the manufacturer's instructions . Three independent biological replicates were used for each assay . First strand complementary DNA ( cDNA ) was synthesized from 2 μg of RNA using Superscript III reverse transcriptase ( Invitrogen ) and oligo ( dT ) . The cDNA was used as template for real time qRT-PCR analysis using SYBR green based detection . qRT-PCR reactions were carried out in triplicate , and melting curves were examined to ensure single products . Results were quantified using the ‘‘delta-delta Ct'' method and normalized to rps3 transcript levels , as previously described [78] . Primer sequences used during the study are provided at the supplemental data ( S5 Table ) . zld cDNA sequence was initially identified by BLAST and included in Fig 1 . Parental RNAi against Rp-zld was performed as previously described [46] .
|
Pioneer transcription factors ( TFs ) are considered the first regulators of chromatin accessibility in fruit flies and vertebrates , modulating the expression of a large number of target genes . In fruit flies , zelda resembles a pioneer TF , being essential during early embryogenesis . However , the evolutionary origins and ancestral functions of zelda remain largely unknown . Through a number of gene silencing , microscopy and evolutionary analysis , the present work shows that zelda is an innovation of the Pancrustacea lineage , governing not only the MZT in the short-germ insect Tribolium castaneum , but also posterior segmentation and post-embryonic patterning of imaginal disc derived structures such as wings , legs and antennae . Further , zelda regulation of posterior segmentation predates the origin of insects with complete metamorphosis ( holometabolous ) , as supported by gene silencing experiments in the kissing bug Rhodnius prolixus . We hypothesize that the emergence of zelda contributed to the evolution of gene regulatory networks and new morphological structures of insects .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"&",
"methods"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"legs",
"morphogenic",
"segmentation",
"limbs",
"(anatomy)",
"animals",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"animal",
"anatomy",
"experimental",
"organism",
"systems",
"epigenetics",
"embryos",
"zoology",
"morphogenesis",
"drosophila",
"research",
"and",
"analysis",
"methods",
"embryology",
"musculoskeletal",
"system",
"genetic",
"interference",
"gene",
"expression",
"beetles",
"wings",
"insects",
"arthropoda",
"biochemistry",
"rna",
"anatomy",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2017
|
Evolution and multiple roles of the Pancrustacea specific transcription factor zelda in insects
|
Insertions and deletions ( indels ) are a major source of genetic variation within species and may result in functional changes to coding or regulatory sequences . In this study we report that an indel polymorphism in the 3’ untranslated region ( UTR ) of the metallothionein gene MtnA is associated with gene expression variation in natural populations of Drosophila melanogaster . A derived allele of MtnA with a 49-bp deletion in the 3' UTR segregates at high frequency in populations outside of sub-Saharan Africa . The frequency of the deletion increases with latitude across multiple continents and approaches 100% in northern Europe . Flies with the deletion have more than 4-fold higher MtnA expression than flies with the ancestral sequence . Using reporter gene constructs in transgenic flies , we show that the 3' UTR deletion significantly contributes to the observed expression difference . Population genetic analyses uncovered signatures of a selective sweep in the MtnA region within populations from northern Europe . We also find that the 3’ UTR deletion is associated with increased oxidative stress tolerance . These results suggest that the 3' UTR deletion has been a target of selection for its ability to confer increased levels of MtnA expression in northern European populations , likely due to a local adaptive advantage of increased oxidative stress tolerance .
Natural populations adapt constantly to their changing environments , with alterations in protein sequences and gene expression providing the main sources of variation upon which natural selection can act . At present , understanding how changes in gene expression contribute to adaptation is one of the major challenges in evolutionary genetics . The fruit fly Drosophila melanogaster has populations distributed throughout the world , with environments ranging from tropical to temperate . On the basis of biogeographical , anatomical and population genetic studies , the center of origin of D . melanogaster has been inferred to be in sub-Saharan Africa [1–3] . Several genomic studies concluded that D . melanogaster underwent a population expansion around 60 , 000 years ago within Africa that set the ground for an out-of-Africa expansion 13 , 000–19 , 000 years ago and the subsequent colonization of Europe and Asia 2 , 000–5 , 000 years ago [4–6] . Because the colonization of new habitats requires that species adapt to new environmental conditions , there has been considerable interest in identifying the genetic and phenotypic changes that occurred during the out-of-Africa expansion of D . melanogaster [7–9] . In order to identify genes that differed in expression between a D . melanogaster population from Europe ( the Netherlands ) and one from sub-Saharan Africa ( Zimbabwe ) , whole-transcriptome comparisons were carried out using adult males and females [10 , 11] , as well as the dissected brains and Malpighian tubules of each sex [12 , 13] . These studies identified several hundred genes that were differentially expressed between the two populations and which represent candidates for adaptive regulatory evolution . One of the candidate genes that showed a large difference in expression between populations in the brains of both sexes was the metallothionein ( MT ) gene Metallothionein A ( MtnA ) . MtnA lies on chromosome arm 3R ( Fig 1 ) and belongs to a gene family of five members that also includes MtnB , MtnC , MtnD and MtnE [14 , 15] . Metallothioneins are present in all eukaryotes and have also been identified in some prokaryotes [16] . In general , MTs are cysteine-rich proteins , a feature that makes them thermostable , and have a strong affinity to metal ions , especially zinc and copper ions [17] . Some of the biological functions that have been described for MTs include: sequestration and dispersion of metal ions; zinc and copper homeostasis; regulation of the biosynthesis of zinc metalloproteins , enzymes and zinc dependent transcription factors; and protection against reactive oxygen species , ionizing radiation and metals [18] . In natural isolates of D . melanogaster , increased MtnA expression has been linked to copy number and insertion and deletion ( indel ) variation and is associated with increased tolerance to heavy metals [19 , 20] . In this paper we show that the expression difference of MtnA between a European and a sub-Saharan African population is not associated with copy number variation , but is associated with a derived 49-bp deletion in the MtnA 3’ untranslated region ( UTR ) . Outside of sub-Saharan Africa , the deletion shows a latitudinal cline in frequency across multiple continents , reaching very high frequencies in northern Europe . Using transgenic reporter genes , we show that the indel polymorphism in the 3’ UTR contributes to the expression difference observed between populations . Furthermore , we use hydrogen peroxide tolerance assays to show that the deletion is associated with increased oxidative stress tolerance . Population genetic analyses indicate that MtnA has been the target of positive selection in non-African populations . Taken together , these results suggest that a cis-regulatory polymorphism in the MtnA 3’ UTR has undergone recent positive selection to increase MtnA expression and oxidative stress tolerance in derived northern populations of D . melanogaster .
A previous RNA-seq study of gene expression in the brain found MtnA to have four times higher expression in a European population ( the Netherlands ) than in a sub-Saharan African population ( Zimbabwe ) [12] . Of the members of the Mtn gene family , only MtnA showed high levels of expression and a significant difference in expression between populations ( Fig 2A ) . To confirm this expression difference , we performed qRT-PCR on RNA extracted from dissected brains of flies from each population following the same pooling strategy used previously [12] . With this approach , we found MtnA to have 5-fold higher expression in the European population than in the African population ( Fig 2B ) . The RNA-seq and qRT-PCR analyses were performed on a "per gene" basis and did not discriminate between the two annotated transcripts of MtnA , which differ only in the length of their 3' UTR ( Fig 1 ) . The MtnA-RA transcript completely overlaps with that of MtnA-RB and contains no unique sequence . The MtnA-RB transcript , however , contains an extra 371 bp at the 3' end that can be used to assess isoform-specific expression . Using RNA-seq data [12] , we found that the MtnA-RB isoform represents only a small proportion of total MtnA expression ( 1 . 50% in the European population and 0 . 13% in the African population ) . Thus , almost all of the observed expression difference in MtnA can be attributed to the MtnA-RA isoform . Although present at very low levels , the MtnA-RB transcript showed much higher expression ( 50-fold ) in Europe than in Africa ( S1 Table ) . Previous studies found copy number variation ( CNV ) for MtnA in natural isolates of D . melanogaster and showed that an increase in copy number was associated with higher MtnA expression [19 , 20] . To determine if CNV could explain the observed expression difference between the European and the African populations , we assayed MtnA copy number in flies of both populations by quantitative PCR . We found no evidence for CNV within or between the populations ( Fig 3 ) . In both populations , MtnA copy number was equal to that of the control single-copy gene RpL32 and was about half that of the nearly-identical paralogs AttA and AttB [21] , which can be co-amplified by the same PCR primers and serve as a positive control . These results indicate that CNV cannot account for the observed variation in MtnA gene expression . To identify cis-regulatory variants that might be responsible for the difference in MtnA expression between European and African flies , we sequenced a 6-kb region encompassing the MtnA transcriptional unit ( Fig 1 ) in 12 lines from the Netherlands ( NL ) and 11 lines from Zimbabwe ( ZK ) . In addition , we quantified MtnA expression in a subset of eight lines from each population in both the brain and the gut by qRT-PCR . Across the 6-kb region , only a polymorphic 49-bp indel and a linked single nucleotide polymorphism ( SNP ) in the MtnA 3’ UTR showed a large difference in frequency between the populations , being this deletion present in 10 of the 12 European lines , but absent in Africa ( Fig 4A ) . This indel was previously observed to segregate in natural populations from North America [20] . A comparison with three outgroup species ( D . sechellia , D . simulans , and D . yakuba ) indicated that the deletion was the derived variant . The qRT-PCR data revealed that the two European lines that lacked the deletion had MtnA expression that was similar to that of the African lines , but much lower than the other European lines . This result held for both brain and gut expression . Taken together , these results suggest that the 3' UTR polymorphism contributes to MtnA expression variation in natural populations . Furthermore , the expression variation is not limited to the brain , but shows a correlated response in at least one other tissue ( Fig 4B ) . To test if the 49-bp deletion in the MtnA 3' UTR has an effect on gene expression , we designed expression constructs in which the MtnA promoter was placed upstream of either a green fluorescent protein ( GFP ) or lacZ reporter gene . Two versions of each reporter gene were made , one with the ancestral MtnA 3' UTR sequence and one with the derived MtnA 3' UTR sequence , which has the 49-bp deletion ( Fig 5A ) . The reporter genes were then introduced into the D . melanogaster genome by PhiC31 site-specific integration [22 , 23] . Our analysis of MtnA expression in the brain and gut indicated that the difference in expression observed between African and European populations is not brain-specific ( Fig 4B ) . This is further supported by the expression of the reporter gene constructs . For the GFP reporter gene , the presence of the 3’ UTR deletion led to increased expression in both the brain and body ( Fig 5B ) , with the difference in expression being 2 . 3-fold and 1 . 75-fold , respectively . A similar result was found for the lacZ reporter gene , where the 3’ UTR deletion led to 1 . 7-fold and 1 . 4-fold higher expression in the head and gut , respectively ( Fig 5C ) . MtnA shows high expression in most D . melanogaster organs , including the fat body , digestive system , Malpighian tubule , and brain [24] . Although it has been documented that MtnA and its paralogs are involved in heavy metal homeostasis and tolerance , it is poorly understood which other functions MtnA might have and in which cells it is expressed . To get a more detailed picture of MtnA expression in the brain , we examined the expression of the GFP reporter gene by confocal imaging of dissected brain tissue ( Fig 6 ) . GFP expression driven by the MtnA promoter is evident in cells that form a mesh-like structure surrounding the brain and in between the neuropiles ( Fig 6 ) . MtnA does not appear to be expressed at a discernible level in neurons , as the cells expressing GFP do not have dendrites or axonal processes . The shape and localization of the cells expressing GFP in the brain suggest that they are glia , which provide neurons with developmental , structural and trophic support as well as with protection against toxic elements [25–27] . In a genome-wide expression profiling study it was found that MtnA is expressed in the astrocyte glial cells of larvae and adults of D . melanogaster [28] . Although we cannot be certain that MtnA expression is limited to the glia in the brain , our results provide direct evidence that MtnA is expressed in cell types other than the copper cells of the midgut and Malpighian tubules , as previously reported [29] . To better characterize the geographical distribution of the indel polymorphism in the MtnA 3' UTR , we used a PCR-based assay to screen ten additional D . melanogaster populations across a latitudinal range spanning from tropical sub-Saharan Africa to northern Europe ( Table 1 ) . We found that the deletion was at very low frequency in sub-Saharan Africa , but nearly fixed in populations from northern Europe . This suggests that , at least outside of the ancestral species range , there is a latitudinal cline in the deletion frequency . Indeed , when the sub-Saharan populations are excluded , there is a highly significant correlation between latitude and deletion frequency ( linear regression; R = 0 . 95 , P = 0 . 0004 ) . This correlation still holds when the sub-Saharan populations are included ( using the absolute value of latitude ) , but is weaker ( R = 0 . 80 , P = 0 . 001 ) . To investigate if the clinal distribution of the MtnA 3’ UTR deletion is present on other continents , we analyzed pooled sequencing ( pool-seq ) data from North America and Australia [30 , 31] . In North America , there is a significant correlation between latitude and deletion frequency ( R = 0 . 94 , P = 0 . 005 ) ( Table 2 ) . A similar pattern was seen in Australia , although data from only two populations were available . The deletion is at a frequency of 42% in Queensland ( latitude 16 S ) and 61% is Tasmania ( latitude 42 S ) . The difference in deletion frequency between the two populations is significant ( Fisher’s exact test , P = 0 . 02 ) . To test for a history of positive selection at the MtnA locus , we performed a population genetic analysis of the 6-kb MtnA region in the original European ( the Netherlands ) and African ( Zimbabwe ) population samples . In addition , we sequenced this region in 12 lines of a Swedish population , in which the 49-bp 3' UTR deletion was at a frequency of 100% ( Table 1 ) . Across the entire region , the Zimbabwean population showed the highest nucleotide diversity , having 1 . 43- and 2 . 50-fold higher values of π than the Dutch and Swedish populations , respectively ( Table 3 ) . Tajima’s D was negative in all three populations , and was significantly negative in both Zimbabwe and the Netherlands ( Table 3 ) . This could reflect a history of past positive or negative selection at this locus , but could also be caused by demographic factors , such as population expansion . A sliding window analysis was performed to determine the distribution of nucleotide diversity ( θ ) ( Fig 7A ) and population differentiation ( Fst ) ( Fig 7B ) across the MtnA region . The region flanking the 3’ UTR indel polymorphism showed very low sequence variation in Zimbabwe and Sweden , but higher variation in the Netherlands . This pattern is due to the fact that the ancestral state of the indel polymorphism is fixed in the Zimbabwean population and the derived state is fixed in the Swedish population . In the Dutch population , the MtnA 3’ UTR is polymorphic for the deletion ( two of the 12 lines have the ancestral state ) . This leads to higher nucleotide diversity than in the Swedish population , because the ancestral , non-deletion alleles contain more SNPs than the derived , deletion alleles . On average , Sweden and Zimbabwe showed the greatest population differentiation , with Fst reaching a peak in the 3’ UTR of MtnA , whereas values of Fst were lowest for the comparison of the Dutch and Swedish populations , indicating that there is very little differentiation between them ( Fig 7b ) . If positive selection has favored the derived MtnA allele ( with the 49-bp 3' UTR deletion ) in northern populations , then in this region of the genome one would expect there to be less variation among chromosomes containing the deletion than among those with the ancestral form of the allele . Indeed , this is what we observe in the Netherlands , where both alleles are segregating . Across the 6-kb region , there are 41 segregating sites within the Dutch population ( Table 3 ) . Among the 10 chromosomes with the deletion , there are 18 segregating sites , while between the two chromosomes lacking the deletion there are 23 segregating sites . This indicates that chromosomes with the deletion , which are in high frequency , shared a much more recent common ancestor . To test if this pattern differs from that expected under neutral evolution , we performed the Hudson's haplotype test ( HHT ) [36] using three different demographic models of the D . melanogaster out-of-Africa bottleneck for neutral simulations . Under the model of Werzner et al . [6] , HHT was significant ( P = 0 . 031 ) . Under the models of Thornton and Andolfatto [35] and Duchen et al . [5] , HHT was marginally significant ( P = 0 . 076 and P = 0 . 094 , respectively ) . These results suggest that neutral evolution and demography are unlikely to explain the observed patterns of DNA sequence variation . To further test if the MtnA locus has experienced recent positive selection in northern Europe , we used the composite likelihood ratio ( CLR ) test to calculate the likelihood of a selective sweep at a given position in the genome , taking into account the recombination rate , the effective population size , and the selection coefficient of the selected mutation [37 , 38] . Within the Dutch population , the CLR statistic shows a peak in the region just adjacent to the MtnA 3' UTR deletion ( Fig 7C ) . This peak was significant when the demographic models of Duchen et al . [5] , Werzner et al . [6] , and Thornton and Adolfatto [35] were used for neutral simulations , which provides compelling evidence for a recent selective sweep at the MtnA locus in the Netherlands population . A similar result was obtained for the Swedish population ( Fig 7D ) , where the CLR statistic was above the 5% significance threshold determined from all three of the bottleneck models , suggesting that the selective sweep was not limited to a single population , but instead affected multiple European populations . To test the possibility that the deletion in the MtnA 3’ UTR might have risen to high frequency as a result of hitchhiking with another linked polymorphism , we examined linkage disequilibrium ( LD ) across a 100 kb region flanking the MtnA locus in the Netherlands population ( S1 Fig ) . The degree of linkage disequilibrium , r2 [39] , was calculated between all pairs of SNPs present in the 100 kb region , excluding singletons . The SNP corresponding to the indel polymorphism ( Fig 4a ) , position 53 of the linkage disequilibrium matrix , is not in significant LD with any of the 94 SNPs present along the 100 kb region analyzed ( S1 Fig ) . These results indicate that the high frequency of the MtnA 3’ UTR deletion cannot be explained by linkage with another positively selected locus . MtnA expression has been linked to increased heavy metal tolerance [19 , 20 , 40] and metallothioneins in general have been associated with protection against oxidative stress [18 , 41] . To test if MtnA plays a role in oxidative stress and/or heavy metal tolerance , we used RNA interference ( RNAi ) to knockdown MtnA expression; these flies , along with their respective controls , were exposed to either hydrogen peroxide or copper sulfate . A knockdown in MtnA expression was significantly associated with increased mortality in the presence of hydrogen peroxide ( P < 0 . 001; Fig 8A ) and copper sulphate ( P = 0 . 026; Fig 9A and 9B ) , although for the latter , this decrease was only significant in females . To further test if the deletion in the MtnA 3’ UTR could be associated with an increase in oxidative stress and/or heavy metal tolerance , a subset of D . melanogaster lines from the Dutch and Malaysian populations , either with or without the deletion , were exposed to hydrogen peroxide and copper sulfate . The 3’ UTR deletion was associated with a significant increase in survival in the presence of hydrogen peroxide in both the Dutch ( P = 0 . 001; Fig 8B ) and Malaysian ( P = 0 . 001; Fig 8B ) populations . The 3’ UTR deletion had no significant effect on survival in the presence of copper sulfate in Dutch and Malaysian females ( P = 0 . 976 and P = 0 . 732 respectively; Fig 9D ) or males ( P = 0 . 578 and P = 0 . 904 respectively; Fig 9C ) . Thus , the deletion in the MtnA 3’ UTR was associated with increased oxidative stress tolerance , but not increased heavy metal tolerance .
Differential expression of MtnA between a European and an African population of D . melanogaster was first detected in a brain-specific RNA-seq analysis [12] . In the present study , we confirm this inter-population expression difference by qRT-PCR and show that it is associated with an indel polymorphism in the MtnA 3’ UTR . We also perform reporter gene experiments to demonstrate that a large proportion of the expression difference can be attributed to this indel polymorphism . The ancestral state of the 3’ UTR contains a 49-bp sequence that is deleted in a derived allele that is present in worldwide populations . The deletion is nearly absent from sub-Saharan Africa , but present in frequencies >80% in northern Europe ( Table 1 ) . The deletion is present at intermediate frequency in Egypt ( 60% ) , Cyprus ( 65% ) and Malaysia ( 45% ) . These findings suggest that positive selection has favored the 3' UTR deletion , at least within northern European populations . This interpretation is supported by population genetic analyses that indicate a recent selective sweep at the MtnA locus in populations from the Netherlands and Sweden ( Fig 7 ) . Furthermore , a clinal relationship between deletion frequency and latitude is also seen in North America and Australia , suggesting that there is a common selection gradient affecting all populations outside of sub-Saharan Africa . Although chromosome arm 3R is known to harbor inversion polymorphisms that vary in frequency with latitude in cosmopolitan populations [42] , we can rule out linkage to a segregating inversion as a cause for the clinal pattern seen for the MtnA 3’ UTR deletion . A previous analysis of the same Dutch population used in our study found that only one of the isofemale lines harbored an inversion on 3R , In ( 3R ) P [43] . This was line NL13 , which is one of the 10 lines with the MtnA 3’ UTR deletion ( Fig 4A ) . Thus , there is no evidence for linkage between the inversion and the deletion . Moreover , the MtnA gene lies 7 Mb outside of the nearest breakpoint of In ( 3R ) P . Using hydrogen peroxide tolerance assays , we found evidence that knocking down MtnA expression decreases oxidative stress tolerance ( Fig 8B ) . The association of the deletion in the MtnA 3’ UTR with increased survival in the presence of hydrogen peroxide ( Fig 8A ) suggests that the deletion has been selectively favored in some environments because it confers increased tolerance to oxidative stress . While cytotoxic reactive oxygen species ( ROSs ) are generated by natural metabolic processes , they can also be introduced via abiotic factors in the environment , such as radiation , UV light or exposure to toxins . The significant correlation between the frequency of the 3' UTR deletion and latitude , coupled with its association with increased oxidative stress tolerance suggests that environmentally induced oxidative stress may vary clinally , with greater stress in northern European environments . Regulation of the oxidative stress response usually occurs via upregulation of antioxidant protective enzymes in response to the binding of a cis-acting antioxidant-responsive element ( ARE ) , which contains a characteristic sequence to which stress-activated transcription factors can bind [41] . A recent example of adaptation to oxidative stress in Drosophila is the insertion of the Bari-Jheh transposable element into the intergenic region of Juvenile Hormone Epoxy Hydrolase ( Jheh ) genes , which adds additional AREs that upregulate two downstream Jheh genes and was associated with increased oxidative stress tolerance [44] . Interestingly , the Bari-Jheh insertion also shows evidence for a partial selective sweep in non-African D . melanogaster [45] , suggesting that oxidative stress may have imposed an important selective constraint on the colonization of Europe . However , the MtnA 3’ UTR deletion cannot mediate its associated increase in oxidative stress tolerance in a similar way , since it does not add any new AREs . Due to their high inducibility in response to heavy metals , metallothioneins have traditionally been thought to play a role as detoxifiers specifically of heavy metals . However , this view has come into question recently , and metallothioneins are now thought to be a part of the general stress response and may function as scavengers of free radicals [41] . The association of the MtnA 3’UTR deletion with increased oxidative stress tolerance ( Fig 8A ) is in line with this more recent view of the role of metallothioneins , while the observed increased mortality after copper exposure in females in which MtnA expression has been knocked down ( Fig 9D ) is in keeping with the more traditional view . However , we found no association between the presence of the deletion and copper tolerance . This may be because the RNAi knockdown results in an MtnA expression level that is much lower than that of naturally occurring alleles , and copper tolerance is only affected when MtnA expression falls below a minimal threshold . The precise mechanisms of how metallothioneins interact with other metal processing systems after their initial binding and help remove excess of heavy metals , remain unclear [41] . At present , the mechanism by which the 3' UTR deletion affects MtnA gene expression is unknown . Although the deletion appears to have an effect on the usage of the MtnA-RB transcript isoform ( S1 Table ) , this isoform is too rare ( <2% of all MtnA transcripts ) to account for the observed 4-fold difference in MtnA expression . Another possibility is that the deleted 3' UTR region contains one or more binding sites for a microRNA ( miRNA ) . miRNAs are short , non-coding RNAs that modulate the expression of genes by inhibiting transcription or inducing mRNA degradation [46] . They are known to bind to a seed region that consists of 6–8 nucleotides in the 3’ UTR of their target mRNA . Post-transcriptional gene expression regulation by miRNAs can result in the fine-tuned regulation of a specific transcript or can cause the complete silencing of a gene in a particular tissue or developmental stage [46–48] . To identify miRNAs that might bind specifically to the 49-bp sequence present in the ancestral form of the MtnA 3’ UTR , we used the UTR predictor [49] . The UTR predictor takes into account the three-dimensional structure of the miRNA and the 3’ UTR , as well as the energetic stability of the miRNA-3’ UTR base-pair binding . The score given by the UTR predictor is an energetic score , with the most negative scores indicating the most probable interactions . Our analysis of the MtnA 3' UTR identified five candidate miRNAs with scores below -6 that had predicted binding sites overlapping with the 49-bp indel region ( Table 4 ) . These candidates should serve as a good starting point for future functional tests of putative miRNA-3' UTR interactions . Genetic variation provides the substrate upon which natural selection acts , resulting in an increase in the frequency of alleles that are beneficial in a given environment . Because changes in gene expression , especially those caused by variation in cis-regulatory elements , are predicted to have fewer pleiotropic effects than changes occurring within coding regions , it has been proposed that they are the most frequent targets of positive selection [50–52] . In contrast to structural changes in protein sequences , changes in gene expression can be specific to a particular a tissue or developmental stage . Our results indicate that the observed variation in MtnA expression is not specific to the brain , as a similar expression pattern is also seen in the gut ( Fig 4 ) . This suggests that the 3' UTR deletion has a general effect on MtnA expression , which is present at high levels in almost all organs of D . melanogaster [24] . However , tissue-specific effects of the difference in MtnA expression cannot be ruled out . As shown in Fig 6 , GFP expression driven by the MtnA promoter in the brain is limited to what seems only one cell type , which according to their morphological and anatomical characteristics , could correspond to glia . It has been reported that glia cells protect neurons and other brain cells from ROS damage caused by oxidative stress [53 , 54] and the fact that MtnA has been found to be expressed in the astrocyte glial cells in larva and adult flies [28] , suggests that MtnA expression in glia could serve as neuronal protection against environmental factors , such as exposure to xenobiotics , that trigger an oxidative stress response [29 , 55–57] . Our functional experiments showing an association between genetic variation in MtnA and oxidative stress tolerance are consistent with MtnA expression in glia providing protection against oxidative stress , which may be especially important in the brain , as neurons are highly susceptible to ROS damage .
This study used isofemale lines from 12 populations of D . melanogaster , including: Zimbabwe ( Lake Kariba ) , Zambia ( Lake Kariba ) , Rwanda ( Gikongoro ) , Cameroon ( Oku ) , Egypt ( Cairo ) , Cyprus ( Nicosia ) , Malaysia ( Kuala Lumpur ) , France ( Lyon ) , Germany ( Munich ) , the Netherlands ( Leiden ) , Denmark ( Aarhus ) and Sweden ( Umeå ) . The lines from Zimbabwe and the Netherlands were the same as those used in previous expression studies [10–12] . Flies from Germany were collected from different locations in the greater Munich area . Flies from Cyprus were collected from a single location near Nicosia . Flies from Denmark were kindly provided by Volker Loeschcke ( Aarhus University ) . Flies from Sweden and Malaysia were kindly provided by Ricardo Wilches and Wolfgang Stephan ( University of Munich ) . The remaining fly lines were collected as part of the Drosophila Population Genomics Project [8] and were kindly provided by John Pool and Charles Langley ( University of California , Davis ) . Flies expressing hairpin RNA targeted against MtnA mRNA under the control of the GAL4/UAS system ( RNAi-MtnA; transformant ID: 105011 ) and the host line used in their creation ( control; transformant ID: 60100 ) were obtained from the Vienna Drosophila Stock Center [58] Act5C/Cyo flies expressing GAL4 under the control of an Act5C driver were kindly provided by Ilona Grunwald Kadow . For tolerance assays , Act5C/Cyo females were crossed to RNAi-MtnA and control males and the progeny ( RNAi-MtnA/Act5C-GAL4 and control/Act5C-GAL4 ) were used in tolerance assays . Using qRT-PCR as described below , MtnA expression was confirmed to be knocked down by 90 . 03% in males and 87 . 58% in females in RNAi-MtnA/Act5C-GAL4 flies in comparison to control/Act5C-GAL4 . Flies were maintained on standard cornmeal-molasses medium at a constant temperature of 22° with a 14 hour light/10 hour dark cycle . Validation of the MtnA expression results obtained from brain RNA-seq data [12] was performed by qRT-PCR using TaqMan probes ( Applied Biosystems , Foster City , California , USA ) . For population-level comparisons , six brains were dissected from males and females of each of the 11 lines from Zimbabwe ( ZK84 , ZK95 , ZK131 , ZK145 , ZK157 , ZK186 , ZK191 , ZK229 , ZK377 , ZK384 , ZK398 ) and five brains were dissected from males and females of each of the 12 lines from the Netherlands ( NL01 , NL02 , NL11 , NL12 , NL13 , NL14 , NL15 , NL16 , NL17 , NL18 , NL19 , NL20 ) . The dissected brains of each population and sex were pooled following the RNA-seq strategy previously described [12] . The above procedure was repeated in two biological replicates for each sex and population . To compare the MtnA expression of individual lines within populations , subsets of eight lines were chosen from Zimbabwe ( ZK84 , ZK95 , ZK131 , ZK145 , ZK157 , ZK186 , ZK377 , ZK384 ) and the Netherlands ( NL01 , NL02 , NL11 , NL12 , NL15 , NL16 , NL17 , NL18 ) . Thirty whole brains and digestive tracts ( from foregut to hindgut ) were dissected per line . Two biological replicates of each line ( each consisting of 30 brains or guts ) were processed . Tissue was dissected from flies 4–6 days old in 1X PBS ( phosphate buffered saline ) . The tissue was stored in RNAlater ( Life Technologies , Carlsbad , CA , USA ) at -80° until RNA extraction . Total RNA extraction and DNase I digestion was performed using the MasterPure RNA Purification Kit ( Epicentre , Madison , WI , USA ) . One microgram of total RNA was reverse transcribed using random primers and SuperScript II reverse transcriptase ( Life Technologies ) following the manufacturer’s instructions . TaqMan gene expression assays ( Applied Biosystems ) were used for MtnA ( Dm02362764_s1 ) and RpL32 ( Dm02151827_g1 ) . qRT-PCR was performed using a Real-Time thermal cycler CFX96 ( Bio-Rad , Hercules , CA , USA ) . Two biological replicates , each with two technical replicates , were processed for each sample . The ΔΔCt method was used to compute the normalized expression of MtnA using the ribosomal protein gene RpL32 as the reference [59] . The paralogous genes AttacinA ( AttA ) and AttacinB ( AttB ) were used as positive controls for CNV assays , because they share 97% nucleotide identity [21] and can be co-amplified with the same primer set . The sequences for AttA and AttB were downloaded from FlyBase [60] and aligned using the ClustalW2 algorithm implemented in SeaView ( version 4 ) [61] . Primers were designed for the second coding exon , where the nucleotide identity of AttA and AttB is 100% . The primer sequences were as follows: forward ( 5’-GGTGCCTCTTTGACCAAAAC-3’ ) and reverse ( 5’-CCAGATTGTGTCTGCCATTG-3’ ) . The ribosomal protein gene RpL32 , which is not known to show CNV , was used as a negative control . The RpL32-specific primers were: forward ( 5’-GACAATCTCCTTGCGCTTCT-3’ ) and reverse ( 5’-AGCTGGAGGTCCTGCTCAT-3’ ) . The primers specific for MtnA were: forward ( 5’-CACTTGACCATCCCATTTCC-3’ ) and reverse ( 5’-GGTCTGCGGCATTCTAGGT-3’ ) . CNV was assessed among 12 lines from the Netherlands and 11 lines from Zimbabwe . Individual DNA extractions were performed separately for three flies of each line and copy number was assessed individually for each fly . Genomic DNA was extracted using the MasterPure DNA Purification Kit ( Epicentre ) . The assessment of CNV from genomic DNA was done with iQ SYBR Green Supermix ( Bio-Rad ) following the manufacturer’s instructions . CNV assays were performed using a Real-Time thermal cycler CFX96 ( Bio-Rad ) . The relative copy numbers of MtnA and AttA/AttB were obtained by the ΔCt method using RpL32 as the reference gene . Approximately 6 kb of the MtnA genomic region , spanning from the second intron of CG12947 to the 3’ UTR of CG8500 ( genome coordinates 3R: 5 , 606 , 733–5 , 612 , 630 ) , were sequenced in 12 Dutch , 11 Zimbabwean and 12 Swedish lines ( Fig 1 ) . The following primer pairs were used ( all 5’ to 3’ ) : GATGGTGGAATACCCTTTGC and AAAGCGGGTTTACCAGTGTG; GTTGGCCTGGCTTAATAACG and ACTGGCACTGGAGCTGTTTC; GCTCTTGCTAGCCATTCTGG and AGAACCCGGCATATAAACGA; GATATGCCCACACCCATACC and GTAGAGGCGCTGCATCTTGT; CACTTGACCATCCCATTTCC and CAAGTCCCCAAAGTGGAGAA; CTTGATTTTGCTGCTGACCA and ATCGCCACGATTATGATTGC; CAGGACAATCAAGCGGAAGT and TTATGAAGCGCAGCACCAGT; GACCCACTCGAATCCGTATC and TGCTTCTTGGTGTCCAGTTG . PCR products were purified with ExoSAP-IT ( Affymetrix , Santa Clara , CA , USA ) and sequenced using BigDye chemistry on a 3730 automated sequencer ( Applied Biosystems ) . Trace files were edited using Sequencher 4 . 9 ( Gene Codes Corporation , Ann Arbor , MI , USA ) and a multiple sequence alignment was generated using the ClustalW2 algorithm in SeaView ( version 4 ) [61] . All sequences have been submitted to GenBank/EMBL under the accession numbers KT008059–KT008093 . For individual flies of the isofemale lines described above , the presence or absence of the MtnA 3' UTR deletion was assessed by performing a two-step PCR ( 35 cycles of 98° for 5 sec . and 60° for 10 sec . ) using the following primers: forward ( 5’-GCCGCAGACCAATTGATTA-3’ ) and reverse ( 5’-TTCTTTCCAGGATGCAAATG-3’ ) . The frequency of the deletion was estimated on an allelic basis , as heterozygous individuals were detected in some populations . Binomial 95% confidence intervals were calculated for the frequency of the deletion using the probit method implemented in R [62] . The strength and significance of the correlation between the frequency of the deletion and latitude was determined using linear regression . To determine the frequency of the MtnA 3’ UTR deletion on other continents , raw pool-seq reads from North America [30] and Australia [31] were downloaded from the National Center for Biotechnology Information ( NCBI ) short read archive ( SRA ) . The reads were mapped to either the ancestral or derived ( with 49-bp deletion ) version of the MtnA 3’ UTR using NextGenMap [63] . Only reads spanning the site of the indel were considered informative . The deletion frequency was estimated as the proportion of informative reads that matched the deletion allele . The 95% confidence interval was estimated using the probit method in R [55] . To test whether the indel polymorphism found in the MtnA 3’ UTR can account for the difference in expression observed between the Dutch and the Zimbabwean populations , we constructed transgenic flies using the phiC31 transgenesis system [23] . Two expression vectors containing a green fluorescent protein ( GFP ) reporter gene were constructed . MtnA 3’ UTR sequences from the Netherlands ( line NL20 ) and Zimbabwe ( line ZK84 ) , corresponding to chromosome arm 3R coordinates 5 , 607 , 448–5 , 611 , 691 , were PCR-amplified with forward ( 5’-TTTCCTCGAACTTGTTCACTTG -3’ ) and reverse ( 5’- GCCCGATGTGACTAGCTCTT -3’ ) primers and cloned into the pCR2 . 1-TOPO vector ( Invitrogen ) . The promoter region of MtnA ( corresponding to genome coordinates 3R: 5 , 607 , 983–5 , 612 , 438 ) , which is identical in the Dutch and the Zimbabwean populations , was also PCR amplified and cloned separately into the pCR2 . 1-TOPO vector using forward ( 5’-GCCGCAGACCAATTGATTA-3’ ) and reverse ( 5’-TTCTTTCCAGGATGCAAATG-3’ ) primers . To generate the GFP expression construct , the MtnA promoter was excised with EcoRI and ligated into the EcoRI site at the 5’ end of GFP in the plasmid pRSET/EmGPP ( Invitrogen ) . Using AvaI and XbaI , the fragment containing the MtnA promoter and GFP was excised from the pRSET/EmGPP plasmid and ligated into the AvaI–XbaI sites proximal to the MtnA 3’ UTR in the pCR2 . 1-TOPO vector . The whole construct ( promoter + GFP + 3’ UTR ) was then excised with XbaI and KpnI and ligated into the XbaI–KpnI sites of the pattB integration vector [23] . For the lacZ constructs , the MtnA promoter was excised from the pCR2 . 1-TOPO vector with EcoRI and ligated into the EcoRI site 5’ of the lacZ coding sequence in the pCMV-SPORT-βgal plasmid ( Life Technologies ) . PCR primers with overhangs containing restriction sites for XhoI and XbaI ( forward 5’- GGTCCGACTCGAGGCGAAATACGGGCAGACATG -3’ and reverse 5’- GGTGCTCTAGAGCTCCATAGAAGACACCGGGAC -3’ ) were used to amplify the MtnA promoter/lacZ fragment and the product was ligated into the XhoI–XbaI sites just upstream of the MtnA 3’ UTR fragment in the pCR2 . 1-TOPO vector . Finally , the whole construct was excised using XbaI and KpnI and ligated into the XbaI–KpnI sites of the pattB vector ( Fig 5 ) . PhiC31 site-specific transgenesis was used to generate flies that differed only in the presence or the absence of the 49–bp sequence in the 3’ UTR of the reporter gene . The M{vas-int . Dm}ZH-2A , M{3xP3-RFP . attP}ZH-51D line was used for embryo microinjections . Microinjection and screening for transformants were carried out by Fly Facility ( Clermont-Ferrand Cedex , France ) and Rainbow Transgenic Flies ( Camarillo , CA , USA ) . The successfully transformed flies were crossed to a yellow , white ( yw ) strain for two generations to eliminate the integrase . Brain tissue was dissected in ice-cold 1X PBS and fixed with PLP ( 8% paraformaldehyde in NaOH and PBS with lysine ( 1 ) -HCl ) for one hour at room temperature as described in [65] . After fixation , the tissue was washed twice for 15 minutes with PBS-0 . 5% Triton X and then incubated for one hour in blocking solution ( 20% donkey serum , 0 . 5% Triton X in PBS ) at room temperature . The primary antibody , mouse anti-disclarge ( Developmental Studies Hybridoma Bank , University of Iowa , USA ) was used at a 1:200 dilution and incubated overnight at 4° Celsius in blocking solution . After washing twice with PBS-0 . 5% Triton X , the tissue was incubated with the secondary antibody , 1:200 anti-rat-CY3 ( Dianova , Hamburg , Germany ) . The brains were mounted in Vectashield mounting medium ( Vector Laboratories , Burlingame , CA , USA ) and scanned using confocal microscopy with a Leica SP5-2 . The images were analyzed using the StackGroom plugin in ImageJ [66] . Summary statistics , including the number of segregating sites ( S ) , number of haplotypes and Tajima’s D [34] were calculated using DnaSP v . 5 . 10 . 1 [67] . The mean pairwise nucleotide diversity ( π ) [33] , Watterson’s [32] estimate of nucleotide diversity ( θ ) and Fst [68] were calculated as described in [5] . Hudson’s haplotype test ( HHT ) was carried out using ms [69] to perform coalescent simulations and psubs [70] to calculate the probability of observing a subset of n sequences containing p or fewer polymorphic sites . The demographic models of Thornton and Andolfatto [35] , Duchen et al . [5] , and Werzner et al . [6] were used to simulate the out-of Africa bottleneck . To test for a selective sweep , a SweepFinder analysis was performed using the SweeD software [38] . The background site frequency spectrum ( SFS ) was calculated for the entire 3R chromosome arm using 11 whole genome sequences from the Netherlands population and one whole genome sequence from the French ( Lyon ) population [8] . The French sequence was included in order to have a constant sample size of 12 sequences for the calculation of the SFS . This approach did not bias the background , as the French sequence did not differ more from the Netherlands sequences than the Netherlands sequences did from each other ( S2 Table , S2a Fig ) . Furthermore , the inclusion of a French line did not lead to a skew in the background SFS ( S2b Fig ) . For the Swedish population , the background SFS of chromosome arm 3R was determined from 12 whole genome sequences from the Umeå population ( S3 Table ) . In order to increase the power of the test , the invariant sites in the alignment were also included [37] . To assess the significance of the composite likelihood ratio ( CLR ) statistic , neutral simulations were performed using ms [69] . In the neutral simulations three demographic models were taken into account [5 , 6 , 35] . These models differ in several parameters , including: the timing of the out-of-Africa bottleneck , the current effective population sizes of the European and African populations , and the ancient demographic history of the African population . For our analyses , it is the estimated time of the out-of-Africa bottleneck that has the largest impact on the results . Duchen et al . [5] infer this bottleneck to have occurred around 19 , 000 years ago , Thornton and Andolfatto [35] around 16 , 000 years ago , and Werzner et al . [6] around 13 , 000 years ago . However , the 95% confidence intervals of the estimates are very wide , ranging from 7 , 359–43 , 000 years ago . Thus , the three estimates are not incompatible with each other . The recombination rate of the MtnA genomic region was obtained from the D . melanogaster recombination rate calculator [71] . A total of 10 , 000 simulations were performed . For each simulation , the maximum value of the CLR statistic was extracted and used to determine the 5% significance threshold . Linkage disequilibrium was calculated between all pairs of SNPs present using Lewontin’s r2 = D2/p1q1p2q2 , where D is the frequency of the haplotypes and p and q represent the allele frequencies [39] . A fragment of ~100 kb flanking the MtnA locus ( 3R: 9 , 732 , 746 . . 9 , 835 , 406 ) was analyzed , with singletons excluded . A Fisher’s exact test was used to assess significance of the r2 values . Copper sulfate and hydrogen peroxide tolerance assays were performed using five D . melanogaster lines containing the MtnA 3’ UTR deletion ( two Dutch and three Malaysian lines ) and three lines without the deletion ( two Malaysian and one Dutch line ) , as well as an MtnA knockdown line ( RNAi-MtnA/Act5C-GAL4 and its control ( control/Act5C-GAL4 ) . Assays were performed at 25°C in tolerance chambers consisting of a plastic vial ( diameter = 25 mm , height = 95 mm ) with compressed cotton at the bottom containing 2 . 5 ml copper sulfate ( Sigma Aldrich ) or hydrogen peroxide ( Sigma Aldrich ) solution supplemented with 5% sucrose and sealed with a cork . Four to six day-old flies were separated by sex and tested in groups of 20 . For each assay , one concentration of copper sulfate ( 50 mM ) or two concentrations of hydrogen peroxide ( 5 or 10% ) were tested with 5–7 replicates per sex and concentration . A control solution containing only sucrose was also tested with 3–5 ( 10–15 for Act5C-GAL4 background ) replicates per sex for each assay . Mortality was recorded as the number of dead flies after 48 ± 1 hours . To determine the effect of the deletion , lines with and without the deletion were compared within each population or background . For copper sulfate assay analysis , t-tests were performed to assess significance . In order to account for potential differences in mortality inherent among the lines , proportional mortality data was scaled by mortality at 0 mM using the formula mortality/ ( 1 + mean mortality at 0mM ) . For hydrogen peroxide assay analysis , the data for each assay and population was fit to a generalized linear model ( GLM ) using concentration , line , sex , and presence of the deletion as factors and a quasibinomial distribution using the glm function in R [62] . The tolerance results for each sex ( S3 Fig ) and the GLM coefficients ( S4–S12 Tables ) are provided as supporting information .
|
Although molecular variation is abundant in natural populations , understanding how this variation affects organismal phenotypes that are subject to natural selection remains a major challenge in the field of evolutionary genetics . Here we show that a deletion mutation in a noncoding region of the Drosophila melanogaster Metallothionein A gene leads to a significant increase in gene expression and increases survival under oxidative stress . The deletion is in high frequency in three distinct geographic regions: in northern European populations , in northern populations along the east coast of North America , and in southern populations along the east coast of Australia . In northern European populations the deletion shows population genetic signatures of recent positive selection . Thus , we provide evidence for a regulatory polymorphism that underlies local adaptation in natural populations .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"invertebrates",
"evolutionary",
"biology",
"3'",
"utr",
"oxidative",
"stress",
"messenger",
"rna",
"geographical",
"locations",
"population",
"genetics",
"reporter",
"genes",
"animals",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"untranslated",
"regions",
"gene",
"types",
"population",
"biology",
"drosophila",
"africa",
"research",
"and",
"analysis",
"methods",
"gene",
"expression",
"zimbabwe",
"insects",
"arthropoda",
"people",
"and",
"places",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"netherlands",
"natural",
"selection",
"genetics",
"biology",
"and",
"life",
"sciences",
"europe",
"evolutionary",
"processes",
"organisms"
] |
2016
|
An Indel Polymorphism in the MtnA 3' Untranslated Region Is Associated with Gene Expression Variation and Local Adaptation in Drosophila melanogaster
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.